XTERM – Change Log

Copyright © © 1997-2021,2022 by Thomas E. Dickey


Here is
the latest version of this file.

It began as a list of the changes that I made for xterm, using
the notes that I added when submitting a patch. You should note
that other changes were made as well, by other people, to fix
bugs and correct ifdef’s for portability. Until mid-2006, most of
these were summarized in the XFree86

  • amend allocation/freeing of scrollback lines, eliminating
    an adjustment for status-line added in patch #371 (report/testcase by Rajeev V.
  • drop double-buffer default from test-packages, since that
    interferes with status-line.
  • add faceSize7 to table in
    charproc.c, overlooked in patch #360 (patch by Vladimir A. Pavlov).
  • use XDrawString rather than
    XDrawString16 for text when the latter is not
  • improve performance for initializing/reusing graphic
    objects in the SIXEL feature.
  • improve error-checking for resize.
  • fix errata in manual page (Debian #988221).
  • extended list of environment variables to purge on startup,
    i.e., in case a user starts xterm from another terminal
  • add comment in cursor.c to explain why
    autowrap (DECAWM) does not apply to data
    saved/restored with save/restore cursor-operations (report by
    Thomas Wolff).
  • add special case in generating the combining-character
    table in wcwidth.c, for Hangul Jamo Extended-B
    (report by Luis Javier Merino).
  • simplify/improve portability of trap commands
    in shell-scripts.
  • update configure options
    --disable-tcap-fkeys and
    to account for changed default settings.
  • rather than ignore empty parameter for setting title, use
    that to reset it to “xterm” for scripting.
  • check for out-of-bounds condition while drawing sixels, and
    quit that operation (report by Nick Black).
  • add fallback definition for PROJECTROOT so
    that a suitable default location for luit will be
    compiled-in on build-servers.
  • implement DEC status-line controls DECSASD/DECSSDT:
    • decode controls, set state
    • implement DECRQSS responses
    • implement indicator-style status-line
    • add configure option
    • add resource indicatorFormat
    • implement host-writable status-line, able to handle
      video-attributes and simple left/right cursor movement.
      This disallows all of the DEC/xterm private modes such as
      switching to/from the alternate screen.
    • move the status-line if the window is resized.
    • Note: the associated screen-resizing does not
      work for the special case of TrueType fonts with
      double-buffering (see note in patch #349).
  • modify configure-script fixup for -Werror
    options to exclude fixup for -Werror=format, e.g.,
  • amend change from patch #370; DEC
    070’s pseudo-code for DECCOLM incorrectly suggests that it
    enables DECLRMM as a side-effect.
  • fix some memory leaks seen using esctest and asan2 (patch
    by Luis Javier Merino).
  • add -a, -c and -d
    options to query-status.pl to improve test for
  • expanded discussion of maximum graphics geometry in
    ctlseqs.ms (suggested by Bon Wong).
  • corrected parameter symbol for DECCARA and
    DECRARA in ctlseqs.ms to indicate
    that multiple parameters may be used (report by Thomas
  • add several test/demo scripts.
  • improve DECCOLM, DECLRMM and
    DECALN referring to pseudo-code in DEC 070.
  • amend a change in patch #348 which caused left/right margin
    mode to be reset when resizing the terminal window (report by
    Valentine Barshak).
  • fix a misnamed macro-parameter (patch by Rajeev V.
  • fix spacing of wideFont when its width is not
    exactly twice as wide as the normal font (report by Rajeev V.
  • suppress loading of italic font in a few places when
    colorITmode is enabled (report/analysis by Rajeev
    V. Pillai).
  • modify XTSMGRAPHICS to return failure status
    if the terminal is not configured to support the corresponding
    ReGIS or SIXEL feature (report by Nick Black, notcurses

    #2252, notcurses #2257).

  • modify DECERA and DECFRA to erase
    corresponding area in SIXEL graphics (patch by Nick Black,
    notcurses #1740,
    vt340test #16).
    Also modify DECSERA.
  • equate visuals for TrueColor and DirectColor (patch by
    Denis Kaganovich).
  • correct computation for pixel value of rgb when using depth
    16 (report by Denis Kaganovich).
  • correct wrapping of VT100-style double-sized characters
    when configured for Unicode wide-characters (report by Luis
    Javier Merino).
  • updated default value for sixelScrolling
    resource to match expected behavior versus DECSDM
    change in patch #369 (report by Grant
  • modify run-tic.sh to work around bug in
    development version of ncurses which was packaged in FreeBSD
  • remove ifdef’s for OPT_COLOR_RES and
  • improve performance over slow connections (report by Harald
  • update cursor if restoring mode for DECTCEM.
  • modify CharWidth macro to ensure that the
    shortcut for Latin-1 is only applied when UTF-8 is not enabled,
    to fix a bug in handling soft-hyphen from patch #334 changes (patch by Martijn van
  • improve terminfo:
    • fill-in function-keys in terminfo which are not Sun/HP
      keyboards using xterm+nopcfkeys
    • add kbeg to xterm+keypad to accommodate termcap
    • add smglp and smgrp to vt420+lrmm, to provide useful
      data for the “tabs” +m
  • support shift-tab in Sun, HP and SCO keyboards.
  • document some legacy features in ctlseqs.ms
    (prompted by discussion with Jimmy Aguilar Mena “Ergus”).
  • add “trim” option to cdXtraScroll and
  • remove support for non-fifo save-lines configuration.
  • extend cdXtraScroll to check if the cursor is
    at the upper-left of the scrolling region when the erasure is
    for the remainder of the screen versus the whole screen
    (prompted by discussion with Jörg Breitbart).
  • add workaround for broken pcre2 package in Debian 10.
  • change screen-refresh call used for DECCARA
    and DECRARA to ensure that trailing blanks which
    are part of the rectangle are repainted (report/analysis by
    Dennis Filder).
  • when resetting the terminal, ensure that the cursor shape
    also is reset, e.g., if DECSCUSR has been used to
    modify the cursor shape for an xterm which was started with the
    underlined cursor option (report/analysis by Luis Javier
  • prevent DECSCUSR from blinking the cursor if
    the cursorBlink resource is “never” (report by
    Vladimir D Seleznev).
  • invert the sense of DECSDM, to correspond with
    VT382 manuals (lsix #41).
  • update tables in wcwidth.c based on Unicode
  • add DefaultOff option to
    RenderFont resource, as part of the
    session-management feature.
  • add auto-scroll-lock feature (patch by Stelios
  • update the window-size information returned via
    TIOCGWINSZ when rows/columns are unchanged but the
    font-size changes (report by Nick Black).
  • improve session-management feature by saving/restoring the
    font settings.
  • update config.guess, config.sub
  • add OSC 22 to allow programs to select
    different pointer cursor at runtime.
  • change configuration for no-return functions to
    use _Noreturn when it is available, because
    clang --analyze does not properly handle the gcc
    noreturn attribute.
  • add cursorTheme resource to provide a way to
    enable or disable the cursor theme feature.
  • modified CopyWait event retries to use shorter
    sleeps, to improve responsiveness (tmux #2556).
  • improve quoting/escaping in demo-scripts per
  • add resizeByPixel resource, to permit
    disabling window manager resizing-hints (patch by Tim
  • corrected printOptsImmediate handling of
    alternate-screen (report by Abhijit Dasgupta).
  • update sample terminfo to more closely match ncurses.
  • add/improve limit-checks for Xlib calls (report by Roman
  • fix a typo in the help-message (report by Tomas
  • correct a compiler-warning fix in patch #352 which allowed sign-extension of
    coordinate values (report by “CismonX”).
  • correct upper-limit for selection buffer, accounting for
    combining characters (report/testcase by Tavis Ormandy).
  • with alwaysHighlight true, xterm does not
    properly track focus. The screen->select
    FOCUS flag remains always on, which prevents
    bellIsUrgent from working, as the urgent
    WM_HINT flag is only set in
    setXUrgency() when the window is not focused. Fix
    this by updating screen->select in
    unselectwindow() regardless of the value of
    always_highlight (patch by Jiri Bohac).
  • improve fix for interaction between SRM and ENQ (report by
    Grant Taylor).
  • build-fix for --with-Xaw3dxft, needed when
    --with-toolbar is omitted (report by Jimmy Olgeni,
    Emanuel Haupt).
  • amend fix for “word” selection in patch #364 to limit that to the
    insert-selectable action, which reads
    data from the screen. This restores the interactive behavior
    where double-clicking on a “word” would make subsequent
    selection extensions by words as well as suppressing some
    boundary-checks (report by David Wolfskill, FreeBSD #253225).
  • add -fc option.
  • correct/improve limit-checks for SRM versus ENQ from
    patch #344 (report by Tom
  • enable XftFont resource in Xaw3dxft
    configuration (patch by Tavis Ormandy).
  • improve quoting/escaping in build-scripts per
  • add libpcre2-posix to the packages tested for
    --with-pcre2 option, needed with Fedora (report by
    Tomas Korbar).
  • correct a typo in manual page, and note that
    KeepClipboard may not be compiled-in (report/patch
    by Sean C Farley).
  • corrected boundary-checks for “word” selection used in
    onNClicks resources (report by Tavis
  • update to autoconf-2.52-20210101, to improve shellcheck
  • improve configure check for desktop categories.
  • disable groff hyphenation in generated html when using
  • change SCS “&4” Cyrillic to non-NRCS, per
    VT520 manual.
  • amend fix for SCS in patch #198 to
    remove DEL rather than converting it to a space
    (report by Thomas Wolff).
  • modify state transitions for selecting character sets to
    eliminate an ambiguity between the “A” used in VT220 versus
  • improve error recovery when selecting characters by
    resetting to US ASCII when no suitable encoding is found, e.g.,
    attempting to use an NRCS sequence when NRCS is not
  • turn off hyphenation in “.txt” conversion,
    prompted by groff changes.
  • add VT5xx ISO Latin-2 (prompted by discussion with Thomas
  • amend change from patch #361 to
    event-handling in CopyWait to handle active-icon
    as a special case (Debian #975687).
  • add call to loadColorTable to get the visual
    information needed to decode BE RGBA32 color format (patch by
    Leandro Lupori).
  • modify the resource-parsing for
    disallowedPasteControls and similar lists of
    names/numbers to recognize “~;” (tilde) for cancelling a given
  • add ENQ, EOT and NUL to default for
  • change default for disallowedPasteControls to
    omit the tab character (suggested by Harald Dunkel).
  • fix swapped height/width in regisScreenSize
    and maxGraphicSize resources using “auto” value,
    from patch #314 (report by Anton
  • revise patch #362 change for
    shift-modifier versus mouse protocol and select/paste by adding
    resource shiftEscape, which can be enabled via a
    control sequence (prompted by discussion with Matthijs van
  • simplify/improve ifdef in trace code for using a separate
    set of files for each run of xterm.
  • add options -r and -t to
    vttests/query-color.pl to show the 6-digit RGB
    code and actual colors.
  • add nel to xterm-basic
  • alter vttests/modify-keys.pl to omit HTML
    reference links to the modified-keys table where no keycode was
  • cleanup of calls to free, removing checks for
    null (Walter Harms).
  • improved mouse-button reporting (prompted by discussion
    with Stephane Chazelas)

    • narrow the scope of the change for shift-key in
      patch #361 to make it apply only
      when the modifyOtherKeys resource is set to 2
      (i.e., “program mode”). Also, when checking the shift-key,
      ignore modifiers other than shift, control and “meta”
    • use the alt/meta modifier information obtained in
      VTInitModifiers to replace a hard-coded
      mod1 used to detect “Meta” for mouse-button
  • reduce SIGWINCH‘s sent to the client by
    filtering out duplicates.
  • improve display when scaleHeight is greater
    than 1:

    • the text-cursor is vertically-centered on the current
      line, rather than only extending below the current line
      (report by Manu Chaturvedi).
    • the built-in line-drawing characters extend to the
      scaled cell-height.
  • fill-in special case for motion-events to match the changes
    for shift-key in pointer-button events from patch #361.
  • treat the return value of strerror as readonly
    (patch by Philipp Klaus Krause).
  • modify event-handling in CopyWait to work
    around hanging while writing large amounts of text to an active
    icon and at the same time iconifying/deiconifying (report by
    Dave Kemper).
  • initialize double-buffer for active-icon window.
  • improve manual page description of scrollbar resources
    (report by Brian Lindholm).
  • correct ifdef for menu entry for active-icon, when
    initializing it with toolbar configuration.
  • add preprocessor option to makefile to set internal
    definition of pixmaps directory to match the
    install-configuration (FreeBSD #250036).
  • cleanup of calls to free, removing checks for
    null (Walter Harms).
  • add fallback actions pointer-button and
    pointer-motion which handle events for the mouse
    control-sequences protocol if the select-related
    translations are omitted with
    *omitTranslation:select (prompted by discussion
    with “Ergus”)
  • amend rule for using shift-key to override mouse-protocol
    for select/paste to limit that feature to mouse-buttons which
    are actually bound to select/paste actions (prompted by
    discussion with “Ergus”).
  • corrected mapping in special case for repainting wrap-marks
    when viewing the scrollback area.
  • use separate GCs for showWrapMarks
    feature, to work around cursor coloring change in patch #345 (report by Paulo Silva de
  • mention decGraphicsID in
    ctlseqs.ms (suggested by Thomas Wolff).
  • modify pixel-coordinate mouse response to use as origin the
    VT100-window rather than the underlying widget, which includes
    the scrollbar (report by Thomas Wolff).
  • add configure option --disable-print-graphics
    (Ross Combs).
  • add test_ptydata to “make check” rule.
  • minor fixes for ctlseqs.ms (report by
    Jean-Marc Bourguet).
  • minor fixes for manpage style (report/patch by
  • correct cleanup from switch between italics/normal font in
    the show/hide cursor functions (report/testcase by Peter
  • integrated patch by Ross Combs:
    • internal renaming of GraphicsID symbols, for
    • adjust logic for VT125, treating it as a ReGIS
  • update configure macros, for compiler-warning fixes.
  • integrated patch by Ross Combs:
    • add control sequence modes for graphics printing.
    • improve conversion to bitmaps for TrueType fonts in
    • add font7 resource, for an “enormous”
      bitmap font.
    • do not reset graphics in a soft-reset.
    • add pointerFont resource, and
      -pf command-line option.
  • improve typography of manual page (patch by Branden
  • amend patch #359 change to
    df-install.in to ignore a command-line assignment
    to DESTDIR if followed by an explicit
    --dir (report by Sven Joachim).
  • add special case in WriteText to allow colors
    8-15 to override colorBDMode (patch by Ingo
  • add utf8Weblike resource, to provide an
    alternate scheme for handling ill-formed UTF-8 sequences
    (adapted from patch by Dan Gohman).
  • improve computation for the number of lines needed to
    scroll-up a SIXEL graphic (report/patch by Ben Wong).
  • correct manpage description for default value of
    disallowWindowOps from changes in xterm #331 (patch by Ben Wong).
  • correct a loop starting-point in
    refresh_graphics from optimization in patch #358 changes (report by Ben Wong).
  • add a new mouse mode 1016, which uses the same format as
    mode 1006, but sends the mouse’s position in pixels (suggested
    by Igor van den Hoven).
  • fix an issue from patch #338
    changes where only the first selection buffer specified in the
    request would be updated using OSC 52 (patch by Michael
  • modify makefile/scripts to allow DESTDIR to
    prefix the target directory for desktop-file-install (report by
    Fred Heitkamp).
  • enable SIXEL feature by default.
  • update config.guess, config.sub
  • correct logic for decodeTerminalID changes in patch #357 (report by “Chartreuse”).
  • modify makefile to use plink.sh when linking
    test-programs, to fix build when using pcre (report by H Merijn
  • build-fix for test_ptydata program (patch by H Merijn
  • several minor optimizations for the ReGIS and SIXEL
    features, improving performance by 10%.
  • add resource decGraphicsID to allow displaying
    graphics when the emulation level would ordinarily disallow
    this (prompted by discussion with Thomas Wolff).
  • add control sequences for fast switching of color palettes:
  • amend change for soft-hyphen from patch #328 to avoid stripping
    replacement-characters which would be shown with malformed or
    overlong UTF-8 input.
  • corrected an error-handling case in
    decodeUtf8, matching a similar fix in patch #268 (report/patch by Dan Gohman).
  • add a test-driver for ptydata.c
  • minor cleanup of macros (adapted from patch by Walter
  • fix some errata in ctlseqs.ms (report by
    Thomas Wolff).
  • allow immediate repaint-on-palette-changed if
    double-buffering is enabled.
  • deprecate codes 10/11 in sgr push controls, changing those
    to 30/31, to avoid confusion with sgr 10-19.
  • modify SGR parameter handling to stop if an
    unrecognized parameter is encountered, to guard against
    malformed or nonstandard sequences (report by Bram
  • modify DECERA color for consistency with other
    erasures/clearing (report by Thomas Wolff).
  • ECH should not be masked by
    DECSCA (report by Thomas Wolff).
  • extend DECFRA and REP to accept
    any “graphic” character rather than just Latin1, etc. (report
    by Thomas Wolff).
  • add -C option to 256colors2.pl and
    88colors2.pl, to demonstrate mixed semicolon/colon separators
    which are implied by ECMA-48.
  • update sample terminfo to reflect the documentation
  • update description of 88/256/direct color in
    ctlseqs.ms to point out that using semicolons is a
    deprecated legacy feature, and standard terminal applications
    should use colons (prompted by discussion with Bram
  • modify configure-check for tgetent to
    conditionally include termcap.h, enabling
    configuration using clang‘s pedantic-errors option
    (report by Dennis Clarke). See
    Other Compatibility in ncurses’
  • remove some unnecessary pointer checks (patch by Walter
  • accept terminal-id and add DA response for VT131,
  • revise fix for Debian #954730,
    which interfered with wheel mouse events (report by Gabriele
  • revise fix for Debian #954730,
    which interfered with wheel mouse events (report by Henri
  • fix typos in documentation (reports by Stephen Hurd, Stefan
  • add mapping for decTerminalID for “100”
    overlooked in patch #354.
  • update tables in wcwidth.c based on Unicode 13.0.0
  • build-fix for “make check” when building out-of-tree
    (report by Sven Joachim).
  • work around performance problems of
    XDrawImageString and
    XDrawImageString16 functions (Debian #954845).
  • add a control sequence which reports xterm’s version (patch
    by Nicholas Marriott, mintty #881).
  • temporarily set numeric locale category to “C” when parsing
    resources, so that scaleHeight and
    faceSize settings do not depend on locale (Debian
  • improve DA/DA2 response by ensuring that the
    decTerminalID maps to one of the known
    identifiers, as well as providing DA2 response for VT241 and
  • terminfo improvements:
    • add (my) comments from ncurses which explain the keypad
    • add vt52+keypad from ncurses
    • use improved xm example for
      xterm+x11mouse, xterm+sm+1006
      from ncurses 6.2 terminfo.src
  • two fixes for left/right wheel mouse event reporting
    (Debian #954730):

    • filter identical button-events
    • correct order of button-range versus protocol type (see
      patch #345)
  • change “make check” makefile-rule to use test-drivers for
    charclass and wcwidth data.
  • quiet “did not find a usable xxx TrueType font”
    warnings by making fontWarnings apply to these
    messages (report by Jim Rees).
  • improve reinitialization of parameter list (report/testcase
    by James Holderness).
  • temporarily set numeric locale category to “C” when
    formatting SVG or XHTML screendumps, to make the radix
    separator used in RGB values consistent (adapted from patch by
    George Kouryachy).
  • add resource forceXftHeight to control whether
    workaround from Debian #880407
    is used.
  • apply updated ascent/descent in workaround from Debian
    to fix a 1-pixel gap in built-in vertical lines
    (report/testcase by Stefan Assmann).
  • improve round-off of scaling for built-in line-drawing
    (prompted by discussion with Stefan Assmann).
  • adjust fonts in svg-icon files to accommodate reduced
    functionality of new pango (report/analysis by YOKOTA
  • improve configure check for X Toolkit library.
  • correct Y-coordinate transformation in
    ClearCurBackground, overlooked in changes for
    patch #334 (report/analysis by Chuck
  • remove --vendor option from test-packages’
    install of desktop files; the feature is badly broken in
  • modify uxterm to make it possible to select
    nonstandard locale C.UTF-8, e.g, if the user’s locale
    is set to “C” (Debian #940626).
  • re-save/tweak “.svg” icon-files to work around breakage in
    toolset since the files were created in patch #283.
  • amend change in patch #352 for
    button-events to fix a case where some followup events were not
    processed soon enough (report/patch by Jimmy Aguilar
  • handle MappingNotify X event, to improve recovery
    when switching keyboard configurations using
    xkbcomp (prompted by discussion with Frank Mosch,
    Debian #661295).
    There is more work needed here, possibly in the X
  • improve discussion of mouse-mode in ctlseqs.ms
    (suggested by Igor van den Hoven).
  • further improve checks for Xft max-advance-width
    to take into account fonts which use two cells for ambiguous
    width characters. Also improve the time used for these checks
    (reports by Yuri Pankov, Frank Mosch).
  • fix a few spelling errors reported by codespell (report by
    Jens Schleusener).
  • modify run-tic.sh to prefer development
    version of ncurses since changes to terminfo file in patch #345 rely upon bug-fixes in ncurses
    (prompted by discussion with Will Senn).
  • adjust fontsize data to handle a minor inconsistency from
    recent Xft versions (Debian #880407,
    adapted from patch by Vincent Lefèvre).
  • add a table to the manual page description of
    forceBoxChars to alert the reader to the special
    characters aside from “line-drawing” which are drawn directly
    when this resource is set (Debian #931305).
  • improve checkXft logic which attempts to
    detect fonts whose max-advance-width is inconsistent
    with the actual glyph widths. For some fonts, it is necessary
    to check additional characters (report/analysis by Jan
  • improve configure-checks for X headers and libraries on
    recent MacOS, which has moved those files under
  • improve portability of iconify/deiconify feature by taking
    into account some window managers which manipulate the EWMH
    _NET_WM_STATE property, adding/removing
    _NET_WM_STATE_HIDDEN rather than actually
    minimizing the window (prompted by discussion with
    Jörg Breitbart).
  • improve workaround from patch #287
    for the -iconic option when configured with
    toolbar by postponing the extra request for minimizing the
    window to the end of menu-initialization.
  • modify xevents special-case for mouse-events
    to include button-events so that the meta key by itself can
    generate button-events (report/analysis by Mattias
  • amend SGR-stack change from patch
    #348 to not associate bold attribute with background color
    (report by Nicholas Marriott).
  • fix copy/paste error in manual page (patch by Larry
  • add definitions in xterm_io.h so that GNU/Hurd
    will use posix_openpty (patch by Samuel
  • build-fix in debug-tracing, for esctest.
  • updated autoconf macros
  • update config.guess
  • correct logic in property_to_string for
    deciding when to fallback from UTF-8 decoding to ISO-8859-1
    decoding, broken in xterm #350
    (FreeBSD #241961).
  • add -report-icons to help-message.
  • improved autoconf macros:
    • CF_ADD_LIBS: the change to filter out
      duplicates caused this to append rather than prepend.
      revise to fix that.
    • CF_GCC_VERSION and
      CF_GCC_WARNINGS: move checks to distinguish
      icc/clang from gcc from the macro which handles the
      –enable-warnings option, to make this work
      without-warnings for the inline-checks.
  • update config.guess, config.sub
  • correct status in XTGETXRES replies when the
    resource was not found.
  • fix some gcc, cppcheck, clang and coverity warnings.
  • guard call to RequestResize from the
    struct-notify event handler to prevent recursion in the
    Xft+buffered workaround in some cases when doing manual
    resizing rather than resizing via escape sequences (reports by
    Stefan Assmann, Mike Thornburg).
  • amend the workaround for Xft+buffered blanking by moving
    the switch to bitmap-fonts to account for differences in font
    metrics between bitmap- and TrueType-fonts (report by Stefan
  • improve the note on the xterm-rep terminfo
    entry (prompted by discussion with Sven Joachim).
  • modify html/svg dump to not ignore zero’d/blank cells.
  • align terminfo file with ncurses, e.g., add
    xterm+osc104 block.
  • improve discussion of modifyOtherKeys in
    manual page and in ctlseqs.ms.
  • add vttests/modify-keys.pl script to
    illustrate the modifyOtherKeys resource (prompted
    by discussion with Bram Moolenaar).
  • various improvements to ctlseqs.ms, as part of
    autogenerating links for the website.
  • update manual page default for saveLines
    resource default value (Branden Robinson, Debian #913815).
  • add command-line option -report-xres to show
    the values of the VT100 widget X resources when initialization
    is complete.
  • add a control sequence which, like tcap-query, allows an
    application to inspect most X resource settings of the VT100
  • adjust some optional features to enable them in the imake
    configuration as they would be by default via the configure
  • add null-pointer checks to improve error recovery when
    bitmap fonts are missing or corrupt (report by Jonne
  • correct the condition for deleting the EWMH window-title
    property, i.e., if UTF-8 titles are disabled, rather than if an
    update to the EWHM property found no change (report by Sven
  • build-fix for the case when configure
    --enable-trace is used without
    --enable-warnings (report by Sven Joachim).
  • fix a few minor bugs found with Coverity.
  • add a check in property_to_string to avoid
    translating UTF8_STRING or
    COMPOUND_TEXT into Latin-1 when UTF-8 encoding is
    active, e.g., when pushing onto the title-stack while switching
    to the alternate screen.
  • build-fix for the --disable-doublechars
    configure option (report by Brian Lindholm).
  • add graphic context to support bold+italics (patch by Quinn
  • document window properties in the manual page.
  • improve title-string feature:
    • if any of allowC1Printable,
      utf8Title or titleModes hint that
      an application might send a title-string encoded in UTF-8,
      check if that is the case, and if it is recodable into
      ISO-8859-1, use that for the ICCCM-style title.
    • check if the title given by a control sequence happens
      to be already encoded in UTF-8, to avoid double-encoding
      (FreeBSD #240393).
    • Make sameName resource work for the EWMH
    • Modify menu-state of utf8Title to be
      consistent with the utf8 source, i.e., setting
      the EWMH properties automatically when UTF-8 is
  • reorganize text-drawing to make it
    possible to investigate using Xft to implement VT100-style
    double-sized characters. While doing this, made a workaround
    for apparent Xft bug which loses its drawable state when
    switching from 132 to 80 columns.
  • improve font-warning messages by showing which are
    derived rather than directly from resource settings
    (suggested by Tomas Korbar). Also filter repeated font-warning
    messages, to accommodate broken X configurations.
  • fix an inconsistency between failure to load derived wide
    font versus failure to load derived wide-bold font (Redhat
    That relies upon the “:unscaled” property which is broken in
    some distributions (however, recent Debian and the BSDs such as
    MacOS work).
  • updated autoconf-252 to check X11R7 include/lib directories
    found on some older configurations.
  • set a graphic-context for border when double-buffering is
    active; to prevent the border color from changing when
    switching to reverse-video.
  • build-fix for --disable-ziconbeep, which
    conflicted with the new double-buffer configuration (report by
    Brian Lindholm).
  • fix loop-limit for lookup of fullscreen
    resource broken in xterm #347 (report
    by Scott Bertilson).
  • update window-manager hints when exiting Tek4014 mode
    (Debian #932569,
    patch by Jonathan Irwin).
  • fix a misformatted printf in
  • add configure check for termios types, to improve
  • ensure that when resetting margins, to also reset
    DECLRMM. This affects DECCOLM,
    DECALN (although DEC STD 070 mentions only
    top/bottom margins), and DECSTR.
  • corrected order of reset/move when setting
    DECCOLM, and make it more consistent by always
    resetting margins, rather than only when the mode is changed
    (report by James Holderness).
  • compile-in double-buffer support by default, changing the
    configure option to set the default resource value for
    buffered to true or false.
  • take in account the reverse-video state when computing the
    filler-color used when clearing the screen in double-buffering
  • correct logic for filtering scrollbar-updates when
    buffered resource is compiled-in but not enabled
    (report by Paul Lampert).
  • improve state saved/restored for cursor-save and SGR stack
  • improve description of 1006 and 1005 mouse modes, to avoid
    implying that they use character-parameters (report by Bryan
  • explain in ctlseqs.ms that some of the numeric
    keypad keys were xterm extensions rather than VT100/VT220
    terminal features (prompted by discussion with Thomas
  • explain in ctlseqs.ms how the DEC windowing
    extension is supported by xterm.
  • add 28 rectangular editing to the primary response
    (suggested by Thomas Wolff).
  • fix a typo, improve wording in ctlseqs.ms
    (Thomas Wolff).
  • fix internal column-parameter when SL or
    SR is used with left/right margins (patch by
    Thomas Wolff).
  • fix off-by-one in VT52 graphics character mapping (patch by
    Thomas Wolff).
  • use _X_UNUSED, etc., when available as a
    fallback for GCC_UNUSED, etc., to reduce compiler
    warnings when building with imake.
  • update config.sub
  • fix a few minor bugs found with Coverity.
  • improve double-buffering for scrollbars combined with
    scrolled text; add bufferedFPS resource to control
    the maximum rate of screen updates (report by Mike
  • improve fixes for DECCRA handling of
    double-width characters (patch by Martin Hostettler).
  • improve discussion of ECMA-48’s typographical error for
    SD in ctlseqs.ms (prompted by report
    by Martin Hostettler).
  • correct off-by-one in parameter limit-check for
    DECCRA (report by Martin Hostettler).
  • modify saveCellData to handle case where
    double-width character is partially copied; that should be
    blanked (report/testcase by Thomas Wolff).
  • add resource buffered to allow
    enabling/disabling double-buffered mode.
  • two fixes for the double-buffer configuration, prompted by
    MacPorts’ switch to double-buffering (patch by Mike Thornburg,
    MacPorts #58313):

    • ensure that the needSwap flag is set after drawing
      TrueType text
    • corrected the drawable-parameter used for the
  • modify ScrnRefresh to ignore a case where the
    left/right halves of a double-width character have been set to
    different video attributes. The attribute to use is in the
    left-half (report/testcase by Thomas Wolff).
  • correct a limit-check for DECCRA in case the
    target lies off-screen (report/testcase by Thomas Wolff).
  • documentation errata (report by Thomas Wolff).
  • reset flags including wraparound and reverse-wrap when
    switching to VT52 mode, while noting that DEC’s standard
    documentation leaves that behavior undefined (report
    by Thomas Wolff).
  • ensure that italic font is turned off on hard/soft resets
    (report by Martin Hostettler).
  • improve responsiveness to X events while processing HTML or
    SVG dumps (report by Martin Hostettler).
  • replace logic in wcwidth.c for detecting
    double-width characters with binary-search table generated
    using updated uniset (report by Robert Ross).
  • add test-driver for wcwidth.c to simplify
    comparison with system’s wcwidth.
  • ensure that window-manager name comparisons work when
    active-icon is enabled, since CSI13t uses
    the window-manager name (report by Glenn Golden, Arch #62818).
  • trim a stray “experimental” in one of the comments about
    the tcap-query feature, which has been a supported feature
    since 2008 (see patch #238).
  • update description of the “default” setting for the
    renderFont resource to match the behavior in
    patch #261 (Debian #862042).
  • account for internalBorder in
    useBorderClipping (report by Robert Ross).
  • update table of unknown-width characters in wcwidth.c based
    on Unicode 12.1.0 (prompted by discussion with Robert
  • improve description of DECSCL versus S7C1T and S8C1T in
  • improve consistency between CSI3t and
    CSI13t, accounting for differences
    between some window managers’ handling of EMWH extents (report
    by Bram Moolenaar).
  • fix a sign-extension when reporting offscreen window
    position (report by Bram Moolenaar).
  • updated FreeBSD test-package from recent /usr/ports
  • workaround in run-tic.sh for HPUX, whose
    mktemp prints the name of a temporary directory
    without creating it.
  • add left/right margin capabilities to terminfo, since the
    default terminal emulation is VT420 (suggested by Thomas
  • modify treatment of reverseVideo for dynamic
    colors to make text foreground/background consistent with the
    cursor color (prompted by discussion with Ben Wong, lsix
  • modify button-handling to make SGR button-release distinct
    from button-press for button-codes starting with 8 (patch by
    Mikulas Patocka).
  • add aliases for the sgr push/pop controls to work around
    language limitations of C# (request by Dan Thompson).
  • correct a typo in setDirectFG, which could
    cause setting of indexed-color after a direct-color to be
  • add resource useBorderClipping as an
    alternative to useClipping (request by Robert
  • improve logic for displaying xterm’s built-in line-drawing
    characters vs missing glyphs, e.g., for U+2409 to U+240D (patch
    by Robert Ross).
  • add vttests/sgrPushPop2.pl to illustrate how
    selective pushes work, as well as to demonstrate push/pop of
    the various color types supported by xterm.
  • fix a special case in XTPOPSGR where
    direct-colors were not checked as a dependency of
    indexed-colors, contrary to documentation.
  • reduce buffer-flushes for OSC 4 and
    OSC 5 color-queries, as well as in the
    queries for dynamic colors.
    Note: this change makes the reply for
    OSC 5 use the “5” in the response; formerly
    it was mapped to ”4” after adding the maximum number of
  • modify vttests/query-color.pl to demonstrate
    multiple queries in one request.
  • update tables of combining and unknown-width characters in
    wcwidth.c based on Unicode 12.0.0.
  • modify response for OSC 5 to be
    consistent with documentation.
  • add vttests/query-dynamic.pl
  • modify vttests/query-color.pl to accept bold,
    underline, etc., keywords to decide whether to use
    OSC 5 rather than
    OSC 4.
  • modify cursor coloring to avoid a case where the cursor
    matched the foreground color while the text was displayed in
    reverse-video (report by Nuno Silva).
  • fix an inconsistency in X10 mouse responses from patch #342 changes; the legacy protocol
    supports only buttons 1-3.
  • reduce the number of buffer-flushes for tcap-query
  • modify tcapquery.pl to demonstrate batch
    queries with -q option.
  • increase response-buffer size, and make it configurable as
    limitResponse (prompted by discussion with Stephen
    P Wall).
  • modified configure script:
    • check/workaround for non-POSIX manipulation of
      predefined symbols
    • check for updated X Toolkit, which uses
  • update config.guess, config.sub
  • provide for secondary-selection, overlooked in patch #338 (report by Emile LeBlanc).
  • documentation errata (patch by Larry Hynes).
  • improve pointer-checks in Tekproc.c
  • revise solution for Debian #919475,
    to allow for struct-notify events occurring as a side-effect of
    toggling to/from full-screen mode.
  • window’s border-size was incorrectly added to position when
    maximizing window (report by Gary Langshaw).
  • add ASCII escape to default for
    disallowedPasteControls (prompted by discussion
    with Martin Hostettler).
  • fix typo in ctlseqs.ms (Robert Ross).
  • implement DEC Cyrillic NRCS (based on screenshots of vttest
    provided by Markus Schmidt).
  • correct control returning sixel geometry maximum versus
    actual size (patch by Ben Wong).
  • improve manual page for resize to clarify that
    resize does not execute the shell commands which
    it generates for setting environment variables.
  • improve memory-management for parser, to handle response
    strings that might occur with ENQ or when
    SRM mode is active (patch by Martin
  • change Comment text in uxterm’s desktop file
    to work around a limitation of GNOME’s GIO application
    #940). The
    documentation reads:

    Searches desktop files for ones that match

    The return value is an array of strvs. Each
    strv contains a list of applications that matched
    search_string with an equal score. The outer list is sorted
    by score so that the first strv contains the
    best-matching applications, and so on. The algorithm for
    determining matches is undefined and may change at any

  • a check of zIconBeep resource was removed in
    double-buffer fix in patch #334. That
    is used to decide whether to handle struct-notify events. When
    xterm does handle the events, it may negotiate a new
    window-size with the window manager. In that case, the
    limitResize resource setting limited the new
    window-size to the screen-size. Amended that by eliminating the
    limit for this special case when double-buffer is configured,
    and restoring the check for zIconBeep when
    double-buffer is not configured (Debian #919475).
  • disallow recursion which could happen if a user configures
    the answerbackString resource with the
    ENQ code and experiments with that code when
    SRM is enabled (report by by Martin
  • add remaining credits in COPYING file.
  • modify run-tic.sh to check if it is using an
    older version of ncurses which does not support large terminal
    descriptions, and if so, remove a non-essential feature to keep
    within the 4096-byte legacy limit.
  • modify run-tic.sh to prefer ncurses6 over
    ncurses5, when available.
  • add COPYING file, and dummy “check” makefile
    rule for testing Arch packages.
  • add/update package scripts for FreeBSD and NetBSD, for
  • update tables of combining and ambiguous-width characters
    in wcwidth.c based on Unicode 11.0.0.
  • modify xterm-new sample terminfo entry to
    correspond to ncurses 6.1, using the SGR 1006 mouse
  • improve manual page discussion of selection ownership
    versus highlighting (prompted by discussion with Martin
  • restore/repair keepClipboard feature which was
    broken by changes in patch #338
    (report by Martin Hostettler).
  • improve documentation for deleteIsDEL
  • modify DECRQM response for private mode 1037 to account for
    cases where the resource deleteIsDEL is initially
    neither true nor false, but dependent upon the keyboard type
    (report/analysis by Martin Hostettler).
  • clear pointers for pattern/fontset after destroying their
    data when switching facename with an escape sequence (report by
    Robert Ross).
  • modify logic for TrueType fallback fonts to match the
    sort-order used by fc-match (report by Robert
  • add resource setting limitFontsets which can
    be used to limit or disable the new TrueType fontset feature
    (request by Robert Ross).
  • limit mouse-button events to 11 buttons in the original
    protocol, and 15 in the extended mouse-protocol (patch by
    Martin Hostettler).
  • correct calculation for buttons past 11 in
    mouse-codes sample script (patch by Martin
  • disable a supplementary check added in patch #341 which interferes with using
    fontconfig to select different-sized bitmap fonts (Arch
  • add options to mouse-codes script to
    demonstrate that the encoding used for mouse events limits
    button numbers to 11.
  • allow mouse button numbers up to 11 using an offset of 128
    for button numbers 8-11 rather than the offset of 64 used for
    wheel-mouse, etc., buttons 4-7 in patch
    #338 (adapted from patch by Martin Hostettler).
  • correct off-by-one right-margin checks for double-width
    character adjustments in DECERA/DECFRA (patch by Martin
  • allocated size of selection buffer was not reset when
    freeing the buffer in patch #338
    (report/analysis by Stefan Assmann).
  • modify the initial pattern passed to fontconfig to disallow
    color bitmap fonts.
  • improve error recovery for a case where fontconfig ignores
    the requested pattern due to override in the user’s font
    configuration (report/testcase by Stefan Assmann).
  • suppress a spurious warning about fontsets when
    initializing menus for the toolbar configuration.
  • fix some strict compiler warnings.
  • improve checks for TrueType font, to help with the case of
    color bitmap fonts, which fontconfig misrepresents (Arch Linux
  • modify CF_GNU_SOURCE macro to treat
    newlib as similar to glibc, fixing a problem
    with the configure checks for pseudoterminals in Cygwin
    overlooked in patch #334 updates for
  • ignore $TERMINFO in the configure script if it
    is set to a non-directory value.
  • updated configure macros:
    • CF_WITH_PCRE2, modified to work with
      Debian’s (mis-numbered) pcre3 package.
    • CF_CC_ENV_FLAGS , report preprocessor
      options in CFLAGS
    • CF_LD_RPATH_OPT , suppress check if
      --disable-rpath option was given.
    • CF_XOPEN_SOURCE, add case for
  • update config.guess, config.sub
  • add a check to ensure that the filename parameter passed to
    XpmReadFileToPixmap is actually a file, since that
    function does not check (report by Ben Nott).
  • amend solution for Debian #758633
    to ensure that replies for bracketed paste are not sent while
    processing a selection for exec-formatted (Debian
  • improve adjustments/clearing for double-width characters in
  • improve adjustments/clearing for double-width characters
    when doing vertical scrolling within left/right margins (patch
    by Martin Hostettler).
  • modify wcwidth to encode a few spacing combining marks as
    regular combining characters.
  • change compiled-in default for saveLines to
    match the resource-file changed in patch
    #192 (Debian #913815).
  • change default faceSize to 8.0, to simplify
    switching back/forth between bitmap fonts and TrueType. This
    was originally (mis)set to 14.0 in patch
  • add fallback support in Xft configuration for missing
    glyphs (prompted by discussion with Mike Burns).
  • fix a problem with using direct-colors for the background
    color when clearing/scrolling (report by Nicholas
  • add a “sources” rule for the makefile’s generated
  • add a short explanation in ctlseqs.ms
    discussing the differences in error-recovery for malformed
    sequences versus unimplemented features, and amend one case for
    consistency (prompted by discussion with Martin
  • adjusted ifdef’s so that the paste64 configure option does
    not automatically enable the readline-mouse configure
  • revert the change which prevented concurrent ownership of
    different selection targets, and instead modify selection
    storage so that different concurrent requests for different
    selection targets will be stored/retrieved independently
    (Debian #901249).
  • remove a check which prevented returning button-codes past
    the documented 4/5 for wheel mouse, and add a script
    mouse-codes which makes a report showing the
    various mouse codes for different button/modifier combinations
    (discussion with Přemysl Janouch).
  • improve display and checksum for DEC Special Graphics by
    mapping 0x5f to 0.
  • add a null-pointer check for table-ending in the
    extended-boolean resource-handling (report by Felix von
  • remove stray ]‘s from INSTALL-file (Larry
  • regenerated configure script (report by Sven Joachim).
  • revise/improve fix for scrolling in margins
    (report/analysis by Martin Hostettler):
  • amend change to default-translations “select” subset to
    ensure that specific key-press actions are listed before
    generic key-press actions (reports by H Merijn Brand, Sven
  • ensure that only one of PRIMARY and CLIPBOARD is owned by
    xterm at a given time (Debian #901249).
  • documentation fixes (report by Lars Krueger):
    • document SL and SR.
    • document DECRPTUI, change number of digits
      to match VT420.
  • revise omitTranslation resource, e.g.,
    splitting “default” into several more useful categories.
  • modify mask-logic to permit control- and
    meta- modifiers to be used in mouse protocol.
  • several fixes for parsing/state (report/testcases by Martin

    • remove an old/unimplemented entry for xterm-title from
      state table.
    • save/restore wrap-flag when scrolling, to fix an
      inconsistency between jump-scrolling and
    • improve a special case of wrapping where the cursor
      starts beyond the right-margin, and wraps at the terminal’s
      last-column before the settable-margins can take
    • if origin mode is set, adjust a restored cursor column
      relative to the left margin.
    • when scrolling a line within left/right margins,
      correct order of the fixups needed when a double-width
      character crosses the margin boundary.
  • review/fix a few cases where the parsing state was not
    reset after completing a control sequence (prompted by report
    by Martin Hostettler).
  • disallow XTCHECKSUM if the corresponding
    window-ops “SetChecksum” is unset.
  • add print-vt-chars.pl script to illustrate the
    NRCS and codepages supported by xterm.
  • add other-sgr.sh script to demonstrate the
    non-VT100 video attributes implemented in patch #305, and added a summary of their
    introduction in ECMA-48 to the control sequences document.
  • add tabs and -tabs keywords to
    ttyModes resource to simplify workaround for
    terminal drivers which are not aware of UTF-8 (FreeBSD #229682).
  • add VT5xx NRCS 7-bit Turkish and DEC Turkish.
  • add VT5xx NRCS 7-bit Hebrew and DEC Hebrew.
  • add VT5xx NRCS 7-bit Greek and DEC Greek.
  • correct a few post-VT220 codepages to make them display
    without setting the NRCS state.
  • extend vt100Graphics resource to include other
    character-sets such as the VT500 codepages in UTF-8 mode.
  • add some VT5xx codepages to help with the discussion of
    96-character sets: ISO Greek Supplemental, ISO
    Hebrew Supplemental, ISO Latin–5 Supplemental,
    and ISO Latin–Cyrillic.
  • improve description of the VT300-controls for designating
    character sets, and fix a typo which indicated that some were
    used for 94-character sets (prompted by discussion with Lars
  • add private control XTCHECKSUM for modifying
  • factor out xterm’s DECRQCRA checksum
    extensions as a resource setting
  • revise calculation of checksum for DECRQCRA to
    match DEC’s terminals (using screenshots of vttest provided by
    Markus Schmidt).
  • restore a fix for memory-checksum setup of
    DECCKSR from patch #315
    incorrectly reverted in patch #334 as
    part of fix for DECRQCRA.
  • add colorInnerBorder resource to make a change
    from patch #334 configurable (reports
    by H Merijn Brand, Gabriele Balducci).
  • modify Imakefile to reflect the fact that
    NetBSD no longer has a working termcap emulation.
  • add resource-setting validShells which can be
    used to augment the system’s /etc/shell (prompted
    by discussion with Paul Lampert).
  • stifle some useless warnings from lintian in
  • add the ncurses extension “RGB” to the responses for the
    termcap-query feature.
  • improved getopts-handling in sample scripts.
  • fix some warnings from gcc8 and clang –analyze.
  • update note about incorrect documentation for
    DECRQSS to include VT525 (report by Markus
  • correct check for default-values in rectangular parsing; a
    zero counts as a missing or default parameter (report/testcase
    by Markus Schmidt).
  • correct some ranges in the ambiguous[] table
    in wcwidth (adapted from patch by KUGA Tsutomu).
  • fix a special case with faint video attribute incorrectly
    combined with default color.
  • add private control XTREPORTSGR for reporting
    video-attributes and color on a rectangle, and script
    report-sgr.pl to demonstrate it.
  • modify some of the markup in ctlseqs.ms to
    work around groff’s reassignment of ASCII punctuation
    characters as documented in groff_char(7).
  • treat ECMA-48 SGR 6 the same as SGR 5.
  • add private controls XTPUSHSGR and
    XTPOPSGR for saving/restoring the current
    video-attributes on a stack (adapted from patch by Dan
  • modify DECRPM response for logging enable to
    indicate its state even when enabling/disabling it is not
  • implement DECSNLS
  • implement DECRSPS
  • improve fill-color for double-buffer configuration, which
    was sharing a graphic context with the cursor.
  • fix a couple of cases where double-sized VT100 characters
    were not clipped, seen in Joe Smith’s VT100 torture
  • color the inner border using the same
    borderColor as the outer border, rather than
    filling with the VT100’s default background.
  • change encoding of “THANKS” to UTF-8.
  • modify handling of DECELR to recognize
    MotionNotify events again, amending a check for
    allowMouseOps which limited that to button-events
    in patch #328 (Olaf Rogalsky).
  • fix some screen-painting problems with left/right margins
    when using insert-line or delete-line (report/testcase by
    Martin Hostettler). also fix similar case with
  • several minor performance improvements using macros, e.g.,
    inline checks for character width.
  • add DECSCPP and DECSLPP to
  • implement DECSCPP.
  • implement DECCIR and DECTABSR
    presentation reports.
  • modify checksum computation for DECRQCRA to
    treat uninitialized cells as blanks rather than nulls.
  • drop custom-entity from HTML-dump, use UTF-8 for  
    (Debian #902381).
  • modify display of non-BMP characters when using bitmap
    fonts to show a null/empty box rather than the replacement
    character which was used in cleanup changes for patch #233 (report by Christian
  • when setting up clipping rectangle for Xft, allow for the
    case where xterm is only displaying a combining character,
    where the base was already written (report by Joshua
  • allow reset Xft’s state if switching font-sizes when
    double-buffering is used.
  • fix repainting, e.g., on resize, when double-buffering is
    used with Xft (patch by Daniel Colascione).
  • correct some interchanged pairs of symbols in
    unicode/keysym.map (Brad Town).
  • improve configure macros CF_GCC_WARNINGS, CF_GNU_SOURCE,
  • modify logFile resource to interpret
    -” as the standard output (adapted patch by Colum
  • improve documentation of Tek4014 menu options; fix a case
    where the Tek4014 window was not displayed before switching
  • re-correct parameter for pid used for DECCKSR,
    DECRQCSR (see xterm #315).
  • hard/soft reset now resets the cursor-blinking state that
    may have been set via escape sequences; the cursor-blinking
    menu-entry feature is unmodified (report by Matthieu
  • update config.guess, config.sub
  • ensure that i18n is enabled if input-method is
  • modify logic for pre-edit to update spot-location while the
    cursor is invisible (Kakoune #1940).
  • change default icon to newer one, “mini.xterm”.
  • install a complete set of icons, to simplify post-install
  • add resource disallowedPasteControls to extend
    filtering of control characters from pastes.
  • add print-on-error action-hook, and document
    action-hooks for dump-html, dump-svg.
  • provide action-hook for print-immediate, which
    was available only as a menu-item (report by Rastislav
  • continue to improve notes in ctlseqs.ms which
    mention where various controls originated.
  • add case to accept ECMA-48’s bogus SD, but
    document the issue in ctlseqs.ms.
  • add control sequence for loading XPM-icon file, using the
    analogous control sequence from shelltool/dtterm.
  • add -report-icons option, to report on
    XPM-icon and title-bar updates.
  • correct a discrepancy between locator-reports for VT220 vs
  • add window-ops control sequences to complement existing
    ones for reporting window-position and text-area size with
    reports for text-area position and window-size.
  • display vt52 graphics mode characters.
  • correct case-statement for window-ops 10, overlooked since
    some window managers equate vertical-maximize and
    horizontal-maximize window hints with full-screen
  • add a GetChecksum item to
  • improve notes in ctlseqs.ms which mention
    where various controls originated.
  • add control sequences for querying the X display size and
    the character size, both in pixels to help eliminate some of
    the constants in esctest.
  • fixes prompted by review of George Nachman’s
    esctest script:

    • add a null-pointer check and a limit-check in
    • correct limit-check for right-margin when processing
      autowrapping if xterm is not built for
    • repair logic in RIS to reset 132-column
      mode when an escape sequence is used to enable
      80/132-column switching.
    • correct array-limit for EWMH working state.
    • further improve limit-checks versus assert’s (see
      patch #315).
    • correct order of check-margins and move-to-left margin
      for DL and IL.
    • modify reverse-wrap to pay attention to top/bottom
      margins as it has done for left/right margins since
      patch #279.
    • modify behavior of DECFI and
      DECBI to not index the screen at the
      left/right edge of the display unless those happen to be
      the left/right margins.
    • add case for VT320 DSR response for keyboard
  • improve check for valid shell path using
    getusershell and endusershell if
    available, in case /etc/shells does not exist
    (suggested by Rainer Orth).
  • modified configure checks for groff and man2html, to use
    the latter by default, but allowing mandoc as a partial
    replacement for former.
  • revisited the inconsistent copyright notices mentioned in
    patch #165, and eliminate the issue as
    mentioned in this copyright
  • fix typo in this change-log (report by Sven Joachim)
  • update config.guess, config.sub
  • add workaround for improper grayscale adjustments made in
    FreeType library, exposed by changes to rounding in 2.8.1,
    which shows up as a gap in line-drawing characters (Debian
  • improve a special case where a non-Unicode font’s
    line-drawing characters were not used, when specifying it via
    the utf8Fonts resource, e.g.,

            -fs 15 \
            -xrm '*VT100.font:12x24' \
            -xrm '*VT100.boldFont:12x24' \
            -xrm '*VT100.utf8Fonts.font: 12x24' \
            -xrm '*VT100.utf8Fonts.boldFont: 12x24' \
  • replace constant 10msec delay for next X event with new
    resource nextEventDelay, and reduce that to 1msec
    to accommodate faster machines than used when
    -hold was implemented in 1999 (Debian #877628).
  • add scroll-to action, which simplifies binding
    a key to scroll to the beginning or end of the saved-lines
    (Debian #880120).
  • add building blocks for alternate screen and/or title-stack
    features in the terminfo file.
  • improve calculations for cell-data size.
  • change configure script to enable XHTML/SVG screen dumps by
  • change configure script to enable 256-colors by
  • update config.guess, config.sub
  • change configure script option for
    --with-man2html to use improved script by
  • add case for private mode 1044 in DECRQM, to report the
    keepClipboard resource setting and corresponding menu
  • fix an inconsistency between private mode 12 (the AT&T
    610 cursor blink) and DECSCUSR: the former relied
    on having the cursorBlink resource set initially
    to enable the escape sequence, while the latter does not.
  • add private modes 13 and 14, as well as resource
    cursorBlinkXOR to allow better control over the
    cursor-blinking state (discussion with Bram Moolenaar).
  • modify the html and svg screen dumps to support direct
  • modify media copy (screen-printing) to support the same SGR
    codes as DECRQSS, including 88/256 indexed color
    as well as direct color.
  • improve options-parsing for query-status.pl script.
  • modify parsing of SGR direct-color control
    making color space identitier optional. The
    corresponding DECRQSS reply always returns an
    empty (default) field for the identifier.
  • add wide-attributes to DECRQSS reply for
  • add private mode 1046 to help with scripting
  • correct expression used for readline-flags in DECRQM; to
    test the current flag rather than information stacked
    within the same variable.
  • correct typo in ctlseqs.ms reference to
    ISO-8613-6 (patch by Mike Frysinger).
  • fix lintian warning for test-package.
  • fix typo in DECRQSS for SGR 48,
    which printed the foreground value for colors past 15. Also use
    colon delimiter for codes 38/48 in response (report by Paul
    LeoNerd Evans).
  • improve workaround for Debian #542434
    by using the font’s maximum width when no ISO-8859-1 glyphs are
    provided (Debian #879936).
  • work around a special case of Xft’s mismanagement
    of its cached data by adding a check before the
    -report-fonts option to ensure that it does not
    use an XftPattern which may have been freed during
    a call to XftFontOpenPattern.
  • improve manual page description of regex
    option for onXClicks resources (report by
    Lukas Mai).
  • add directColor resource.
  • additional manpage macro cleanup (Branden Robinson, Debian
  • add optional support for direct-colors (adapted from patch
    by anonymous “Nibby Nebbulous”).
  • improve legacy/NRC character set mapping (patch by Thomas
    Wolff), e.g.,

    • enable alternate NRC set designators for French and
      French Canadian, ‘9’ and ‘f’ respectively, as documented in
      ctlseqs. (‘9’ is documented for VT510, ‘f’ is a Kermit
    • correct the Unicode value in the DEC Technical
      table to show capital delta.
    • referring to


      add entries for the DEC Supplemental Graphics
      table to display 0x28/0xa8 as ¤ and 0x5d/0xdd as Ÿ.

    • referring to


      alter the Unicode values used for 0x2b through 0x2c to
      use curly braces to work with the “middle” parts
      displayed with 0x2f and 0x30.

    • modify the VT220 “Supplemental” table, giving a hint
      that it was probably meant to be “Supplemental Graphics”
      and is the same as VT320’s table.
  • quiet a few font-warnings when a derived fontname cannot be
    opened, overlooked in refactoring of font resources in patch #328 (initially reported on Cygwin
    mailing list, with followup).
  • correct error response for DECRQSS broken in
    cleanup of Coverity reports in patch
    #288 (reports by Bram Moolenaar and IWAMOTO Kouichi).
  • improve DECRPM responses by returning mode
    not recognized for modes which may not be settable due to
    the selected decTerminalID resource (report by
    IWAMOTO Kouichi).
  • correct logic for print-immediate action, and
    enable corresponding menu entry (patch by Lauri
  • add configure option --with-pcre2 (patch by
    David Michael).
  • fix a misspelled subsection title in ctlseqs.ms and add a
    note regarding blink which was rendered as bold in X11R6.
    Blinking text was implemented in Patch
  • fix typos in xterm.man (patches by Sven
    Joachim, Larry Hynes).
  • fix typography in xterm.man (patch by Bjarni
    Ingi Gislason, Debian #869248).
  • fix typo in INSTALL (Larry Hynes).
  • add xterm-direct terminal description based on
    changes introduced in patch #277, and
    relying upon ncurses RGB
  • modify xterm-new terminal description to use
    ECMA-48 REP, reflecting its use in xterm since
    patch #32 (1996).
  • clarify comment in ctlseqs.ms regarding blink: it has been
    part of xterm since patch #60
  • update ftp URLs in documentation.
  • updates for ReGIS (Ross Combs):
    • remove redundant text command error check which broke
      T(B) and T(E).
    • retain the loading alphabet number across multiple “L”
    • add S(T) delay handler.
    • fix some color handling error messages.
    • add stubbed-out macrograph handling.
    • use fragment_remaining() and
      fragment_consumed() instead of manually
      checking position / length in various places.
    • rename some local variables in string / extent / option
    • wrap some long lines.
    • move macrograph command handling out of the
  • add a summary of the italic fonts loaded to
    -report-fonts option.
  • modify the font-lookup for italics to allow for
    -i-” if no match is found with slant
    -o-” (prompted by patch by Ben Wong).
  • change default values for mkSamplePass and
    mkSampleSize to reflect generally-improved locale
    support in various operating systems (FreeBSD #219800).
  • modify wcwidth.c to return -1 for non-Unicode
    values, and adjust a couple of blocks to better match
    assumptions about ambiguous-width characters in other
    implementations. Also modify wcwidth.c to support
    configurable soft-hyphen, so there is no drawback to using this
    version rather than a system wcwidth.
  • amend change made in patch #328
    for cursor-visibility to handle case where an application is
    updating the reverse-video state (FreeBSD #219800).
  • update tables of combining and ambiguous-width characters
    in wcwidth.c based on Unicode 10.0.0.
  • build-fix for --enable-sixel-graphics without
    --enable-regis-graphics (reports by Sven Joachim,
    FreeBSD #219945).
  • add control sequences for reading the Sixel and ReGIS
    graphics sizes (suggested by Ben Wong).
  • add a workaround for wcwidth returning -1 for
    characters which should have been printable (FreeBSD #219800).
  • fix a bug in font initialization from patch #328 (FreeBSD #219800).
  • fix a special case in HideCursor which
    assigned a bold font to the slot used for normal font in
    changes for italics in patch #307
    (Debian #858304).
  • updates for ReGIS (Ross Combs):
    • Strings specified with no command are used as
      “comments”. Print these in the log when tracing.
    • Catch attempts to use “alternate display” mode (AKA
      “blink”) from the GIGI, but do not implement it.
    • The T(M) command should only multiply the
      height by 10, not 20.
    • Make the S(E) command reset more state
      than just the screen contents.
    • Remove two rotation variables which were only being
    • Numerous minor fixes and comment updates in the
      R command.
    • Unknown R command option names trigger an
      empty response.
    • Fix the output position after printing rotated text (it
      was missing the the sign before).
    • Fix the position change with pixelvectors and rotated
      text (the rotation transform was not being applied).
    • Update the TODO list and remove a verification
      FIXME (slanted text positioning is correct as
    • Emulate the approximately 1.4x enlargement for text
      which isn’t rotated at right angles.
    • Only update the color planes specified in the plane
      mask (the W command’s F
  • fix a bug introduced by the changes to font information in
    patch #328. When processing the
    “checkfont” option of the locale resource, the
    program referred to the request data, to an array
    which was only allocated in the new/result widget
    (report by H Merijn Brand).
  • fix a missing assignment initialization to make the
    utf8 resource control whether escape sequences to
    enable/disable UTF-8 mode are allowed.
  • revise parser for charClass resource, making
    these improvements:

    • accept octal and hexadecimal values
    • allow embedded whitespace
    • allow the class after colon to be optional, e.g., to
      clear class settings for a range of characters.
  • add command-line option
  • fix most lintian warnings about test-package
  • add eraseSavedLines resource.
  • document DECSED 3 in
    ctlseqs.ms (report by Ben Longmans).
  • improve integration between configure-events and updates
    for reported screensize, in particular when switching between
    vt100 and tek4014 modes.
  • modify selection-highlighting of reverse-video text to keep
    that distinct, e.g., by reversing the selection foreground and
    background colors as one would expect. This fixes a “useless”
    case in the description of
  • improve fix for Debian #759734,
    addressing a case where non-colored cursor would be invisible
    against reverse-video (see patch
  • updates for ReGIS (Ross Combs):
    • the “H” option of the “T” command should multiply by
      10, not 20.
    • display unknown glyphs as a solid block.
    • given a succession of text-direction options, use the
    • fix the direction of ReGIS slanted text so that
      negative values produce oblique output .
    • fix the ReGIS text direction option to only rotate
      characters when no following size option is used.
  • update terminfo to better match corresponding entries in
    ncurses, e.g., u8 pattern to match the VT220,
    VT420, etc., primary responses, as well as adding
    smxx and rmxx.
  • fixes from Jörg Sommer:
    • corrected a trace-message regarding maximum
      graphics-size; it used the similar ReGIS maximum size which
      might not be configured.
    • in do_select_regex, clear selection if
      there is no match. If the regex does not match anything
      around the cursor, the selection returned must be empty,
      otherwise the whole line is treated as a match. This way
      the command defined by exec-selectable will
      not be executed if there is no match.
    • modify limit in do_select_regex to include
      the character at the cursor in the match, making it easier
      to type something and then hit the key to trigger
    • If exec-selectable or
      insert-selection is triggered by a mouse
      button click, the position of the mouse pointer should be
      used. This makes it easier to address any position on the
      window and it makes it possible to use the mouse, e.g., for
      applications such as mutt where you cannot move the
  • modify DECRC to save/restore xterm’s
    last-column flag used to control wrapping behavior rather than
    manipulating DECAWM (report/analysis by Mattias
  • add configure option --enable-terminfo-env to
    use the value set by --with-own-terminfo for the
    $TERMINFO environment variable. That variable was
    set automatically for HPUX, but would be useful in other
    systems, e.g., for Solaris (request by Jeff Wieland).
  • fix a race condition when setting up a signal handler to
    timeout if opening /dev/tty hangs (patch by Tobias
  • review/cleanup resources which were not in the manual page
    (report by Maxwell Anselm):

    • add manual page description as needed.
    • drop resource name for menuBar, as
    • modify fallback numeric value for
      regisScreenSize resource to match that for
  • updated configure macros CF_ADD_CFLAGS,
    other program-changes.
  • update config.guess, config.sub
  • change “maximum screensize” assumed by resize
    to 9999×9999, to accommodate people using the
    Unreadable font.
  • drop Utility from default value of
    --with-desktop-category (Debian #780176).
  • widen the configure script pattern used for finding related
    “.desktop” files, including “Terminal”
  • several minor improvements to font utility functions:
    • provide for later modification to implement font-sets
      by parsing the font resources as comma-separated
    • parse -fn and -fa similarly,
      using “x:” and “xft:” prefixes for font name/family strings
      to distinguish between XLFD and Xft font
    • use loops to iterate over font classes
    • use getters/setters for font data to allow for
      on-demand lookups.
    • make the debugging trace for missing glyph less verbose
      since that interferes with the -report-fonts
    • refactor xtermLoadFont to make it clearer
      how some fonts are derived from others, e.g., bold,
    • make the triggering and suppressing of font-warnings
      more consistent by storing the last state in the
    • reduce font-warnings by checking for repeated
  • add vttests/query-status.pl
  • add vttests/closest-rgb
  • add special case for displaying soft-hyphen if it happens
    to fall at the right margin, and omitting similar case such as
    the BIDI markers, where a zero-width character is neither a
    control character nor a combining character (Debian #844325).
  • modify logic for OSC 52, manipulate
    selection data, to update the selection-time to include
    the latest X events. This fixes some cases where the selection
    was invalid, e.g., after an event due to focus-follows-mouse
    (report/testcase by Stephane Chauveau).
  • revise macro CastMallocN as new macro
    TextAlloc to make explicit use of
    sizeof(char) (prompted by patch by Cade
  • add “Mouse Ops” menu entry and related resources
    to allow runtime disabling/enabling of the mouse protocol
    escape sequences (discussion with Bob Proulx).
  • improve discussion of mouse actions versus protocol in the
    manual (discussion with Bob Proulx).
  • improve discussion of environment variables in the manual,
    pointing out where some features (such as termcap and the
    System5 COLUMNS and LINES variables)
    are used rarely, mainly to support legacy applications.
  • add examples of translations resource for
    select/paste, and for font-size changes to the manual.
  • minor reordering of some entries in ctlseqs.ms
    for consistency (report by Arran Ubels).
  • add -s option to 256colors2.pl
    and 88colors2.pl, to demonstrate modifying the
    “system” colors 0–15.
  • omit XFT_SPACING property from call to
    , to work around a bug in
    fontconfig for handling Google Go fonts, whose names
    sort in an order not expected by fontconfig, causing
    the request for a monospaced font to return italics,

    $ fc-match 'Go Mono:spacing=monospace'
    Go-Mono-Italic.ttf: "Go Mono" "Italic"
    (report by Giacomo Boffi on Stackoverflow).
  • modify minstall.in to improve a workaround
    added to the manual page in patch #182
    to avoid having the C preprocessor used in the imake
    configuration strip out the comments in the character classes
    section (reports by Ted Unangst, Anthony J Bentley).
  • add a check in the function which handles end-of-line
    wrapping to ensure that C1 controls are allocated one column
    when allowC1Printable is set (Debian #738794).
  • use consistent error-checking after strtol
    calls, fixes a case where a query with OSC 6 did not ensure
    there was a valid color number (report by Alex Smith).
  • add -baudrate option, for testing
  • always generate the CASE_xxx symbols
    in VTparse.h and Tekparse.h, as part
    of a change to improve debug-logging. This makes the build
    always depend upon awk.
  • modify allowC1Printable to disallow codes
    160-254 as being equivalent to codes 32-126 when parsing escape
    sequences (Debian #839220).
  • amend fix from patch #326 for
    TrueType fonts to exclude the hidden character used for
    double-width cells (report by Grady Martin).
  • fix a typo in ctlseqs.ms
  • updated appdata file (report by Richard Hughes).
  • improve discussion of the different terminal emulations
    provided by xterm in the manual page.
  • add examples of setting the icon title with/without the
    window title in the manual (Debian #833984).
  • correct a limit-check when using a numeric value for
    extended Booleans e.g., *fullscreen:3 rather than
    a name such as *fullscreen:never.
  • add action allow-bold-fonts
  • improved formatting fixes for manual page, using script to
    find mismatches in spelling of resources, actions and menu
  • improve documentation of logging resources.
  • fix a special case of flickering cursor by adding
    GraphicsExpose to the list of event types that
    should not trigger making the mouse cursor visible (patch by
    Joe Peterson).
  • correct initialization of line-drawing in VT52-mode,
    overlooked in changes for patch #297
    (report/patch by Ben Wiley Sittler).
  • minor clarification of form-feed versus line-feed in
    ctlseqs.ms (suggested by David Kemper).
  • amend fix for Debian #738794
    to restore a check for missing characters which are
    not combining characters. Also fill in a corresponding
    special case for TrueType fonts (Debian #827905).
  • improve manual page discussion of function keys (discussion
    with Ross Combs).
  • further improve fix for Debian #545220
    in patch #248, to avoid conflict with
    combining characters (Debian #738794).
  • improve -hold option to avoid 100% CPU usage
    with NetBSD after closing the shell, which makes subsequent
    checks for X input events fail (prompted by patch by Pierre
  • review #ifdef statements, listed those which
    do not have a configure option in
  • fixed one case where the menu sensistivity for
    Print-All Immediately and Print-All on Error
    was not properly ifdef’d.
  • modify terminfo entries for 16-, 88- and 256-color to reset
    palette with rs1 capability.
  • accept legacy value of -kt as
    synonym for the oldXtermFKeys resource, and extend
    the full-reset logic to use the keyboard type set via
  • modify ioctl calls for I_PUSH to
    first check if the module has been added, using
    I_FIND. This is needed for newer Solaris libraries
    with c11 support (adapted from patch by Alan Coopersmith).
  • add check in getXtermCombining to ensure that
    combining characters were allocated (report by Tor
  • add configure option --without-xinerama to
    allow suppressing the extension (Gentoo #580936).
  • update keysym2ucs.c based on Unicode 9.0
  • fixed most cppcheck --enable=all warnings,
    including for style (prompted by report by David Binderman).
    There was one bug-fix:

    • RGB least-squares computation in
      allocateClosestRGB used only one ordinate
  • add cppcheck to lint-like programs in
    configure script and makefile.
  • fix some const correctness issues and point out an array
    lifetime issue (Ross Combs).
  • updates for SIXEL and ReGIS (Ross Combs):
    • Sixel drawing should still happen after an error if
      some commands have been processed
    • Tiny steps toward reporting ReGIS input
  • updated configure macro CF_LD_RPATH_OPT from
    ncurses changes.
  • provide alternate fix for Juha Nurmela’s report by turning
    on POSIX signals if _POSIX_C_SOURCE is at least
    one. This is done to improve the behavior if a direct child
    process receives a STOP signal.
  • revert change to CF_POSIX_C_SOURCE from
    patch #323 (reports by Ashish Shukla,
    Christian Weisgerber).
  • updated appdata and desktop files to add keywords (report
    by Richard Hughes).
  • modify configure macro CF_POSIX_C_SOURCE to
    accommodate some systems, e.g., FreeBSD, which have separated
    their implementation specific preprocessor symbols from the
    POSIX ones, requiring both to be defined (report by Juha
  • change the directory to the current working dir of the
    child process in exec-formatted and exec-selectable actions
    (patch by Alexander Pohoyda).
  • correct typo in xterm.man (patch by Larry Hynes).
  • correct typo in ctlseqs.ms (report by Shriramana
  • add feature to support XHTML and SVG screen dumps (patch by
    Jens Schweikhardt).
  • correct response to DECRQSS when terminal id is less than
    400 (patch by Iwamoto Kouichi).
  • fix regression due to incorrect fix for compiler warning
    when allocating storage for /etc/shells (reports
    by Ashish Shukla, Debian #809646).
  • add resource keepClipboard, escape sequence
    and action keep-clipboard.
  • add optional feature to capture text copied to clipboard at
    the time of copying rather than at the time the clipboard
    contents are requested for pasting (patch by Milan
  • improve a special case where the -e option was
    used to pass a single-quoted command via luit, by
    wrapping it in a “sh -c” (report by Keith Hedger).
  • minor fix for type-cleanliness when allocating storage for
    /etc/shells (Tobias Stoeckmann).
  • fix a typo in manual page (Dan Church).
  • fix minor file-descriptor leak; after calling
    openpty, the slave’s file descriptor is not needed
    (report by Juha Nurmela).
  • editorial change to ctlseqs.ms (report by David
  • minor updates for autoconf macros.
  • update config.guess, config.sub
  • correct documentation for %t format (report by
    Martin Tournoij).
  • add %R for consistency to
    exec-formatted() and
    insert-formatted() actions.
  • add %r format to exec-formatted()
    and insert-formatted() actions (patch by Martin
  • Amend fix for Debian #794201
    (report forwarded from Christian Jachmann by Emanuel Haupt,
    Debian #797008).
  • add a section to ctlseqs.ms discussing control
    sequences and ECMA-48, to explain why C1 controls do not occur
    in the decoded characters from UTF-8 byte streams (prompted by
    discussion with Poul-Henning Kamp).
  • modify check for cursor-theme from patch #301 to also check if the resource
    Xcursor.theme is set to a nonempty value before
    fallback to xterm’s own dummy theme (request by Robert
  • explain in ctlseqs.ms that some keys which
    normally send SS3-prefixes are changed to CSI-prefixes if
    key-modifiers are passed as parameters (report by George
  • correct double-free of font information when multiple
    problems are found, particularly for the wide fonts loaded via
    the -wc option (report/testcase by Nelson
  • make configure option --enable-builtin-xpms
    actually work (report by William Bulley).
  • correct combination of -ls and -e
    options for utempter configuration (Debian #794201).
  • NetBSD build-fix for OPT_RENDERFONT versus OPT_SHIFT_FONTS
    (patch by Matthew Green, forwarded by Thomas Klausner).
  • fix a few minor bugs found with Coverity.
  • update pixelvector handling (Ross Combs):
    • split pixel-based and coord-based functions with common
      parts factored out to a “raw” function
    • add a “step” variant which loads a single PV digit
  • make the graphic dirty upon resize or clear, fixing some
    missing refreshes (patch by Ross Combs)
  • fixes scrolling to use user coordinates and move in the
    correct direction (patch by Ross Combs)
  • several fixes/improvements for ReGIS whitespace and
    page-handling (patch by Ross Combs).
  • rename new(er) configure option
    --enable-initial-erase to
    --enable-pty-erase, since that conflicted with an
    existing option (report by Jens Schweikhardt).
  • fix cut/paste error in --with-builtin-xpms
    configure option (reports by Jens Schweikhardt, Ross
  • fix minor formatting issue in xterm manual (report by Jens
  • eliminate unnecessary “#(vi” markers in autoconf macros by
    using “(” to begin case-statement cases (suggested by Jens
  • minor fixes to improve const usage (Ross
  • adapt example for exec-formatted from
    Lukáš Zapletal’s webpage to manual (Debian #780008).
  • add a short usage section to the xterm manual, including
    notes on setting the window title (Debian #742477).
  • revise a change made to ICH in patch #314 to address limit-checks
    (reports/testcases by Zoltán Kéri and Joe Peterson, also
    reported by Christian Weisgerber).
  • revert change to make DCH honor top/bottom
    scrolling margins. For the record, both the VT420 and VT520
    reference manuals incorrectly state that “DCH has no effect
    outside the scrolling margins.”
  • fix regression in DCH introduced in adjustment
    for limit-check in patch #315
    (report/testcase by Zoltán Kéri, also reported by Christian
  • correct default value for configure option
    --enable-initial-erase; it should be “False”, but
    in patch #315 it was “Maybe”, which
    differed from default set in patch
  • change default for --enable-narrowproto for
    Cygwin to “no” to better match contemporary configurations
    (report by Jens Schweikhardt, prompted by
  • provide OSC 106 as an alternative to
    OSC 6, avoiding conflict with one of
    Terminal.app’s undocumented escape sequences (report by Egmont
  • fix an off-by-one in comparison when autowrap and
    left/right margins are combined (report by George
  • modify DECBI and DECFI to work
    outside the left/right margins, since the vague VT520 reference
    manual could support that interpretation (suggested by George
  • correct response for DECSCUSR in
    DECRQSS (report/analysis by George Nachman).
  • add DECSLRM to list for DECRQSS
    in ctlseqs.ms (report by George Nachman).
  • improve limit-checks versus assert’s in the debugging
    version (reports by George Nachman).
  • correct a problem with multiple writes to the right-margin
    when autowrap is turned off (report by George Nachman, also
    reported in Gnome #744819
    by Ulf Magnusson).
  • modify DCH to honor top/bottom and left/right
    margins (report by George Nachman).
  • correct off-by-one in comparison so that DL
    honors left/right margins (report by George Nachman).
  • correct logic of DECSCL, which always set
    8-bit controls as a side-effect (report by George
  • correct loop limit for DECSED 1, when
    DECSCA is enabled (report by George Nachman).
  • fix a entry in csi_table for “$”, needed for
    CSI [ $ z (report by
    George Nachman).
  • remove restriction from patch #279
    changes on DECIC and DECDC to
    left-right mode (report by George Nachman).
  • correct PID value returned in response to DECRQCRA
    (report/analysis by George Nachman).
  • fix an ifdef’ing problem, where
    --disable-dec-locator would turn off logic needed
    for DECIC and DECDC (report by George
  • correct order of parameters in report for
    CSI 1 9 t, which gives the
    screensize in characters (report by George Nachman).
  • add null pointer checks for a case in DECCRA where part of
    the target rectangle lies outside the screen limits (report by
    George Nachman).
  • remove extra “6” for selective erase from DA1 response
    (report by George Nachman).
  • support Xinerama screen specification in geometry parsing
    (patch by Nicolas George).
  • add configure option --enable-initial-erase to
    set ptyInitialErase consistently with manpage, as
    well as adding a table in manpage to clarify the relationship
    between the resources related to the backarrow key (Debian
  • add configure --enable-builtin-xpms to
    simplify compiling-in the icons introduced in patch #284 (prompted by discussion with
    William Bulley).
  • fix/improve autoconf macros:
    • CF_WITH_APP_DEFAULTS, add paths for OSX
    • CF_WITH_MAN2HTML, configure option “–with-man2html”
      provides “man2html” as alternative to groff’s man/html
  • fix some minor issues in manpage (Jens Schweikhardt).
  • fix a minor bug in the termcap-specific version of
    resize when the lines or columns values are at the
    very end of an incomplete termcap string (prompted by coverity
  • improve paste64 feature (report by Olaf Rogalsky)
    • fix “p;” command in
      paste64.pl example.
    • modify _OwnSelection to allow the paste64
      control to cause primary/clipboard selections to be
      disowned and cleared.
    • omit an extra “=” used in padding, to make the base64
      response a multiple of four characters.
    • modify _ConvertSelectionHelper to not stop
      converting on embedded nulls in the primary or clipboard
      selection. That makes the behavior consistent with
      cut-buffers. Nulls are not passed on to the application
      because they are used for separators in the X library calls
      that convert XTextProperty values.
  • add configure option --with-man2html to allow
    an alternative to groff’s man/html conversion.
  • review and modify resource-settings which can be set via
    control sequences, etc:

    • add OSC 6, to enable/disable colors
      set via OSC 5, without changing the color
      assignments (prompted by patch by Ingo Brückl).
    • add command-line option -itc for
    • add colorIT and colorITMode
      for italics, like colorBD, etc.
    • guard allowPasteControls against
      modification by editres.
  • change passedPty from fixed-length to
    allocated to ensure that it is long enough to hold the
    -S option value (report by Ben Longbons).
  • improve ReGIS graphics initialization (Ross Combs):
    • add a new string resource to set the default ReGIS
    • fix a caching problem where the font name wasn’t taken
      into consideration when looking up cached metrics.
    • enables use of fractional values in coordinates (which
      in turn allows setting the coordinates to something like
    • doubles the minimum ReGIS graphic to 200×200
    • set the default ReGIS image size to “auto”
    • fix some swapped width/height parameters in TRACE
    • add support for scaling and mirrored coordinate systems
      to the screen address command
    • sets a minimum ReGIS graphics size of 100×100 —
      smaller address ranges will be scaled up
    • reset the image if either the width or height is zero
      (which should happen together or not at all, but switching
      the logic gets rid of some parentheses)
    • adds support for alternate screen coordinates (just
      offsets, not inverted coordinates or scaling)
    • keeps the graphics state across ReGIS calls unless P1
      or P3 entry codes are used
    • changes loops to continue after parsing whitespace to
      allow trailing whitespace
    • fixes some incorrect character casts
  • add regisScreenSize resource setting to allow
    custom screensizes for ReGIS graphics (prompted by discussion
    with Scott Froebe).
  • fix some minor issues in manpage (Jens Schweikhardt).
  • improve ReGIS font-handling (Ross Combs):
    • remove “random junk” generation for unknown
    • add xterm extension to load user glyphs for alphabet
    • add xterm extension to specify loading an alphabet from
      a font by name
    • fix bug with use of empty alphabet slots which caused
      slot 0 to be clobbered
    • update/clarify ReGIS-related comments
    • add DECprint extension for printing/clearing the screen
      (only clearing is performed)
  • improve ReGIS colorspec conversion (Ross Combs):
    • avoid some floating point math for colorspace
    • move two copies of colorspec parsing to a single
    • change colorspec parsing to handle components being out
      of order or with spaces and commas between them
    • add support for an extension where RGB components can
      be given instead of HLS
  • improve use of const, e.g., for the LineData
    pointers (patch by Ross Combs).
  • clip of graphics that overlap the scrollback buffer and alt
    screen (patch by Ross Combs).
  • amend change to focus-change events in patch #287 to limit it to suppressing the
    reset of the urgency-hint, because some useful
    FocusOut events were lost (report by Joe
  • modify dpkg build-script to configure xterm-dev as an
    x-terminal-emulator alternative.
  • extend the --with-xterm-symlink feature to
    make symbolic links for the other programs and scripts.
  • minor fix to vttests/resize.pl to avoid
    undefined variable when receiving broken/missing response,
    e.g., from mrxvt.
  • change the way that ReGIS graphics are refreshed (patch by
    Ross Combs):

    • it pre-composes any overlapping graphics to avoid
      unnecessary and flickering draws
    • it adds a three-dimensional lookup table for RGB ->
      pixel values and uses this across all graphics
    • it draws in line segments when there is a run of the
      same color
    • in cases where there are no transparent parts, it uses
      an XImage to draw
    • it hooks into the double-buffering support if it is
      enabled (though this doesn’t seem to get rid of all

    This has the effect of reducing flicker and speeding up

  • undo change to do_select_regex() in patch #311 (reports by H Merijn Brand,
    Sven-Haegar Koch, Debian #762978).
  • mention xclip in manpage as an alternative
    workaround for copying clipboard data, noting that
    selectToClipboard (see patch
    #209) is the recommended approach (Debian #639094).
  • correct comparison in do_select_regex() of working position
    against starting column.
  • correct initialization for regular-expression feature of
    exec-selectable and insert-selectable
    actions (Debian #758633).
  • modify logic for exec-formatted and
    insert-formatted actions to ensure that the
    formatting occurs just after the selection is received (Debian
  • account for state of reverse-video in special case of
    cursor coloring (Debian #759734).
  • fix a sign-extension problem in ReGIS support, as well as
    correcting a case where ignored-characters were not really
    ignored (patch by Ross Combs).
  • Enable the “Escape Sequence” menu entry when an OSC 50
    is received. Also disable it if the escape sequence specifies
    no font (Debian #760208).
  • improve fix for the fontsel menu entry from
    patch #304: because the recovery used
    the “current font”, it would fail if one first selected a valid
    font, then an invalid font (Debian #760207).
  • correct an off-by-one in limit-check for
    ScrnLimitChar function (report by Egmont
  • update package/freebsd files based on xterm #309 port.
  • adapt changes from Minux3.2 which ported xterm #197, to allow building on that platform
    both using imake as well as using the configure
  • update precompose.c based on Unicode 7.0
  • build-fix for --with-Xaw3dxft option (report
    by Emanuel Haupt).
  • modify configure script to work around debris left by
    XQuartz upgrades.
  • modify test-package scripts to enable ReGIS graphics.
  • improvements/additions to ReGIS graphics, e.g., drawing
    text (Ross Combs).
  • add --with-Xaw3dxft, to link with Xaw 3d xft
    library (Stephen P Wall).
  • amend reset for cursor-shape to take into account
    cursorUnderline resource (report by Christian
  • update config.guess, config.sub
  • corrected font used for clipping double-width bitmap fonts
    from patch #307 changes to work around
    mis-scaled fonts (Debian #752947).
  • improve wording of a warning message.
  • fill background for top/bottom parts of
    double-width/double-height characters, to cover occasional gaps
    due to bitmap font-scaling (report by Egmont Koblinger).
  • amend resets for keyboard-type, pointer-mode and
    title-modes from xterm #305 changes to
    account for resource settings (Debian #751351).
  • modify printAttributes feature to include new SGR
  • modify cursor show/hide logic to work with italic
  • do clipping and filling for a case with bitmap-fonts and
    italics when the font server returns only a “close” match
    (report/testcase by Egmont Koblinger).
  • adjust loop comparison for underlining to display
    underlines on 5×8 font as in patch
    #304 (Debian #750733).
  • amend fix for Freedesktop #15979
    (Debian #750733).
  • amend xterm.appdata.xml, making it validate with current
  • fixes for display-errors in new SGR codes (report by Egmont
  • add xterm.appdata.xml (request by Richard
    Hughes). By the way, none of the released validators
    yet handle <licence> tags (or
    <metadata_license>, etc).
  • fix memory leaks in bitmap font-name caching.
  • other changes to hard-reset behavior:
    • reset keyboard-type on hard-reset.
    • reset pointer-mode (pointerMode) on
    • reset title-modes (titleModes) on
    • reset cursor-shape on hard or soft-reset
    • reset LEDs on hard-reset.
    • reset DECSACE on hard-reset.
  • reset bracketed-paste mode on hard-reset (suggested by
    Egmont Koblinger).
  • modify UTF-8 decoder to account for
    allowC1Printable resource.
  • revise getXftColor to ensure that its sequence
    number will not overflow in long-running sessions.
  • minor fix to selection to prevent hidden character from
    being returned in selection data (report by Egmont
  • add -sh to help-message.
  • amend check for incomplete fonts to always decide that a
    character is missing if the font lacks per-character metrics
    (prompted by Freedesktop #15979).
  • minor reorganization to implement “filler” SGR features.
    There are no established applications which rely upon these;
    some people find them amusing.

    • separate bits used to manage drawing state from
    • implement SGR codes 2, 3, 9, 21 and their corresponding
    • add configure option --disable-wide-attrs
      to disable the feature.
  • additional changes for ReGIS support (Ross Combs):
    • fix some arc drawing bugs and add support for pattern
  • add debugging check for parsing table entries, fixed a few
  • handle utmp/utmpx strings consistently with new function
    that null-pads unused locations.
  • build-fix for imake configuration with FreeBSD 9 and later,
    to account for replacing utmp with
  • improve handling of the fontsel
    (font-selection) menu item:

    • initialize the menu entry differently if the toolbar
      configuration is used, i.e., enabling it without attempting
      to use the current selection as a font name. Because of the
      way in which the toolbar is setup, this attempt usually
      failed, and would result in a warning.
    • if the menu action fails, e.g., no font can be loaded
      for the given selected name, then recover by resuming with
      the current font. Before, the recovery went to the default
      font. That was done to work around missing bitmap font in
      patch #276.
  • document OSC 119 in the control sequences listing
    (report by Egmont Koblinger).
  • modify special handling of XF86Paste or
    SunPaste keysyms (see patch
    #227) to use the standard shift/lock modifiers in case
    those are, for example, used in level 3 in an XKB file (patch
    by Matthieu Herrb).
  • configure macro updates:
    • drop the -no-gcc option from
    • modify CF_XOPEN_SOURCE to suppress the check for
      defining _XOPEN_SOURCE on Solaris
  • reduce bss storage used for ReGIS/Sixel modules, as well as
    cleanup storage usage in other modules.
  • initial changes for ReGIS support (Ross Combs):
    • document control-sequences
    • separate ifdef’s for sixel and color-register
    • add configuration for VT382
  • improve check for /etc/shells to allow its
    entries to be symbolic links.
  • fix a regression in patch #302
    fixes relating to interaction of SHELL and
    XTERM_SHELL variables (Debian #740919).
  • improve manpage discussion of use of
    /etc/shells for validating the SHELL
    environment variable.
  • further improve the approximation for color map by ignoring
    bits which would not be applied by the rgb masks for the
    current visual (report by Joe Peterson).
  • add -report-colors to usage-message (report by
    Joe Peterson).
  • amend approximation for color map introduced in patch #277 to make that apply only to
    TrueColor (adapted from patch by Joe
  • add -report-colors option, to show colors as
    they are allocated.
  • fix some gcc -Wcast-qual warnings.
  • correct app-defaults location for CentOS-4 in
  • improve check in minstall.sh for X manpage’s
    section number; add sample dependencies for the relevant
    documentation package to packaging test-scripts (prompted by
    Novell #779474).
  • modify sample build-scripts for dpkg and rpm to ensure that
    vttests scripts are uncompressed and
  • modify Perl scripts in vttests to use
    env to find Perl rather than relying on
    /usr/bin/perl This makes the sample scripts usable
    with NetBSD and other systems using pkgsrc.
  • amend check from patch #301 for
    valid shell to permit existing SHELL value to
    specify the shell; clarify the behavior in the manpage (report
    by Emmanuel Thomé).
  • amend change to wide-character processing from patch #257 to allow switching to/from UTF-8
    encoding if the utf8 resource was not set to
    “always” (report by Honza Maly).
  • fix incorrect free from patch #301
    changes, if explicit program without -e option is
    a valid shell (patch by Romain Francoise).
  • OSC 104 and 105 did not accept parameter to reset a
    specific color (patch by Egmont Koblinger).
  • add configure check for initgroups, and use
    that function to complement its use of setgid
    (prompted by patch by Miroslav Lichvar).
  • improve manpage description of scrollbar translations
    versus the vt100 translations (Debian #723573).
  • add a few paragraphs to manpage explaining the implications
    of xterm’s reliance on X Toolkit for command-line parsing.
  • modify cursor-theme logic to check if the environment
    variable XCURSOR_THEME is set to a nonempty value
    before constructing a dummy them (prompted by similar check
    added in OpenBSD CVS).
  • add check for a case where TrueType bold font is missing a
    glyph where the non-bold font has it. Temporarily switch to the
    non-bold font to draw the glyph (report by David
  • minor documentation fixes for ctlseqs.ms
  • only set SHELL environment variable to
    programs found in /etc/shells (prompted by
    patch/report by Al Poole).
  • build-fix, when --disable-wide-chars
    --disable-boxchar is used (prompted by patch by
    Andrey Panov).
  • amend workaround for Xcursor library from
    patch #298; the suggested solution for
    Debian #466704
    fell afoul of a bug in Xcursor which treated an
    empty value for Inherits= as a valid theme—and
    resulted in a stack overflow due to unbounded recursion.
    Certain environments such as XFCE4 exposed the bug. A related
    report in Debian #531679
    was closed without following the recommendation to limit stack
    recursion in this library (reported by Sergey V Dyatko, Debian
  • fix regression in line-drawing from patch #297 changes for NRCS exposed by change
    to assumeAllChars resource in patch #298 (report by Markus Waldeck).
  • quiet a warning about ambiguity when -h
    command-line option is parsed, since -help is used
    rather than the options mentioned in the message. This warning
    was caused by fixes in patch
  • remove an extra check for keypad-mode which interfered with
    the patch #280
    modifyKeyboard feature for the numeric keypad
    (report by David Conners).
  • change default for assumeAllChars resource
    (Debian #725682).
    There are unresolved issues with Terminus fonts, e.g., Debian
  • add -report-fonts command-line option and
    reportFonts resource to optionally report the
    fonts which are loaded, and their metrics such as the number of
    missing glyphs.
  • work around Xcursor library to make
    pointerColor resource work as documented (Debian
    analysis by Vincent Lefèvre).
  • amend change from patch #297 for
    ASCII-equivalents to ensure that missing characters which
    happen to be double-width are padded to the expected number of
    columns. Also, correct the choice between normal/wide Xft fonts
    used when checking for missing characters (reported by WU Yue,
    Debian #728949,
    Gentoo #491334).
  • modify sixel-graphics drawing to account for scrollbar
    width (report by Mario Edelmann).
  • fix remaining issue with DECNRCM; the British
    character set was confused with ISO Latin-1 (report by Hayaki
  • build-fix for --disable-boxchars configure
  • updated configure macros, from ongoing work on lynx and
  • update config.guess, config.sub
  • modify check for missing cells in bitmap font to work
    around terminus 10646 encoding, which is mostly missing, add
    assumeAllChars resource to provide the older
  • modify macros used to check for missing cells in bitmap
    fonts to pick up a long-overlooked improvement made to xfd in
  • improve workaround for groff versus ASCII-equivalents;
    patch #185 had overlooked Xft
    configuration (reported by anonymous user on Arch Linux
  • review/extend DECNRCM support (prompted by
    report by Hayaki Saito asking about the two “A” codes).

    • make pasting of DECNRCM data work by translating the
      pasted data into the selected encoding.
    • limit mode changes to VT220 and up.
    • ignore SCS for National Replacement Characters if
      DECNRCM is not set.
    • add tables and logic to decode SCS controls ending with
      “%” and one additional character. This expects the parsing
      of SCS to VT320.
    • also added parsing for SCS of DEC Supplemental and DEC
      Technical, which are for VT2xx and VT3xx respectively.
    • implement VT220 Multinational character set “GR” aka
      “DEC Supplemental Graphic”.
    • DEC Technical character set based on vt100.net
    • implement Portuguese NRC.
    • provide for temporarily switching to/from NRC mode from
      UTF-8 mode when DECNRCM is set or reset.
  • minor fix to assert’s found while testing examples from
  • modify DECLRMM to not update the cursor
    position (report by Iwamoto Kouichi forwarded by Hayaki
  • modify cursor-position report to take origin-mode into
    account (report by Hayaki Saito, see also iTerm2 pull
    request 129).
  • correct off-by-one comparison in when handling
    carriage-return at left margin (patch by Iwamoto Kouichi,
    forwarded by Hayaki Saito).
  • improve handling of faceName resource when a
    “size=” property is embedded in it, by using that size to
    replace the default faceSize resource. This lets
    xterm honor the other faceSize resources (report by Jens
  • improve OSC 52 selection-data by setting its timestamp
    to correspond to the most-recent event received by xterm, to
    ensure that calls to XtOwnSelect succeed even if
    the selection is manipulated solely through the keyboard
    (prompted by patch by Richard Tollerton).
  • change configure-script defaults for “luit” and
    “wide-chars” options to match the values used in the imake
    configuration, to lessen user surprise if they are accustomed
    to building xterm using imake(report by Andries E
  • improve the imake-related configuration’s check if xterm is
    started in a locale that uses UTF-8 encoding. Before, this
    checked the locale settings only for “UTF-8”, now it checks
    ignoring case for “UTF-8” and “UTF-8” (report by Andries E
  • improve experimental sixel feature (Ross Combs).
  • modify delimiter in configure macro used to add a library
    after a given library to avoid confusion with -Wl
    pass-through options to linker (NetBSD
  • fix quoting in expression used to trim extra libraries from
    patch #295 changes (report by Thomas
  • adjust order of libraries to work around incomplete symbol
    resolution when using --as-needed option
    (ArchLinux #36047).
  • minor fix for plink.sh, for IRIX64.
  • add note in ctlseqs.ms for CPR response pointing out a
    potential point of confusion with the string sent by a modified
    F1 key (suggested by Bram Moolenaar).
  • improve plink.sh by checking if the linker
    supports the --as-needed option (suggested by
    David Philippi).
  • improve clearing for private mode 1049 when switching to
    the alternate screen (Debian #711758).
  • amend fix from patch #292 by
    removing now-unneeded chunk which introduced problem in
    selection (Debian #714527).
  • add definition to optionally override compiled-in class for
    xterm from makefile, needed for test-packages.
  • add configure option
  • add experimental support for sixel graphics (Ross
  • modify sample xterm.spec to use newer icon
  • add configure option --with-icon-symlink to
    work around systems which map icon requests for to a single
    “xterm” icon, but neglect to install the icon needed for window
    decorations (report by H Merijn Brand).
  • improve parameterizing of sample xterm.spec
  • amend fix for printer from patch
    #280, removing a reset of the signal handler for
    SIGCHLD (report by Joe Julian).
  • set environment variable XTERM_FILTER if a
    locale-filter is used.
  • enable DEBUG logic when
    --enable-trace configure option is given.
  • improve description of initialFont,
    set-vt-font and set-tex-text in
    manpage (Debian #707899).
  • fix regression from patch #292;
    selecting a word that ended at the right margin without
    wrapping would not select the last cell (report by Christian
  • add limit-check when double-click selects a word; if the
    saved-lines were scrolled back and the word selected wrapped at
    the screen’s lower right corner, an assertion was triggered
    (patch by Taketo Kabe).
  • correct limit-check in unsaveEditBufLines,
    fixing a case where enlarging the terminal size could show a
    blank line where there actually is available text (patch by
    Bertram Felgenhauer).
  • add PasteControls feature to
    allowWindowOps which by default disallows pasting
    control characters other than formatting such as carriage
    return (discussion with Hayaki Saito).
  • fix typo in ctlseqs.ms for DECRQM response (patch by
    Emanuele Giaquinta).
  • update default shown in configure --help
    message for --with-terminal-id option (patch by
    Andres Perera).
  • rename script/preprocessor variables used for substituting
    the backarrowKey default value for consistency
    (patch by Andres Perera).
  • set umask to 077 before creating debugging-trace
  • add configure option --disable-openpty to
    control whether openpty() may be used in preference to
  • improve workaround from patch #279
    for BSD systems which do not complete initialization until both
    sides of the pseudoterminal are opened, by preferring openpty()
    function over posix_openpt() when available (report by
    Christian Weisgerber).
  • add validity check for xterm widget parameter to
    AlternateScroll function, needed to handle wheel
    mouse events in the scrollbar area since patch #282 changes which introduced
    alternateScroll feature (Redhat #874327).
  • revert of patch #282 change was
    incomplete (report by Jim Reisert).
  • fix typo in manpage (report by Vincent Lefèvre).
  • revert the patch #282 change which
    restored “lost” text after shrinking/growing the screen size.
    If the screen was updated between the two resizing operations,
    unexpected text might be shown at the end (report/testcase by
    Joe Peterson).
  • reverse the arrow-keys sent for
    alternateScroll to match the usage in browsers
    (suggested by Dieter Roelants).
  • modify mouse reports to consistently indicate motion events
    whether or not a button is pressed. Old behavior (since
    patch #127 modified encoding for wheel
    mouse) showed only that the button was released if no button
    had been pressed (report by Andy Koppe).
  • correct and improve behavior for SGR 1006 mouse protocol on
    button release in any-event mode. Initial implementation in
    patch #277 left internal state showing
    the last button pressed when reporting motion events. It now
    keeps track of all buttons which have been pressed and
    released, showing the lowest remaining button by number in the
    response (reports by Hayaki Saito, Andy Koppe).
  • fix inconsistency between TermColors and
    OscTextColors enums which interchanged values for
    tektronix cursor color and highlight foreground in control
    sequences in patch #225 (report/patch
    by Peder Stray).
  • fix inconsistent use of noreturn-attribute in
    Exit function which interfered with clang
  • remove incorrect free used in computation of
    XTERM_SHELL variable, from Coverity fixes in
    patch #288 (FreeBSD #175782).
  • add --with-valgrind option to configure script
    for consistency with my other programs.
  • update table of ambiguous width characters in wcwidth.c
    based on Unicode 6.2.0
  • update table of combining characters in wcwidth.c based on
    Unicode 6.2.0
  • update precompose.c based on Unicode 6.2.0
  • fix a special case in ShowCursor where the
    foreground and background colors of the current position are
    the same. In that case, choose the further of the window’s
    foreground and background colors for the cursor color.
  • modify alternateScroll feature added in
    patch #282 to use either
    CSI or SS3 according to the cursor
    keys application mode setting, to simplify using it in vi, etc.
    (suggested by Dietar Roelants).
  • revise manpage for resize to clarify the
    changes which resize may make to the terminal
    settings as well as to the terminal itself.
  • remove code such as struct ttysize, used for
    SunOS 3 and 4.
  • modify configure script and makefile to check for
    groff and suppress groff-specific rules if groff
    is not found.
  • modify ctlseqs.ms and makefile to work around bugs in
    grohtml which cause “.png” files to be truncated
    when producing html documentation.
  • modify suffix rules for “make docs” to make the names of
    the generated “.png” files predictable, helping to work around
    longstanding problems with this feature of groff—none of its
    releases work for all of xterm’s tables..
  • correct typo in ctlseqs.ms for DECIC and DECDC; the
    character preceding the final “~” is an ASCII
    single-quote “'” rather than a space (report by
    Paul LeoNerd Evans).
  • update CF_GCC_VERSION macro, handling both Debian and
  • improve configure script checks for the
    --with-desktop-category option. After patch #280 changes, if no value was given for
    this option, and no existing desktop files found for
    comparison, the incorrect “auto” value was passed into the
    generated desktop file (report by Julien Cristau).
  • modify sample build-scripts to disable check for
    imake due to code-rot in Xorg.
  • add -v command-line option to
    resize, reporting the same version string as
    xterm and the two shell-wrappers.
  • make name-transformation apply to other occurrences of
    resize, other manpages.
  • special-case the name-transformation in xterm’s manpage in
    the NAME section so that the result works with
    makewhatis (report by Julien Cristau).
  • modify install-rules for manpages to put each program’s
    respective transformed name into the header rather than
  • remove deprecated files: proto.h, os2main.c
  • minor fixes based on Coverity scan, including:
    • correct caching of Atom value for the font menu’s
      “Selection” entry.
    • ensure that utmp/utmpx calls do not depend upon
      trailing nulls in the ut_id and
      ut_line values.
  • fix a case where in_put() loop could continue when it has
    found an X event rather than the pty, causing the pty-read to
    be delayed (patch by Balazs Kezes).
  • add option to pointerMode resource and
    corresponding control sequences to allow text-cursor to remain
    hidden if the pointer leaves/enters the window via accelerator
    keys. Ordinary motion within the window restores the pointer
    (prompted by patch by Balazs Kezes).
  • remove special case of modifiers for the
    Mode_switch and XK_ISO_Level3_Shift
    from patch #223. Verified that the
    conflict which this was intended to resolve does not happen for
    normal compose- or modeswitch-handling (prompted by patch by
    Balazs Kezes, Debian #638694).
  • modify the handling of focus-change events to ignore those
    whose detail is NotifyPointer, since those are
    sent in addition to focus-change events directed to the old/new
    windows having focus. In particular, this prevents the
    urgency-hint from being reset inadvertently (prompted by patch
    by Balazs Kezes).
  • modify description of iconName in manpage to
    make it clearer that this resource does not specify the
    filename of an icon (discussion with Ian Collier).
  • modify iconHint resource to allow suppressing
    the built-in icon by setting this to “none” (discussion with
    Ian Collier).
  • make -iconic option work with the toolbar
  • correct initialization of window manager hints used to set
    the icon-pixmap. That detail from patch
    #282 overwrote state used to control the
    -iconic command-line option (report by Ian
  • fix paste64 feature for the case where the selection comes
    from the current terminal. The selection callback can happen
    after the reply is partly built, making it necessary to flush
    the reply-buffer to keep things in the proper order (report by
    Ailin Nemui).
  • reject command-line options which are longer than the
    options in xterm’s table, to avoid mismatches (Redhat #875305).
  • correct attribution for patch #282
  • fix minstall.in to work with “make docs” rule,
    so that patch number appears in the corresponding generated
  • modify minstall.in to keep the name shown in
    the heading consistent with any renaming, e.g., for
  • check for misconfigured printerCommand
    resource on the first use, warn and disable it if it does not
    specify an executable command (Debian #691642).
  • improve check for window-manager name needed to establish
    usable default for activeIcon resource. This works around a
    scenario where gdm does incomplete cleanup,
    leaving window properties that refer to windows which no longer
    exist (Redhat #869959).
  • add/adapt FreeBSD ports files for test-building.
  • modify minstall.sh to use the patch-number and
    date in the manpage footer.
  • add check for failure to allocate the fg/bg colors at
    startup, e.g., misconfiguration. Work around by setting both to
    the default colors (suggested by Scott Bertilson).
  • amend change for patch #280 which
    added modifyKeyboard, to make
    modifyOtherKeys work (report by Ailin Nemui).
  • add query-colors.pl example for
    OSC 4.
  • correct mis-applied fix for minstall.sh (report by Miroslav
  • amend configurability changes for icons from patch #283, moving the new functionality to
    the iconHint resource to avoid conflict with
    existing uses of the iconName resource in the
    Shell widget (report by Emanuel Haupt).
  • fix documentation errata reported by Miroslav Lichvar:
    • correct manpage default for allowTcapOps
      (see patch #243
    • modify documentation for configure
      --enable-backarrow-key to match the script
      (see patch #280).
    • fix typo in minstall.sh in patch #283, which made the leading-cap
      “Xterm” become “xterm” (patch by Miroslav).
  • fix two typos in configure.in from patch #283 changes; one broke the
    --regex configure option (report by H Merijn
  • modify makefile and script to generate appropriate renaming
    s/xterm/xterm-dev/ for test-package’s manpage.
  • improve makefile rules for installing/uninstalling pixmap
    files, to include the newer mini- and filled- variants.
  • clarify change in manpage regarding patch #282 change to always set a
    window-decoration icon, even if iconName is not
  • add a fourth flavor of window-decoration icons,
    “filled-xterm” for a monochrome xterm with filled
  • use shape-mask to improve contrast of the window-decoration
  • minor changes to icon colors to improve contrast on
    bright-mud backgrounds.
  • drop configure checks for memmove versus bcopy, and
    strerror function; all currently tested platforms support
  • cleanup configure macros, as done in byacc 20121003.
  • improve configurability of the built-in icon selectable via
    the iconName resource.
  • update config.guess, config.sub
  • improve configure check for
    XkbKeycodeToKeysym, fixing a regression on some
    older systems in the patch #280
  • add clarification in manpage for
    alwaysBoldMode and veryBoldColors
    resources (prompted by Derek Martin question about using the
    former with TrueType fonts).
  • add alternateScroll resource and corresponding
    control sequences which modify the scroll-forw and
    scroll-back actions: when the alternate screen is
    displayed, wheel mouse up/down will send cursor keys (Debian
  • improve rendering for the case when a Unicode character is
    absent in the bold font but present in the normal font by
    temporarily falling back to the normal font (Debian #359006,
    Debian #408666).
  • provide configure option
    --enable-double-buffer and ifdef’s to allow
    comparisons with/without double-buffering. The default uses
    normal buffering.
  • add feature to optionally scroll current page before
    clearing, controlled by resource setting
    cdXtraScroll (prompted by patch by Balazs
  • integrated patches from Arch Linux forum posting (patches
    by Balazs Kezes):

    • modify reallocation limit of line-data when resizing
      screen so that “lost” text will be restored if the screen
      is first shrunk and then grown.
    • modify output to use double-buffering to reduce
  • modify scroll-lock action to handle
    on/off/toggle keywords like other actions. Previously the
    optional parameter was interpreted as an integer.
  • extend DECSCUSR to provide a way to set the
    cursor to a vertical bar (patch by Paul Bolle).
  • add -8, -c, -d and
    -u options to 88colors2.pl and
    256colors2.pl scripts. No current terminal
    emulator recognizes C1 controls while handling UTF-8
    encoding—see notes in patch #119 and
    patch #109 regarding
    utf8controls—but the -u and
    -8 options are useful for demonstrating this
  • improve speed when changing color palettes using OSC
    controls by deferring repaint while followup controls might be
    additional color palette changes.
  • modify SGR 38 and
    SGR 48 to accept RGB index, matching the
    closest entry in xterm’s palette.
  • extend SGR 38 and
    SGR 48 to accept colon as parameter separator
    (request by Paul LeoNerd Evans).
  • modify fullscreen action to allow it to toggle
    the full-screen mode rather than only set or unset it (prompted
    by Eeri Kask’s suggestion to support
    _NET_WM_TOGGLE). This implementation does not yet
    use _NET_WM_TOGGLE.
  • add support for _NET_WM_STATE_MAXIMIZED_VERT
    (suggested by Eeri Kask).
  • make fullscreen mode work with the tek4014 window. This
    change makes only the currently active window to be changed to
    fullscreen rather than changing both windows.
  • modify TekClear function to discard pending
    output before clearing the tek4014 display (patch by Kevin
  • fix deleteIsDEL feature, broken in patch #280 changes (report by Mike Thornburg,
    forwarded by Jeremy Huddleston).
  • fix a special case of wrapping double-width characters
    (report/test-case by Ken Winstein).
  • add E3 extended capability to
    xterm-basic to match ncurses 20120728
  • fix incorrect transformation of row-parameter for
    ClearCurBackground in patch
    #279 which caused incorrect painting in some scrolling
    scenarios (report by Chuck Silvers).
  • fix typo in manpage description of +maximized
    option (Paul Maier).
  • change default for configure option
    --with-xpm, turning it on unless requested
  • add configure option --with-icon-name which
    can be used to override the default icon used in
  • extend activeIcon resource to provide
    default value which tells xterm to determine the
    window manager’s name and automatically enable the feature for
    fvwm and window maker. The explicit true/false values can be
    used to override this behavior.
  • add “mini” xterm icon, which scales better than the
    standard xterm icon, e.g., for the gnome/kde “panel”
  • extend iconName resource to search for icon
    named according to resource value and use that if found. If not
    found, use compiled-in pixmap. Both set the WM_ICON_NAME
    property which is used by various window managers, usually for
    window decoration.
  • updated autoconf patch, adding support for –datarootdir,
    which changes the default location of manpages.
  • further improved autoconf macros for configuring icons,
    from vile 9.8h
  • remove “$(srcdir)/” from install-icon rule to fix builds
    when –srcdir configuration option is used.
  • fix for configure script’s search for icon file when it is
    not found in the current directory, e.g., when building out of
    tree (patch by Thierry Reding).
  • undo the XK_Home / XK_End change from patch #280 which broke the non-VT220 keyboard
    for those keys (report by H Merijn Brand).
  • add configure check to work around warning from
    desktop-file-install for “Encoding”, noting that it emits
    deprecation warnings without providing its version
  • add configure option --with-icon-theme to tell
    xterm whether to install into an icon theme, defaulting to
    “hicolor” (Redhat #755206,
    Redhat #799614
    and Redhat #804279).
  • change --with-icondir to look for “icons”
    directory rather than “pixmaps” by default. Add new option
    --with-pixmapdir to allow these two uses to be
    installed concurrently.
  • modify configure options --with-app-defaults
    and --with-icondir to accept “auto” as value to
    look for existing directory from well-known locations.
  • add missing codes for secondary DA response which tell the
    type of terminal as in the primary DA response.
  • update primary DA response for VT420 to include user
    windows and horizontal scrolling.
  • add modifyKeyboard resource to support
    keypad-modifier changes per request by Thomas Wolff.
  • add XK_Home / XK_End to editing-keypad checks
  • add back-tab key kB to termcap entries to
    match the terminfo.
  • update terminfo to match corresponding entries in ncurses,
    e.g., add XT and adjust sgr string
    for xterm-bold entry.
  • add configure options to predefine the most commonly
    customized resources for terminal-settings, use those to
    substitute into the manpage the corresponding compiled-in
    values: altSendsEscape, backarrowKey,
    backarrowKeyIsErase, deleteIsDEL,
    metaSendsEscape (Redhat #819588).
  • change default emulation level to VT420 (discussion with
    Ailin Nemui).
  • improve logic for DECSCL; patch
    279 partly enforced limitation of this to higher-level
    emulations. This patch finishes that change.
  • improve check for base of combining characters, preventing
    combination when cursor movement has intervened. Mosh’s webpage
    gives two different examples of this, with other causes
    ascribed to the behavior.
  • modify IL/DL to set cursor to first column on row for
    better VT102/VT220 compatibility (prompted by code review of
  • separate state-table entries for DECSC/DECRC from the
    ANSI.SYS save/restore cursor, since the latter conflicts with
    DECSLRM when DECLRMM is enabled since patch #279 (report/analysis by Ailin
  • fixes to avoid using SIGCHLD handler of main
    program in printer’s subprocess (report by Joe Julian).
  • correct restore-parameter for mouse-modes from patch #279 changes.
  • add null-pointer checks to input-method caching added in
    patch #277 to fix a problem in the
    exposure code, when deselecting a window (report by Kriston
  • unset DESKTOP_STARTUP_ID environment variable
    to lessen confusion on the part of GTK applications which use
    the variable without checking it (report by John Little).
  • fill in missing cases in the save/restore modes feature
    (report by Thomas Wolff).
  • add check to ensure that combining characters are
    precomposed in the order given (report/analysis by Andries E
  • improve workability check for posix_openpt(), to take into
    account BSD systems which do not complete initialization until
    both sides of the pseudoterminal are opened (report by
    Christian Weisgerber).
  • amend fix for Debian #650291
    in patch #277 changes to account for
    different data returned by vnc4server (Debian #670638).
  • add check in DECCRA operation to make copies of blinking
    text also blink by updating a line-level flag.
  • modify rectangle operations to work with DECOM.
  • modify DECSERA to use the current protected state rather
    than preserving it.
  • add precompose resource to allow storing
    character data in Normalized Form D as described in http://unicode.org/reports/tr15/
    (report/discussion with Andries E Brouwer).
  • modify CBT, other ISO-6429 controls which are used by VT520
    to work with DECOM.
  • add HPR and VPR controls, ISO 6429 cursor movement used in
  • remove “linux” restriction for IUTF8 ifdef in main.c
    (prompted by Matthew Dempsky posting on
  • implement the remaining VT420-level device status
  • change DECXCPR to return page 1, rather than 0.
  • improve VT-level checks on reporting functions, such as
  • change limit on decTerminalID to 525.
  • fixes to improve vttest vt52 screen when running as a

    • add checks for some VT2xx and up controls for
      consistency: DECSCL, S1C8T, S1C7T.
    • modify DECRQSS return for DECSCL to only return that
      when running as VT2xx and up.
    • when exiting from VT52-mode, resume in VT100 level
      rather than the level before starting VT52-mode.
  • implement DECLRMM, DECSLRMM and DECNCSM (prompted by
    discussions with Ailin Nemui and Paul LeoNerd Evans). This
    modifies several controls to obey top/bottom and left/right
  • correct macro definition used for testing modes used in
    ANSI/DEC request-mode controls introduced in patch #262.
  • modify x_getlogin to check $LOGNAME and $USER
    before fallback to getlogin, so that user’s choice
    for these variables can be carried forward to the xterm process
    (Debian #611487).
  • document DECSCUSR response for DECRQSS in ctlseqs.ms
  • add zIconTitleFormat resource to allow
    customizing the “*** ” prefixed to the icon title when the
    zIconBeep feature is activated (request by Thomas
  • modify DECSCUSR to update the same internal variable as
    RM/SM 12, and document in manpage the two variables used for
    controlling blinking cursor (report by Paul LeoNerd
  • correct response data for DECSCUSR in reply for DECRQSS
    response, which had inverted the blink-value (reports by Ailin
    Nemui, Paul LeoNerd Evans).
  • add a null-pointer check in OkPasswd macro to
    fix a problem in resize with Fedora 17 and a
    serial console (report/patch by Daniel Drake).
  • add workaround for Mac OS X, which loses the window size of
    a pseudo-terminal when the tty device is opened
    (report/analysis by Egmont Koblinger).
  • updated configure check for workable
    posix_openpt versus grantpt, from
    luit fixes.
  • modify DECIC/DECDC/DECBI/DECFI implemented in patch #277 to enable them only in VT4xx mode
    and up (report by Ailin Nemui).
  • update config.guess, config.sub
  • correct initialization for eightBitMeta
    resource (FreeBSD #164101).
  • make special check for Darwin 9 (and lower) to not
    use posix_openpt (report by Christian Ebert).
  • minor fixes (adding “.ne” directives) to reduce the cases
    where groff’s utility for generating images for tables dumps
    core. Most versions of groff (I’ve found 1.19.2 to be the most
    stable) dump core when attempting to report that there are too
    few lines on a page for a table to be shown.
  • remove special case for ISC pseudo-terminals which attempts
    to open the pty in two different ways.
  • move call to grantpt before asking utempter to
    add a record, to work with kFreeBSD which does not update the
    terminal’s ownership until this point (Debian #652907).
  • document limitation of XIM interface in manpage (Debian
  • cleanup error reporting with new xtermWarning
  • add configure option --disable-selection-ops
    to make the new actions optional.
  • add four new actions for making the selection or data
    directly copied from the screen (prompted by discussion in
    Debian #637001,
    as well as report by Arjen van Tol):

    • exec-formatted
    • exec-selectable
    • insert-formatted
    • insert-selectable
  • add visualBellLine resource to allow
    visualBell to flash only the current line
    (prompted by patch by Gertjan Halkes).
  • add eightBitMeta resource to control the
    features which modify or interpret the eighth bit of a key when
    the meta modifier key is pressed (prompted by Debian #326200).
  • improve discussion of eightBitInput in the
    manpage (prompted by Debian #326200).
  • correct logic for alt-sends-escape action,
    overlooked when implementing altSendsEscape
    resource, which still used eightBitInput resource
  • add a workaround for XAllocColor(), which does
    not actually allocate “a read-only colormap entry corresponding
    to the closest RGB value supported by the hardware”, but rather
    a rough approximation (Debian #650291).
  • undo parameter checks for RequestResize()
    added in patch #251 and amended in
    patch #270, because zeros also are
    special cases (report by John S Urban).
  • modify some test-scripts to use /bin/echo
    rather than the shell’s possibly-builtin echo, to
    work around broken configuration on Mac OS X, i.e., neither
    honoring the option nor flagging an error.
  • add SGR 1006, as a better technical solution than SGR 1015:
    • the responses will not be confused with line-deletion
      and scrolling controls.
    • the button encoding is a little simpler, since it does
      not add an unnecessary 32 because the integer parameter
      does not have to be represented as a printable
    • the control responses for pressing and releasing a
      mouse button differ, allowing an application to tell which
      button was released.

    Besides these improvements, in discussion, it was noted
    that urxvt’s implementation of 1005 is incorrect, relying
    upon a locale that provides UTF-8 encoding. In contrast,
    vttest demonstrates a correct decoding, independent of

  • add support for urxvt SGR 1015 to address shortcoming of
    SGR 1005 with luit (patch by Egmont Koblinger).
  • add ISO and DEC controls useful for left/right
  • add some changes for OpenBSD and MirBSD (adapted from patch
    by Thorsten Glaser):

    • disable search for non-Unix96 ptys.
    • fix a gcc warning in
    • modify Imakefile to install xterm setgid
      to utmp.
  • add/use/prefer posix_openpt() for opening
  • modify special errno handling case in ptydata.c from
    patch #158 to allow for the
    possibility that any platform may have special cases where
    “/dev/tty” is absent. For example, this can happen in a FreeBSD
    jail (patch by David Wolfskill).
  • add keyboard logic to map shift-tab into XK_ISO_Left_Tab,
    which is usually, not always, done by the X keyboard
  • portability fixes for some configure macros:
  • modify clipping limits for TrueType fonts to account for
    the scaleHeight resource setting, to work around
    another problem due to recent FreeType changes. In this case,
    the DejaVu Sans Mono set to pixelsize=13.5 is
    truncated because the font descent is reduced by FreeType to
    match an incorrect height metric (report by Adam Lee).
  • improve recovery when bitmap fonts are not installed, e.g.,
    so that switching font-sizes works for TrueType fonts.
  • modify lookup for XTERM_SHELL feature to allow
    relative pathnames.
  • modify abbreviation disambiguation check for command-line
    parameters to account for -geometry, whose
    parameter may begin with “+” or “-” (report by Scott
  • add ash, zsh to known shells for
  • modify resize to reuse the logic from
    xterm which determines the actual logon-user’s
    shell if $SHELL is not set.
  • revert the unsetenv("SHELL") added in patch #272.
  • an unsetenv("SHELL") added in patch #272 to help ensure that
    luit would get the user’s shell consistently did
    not work as expected for cases where multiple names are in the
    password-file for a given uid. That was because changes in
    patch #157 to handle this situation
    did not take into account that repeated calls to
    getpwnam and getpwuid return a
    pointer to the same static buffer. Fixed the older logic to
    work as intended, by ensuring that the passwd-data from each
    call is stored separately (report by Paul Keusemann).
  • adjust ifdef’s for putenv and
    unsetenv in case only one of those is provided on
    a given platform.
  • correct comparison used in ExposeContains
    macro from patch #274 changes, to
    handle window-dragging (patch by Todd Eigenschink).
  • portability fixes for cygwin: do not define
    SVR4, and work around nonstandard header
  • ifdef’d use of unsetenv from patch #273 changes to work with Solaris 9
    (report by Waldemar Rachwal).
  • modify logic for XtAppPending to merge
    adjacent Expose and ConfigureNotify
    events which are redundant (report by Edward McGuire).
  • fix an unneeded warning message when -r option
    is given.
  • remove a few redundant entries from table used for helping
    abbreviation-checking of command-line options, makes
    -geom work again after patch
    #272 changes as an abbreviation of
  • add scaleHeight resource and command-line
    option -sh as workaround for some
    font-configurations broken by changes in FreeType 2.4.6 (report
    by Miroslav Hodak).
  • portability fixes for some configure macros:
  • add configure option --with-freetype-config to
    improve selection over the plethora of configuration options
    which freetype has so far provided.
  • build-fix for configure --enable-load-vt-fonts
    when --enable-widec is not specified.
  • build-fixes for suppressing various features, needed after
    changes in patches 270, 271 and 272 (report by Brian
  • build-fix for out-of-tree “make docs” rule.
  • correct a typo in x_strdup, from patch #198 changes.
  • correct initialization for -e option, broken
    in patch #272 changes (report by
    Gabriele Balducci).
  • build-fix for out-of-tree builds to address
    minstall script changes in patch #272 (patch by Thierry Reding).
  • document limitation of Gtk in connection with
    xterm’s -into option, in the manpage (Ubuntu
  • improve -into by checking for and using the
    size of the window within which xterm is embedded, overriding
    other clues.
  • modify logic for localeFilter resource to
    allow that to include command-line options of luit.
  • improve -into by checking for invalid
    window-id, and allowing hexadecimal/decimal/octal values.
  • improve keepSelection, adding the case where
    the highlighting is cleared, overlooked in patch #230 (patch by Marco Peereboom).
  • improve command-line parsing to make abbreviate options
    work consistently across xterm-specific versus standard X
    toolkit options, and report cases where an abbreviated option
    happens to be ambiguous. In particular, -d now
    works as an abbreviation for -display.
  • fix regression in command-line parsing introduced in
    patch #271 changes for Debian #629358,
    (Debian #637910).
  • split-out new termcap/terminfo building block
    xterm+kbs for configurability.
  • modify terminfo file to reflect changes in
    ncurses for xterm-16color and
  • modify minstall.sh, etc., to reflect the
    default default class, $TERM and
    decTerminalID values.
  • reword resize manpage to reflect the fact that
    $TERMCAP is not set on all systems (patch by Alan
  • work around combined Xaw6/Xaw7 package in DragonFlyBSD
    which omits the usual symbolic link to the preferred library
  • further improve build-fix for termcap systems by checking
    for some which are only partial implementations, e.g., termcap
    2.08 in CentOS 5.2
  • omit permissions adjustments to pty on exit except for
    pre-Unix98 ptys, since modern implementations handle this
    (report by Sean C Farley).
  • modify logic for switching fonts between UTF-8 and
    non-UTF-8 encoding to not merge the derivable bold-, wide- and
    widebold values from the VT100 fonts. Also suppress warning
    when not using UTF-8 fonts if wide- and widebold-fonts cannot
    be derived or otherwise loaded (report by Werner
  • modify menu-creation to suppress entries which will never
    be used in the current configuration, rather than simply
    disabling them.
  • add resource printModeImmediate and menu item
    to allow print of screen plus saved lines to a file.
  • add menu item to allow runtime enable/disable of the
    printFileOnXError feature.
  • append a timestamp to filename used in
    printFileOnXError feature, and restrict its
    permissions (request by Vincent Lefèvre).
  • add a check when cancelling cursor-blinking, in case the
    cursor is blinked off. Fix so that the cursor is repainted
    without waiting for other events, e.g., keypress (report by
    Ailin Nemui).
  • add configure --with-app-class option, to
    simplify building “xterm-dev” packages with filenames that do
    not conflict with conventional “xterm” packages.
  • corrected logic flow for DECSCL, which prevented the
    updated operating level from being reported via DECRQSS (report
    by Ailin Nemui).
  • corrected default for brokenStringTerm
    resource to match manpage.
  • add vttests/dynamic.pl
  • add runtime check for locale not supported by X libraries,
    and fallback to XA_STRING in this case (request by Bryan
  • fix a special case in configure script after no FreeType
    libraries are found. The script was proceeding to check for a
    usable configuration.
  • add xterm+tmux building block to terminfo
    (adapted from changes proposed by Ailin Nemui and Nicholas
  • improve discussion of faceName resource in
    manpage (adapted from suggestions by Jens Schweikhardt).
  • correct mapping of shifted up/down cursor-keys in termcap
    function-keys mode, i.e., resource
    tcapFunctionKeys (patch by Gertjan Halkes).
  • update AIX case in CF_XOPEN_SOURCE configure
    macro to add release 7.x.
  • modify ifdef’s in xterm_io.h for
    __hpux to force that to use the hacked SYSV
    support in that file. This fixes a problem with a non-blocking
    socket call (patch by Paul Lampert).
  • improve filtering of desktop category scanning, to exclude
  • modify configure script to work with systems that have both
    ncurses (or other terminfo) as well as a real termcap library,
    e.g., Slackware (report by Andrew Watts).
  • modify configure script to work around special case where
    user’s environment adds compiler flags to the CC
    variable (prompted by report by Paul Lampert).
  • amend change for Debian #110226
    so that “-h” or “-v” options cause an exit, rather than simply
    printing to stdout while the window is displayed (Debian
  • add response for DECRQSS which gives the
    setting for DECSCUSR.
  • modify AllocateTermColor() to handle
    XtDefaultForeground and
    XtDefaultBackground, which are not recognized by
    XParseColor. For example, this fixes the use of
    OSC 112 when no explicit cursor color was set (report by
    Ailin Nemui).
  • handle special-case of KeyPress translated to
    popup-menu action (Ubuntu #756273).
  • build-fix, e.g., for using imake on platforms
    which use the termcap library.
  • modify utf8 resource to accept a name.
  • mention default for fontWarnings in manpage
    (report by Werner Scheinast).
  • split “UTF-8” menu entry into “UTF-8 Encoding” and “UTF-8
    Fonts” (prompted by discussion with Werner Scheinast).
  • gray-out font-menu entries when a font fails to load, e.g.,
    a bitmap font is not installed.
  • improve behavior when there is no app-defaults file:
    • set the toolBar resource to false
    • gray-out the font menu entries where no resource is
  • add configure option --with-desktop-category
    to allow customization of the “.desktop” files.
  • build-fix for the install-desktop
    makefile-rule, when the source/build directories differ (patch
    by Loïc Minier).
  • add menu entry and corresponding resource which can be used
    to suppress all bold-fonts (discussion with Jan
  • make internal line-size value consistent with allocated
    sizes of character and related arrays to ensure that bulk
    copying of line data, e.g., in scrolling, accounts for the
    padding used for pointer alignment (adapted from patch by
    Rajesh Mandalemula, also reported by Ali Bahar).
  • widen ifdef for screen-resizing logic from patch #176 changes to send
    SIGWINCH to process group to include any system
    supporting ioctl(*,TIOCGPGRP,*) (prompted by
    linux-specific patch in OpenSUSE rpm package).
  • add printModeOnXError and
    printFileOnXError resources, which allow the user
    to specify that xterm will write the contents of its screen to
    a file if it is exiting due to an X error (Debian #280457).
  • restore logic that made reverse-video apply to the
    scrollbar’s foreground/background, broken in patch #158 (report by Bryan Ischo).
  • amend some of the window operations parameter-checks added
    in patch #251 for the push/pop title
    feature, to allow the parameters used for window resizing to be
    -1’s, which makes the corresponding values ignored rather than
    using the window’s maximum width/height (report by Noah
  • correct order of initialization for
    translations vs fullscreen resources
    to enable a special case which omits the Alt-Enter translation
    when fullscreen is disabled (Debian #612978).
  • update config.guess, config.sub
  • build-fixes for imake (report by Heiko
  • modify autoconf macro CF_PKG_CONFIG to work with
    cross-compile environments (patch by Thierry Reding).
  • modify MapToColorMode() to favor bold over
    underline, matching the precedence used before patch #252 (report/analysis by Nicolas
  • add omitTranslation resource, which can be
    used to suppress the default translations for these features:

    • fullscreen
    • scroll-lock
    • shift-fonts
    • wheel-mouse
  • make the fullscreen feature configurable (Debian #612978)
    • add it to the configurable list
    • add command-line option -fullscreen to
      allow the feature to be enabled at startup.
    • add resource fullscreen to control whether
      the feature is active or may be enabled.
  • modify probe_netwm_fullscreen_capability for
    64-bit machines. Contrary to XGetWindowProperty
    manpage, that function returns 32-bit data packed as
  • eliminate copy of name resource, which was
    otherwise used only to give the terminal-description name for
    the tcap-query feature. Use the actual $TERM value
    instead, as derived from termName resource,
  • eliminate an old inconsistency with error messages, some
    used the -name option, while others used
    argv[0]. The latter is now used consistently.
  • improve configure check for rpath-hack, to improve builds
    on systems where gcc will not search /usr/local/lib, etc.
  • build-fix for Xaw3d configuration (report by H Merijn
  • update config.guess, config.sub
  • fix an inconsistency of the “Enable Reverse Video” checkbox
    in the VT Options menu. This also removes a special case added
    in patch #217 which limited the effect
    of the reverseVideo resource (Debian #603808).
  • amend decoding of misformed UTF-8 sequences to avoid
    absorbing valid characters as documented in Unicode 6.0 section
    3.9 (report by Keith Winstein).
  • do not set urgency hint when window already has focus
    (patch by Dimitrios Christidis).
  • amend extended mouse-coordinate mode from patch #262 changes to include the Cb
    button-code, which also may be greater than 127 (report by
    Ailin Nemui).
  • ensure that underline-cursor is visible when an application
    happens to set the background color (report by Christian
  • add feature for
    full-screen toggling using either Alt-Enter or a menu
    selection (integrated patch by Dave Simmons).
  • add missing logic to handle reallocation of FIFO index for
    the “UTF-8” menu entry (report by David Holland, NetBSD


  • add makefile rules docs-ctlseqs,
    docs-xterm, etc.
  • correct typo in description of DECRPM in control sequences
    document (report by Ailin Nemui).
  • minor formatting changes to ctlseqs.ms to simplify a script
    which extracts the feature information. See the results in

    Comparing versions, by counting controls in the xterm

  • add docs-clean makefile rule.
  • add copy-selection action (request by Timo
    Juhani Lindfors, Debian #588785).
  • trim leading/trailing blanks from string used for
    “Selection” font-menu data.
  • trim leading/trailing blanks from color resource
  • configure script improvements:
    • add workaround for removal of X11 dependency from Xt’s
      package file (report by Robert Hooker).
    • add workaround for removal of fontconfig dependency
      from Xft’s package file (report by Jeremy Huddleston).
    • add workaround for removal of Xmu dependency from Xaw’s
      package file (report by Jeremy Huddleston).
    • improve workaround in CF_X_TOOLKIT macro,
      checking for other possible packages where Xt’s
      dependencies may be given.
    • prefer ${name:=value} to
      ${name-value}, since recent bash changes break
      legacy support for that feature.
  • add rpm and dpkg scripts, for testing.
  • more fixes for Debian #600707
    (report by Cyril Brulebois).
  • fix a regression in fontname logic from patch #263 changes (Debian #600707,
    reported by Vincent Lefèvre).
  • revert modification of any-event/any-button protocol from
    patch #263 changes. It interferes with
    selection using a shifted mouse button (reports by Neil Bird,
    Bram Moolenaar).
  • replace a null-pointer check with check for empty string in
    xtermOpenFont, to eliminate a warning message from
    patch #263 changes.
  • build-fix for patch #263 when
    toolbar is not configured (patch by Chris Clayton) (reports by
    Robby Workman, David Wood).
  • corrected initialization of “misc” resource values, to
    ensure that xterm has allocated a copy of strings which may not
    have been malloc’d by the X library (Debian #600129).
  • modify handling of any-event/any-button mouse protocol; it
    now is active with any combination of key-modifiers.
  • add debugging feature showWrapMarks, which
    marks lines which xterm knows are wrapped, showing where a
    double-click will select past the end of a line.
  • build-fix to address change in include-guards for
    Xlib.h in ongoing Xorg edits (patch by Jeremy
  • improve pointerMode by continuing to watch for
    motion events after mouse tracking is disabled if the pointer
    is hidden (Debian #594856).
  • further extend initialization for active-icon font to check
    if the font was not loaded successfully, to retry with
    font1, or as even (if TrueType fonts are used) to
    use a TrueType font. The retries are to help with cases as in
    patch #241 where the bitmap fonts are
    not available.
  • fix special case of active-icon used when TrueType font is
    specified for the xterm window, from patch
    #261 change. In that case, the default font’s size was used
    for layout of the active icon’s window (Debian #591265).
  • fix a case where changing the cursor color via escape
    sequences did not immediately update the screen (report by
    Andreas Wagner).
  • implement ANSI and DEC request-mode control sequences. The
    latter includes the xterm-specific private modes such as the
    mouse mode. The feature is ifdef’d with the rectangle
    operations since its decoding overlaps that feature.
  • correct typo in ctlseqs.ms for response of OSC 21 (patch by
    Kevin Schoedel).
  • improve discussion of mouse tracking in ctlseqs.ms
  • increase an array limit used in reporting mouse events
    (report by Ryan Johnson).
  • add extended mouse-coordinates mode, allowing up to
    2015×2015 windows, using UTF-8 encoding (patch by Ryan
  • modify manpage hyphens to conform with
  • fix regression in renderFont logic, from
    patch #260 changes (report by Joseph
  • modify plink.sh to work around problem linking
    to recent PCRE libraries.
  • extend renderFont resource to allow deferred
    switch to TrueType fonts without affecting existing resource
    settings (Debian #585620).
  • modify configure macro CF_X_TOOLKIT to work around omission
    of ICE library from “.pc” file (report by Miroslav
  • change configure script default for
    --enable-broken-st i.e., the
    brokenStringTerm feature) to normally enable it.
    If the corresponding resource is enabled, this feature
    eliminates an apparent freeze of xterm when sending mis-encoded
    data to the screen (Debian #584801).
  • document in manpage some actions which were overlooked:
    • readline-button
    • scroll-lock
    • set-8-bit-control
  • undo a change to limit-check in ScrnRefresh in
    patch #257, which broke
    fastScroll feature (Debian #584841).
  • modify handling of brokenLinuxOSC and
    brokenStringTerm to also sound the bell.
  • add control/D and control/Q to controls which will cause
    early exit from control string per
    brokenStringTerm resource.
  • improve documentation of brokenStringTerm
    resource in manpage.
  • modify configure check for luit to include new
    aliases for the program (xterm-filter and
  • add workaround in xtermClearLEDs() to account
    for Xkb’s override, making vttest’s LED demo reset the scroll
  • filter out client-message events when deciding whether to
    hide cursor, e.g., when using SCIM (patch/report by anonymous
  • improve description of -bd option in manpage
    (report by Guy Daniel Clotilde).
  • modify configure checks for PCRE and other libraries to use
    pkg-config, if available.
  • amend change from patch #252, to
    take veryColorColors resource into account when
    checking colorBDMode resource for TrueType fonts
    (report by anonymous user).
  • add vttests/query-fonts.pl script for
    demonstrating the OSC 50 font query.
  • improve manpage discussion of Scroll Lock feature.
  • improve configure macros CF_GCC_VERSION and
  • fix warnings for “clang –analyze”.
  • change default for allowScrollLock resource to
    false, noting that the supposedly unused key has
    been useful for various rebindings (Debian #580946).
  • add pointer-checks in ScrnRefresh to fix a
    case in rapid scrolling where an empty record is fetched from
    the scrollback FIFO.
  • correct ctlseqs.ms description of OSC 17 and OSC 19 (patch
    by Emanuele Giaquinta).
  • corrected logic for menuLocale resource; the
    setlocale function returns the original locale
    only when querying.
  • improve filtering of translations resource, narrowing the
    scope of the alwaysUseMods to address only the
    translations that would cause a key to be sent to the host
    (report by Andrew Gaylard).
  • change default value of menuLocale resource to
    “C”, to work around longstanding Xorg
  • modify handling of scrollKey feature to ignore
    XON/XOFF keys.
  • implement scroll-lock feature.
  • revise memory allocation in UTF8toLatin1() to fix an
    out-of-bounds index (Mandriva #54531).
  • compute value for first wide-character rather than assuming
    it is 256, fixes problem with -cjk_width
    introduced in patches 242 and 249 (report by Thomas Wolff).
  • improve configure script:
    • corrected check for _XOPEN_SOURCE for
    • when possible, add rpath option for libraries in
      unusual places
    • add configure option --disable-rpath-hack
      to control whether the rpath option can be added.
  • modify AllocateTermColor() to separate
    initialization from control sequences, fixing problem from
    patch #254 changes where enabling
    allowSendEvents resource prevents setting cursor
    color on command-line (Debian #572928).
  • amend logic from patch #185 to not
    reallocate cell-array if processing
    ESC % G to switch from UTF-8 if already
    in ISO-8859-1 character set (report by Michael Koehne).
  • fix to avoid calling XmuInternStrings() with
    zero count (report by Johan Bockgård).
  • fix build when –disable-ansi-color configure option is
  • fix build when neither OPT_TCAP_QUERY or OPT_TCAP_FKEYS is
    defined (patch by Matthieu Herrb)
  • add TerminalEmulator to desktop category files.
  • modify sinstall.sh to ignore the “.” appended
    to permissions by selinux.
  • change app-defaults organization, installing UXTerm-color
    and KOI8XTerm-color for consistent behavior regarding
    customization: color (prompted by discussion
    in Ubuntu #421261).
  • fix typo in minstall.sh from patch #255 changes, and add case for
    /var/run needed for full path of utmp (report by
    Julien Cristau).
  • minor fix to xterm manpage, remove a comment stating that
    margin bell can be changed via the VT Options menu. That was
    replaced in patch #225
  • add a “docs” rule to makefile.
  • fix initialization of Atom used for XkbBell
    feature from patch 243 changes. Unlike
    the other calls to XInternAtom(), in this case the
    flag telling X to create the Atom was unset (patch by Chris
  • rename install.sh to install-sh
    in case suffix-rules might interfere.
  • extend range for convertToUTF8 function to
    full 31-bits, to use with printing, etc.
  • improve manpage by checking for actual locations of
    utmp/wtmp files (Debian #562640).
  • modify configure macro CF_XOPEN_SOURCE to remove -D’s
    before adding the same name rather than relying on -U’s, to
    reduce redefinition warnings for some platforms that have
    conflicting definitions in headers.
  • correct logic used to switch to alternate screen using
    FIFO-lines configuration (Debian #565772).
  • update config.guess, config.sub
  • add a configure-check to eliminate install-ti
    rule from Makefile when the system has no tic (terminfo
    compiler) program. This lets one use the
    install-full rule more consistently.
  • amend change to WriteText() function in patch #252 to take into account the
    colorAttrMode resource (report by Krzysztof
  • document titleModes resource in manpage, added
    in patch #252.
  • modify tcap-query table entries for shifted up/down cursor
    keys to match ncurses convention.
  • improve lookup of termcap-query data, allowing for
    duplicate keycodes versus missing entries.
  • add control sequence which can be used to modify the
    terminal data used for the termcap-keyboard.
  • improve portability of tcap-query feature, using terminfo
    functions in preference to termcap on systems having
  • improve font-setting/querying control
    (OSC 50):

    • when TrueType font is selected, the TrueType
      faceName will be set, rather than the bitmap
    • when TrueType font is selected, querying returns the
      name of the TrueType font.
    • querying a font recognizes the relative-font convention
      that setting a font could use.
  • add menu-entry for allowColorOps.
  • add new resources for fine-tuning menu entries:
    allowColorOps, disallowedColorOps,
    disallowedFontOps and
  • correct logic for disabling the “TrueType Fonts” menu item;
    it was not ensuring that the faceName resource
    value was non-empty.
  • implement VT520-style controls DECSMBV and
    DECSWBV for setting the margin- and warning-bell
  • fix a minor error from patch #243
    which made the zIconBeep feature use a minor-error tone rather
    than an informational tone.
  • add a null-pointer check for the case where
    renderFont resource is true, but
    faceName resource is unset, used in logic to strip
    “xft:” prefix from patch #251 changes
    (patch by Michael Riepe).
  • add special case to configure CF_XOPEN_SOURCE macro to use
    extensions on Darwin (patch by Dennis Preiser).
  • improve configure checks for regular expressions header and
  • update config.guess, config.sub
  • add a null-pointer check in getPrinterFlags()
    per changes in patch #252.
  • add a null-pointer check, needed in UTF-8 mode for Xft
    fonts after changes in patch #252
    (patch by Alan Coopersmith).
  • correct size when clearing struct for tek4014 (patch by
    Jochen Voss).
  • parenthesize expression in MoreRows macro,
    fixing a limit check added in patch
    #251 (Debian #560039).
  • modify title-querying logic to support retrieval of titles
    encoded using UTF-8.
  • add new “title-modes” control sequence for controlling
    whether window/icon titles can be set or queried using UTF-8,
    optionally encoded in a hexadecimal string.
  • use mkdir -p rather than
    mkdirs.sh (prompted by discussion of mawk by
    Aleksey Cheusov).
  • add appropriate copyright/license notices to the bulk of
    files lacking same, and modify to use identical terms in others
    which used different wording. The main exceptions are the files
    contributed by Markus Kuhn, who appears to prefer public domain
    distribution, noting that I have modified/improved several of
    these without changing the terms of distribution (request by
    Jari Aalto).
  • add control sequences for resetting the “dynamic” colors to
    their default values.
  • add control sequences for resetting the “special” colors to
    their default values.
  • add control sequences for setting the “special” colors such
    as colorBD (bold).
  • add overlooked case for setting highlight foreground color
    with the dynamic-colors control.
  • add OSC 104, for resetting ANSI/16/88/256
    colors to default.
  • reset ANSI/16/88/256 colors to default in soft/hard reset
  • strip “xft:” prefix from faceName and
    faceNameDoublesize resource values,
  • add DECSCUSR (discussion with Andy
  • add check/warning on failure to load font, to see if the
    name looks like an Xft pattern rather than XLFD. This can
    happen if someone happens to modify their X resource settings
    for programs that read data from xterm’s namespace.
  • modify handling of print and
    print-everything actions to allow the various
    printer flags to be overridden by supplying parameters.
  • add resource printerNewLine (request by Ovidiu
  • minor cleanup, finish using TScreenOf() and
    TekScreenOf() macros introduced in patch #224.
  • improve checks in ShowCursor/HideCursor to get the
    background color, particularly when
    highlightReverse resource is used (reports by Jan
    Engelhardt, Christian Weisgerber).
  • correct checkVeryBoldAttr to omit comparison
    of foreground color to special color values used for
    color<XX> resources (report by Jan
  • make colorBDMode and colorULMode
    resources work with TrueType configuration (report by Jan
  • add window-ops controls to push/pop icon and/or window
    labels on a stack.
  • minor fixes to align termcap file with terminfo.
  • add resource disallowedWindowOps, to allow
    fine-tuning of features to suppress with the
    allowWindowOps resource (prompted by discussion
    with Bram Moolenaar).
  • add makefile rules for resize-manpage to pdf, etc.
  • further improve limit-checks in select/paste (Mandriva
  • add check and error-message for fonts that have no
    printable values in the ISO-8859-1 range (Debian #542434).
  • some compiler-warning cleanup, in particular workaround for
    defective implementation of gcc’s attribute warn_unused_result
    (report by Bram Moolenaar).
  • improve estimate of single-column width for packed TrueType
    fonts by ignoring extents for codes 127 and 159.
  • improve line-drawing for TrueType fonts which happen to
    have defined glyphs which are not line-drawing in
    0..31 by assuming they’re not, and just checking the existence
    of the Unicode codepoints. This makes it more likely that the
    user can override a misconfigured font using the “Line-Drawing
    Characters” menu entry.
  • limit minimum cell-width for packed font to maximum-advance
    reported by Xft. Some fonts are wider than that, even in the
    Latin-1 range (Debian #550497).
  • add list of direct-contributors in “THANKS” file.
  • stylistic changes to this file to help scripted extraction
    of list of contributors.
  • correct off-by-one in okPosition fix from
    patch #249 limit-checks which
    prevented double-click selection on the bottom line of the
    screen (reports by Rajeev V. Pillai, Debian #550368).
  • add -q option to
    vttests/256colors2.pl and
    vttests/88colors2.pl to demonstrate bulk
    initialization of color palette.
  • improve the workaround from patch #188 by enabling
    resources for the 88-color model.
  • document in xterm manpage the limited availability of
    resources color16 to color255 as
    noted in patch #188 (Ubuntu #438850).
  • change default for allowWindowOps resource to
  • add limit-checks for result of visual_width() function,
    needed from patch #242 and exposed by
    #244 changes (Debian #548321).
  • improve limit-checks in select/paste.
  • fix a remaining bug from patch
    #230 changes for displaying multi-column characters in a
    proportional font (report by Chris Jones).
  • add new resource forcePackedFont and menu entry “Packed
    Font” to control whether to use the font’s minimum (default) or
    maximum width when those differ. The workaround which xterm
    uses to accommodate proportional fonts is not necessary with
    certain fonts such as unifont which happen to store a mixture
    of multicolumn glyphs (report by Chris Jones).
  • fix an (old) bug which did not restart the timer for
    blinking text if the only blinking text was temporarily
    scrolled out of view, e.g., using the scrollbar.
  • fix an (old) flaw in the delete-line operation where the
    text which is scrolled into view while the display is scrolled
    up was not repainted.
  • improve delete-line and insert-line operations, retaining
    selection when the selection does not intersect the
    deleted/inserted lines.
  • fix an (old) off-by-one error when an application cleared
    above the cursor position while the display was scrolled up,
    that would leave an extra line of text uncleared.
  • fix a similar problem where the double-size attribute would
    not be reset when clearing the screen while the display was
    scrolled up.
  • fix an indexing error which would occur if an application
    cleared a line while the display was scrolled up and was also
    in UTF-8 mode (Redhat #524503).
    The error was from patch #228 but more
    visible after changes from patch
  • fix an overlooked adjustment for selecting double-width
    characters in the narrow-character configuration.
  • eliminate uses of XTERM_CELL and
    XTERM_CELLC where an appropriate
    LineData pointer is available.
  • correct expression in okScrnRow macro, making
    selections give the length of lines below the visible screen
    (report by Stuart Henderson).
  • correct logic used to improve performance of missing-glyph
    check, which did not handle line-drawing characters (Debian
  • add ifdef’s and check for openpty() on DragonFly (patch by
    Alex Hornung).
  • correct calculation for size of line’s data block, which
    was sometimes off-by-one when configured using
    --enable-16bit-chars (report by H Merijn
  • fix indexing error in print-everything feature from
    patch #246 (patch by Ovidiu
  • remove obsolete logic for saving/restoring wrapping flags,
    which did not work on 64-bit platform. Wrapping flags (stored
    in the line-index) are now copied with line-data (Debian
  • modify comments in app-defaults files to avoid problem with
    C preprocessor used by xrdb (Debian #541603).
  • restore special case in makeColorPair, needed
    for colorBDMode resource (Debian #541089).
  • correct SetLineFlags() macro, broken in
    patch #244 when recoding to avoid
    gcc-specific bitfields (Debian #541236).
  • modify initialization of screen buffers to ensure that
    pointers align to int-boundaries. This fixes a problem
    introduced in patch #244 where the
    color- and character-arrays (stored after the video-attributes
    in each row) might be misaligned (report by Rajeev V
  • add limit-check in ScrnRefresh for handling
    saved-lines from the circular buffer which are repainted on a
    screen whose width has increased. To improve performance,
    circular buffer entries are not resized (report by Rajeev V
  • correct type for CellColor (a late change in
    patch #244 to avoid gcc-specifc enums
    made that unsigned rather than unsigned short, for the
    256-color option).
  • fix typo in configure option –enable-16bit-chars (report
    by Rajeev V. Pillai).
  • correct a special case in saving FIFO-lines from patch #244. If the screen was shrunk, xterm
    used the wrong amount for copying to FIFO-lines, and then used
    this amount to adjust the current row on the screen. That was
    both a visible defect (Debian #541109)
    as well as a potential addressing error (Debian #541132,
    Debian #541160,
    and Debian #541236).
  • add clarification in xterm manual about the various
    allowXXXOps resources, which are disabled
    when the allowSendEvents resource is active (patch
    by Julien Cristau, Debian #531597).
  • refactored storage of saved-lines, providing a configure
    option to manage them as a FIFO (actually a circular buffer),
    improving performance. Added configure option
    --enable-fifo-lines to enable/disable the new
    feature (it is enabled by default).
  • added fastScroll resource, to amuse people who
    measure terminal emulator performance by cat‘ing
    large files to the screen.
  • modify check in readPtyData from return values
    to provide exit on zero-bytes read from pty for FreeBSD, or
    eliminate high-CPU in “xterm-hold” processing (discussion with
    Ulrich Spoerlein, FreeBSD ports/136686). The check was
    originally modified to combine negative/zero values in
    XFree86-3.1.2E, 1996/05/06.
  • add configure option --enable-16bit-chars to
    provide wide-characters with 16-bits (rather than the default
  • add retryInputMethod resource to allow
    configuring out the retries xterm uses to connect to
    non-responsive XIM server, to work around defective X
    configurations as noted in
    NetBSD mailing list.
  • make regular-expression selection work for VT100
    double-sized characters.
  • improve layout when drawing missing characters in a
    proportional font, e.g., as boxes, to take into account whether
    they are double-width (report by Guilbert Stabilo on
  • add capability for keypad-center (kb2/KA2) to termcap entry
    for xterm-new, as well as xterm-8bit, xterm-sun and xterm-vt220
    (FreeBSD conf/136336).
  • change default for keepSelection resource to
    true (prompted by discussion with David Muir Sharnoff).
  • remove a limit-check in ptydata.c, allowing Unicode values
    past 64k to be displayed using TrueType fonts (Debian #458432).
  • remove a vt52-specific ifdef to allow mapping F1-F4 to
    PF1-PF4 when vt52 support is not compiled (report by Olaf
    ‘Rhialto’ Seibert).
  • save/restore line-wrapping flags when converting from
    ISO-8859-1 encoding to UTF-8 encoding, as well as when resizing
  • remove extra adjustment of position in fix for Debian
  • modify default check for mkWidth resource to
    check for line-drawing characters, which are categorized as
    double-width in Solaris 10 (report by Sebastian Kayser).
  • add “print-everything” action (patch by Ovidiu
  • start refactoring scrollback data using new getLineData()
  • demote recent change to Debian #252873
    fix to experimental, ifdef’d out as EXP_BOGUS_FG
    (Debian #522141).
  • work around groff mapping of ASCII quotes using macros
    (requested by Reuben Thomas based on Colin Watson advice, fixes
    Debian #378700).
  • correct symbol used for default of
    allowWindowOps which was
    DEF_ALLOW_FONT rather than
    DEF_ALLOW_WINDOW (report by Matthieu Herrb).
  • amend fix for tek4014 from patch
    #243 to make it only apply to the Tek Options menu.
  • revert change to default for allowTcapOps
    (request by Bram Moolenaar).
  • reallocate result returned by xtermEnvLocale()
    to avoid reference to freed memory after handling
    menuLocale resource.
  • fix an old (X11R5) bug in tek4014 for switching
  • add resource defaultString to make
    configurable the use of “#” when pastes of UTF-8 text fail due
    to limitations in the current locale settings.
  • make the set of selection target Atom’s configurable by two
    new resources eightBitSelectTypes and
    utf8SelectTypes, e.g., to use the
    TEXT Atom in preference to
    UTF8_STRING (discussion with Stanislav Sedov
    regarding koi8rxterm and the FreeBSD port).
  • modify handling of TARGETS Atom by making it
    return exactly the set of targets as those which xterm is
    currently providing.
  • set MANPAGER and PAGER explicitly
    to /bin/cat in minstall.sh to work around
    /etc/man.conf’s with those variables already set (report by
    Марьясин Семён).
  • improve error-checking of tcap-query parser.
  • add check for keyboard tcap), which ensures
    that terminal descriptions containing the same string for
    shifted/unshifted keys will be seen by tcap-query as only the
    unshifted key. (This would only happen with an incorrect
    terminal description).
  • fix conversion for input event-state to modifier-parameter
    which made tcap-query feature not work with
    tcapFunctionKeys (keyboard type
  • add “DEF_ALLOW_XXX” definitions to main.h to allow
    overriding the default compiled-in values for
    “allowxxx” resources.
  • remove check on bell-percentage added in patch #242, which disallowed zero/negative
    values (Redhat Bugzilla #487829).
  • fix configure check for XkbBell and provide
    appropriate parameter for it.
  • fix a caching problem with double-size fonts versus reverse
    video that could cause core dump.
  • repair double-size fonts from workaround used in patch #240.
  • add new section to the VT Fonts menu which allows enabling
    or disabling the font, termcap (tcap-query), title and window
  • add fontWarnings resource, to control whether
    to show warnings on failure to load a font.
  • improve warnings for unloadable fonts introduced in
    patch #240 by limiting those to the
    cases where a font would be specified directly by a resource
    setting rather than a derived fontname.
  • further amend fix for Debian #252873
    from patch #197 to treat a blank cell
    which does not have both foreground and background colored as a
    non-colored cell. This improves a special case where the cursor
    is on a blank cell which had foreground color scrolled in
    (report by Miroslav Lichvar).

    Also add the same logic when hiding cursor, so the outline
    matches the in-focus cursor.

  • modify internals to reduce places PAIRED_CHARS() is used,
    making WriteText() and ScrnWriteText() accept IChar array, as
    well as providing a wrapper for drawXtermText().
  • change default XIM font from “*” to “fixed” to improve
    startup time in zh_CN.UTF-8 locale (Mike Fabian, SuSE Bugzilla
  • typo in #240 log (Slava Semushin)
  • improve checks for missing bitmap fonts, fallback to
    “fixed” as needed to work around broken font-packages (report
    by Jacek Luczak).
  • fix breakage from patch #240
    changes for xtermAddInput() (patches by Jeff Chua,
    Julien Cristau).
  • use plink.sh for linking xterm (suggested by
    Larry Doolittle).
  • add resource descriptions for input method to xterm
  • update configure script; consistently append to $CFLAGS
    rather than prepend.
  • add install-scripts rule to makefile, to allow
    koi8rxterm and uxterm scripts to be altered independently of
  • add -maximized command-line option and
    corresponding resource (prompted by alt.os.linux newsgroup
  • modify translations of scrollbar widget using
    xtermAddInput() (see patch
    #181) to accept the actions that the vt100 widget accepts,
    such as shift-insert to perform a paste operation (request by
    Martin Zwickel).
  • change default for allowTcapsOps resource to
    false, since it causes unexpected behavior for vim users with
  • update config.guess, config.sub
  • correct a cast in input.c, which broke translation of
    numeric keypad codes to pageup, pagedown, etc., on 64-bit
    platform (Debian #511138,
    report by Larry Doolittle).
  • update configure macro CF_XOPEN_SOURCE for AIX 6.x and Mint
  • reset the screen wrapping-flag at the end of
    ClearRight to fix an occasional case where the
    last character of a scrolled and wrapped line would be cleared
    (patch by Joe Peterson).
  • modify to use POSIX coding for comparing resource settings
    such as locale, to work with locales such as
    Turkish (report by M Vefa Bicakci).
  • turn on configure paste64 feature by default
    (request by Jean-Philippe Bernardy). It is runtime
    enabled/disabled with allowWindowOps.
  • turn on configure tcap-query feature by
    default, add resource allowTcapOps to make this
    runtime enabled/disabled.
  • make OSC 3 (change X property, from patch #110) subject to
    allowWindowOps resource.
  • make VT220 DSR responses inactive in
  • make DECUDK feature inactive in
  • respond to incorrectly formatted DECRQSS with
    a cancel.
  • add allowFontOps resource to allow the
    fontsize-switching and font query/set control sequences to be
    enabled/disabled (prompted by Debian #510030).
  • some code cleanup based on gcc 4.x
    -Wconversion warnings in button.c and
  • modify tcap-query feature to not return data
    for shifted cursor-keys when the keyboard type is set to vt220,
    since returning the same string for shifted/unshifted keys may
    confuse some applications (Gentoo #212546).
  • improve usability of TrueType fonts by making the font-size
    switching for shifted keypad plus/minus use the
    faceSize resources to determine the order of fonts
    (when TrueType fonts are used) rather than the bitmap fonts,
    since their sizes may not be in the same order (report by H
    Merijn Brand)
  • remove an optimization of ConfigureNotify
    events from patch #236 which seems to
    interfere with passing SIGWINCH to applications
    (Gentoo #233836).
  • modify handling of altSendsEscape to reset the
    eightBitInput mode, like
    metaSendsEscape (patch by Ted Phelps).
  • add feature to show the text-cursor as an underline rather
    than a box, plus command-line options -uc and
    +uc and resource cursorUnderLine to
    control the feature (patch by Paul Lampert).
  • update config.guess, config.sub
  • correct memory reallocation when handling a paste of UTF-8
    text from patch #225 changes
    (report/patch by Max Mikhanosha).
  • correct allocation of temporary buffer in
    xtermFindShell in case the user’s
    $PATH contains no “:” (report/analysis by Victor
    Stinner, Freedesktop.Org Bugzilla #16790).
  • modify CF_XOPEN_SOURCE to add case for DragonFly BSD, to
    fix new compile problem exposed by fix for fd_mask (patch by
    Hasso Tepper).
  • add configure-check for ncurses
    use_extended_names, (report by Martin
  • correct computation for toolbar height; layout manager
    already takes into account borderWidth
  • implement VT320-style SCS (select character set) for ISO
    Latin-1 supplemental.
  • fixes for vt100-style character sets in UTF-8 mode (Ubuntu
  • fix to make luit work with xterm’s
    -ls option (report/patch by Marius Tolzmann).
  • update config.guess, config.sub
  • add control sequences for some of the recent resource/menu

    • altSendsEscape (private mode 1039)
    • keepSelection (private mode 1040)
    • selectToClipboard (private mode 1041)
    • bellIsUrgent (private mode 1042)
    • popOnBell (private mode 1043)
  • add resource formatOtherKeys to provide an
    alternate escape sequence format for the
    modifyOtherKeys resource (request by Paul LeoNerd
  • adjust saved-cursor position if the window is resized while
    displaying the alternate screen (Novell #196880,
    Debian #383384).
  • improve pointer-checks to fix a bug exposed by resizing
    during initialization under StumpWM window manager (Fedora
    Bugzilla #437928).
  • modify unselectwindow() to ensure that the
    mouse pointer is not hidden after xterm loses focus (report by
    Jeremy Huddleston).
  • add special check for fd_mask on Mac OS X (report by Jeremy
  • add dylib to autoconf’s suffix list used for
    checking the result from xmkmf, to work with Mac
    OS X (report by Jeremy Huddleston).
  • correct initialization of bold- and wide-, wide-bold fonts
    which may be set via the utf8Fonts subresource
    (Debian #347790).
  • modify sinstall.sh to use POSIX locale to
    bypass GNU ls changes to date-format.
  • improved/refined changes for closing bitmap font (patch by
    Andrea Odetti).
  • improve resize computation for situations where the
    negotiation fails, by invoking the xterm widget’s core-class
    resize method (Debian #365602,
    patch by Jim Paris).
  • restore initialization of terminal’s VMIN and
    VTIME settings, from patch
    #232 changes (patch by Matthieu Herrb).
  • add configure check for ttydefaults.h, include
    if needed, e.g., for systems where defining
    _POSIX_C_SOURCE, etc., prevents it from being
    included via termios.h (prompted by comments by
    Robert Delius Royar and Jeremy Huddleston).
  • bug-fixes for bugs.opensolaris.org (patches from Alan

    fix a typo in manpage
    xterm doesn’t properly set ut_syslen
    left-left-right misinterpreted as triple click
  • minor optimization to tab-initialization (patch by Németh
  • fix a case where an incorrect font was freed during
    initialization from patch #232 changes
    (patch by Andrea Odetti).
  • improve comparison used in SameFont function
    for GC-caching (Julien Cristau).
  • correct macro name used for default CKILL
    definition to work with Mac OS X (report by Jeremy
  • corrected logic in a font-cache used for reverse-video
    (Debian #404079).
  • add control sequence to alter pointerMode at
  • add limit-checks for rectangle operation parameters (report
    by Martin Pirker).
  • modify minstall.sh to suppress
    $MANPAGER and $PAGER environment
    variables, which may interfere with redirecting output of
    man to a shell variable (report/patch by Zdenek
  • do not try to hide mouse pointer in the tek4014 window,
    fixes broken “-t” option at startup from patch #230 changes (report by Robert K.
  • correct datatype used when drawing tek4014 data using xterm
    compiled for wide-characters, on big-endian machines (reports
    by Jeremy Huddleston, Harald Hanche-Olsen, Martin Costabel,
    Merle Reinhart).
  • modify to cache the font-names along with the bitmap font
    data, to improve comparison of fonts.
  • modify to allow building with configure options
    --disable-ansi-color and
    --disable-leaks (Debian #459817,
    report/patch by Németh Márton).
  • modify to allow building with configure options
    --enable-wide-chars and
    --disable-c1-print (Debian #459816,
    report/patch by Németh Márton).
  • add pointerMode resource to control whether
    and when the pointer cursor is hidden as the user types.
  • simplify initialization of ttyMode- and related characters
    using a table.
  • modify initialization-logic for stty values that correspond
    to ltchars structure and the BSD TIOCSLTC ioctl (susp, dsusp,
    rprnt flush, werase, lnext). These were reset to constants for
    both termios and legacy interfaces immediately after asking the
    system for the existing values since X11R6.1, rather than using
    them to provide inherited values. While the legacy interface
    has some constraints, e.g., on HPUX, the POSIX or termios
    interface should not. Your shell may reset these anyway
    (prompted by patch by Ed Schouten).
  • improve logic for hiding/displaying pointer-cursor (report
    by Mark Brukhartz).
  • add limit-checks to tabs.c, increase maximum column for
    setting tab-stops from 320 to 1024 (report by Németh
  • correct length, i.e., number of types of selection targets,
    computed by ConvertSelection() when not handling
    wide characters for the XA_TARGETS() case. This
    leaves an extra Atom on the end of the list which is not
    handled by Java
    applications. The problem was introduced in patch #151 (report by David Wood).
  • undo change to getXtermCell() from patch #230
    using PACK_PAIR macro (Debian #459014,
    analysis by Caetano Jimenez Carezzato).
  • minor documentation fixes (patch by Slava Semushin)
  • add makefile actions to install KOI8RXTerm app-defaults
    file (patch by Julien Cristau).
  • add quietGrab resource, which when true,
    suppresses cursor repainting when NotifyGrab and
    NotifyUngrab event types are received during
    change of focus (request by Nicolas George).
  • do not treat Unicode BIDI control characters as combining
    characters (Debian #457634).
  • add koi8rxterm, from Debian.
  • add manpage for uxterm, from Debian (Ubuntu
    Debian #438645)
  • remove “.xpm” suffixes from Icon filenames in desktop files
    since it confuses some lookups following the
    Icon Theme Specification (report by Slava Semushin)
  • correct width-calculation used for adjusting proportional
    fonts, to work with wide-characters (Debian #441354).
  • fixes/improvements for double-size characters:
    • correct old clipping calculation which used total
      height of glyphs where ascent was needed.
    • if bold font is unavailable, fall back to normal
    • adjust to “work” with Xft (which does not support
      double-width single-height characters).
    • restore reset of doublesize for a line when it is
      cleared, broken in patch
  • modify logic for forceBoxChars resource when
    using TrueType fonts to be consistent with bitmap fonts
  • modify logic for forceBoxChars resource to
    make the “Line-Drawing Characters” menu entry use xterm’s
    line-drawing characters even asked to draw wide line-drawing
    characters which are available in the font.
  • modify rectangle-support functions to preserve colors when
    filling/erasing to match WRQ Reflection behavior (request by
    Enzo Toscano).
  • add getopt-parsing to tcapquery.pl, including feature to
    test the extended cursor/editing keys.
  • make missing double-width glyphs display as double-width
    (Debian #456236).
  • change tcap-fkeys and rectangles
    configure options to enable them by default.
  • hide the mouse pointer while user is typing (request by
    Rodolfo Borges).
  • extend configure options --enable-tcap-query
    and --enable-tcap-fkeys to send cursor- and
    editing-keypad keys modified according to the keyboard (or
    termcap) selection for shift, alt, control, meta.
  • modify kdch1 in termcap, e.g.,
    xterm-r6 to match the terminfo file.
  • add -hm option to turn
    highlightColorMode on or off.
  • add highlightColorMode resource to separate
    the new (since patch #225)
    highlighting with both text- and background-colors (prompted by
    report/example by Thomas Wolff).
  • add Keep Selection menu entry to turn the
    keepSelection resource on/off at runtime.
  • add keepSelection resource, which when
    enabled, tells xterm to retain the X selection even after it
    stops highlighting it (patch by Sergey Vlasov).
  • extend the CSI > n sequence to
    allow disabling all types of modified-keys that the
    CSI > m sequence affects.
  • move include for <xtermcap.h> in
    resize.c to avoid redefinition of
    termios structure on OpenSolaris (report by Rahul
    Gopinathan Nair).
  • extend terminfo building blocks for modified editing keys
    to include all six keys.
  • synchronize terminfo with ncurses (report by Stephane

    • equate xterm-xfree86 and
    • add ncurses extensions OTbs, AX, for termcap
    • make old/legacy entries such as xterm-24,
      xterm-65 and aliases xterms,
      vs100 inherit from
    • make xterm-r5 and xterm-r6
      the same, ignoring historical errors in X Consortium’s
  • fix an ifdef in logic for selecting regular expressions
    while in a narrow-character locale (Debian #449227).
  • override locale in minstall.sh; change in patch #226 does not work in UTF-8 locale
    (report by Zdenek Sekera).
  • undo an incorrect fix for a memory leak in patch #209 (Debian #435858).
  • modify configure script to permit combining
    --with-utempter and --enable-setuid,
    e.g., for using xterm with the utempter library on FreeBSD
    (report by Andriy Gapon).
  • modify “Quit” menu entry to override the -hold
    command-line option.
  • add a check in the startup error-reporting to avoid writing
    to pipe when it has not been opened. In that case, report
    errors directly to the standard error.
  • add OPT_READLINE definition to xtermcfg.hin, overlooked in
    patch #205 (report by Kalle Olavi
  • modify 88colors2.pl and
    256colors2.pl, adding -r option to
    reverse the palettes for the extended colors.
  • check for partial overwrite or deletion of multi-column
    characters in several cases, e.g., insert-character,
    delete-character, etc., and fill the remainder of the cells
    used by the multi-column characters affected with blanks.
  • correct character-class codes in wide-character mode for
    characters 215, 247 (see also patch
  • fix missing assignment for UTF-8 parsing in widget
    initialization (Debian #403360).
  • correct index expression used to set line-wrapping flag,
    making selection from scrollback work consistently (Debian
    report by Vincent Lefèvre).
  • amend changes to handshake in patch
    #226 to accommodate Solaris, which relies on the extra
    setting of the terminal size after I/O initialization. Do this
    by adding new resource ptySttySize, which is false
    for Linux and MacOS X, i.e., true for for Solaris and other
    SVR4 platforms, as well as FreeBSD (reports by David Wood,
    Renato Botelho).
  • check for X events after cursor-left, and carriage return,
    consistent with indexing operations (comments by Vegard Nossum
    and Ingo Molnar on a mailing-list).
  • initialize the .keyboard structure, needed for
    some platforms (such as Solaris) after patch #227 fixes for keysyms (patch by David
  • update config.guess, config.sub
  • exclude the Intel compiler from the extra gcc warning
    options added in patch #226.
  • modify change for Debian #422521
    from patch #226 to work with
    configurations where the X server does not recognize the
    XF86Paste or SunPaste keysyms (report
    by Paolo Pumilia).
  • fix incorrect free in fix for “Selection” menu entry in
    patch #226.
  • add configure check to use
    -Wno-unknown-pragmas for Solaris, and a few other
    new gcc warning options to address regressions in its warning
  • add sample desktop and icon files, along with configure
    options for manipulating them.
  • fix an infinite loop when showing a 2-column character in a
    1-column screen (Debian #426863).
  • add XF86Paste and SunPaste to the
    default translations (Debian #422521,
    patch by Bernhard R Link).
  • update wcwidth.c (partly based on/prompted by mailing list
  • update/improve keysym2ucs.c based on Unicode 5.0.1d3 and
    Xorg 7.1 keysymdef.h file.
  • improve gen-pc-fkeys.pl, making it show all of
    the building-block terminfo entries used by xterm.
  • correct strings used for modifiers applied to F1-F4 in
    xterm+pcf1 terminfo entry.
  • improve session management by adding the ICE connection
    number to the select call rather than waking up
    once per second (patch by Miroslav Lichvar).
  • add environment variable $XTERM_LOCALE to help
    with shell initialization, e.g., to set a specific locale for
    xterm on systems where a global locale is set in the shell
    startup scripts.
  • add resource settings mkSampleSize and
    mkSamplePass to modify the mkWidth
    added in patch #201. In patch #202, xterm would also check if the
    system’s wcwidth matched its built-in tables “well
    enough” to be used when mkWidth was false, and if
    not would use the built-in tables anyway. These resources allow
    the user to customize the decision (prompted by comments by
    Emanuele Giaquinta)
  • modify logic which resets/updates the screensize on the
    child process side of the pseudo-terminal to do this only if a
    successful handshake was received, e.g., as determined by the
    waitForMap resource (prompted by reports by
    Emanuele Giaquinta and Bernhard R Link, but see also patch #177 and patch
  • improve permissions logic when closing pseudo-terminal
    (Debian #12261,
    patch by Nathanael Nerode, analysis by Richard Braakman).
  • add resource highlightReverse which controls
    whether xterm will allow selection highlighting to hide
    reverse-video or use the older behavior which inverts the
    foreground and background colors when selected text with
    reverse-video attribute (report by Adam M Costello, comments by
    Victor Vaile).
  • restore a special case for cursor-color in
    ReverseVideo() lost in rewrite for patch #224 (report by Adam Sulmicki).
  • correct initialization for menu entry for
    bellIsUrgent (patch by Emanuele Giaquinta).
  • correct length of underlining for double-width Xft fonts
    (report by Shidai Liu “Leo”).
  • correct clipping for double-width Xft fonts (report by
    Shidai Liu “Leo”).
  • modify initialization for italicULMode to
    avoid XftPatternBuild reusing bold attributes
    (report by Shidai Liu “Leo”).
  • add a check in case someone tries to call the
    popup-menu() action on a menu which is not
    initialized (Debian #426364).
  • improve CF_IMAKE_CFLAGS to work with Solaris sed (report by
    Peter Bray).
  • improve guess for X’s manpage section in
    minstall.sh, (prompted by comment by Miroslav
  • modify minstall.sh to handle manpage patch
    with more than one dot (patch by Miroslav Lichvar).
  • fix configure message for –disable-setgid option (patch by
    Miroslav Lichvar).
  • add allowTitleOps resource to allow users to
    prevent the title- and icon-names from changing (request by
    John Bashinski).
  • fix “spawn-new-terminal” action, for the case where no
    parameter is passed to the action (patch by Daniel
  • fix error-checking on internal font switching for
    “Selection” menu entry (Debian #421523).
  • amend select/paste change from patch
    #225 by limiting it to non-UTF-8/non-KOI8-R encoding
    (Debian #420974).
  • add workaround for groff “.URL” codes which are not present
    in some commonly-used bitmap fonts (Debian #418324).
  • add useClipping resource to allow clipping to
    be disabled.
  • use XftDrawSetClipRectangles to work around
    Xft pixel-trash (report by Reuben Thomas).
  • add configure option --enable-tcap-fkeys, and
    resource tcapFunctionKeys, which can be used to
    tell xterm to use function-key definitions from the termcap (or
    terminfo) which it uses to set $TERM on startup.
  • add resources altIsNotMeta and
    altSendsEscape to allow one to use Alt-keys like
    the meta-key even if they are bound to different keycodes
    (prompted by discussion with Daniel Jacobowitz).
  • revert a change from patch #216
    that unnecessarily made the meta modifier override the
    eightBitInput resource if the alt- and
    meta-modifiers happened to overlap (report/patch by Daniel
  • correct associated font for active icon for colored text
    (broken in patch #224).
  • correct ifdef’s for Darwin (patch by Emanuele
  • add highlightTextColor resource, and options
    -selfg, -selbg like xwsh
    (adapted from patch by Victor Vaile).
  • revise find_closest_color() function to
    address concern about borrowing from Tcl/Tk
    (request by Dan McNichol).
  • add “spawn-new-terminal” action, which can be assigned to
    key translation, allowing one to spawn a new copy of xterm
    using the current process’s working directory (adapted from
    patch by Daniel Colascione).
  • improve select/paste between UTF-8 and Latin1 xterms by
    adapting the translations from patch
    #185. Extend that to include Unicode fullwidth forms
    FF00-FF5E. Also modify select/paste of DEC line-drawing
    characters in Latin1 mode to use ASCII characters.
  • add “Enable Bell Urgency” to VT Options menu, removed
    “Enable Margin Bell”.
  • add bellIsUrgent resource to control whether
    the Urgency hint is set/reset.
  • modify to set Urgency window manager hint on bell, reset it
    on Focus-In event (patch by Emanuele Giaquinta).
  • add --disable-setgid configure option (request
    by Miroslav Lichvar).
  • fix a possible infinite loop in last change to
    dabbrev-expand() (patch by Emanuele
  • modify initialization to set the pty erase value if the
    erase is set in the ttyModes
    resource. This overrides the ptyInitialErase
    setting (request by Lluís Batlle i Rossell).
  • add initialFont resource to xterm widget, like
    tek-widget (Debian #299669).
  • amend change to boldMode from patch #223 for Debian #347790.
    As noted in Debian #412599,
    that made xterm no longer match the documented behavior. Add
    new resource alwaysBoldMode to allow overriding
    the comparison between normal/bold fonts when deciding whether
    to use overstriking to simulate bold fonts.
  • restore background color in ClearCurBackground(), omitted
    in changes for patch #223 (report by
    Miroslav Lichvar).
  • correct logic for repainting double-width TrueType
    characters (prompted by test-case for Novell #246573).
  • add a check to avoid trying to repeat a multibyte character
    (report by Sami Farin).
  • modify parameter to XftNameParse() to select
    wide face-name as needed, to make -fd option work
    (patch by Mike Fabian, Novell #246573).
  • correct logic for mouse highlight tracking’s abort
    sequence, broken in a restructuring modification from patch #224 (report by Thomas Wolff).
  • revert the simplification of blinking cursor, since that
    broke the xor’ing introduced in patch
    #193 (report by Thomas Wolff).
  • simplify code for set/reset mode for blinking cursor (patch
    by Emanuele Giaquinta).
  • modify dabbrev-expand() to restart after the
    last match (patch by Emanuele Giaquinta).
  • add control sequences for enabling/disabling focus in/out
    event reporting (request by Bram Moolenaar).
  • improve startup performance of menus by adding resource
    setting menuLocale which can be set to override
    X’s lengthy initialization of fontsets – which are seldom used
    for the Xaw popup menus (adapted from patch by
    Dave Coffin).
  • modify do_precomposition() function and
    make-precompose.sh to handle 21-bit codes vs the
    16-bit codes those were written for, and fix a few mis-sorted
    codes (patch by Thomas Wolff).
  • handle special case in -cjk_width which
    unexpectedly caused a character’s width to change when a
    combining character (patch by Thomas Wolff)
  • fix build for GNU/KFreeBSD (Debian #40111).
  • consolidate GC creation/updating into a single module to
    reduce GC manipulation (prompted by Debian #389476,
    though a complete fix would involve optimizing the scrolling
  • fix ifdef of xtermCellWidth(), which broke for a case
    without TrueType and without wide-character support (report by
    Martin Pirker).
  • undo a comparison in
    handle_translated_exposure which did not work due
    to X server optimization. That caused the inner border to be
    repainted with unexpected colors when handling a repaint, e.g.,
    after switching to/from another workspace (Debian #401726,
    Redhat Bugzilla #223027).
  • allow -cr option to override cursor color when
    -ah option is used (Debian #406502).
  • add a note in xterm’s manpage explaining that the
    -bw (or -w) option is only used by
    the window manager, if at all (Debian #405043).
  • make Selection of VT Fonts
    work with selectToClipboard resource.
  • correct length calculation for Selection entry
    of VT Fonts menu, broken since it ignored the
    actual selection length since X11R4.
  • fixes for fontsize changes with -fa option
    (Redhat Bugzilla #222340).
  • add --enable-rectangles configure option
    (request by Martin Pirker).
  • correct default value for --with-symlink
    configure option.
  • fixes configure script macros that use
  • modify configure script to provide support for pre-package
    config versions of Xft aka “FreeType”.
  • add the Xaw scrollbar translations resource to the xterm
    manpage, add an example showing how to change the mouse button
    assignments (Debian #382225).
  • amend a change from patch #216,
    which omitted modifiers for control, meta, etc., if they were
    mixed with any other modifiers. The intent of the change was to
    avoid confusion with XK_Mode_switch and
    XK_ISO_Level3_Shift; the check is now done
    explicitly (report by Daniel Jacobowitz).
  • interpret a negative value for
    modifyCursorKeys or
    modifyFunctionKeys resources to disable the
    respective features (prompted by Novell #220728).
  • amend cell-width computation for FreeType from patch #217, which did not work for VT100
    line-drawing characters (Debian #399638,
    Gentoo #147111).
  • amend a change from patch #216,
    which made alt-modifier on a cursor-key send a modifier
    parameter (Novell #220728).
  • correct an off-by-one that made DECCRA not work
    (report/patch by Martin Pirker).
  • revert an optimization in SGR_Foreground() and
    SGR_Background from patch #209 (Debian
    analysis by Pierre Lombard).
  • fix for boldMode (Debian #347790,
    patch by Tim Pope).
  • amend fix for -iconic in patch #208, which broke the positioning part
    of -geom with toolbar configuration.
  • fix to prevent indexing error in regular expressions (patch
    by Dennis Schneider).
  • fixes to make the internalBorder area not
    change color due to reverseVideo and/or related exposure
    events. The latter was a very old bug exposed in patch #196 (report by Neil Hoggarth, also
    Debian #397624).
  • minor optimization for recoloring cursor via dynamic
  • fix a bug caused by restructuring of tek4014 widget (report
    by Paul Schenkeveld)
  • fix for regular expressions: the code which converted the
    column offset within a line did not check properly for the end
    of a line, and if allowed to match the whole line, would select
    the beginning of the following line. If the following line were
    empty, in turn it would select from the next, etc. (report by
    Sean Reifschneider).
  • minor optimization of color allocation to avoid repainting
    the screen if the corresponding color had not been
  • fixes for vttests/256colors.pl (report by Egmont
  • add terminfo building block entries for modifiers of the
    6-key editing keypad.
  • fix for initialization of tek4014 which broke on QNX
  • make “xterm -t -iconic” work as expected, i.e., start in
    tek4014 mode, but iconified.
  • some internal restructuring to separate data for vt100 and
    tek4014 widgets.
  • fix a few cases where form-events would be seen by the
    vt100 widget when built with the toolbar configuration.
  • change behavior when encountering an illegal character in a
    title string. Rather than reject the string, translate illegal
    characters into “?” and use the string (requested by Thomas
  • improve checks for nonprinting characters in title strings
    (report by Samuel Thibault).
  • correct typo in menu labels, changing
    print-redirect to print-redir, making
    it follow the manpage and match the usage for the actions table
    (report by Samuel Thibault).
  • correct a typo that prevents building with some older
    systems such as Solaris 2.6 (report by Julian Bridle).
  • minor improvements to FreeType font layout and
  • add a check in the ptyInitialErase logic to ensure that the
    termcap was read (Redhat Bugzilla #201246).
  • limit changes for reverse-video from patch #216 to cases where the reverse-video
    command-line option is used (report by Zdenek Sekera).
  • correct bitmap-derived pointsizes for TrueType fonts; they
    should be proportional to the square root of the area of the
    bitmap fonts.
  • add resources to specify pointsize of TrueType fonts
    (request by Reuben Thomas).
  • improve install of terminfo by filtering out harmless
    messages related to extended capabilities. At the same time,
    use ncurses tic to compile the extended capabilities if
    possible (report by Zdenek Sekera).
  • update “xterm+pcfkeys” terminfo entry to correspond to
    patch #216.
  • improve handling of ConfigureNotify events by
    checking if there are further events in the queue which
    obsolete the current one (adapted from rxvt 2.7.5). Only the
    normal normal (non-toolbar) configuration is addressed in this
  • several changes to terminfo:
    • incorporate some minor changes from ncurses to help
      keep these synchronized:

      modify sgr/sgr0 in xterm-new to improve tgetent’s
      derived “me”.
      remove ncv flag from xterm-16color
      improve xterm-256color by combining the 16-color
      setaf/setab strings with SGR 48, and cancelling the
      setf/setb strings.
    • use extended function-key definitions for xterm-sun,
      xterm-sco entries.
    • add terminfo building-blocks corresponding to the
      modifyFunctionKeys and
      modifyCursorKeys resources.
  • ifdef’d Sun function-key feature to make it optional, like
    HP and SCO.
  • extend table for termcap-query feature through F63, and
    updated tcapquery.pl to match.
  • modify logic for function-key input processing to allow
    function key numbers which are constructed by control- and
    shift-modifiers to extend beyond X’s hardcoded limit of
  • add control sequence to set or reset the
    eightBitInput resource.
  • change default resource modifyFunctionKeys to
    2 to avoid sending SS3 with parameters (report by Kalle Olavi
  • add control sequences for setting and resetting the values
    of the modifyCursorKeys,
    modifyFunctionKeys and
    modifyOtherKeys resources.
  • add modifyFunctionKeys resource like
    modifyCursorKeys. Setting this to zero allows one
    to use the control- and shift-modifiers to construct function
    key strings, for terminals using many function keys, e.g., for
    xterm-sun or xterm-sco.
  • modify screen responses and function-key logic to reduce
    the number of writes made, to make it less likely that an
    application would read only part of a function-key in a read
    operation (suggested by John E Urbanczyk).
  • add combiningChars resource, which allows the
    user to specify the maximum number of combining characters that
    xterm will store for each cell in wide-character mode (prompted
    by request by Markus Kuhn to increase the limit from 2).
  • improve logic in metaSendsEscape in case the
    Alt- and Meta-keys are mapped to different modifiers. That
    allows one to use the Alt-key for shifting in the
    eightBitInput and use a Meta-key modifier to
    prefix the result with an <ESC>.
  • improve modifyCursorKeys logic to prevent it
    from modifying codes where eightBitInput or
    metaSendsEscape are set (request by Dan
  • improve modifyCursorKeys logic to prevent it
    from changing the user input when other modifiers such as AltGr
    are used (report by Thomas Wolff).
  • extend modifyCursorKeys to include the numeric
    keypad when in application mode.
  • improve -reverse (-rv) option
    (patch by Jason Vas Dias, Redhat #189161).
  • workaround for color resources on Fedora-5 which made the
    toolbar colors inconsistent.
  • corrected calls used for --disable-setuid
    option, add debugging traces to help diagnose this area.
  • modify configure script to ensure that
    USE_UTMP_SETGID is defined only if the check for
    POSIX saved-ids succeeds, or corresponds to one of the BSD
    systems known to have a workable setegid
  • extend configure check for POSIX saved-ids to include BSD
  • modify CF_SYSV configure macro to work with gcc on HPUX
    10.20, whose broken <term.h> relies on
    including <termios.h>.
  • change order of setuid/setgid ifdef checks in Imakefile to
    make it simpler to produce a setgid install, allowing the
    latter to override the former (prompted by an XFree86
  • adapt a fix for setgid support from subsequent XFree86
    changes (patch by Emanuele Giaquinta).
  • further improve setgid support (patches by Emanuele
  • improve setgid support by exploiting the saved-ids feature
    on which it relies (patch by Emanuele Giaquinta).
  • make the modifyOtherKeys resource disabled by
    default since the intermediate setting altered some common
    bindings (report by Emanuele Giaquinta).
  • modify makefile rule for ctlseqs.txt to strip
    backspace/overstrikes, and add a copy of ctlseqs.txt to the
    source tarballs (requests by Bram Moolenaar, Emanuele
  • add modifyOtherKeys resource, analogous to the
    modifyCursorKeys resource. This applies to keys
    that normally would transmit nothing when a given modifier is
    applied (request by Dan Nicolaescu).
  • add default to the acceptable values for the
    keyboardType resource, allowing the
    -kt command-line option to override app-defaults
    resource settings of sunKeyboard, etc.
  • correct ifdef’s for XkbBell() which used the
    header file but not the corresponding function in patch #175 (report by Zach Beane).
  • add xterm manpage to the --with-symlink
  • regenerated configure script to omit some debugging
    artifacts of the SIGWINCH test.
  • fix install-man rule in
    Makefile.in to avoid including the
    $(DESTDIR) value in substitutions made on the
    manpages (patch by Emanuele Giaquinta).
  • improved fix for cursor deallocation (Redhat #186935,
    patch by Jason Vas Dias).
  • improve checks for setuid/setgid operation to accommodate
    limited resource management in some kernels (Gentoo #193238).
  • update config.guess, config.sub
  • minor optimization to cell layout; will use that to
    simplify some logic and allow for more than two combining
    characters in another patch.
  • add utf8Latin1 resource to make optional the
    feature from patch #209 which allowed
    ISO-8859-1 fonts to be used in cases where a wide font was
    given. This would only work for the special case where the user
    normally used Latin-1 and wanted some wide characters (report
    by Rostislav Krasny).
  • add utf8Title resource to manpage (Novell
  • fix typo in table entry for utf8Title resource
    which made it treated as an integer rather than boolean (patch
    by Mike Fabian, Novell #52655).
  • improve description of forceBoxChars in
    manpage (discussion with Joe Wells).
  • set checkmark for “Select To Clipboard” menu entry on
    startup, from resource setting.
  • update manpage to note that
    *customization:color is not needed, though it is
    useful (Redhat #188034).
  • add underscore to sample pattern for URL in
    XTerm.ad (Redhat #188037).
  • modify install rule for manpage to use the configured
    app-defaults directory (Redhat #188031).
  • add checks in releaseCursorGCs() to ensure
    GC’s are distinct, needed since memory leak changes for
    patch #208 (Redhat #186935,
    patch by Jason Vas Dias).
  • modify Imakefile definitions to allow
    imake to set InstallXtermSetGID. Add
    symbol InstGidFlags.
  • add configure script check to ensure that SIGWINCH if
    defined even when headers undefine this as a side-effect of
    _POSIX_C_SOURCE, etc.
  • resync with XFree86 CVS
    • ifdef-out chmod of terminal device for OS/2 (XFree86
      Frank Giessler).
    • move new Imakefile chunk which defines
      $(CSGIDFLAGS) before DEFINES, e.g., for IRIX64
      (Marc La France).
  • modify prefix/suffix transformation in makefile to make
    uxterm install properly on Cygwin.
  • restore the sizeof-comparison for struct
    added in patch #208,
    rendered (mostly) obsolete in patch
    #210, since an imake configuration can still turn that code
    on (prompted by comment by David Dawes).
  • remove feature for FreeBSDArchitecture from
    Imakefile which would install xterm using the
    utmp group since a standard FreeBSD system has no
    such group, though it would support this configuration. FreeBSD
    packager uses the configure script, which can make
    the appropriate check (prompted by comment by David
  • fixes for gcc redefinition warnings when built using imake
    (reports by David Dawes, Marc La France).
  • add configure check for workaround in patch #208 for Solaris’ inconsistent support
    for struct lastlog to quiet compiler warnings (prompted by
    David Dawes commit message for XFree86).
  • fixes for BASE64 selection manipulation, e.g., ensure that
    the string terminator is written after cut-buffer data (report
    by Joe Allen).
  • add vttests/paste64.pl script to test experimental option
    for setting/getting selection data.
  • add some error-checking in fontutils.c (prompted by David
    Dawes commit message for XFree86).
  • change xterm manpage to show the actual color resource
    names XtDefaultForeground and
    XtDefaultBackground rather than black and white
    (prompted by discussion with Emanuele Giaquinta).
  • add utf8Title resource and menu entry,
    allowing the user to control whether title strings are
    interpreted as ISO-8859-1 or UTF-8 encoding (Novell #52655,
    Novell #113206).
  • change order of ifdef’s for utempter versus platforms in
    Imakefile to ensure a proper value for
    InstallXtermSetGID (prompted by Marc La France
    commit message for XFree86).
  • update the cursor GCs when menu entry for reverse video is
    toggled. (Redhat #183993).
  • a memory-leak fix for set_cursor_gcs() in
    patch #208 broke the -cm
    option (Redhat #182382).
  • remove setf and setb from
    xterm-256color terminfo entry to avoid issues with
    applications that ignore the ANSI strings (report by Emanuele
  • remove obsolete references to “Xorg” to avoid confusion
    with “X.org” (prompted by David Dawes commit message for
  • improve initial layout performance for the toolbar
    configuration by precalculating the height of the toolbar.
  • remove menu entry for “Enable Curses Emulation” (to make
    room for “Select To Clipboard”).
  • add resource selectToClipboard, action
    set-select and a menu entry to allow users to
    switch between PRIMARY and CLIPBOARD for select/paste.
  • allow cursor to have the same color as foreground (text),
    since it is rendered as reverse (Debian #350664).
  • amend change for loading utf8Fonts resource
    from patch #204 to allow an ISO-8859-1
    “normal” font to be combined with an ISO-10646 font if the
    latter is given via the -fw option or its
    corresponding resource value (prompted by comment in Novell
  • add TIOCSCTTY ioctl() ifdef’d for
    __GNU__ to set the controlling terminal (Debian
  • add configure option --disable-leaks.
  • add a check in Bell to ensure that the VT100
    widget is realized, since it may be called by
    xtermLoadFont given an incorrect font resource
    (Redhat #180450).
  • improve initialization of wide-bold font by not using the
    derived value when it happens to have few glyphs (report by
    Mboso Sampson).
  • improve dynamic colors by not repainting the screen if only
    the cursor color changes (request by Bram Moolenaar).
  • fix menu initialization for tek4014 window broken in
    patch #206.
  • typo in manpage (Debian #351425).
  • correct loop logic in dotext to prevent an
    infinite loop if a wide character was just at the right margin
    when wrapping was disabled (report by Serge van den Boom).
  • initialize the saved-cursor data so a restore-cursor
    operation without a preceding save-cursor operation will not
    modify the foreground color.
  • add a new selection feature: regular expressions, and new
    resources which specify what happens on multiple mouse clicks:
    on2Clicks, on3Clicks,
    on4Clicks and on5Clicks.
  • revert the XTerm.ad change from patch #208 (Debian #348384).
  • fixes for the print-window to make it handle combining
    characters in UTF-8 mode.
  • add check to ensure that double-clicking to extend
    selection will not extend it into the scrollback area. This
    still allows users to scroll back and select text (Debian
  • add configure option --with-tty-group to help
    work around deficiencies in packager’s build environment
    (Debian #349142).
  • fix typo in configure script option
    --disable-narrowproto (report by David Martínez
  • amend CF_X_FREETYPE changes for pkg-config
    check to ensure that Xft is known to that tool
    (report and patch by Emanuele Giaquinta).
  • add an ifdef to allow ./configure
    to work, broken by new code for
    testing memory leaks in patch #208
    (report by Brian Lindholm).
  • add charClass resource to
    XTerm.ad and UXTerm.ad which
    simplifies selecting URLs (prompted by Kirill Ponomarew).
  • modify install rule for uxterm to account for
    the --program-prefix, etc., options by invoking
    the transformed name of xterm rather than simply “xterm”.
  • modify configure --with-app-defaults option to
    allow --without-app-defaults.
  • add configure --with-symlink option to make it
    simpler to install successive versions of xterm
    renamed using the --program-prefix, etc., options
    with a symbolic link pointing to the most recent.
  • make -iconic option work with toolbar
    configuration (Gentoo #113604).
  • suppress lastlog feature for 64-bit Solaris
    configuration since that platform provides only a 32-bit
    interface (report by Peter Bray).
  • fix memory reallocation for the “UTF-8” menu entry when
    xterm is started using the +u8 option (reported by
    Jan Willem Stumpel).
  • add a configure check for the actual path of
    luit, to work around broken imake
  • add a configure check for pkg-config (Redhat
    Bugzilla #173541).
  • modify install-rule for uxterm to omit
    executable suffix, e.g., for Cygwin.
  • modify sinstall.sh to work around quirks in
    Cygwin configuration.
  • enable lastlogx support for NetBSD (was added,
    but not enabled in patch #186).
  • work around broken lastlog.h in glibc 2.3.5,
    which includes utmp.h.
  • revert part of recent XFree86 Imakefile change, restoring
    the -I. needed for xmkmf builds of
    xterm (XFree86 Bugzilla #1633, reports by Alexander Pohoyda,
    Matthieu Herrb).
  • change compiled-in default for printerCommand
    resource to an empty string. People who want to use the printer
    should be able to read the manual (Debian #311490).
  • modify Imakefile to work around old problems
    in imake configuration to allow test-builds using
    xmkmf on Linux. (This was not noticed since
    several releases had broken definitions relating to Xft which
    were harder to work around).
  • link resize for SCO platforms (Kean
  • add configure --with-app-defaults option to
    allow app-defaults directory for install-rules to be
  • remove default translations for
    dabbrev-expand() due to conflicts with existing
    keyboard arrangements.
  • remove redundant check for _NET_WM_PID (report
    by Emanuele Giaquinta).
  • set icon border width explicitly to work around fvwm
    problem with active icon resizing (report by Steve Morris,
    analysis by Dominik Vogt).
  • modify resource files to make the font-resources a little
    more specific, e.g., changing “*VT100*” to “*VT100.”, to make
    the distinction between VT100.font and VT100.utf8Fonts.font
    sharper, in case a packager modifies one of those.
  • expanded comments in UXTerm.ad regarding the
    font resources (Debian #319179).
  • add --enable-narrowproto configure option to
    accommodate X.org “modular” build (report by Stephan Hermann,
    Gentoo #17220).
  • fix typo in xterm.man description of +wf
    (patch by Tobias Stoeckmann).
  • add scrollBarBorder resource (request by Floyd
    L Davidson).
  • modify xterm-new terminfo entry to use
    capabilities for shifted scroll forward/reverse as shifted
    cursor up/down.
  • correct updating of checkmark for toolbar entry in popup
    menu (report by Emanuele Giaquinta).
  • fix ifdef’s to allow compiling with toolbar and without
    tek4014 (patch by Emanuele Giaquinta).
  • use openpty() for Darwin port (patch by
    Emanuele Giaquinta).
  • fix Gentoo #90697 a
    different way, postponing the logic in
    SetupToolbar until the toolbar is actually needed,
    i.e., the +tb option is handled as expected.
  • revert fix made in patch #203 for
    Gentoo #90697.
    That introduced a problem with the control mouse click-popups,
    while most of the performance problems can be resolved by
    restricting the menu fonts (report by Emanuele Giaquinta).
  • fix a file-descriptor leak when calling
    openpty() (OpenBSD system/4561).
  • make a special case of resizing work like vt100: a hard
    reset also resets the 132/80 mode. The code to support this was
    present since X11R5, but not used because the corresponding
    initial state of the -132 option was not
  • restore window manager hints after
    XtMakeResizeRequest() calls. One instance from
    patch #205 resulted in the window
    manager displaying pixels rather than than characters after
    selecting a different font size (FreeBSD ports/87424).
  • remove special case for Darwin in CF_XOPEN_SOURCE (Emanuele
  • modify parsing of control sequence CSI T
    to allow scroll-down to be sent while mouse tracking is enabled
    (request by D Hugh Redelmeier).
  • correct termcap “me” (mode-end) string so it does not
    modify the alternate character set (report by Andrey
  • correct size-comparison in HandleInterpret() broken in
    changes from patch #201 to allocate
    input buffer (Debian #334317).
  • amend adjustments for scrollbar layout from patch #204 to make this apply only to the
    toolbar configuration. In the non-toolbar configuration, the
    resulting scrollbar was shifted by its borderwidth (report by
    Matthieu Herrb).
  • add select-cursor-extend() action.
  • fix some broken href’s in xterm.log.html, and typo in the
    INSTALL file (report/patch by David Martínez Moreno).
  • modify Imakefile to use setgid mode for
    installing with Linux, OpenBSD and FreeBSD.
  • add configure –with-setuid and –with-reference options to
    allow packagers more flexibility in customizing install
  • generalize and make optional (configure –with-utmp-setgid)
    the change made for XFree86 Bugzilla #878 in patch #205 (FreeBSD bug report
  • correct a typo in CF_FUNC_TGETENT introduced in patch #198 fix for Gentoo #69926.
  • implement logic in termcap query to process multiple
    parameters as documented in ctlseqs.ms
  • fix buffer size used for termcap query, which was not long
    enough for the terminfo “colors” name (patch by Bradd W.
  • add configure option –enable-readline-mouse, which turns
    on the experimental OPT_READLINE code (patch by Ilya
  • for FreeBSD, drop setuid privileges after startup (XFree86
    Bugzilla #878, report/patch by Alexander Pohoyda)
  • add menu entry (alt-esc) and corresponding action
    (alt-sends-esc) to toggle the eightBitInput
    resource setting.
  • generate configure script with autoconf 2.52 (patched) to
    use the feature therein which forces the script’s locale to
    POSIX (Gentoo Bugzilla #105369).
  • modify computation of rows/columns on resize to avoid
    extending beyond the given limits, e.g., if resizing in
    response to a “maximize” in Gnome or KDE which do not use the
    window manager hints for this case (Debian #289123,
    Novell #61153).
  • modifications to work with z/OS 1.4 (Paul Giordano).
  • improve error-reporting when chown/chmod of the
    pseudo-terminal fails, e.g., if a copy of xterm which was
    designed to work with old-style pseudo-terminals is not
    installed setuid or setgid (report by Jeremy C. Reed).
  • patches from Emanuele Giaquinta:
    • ctlseqs.ms says that primary and secondary DA accept a
      nonzero parameter, which is incorrect. Also modify code to
      agree with this.
    • correct a comment in 88colres.pl
    • ctlseqs.ms says that the DEC Set/Reset control
      sequences for the backarrow key make it send DEL/BS,
      respectively, while it’s the opposite.
    • update the menu entries for the metaSendsEscape,
      deleteIsDEL and numLock resources when changing them with
      the corresponding DEC Set/Reset control sequences.
  • fixes for configure script:
    • add special case for QNX, defining _QNX_SOURCE.
    • check for preprocessors which do not perform -U and -D
      options in the given order.
    • improve macro to determine gcc version
    • improve check for Intel compiler and related warning
  • update config.guess, config.sub
  • improve fix from patch #198 for
    Cleanup() by ensuring it is not called from the
    SIGCHLD handler (patch by Todd Miller (OpenBSD CVS)).
  • eliminate a retry for a better-matching bold font, to work
    around recent font server changes.
  • fixes for Novell #113277:
    • specify weight for wide font which may be derived from
      normal fontname.
    • cache the derived wide- and widebolt-fontnames.
  • workaround for Gentoo Bugzilla #100728.
  • add menubar’s border width to layout computation. Normally
    this is zero, but patterns such as
    XTerm*borderWidth:1 would give poor layout for the
    toolbar configuration.
  • work around quirk in shell which allowed user to “run”
    uxterm script when there was no UTF-8 locale installed by
    entering “sh -x uxterm”. Also popup an xmessage to show the
    problem for users who run this via a GUI (Debian #318513).
  • disallow changes to fonts, toolbar and scrollbar when the
    vt100 window is iconified, to simplify management of the active
  • fixes to make -geom option work properly with
    the toolbar configuration (Gentoo #90717,
    Gentoo #91967).
  • minor improvements to scrollbar layout: ensure that the
    scrollbar border is zero if the vt100 border is zero.
  • improve initialization due to utf8 resource by
    loading the utf8Fonts resource in the case where
    locale resource is false. Also in this case, do
    not disable switching UTF-8 mode on/off.
  • minor optimization of TrueType font-loading, loads italic
    font only when needed.
  • correct variable used to store temporary result from lookup
    of bold font, when initializing the wide-bold-font data and no
    immediate match is found by asking for a bold variant of the
    wide-font. The result was to use overstriking rather than the
    actual bold font (Debian #318162,
    patch by Eugene Konev).
  • add checks to ensure vt100 widget is realized when it might
    be updated via actions handled from a tek4014-only
  • modify initialization of allowSendEvents and
    allowWindowOps to prevent modification with the
    editres protocol.
  • fix compiler warning for NetBSD by including util.h in
    main.c (XFree86 Bugzilla #1596).
  • fix to build on Darwin 8.x, which no longer provides
    setpgrp() (patch by Min Sik Kim).
  • adapted fixes for Legend (SCO) from diffs attached to
    Freedesktop.Org Bugzilla #3180.
  • fix typo in xterm manpage description of -ls
    option (Freedesktop.Org Bugzilla #3543).
  • add support for interpreting the underline attribute as an
    italic font in Xft mode (patch by Chuck Blake).
  • improve initialization of toolbar so that individual
    pulldown menus are initialized on demand, as they are in the
    scenario where they are invoked as popup menus (Gentoo Bugzilla
  • fix logic in find_utmp, which did not reset
    result in getutid(), causing an infinite loop in
    some conditions (report by Emil Mikulic).
  • set the _NET_WM_PID property (Gentoo Bugzilla
  • modify ifdef’s in ptyx.h and xterm_io.h to build with
    DragonFly (patch by Jeroen Ruigrok).
  • change default values for minBufSize and
    maxBufSize to 4096 and 32768 respectively so that
    the initial read request will match the value from before
    changes to use sched_yield().
  • make paste of UTF-8 faster for Western character sets by
    checking range of incoming data (patch by Joe Allen).
  • add experimental option to allow applications to get or set
    the selection data as a BASE64 string (adapted from patch by
    Joe Allen).
  • fix an off-by-one error parsing -S/nn option
    (Debian #311438,
    report/fix by Peter Chubb).
  • fix an initialization bug from patch
    #201 that broke logging (report by Rodney Thayer).
  • amend change to command-line processing in patch #201 to avoid conflict with
    -e option (report by Servatius Brandt).
  • suppress configure check for _XOPEN_SOURCE on
  • update config.guess, config.sub
  • add extended shift- and control-modifier cursor keys to
    “xterm+pcfkeys” terminfo entry to correspond to ncurses
    20050430 patch.
  • fix a rare case where text would be written with the wrong
    colors because output of scrolled text would reset the colors
    and the new text would be written with the same colors
    (report/testcase by Thomas Glanzmann).
  • rename $CMD variable in plink.sh
    because it is a reserved symbol in bash 3.00.16 (report by Ted
  • add environment variables $XTERM_SHELL and
    $XTERM_VERSION (request by Zdenek Sekera).
  • corrected workaround for background color of menubar, which
    gave a too-broad expression, coloring the background of the
    VT100 widget unless overridden by another resource
  • correct an error in the logic which decides when
    sched_yield() is run; it would occasionally hang
    when contending with other pseudo-terminal applications such as
    screen (report by Kirill Ponomarew).
  • modify initialization to decide whether to default to
    built-in wcwidth() versus system’s version based
    on the starting locale and whether the system’s version is poor
    quality (suggested by Bram Moolenaar).
  • update table for mk_width() from UnicodeData
    4.1.0 using Markus Kuhn’s uniset script.
  • improve resource files to show how the menubar and popup
    menus can be colored (prompted by report by Joe Wells).
  • modify parsing of OSC (and SOS, etc), strings so their
    contents are not interpreted as UTF-8. This allows non-ASCII
    title strings to be set, provided that the window manager
    complies (report by Thomas Wolff).
  • improved some of the built-in line-drawing glyphs.
  • correct color of “box” character drawn for line-drawing
    glyph 1 (report by Nicolas George).
  • improve behavior when switching to UTF-8 mode after startup
    so xterm will check if the current fonts are already wide
    (ISO10646-1). If they are not, xterm will use the
    utf8Fonts subresource to load appropriate fonts
    (request by Bram Moolenaar).
  • modify logic for setting title-string so it applies to the
    current widget rather than the vt100 widget.
  • modify initialization for wide-bold fontname, to search for
    one if none is given (report by Michael Schroeder).
  • add resource mkWidth and command-line option
    -mk_width to control whether xterm uses the
    built-in version of wcwidth().
  • add resource settings for minimum/maximum input buffer
    size, and call to sched_yield to improve
    performance with newer Linux kernels (adapted from patch by
    Nicolas George).
  • correct computation of width for wide characters with the
    invisible attribute (report by Thomas Wolff).
  • modify interaction between +u8 and
    locale resource to allow the command-line option
    to override the resource (requested by Thomas Wolff).
  • add a limit check for scrolling margins in a one-line
    screen, overlooked in fixes for patch
    #198 (Debian #297430).
  • correct treatment of iconBorderWidth for
    resizing an active-icon, and its description in manpage (Debian
  • modify configure script --disable-imake to use
    the script’s definitions anyway if it cannot detect imake
    (prompted by FreeBSD bug 77408).
  • ignore error in the I/O initialization that tries to set
    the tty to 7-bit input for the case where
    eightBitInput resource is false (Debian #298551).
  • modify command-processing to accept an optional parameter
    that tells xterm which shell program to use (request by Zdenek
  • add simpler resource keyboardType which, when
    set, overrides the individual keyboard-type resources and
    eliminates the possibility of conflict between them.
  • add initialization for scoFunctionKeys
    resource (report by Rick K).
  • correct logic of ReallocateBufOffsets() which
    did not copy the content of the old screen buffer to the proper
    location, making a repaint clear after switching to
    wide-character mode (report by Bram Moolenaar):
  • implement the remaining pieces to make xterm allocate cells
    for wide-characters when the “UTF-8” menu entry is selected.
    Also, load the UTF-8 font when that menu entry is selected, or
    the escape sequence for UTF-8 mode is received (report by Bram
  • add command-line options (-tb,
    +tb) and resource toolBar to allow
    menu/toolbar to disabled or enabled at startup (prompted by
    reports by Joe Wells).
  • correct typo in configure script’s –enable-dec-locator
    option (report by Bram Moolenaar).
  • increase color pairs value for xterm-256color and
    xterm-88color to match ncurses, which has an experimental
    option to support this.
  • modify ifdef’s to make AIX use termios rather than termio;
    the struct sizes for the two were not the same.
  • improve CF_WITH_IMAKE_CFLAGS configure macro script for
    OSMAJORVERSION and OSMINORVERSION values, e.g., for Tru64 and
  • modify ifdef to define USE_POSIX_TERMIOS for Darwin (patch
    by Min Sik Kim).
  • modify find_utmp() to initialize the whole utmpx struct
    (except ut_id), since that is needed for OSF1 4.0D to prevent
    an infinite loop on exit.
  • add configure check before adding -D_POSIX_SOURCE since
    some platforms predefine it, e.g., cygwin.
  • add simplified sed expressions in CF_IMAKE_CFLAGS configure
    script macro to ensure value for PROJECTROOT is quoted on
    Solaris, i.e., when nested \( and \)
    are not interpreted correctly.
  • correct DEC rectangle operations to reset state after
    completing the operations.
  • modify CASE_ST handling in charproc.c to ensure that the
    parse state is reset even if xterm is not currently processing
    an OSC or other string (patch by Johnny Billquist forwarded by
    Matthias Scheler, NetBSD xsrc/29003).
  • fix OS/2 build for innotek_libc (patch by David Yeo).
  • fix a regression from patch #197
    fix for Debian #277832
    which disowned the selection if it was scrolled, e.g., by the
    user pressing return at the bottom of the screen (Debian
  • move the warning/exit for missing $DISPLAY into the error
    handler in case -display is given, and the connect
    fails for some other reason.
  • instead of setting $DISPLAY, check for the
    unset variable and warn/exit on this condition (comments by H
    Merijn Brand and Bernhard R Link).
  • fix a typo in Imakefile from patch
    #198 (reports/patches by Stefan Dirsch, Mike Castle).
  • set $DISPLAY to “:0” if it is not set.
  • add utmpDisplayId resource to allow users to
    control whether the display identifier (display number and
    screen number) are retained in the connection information
    recorded in utmp (discussion with Edoardo Tirtarahardja).
  • add bellOnReset resource to allow users to
    disable bell which sounds on hard reset since patch #183 changes to DECSCL (discussion with
    Danek Duvall).
  • improve $WINDOWID for configuration with
    toolbar by making it refer to the top-level shell rather than
    the parent of the current window. For that case, the parent is
    a form widget, which does not have a name, which made the
    $WINDOWID not very useful as a parameter for
    xwininfo (suggested by Dave Bodenstab).
  • fix a typo in WhichVFont() macro from patch #197 changes which broke the
    –disable-active-icon configuration (report by Ralf S.
  • improve some limit checks (Gentoo Bugzilla #75604).
  • add –disable-setuid option to configure script (Gentoo
    Bugzilla #76543).
  • add –disable-full-tgetent option to configure script,
    allowing one to ignore a termcap library in favor of
    ncurses/curses (Gentoo Bugzilla #69926).
  • modify configure script to choose useful warning options
    for Intel version 8.0 compiler.
  • update config.guess, config.sub
  • make active-icon work properly when TrueType fonts are used
    (Debian #286068).
  • correct change from patch #157
    which uses getlogin() to check for an alias; the storage used
    for the related getpwuid() call was overwritten by the data
    used for comparison (patch by Per Hedeland).
  • correct case of SCS for character set 0 (line-drawing) to
    allow it to be selected into GR.
  • fix a file-descriptor leak (Redhat Bugzilla #139597).
  • modify creat_as() to only fork if xterm is
    actually running as setuid and/or setgid. This works around a
    Cygwin bug which hangs when logging is enabled and makes xterm
    a little faster for systems using interfaces such as utempter
    (report by Al Goodman).
  • modify Cleanup() to avoid operations such as X
    calls that might use unsafe functions when it is called by a
    signal handler (report by Michiel Boland).
  • fix bugs in patch #191 and in SRM
    changes from patch #197 that broke
    DECSET 38: switch to Tek4014 emulation (report by Dave
  • fix for manpage escapes (Marc La France).
  • improve on IRIX-specific change for Imakefile in XFree86
    CVS versus resize linking against termcap library to reflect
    definitions for USE_TERMCAP in resize.c
  • modify configure script to remove empty “-DPROJECTROOT=”
    definition which resulted unusable values for luit’s default
  • update precompose.c based on Unicode 4.0.1
  • several minor fixes based on Intel compiler warnings.
  • change default translations so a BtnDown which
    is not recognized is simply ignored rather than emitting a
    bell. That makes it less obtrusive when the user tries to use a
    mouse which provides more capabilities than the X mouse driver
    supports, e.g., one with a horizontal scroll wheel (Debian
  • note in xterm’s manpage that translations is
    not specific to xterm (Debian #278897).
  • modify uxterm script to use locale program to verify if the
    derived locale is installed (Debian #246398).
  • correct font handling for active icon when in UTF-8 mode
    (report by Paolo Liberatore).
  • make active-icon and toolbar configurations work
  • modify the criteria for disowning primary selection.
    Previously, this happened anytime the cursor was moved before
    the end of the selection. That would ensure that any
    insert/delete of char or line, as well as scrolling, would
    disown the selection. The new criteria change this to checking
    if the operations would modify the data which is highlighted
    (Debian #277832).
  • reimplement DECALN with functions for vt420
    rectangles, fixes selection for this case.
  • implement vt420 rectangle operations.
  • add parsing, for debug/test of vt220 soft-fonts.
  • add menu entry, actions and escape sequence to allow
    enabling/disabling toolbars at runtime.
  • improve rendering for Xft, allow it to draw non-linedrawing
    characters such as “pi”, which were drawn from internal tables
    with patch #180 (Freedesktop.org
    Bugzilla #1260).
  • add configure option --enable-mini-luit,
    ifdef’d the mini-luit feature with
  • add mini-luit feature, which supports Latin9 directly
    rather than via luit, provided that Unicode fonts are used
    (Freedesktop.org Bugzilla #1571,
    request by Stefan Dirsch, patch by Michael Schroeder).
  • for Linux, if IUTF8 is defined, e.g., on recent 2.6.x
    kernels, set the corresponding flag for the slave pty, to
    enable UTF-8 interpretation of backspace in cooked mode
    (Freedesktop.org Bugzilla #1578,
    request by Stefan Dirsch).
  • modify faceSize resource to use a
    floating-point internal value (adapted from patch by Sam
  • correct handling selection of tabs over cleared space in
    UTF-8 mode. In this special case, xterm represents the
    whitespace with a null rather than a space character (Debian
  • amend fix for infinite loop from patch
    #192 to check if there is wrapped text to output in a
    following iteration (Debian #273202).
  • fixes ifdef’d with __INTERIX to allow building
    with Interix (Windows Services for UNIX) 3.5 using the xlibs
    libraries from freedesktop.org (patch by Min Sik Kim).
  • amend solution for Debian #252873,
    Debian #260471
    from patch #194 by making the cursor
    not explicitly colored if only the foreground color is set, and
    the cursor is on a blank space (Debian #275473).
  • correct logic for send/receive mode (SRM) with regard to
    control characters.
  • fix masking of invisible text in wide-character mode, which
    did not work for line-drawing characters.
  • incorporate CF_XOPEN_SOURCE into configure script,
    replacing CF_GNU_SOURCE.
  • add a special case to configure script to ignore NetBSD’s
    grantpt() which was recently added (but not part
    of a release), until someone is able to ensure that xterm can
    use it (report by Min Sik Kim).
  • clear the buffer returned by getutid(). This fixes an
    infinite loop on some platforms introduced in patch #193 by Debian #256468
    fix (report by David Ellement).
  • correct length used for blinking text, make the last column
    blink (patch by Alexander V Lukyanov).
  • start changes to make doublesize characters work with
    TrueType fonts (see patch #44).
  • trim leading/trailing blanks from color resources as done
    for other strings in patch #167.
  • fixes for showBlinkAsBold resource (report by
    Christoph Berg).
  • change clearing operations so foreground color attribute is
    not set. Usually this is benign, but in some cases when the
    cursor color is not set explicitly, the cursor would show this
    color (Debian #252873,
    Debian #260471).
  • add extra state (cursor-moved) to guard against deciding
    that the cursor did not require repainting after an indexing
    operation that leaves the cursor in the same location on the
  • fix a case where a full-screen indexing operation would not
    restore the cursor-busy state.
  • fix a repainting bug introduced in patch #180: when using a font lacking
    line-drawing characters, a repaint of the screen could skip
    horizontally an extra amount after filling in the missing
    character (reports by Nicolas George, Hans de Goede, Redhat
    Bugzilla #128341).
  • rename terminfo fragment “xterm-pc-fkeys” to
    “xterm+pcfkeys” for consistency with ncurses.
  • fix for wide-character selection from OpenBSD CVS
    (report/patch by Matthieu Herrb).
  • modify initialization and cleanup of utmp data to also
    compare the ut_line member (Debian #256468).
  • modify check on focus-change to ignore
    FocusOut events generated by
    XGrabKeyboard. This fixes a case where the text
    cursor would act as if focus were lost when selecting the
    “Secure Keyboard” menu option.
  • add gen-pc-fkeys.pl script, use that to
    generate terminfo fragment corresponding to the pc-style
    function keys.
  • fix a case where the checkmark by the “VT220 Keyboard” menu
    entry was not set on startup.
  • separated two methods for making the cursor blink (menu and
    escape sequence) by using the menu as the primary method and
    XOR’ing the state of the escape sequence against that.
  • modify logic for enabling blinking cursor via escape
    sequence as well as the related save/restore operations so this
    is only available if the cursorBlink resource was
    set on startup.
  • change resource settings for color4 and color12, add some
    discussion in XTerm-col.ad (Debian #241717).
  • add a note in xterm manpage discussing the difference
    between alt- and meta-keys, and the way the latter is used in
    the eightBitInput resource.
  • add a note in xterm manpage regarding possible conflict
    between resource settings for xterm.vt100.font and
    xterm.vt100.utf8Fonts.font (Debian #254650).
  • add compile-time customization of
    backarrowKeyIsErase and
    ptyInitialErase default resource values (adapted
    from OpenBSD CVS).
  • change parameter of FIONREAD
    ioctl() call from long to int (discussion on
    tech-x11@netbsd.org regarding LP64 by John Heasley and Matthias
  • modify configure script options for Athena widgets to work
    as expected for “–without-Xaw3d”, etc., (Gentoo Bugzilla
  • add case to uxterm to accommodate locales ending with
    “@euro”, e.g., fr_FR.UTF-8@euro (Debian #255197,
    report/analysis by Matthieu Lagouge).
  • add special case for VT100 graphic’s “box” character
    (discussion with Ben Armstrong).
  • add missing initialization for bitmap-font sizes needed to
    make fonts menu work with TrueType fonts (report by Ben
  • save the fontnames for bold fonts that are derived from
    normal fonts, or from the boldFont resource, so
    the same value is restored when switching with the VT Fonts
    menu (Debian #256086).
  • fix manpage preprocessing (Marc La France).
  • fix typo in manpage’s description of character classes
    (Debian #257073).
  • modify terminfo to accommodate luit, which relies on G1
    being used via an ISO-2022 escape sequence (Debian #254316,
    analysis by Juliusz Chroboczek).
  • modify Makefile.in rule for ctlseqs.txt to work around
    groff SGR misfeature.
  • modify XTerm.ad to set saveLines
    default to 1024 (Redhat Bugzilla #127132).
  • add a limit-check in dotext() to prevent
    infinite loop in a corner case of UTF-8 configuration.
  • update config.guess, config.sub
  • correct options parsing for -into option so it
    can be combined with -e (Redhat Bugzilla #124518,
    report/patch by James Armstrong).
  • fix ifdef’s for OPT_COLOR_RES2 so that the
    fake resource table introduced in patch
    #188 is not compiled if it is empty. This happened to work
    with gcc (report by Joel Konkle-Parker).
  • reorganize ptydata.c to use one input buffer shared between
    the VTxxx and tek4014 emulators. In the new scheme, UTF-8
    decoding is performed on characters as they are needed by the
    state machine rather than on buffers as they are read.
  • work around change in quoting of PROJECTROOT symbol when
    using configure script, from changes made in patch #187 for CF_IMAKE_CFLAGS (Gentoo
    Bugzilla #50982).
  • correct state for values in the range 128-159 after
    translating from UTF-8: this should be ignored unless the user
    overrides it with the allowC1Printable resource
    (reported by Simon Strandgaard).
  • add configure-script check for nl_langinfo(CODESET), use
    this to replace check of environment variables for UTF-8.
  • change Makefile.in rules for 256colres.h and 88colres.h so
    they are only made if they do not exist. The maintainer-clean
    rule will remove these files; they require perl to be
  • remove check in configure script –enable-toolbar that
    suppressed this option when building with Xaw7.
  • apply fixes to Tektronix widget used for VT100 widget to
    make toolbar work with Xaw7 (XFree86 4.x).
  • do not call xim_real_init() if openIm resource
    is false (Debian #249025).
  • minor improvements to built-in line-drawing.
  • fix a few portability issues with dynamic abbreviation
    support, i.e., did not compile on Tru64.
  • modify constraints in form used to layout toolbar, to work
    with newer Xaw in XFree86 4.x.
  • correct table entry for DEL in the ground state, which
    marked it as a printable character from patch #171 (report by D Hugh Redelmeier).
  • improve fix in patch #186 for
    failure in xim_real_init() by adding a sleep.
  • fix a typo in os2main.c (XFree86 Bugzilla #1358,
    report/patch by Frank Giessler).
  • make escape sequence reporting dynamic colors consistent
    with the logic that sets it; choosing the opposite color when
    reverse video is set (XFree86 Bugzilla #1361, reported by Bradd
    W Szonye).
  • modify initialization of 256- and 88-colors so that colors
    beyond 16 are normally not X resources. This works around a
    hard-coded limit in Xt which breaks xterm when 256-colors and
    luit are both configured (report by Noah Friedman).
  • remove ncv from xterm-256color
    terminfo entry since it is no longer needed (report by Eli
  • add “erase2” and “eol2” keywords to ttyModes
    resource, for recent/current FreeBSD.
  • improve ifdef’s for utempter library to omit direct calls
    to setutent() or getutent() (adapted
    from patch by Christian Biere).
  • add dynamic abbreviation support like Emacs (patch by
    Tomasz Cholewo). This is ifdef’d with OPT_DABBREV,
    and enabled via the configure script
    --enable-dabbrev option.
  • fix problem responding to session management events, e.g.,
    which would make logging out very slow (patch by Eddy De Greef,
    Debian #233883).
  • work around a newer bug in toolbar with XFree86 by forcing
    menus to be fully initialized at startup. In older versions of
    XFree86 and X11R6, it was possible to delay initialization of
    the menu contents until it was popped up. (The longstanding bug
    with XFree86 4.x layout for toolbar still exists, though).
  • modify xtermAddInput to work around core dump
    on IRIX64 when initializing scrollbar translations if toolbar
    is compiled-in.
  • fix some minor conflicts in the 2-character entry names in
    the termcap file.
  • change xterm version string to use
    __vendorversion__ where that is available, and
    “XTerm” otherwise.
  • improve description of utf8 resource in
    manpage (Debian #179407).
  • modify configure macros CF_IMAKE_CFLAGS and CF_ADD_CFLAGS
    to handle -D options that define string values,
    e.g., for XVENDORNAME.
  • modify configure macro CF_IMAKE_CFLAGS to allow (if
    $PATH is set accordingly) to use the
    xmkmf script within an X build tree.
  • add missing #undef OPT_SESSION_MGT to
    xtermcfg.hin to make the configure script’s
    –disable-session-mgt option work.
  • update config.guess, config.sub
  • change reset on DECSCL to a soft-reset
    (appears some DEC manuals have errors).
  • add a section to ctlseqs.ms elaborating on normal/alternate
  • modify initialization of Acolors[] so that any
    XtDefaultForeground or XtDefaultBackground values are
    translated as in the Tcolors[], to use the window’s
    foreground/background colors. This affects the
    colorBD and similar resources which normally have
    no explicit color assigned.
  • modify initialization of Tektronix window so that control
    sequences setting its color before the window is popped up will
    apply to its initial colors.
  • add control sequence to set Tektronix window’s
  • modify initialization of terminal colors, e.g., mouse
    pointer and text cursor, to treat
    XtDefaultForeground and
    XtDefaultBackground values as the actual
    foreground and background colors of the terminal rather than
    white and black (Debian #241717).
  • remove an incorrect comparison against
    PTYCHARLEN in parsing the -S option
    (report by Michael B Taylor).
  • minor restructuring of terminfo/termcap files, having noted
    some packager’s customizations which caused the structure to be
  • eliminate an isolated use of MIN/MAX in charproc.c
  • replace XtExtdefaultfont and
    XtExtdefaultbackground by their more familiar
    equivalents XtDefaultFont and
  • replace ifdef’s using SCO, sco
    and SCO325 with __SCO__ (XFree86
    Bugzilla #1301, Kean Johnston). But ensure that it still builds
    on platforms where this symbol is not defined.
  • add scoFunctionKeys resource, to match
  • update manpage discussion of menus and related
  • enable utmpx support for NetBSD 1.6C and newer (patch by
    Matthias Scheler).
  • add a note in the manpage discussing xterm’s treatment of
    open file-descriptors (request by Dan Shearer).
  • modify Help() to make “xterm -h” write to
    standard output rather than standard error (patch by Bram
  • check for type of failure in xim_real_init() to avoid
    looping when the problem is an unsupported input method rather
    than a failure to connect to the XIM server. Problem was
    introduced in patch #175 (XFree86
    Bugzilla #1306).
  • modify Imakefile to remove dependency of
    “install” target on the xterm executable to make
    installs from tree without attempting to rebuild anything
    (commit by David Dawes, report/patch from Lee Olsen).
  • remove call to ShowCursor from
    SetCursorBlink() since that is redundant, and can
    cause display glitches if the cursor is already blinking
    (XFree86 Bugzilla #1158, patch/report by Andreas Schwab).
  • fix tcap-query logic for the backspace key (XFree86
    Bugzilla #1233, report/patch by Anton Kovalenko).
  • add test-scripts resize.pl and tcapquery.pl
  • add translation to ASCII of commonly-used characters that
    groff translates to Unicode, when the font in use does not
    provide the corresponding glyphs (Debian #219551).
  • modify RequestMaximize(), which performs maximize/restore
    via control sequences, to account for window-frame (patch by
    Jess Thrysoee).
  • improve pattern used in uxterm to check for
    UTF-8 locale, e.g., for HPUX (patch by H Merijn Brand).
  • add -fd option and resource
    faceNameDoublesize to specify double-wide fonts
    with Xft (adapted from patch by Zarick Lau).
  • change a couple of resource classes from “Boolean” to
    specific values: freeBoldBox,
  • add resource showMissingGlyphs to outline
    places on the screen where a font lacks the corresponding
  • add resource showBlinkAsBold to control
    whether blinking text should be shown as bold or actual
    blinking text.
  • improve logic in ShowCursor() to avoid repainting the
    cursor when it is already visible, e.g., in response to the
    DECTCEM escape sequence. This also addresses XFree86 Bugzilla
  • implement blinking text, using the timer for blinking
  • modify FreeType support to allow resizing the font, in the
    same ways the window can be resized if fixed fonts are used.
    The relative font sizes are derived from the fixed font
  • add menu items and corresponding actions for switching
    on/off the UTF-8 mode and Xft (TrueType) support.
  • add logic to handle switching UTF-8 mode on/off. It worked
    if the escape sequences were flushed, but if data was mixed in
    with the same write, some were not handled properly (report by
    Nicolas George).
  • modify to allow turning UTF-8 mode on/off via escape
    sequence even if -wc option was not given at
    startup (patch by Peter Berg Larsen).
  • amend fix for XFree86 Bugzilla #981, adjusting for
    savedlines value (report by Tim Adye).
  • fix a typo in computing relative font size (Jess
  • improve configure-script checks for FreeType and related
    libraries, using xft-config or
    freetype-config scripts when available.
  • fix configure-script check for SYSV definition
    by ensuring whether sys_errlist[] is declared, and
    by modifying the test program to include
    X11/Intrinsic.h to check that wchar_t
    is declared consistently (report by H Merijn Brand).
  • correct logic for configurations that may attempt to open
    both old/new-style pseudoterminals. In this case, the old-style
    are preferred. Logic was broken in patch
    #145 (XFree86 Bugzilla #997, report/patch by Kean
  • modify handling of eightBitInput resource in
    UTF-8 mode to translate the value into UTF-8. Otherwise an
    illegal UTF-8 code is sent to the application (report by Bram
  • modify uxterm script to interpret help and
    version options so xterm does not always create a window when
    the user requests this information (Debian #223926).
  • add a limit check to ScrnTstWrapped() (XFree86 Bugzilla
  • modify DECSCL to perform a hard reset (RIS) as per DEC
    manuals. Extended DECSCL to accept parameters for vt4XX and
    vt5xx terminals.
  • correct logic for ANSI conformance level escape sequences,
    which were confused with DEC conformance level escape
  • correct state for vt52 shift-in/shift-out, which was not
    reset properly after patch #171.
  • correct handling of graphics characters for vt52 mode,
    which did not display line-drawing characters after G1 fix from
    patch #182.
  • fixes for configure script to work with current Cygwin
    headers and libraries.
  • modify Imakefile to put the current directory
    at the beginning of the include searchpath (patch by David
  • correct logic for metaSendsEscape resource to
    allow for the meta-right key to be tested. Improve check for
    meta/alt modifiers by dropping the assumption that a keysym is
    associated with only one modifier, and by ignoring NoSymbol
    entries in the xmodmap data (XFree86 Bugzilla #924, patch by
    David Dawes).
  • correct wrapping logic for line-drawing characters written
    in non-UTF-8 mode of xterm build for wide-characters (XFree86
    Bugzilla #918, report/patch by Jürgen Keil).
  • improve manpage description of resources, in particular the
    utf8Fonts class (XFree86 Bugzilla #905).
  • add definitions to compile with glibc-based GNU/Hurd,
    GNU/KFreeBSD and GNU/KNetBSD (XFree86 Bugzilla #893).
  • compiler-warning fixes (patch by Christian Biere).
  • add README.i18n (Tomohiro Kubota).
  • correct initialization of G1 character set mapping, which
    used line-drawing set as a result of confusion between “DEC
    Supplemental Graphic” and “DEC Special Graphic” character sets
    (see patch #34 regarding DECSTR).
  • correct ifdef in main.c for variable utret (patch by
    Bernhard Rosenkraenzer).
  • document in xterm’s manpage how to use XFree86 “:unscaled”
    keyword to suppress scaling of bold fonts.
  • when deriving bold fontname from normal fontname, use the
    normal font’s average width to avoid for example selecting
    7x13bold from an 8×13 normal font (Debian #107769).
  • implement boldMode for wide-character logic in
    drawXtermText() (report by Michael
  • modify UXTerm.ad resource file to include
    “XTerm-color” rather than “XTerm”, in case the latter file
    contains no color resource definitions, e.g., after patch #180.
  • add action load-vt-fonts() and configure
    option --enable-load-vt-fonts which allows users
    to define additional sets of VT-fonts which can be loaded at
  • add logic to wide-character support which attempts to load
    fonts specified by utf8Fonts subresources at
    startup. The subresources have the same names as the fonts
    which they replace, e.g., font,
    font1, etc., so that the ISO-10646-1 fonts can be
    specified in the XTerm app-defaults file (adapted
    from patch by Tomohiro Kubota).
  • improve Set Font escape string handling by making the
    relative settings apply to the size of the font, as in shift
    keypad plus/minus rather than the menu index.
  • simplify parameter passing for the set-vt-font
    action and related code.
  • cleanup some include-ordering, moving some recently-added
    hardcoded stuff into xterm.h where it will not interfere with
    the configure script.
  • modify xtermAddInput() (see note on
    augmentation in patch #158) to use the
    complete set of default keyboard translations so that one can
    use shifted pageup, wheel mouse, etc., while the mouse pointer
    is over the scrollbar. (Debian #178812).
  • make save/restore mode controls apply to show/blink cursor
  • add escape sequence to start/stop blinking cursor, which
    allows implementing cvvis terminfo capability
    (request by Nate Bargmann).
  • add indp and rin to terminfo
  • fix an out-of-bounds array reference in ScrnRefresh() for
    wide characters (report by Dan Harnett, patch by Todd Miller).
    This bug dates from patch #141.
  • several fixes for rendering using Xft via option

    • translate Unicode values, i.e,. from UTF-8 output to
      xterm, for line-drawing to xterm’s internal code, etc.,
      since TrueType fonts generally do not have either set of
      line-drawing glyphs. xterm can draw these directly (report
      by Abigail Brady).
    • pass 16-bit values rather than 8-bit values to
      xtermXftDrawString() to allow for wide-characters.
    • remove spurious check for colorBDMode
      resource in logic that does bold fonts.
    • implement underlining.
  • patches by Ilya Zakharevich:
    • improve drawXtermText() by making the recursive calls
      communicate through arguments, and not through
      saving/restoring global variables.
    • make double-width characters work with -u8
    • modify lookup of double-sized fonts by checking for a
      match ignoring x/y resolution if the first check
  • make height of TrueType fonts match ascent+descent (patch
    by Keith Packard).
  • correct configure-script check for imake $CFLAGS, which did
    not check properly if imake was not available.
  • correct install rule in Makefile.in for
    uxterm, which was attempting to strip the script
    (newsgroup posting by Fernan Aguero).
  • correct AF/AB strings in termcap for xterm-256color and
    xterm-88color entries (report by Josh Howard).
  • update wcwidth.c to incorporate changes from Markus Kuhn’s
    2003-05-20 (Unicode 4.0) version of that file.
  • modify scroll-back and
    scroll-forw actions to accept an adjustment value,

                  scroll-back(1, page-2)

    to scroll back by 2 lines less than a page (patch by Greg

  • use color resource setting from Debian package for xterm
    VT100 widget, since the choice of blues provides better
  • remove color resources from XTerm.ad, leaving
    them only in XTerm-col.ad (prompted by Debian
    package for xterm).
  • correct configure script option –enable-pty-handshake
    (report by Paul Gilmartin).
  • add visualBellDelay resource to modify the
    length of time used for visual bell, for very slow displays or
    very fast computers (reports by Ingo van Lil and Doug
  • correct logic for initializing dynamic highlight color
    (patch by Jess Thrysoee).
  • add a check for non-zero size in call to
    XCopyArea() to accommodate a Solaris bug.
  • correct typo in example for character classes in xterm
    manpage (Debian #198910).
  • link xterm with bind_on_load option on Darwin to work
    around a deadlock in the dynamic loader when a signal is
    received while the dynamic loader is looking up symbols
    (patches by Rob Braun, Torrey Lyons).
  • modify configure script to avoid using “head -1”.
  • update config.guess, config.sub
  • modify ifdef’s to work around inclusion of types FcChar32
    and XftCharSpec with FreeType 2.0 (see patch #175).
  • modify the predictable version of the generated logfile
    name (see patch #171) to append the
    process-id rather than a random value.
  • resync with XFree86 CVS
    • Enable SCO function keys in xterm
    • Make Delete key send DEL by default on SCO in
  • modify ifdef’s for WTMPX_FILE to allow building on cygwin
  • change the default of pty-handshaking configure option (and
    related default for imake) to assume this feature is
  • add ifdef’s so configure option for XawPlus library works
    (patch by Pavel Roskin).
  • add $(MAIN_DEFINES) to the variables checked
    by the configure script that may contain information set by
    imake. This allows xterm to build/run properly using the
    configure script on IRIX64.
  • add configure check for “.exe” suffix on cygwin.
  • modify default for configure
    --enable-pty-handshake option to enable it for
    Solaris (report by Nelson Beebe).
  • modify in_put() function to call
    PreeditPosition() only when the cursor’s position
    has changed. This addresses report by Stefan Baums of high CPU
    usage while xterm with chinput are running in the background
    (patch by Yong Li).
  • improved explanation of -ls conflict with
    -e option in xterm manpage (adapted from comments
    by Henning Makholm).
  • correct comment in terminfo file regarding modifier used
    for kDC (Debian #189764,
    report by Henning Makholm).
  • correct/extend some of the keypad description in ctlseqs.ms
    (report by Henning Makholm).
  • correct keypad-mapping table in input.c so
    XK_KP_Equal works (report by Henning
  • modified to work with CJK double-width (bi-width/monospace)
    fonts. They’re similar to Roman Czbora and David Starner’s
    gnuunifonts in that Latin letters and numbers have all the same
    width which is exactly the half of the width of East Asian
    characters. (patch by Jungshik Shin
  • add configure option --enable-broken-osc and
    resource brokenLinuxOSC to accommodate scripts
    which do not distinguish between running in the Linux console
    and running in X. Linux console recognizes malformed control
    strings which start with an OSC, but are
    fixed-length, with no terminator.
  • add configure option --enable-broken-st and
    resource brokenStringTerm to allow user to revert
    one part of the parsing table corrections from patch #171. (reports by Matthias Scheler and
    Kirill Ponomarew indicate that someone’s network firmware sends
    an <escape>X).
  • modify configure --disable-imake to provide
    values for OSMAJORVERSION,
    NARROWPROTO (report by Heiko Schlichting).
  • correct vttests/16colors.sh, which omitted the
    $SUF variable in output strings (patch by Paul
  • modify shell scripts in vttests directory to
    attempt to use named signals in the trap
    statement, making this portable to OS/390 (report by Paul
    Gilmartin). Tested on SunOS 4.1.4, which implements only
  • fix definition of USE_HANDSHAKE, must be
    numeric (reported by Jens Schleusener).
  • add configure option --enable-pty-handshake to
    allow one to compile-in support for the pty handshaking logic,
    and resource ptyHandshake to enable or disable it
    (suggested by Ian Collier).
  • restore USE_HANDSHAKE ifdef removed in
    patch #159 to address Debian #39964.
  • move ifdef’s for ttysize/winsize into xterm_io.h
  • simplify loop on tgetent, check if the
    successive entries in the lookup table are the same. That makes
    xterm do one less lookup if there is no “xterm” entry in the
    termcap file (report by Derek Martin).
  • correct manpage discussion of $TERM and
    $TERMCAP variables, which omitted the Tektronix
    emulation and a note of the final “dumb” fallback.
  • broaden ifdef for
    XRegisterIMInstantiateCallback() from patch #175 (XIM fix) to exclude non-XFree86
    (report by Nelson Beebe indicates this breaks for Solaris, IRIX
    and OSF/1).
  • correct resource-size for iconFont (this was
    added by X11R6.3).
  • improve configure check for XKB bell extension, to work
    around inconsistent implementation of this feature (reports by
    Nelson Beebe, Kriston Rehberg and David Ellement).
  • modify configure-check for tty group to be
    less strict in batch mode (report by Nelson Beebe).
  • modify to allow building with g++, to use its compiler
    warnings (suggested by Nelson Beebe).
  • modify dec2ucs[] table to reflect newer codes
    available for scanlines 1, 3, 7, 9 (report by Michael
  • add configure option for XawPlus library.
  • fix a SIGSEGV which could occur if xterm is connecting to
    XIM server, and the XIM server is destroyed (patch by Nam
  • modify to use built-in line-drawing characters for Xft
    fonts (patch by Andrew Tipton).
  • make menu reflect the state of the tekInhibit
  • make signalInhibit resource work, i.e.,
    disable the menu entries that would send signals to, or exit
    xterm. This was probably broken in X11R5 when logging was
    disabled (report by Sven Mascheck).
  • changed classes of colorBDMode and similar
    resources that override colors when a video attribute is set to
    ColorAttrMode, to make them distinct from
    ColorMode. This avoids an unexpected rendering of
    reverse video, for example (report by Paul Fox).
  • changed class of veryBoldColors to
    VeryBoldColors, since ColorMode is
    associated with boolean resources.
  • add option -k8 and resource
    allowC1Printable to allow users of non-VTxxx
    character sets such as KOI-8 to treat the C1 control area
    (character codes 128-159) as printable rather than control
  • add a null-pointer check for return-value of
    ptsname() in HPUX-specific code (report by David
  • revise the ifdef’s used for XKB bell support.
    The code was using a nonstandard call
    XkbStdBell(). Changed to use
    XkbBell() (based on patch by
  • add a null-pointer check in xtermLoadFont() in
    case there is no wide-bold font (Nam SungHyun).
  • change Makefile.in to use autoconf’s
    bindir, libdir and
    mandir variables (report by Nam SungHyun).
  • add le to termcap xterm-basic
    entry. Though missing from older termcaps for xterm, some
    applications check for it (report by Matthias Buelow).
  • modify uxterm script to strip modifiers such
    as “@euro” from the locale setting before adding “.UTF-8”
    (Debian #179929).
  • modify the remaining places where tek4014 emulation uses
    XDefineCursor(), to make it work as originally
    implemented, e.g., when switching back to alpha mode. Added
    test-screen in vttest to test this feature properly.
  • work-around for XFree86 bug which made
    XDefineCursor() on a shell-window no longer work.
    The tek4014 emulation used this. Use the next lower window
    (report by Karl Rudolf Bauchspiess).
  • add a resource setting allowWindowOps to
    control whether the extended window operations should be
    allowed, e.g., resize, iconify, report window attributes. This
    is to accommodate people who are not capable of using a pager
    to view log-files.
  • reset mouse mode to normal on a full reset. This does not
    apply to mouse hilite tracking mode, of course (see
  • add a time-delay at the point where mouse hilite tracking
    mode choses to not handle X events, to avoid runaway CPU usage
    (report by D Hugh Redelmeier).
  • check for illegal character in DECUDK string, quit if
  • fixes to make repainting of 256-color example work properly
    (reports by Abigail Brady and Scott A Crosby).

    • set flag in AllocateAnsiColor() to ensure
      the color is allocated once only.
    • fix check in ScrnRefresh, which was
      comparing background colors only if the ANSI foreground
      colors also were set.
  • merge Error() calls and some
    exit() calls into SysError(), and
    change that to add the brief explanation for each error code
    which is provided in the manpage. Change a few
    SysError(), calls to avoid using code 1, to avoid
    confusion with exit status from places that do not use
  • simplify logic used to open a debug logfile as the standard
  • modify the -e option so that if it fails,
    xterm will check if only one argument follows, e.g., it was
    quoted, and then retry using sh -c.
  • modify parsing of DECUDK string parameter to allow a comma
    between pairs of hexadecimal digits (Ray Neuman
    <raymond@one.com.au> reports that “real” terminals accept
    this; perhaps they simply ignore unexpected characters).
  • fix a few problems with the $TERMCAP string
    generated by resize:

    • for Bourne shell, add an export command.
      This was missing as far back as X11R5.
    • escape exclamation marks, used in xterm’s reset
    • translate literal \177\ to “^?”.
  • improve configure check for tgetent() to work
    when $TERMCAP has been set to a specific
  • modify minstall.sh to use “%” rather than “@”,
    to avoid problems with AFS (report by Zdenek Sekera).
  • list fatal error codes from error.h in the
    manpage, remove unused codes in error.h.
  • use more explicit wording for manpage list of color
    resources, since at least one user confused the generic names
    such as “black” with the names in rgb.txt (the latter are now
  • minor fix to description of 1003 mouse mode in ctlseqs.ms
    (Larry Riedel).
  • modify parser tables to improve detection of malformed
    control sequences, making xterm behave more like a real DEC
    terminal (patch by Paul Williams).
  • update comment in input.c to document
    Meta as a modifier for escape sequences (patch by
    D Roland Walker).
  • add ifdef’d code for logging option which adds the hostname
    and a timestamp to the generated logfile name. The ifdef’s are
    setup with the configure script (patch by Nelson Beebe).
  • remove xevents() call from the end of
    BlinkCursor(), to fix an occasional problem which
    caused xterm to pause until a key was pressed. The reason for
    this was that there was no check to ensure that there really
    were events for xevents() to process. (patch by
    Semen A Ustimenko <semenu@FreeBSD.org>).
  • remove unused mode-params from open() calls
    that do not create a file.
  • modify configure script to put new items first on
    $CPPFLAGS and $CFLAGS to avoid
    conflict with environment’s -I and -D
  • update config.guess, config.sub
  • resync with XFree86 CVS
    • fix va_args glitches for xterm/libfontconfig: 0 ==
      (void*)0 isn’t true for all platforms (Egbert Eich).
    • initialise ProgramName in xterm’s main before
      referencing it (XFree86 #5473,
      Peter Valchev).
    • some cleanup of Imakefile ifdef’s (Marc La
  • correct an off-by-one allocating data for sorted help
  • modify configure script to check for Xpm library, on which
    XFree86 Xaw library depends.
  • update config.guess, config.sub
  • modify wording of some options in help message to make them
    use -/+ consistently with respect to “on/off” or
  • sort options list which is displayed in help- and
    syntax-messages at runtime to simplify maintenance.
  • remove support for Amoeba and Minix (Juliusz Chroboczek
    noted it was removed from XFree86 server; there have been no
    users since 1996).
  • add configure script option
    --disable-session-mgt to control whether the
    session management code should be compiled-in (request by H
    Merijn Brand).
  • ifdef’d the session-management changes with
    OPT_SESSION_MGT, to accommodate X11R5 which
    predates the related definitions.
  • fix decode_keyvalue(), which did not properly
    parse multiple settings as needed for the ttyModes
    resource, since it did not skip over the parsed data.
  • fix an option-parsing conflict between -class
    and -cjk_width (Nam SungHyun)
  • add a missing null in XtVaSetValues() call
    used in Cleanup() (Nam SungHyun)
  • improve data reported for control sequence that requests
    window position by taking into account the window decorations,
    e.g., border and title (patch by Jess Thrysoee
  • add -cjk_width and corresponding resource
    cjkWidth (patch by Jungshik Shin
  • add -into option, for embedding xterm in a
    Tcl/Tk application (patch by George Peter Staplin
  • add simple session management (XSM) client capabilities to
    xterm. So a session manager such as, e.g., xsm, should be able
    to respawn or kill xterms without the help of an SM proxy (like
    smproxy), which is a hack and almost always buggy (patch by
    David Madore).
  • fix conflict between ifdef’s for
    button.c (reported by Ilya Zakharevich).
  • fix for inconsistent use of struct utmp versus
    struct utmpx introduced in patch #167 (patches by Paul Gilmartin, Marc La
  • modify logic for metaSendsEscape to allow it
    to work in a setup where the Meta-key is not recognized as a
    modifier by the key-translations logic (discussion of
    eightBitInput with H.J.Lu). Note however that Vincent Lefèvre
    reported this in January).
  • use null pointer values consistently, rather than literal
    “0”, for ending variable-length argument lists, e.g., for
    execlp() (based on patch by Matthieu Herrb).
  • correct logic of ChangeAnsiColorRequest(),
    which would do a screen repaint after replying to a request for
  • improve scripts in vttests to work with systems whose
    shells support echo -n and have, as does Debian,
    an unrelated print utility.
  • add vttests/acolors.sh to demonstrate OSC 4, which queries
    or sets ANSI colors.
  • fixes for ctlseqs.ms (Pavel Roskin, Ilya Zakharevich).
  • typos in xterm manpage (Jens Schweikhardt)
  • remove a redundant GCC_UNUSED from InitPopup() (Nam
  • correct ifdef’s for USE_TERMCAP to match
    cygwin configuration.
  • fix several places in ctlseqs.ms which had no boxes around
    the literal text (report by Ilya Zakharevich).
  • extend mouse support for readline, ifdef’d with
    OPT_READLINE (integrated patch from Ilya Zakharevich).
  • modify terminfo description to match default for
    modifyCursorKeys resource.
  • add modifyCursorKeys resource to control how
    the shift- and similar modifiers are used to make a cursor
    escape sequence. The default makes a modified escape sequence
    always start with CSI and puts the modifier as the second
    parameter, to avoid confusing applications that would interpret
    the first parameter as a repeat count. The original behavior
    can be obtained by setting the resource to 0 (newsgroup
    discussion with Stephen J Turnbull, Jeffrey Altman).
  • correct missing initializations for
    appdefaultCursor and appdefaultKeypad
  • add configure option --enable-luit and ifdef’d
    the luit-related code with OPT_LUIT_PROG.
  • modify xterm to invoke luit (integrated patch by Tomohiro
  • update wcwidth.c to match Markus Kuhn’s 2002-05-18
  • correct limit-checking in ComputeSelect() to
    handle selections that extend off the visible area; rather than
    modify the parameters to TrackText(), use
    ScrollSelection() to update the highlighting
    limits. (reported by Yegappan Lakshmanan and Nelson Beebe,
    patch by Alexander V Lukyanov).
  • correct manpage description of tiXtraScroll
    resource (reported by Tony Finch).
  • changes from OpenBSD:
    • Make xterm setgid utmp to be able to update utmp even
      with root privileges revoked.
    • If not updating utmp, revoke group privileges totally
  • changes from NetBSD:
    • Check that the return value from ttyslot() is greater
      than 0 before writing the utmp file. Fixes w’s “w: Stale
      utmp entry: <user> <tty> <pty>”
    • Use openpty() to deal with new pty naming scheme.
  • add print-redir action and menu entry to allow
    user to switch terminal in/out of printer controller mode. It
    appears from the manual that a real vt220 would not switch back
    to normal mode, so another mechanism is used (addresses Debian
  • check if printerCommand resource string is
    empty, use this to allow user to disable printer function.
  • trim trailing blanks from resource strings.
  • check return value from ptsname(), which may
    return null for example if someone has changed the permissions
    of /dev/pts to zero (Debian #121899).
  • modify OS/2 version to use __UNIXOS2__
    definition rather than __EMX__, related cleanup
    (patch by Holger Veit).
  • used modified indent 2.0 (patch 20020428) to
    reformat most of the C source files, to simplify
  • fix a couple of places where there were leading tabs on
    symbol-definition lines in Imakefile (Marc La France from
    report by Tony Finch)
  • add imake variable (TraceXTerm) to allow building debug
    version (Egbert Eich).
  • correct a bug in selection: double clicking on a word which
    was partly scrolled off the screen may select that text (report
    by Vincent Lefèvre <vincent@vinc17.org>).
  • implement veryBoldColors resource to control
    whether the corresponding video attribute such as
    bold is displayed when using
    colorBDMode, etc. (request by Josh Howard
  • define escape sequences for function keys F21-F35 (patch by
    Stephen P Wall).
  • change the colors for the 256-color model, making them less
    skewed toward black (patch by Stephen P Wall).
  • add vt100Graphics resource (see patch #115, based on discussion with Glenn
  • ifdef’d Xaw/Xaw3d/neXtaw includes separately to avoid
    potential incompatibilities between these flavors of Athena
  • add configure check for XFree86 4.x Xaw library, whose
    geometry management is broken, to avoid trying to use it for
    toolbar configuration.
  • updates to configure script from vile and lynx to allow
    configure.in to be compiled with autoconf 2.5x
  • add a check for null pointer return by
    ptsname() (newsgroup posting from Mike Silva
  • fill in a few details needed to allow UTF-8 mode to switch
    on/off after startup. This requires that wideChars
    resource be set. (based on comments in 4 Aug 2001 by Alexey
    Marinichev <lyosha@lyosha.2y.net>).
  • remove duplicate install rules that make directories, e.g.,
    so installing manpage will not create app-defaults
  • add comment in in do_osc(), reserving cases 30
    and 31 for for Konsole (request by Stephan Binner
  • fixes for ctlseqs.ms and xterm.man (patch by Werner Lemberg
  • add check for monochrome display, disabling colorMode in
    that case (fixes Debian #134130).
  • resync with XFree86 CVS
    • Fix some xterm build warnings on *BSD (patch by David
    • Only use SA_RESTART in xterm when it’s available (patch
      by Frank Liu).
    • Fix incorrect code in signal handlers in most of the
      clients, xterm and xdm not done yet (patch by Matthieu
  • modify uxterm script to strip encoding part
    from environment variable before adding “.UTF-8” (based on
    Debian #125947,
    but using a more portable solution).
  • add an assignment statement in VTInitialize() to make
    awaitInput resource work.
  • use new macros init_Bres(), etc., in VTInitialize() to add
    trace of the initialization of resources.
  • modify checks for $LC_ALL, related environment variables to
    ensure the resulting strings are nonempty (report by Markus
  • add an ifdef in charproc.c for num_ptrs variable in case
    all configure options are disabled.
  • modify definition of getXtermBackground() to avoid negative
    array index warning on Tru64 (report by Jeremie Petit).
  • improve fix from patch #165 (still Debian #117184,
    report by Matt Zimmerman <mdz@debian.org>)
  • correct install rule for uxterm in Makefile.in, to handle
    the case where building in a different directory than xterm’s
    source (patch by Paul Gilmartin).
  • documented ANSI.SYS-style cursor save/restore escape
    sequences in ctlseqs.ms, which are in xterm since X11R5.
  • correct two entries in the default charClass table, which
    did not follow the manpage comment about the character number
    corresponding to the class (patch by Marc Bevand
  • fix a couple of typos in comments in the app-defaults files
    (David Krause <xfree86@davidkrause.com>).
  • resync with XFree86 CVS
    • update language of copyrights in some files to reflect
      the fact that they were reassigned from X Consortium to The
      Open Group in 1998. Note that this xterm source is derived
      from the 1996 version from X Consortium, does not
      incorporate changes made by X Consortium or The Open Group
      after that date, hence we do not add The Open Group’s 1998
      copyright date to related files.
    • save/restore errno in signal catcher
      (patch by Matthieu Herrb).
    • modify UXTerm.ad’s font5 resource so that xterm can
      display double width characters using a font distributed
      with XFree86 (Tomohiro Kubota).
  • correct a case where ptyInitialErase and
    backarrowKeyIsErase resources combine to set
    DECBKM mode, but a reset command
    would not reset xterm to that state, making the erase character
    revert to ^H (Debian #117184)
  • correct ifdef’s for __QNX__ and
    USE_SYSV_PGRP in main.c call to
    tcsetpgrp, which broke bash behavior around
    patch #140 (report/patch by Frank Liu
  • modify trace.c to fix missing definition of GCC_UNUSED for
    compilers other than gcc when configured for trace code
    (report/patch by Paul Gilmartin).
  • change format in TraceOptions() to use long rather than
    int, since the latter could lose precision on 64-bit machines
    (report/patch by Nelson Beebe).
  • modify xterm manual page and minstall.sh to allow imake
    rules to define location of app-defaults directory (Debian
  • review/update list of conflicting preprocessor symbols to
    remove from $CPPFLAGS at the end of the configure script. In
    particular, this allows one to configure xterm without the
    utempter library on Redhat 7.1 (report/patch by Adam
  • correct logic that processes -class option, so
    that a following -e option is handled (Debian
  • improve options-decoding to allow -version and
    -help options to be combined (Debian #110226).
  • add a 10 millisecond delay in event loop when processing
    -hold option, to avoid using too much CPU time
    (Debian #116213).
  • prefix final program execution in uxterm with
    “exec” to avoid a useless shell hanging around (Christian
  • modify logic that resets keypad application mode to avoid
    doing this if there is no modifier associated with the Num_Lock
    keysym (report by John E Davis <davis@space.mit.edu> and
    Alan W Irwin <irwin@beluga.phys.uvic.ca>).
  • add built-in translation for Control/KP_Separator to
    KP_Subtract, to accommodate users who wish to use xmodmap to
    reassign the top row of the numeric keypad.
  • correct Imakefile install-rule for uxname
    script (reported by Nam SungHyun <namsh@lge.com>).
  • resync with XFree86 CVS – correct typo in
    <ncurses/term.h> ifdef.
  • modify logic in main.c (see patch
    #145) to avoid generating the same identifier, for example,
    for /dev/tty1 and /dev/pts/1, which is used to denote an entry
    in the utmp file (Debian bug report #84676),
    A similar fix was also sent by Jerome Borsboom
    <borsboom@westbrabant.net> in May, but I overlooked it
    when reviewing bug reports.
  • add configure check for <ncurses/term.h>
    to get rid of hardcoded __CYGWIN__ ifdef in

    NOTE: The CYGWIN port should not be linking
    resize with ncurses. It appears that the only
    reason it is, is because both the ncurses and termcap ports
    on that platform are badly misconfigured (essential pieces
    have been removed, etc). Checking for
    <ncurses/term.h> does not hurt anything,
    since there are some correct installations that are set up
    that way.

  • add a sample uxterm script, which uses the
    UXTerm application defaults for UTF-8
  • undo change to xterm application defaults, since this
    introduced an unnecessary incompatibility. The intended
    functionality was already addressed by the UXTerm app-defaults file.
  • resync with XFree86 CVS
    • Modified xterm app default to use LFD fontnames instead
      of old type (Michael Schroeder).
    • Fix xterm when XIM is disabled – caused a segfault
      (Tomohiro Kubota).
  • remove an ifdef for USE_HANDSHAKE added in
    patch #158 from the second TIOCSSIZE ioctl call in
    the initialization code. Paul Gilmartin reports that for
    Solaris 2.6 (sparc), stty does not show that xterm sets rows
    and columns unless the second ioctl is executed.
  • correct treatment of empty parameter list for some OSC
    strings (report by Sami Farin <sfarin@ratol.fi>).
  • augment key translations for scrollbar widget to ensure
    that keystrokes intended for the text area are not lost if the
    mouse pointer happens to fall on the scrollbar. This can
    happen, for instance, if the user’s app-defaults
    or .Xdefaults file contains a translations
    resource, though it may also happen through unrelated resource
    settings: probably a bug in libXt (reported by Paul Fox
    <pgf@foxharp.boston.ma.us> and Dmitry Yu. Bolkhovityanov
  • widen ifdef’s in xterm.h to define
    __EXTENSIONS__ on Solaris, for compilers other
    than gcc. This was added along _POSIX_C_SOURCE
    with in patch #151, but is not
    sufficient since Sun’s compiler does not define
    __STDC__ by default (report by Matthias Scheler
  • modify xterm to obtain the closest matching color if an
    exact color is not available, e.g., on 8-bit displays (patch by
    Stephen P Wall)
  • implement “OverTheSpot” preedit type of XIM input (patch by
    Tomohiro Kubota):

    • This preedit type is a reasonable compromise between
      simpleness of implementation and usefulness. “Root” preedit
      type is easy to implement but not useful; “OnTheSpot”
      preedit type is the best in the point of view of user
      interface but very complex. And more, “OverTheSpot” is the
      preedit type which the most XIM servers support. Thus, to
      achieve reasonable usability, support of “OverTheSpot”
      preedit type is needed.
    • In “OverTheSpot” preedit type, the preedit
      (preconversion) string is displayed at the position of the
      cursor. On the other hand, it is XIM server’s
      responsibility to display the preedit string. Thus, it is
      needed for XIM client (here XTerm) to inform the XIM server
      of the cursor position. And more, to achieve good visual
      proportion, preedit string must be written using proper
      font. Thus, it is XIM client’s responsibility to inform the
      XIM server of the proper font. The font must be supplied by
      “fontset”. Fontset is a set of fonts with charsets which
      are specified by the current locale. Since XTerm uses
      ISO10646 fonts regardless of the current locale, the fonts
      for XIM must be prepared separately. It is difficult to
      prepare fonts which are similar to XTerm font. Thus, my
      patch uses a simple way – the default font is “*” which
      matches every fonts and X library will automatically choose
      fonts with proper charsets. I added “-fx” command option
      and “ximFont” resource to override this default font
    • changed the definition of “OverTheSpot” preedit type
      from XIMPreeditPosition|XIMStatusArea to
      XIMPreeditPosition|XIMStatusNothing. This matches the
      behavior of other programs such as Rxvt, Kterm, Gedit.
    • A tiny XIM bugfix is also included. By calling
      XSetLocaleModifiers() with parameter of “”, it can consider
      XMODIFIERS environmental variable which is a standard way
      for users to specify XIM server to be used.
  • adjust configure script to accommodate repackaging of
    keysym2ucs.c as include-file for xutf8.c (patch by Tomohiro
  • add the UXTerm app-defaults file to makefile
    install rules (request by Juliusz Chroboczek).
  • correct logic in get_termcap(), which returned
    false if the tgetent() call indicated that xterm
    was linked with terminfo. Although the $TERMCAP
    variable cannot be adjusted in this case, xterm still needs
    information from this call to extract data to initialize the
    erase-mode when the ptyInitialErase resource is
  • modify ScrollbarReverseVideo() function to
    cache the original border color, so it can restore that when an
    application flashes the screen (Bugzilla #38872).
  • resync with XFree86 CVS
    • remove ifdef’s for X_NOT_STDC_ENV (patch by David
    • add Cygwin to special errno handling case in ptydata.c
      (Alan Hourihane)
    • set screen size earlier in initialization, to address a
      race with window manager resizing its clients (Keith
  • clean up button.c and input.c (patch by Juliusz
    Chroboczek). The basic idea is to use a single set of APIs in
    the XTerm core, and put a set of workarounds in a separate
    file. As you will notice, this drastically simplifies parts of
    the code; in particular, the only remaining ifdefs related to
    the differences between Xutf8 and legacy systems are related to
    the selection provider logic. This defines two new files,
    xutf8.h and xutf8.c, provide some half-hearted but fully
    portable emulation for the three Xutf8* functions that XTerm
    uses. Putting these in a separate file will allow people to
    experiment with more complex versions without making the core
    of XTerm more difficult to maintain. The functionality of the
    emulation is as follows.

    • Xutf8TextListToTextProperty fully handles XStringstyle
      and XUTF8StringStyle. It will only generate STRING for
      XStdICCTextStyle, and will only generate Latin-1 in
      XCompoundTextStyle (but label it as COMPOUND_TEXT, as the
      spec requires). (I have hesitated to make this function
      fail for XStdICCTextStyle and XCompoundTextStyle; this
      might be a smart thing to do, in the hope that the
      selection requestor will try UTF8_STRING afterwards;
    • Xutf8TextPropertyToTextList fully handles STRING and
      UTF8_STRING. It systematically fails for
    • Xutf8LookupString will properly handle single-keystroke
      input for the keysyms covered by Markus’ keysym2ucs
      function. It will not handle either compose key input or
      external input methods.
  • add an ifdef for SunXK_F36, used in
    xtermcapKeycode() (patch by Mark Waggoner
  • modify check for login name to try getlogin()
    and $LOGNAME, $USER environment variables to detect if the user
    has logged in under an alias, i.e., an account with the same
    user id but a different name (Debian bug report #70084).
  • add resource tiXtraScroll, which can be used
    to preserve the screen contents in the scrollback rather than
    erasing it when starting a fullscreen application such as
    vi (patch by Ken Martin
  • two patches by Denis Zaitsev <zzz@cd-club.ru>:
    • added forceBoxChars resource to control
      the “line-drawing characters” option.
    • added freeBoldBox resource, which, when
      set true, suppresses check in
      same_font_size(), so xterm does not attempt to
      ensure that the bold font is the same size as the normal
  • change the color class for ANSI colors and similar ones
    such as bold-color, etc., to follow the convention that the
    classname is the instance name with a leading capital (e.g.,
    color0 and Color0 rather than
    color0 and Foreground). There was
    little use for Foreground as a class other than to create
    occasional confusion and bug reports, most recently by Nelson
    Beebe who reports that it disables colors in xterm but not on
    Linux, which is probably due to a difference in resource
    evaluation order. A configure script option
    (–disable-color-class) is provided for anyone who did use the
    older behavior.
  • fix a signal-handling bug. When running xterm with
    ksh or similar shell such as bash 2.05, xterm will
    hang and not respond to keystrokes after the user types
    suspend. Additionally the popup menus do not
    respond, so it is not possible to send SIGCONT to the shell.
    What is happening is that the shell is sending itself SIGSTOP,
    and it is being stopped. Per specification it also means that
    the operating system is sending SIGCHLD to parent to the bash
    (XTERM). Now the issue is that xterm assumes it cannot happen
    and it expects to receive SIGCHLD only upon
    termination of its children. This causes a deadlock
    with xterm waiting for the child to die, and the child waiting
    for SIGCONT. (report/patch by Adam Sulmicki), analysis/testing
    by Sven Mascheck
  • correct some inconsistent checks for
    XtReleaseGC() calls in
    xtermLoadFont() (patch by Nam SungHyun
  • updated wcwidth.c and keysym2ucs.c to versions dated
    2001/1/12 and 2001/4/18, respectively from
  • correction to change from patch
    #90, which was intended to reset the saved cursor position
    for normal/alternate screens on a soft reset, but actually
    wiped out all of the saved cursor information. Just reset the
    saved position for the current screen (report by Michael
  • modify ShowCursor() and HideCursor() so that if the cursor
    points to the second part of a wide character, make the
    show/hide operate on the first position of the wide character
    (patch by Sven Verdoolaege).
  • scrolling in the alternate screen just before switching
    back to the main screen can cause extra blank lines to be
    inserted into the scrollback buffer (patch by Paul Vojta
  • resync with XFree86 4.1.0:
    • Use TermcapLibrary as -lncurses instead of -ltermcap,
      fixes problem building xterm/resize on Cygwin/XFree86
      (Harold Hunt).
    • Install xterm.termcap and xterm.terminfo when
      installing xterm (Torrey T. Lyons).
    • Fix some build issues on Cygwin/XFree86 (Suhaib
    • Define CBAUD, when it’s missing in xterm, on LynxOS
      (Stuart Lissaman).
  • change order of selection-target types to make 8-bit xterm
    prefer UTF8_STRING to COMPOUND_TEXT (patch by Juliusz
  • document -fa, -fs command-line
    options and faceName, faceSize
    resources which are used by the freetype library support.
  • if configure script finds freetype libraries, but imake
    definitions do not have the XRENDERFONT definition, define it
  • modify configure script check for freetype libraries to
    include <Xlib.h>, since an older version of
    the related headers relies on this (patch by Adam
  • correct return type of in_put() from patch #153 changes, which left it not wide
    enough for UTF-8 (patch by Bruno Haible).
  • undo check for return value from pututline
    (used for debugging) since that function does not return a
    value on Slackware 3.6.
  • correct length in ScreenWrite, when rendering invisible
    text (patch by Sven Verdoolaege <skimo@kotnet.org>).
  • fixes/improvements for the i18nSelection
    resource from patch #153, by Bruno Haible:

    • add missing initialization for
      i18nSelection resource.
    • split-out the non-ICCM aspect of the
      i18nSelection resource as a new resource,
  • add configure check for <time.h> and
    <sys/time.h>, to allow for Unixware 7,
    which requires both. If the configure script is not used, only
    <time.h> will be included as before (report
    by Thanh Ma).
  • fix redefinition of dup2,
    getutent, getutid,
    getutline and sleep functions on
    Unixware 7 (report by Thanh Ma
  • add a fall-back definition for __hpux, which
    is apparently not defined in some compilers on HPUX 11.0
    (reported by Clint Olsen).
  • change VAL_INITIAL_ERASE, which is used as a fallback for
    the "kb" termcap string to 8, since that matches
    the xterm terminal description (request by Alexander V
  • correct an off-by-one in ClearInLine, which
    caused the erase-characters (ECH) control to display
    incorrectly (patch by Alexander V Lukyanov).
  • correct escape sequences shown in terminfo for shifted
    editing keys. The modifier code was for the control key rather
    than the shift key.
  • increase PTYCHARLEN to 8 for os390, add some debugging
    traces for UTMP functions (patch by Paul Gilmartin).
  • correct an misplaced brace in
    SelectionReceived() (patch by Bruno Haible).
  • correct an assignment dropped in
    SelectionReceived() which made it not compile for
    wide-characters combined with debug traces.
  • correct typo, clarify description of 88- and 256-color
    controls (report by Bram Moolenaar).
  • correct a typo in ctlseqs.ms which caused DEC Locator
    control sequences using single quote “‘” display grave “`”
    instead (reported by Paul Williams, apparently in patch
  • correct index in inner loop in VTInitI18N()
    from patch #151 changes, which resulted in infinite loop under
    some conditions (report/analysis by Paul Gilmartin).
  • remove spurious “%|” from terminfo sgr capabilities
    (report/analysis by Adam M Costello, Debian #89222).
  • add shell script to adjust list of dynamic libraries linked
    by resize when building with the standalone
    configure script. Otherwise it uses the same list as for xterm,
    which is excessive.
  • fix a few compiler warnings reported by the 20010305 gcc
  • alter the behaviour of selections in XTerm (patch by
    Juliusz Chroboczek). It is believed to follow the ICCCM +
    UTF8_STRING to the letter, both in UTF-8 and in eight-bit mode.
    From his description:

    • When compiled against XFree86 4.0.2, the patched XTerm
      will make the selection available as COMPOUND_TEXT, STRING
      or UTF8_STRING in both modes. It will request selections in
      the following order:

      eight-bit mode: TEXT, COMPOUND_TEXT, UTF8_STRING,

    • When compiled against an earlier version of XFree86 (or
      compiled with OPT_UTF8_API=0), it will still obey the
      ICCCM, but will neither request nor provide UTF8_STRING in
      eight-bit mode, and neither TEXT nor COMPOUND_TEXT in UTF-8
    • For compatibility with previous versions of XTerm, a
      resource i18nSelections is provided, please
      see the manpage for details. However, due to an unexpected
      combination of ICCCM extensions by various bits and pieces
      of the libraries, interacting with previous versions of
      XTerm will work in many locales even without this flag.
      (Please do not set this resource to true by default, as
      this would violate the ICCCM.)
  • improved error checking/reporting in VTInitI18N(), ensuring
    that lack of input method styles is treated distinctly from a
  • remove an incorrect ifdef from patch #141 which suppressed
    overstriking to simulate bold font when xterm was compiled to
    support wide characters (report/analysis by Adam M Costello
    <amc@cs.berkeley.edu>, Debian #76404,
    Debian #77575).
  • modify RequestResize() function to
    save/restore window manager hints. Its call to
    XtMakeResizeRequest() had the undesirable
    side-effect of clearing window manager hints, e.g., when
    switching to 132-columns via DECCOLM escape sequence. Window
    manager hints make it simple to resize xterm in terms of
    character cells rather than pixels (reports by Christian
    Weisgerber, Debian #79939).

    This applies to the normal configuration. When built with
    toolbar support, the hints are applied to a different widget
    level (more work will be needed to make hints work with the

  • fix a redefinition warning for resize.c on OpenBSD (patch
    by Christian Weisgerber).
  • change resource corresponding to -T option to
    match Xt library’s -title, i.e.,
    .title rather than *title so the
    command-line options are interchangeable as documented (Debian
    bug report #68843).
  • add script used from autoconf’d makefile for installing
    manpages, since recent XFree86 changing the way section numbers
    are represented makes the install dependent on extra
  • add configure check for freetype libraries and configure
    option –disable-freetype to override this feature.
  • modify some configure script macros to avoid using
    changequote(), which has been rendered useless in the latest
    autoconf alpha 2.49c
  • update config.guess, config.sub to 2001-2-13
  • remove redundant/contradictory __CYGWIN__ definitions from
  • correct manpage typo introduced by X11R6.5 resync.

Changes from XFree86 4.0.2:

  • add definition of _POSIX_C_SOURCE for Solaris
    to make this compile with gcc -ansi -pedantic
    (report by <mark@zang.com>.
  • move the binding for shifted keypad plus/minus, which
    invokes the larger-vt-font() and
    smaller-vt-font() actions, respectively, into the
    translations resource (suggested by Marius Gedminas
  • modify configure script to support the
    --program-prefix, --program-suffix
    and --program-transform-name options (request by
    Alison Winters <alison@mirrabooka.com>).
  • fixes for wide/combining characters (Robert Brady):
    • fix a scrolling / combining characters display
    • fix a problem with double-width characters where if the
      primary font had no box-drawing characters, the right hand
      half of double-width characters was erased (reported by Yao
      Zhang <yzhang@sharemedia.com>).
    • fix special case of null byte for key handling in UTF-8
  • modify logic that compares sizes of normal and bold fonts
    to be more forgiving of the font server’s choice of bold font
    which must match the normal font’s size. Now
    same_font_size() compares the height of the fonts
    rather than individually ascent and descent, and allows the
    bold font to be one pixel smaller than the normal font
    (addresses a report by Alan Citterman <alan@mticket.com>,
    who says that something in patches #146 to #148 made
    xterm more likely to overstrike bold fonts, and indirectly
    Debian bug report #76404,
    which reports the opposite).
  • make configure script use $CFLAGS and $CPPFLAGS
    consistently, including removing a chunk from configure.in
    which attempted to save/restore $CPPFLAGS while processing
    value set by the –x-includes option, but lost values set in an
    intervening AC_CHECK_HEADERS. This change modifies macros
    removing variables IMAKE_CFLAGS and X_CFLAGS from the generated
    makefile (the AC_CHECK_HEADERS problem was reported by Albert
    Chin-A-Young <china@thewrittenword.com<).
  • correct a comparison in SELECTWORD case of ComputeSelect(),
    which resulted in a word-selection wrapping past the first
    column without checking the first column’s character class
    (reported by Christian Lacunza
  • correct a logic in UTF-8 mode for selecting double-width
    characters; a combining character was omitted (patch by Markus
  • add feature to pop (raise) window when a bell is received
    (patch by Gael Roualland
  • add __NetBSD__ and __OpenBSD__ to special-case in
    xterm_io.h for USE_POSIX_TERMIOS definition (patch by Christian
  • move special-case HPUX include for <sys/bsdtty.h> to
    xterm_io.h to define TIOCSLTC, making HAS_LTCHARS defined for
    HPUX 10.20 (report by Bruno Betro).
  • restructured includes for termios.h,
    termio.h and related definitions for main.c,
    os2main.c, screen.c and resize.c so they will share equivalent
    definitions in a new header xterm_io.h. This is
    intended to solve some problems mainly for HPUX which appear to
    arise from inconsistent definitions for SIGWINCH- and
    HAS_LTCHARS-related symbols (reports by Bruno Betro, Jeremie
    Petit and Clint Olsen).
  • improve usability of double-width fonts by allowing normal
    fonts to be given as double-width (from a patch by Fabrice
    Bellard <bellard@email.enst.fr>).
  • correct a few compiler warnings in TRACE() macros for
    signed/unsigned variable differences (reported by Clint
  • make configure script use $CFLAGS and $CPPFLAGS more
    consistently, i.e., by using CF_ADD_CFLAGS in CF_ANSI_CC
  • expanded description of environment variables in
  • modify OPT_TCAP_QUERY feature to always return the termcap
    or terminfo capability string in the response, and to
    read/write the names in hexadecimal form to allow for the
    special case of termcap’s k; name (patch by Bram
  • add OPT_SAME_NAME and OPT_TCAP_QUERY to xtermcfg.hin, so
    the corresponding configure options work (patch by Bram
  • resync with XFree86 4.0.1g:
    • Rewrite Xft library for Render extension/core text and
      font management Change xterm to use new interface (Keith
  • document logfile options in man-page.
  • correct spelling of -samename option in help
  • add configure script option
    --enable-tcap-query (request by Bram
  • add a “Co” or “colors” entry to the OPT_TCAP_QUERY feature
    (patch by Bram Moolenaar).
  • patch by Kiyokazu Suto <suto@ks-and-ks.ne.jp>:
    • add support for bold font for double width characters.
      The font name may be specified with the command line option
      -fwb or with resource
      wideBoldFont (class
    • correct underlining of double width character string,
      which was drawn only half width.
    • correct binary search of precomposed character table,
      which may return wrong result when int is just
      32 bits.
  • some changes to align terminfo with ncurses 5.2:
    • remove xtermm description, retaining xterm-mono since
      the former conflicts with ncurses.
    • modify initialization and reset strings to avoid
      putting the save/restore cursor operations bracketing
      changes to video attributes, since the changes could be
      lost when the cursor is restored. This affects xterm-r6 and
      xterm-8bit (the xterm-xfree86 entry uses the soft-reset
      feature which resets scrolling margins and origin mode
      without requiring us to save/restore the cursor
    • make a few entries explicitly inherit from
      xterm-xfree86 rather than xterm: xterm-rep, xterm-xmc,
  • ensure that sign-extension does not affect
    ctype macros by using CharOf() macro
    to coerce the parameter to an unsigned char.
  • resync with XFree86 4.0.1d:
    • Add primitive support in xterm for Xft based fonts
      (Keith Packard). The changes are ifdef’d with
  • correct implementation of ptyInitialErase: the
    value assigned to initial_erase was for the
    control terminal, which is correct as far as it goes. But there
    was no following test for the pseudo-terminal’s erase value,
    which would overwrite the default obtained from the control
    terminal (reported by Christian Weisgerber
  • modify check for printable-characters in OSC string to use
    xterm’s ansi_table rather than isprint() macro, to isolate this
    check from locale settings. This fixes a problem using 8-bit
    characters to set the title (reported by Ricardas Cepas
  • modify sample scripts to check for printf before print,
    since the latter is not as well standardized (e.g., on
  • updated config.sub, config.guess to reflect changes on
  • correct two instances overlooked from patch #141 which
    assumed UTF-8 mode without checking, causing a core dump in
    non-UTF-8 mode (one instance was reported by Tommi Virtanen
  • correct a problem selecting from the scrollback buffer in
    UTF-8 mode by changing remaining getXtermCell() calls to
    XTERM_CELL() as in the non-UTF-8 cases (report by Markus Kuhn,
    patch by Robert Brady).
  • changes for UTF-8 configuration (Robert Brady):
    • doublewide characters don’t lose their accents when the
      cursor moves onto or from them (a visual bug)
    • fix logic in addXtermCombining, which mean that if the
      low byte of a cell’s first combining char was 0, a
      following combining char would go into combining slot 1,
      not 2.
    • modify logic for cut-buffers so UTF-8 data is first
      converted to Latin1.
    • collapse surrogates, 0xfffe, 0xffff to UCS_REPL.
    • modify to allow xterm to to show combining characters
      attached to doublewidth characters.
    • correct bug in linewrap with -u8 option (reported by
      Andreas Koenig <andreas.koenig@anima.de>).
  • several changes to PTY logic (based on request by Tim Ryan

    • modify treatment of -S option to to make
      it work with Unix98 PTY’s.
    • restore sense of IsPts flag in get_pty(),
      which was lost in Unix98 changes.
    • use new functions my_pty_id() and
      my_pty_name() to simplify/fix strings used for
      utmp, wtmp identifiers.
    • simplify get_pty() function, making it
      have a single return point so its inputs/outputs can be
  • update config.guess and config.sub and scripts to my
    20000819 patch, adding cases for OS/2 EMX.
  • add special case for os390 compiler options to configure
    script (patch by Paul Gilmartin)
  • fix some unused-variable compiler warnings (reported by
    Zdenek Sekera).
  • split-out some string functions into xstrings.c, to use
    them more consistently among main.c, os2main.c and
  • align termcap/terminfo files, adding entries to make them
    match. The termcap entries are necessarily less complete than
    the terminfo, to fit within 1023 character per entry.
  • add terminfo entry for xterm-sco (SCO function keys).
  • modify same_font_name() to properly handle
    wildcard introduced in bold_font_name(), making
    comparison for different fonts succeed when only the normal
    font is specified. This is needed to decide if 1-pixel offset
    should be used. (reported in a newsgroup by Bart Oldeman
  • correct preprocessor line for OPT_WIDE_CHARS in
    drawXtermText() from patch #141 which resulted in overstriking
    for bold fonts not working.
  • correct Imakefile from 4.0.1c resync so UTF-8 modules are
    in UTF8SRC, UTF8OBJ lists, allowing build without UTF-8
  • remove a spurious assignment in ScreenWrite() from Robert
    Brady’s patch which set a null at the “end” of the buffer to be
    written. That made the autowrap feature write a blank in the
    first column for the non-UTF-8 configuration, rather than the
    actual character (reported by Alan Citterman
  • add a check to ensure that -class command-line
    option is not confused with -c (reported by Paul
    Townsend <aab@aab.cc.purdue.edu>).
  • correction to precompose scripts, so 0061 + 0300 will now
    be really be displayed as 00C0 (patch by Robert Brady
  • correct macro ClassSelects() in button.c, used to hide
    ifdef’s for OPT_WIDE_CHARS in patch #141 (reported by Andreas
    Paul <paula@informatik.tu-muenchen.de>).
  • change wcwidth.h to include stddef.h rather than wchar.h,
    which is not present on OpenBSD and FreeBSD (reported by
    Christian Weisgerber and Bram Moolenaar).
  • newer config.sub and config.guess, from lynx
  • changes for doublewidth and combining characters, from
    http://www.ecs.soton.ac.uk/~rwb197/xterm/ (integrated patch by
    Robert Brady). In this context, doublewidth refers to 16-bit
    character sets which may have glyphs occupying two cells.
  • add command-line option -class, which allows
    one to override xterm’s resource class. Also add resource file
    UXTerm.ad, which simplifies using xterm for both
    8-bit character sets and UTF-8.
  • fixes/improvements to OPT_TCAP_QUERY logic (patches by Bram
    Moolenaar, Stephen P Wall).
  • modify Makefile.in to circumvent GNU make’s built-in suffix
    rule for “.sh” which confuses install.sh with the “make
    install” target (report/patch by Paul Gilmartin).
  • implement an experimental control sequence which an
    application may use to query the terminal to determine what
    sequence of characters it would send for a given function key.
    This is ifdef’d with OPT_TCAP_QUERY (request by Bram
  • add /usr/local to search path in CF_X_ATHENA configure
    macro to build with Xaw3d on OpenBSD (patch by Christian
  • add missing #undef HAVE_TERMIO_C_ISPEED to
    xtermcfg.hin, omitted in patch #133 fix for IRIX 6.5 baudrate
    (report by Alain Filbois <Alain.Filbois@loria.fr>).
  • correct a few errors in xterm.man: font in filename example
    for Tektronix emulation, and description of -nul/+nul
    command-line options (report by Eric Fischer
  • update config.guess and config.sub, from tin and lynx.
  • back out change to -name command-line option,
    restoring its original behavior (as noted by David Madore, the
    correct interpretation of this option is the application
    instance rather than the application class).
  • workaround for fixed fonts which are generated from Unicode
    fonts: they omit glyphs for some xterm’s less-used line-drawing
    characters, which caused xterm to set a flag telling it to use
    only its internal line-drawing characters. Do not set the flag
    (it can be set from the popup menu), and xterm will generate
    only the line-drawing glyphs which actually are missing.
    Otherwise, when used for large fonts, xterm may generate a
    2-pixel wide line, which can leave dots on the screen.
  • restore first line of 256colres.pl, omitted in 4.0c
  • make command-line -name option work as
    documented. Apparently this was lost in X11R5 when coding to
    use XtAppInitialize.
  • limit numeric parameters of control sequences to 65535 to
    simplify checks for numeric overflow.
  • change index into UDK list to unsigned to guard against
    numeric overflow making the index negative (Taneli Huuskonen
  • change sun function-keys resource name to
    sunFunctionKeys to work around redefinition of the
    token sun by xrdb on Solaris. Similarly, renamed
    resource sun keyboard to
    sunKeyboard (Stephen P Wall).
  • change similar resource names for HP and SCO to avoid
    potential conflict with xrdb symbols on other systems, as well
    as for consistency.
  • reorganized the install targets in the autoconf’d Makefile,
    adding install-app, install-bin, install-dirs and install-man.
    The app-defaults class can be overridden by setting the make
    variable ‘CLASS’, simplifying customization of xterm as a
    Unicode terminal, e.g., CLASS=UXTerm.
  • add limit checks to ClearInLine(),
    ScrnInsertChar(), ScrnDeleteChar() to
    correct potential out-of-bounds indexing (prompted by Debian
    bug report #64713,
    which reported a problem with ICH escape sequences).
  • updates to config.sub and config.guess Kevin Buettner
    <kev@primenet.com> for elf64_ia64 Bernd Kuemmerlen
    <bkuemmer@mevis.de> and MacOS X.
  • for os390, add check for errno set to ENODEV on failure to
    open /dev/tty when there is no controlling terminal (patch by
    Paul Gilmartin).
  • fixes for building on Digital Unix 4.0 and AIX 4.2 (patch
    from H Merijn Brand <h.m.brand@hccnet.nl>).
  • modify DECRQSS reply for DECSCL to additionally report if
    the terminal is set for 8-bit controls.
  • add a resource (limitResize) limiting resizing
    via the CSI 4 t and CSI 8 t sequences.
  • ignore out-of-bounds resize requests, i.e., where
    sign-extension or truncation of the parameters would
  • remove code introduced in #134 which made some backgrounds
  • minor correction to format of updated $TERMCAP when adding
    kb capability for ptyInitialErase
  • improved test for SVR4 definition.
  • update URL’s and mailing addresses, moved to
    http://dickey.his.com and dickey@herndon4.his.com
  • correct missing quotes in CF_TTY_GROUP configure script
    macro in case the script is run in batch mode.
  • modify ownership-check of log file to ignore the group
    ownership. Otherwise xterm cannot create logfiles in
    directories with set-gid permissions.
  • simplify the logic that reads termcap data.
  • add fallback definition for B9600 in case line speed
    definition for 38400 is missing (report by Jack J Woehr
    <jwoehr@ibm.net>, for OpenBSD 2.6).
  • fix: Set highlightColor, and select a region containing the
    text cursor. If the window loses focus, the cursor becomes
    hollow, with the region inside the cursor being
    background/foreground, unlike the rest of the selection, which
    is foreground/highlight (patch by Ross Paterson
  • add configure script tests to define SVR4, SYSV and
    USE_POSIX_WAIT, which enables xterm to compile on Solaris 7 and
    SCO Openserver without imake, though there are still a few
    features for the latter which require sco to be
  • patches from Stephen P Wall:
    • add support for two Sun-specific function keys. These
      keys are labeled F11 and F12 on Sun Type 5 keyboards, but
      return SunXK_F36 and SunXK_F37. Support will only be
      compiled in if the header file <X11/Sunkeysym.h>
      exists and contains the appropriate symbol definitions. The
      keycodes for the DEC keycodes were arbitrary unused codes,
      but the ones for the Sun keycodes are what cmdtool and
      shelltool actually send.
    • add colorRV and colorRVMode
      resources to allow specifying a color to use for reverse
      video, similar to the existing UL, BD, and BL modes.
    • add alwaysUseMods resource, to override
      check if alt or meta modifiers
      are used in translations resource. Revamped
      the code to calculate the modifier value, and included Meta
      if alwaysUseMods is TRUE, using values 9-16.
  • fixes for os390 (Paul Gilmartin):
    • regularize the definition of CONTROL() and remove an
      acknowledged “trial and error” table.
    • translate “^?” into A2E(0177) which is the EBCDIC “DEL”
      rather than plain 0177 which is the EBCDIC quotation
    • modify xtermMissingChar() so that EBCDIC
      codes 128-159 are not rendered as blanks by X server
      running on Solaris, which sees those as control
    • make debugging traces (configure –enable-trace) work
      properly with EBCDIC.
  • add substitutions in autoconf’d Makefile for CPPFLAGS,
    LDFLAGS and AWK (reported by Neil Bird).
  • correct uninitialized childstat variable from
    patch #131 in creat_as() function, which caused
    logging to not work on Solaris, whose waitpid()
    function does not initialize its parameter. Add check for EINTR
    on return from waitpid() as well (reported by Neil
    Bird <neil.bird@rdel.co.uk>).
  • remove a redundant check for working setuid()
    function introduced in patch #132. Greg Smith reports that this
    does not work as intended on os390.
  • change line speed from 9600bd to 38400bd, to accommodate
    people who mistakenly use $TERM set to vt100, to
    reduce the effect of padding associated with this terminal
  • add configure script check for IRIX 6.5’s redefinition of
    baud rates associated with struct termio, to
    correct a situation where the baud rate was initialized to zero
    (reported by Andrew Isaacson <adi@lcse.umn.edu>).
  • remove unused configure script check for VDISABLE.
  • undo an incorrect change from patch #113 caused the right
    scrollbar to be positioned incorrectly when re-enabling it
    (analysis by D Roland Walker).
  • add ctrlFKeys resource, replacing constant for adjusting
    control-F1 to control-F12 to VT220-style F10-F20. The resource
    changes the constant 12 to a default value of 10 (request by
    Jim Knoble <jmknoble@pobox.com>).
  • correct ifdef’ing for conflict between definitions in AIX’s
    <sys/select.h> and <X11/Xpoll.h> (reported by Clint
  • add checks for return-values of getutid(), initgroups() and
    setuid() in main.c, as well as modifying ifdef’s for __osf__ to
    include tty-group and WTMP logic (adapted from patch by Paul
    Szabo <psz@maths.usyd.edu.au>)
  • modify resize.c to build and work on os390 (uses EBCDIC)
    (adapted from patch by Phil Sidler
  • use Ires(), Bres() and Sres() macros to simplify resource
    list in charproc.c
  • resync with XFree86 4.0a:
    • correct a typo in os2main.c (Holger Veit, from
  • improve logfile security (integrated patch by Branden

    • make the creat_as() function more strict
      by using O_EXCL rather than O_APPEND.
    • fixes to make DEBUG ifdef’s compile/run, including
      making the debug logfile more unique by appending a
      timestamp to its name.
    • include <term.h> in resize.c, to fix a
      missing-prototype warning.
  • modified creat_as() a little more, retaining
    the ability to append to a logfile If the user specifies the
    name. Also, check if the opened file (which patch #130 ensures is owned by the effective
    user) is not writable by other users.
  • use creat_as() logic to make tek4014
    screen-copy more secure (noted by Branden Robinson).
  • ifdef’d some of Branden’s changes to build/work on older
  • correct missing initialization of the .mode
    flag in ColorRes struct, from patch #129. This
    worked on Linux because malloc() zeroes memory on
    that platform (reported by Christian Weisgerber).
  • modify logic for deleteIsDEL resource so it
    has internally 3 states: unspecified, true and false. If
    unspecified, the keyboard type determines whether the Delete
    key transmits <ESC>[3~ or \177,
    and the popup menu entry reflects the internal state.
    Otherwise, the popup menu entry overrides the keyboard type
    (suggested by Dr Werner Fink, to make it simpler to set
    resources that imitate the legacy X11R6 xterm).
  • modify scroll-forw() and
    scroll-back() actions, adding a third parameter
    which will direct xterm to ignore the action when mouse
    reporting is enabled. This is needed for the wheel mouse to be
    used to report to the application rather than scroll the
  • add menu entry and action to allow disabling xterm’s
    assumption that the current font contains line-drawing
    characters if the font cells 1-31 are nonempty. Some fonts may
    have other characters (reported by Bruno Betro
  • add a check in creat_as() to ensure that the
    user really owns the logfile that has been opened.
  • add logic to implement SCO function-keys. (This is really
    incomplete, since I intend to revisit this and make xterm able
    to emulate scoansi better than just the function-keys).
  • add configure script option –enable-sco-fkeys, minor
    related fixes (patch by Dr Werner Fink).
  • fix typos in ctlseqs.ms (reported by Bram Moolenaar)
  • fix typo in sinstall.sh default for $TST_PROG (reported by
    Paul Gilmartin <pg@sweng.stortek.com>)
  • improve initialization of ANSI colors by delaying
    allocation until each color is first used.
  • remove ifdef that prevented colorBD/colorUL/colorBL
    resources from working when 256-color configuration was built
    (reported by Todd Larason).
  • fix some minor inconsistencies in terminfo (Debian #58530).
  • correct logic for oldXtermFKeys resource,
    fixes a core dump when attempting to set it from the
    command-line (reported by Dr Werner Fink).
  • correct ifdef for meta-sends-escap so configure
    –disable-num-lock builds.
  • add resource, popup menu entry and control sequence to
    allow changing the Delete key to send either DEL or the
    VT220-style Remove escape sequence.
  • remove logic for metaSendsEscape that would allow xterm to
    send <ESC>[3~ before a function key that
    would begin with <ESC>[3~ (request by
    Christian Weisgerber).
  • add missing action and documentation for
  • correct a few typos in ctlseqs.ms (incorrect code for Cyan
  • ensure that xterm will automatically activate the UTF-8
    mode whenever the name of the locale environment variable
    suggests that a UTF-8 locale is in use (patch by Markus Kuhn).
    This will help that by simply setting LC_CTYPE an entire system
    can be switched over to UTF-8, without users having to remember
    the UTF-8 command line options (“-u8”, etc.) of the various
    applications. Command line options and X resource entries can
    still be used to override this default choice.
  • add old function-keys control sequences and popup menu
    entry, for compatibility with legacy X11R6 xterm.
  • revert translation of editing keypad “Delete” key to legacy
  • simplify the color-resource data expressions with macro
    COLOR_RES, for later use in restructuring color
  • change encoding of wheel mouse (buttons 4 and 5) to avoid
    conflict with legacy mouse modifiers (suggested by Bram
  • make this work on OpenVMS (integrated patch by David
  • rename some functions, e.g., Index to
    xtermIndex so that ports such as VMS which link
    externals ignoring case will not have library conflicts
    (reported by David Mathog).
  • correct logic of do_reversevideo(), which did
    not update the corresponding popup menu check mark (reported by
    David Mathog, this was a detail overlooked in patch #94).
  • change TRACE macro so semicolon is not within definition,
    making indent and similar programs work better.
  • add depend rule to Makefile.in
  • modify logic of boldColors resource to suppress it if an
    extended color control has been used, e.g., for 88-color or
    256-color mode (patch by Todd Larason).
  • revise logic that handles menus and input translation for
    keyboard type so only one can be selected at a time.
  • restore kdch1=\177 for the Sun function-key type, and make
    the the Delete key send DEL (\177) if the oldXtermFKeys
    resource is set.
  • rephrase logic and ifdef’s for POSIX VDISABLE to avoid
    preprocessor expression that will not compile on NetBSD/x86
    1.4.1 (reported by Takaaki Nomura
  • change coding of editing keypad’s “Delete” key to
    <escape>[3~, in the default (Sun/PC) keyboard mode. This
    makes the terminfo kdch1 capability independent of the coding
    of the backarrow key, which sends either backspace (8) or DEL
    (127). The reason for doing this (compatibility with the
    screen program) outweighs the choice of DEL (127)
    which was used in X11R5/X11R6 xterms. The screen
    program translates whatever matches kdch1 into
    <escape>[3~, even if it happens to be the stty erase
  • add encoding for control/?, to work around xmodmap or key
    translations which may confuse backspace and delete. A
    control/? will send DEL (127), and a control/H will of course
    send backspace (8).
  • add encoding for kcbt to <escape>[Z (fixes Debian
  • minor correction to logic that encodes Sun and DEC function
    keys to avoid sending an escape sequence if the key symbol is
    not found in xterm’s lookup table.
  • simplify ifdef’s in main.c for POSIX VDISABLE so the “^-”
    pattern is more likely to be implemented when imake
    configures xterm (fixes Debian #55105).
  • change manpage to make it clearer what codes are sent by
    the backarrow key.
  • add a note reserving OSC 51 for use in Emacs shell (request
    by Rob Mayoff <mayoff@dqd.com>).
  • correct a missing backslash in xterm-vt220 termcap.
  • cleanup remaining quoted includes, preferring bracketed
  • minor configure-script macro updates from tin and
  • add configure-script option for using utempter library,
    adapted from Redhat 6.1 patch for XFree86 3.3.5
  • resync with XFree86 3.9.17a:
    • correction to QNX support (Frank Guangxin Liu)
    • some cosmetic changes that did not correct any reported
  • move the suggested wheel-mouse button translations into
    charproc.c to simplify customization. Correct some minor logic
    errors in the support for buttons 4 and 5, used for wheel mice
    (reported by Bram Moolenaar).
  • implement metaSendsEscape resource, with
    corresponding control sequence and menu entry. Like
    eightBitInput, this causes xterm to send ESC prefixing
    the given key, but applies to all keys and is independent of
    the 8-bit/7-bit terminal setting. (requests by Alexander V
    Lukyanov and Marc Feeley).
  • correct potential indexing with negative subscript in
    udk_lookup(), (reported by Ian Collier
  • modify configure script that sets TERMINFO_DIR to use
    ${prefix} rather than /usr if the –prefix option was specified
    (request by Zdenek Sekera <zs@sgi.com>).
  • modify checks for repeat-character control sequence to test
    the character class against xterm’s state table, rather than
    the isprint() macro (patch by Alexander V Lukyanov).
  • change several functions to macros to improve speed
  • two corrections to simulation of bold font via

    • use clipping to avoid leaving trash at end of the text,
    • add brackets so wide-character logic does not
      fall-through into the overstriking logic (reported by Marc
      Feeley <feeley@IRO.UMontreal.CA>)
  • add several entries to termcap file to make it have the
    same set of aliases as the terminfo file.
  • scale the color values used for xterm-256color terminfo
    entry to 0..1000, as expected by ncurses.
  • change xterm-r6 terminfo definitions for F1-F4 to match
  • Add QNX/Neutrino support (Frank Guangxin Liu
  • change label on “Sun/PC Keyboard” popup menu entry to
    “VT220 Keyboard”, since the checked state corresponds to VT220
    rather than Sun/PC.
  • add configure test CF_UTMP_UT_XSTATUS to handle the
    variants of utmp exit status (reports by Dave Ellement, Jeremie
    Petit, Tomas Vanhala).
  • amend treatment of ALT key (see patch
    #94) so that if ALT is used as a modifier in key
    translations, then no parameter will be sent in escape
    sequences for Sun/PC function keys (request by Dr Werner
  • modify default for OPT_I18N_SUPPORT to assume that
    XtSetLanguageProc() is available in X11R5.

    Caveat: XtSetLanguageProc() was added fairly
    late in the X11R5 patches, and some vendors shipped buggy
    versions of this function (request by Tomas Vanhala).

  • correct configure macro CF_SYSV_UTMP to test-link with
    functions consistent with the header, e.g., getutent() for
    utmp.h and getutxent() for utmpx.h (reported by Greg
  • modify terminfo entry for xterm-xfree86 to
    reflect modifiers for shift and control (from a patch by
    Alexander V Lukyanov).
  • modify terminfo entry for xterm-sun to match
    the function-key definitions in ncurses. The pageup/pagedown
    and related function keys correspond to the Sun keyboard, which
    does not necessarily correspond with X’s notion of those
  • modify treatment of XK_Delete keysym so it transmits
    parameterized VT220-style <ESC>[3~ if
    modifiers (shift, control alt) are given (request by Alexander
    V Lukyanov).
  • corrected misspelled resource name in command-line option
    for HP function keys.
  • refine the change to SGR_Background() in patch
    #119, by not flushing the pending scrolling operation if the
    background color is not actually changing. This combination
    occurs when using color-ls to display a long listing, since
    each line ends with an SGR0 which affects only the foreground
    color. The unnecessary flushing made it noticeably slower
    (reported by D Roland Walker <walker@pobox.com>).
  • remove obsolete documentation about modifiers which can be
    returned in mouse tracking mode, and modify logic to ignore
    modifiers other than the existing ones, e.g., NumLock (prompted
    by discussions with Christian Weisgerber and Brad Pepers
  • use free bit from obsolete shift-modifier coding of mouse
    tracking button events to encode buttons 4 and 5, e.g., for a
    wheel mouse (requests by Brad Pepers and Bram Moolenaar).
  • correct a place where the ptyInitialErase logic did not set
    the backarrowKey state, and modify it further to use tgetstr()
    rather than parse the termcap data returned from tgetent() so
    the ptyInitialErase logic will work when xterm is linked with a
    terminfo library (based on a patch by Dr Werner Fink
  • fix definition of HAVE_UTMP_UT_HOST for ISC configuration
    (patch by Michael Rohleder).
  • improve configure script’s utmp tests (based on reports by
    Greg Smith for os390, and David Ellement for HPUX).
  • modify sinstall.sh to use uid=0 rather than ‘root’ to
    determine if the installer is privileged. This is needed on
    some systems since more than one account may be privileged
    (report by Greg Smith).
  • add an application resource, messages (and a
    corresponding -/+mesg option) which controls the initial
    permission on the terminal: if messages is set to true (the
    default), behavior is as without the patch; if it is set to
    false (as per -mesg), the terminal is opened in mode 0600, thus
    producing the effect of the mesg n command. This
    is useful for users who want to redirect all their messages to
    one particular xterm: it is more pleasant to do this with xterm
    resources than with explicit calls to the mesg
    program (patch, description by David Madore
  • add responses (DA and DSR) for DEC locator mode
  • add coding for ANSI color to DA response
  • implement UTF-8 translation for Media Copy (print)
  • implement vt320 control sequences for Print Composed Main
    Display and for Print All Pages. The latter directs xterm to
    print the current screen as well as the scrollback buffer.
  • correct error in _GNU_SOURCE configure test, which left it
    always defined.
  • add more information, i.e., with strerror for some system
    calls in the main program which may fail due to insufficient
    permissions (prompted by a problem report for -C
    by Jeremie Petit <Jeremie.Petit@digital.com>).
  • add workaround for conflict between <X11/Xpoll.h> and
    <sys/select.h> on AIX 4.3 (Richard Griswold
  • add configure script test to resolve conflict between
    between <X11/Xpoll.h> and <sys/select.h>
  • modify translation of UTF-8 sequences to reject “overly
    long” variations (patch by Markus Kuhn).
  • remove utf8controls resource, since Markus’
    change removes the corresponding logic.
  • correct a case where colors were not rendered properly.
    This happened when an application inserted several lines, then
    changed colors. If this was done all in one write, then there
    would be no intervening refresh, and the new color was applied
    to the pending scrolling operation which was awaiting the next
    refresh (reported by Stephane Chazelas
  • refine configure test for utmp versus utmpx, to build on
    HP-UX 10.x (reported by David Ellement).
  • move the configure check for const after the
    check for ANSI C compiler options, since those may be required
    to make it work properly on HP-UX.
  • add configure test for defining
  • correction to -hold option, ensure that if data is
    already in the output buffer that it will be displayed before
    closing the PTY.
  • move the configure checks for setuid install of xterm into
    a script to avoid installing it setuid’d to a non-root user
    (reported by Adam Sulmicki <adam@cfar.umd.edu>).
  • correct configure script’s check for termcap.h to avoid
    using ncurses’ version of it on systems that have a working
    tgetent() function. This is needed to make
    resize work properly.
  • fix some typography in ctlseqs.ms description of DEC
    locator events (reported by Stephen P Wall).
  • change order of tests in configure script for utmp and
    utmpx to test utmpx first, to compile on Solaris (reported by
    Leena Heino <liinu@uta.fi> and Patrik Hagglund
  • add a configure test for utmp.ut_xtime, needed for SCO
    Openserver, and for lastlog, needed for Redhat 6.0, to refine
    the utmp/utmpx auto-configure.
  • remove a spurious comma in an #undef (reported by David
    Green <greendjf@cvhp152.marconicomms.com> and David
    Ellement <ellement@sdd.hp.com> both on HP-UX, whose
    compiler does care about syntax).
  • change ifdef’s using __CYGWIN32__ to __CYGWIN__ (reported
    by Suhaib M. Siddiqi <Ssiddiqi@InspirePharm.Com>, who is
    told that the next Cygnus release will drop that symbol in
    their next release).
  • minor cleanup of ifdef’s for makeColorPair (patch by
    Stephen P Wall).
  • work around problem observed in XFree86 3.3.5 (patch by
    Alexander V Lukyanov, Redhat #5419).
    From the problem report:

    xterm consumes cpu when selecting text with mouse
    (holding down left mouse button) and when a program working
    under the xterm outputs something to stdout.

    Easy way to reproduce:
    while :; do echo aaa; sleep 1; done
    (while this runs, select text and hold down left mouse
    button) watch cpu load.
    release left mouse button – spinning stops.

  • modify warning if change-ownership of PTY fails; some
    configurations may not happen to have old-style pty’s (reported
    by Bob Maynard).
  • improve check in configure script for group ownership of
    installed xterm; some platforms use the -g option of ls to
    toggle group off rather than on (reported by Greg Smith).
  • minor improvement to toolbar geometry, to make it not
    resizable. This still is not satisfactory (but is usable) since
    the toolbar overlaps the xterm widget if the window is resized
    to make it smaller. It appears that some work is needed for the
    xterm widget’s geometry management to make it function
  • implement configure script tests for utmp, tty group.
  • implement -hold option, allowing users to retain
    the window after a shell has exited (this is recently an FAQ,
    but prompted by a comment by Joachim Plaettner-Hochwarth
    <plaettner@aem.umn.edu> in comp.unix.programmer, that the
    IRIX winterm provides this option).
  • add support for DEC Locator control sequences for xterm
    (integrated patch by Stephen P Wall):

                    DECEFR, "Enable Filter Rectangle"
                    DECELR, "Enable Locator Reports"
                    DECSLE, "Select Locator Events"
                    DECRQLP, "Request Locator Position"

    This allows the xterm mouse to be used with applications
    that use the DEC Locator sequences, such as VAX Tpu, or SMG$
    based applications.

  • improve print action (patch by Matthias Baake). From his

    • Bug 1
      Underlined text is preceded by ESC [0;2m . This should
      be ESC [0;4m , ESC [2m doesn’t seem to have
      any effect. (print.c/send_SGR)
    • Bug 2
      The check for the last non-empty column (while (last >
      0) … in print.c/printLine) omits the rightmost column of
      the screen, the loop must start with last =
      screen->max_col+1 instead of last =
    • Bug 3
      Any attributes of the first character (and of all
      immediately following characters with the same attributes)
      are ignored. The variable attr (print.c/printLine) should
      be initialized with 0, not with *a & SGR_MASK.
  • integrated changes by Stephen P Wall to implement an
    88-color model for systems where 256-colors cannot be
  • when 256-color configuration is compiled, colored bold and
    underlining is not available; ifdef’d to avoid possible odd
    effects in this case (reported by Stephen P Wall).
  • add resource cacheDoublesize, to limit the caching
    of font information for double-sized characters. This addresses
    a problem reported by Aryeh Koenigsberg for X terminals with
    limited font memory.
  • modify treatment of line-drawing characters in UTF-8 mode
    so that the Unicode values are used rather than the C0 codes
    for storing the translated characters (request by Markus
  • add configure script checks for Athena headers and
    libraries under /usr/contrib to work on HPUX (reported by
    several people: David Nixon <djn@csc.liv.ac.uk> Aryeh
    Koenigsberg <aryeh.koenigsberg@telrad.co.il> Johannes
    Mähner <johanm@camline.com> Andrew Gaylard
  • add check to configure script if xterm is installed setgid
    rather than setuid, since wtmp and utmp may be installed with
    group-writable permissions other than root (based on Debian bug
    report #7112
    by Bo Branten <bosse@ing.umu.se>).
  • rewrote logic that removes data from termcap entry, e.g.,
    for titeInhibit, to make it less likely to remove the wrong
  • correct logic which checks for missing characters used for
    line drawing. The 0 character was tested unnecessarily, leading
    to some inefficiency when rendering.
  • change termcap capability which is used as input or output
    of ptyInitialErase logic from kD to kb.
    Christian Weisgerber <naddy@mips.rhein-neckar.de> pointed
    out in effect that kD (in terminfo kdch1)
    should correspond to the control sequence for dch1,
    which deletes from the current position toward the right.
  • check for failure to change ownership of the PTY device and
    warn when xterm is running setuid’d to root. This was reported
    to happen on the FreeBSD/NetBSD/OpenBSD systems as a result of
    the chflags() call.
  • add xterm-noapp terminfo entry to illustate a nominally
    bash-compatible terminal description.

Several fixes. The main one is a first draft of pulldown
menus. It’s not complete (I have some minor/annoying geometry
problems to correct), but is usable. Once it’s complete I plan to
add a menu to support additional selections that won’t fit on the
current popup menus. Also, this fits into my longterm plan to
allow configuring with Motif libraries.

  • correct error in input conversion for NRC mode (reported by
    Stefan Traby <stefan@sime.com>).
  • fix initialization of num_lock data in Misc struct (since
    those fields of the reference widget aren’t initialized), and
    add logic to deal with XVision whose NumLock key does not
    transmit but does alter keypad state.
  • correct a missing return-value in get_pty(), for SCO
  • add E2A fix for backspace (patch by Greg Smith), i.e.,
    #define VAL_INITIAL_ERASE A2E(127)
  • correct foreground color within cursor outline when the
    window is unfocused (reported by Stephane Chazelas).
  • corrected position of scrollbar set in ResizeScrollBar(),
    which left it positioned incorrectly if the right scrollbar
    were enabled from the popup menu but was not initially enabled.

    I have noticed some additional problems with
    right-scrollbar on X11R5 which I will correct later.

  • integrate changes by Jean-Claude Michot for QNX from
    XFree86 3.3.4c
  • add resources menuBar/MenuBar, menuHeight/MenuHeight for
    later use in toolbar geometry.
  • add configure option –enable-toolbar
  • add missing definition to make ziconbeep logic not
    compile-in when configure script disables it.
  • add configure-check for input-method support in X
  • add null-pointer check to FlushLog(), fixes a core dump
    when both -l and -lf options are used when xterm is configured
    with wide-character support.
  • remove “ISO” case for SD, which was due to a typographical
    error in ECMA-48 (reported by Paul Williams
    <paul@celigne.co.uk> for vttest).
  • add “FILES” section to manpage.
  • generate header file to initialize default resources for
    colors 16-255.
  • fixes for os390 (Greg Smith):
    • add README.os390
    • use the pty_search() function to find an available
      pty/tty pair.
    • move E2A() call out of getXtermCell() to SaveText()
      function so it will be available in all
  • patches by Todd Larason:
    • enable SGR 48 5 in ISO color mode, not just 256 color
    • change configure’s –disable-256-color option to
      –enable-256-color, to match its effect
    • fix OSC 4 xx ? (report ansi color) to report a string
      which can actually be used to set the color back
    • fix OSC 4 xx yy (change ansi color) to not allow
      setting colors > 15 in 16 color mode
    • simplify the COLOR_ settings a bit in ptyx.h, along
      with setting NUM_ANSI_COLORS needed for #1 and #4
    • correct string-terminator code passed for reference to
      OSC responses; when in 7-bit mode, only the final byte of
      ST was seen.
    • Allow multiple color #;name pairs in OSC 4, and
      document changes to match.
  • add control sequences for specifying the RGB value of the
    ANSI colors, and for configuring with 256 colors (patch by Todd
    Larason <jtl@molehill.org>). I made the default
    configuration to 16-colors, because xterm uses the default
    color map (which has only 256 colors).
  • correct an error in DCS $ q m reporting for colors 8-15
    (Todd Larason).
  • add test/demo script for double size characters. Used this
    to test/correct display of double size characters that should
    wrap, underlined double size characters.
  • increased cache size for double size fonts to 8, to allow
    for both normal and bold fonts (discussion with Aryeh
    Koenigsberg <aryeh.koenigsberg@telrad.co.il>).
  • add fixes for port to OS390 aka MVS (integrated patch from
    Greg Smith <rys@epaibm.rtpnc.epa.gov>). OS390 uses EBCDIC
    rather than ASCII.
  • correct an off-by-one in binary search limits in
    keysym2ucs.c (Markus Kuhn).
  • implement logging for UTF-8 mode. The output is written in
    UTF-8 form.
  • If colorMode is enabled by default, compile-in default
    resources to match the colors listed in XTerm-col.ad (this
    should fix a longstanding FAQ).
  • added new OSC 3 ; PROPNAME=VALUE ST
    escape sequence to set an arbitrary X property on the top level
    xterm X11 window. Omit “=VALUE” to delete the X property (patch
    by Greg Badros <gjb@cs.washington.edu>).
  • change internal flag used for utf8controls
    resource so we allow 31-bit range of characters (suggested by
    Thomas Wolff).
  • add check for 16-bit characters in OSC strings, change them
    to ‘?’ (reported by Thomas Wolff).
  • modify logic of same_font_name() to avoid trying to
    interpret both parameters as wildcard patterns. That does not
    (cannot) work, and in some instances the font server will
    return unresolved wildcards for the normal or bold fontnames,
    making them match inadvertently, triggering the fallback
    overstrike logic (reported by Tim Adye).
  • correct range-check from patch #108, which resulted in not
    being able to select from the scrollback buffer (reported by
    Tim Adye <T.J.Adye@rl.ac.uk>)
  • correct “no available ptys” problem with Cygwin B20.1
    (patch by Tim Adye).
  • modified install-ti rule in Makefile.in to allow override
    of the terminfo directory when doing a “make install”, i.e., by
    assigning to TERMINFO_DIR (request by Zdenek Sekera
  • added install-full rule to Makefile.in
  • resync mkdirs.sh and install.sh scripts against current
  • implement $(DESTDIR) in Makefile.in, making it simple to
    install xterm and associated files into a directory just for
    packaging a binary release (suggested by CaT
  • change IChar type to unsigned, rather than unsigned short,
    making room for a flag to keep with 16-bit characters to
    prevent them from being interpreted as C0 or C1 controls
    (reported by Thomas Wolff).
  • correct a typo from patch #107, incorrect array name, in
    the filterUTF8 function (patch by Bruno Haible
  • add utf8controls resource to specify whether xterm
    should interpret 16-bit characters unpacked from UTF-8 form as
    control characters if they happen to fall into that range. This
    behavior is left unspecified by the Unicode standard (request
    by Thomas Wolff).
  • modify handling of OSC to recover if application sends
    16-bit characters with codes above 255. We cannot display them
    as is, but translate out-of-range characters to a ‘?’ (reported
    by Thomas Wolff).
  • add a range-check to LastTextCol(), to guard against
    indexing before the beginning of the scrollback buffer. This
    appears to happen with certain fonts under X11R5 (reported by
    Stephane Chazelas <Stephane_Chazelas@Raytheon.com>).
  • implement resource boldMode, to allow disabling
    the simulation of bold fonts when the bold and normal fonts are
    not different (requested by Will Day
  • change the atom “UTF-8” to “UTF8_STRING”, and fixes a few
    bugs in the UTF-8 selection (patch by Juliusz Chroboczek).
  • correct logic of binary-search in keysym2ucs.c (patch by
    Markus Kuhn).
  • add special interpretation of keysym codes above 0x1000000
    as the corresponding UCS value plus 0x1000000 (patch by Markus
  • Two changes from Stephen P Wall. From his description:

    The first change is simple – I added ESC[3J to erase the
    stored lines above the screen. That’s what the changes to
    util.c and ctlseqs.ms are.

    The second change is to get the blinking cursor working.
    I took out the cursorBlinkTime resource, and put in
    cursorBlink (Boolean), cursorOnTime (time cursor is on in
    msecs) and cursorOffTime, and added a cursorblink item to
    the vtMenu to enable/disable it.

  • improve selection (integrated patch by Juliusz Chroboczek).
    From his description:

    With this patch, selection conversion works

            ISO 8859-1 xterm -> ISO 8859-1 xterm (transferred as STRING);
            ISO 8859-1 xterm -> UTF-8 xterm (transferred as STRING);
            UTF-8 xterm -> ISO 8859-1 xterm (transferred as STRING);
            UTF-8 xterm -> UTF-8 xterm (transferred as UTF-8).

    It will not work properly if one xterm is in, say, ISO
    8859-2. Actually, for this case xterm breaks the ICCCM
    routinely (sending ISO 8859-2 data as STRING), so I
    wouldn’t worry too much about it. I have not changed the
    behaviour in eight-bit mode in any way.

  • remove duplicate fix for OpenBSD in resize.c
  • correct logic in ScreenWrite, which did not reset the high
    bytes of 16-bit characters when overwriting them with a 8-bit
    character string (reported by Thomas Wolff).
  • provide limited support for input of UTF-8 16-bit data by a
    lookup table (integrated patch by Markus Kuhn).
  • correct check in non_blank_line to ensure we’re in
    wide-character mode before looking at the high bytes, for
  • implement new resource trimSelection, which allows
    xterm to trim trailing blanks from selected lines. This does
    not affect the highlighting. (reported by several people using
    mutt, including Hans Morten Kind <edphk@uib.no>,
    Jeremy <jeremy@exit109.com> and (Michael Fuller
  • include term.h in resize for OpenBSD (patch by Matthieu
  • correct logic for UTF-8 in functions that hide and show the
    cursor; it was displaying a space whenever the low byte of the
    character at the cursor position was zero (reported by Thomas
    Wolff <Thomas.Wolff@icn.siemens.de>).

This is a resync patch against XFree86 3.9Pn, reflecting
changes which were submitted by Branden Robinson, who worked with
Wichert Akkerman <wichert@cs.leidenuniv.nl> to set up
ifdef’s to handle GNU libc 2.1, and use getpt() which lets xterm
avoid having to know the actual PTY name.

I have also added a few fixes for signed/unsigned mismatches,
and corrected a problem in the configure script (the UTF-8 code
was always configured since the Imakefile defines this).

  • correct selection logic: I omitted an offset that accounts
    for the distance into the scrollback buffer when rewriting this
    for patch #101. Also fixed a similar
    problem for selecting double size characters from the
    scrollback buffer (first was reported by D Roland Walker
  • improved support for Unix98 PTY’s, using patch in Debian
    bug report #35650,
    by J.H.M. Dassen <jdassen@wi.leidenuniv.nl>. From the
    patch description:

    • No longer links xterm against libutil on a glibc2.1
      system. libutil is a compatibility library and should
      only be used when necessary. Rather than having get_pty()
      use openpty() from this compatibility library, use the
      real UNIX98 pty support in get_pty() (open()ing the
      master pty, through getpt() if available (glibc
      extension)). Use openpty() only under glibc2.0.
    • GNU libc2 is not Linux-specific; already it runs on
      the Hurd. It provides the UNIX98 pty functions (plus the
      getpt() extension), regardless of the underlying OS.
      Changed two constructs to look for GNU libc2 only, not
      GNU libc2 on Linux.
  • improve font configuration, by checking if the user’s
    resource settings for normal and bold fonts give the same font
    sizes (fixes problem reported by Peter Waltenberg
  • revert change to openpty call in patch
    #101; this causes xterm to fail on DEC-Alpha OSF/1 4.0B
    (reported by H Merijn Brand).
  • modify print.c to move include of stdio.h after the
    autoconf’d xtermcfg.h since HP’s ANSI C compiler otherwise sees
    inconsistent prototype for getopt, probably due to problem with
    const (reported by H Merijn Brand).
  • moved includes and definitions for ‘select’ from data.h to
    xterm.h to fix problem introduced by prototype for getPtyData,
    part of UTF-8 changes (reported by Jens Schleusener
  • added “Meta <Btn2Down>:clear-saved-lines()” default
    translation (patch by H Merijn Brand)
  • fixes to configure script and ifdef’s in main.c to build on
    a HP9000/D390 (hppa-2.0w) running HP-UX 11.00 (64 bit) with
    egcs 1.1.2 and HP’s ANSI C compiler (patch by H Merijn Brand
  • add more parentheses in ifdef’s (patch by Bob
  • eliminate conflicting definitions for USE_TERMINFO in
    resize.c (reported by Jeremy Buhler).
  • change openpty call to pass NULL rather than ttydev
    parameter, since that was used only to estimate the length of
    the corresponding data, and may not really be long enough
    (reported by Andreas Jaeger)
  • update description in xterm manpage for character class
    table, which said it handles only 7-bit codes.
  • correct a typo in ScrnDeleteChar() which made it not clear
    the high byte of wide-character data.
  • add logic to convert selection to UTF-8 form when
    appropriate. This makes select/paste “work”, but further work
    is needed to make UTF-8 recognized as a locale in Xlib.
  • correct right-limit check when selecting double-width
  • change default answerback response to an empty string.
  • Correct a typo in the default resource value for
    backarrowKeyIsErase: it was always true (reported by Bram
  • improve configure script’s test if the installed xterm is
    setuid, in case that is a symbolic link.
  • correct “install-ti” rule in Makefile.in, by not setting a
    blank $TERMINFO value. That is interpreted as “.” by ncurses’
  • correct logic that computes num_ptrs count of the number of
    indices into the screen buffer. This is the maximum of the
    colors and character-set indices; was incorrect in patch
  • correct argument type for sigsetjmp, incidental change in
    patch #96’s Unix88 PTY patch (reported by Bram Moolenaar).
  • correct description of secondary DA in ctlseqs.ms (reported
    by Bram Moolenaar).
  • decouple the backarrowKey and ptyInitialErase resources by
    adding a new resource backarrowKeyIsErase, to accommodate
    people using applications which have hardcoded tests for
    characters 8 and 127 rather than relying on the stty
  • modify the UTF-8 decoder so that all possible illegal UTF-8
    sequences are properly represented by U+FFFD. This should be
    very helpful for developers of code that output UTF-8 strings
    for debugging. See the file utf-8-test.txt in
    http://www.cl.cam.ac.uk/~mgk25/download/ucs-fonts.tar.gz for a
    demonstration text that contains numerous illegal UTF-8 values.
    (patch by Markus Kuhn).
  • correct a place in ScrnRefresh where I was filling the high
    byte of a wide character with a space rather than a null
    (reported by Markus Kuhn).
  • correct data manipulation in unparseputc(), broken for
    little-endian machines by patch #97’s UTF-8 changes. This meant
    that keyboard input on SunOS did not work, though Linux i386
    was fine.
  • modify initialization for backarrowKey and logic for
    initial-erase to prevent the initial-erase from overriding an
    explicitly set backarrowKey resource (reported by Vikas
  • add a missing null-pointer check in ScrnRefresh, for the
    wide-characters configuration.
  • add configure script test for -lutil, needed for openpty
    call when configuring xterm for Glibc-2.1 and Unix98 PTY’s
    (first reported by Martin Lorentz”
  • completely parenthesize ifdef expressions for Glibc
    (suggested by Bob Maynard).
  • add initial-erase options (-ie, +ie) to help message
    (reported by Vikas Agnihotri).
  • remove duplicate definition of USE_USG_PTYS (reported by
    Jeremy Buhler).
  • change termcap kD and terminfo kdch1 to a
    DEL (\177). I overlooked this when separating the styles of
    keyboard with the sunKeyboard resources in patch #94,
    so that it normally matches the value of the stty erase

    • Reported by Jae Gangemi
      <jgangemi@ccf.rutgers.edu>, this caused emacs to not
      process the DEL properly, combining it with succeeding
    • This does not appear to be related to a problem which I
      have found with screen, which translates the stty
      erase into the termcap kD or terminfo
      kdch1 value (depending on how it is linked) if the
      $TERMCAP variable is set when screen is
  • add command-line options for enabling UTF-8 mode: -u8 and
    +u8. The more obvious -utf8 and +utf8 would conflict with
    xterm’s -ut and +ut (utmp) options. The UTF-8 changes were
    requested by Markus Kuhn <Markus.Kuhn@cl.cam.ac.uk>. This
    patch does not complete UTF-8 implementation, but makes it
    usable, i.e., display and refresh work, and I am able to
    display the test cases which Markus provides. More work is
    needed to complete this feature:

    • the control sequences for switching in/out of UTF-8
      mode are partly implemented (don’t use them). Similarly,
      the switching between vt100 and tek4014 emulations when
      UTF-8 mode is enabled will not work properly.

      You must use the -u8 command line option to use this
      feature, as well as compile with the OPT_WIDE_CHARS

    • cut/paste only copies 8-bit characters.
    • logging is disabled in the wide-character
    • printing only writes 8-bit characters.
    • input only does 8-bit characters. This is the area that
      I know least about.
  • modify Makefile.in to work with configure script’s –srcdir
    option. (patch by Jeremy Buhler
  • add checks for ‘echo -n’ equivalent for 8colors.sh and
    16colors.sh scripts (reported by Vikas Agnihotri).
  • improve logic that looks for bold fonts to allow for
    wildcards in the specification for normal fonts, and to ensure
    that if a bold font is specified as normal, that xterm will
    simulate a bold version of that using a one-pixel offset
    overstrike (reported by Henrik Harmsen
  • correct horizontal spacing of double width line-drawing
    characters that xterm simulates.
  • improve support for Unix98 PTY’s, using patch in Debian bug
    report #35650,
    by Topi Miettinen <Topi.Miettinen@medialab.sonera.fi>.
    Andreas Jaeger says this also corrects a permissions problem
    reported by cat@zip.net.au
  • modify initial-erase logic to ensure that
    ttyModes resource overrides it.
  • modify primary DA response to allow a ‘1’ parameter.
  • add printer and national replacement character sets to
    VT220 primary DA response.
  • document primary and secondary DA responses in
  • use the patch number (e.g., 95) in the secondary DA
    response, providing user applications a means of determining
    the version of xterm for feature comparison (request by Bram
  • make xterm respond to secondary DA when the decTerminalID
    is set for VT100.
  • limit user-defined keys (DECUDK) to VT220-style keyboard
    when sunKeyBoard resource is true.
  • modify ifdef’s for Linux-2.2.x with Glibc-2.1 to work with
    Glibc-2.1 and no Unix98 PTY support (patch from Andreas Jaeger
  • add optional feature (resource and command-line options) to
    make xterm use the PTY’s sense of erase character on startup,
    rather than requiring it to be \177, or set the PTY’s erase
    character to match xterm’s configuration. Note that while
    $TERMCAP is modified to reflect the actual configuration, the
    terminfo kdch1 string is not (request by Dirk H Hohndel
  • improve scripts in vttests to work with newer shells that
    do not use ‘echo -n’.
  • add fonts.sh example script
  • correct inequality in handling of “#1” font
  • correct call to XGetWMNormalHints() used for computing
    maximum screen size; the size hints may not have been set.
  • begin implementation of support for wide-characters
    (configure option –enable-wide-chars defines OPT_WIDE_CHARS,
    invoke xterm with -wc option to activate this feature). This
    patch optionally widens internal data structures, invokes the
    16-bit text output rather than the 8-bit version and adds some
  • further fixes for terminfo: ka1, ka3, etc., differ between
    the default xterm-xfree86 and xterm-vt220 entries.
  • change default (with sunKeyboard resource false) behavior
    of the editing keypad “Delete” to send a 127, like xterm-r6.
    The VT220-style <ESC>[3~ is sent when
    sunKeyboard is true (reported by Tomas Vanhala).
  • add parameters to function keys to indicate if shift,
    control or alt are set. The codes are based on a description of
    a DEC VT510 with a PC keyboard, from Jeffrey Altman
  • add control sequence 1035, set-num-lock action and num-lock
    menu entry to control the use of the NumLock and Alt keys for
    the Sun/PC and VT220 keyboard extensions.
  • implement DECSET/DECRST numeric keypad (DECNKM) mode.
  • modify terminfo and termcap to use recommended “X Window”
    or “X11” names rather than “X Windows” (reported by Tomas
  • suppress translation of shifted keypad “+” when sunKeyboard
    is true.
  • workaround unexpected behavior (perhaps bug) in
    XmbLookupString, which returns trash in the string buffer for
    numlock and control-key combined with keypad-keys.
  • modify ScrollBarReverseVideo() to keep scrollbar border
    visible when reverse video is toggled.
  • correct missing case for parameter 17 (set highlight color)
    in dynamic colors control sequences.
  • extend dynamic colors control sequences to allow users to
    determine the colors and font which are currently active.
  • minor tweak to OSC responses, to use BEL if the application
    used that to end the request, rather than ST. This works better
    with shell scripts, which may not handle an
    <ESC>backslash very well.
  • separate menu settings for reverse video from that done
    under program control.
  • corrected ifdef’s for menus, which did not allow tek4014 to
    be suppressed properly (reported by Clint Olsen).
  • changes for Linux-2.2.x with GLibc-2.1 and /dev/ptmx
    support (integrated patch from Pavel Roskin):

    • main.c and resize.c were using different rules to
      determine whether ATT should be defined (actually
      USE_USG_PTYS is more apt).
    • copy definitions from main.c to resize.c to prevent
      sys/stream.h and sys/ptem.h from inclusion when SYSV is not
    • define CNUL if not already defined like other
    • /dev/tty does exist in Linux, but it doesn’t mean, it
      should be used. Therefore EACCES is now an acceptable
    • ifdef’d several calls such as ioctl (ptyfd,
      I_PUSH, "ptem")
      to build on Linux (I_PUSH is not
      defined when sys/stropts.h is not included).
  • initialize second “ltc” variable in main.c (reported by
    David Dawes).
  • provide definition for USE_USG_PTYS in screen.c
  • add resource-files to install rule in standalone
  • add sample scripts to illustrate titlebar controls,
    resizing and colors.

Here are several fixes and minor enhancements. The chief ones
are the fixes for NumLock mode and reverse video, since we had
become used to working around the problems.

  • remove kfnd/kll/kslt strings from terminfo, because curses
    applications do not necessarily return khome/kend pairs
    (reported by Vikas Agnihotri).
  • implement NumLock resource which overrides the keyboard
    tables for the special case of keypad keys. This is a problem
    introduced in xterm by X11R6 changes, i.e., an ambiguity which
    in effect discarded vt100 keypad support.
  • modify Sun/PC keyboard mode to extend this (emulation of
    DEC vt100 keypad) to the remainder of the numeric keypad. Now,
    the default operating mode of xterm uses the keyboard tables
    as-is (except if the NumLock mode overrides), but provides good
    vt100 keypad compatibility if the Sun/PC keyboard menu item is
  • separate command-line settings for reverse video from that
    done under program control. This is a problem which was
    introduced by X11R6. Though correct, most users are
    confused by allowing the reset command to undo the effect of
    the command-line -rv option.
  • add description of function keys, keypad and cursor keys to
  • add terminfo entries for xterm-vt52, xterm-sun and
  • correct typo (missing case value) for DECSET 35,
    enable/disable shifted keypad action and a few compiler
    warnings (reported by Zdenek Sekera <zs@sgi.com>).
  • correct reporting of color values 8-15 in DECRQSS (reported
    by Vikas Agnihotri).
  • modify parsing of ttyModes resource to recognize “^-” as
    “undef” (requested by Tomas Vanhala).
  • integrate/extend changes to add iconify/maximize actions
    (from Edward S. Arthur <eda@ultranet.com>).
  • add control sequences for maximizing/restoring window, and
    for reporting maximum screen size.
  • add ‘interpret’ action, to support local function-key
    interpretation. Used properly, this makes most of the
    specialized actions of xterm redundant.
  • add control sequence private modes 1051, 1052 and 1052, for
    setting the Sun and HP function key modes, and for setting the
    Sun/PC keyboard mode.
  • add configure option –disable-maximize
  • add configure option –disable-num-lcok
  • extend descriptions of configure script options in
  • increase buffer size for tgetent (i.e., termcap) to 1500.
    This fixes a problem where screen is built using
    ncurses or GNU termcap and xterm is built using the
    standard termcap interface. The former does not limit the
    termcap size, while the latter is assumed to be no longer than
    1023 characters. The screen program’s termcap entry is
    about 1200 characters long.
  • change update_menu_item() to a function, to simplify
    debugging. This also reduces the executable by 4Kb.
  • add control sequences for DECSET 30, 1010, 1011 like rxvt
    (enable or disable some features that were only settable via
    resources or command line arguments).
  • add control sequence for DECSET 35, which enables/disables
    the shifted keypad functions.
  • add support for switching font sizes, by stepping through
    the font menu using shifted keypad plus and minus.
  • correct missing initialization of tekInhibit and tekSmall
  • correct ifdef’s in charproc.c for XtNgeometry and
    XtCGeometry (reported by Bram Moolenaar).
  • Implement logic to translate input characters which are
    mapped when in vt220 National Replacement Character mode
    (requested by Tomas Vanhala).
  • Resync configure scripts with my patches to autoconf
  • Change order of -lXmu and -lXext to accommodate cygwin32
    (reported by Vikas Agnihotri).
  • Add “-ti” option to set terminal emulation level from
    command line rather than via resource.
  • Simplify some of the preprocessor logic using #elif.

This implements several small fixes and enhancements. The
chief one implements fallback support for line drawing characters
with fonts that do not include those characters. But I
implemented that last.

  • If any of the glyphs in positions 0-31 (used by xterm to
    implement the VT100 alternate character set) are zero-sized
    (i.e., missing), xterm will reserve a normal space for the
    glyph when drawing. I implemented a simple stroke-drawing
    function to draw the line-drawing characters and a couple of
    the other (simpler) characters such as diamond. (This was
    suggested by Vikas Agnihotri).
  • Modify the computation of doublesize characters to work
    around font servers which shift the scaled characters up/down
    or do not give the correct width.
  • Add popup menu item and corresponding resource settings to
    disable the font-scaling logic for doublesize characters to
    work around (older) font servers which simply do not draw the
    fonts scaled to the size that they said they would. I see this
    problem on a SunOS system running X11R5; the X11R6 servers seem
    well behaved.
  • Implement new escape sequence, private mode 1049, which
    combines the switch to/from alternate screen mode with screen
    clearing and cursor save/restore. Unlike the existing escape
    sequence, this clears the alternate screen when switching to it
    rather than when switching to the normal screen, thus retaining
    the alternate screen contents for select/paste operations.

    When I implemented the popup menu entry to toggle between
    the normal and alternate screens, I considered only pasting
    from the normal screen to the alternate; this improvement
    allows either direction.

  • Changed the termcap and terminfo for xterm-xfree86 and
    xterm-8bit to use the new 1049 private mode.
  • Modify the logic which switches between normal and
    alternate screens so that the save/restore cursor operations
    apply only to the current screen. That means that applications
    which use the terminfo smcur/rmcur or termcap ti/te
    capabilities will restore the cursor to the original position
    on the normal screen rather than to the most recent place where
    a save-cursor operation was performed.

    I note that a real VT100 terminal would not behave in this
    way, but it is a moot point since the VT100 does not
    implement alternate screen, and therefore the save/restore
    cursor sequence would not be used in this context. I reviewed
    the logic which switches between normal and alternate screens
    based on some recent newsgroup postings as well as a proposed
    patch in the Debian group which attempts to do this (the
    patch has a bug, however, so I did not use it).

  • Add popup menu entry for toggling the titeInhibit
  • Add new resource answerbackString, which overrides the
    default “xterm” returned by xterm when responding to an ENQ
    (control/E) character (request by Rajesh Vaidheeswarran
  • Add new resource keyboardDialect for setting the NRC
    display character set (request by Tomas Vanhala, who notes that
    I should add logic to translate the keyboard as well).
  • Add new command-line option and corresponding resources for
    making xterm generate escape sequences compatible with HP
    terminals. Like the existing Sun escape sequences, this is
    available as a popup menu item. It is enabled by the configure
    script with the option –enable-hp-fkeys (requested by Toni
    Mueller <sales@oeko.net>).
  • Add configure script option –disable-boxchars to disable
    the fallback support for line drawing characters.
  • Fix ifdef’s for configure script –disable-ansi-color, and
    reviewed all configure options to ensure that all can be
    enabled/disabled appropriately.
  • Update config.guess and config.sub, from Lynx.

This patch completes the implementation of double-sized
character support for the VT100 emulation, and fixes a few minor

  • corrected the cursor position in HideCursor, which did not
    multiply the column by two when in doublesize mode. This bug,
    which did not appear in normal use, dates back to my original
    changes to partly implement
    double-sized characters. I noticed it when cat’ing a typescript
    from vttest’s double-sized character test.
  • ensure that the current line is repainted when switching
    between single and double width characters.
  • reduce the number of bits used for double-sized character
    coding from 3 to 2, to make more room for soft-font codes.
  • copy newer ifdef’s from the XFree86 3.3.3 release’s main.c,
    which address details of glibc and powerpc.
  • moved definition of DECL_ERRNO in xterm.h to match XFree86
  • modify resize to remove the ifdef on SVr4 that
    suppressed printing the script for $LINES and $COLUMNS.
    Solaris’ resize utility does this; suppressing the behavior is

I tested the double-sized characters using vttest and the
xfsft patch. These fonts worked reasonably well:


The iso8859 font does not include box characters, of course,
but looks good.

This refines my patch #85 by checking
for a case where the font server returns a bold font that does
not quite correspond to the normal font. When I asked for a bold
version of the 6×12 font, the font server returned a near
equivalent where the ascent and descent values did not match,
causing xterm to leave lines across the display. This patch makes
xterm reject that type of mismatch, falling back to the old font

This corrects a problem reported by Stefan Dalibor. My table
entries for the printerExtent and
printerFormFeed resources used the wrong types for
sizeof, causing the printer to not work properly on some
platforms because the printAttributes resource was
overwritten (e.g., Digital Unix with DEC Alpha).

I reviewed all of the resource table entries and fixed a
potential problem with resizeGravity, which was typed as
int rather than XtGravity.

A small fix: when switching to/from the continuous mouse
reporting mode, the event mask was incorrect if xterm was built
under X11R5 because the original data was saved under an ifdef
for active-icon.

Some cleanup (I noticed the pixel droppings a few months ago
after getting a new 17″ monitor):

  • split-out the functions that load/reload fonts for xterm as
    new file fontutils.c to extend them to support automatic
    loading of bold font corresponding to the normal font. Adapted
    logic from EMU 1.3 for this purpose. This fixes most
    occurrences of pixel droppings from bold characters.
  • modified by renaming variables and adding casts to permit
    compile with g++.
  • modify headers so they can all be compiled without order
    dependencies, and use bracketed rather than quoting includes to
    allow compile from separate directory.
  • renamed xtermm terminfo entry to xterm-mono, to avoid
    conflict with Solaris entry now in ncurses.

Several small fixes and enhancements, including a patch from
Bjorn Helgaas:

  • correct initialization of TERMCAP environment variable,
    which was frequently set to garbage on non-Linux systems
    because “termcap” and “newtc” contained random data from the
  • remove the LINEWRAPPED attribute from lines as they are
    cleared. Previously, the attribute was set but never
  • make word and line selections work even when lines are
    wrapped by xterm. Previously, selections were limited to one
    screen line.

as well as

  • modify logic for line-wrapping to reset the associated flag
    if the application deletes a character. This fixes a problem
    reported by Bjorn Helgaas where word/line selection would still
    wrap even after deleting characters from the first line.
  • add kll/kH capabilities to xterm-xfree86 entries in termcap
    and terminfo (request by Michael Schroeder).
  • shorten descriptions in termcap to make resolved entries
    all shorter than 1023 character limit.
  • use DECSTR control sequence to shorten initialization and
    reset strings in xterm-xfree86 termcap and terminfo.
  • use rmul/smul and rmir/smir in xterm-r5 terminfo to match
  • correct typos in initialization and reset strings in
    xterm-r5 termcap and terminfo.
  • disable special translations of key symbols (the backarrow
    key, the editing keypad and the keypad “+”) when a modifier
    other than shift, control or numlock is used. This makes xterm
    handle the meta key as expected by people using emacs (reported
    by Pete Harlan <harlan@pointofchoice.com> in
    linux.debian.user newsgroup).

Merge changes through patch #83 with
the 3.3.2e version.

This follows a change from XFree86 CVS which adds checks for
non-null return from malloc (patch by Matthieu Herrb).

Several small fixes and enhancements:

  • add configure test to infer if xterm should be installed
    setuid based on previously installed xterm (reported by Stephen
    Marley and Stefan Dalibor).
  • implement quasi-continuous mouse reporting (integrated
    patch by Jason Bacon).
  • correct control sequences transmitted by function keys F1
    to F4 when sunFunctionKeys resource is true (it was still using
    the VT100 control sequences).
  • modify handling of backarrow key so that the control
    modifier toggles the backspace/delete interpretation set by the
    backarrowKey resource.
  • limit the row and column values used to report mouse

From resync with XFree86 3.9Nb:

  • add support for the VSTATUS control character (patch by
    Robert Earl <rearl@teleport.com>).

Bug fix for patch #81:

  • remove an #undef for NBBY, which caused compile problem on
    FreeBSD (reported by David Dawes).

More bug fixes:

  • fix conflicting ifdef’s for TIOCSLTC when building with
    Linux on AXP aka DEC Alpha (reported by Robin Cutshaw).
  • correction to patch #79, move assignment for
    *utptr->ut_user = 0; back into non-SVR4 ifdef in main.c,
    since this clobbers username on Solaris 2.5.1 (analysis by Will
    Day <willday@rom.oit.gatech.edu>).
  • corrected khome/kend in xterm-8bit terminfo
  • improve cursor color by making it always the reverse of
    fg/bg unless the cursorColor resource is set, i.e., to
    something other than the default foreground (reported by Vikas
    Agnihotri and Bram Moolenaar).
  • minor fixes for compiler warnings, including a syntax error
    in the AMOEBA ifdef’s.

From resync with XFree86 3.9Aj:

  • correction to patch #73, supply missing #else for fallback
    definitions of size_t, time_t (Robin Cutshaw

This fixes a couple of items leftover or introduced by patch

  • take out the logic that suppresses editing-keypad in vt100
    mode (those keys are too valuable to give up for a fine point
    of emulation).
  • corrected a couple of places in terminfo where I missed
    using the vt220-style editing-keypad codes for Find/Select
    (mapped to Home/End).
  • checked termcap file against terminfo, fix several places
    where it was inconsistent.

This patch fixes several small bugs:

  • use X_EXTRA_LIBS in standalone Makefile.in (patch by Tomas
  • add new resource ‘oldXtermFKeys’ which provides backward
    compatibility for F1-F4 control sequences with TOG’s
  • determine the server’s default foreground/background Pixel
    values, needed if the -flipPixels X server option is used
    (reported by David Dawes).
  • correct logic for F1-F4 function keys so that they generate
    VT100 compatible escape sequences in VT220 mode, since that is
    what the numeric keypad is supposed to do (reported by Ron
    Johnson, Jr. <ronjohn@communique.net>).
  • modify logic for editing keypad to work more like VT220:
    don’t pass those codes in VT100-mode unless oldXtermFKeys mode
    is set.
  • correct an interaction with the editing-keypad logic that
    prevented backarrow key from sending a 127 for the delete key
    (the ‘remove’ escape sequence was being sent instead).
  • use return-value from getuid() rather than the parameter,
    to work on systems which do not update the latter (patch by
    Kevin Buhr <buhr@mozart.stat.wisc.edu>)
  • correct scrollbar border color when toggling to/from secure
    keyboard mode (patch by Jeff Uphoff
  • modifications to terminfo file:
    • rename description to xterm-xfree86, adding an entry
      ‘xterm’ which is derived from xterm-xfree86 to simplify
    • change string for kdch1 from \177 to \E[3~, to address
      complaints from Debian developers. (This applies to the key
      labeled “Delete”, and does not affect the backarrow
    • add user-strings u6, u7, u8, u9 to entries to make them
      work with Daniel Weaver’s “tack” program.
    • modify xterm-24, xterm-bold, xterm-boldso entries to
      disentangle them from explicit dependency upon
    • add generic ‘xterm-color’ entry.
    • minor corrections to xterm-xfree86, xterm-8bit,
      xterm-r5 and xterm-r6 entries.

Again, most of the bulk of this patch is for ANSI conversion.
I used the IRIX compiler’s -wlint option to find the remaining
functions that use K&R syntax (since gcc does not do this
properly). Also, I changed the logic in the print code slightly
to close unwanted files before opening a pipe to the printer.
This may help in some configurations where the line printer hangs
until xterm closes the pipe.

Most of the bulk of this patch is to convert the extended C
(K&R functions with prototypes) to ANSI. I verified that on
Linux by comparing object files, to ensure that I did not, for
example, interchange parameters in the function declarations.

The rest of the patch fixes several minor bugs, and adds a few

  • back-out my use of ncurses “captoinfo -f” option
    (if/then/else/endif) formatting, since ncurses did not
    correctly filter embedded newlines in terminfo capability
    strings until _after_ ncurses 4.1, thus corrupting
    setf/setb/sgr strings (reported by Darren Hiebert and
  • document SGR 8, 28 (invisible/visible), add corresponding
    capabilities to terminfo description.

    I would also add the ‘prot protected capability, but the
    control sequences for that would not be recognized or
    properly ignored by the older xterm programs.

  • modify ChangeGroup to not suppress a null-resource, but
    treat it as an empty string (recommended by Stefan
  • add printerAutoClose resource to control whether printer is
    closed when going offline. More than one person reports
    problems (on Solaris 2.6 and Digital Unix 4.0) getting the
    printer to proceed unless xterm exits; I think that it is a
    problem flushing the pipe. Closing it ought to flush it.
  • adapt TOG fix-3 to in HandleKeymapChange and VTInitI18N
    (but adapt XtStackAlloc/XtStackFree for clarity, as well as
    fixing memory leak).
  • change calls on FillCurBackground to ClearCurBackground, in
    effect using XClearArea rather than XFillRectangle when
    clearing as a side-effect of scrolling and insertion (patch by
    Alexander V Lukyanov)
  • correct some places where insert/delete did not _move_ the
    color attributes, using memmove (patch by Alexander V Lukyanov
  • add ifdef’s for __CYGWIN32__, for port to cygnus version
    B19.1 (patch by Andrew Sumner
  • remove #define for hpux that turned on USE_SYSV_ENVVARS,
    since this causes some applications (such as ‘less’) to get
    incorrect $LINES and $COLUMNS values (reported by Clint
  • modify behavior for HP-UX, to set the “reserved” process
    group controls to _POSIX_VDISABLE so the TIOCSLTC ioctl does
    not produce an error (patch by Ben Yoshino
  • correct length of underlining, reducing it by one pixel to
    avoid leaving a dot when the application does not clear the
    whole screen I noticed this only recently myself (and other
    people as well); it depends on the font chosen. A 6×10 font
    shows it, but the other fonts from XTerm.ad do not. I also see
    a similar problem with the emulation of bold fonts. (patch by
    Sergei Laskavy <Laskavy@cs.msu.su>).
  • add include for Xos.h to xterm.h, to ensure proper
    definitions (reported by Holger Veit).
  • update config.guess to recognize Unixware 2.1 and 7 (patch
    by Mike Hopkirk <hops@sco.com>).

A fix for the print-window function, and some minor

  • modify logic that closes pipe in the print-window function
    to not use pclose, which does not work on all systems, since I
    did not open the pipe with popen (reported by Stefan
  • correct name of $(EXTRA_LOAD_FLAGS) imake variable
    (reported by Stefan Dalibor).
  • guard logic in ChangeGroup() function against null pointer
    (reported by Stefan Dalibor).
  • ensure that menu entry for toggling sunKeyboard resource is
    initialized (reported by Branden Robinson
  • gcc unused-variable warnings (reported by Bernd Ernesti
  • rename global variable ‘buffer’ to ‘VTbuffer’

This incorporates fixes from several people, as well as some
improvements that I made based on feedback from users:

  • add option to standalone configure script,
    “–with-own-terminfo-dir”, which causes xterm to pass a
    predefined value of $TERMINFO to the application. Also use this
    value in install rule “make install-ti”, for the standalone
    configure script. (patch & feedback from Tor Lillqvist
  • change ifdef’s for “hpux” to “__hpux” (patch by Tor
    Lillqvist). The imake configuration uses the former, but the
    latter is predefined by HP’s compilers.
  • define USE_SYSV_ENVVARS for HP-UX, curses does not use
    $TERMCAP (patch by Tor Lillqvist).
  • on HP-UX, use the /dev/ptym/clone device to allocate PTY’s.
    Works both on HP-UX 9 and 10. (patch by Tor Lillqvist).
  • modify configure script to obtain the $(EXTRA_LOADFLAGS)
    value from imake, needed for compiling under DEC OSF/1
    (reported by Stefan Dalibor)
  • modify print-window action so that the printer is closed
    after printing a window, unless the printer was already opened
    (i.e., by an application running in the window). This makes the
    printing complete without having to exit xterm, since not all
    systems treat fflush through a pipe very well (reported by
    Stefan Dalibor)
  • modify printing code so that the ^M’s are not printed if
    the printAttribute resource is set to zero (request by Stefan
  • correct typo in xterm.man, for the default value of
    printAttribute (reported by Stefan Dalibor).
  • correct printing code so that the alternate character set
    is printed, like other attributes, with escape sequences.
  • correct termcap description, removing spurious ‘m’
    character from the ‘op’ capability (reported by Greg Woods
  • correct trace code so that the initial timestamp is
  • correct/modify utmp data for the Linux glibc 2
    configuration, so that the ut_line member is set before using
    it to update wtmp (patch by Bill Nottingham

This corrects a couple of recent bugs and adds a new

  • modify definition of TRACE_CHILD so that it does not
    conflict with ifdef’s for USE_USG_PTYS in main.c (reported by
    Vikas Agnihotri <VikasA@att.com> and Stefan
  • correct ownership of file (actually pipe) written by the
    print controls. When undoing the setuid changes in patch #69, I
    overlooked this. Now xterm forks a process which resets setuid
    and routes the printer data as the real user.
  • add a new resource, “printAttributes”, which controls
    whether color attributes (or any attributes) are sent to the

This patch does the following:

  • implements a print-window facility for xterm. Though useful
    in itself, I added it as part of some debugging which I am
    doing. (Occasionally during an exposure event xterm fails to
    restore the proper colors for highlighted text, and printing
    the screen will show the internal states nicely).

    The print facility now also displays the color
    information. (I will probably make this a resource).

  • modified the xterm-16color terminfo description to use the
    setaf and setab strings, which is a little more efficient
    (requested by Stephen Marley).
  • reduced some clutter of the ifdef/includes (e.g., stdlib.h,
    unistd.h, errno) making the configure script test for these.

    I removed the symbols Size_t and Time_t, since they no
    longer serve a useful purpose.

  • added some debugging traces to show the initial screen
    size, and the success/failure of the ioctl calls pass handle
    window resizing events to the application (for Clint Olsen).

    I also modified the trace code to produce two files since
    some of the information is produced by the child process. So
    now they are Trace-parent.out and Trace-child.out

  • updated configure test for ANSI compiler options to handle
    a special case for HP-UX 10.x (reported by Clint Olsen).

This is a patch from Chris Siebenmann
<cks@hawkwind.utcs.toronto.edu>, which I have cleaned up a
little, and integrated into the configure script.

From his description:

This set of patches is the latest incarnation of patches
originally written by Ian! D Allen, then of the University of
Waterloo and now of who knows where; I have been carrying them
forward from xterm to xterm ever since about X11R4. What they
do is add an option so that when an iconified xterm receives
output it prepends ‘*** ‘ to its icon title and (optionally)
beeps the bell; deiconifying the xterm removes the ‘*** ‘. Over
the years I’ve found this to be incredibly convenient for
monitoring all sorts of low-activity things.

This patch is a slightly modified version of one by Richard
Braakman, which prevents buffer overflow in the input-method and
preedit-type parsing in xterm. I changed a couple of details to
make the code more maintainable, and looked for similar things –
copying into a fixed-size buffer (found none, though I did spot
an unused variable).

This patch corrects some minor bugs in xterm, and fills in
some more details in the VT220 emulation:

  • modify Imakefile to use SpecialCObjectRule for menu and
    data modules, making xterm build properly if logging is enabled
    (patch by Scott Sewall <scott@iprg.nokia.com>).
  • restore “ich” capability to terminfo entries, since Michael
    Schroeder points out that “ich1” is the one which is the
    problem in Solaris vi.
  • correct color of cursor, which would disappear on
    inverse-video since the initialization did not compare the
    cursor color against both foreground and background (reported
    by Olivier Calle).
  • correct abbreviation-test for -version and -help options,
    making command-line option -vb work (reported by Stefan
  • correct a dependency between ifdef’s for OPT_I18N_SUPPORT
    and OPT_INPUT_METHOD (reported by Stefan Dalibor).
  • split-out character-set translations as new module
    charsets.c, to implement VT220 national replacement character
    sets (a mode where some of the characters are displayed as
    specific European glyphs). This relies on xterm using an ISO
    8859-1 font (approach suggested by Kenneth R. Robinette
  • correct a missing increment, which made SS2 and SS3
    controls repeat the shifted character.
  • add xterm-nrc to terminfo to illustrate the VT220 national
    replacement character sets.
  • reformat terminfo using new ncurses tic option “-f”, which
    makes if/then/else/endif expressions easier to read.

This corrects the problem in xterm with utmp, reported by
Olivier Calle <olivier@tc.fluke.com> as well as some other
people after the XFree86 3.3.2 release.

The error was introduced by my patch
#53 in XFree86 3.9r (October 1997).

The correction removes the reset of setuid/setgid before the
main event loop, and solves the problem which I had worked around
in a different manner: for some reason, the particular waitpid()
call in creat_as() hangs, does not return when the child process
exits. I changed the ifdef’s to force the Linux configuration to
use wait() instead. This call appears to work properly on other
platforms such as SunOS and Solaris.

I also added tests in the standalone configure script to check
for the existence of waitpid().

This corrects another problem with the logic for
highlightColor resource. As reported by David Dawes:

I’ve just noticed a problem with with the “inverse” control
sequence (ESC[7m) with the 3.3.2 xterm. What it does is sets
the background black, and the text the usual foreground colour
rather than simply swapping the foreground/background.

This appears to be because when I added logic to check that
the highlightColor was distinct from foreground and background
colors, I did not add a further check to see that it was not
black (I would make an explicit check for the resource not being
set, but see nothing definite in the headers that would let me
reference an explicitly-undefined Pixel value, though there are
some implications in xpm.h). But this should work.

This patch does the following:

  • improve test for highlightColor so that xterm does not
    attempt to use that unless it is different from the foreground
    and background colors (reported by Stefan Dalibor

    I could not reproduce this until I noticed that the -rev
    option exposed it nicely.

  • remove ich and ich1 from xterm and xterm-8bit terminfo
    descriptions. SVr4 terminfo documents that you must not combine
    these with smir/rmir, but ncurses allows it, opening a hole.
    Solaris 2.6 vi does not work properly if it is using a terminfo
    description when these are combined, since it was written to
    accommodate old terminals that required it (reported by Stephen
  • restore 1-pixel overlap of scrollbar border with left edge
    of window (reported by Jim Burmeister
  • add a configure option, –enable-logfile-exec, which allows
    the user to specify a pipe for logfile. This defines
  • makes the ALLOWLOGFILEEXEC code compile & run, if
  • minor documentation nits.

This fixes the problem reported with failure to build the
resize program (BSDI, OSF/1) due to not having
<termcap.h>, by changing the ifdef to one that would be set
only if the file exists (not currently specified, except by the
standalone configure script). We do not really need to include
<termcap.h> to build, but only for a clean compile, since
it may declare the tgetent() prototype. I also updated the
man-page for resize, since I had recently noticed that
it can be used to resize xterm using the “Sun” control
sequences option.

This is a small patch to xterm’s 8-bit terminal description. I
noticed while testing ncurses that I had specified the wrong
(VT100-style) codes for the F1-F4. Xterm only uses VT220-style
function keys in 8-bit mode.

This fixes the following problems with xterm:

  • save/restore the ANSI foreground and background colors with
    the other visible attributes in the save-cursor and
    restore-cursor operations. This works around a problem with
    vim, which apparently assumed that switching between the normal
    and alternate screens resets the colors (reported by Jim Battle

    It would be nice to implement save/restore cursor as a
    stack (and solve this type of problem completely), but that
    would lead to incompatibility with applications which assume
    they are running with a VT100 or VT220.

  • corrects behavior of a restore-cursor operation which does
    not follow a save-cursor (it is supposed to set the character
    sets to a known initial state).
  • extends the sunKeyboard resource (and menu toggle) to
    modify the home, end and delete keys on a Sun or PC editing
    keypad, making them generate codes compatible with DEC VT220’s
    Find, Select and Remove keys.
  • corrects a length in checking command-line options, which
    caused the “-help” message to not work when X was not
  • add some detail to the man-page (requested by Jason Bacon
  • modify the standalone configure script to ignore the broken
    nsl and socket libraries on IRIX 6.2 (the ones on 5.2 also are
    broken, so this change just widens the check for the system
    version number).

This is another patch from Bjorn Helgaas
<helgaas@rsn.hp.com>, which I’ve reviewed (and learned
some). Following are his notes:

I poked around some more and finally got xterm-62 to build and
run cleanly on HP-UX 10.20. Here are the patches. They look sort
of ugly, so here’s a little explanation:

  • aclocal.m4: Removed side effects from the AC_CACHE_VAL
    commands in CF_FUNC_TGETENT. Previously, LIBS was set inside
    AC_CACHE_VAL, which worked fine the first time configure was
    run, but failed if there were cached values.
  • aclocal.m4: Added temporary setting of LIBS before
    AC_TRY_LINK in CF_FUNC_TGETENT. Previously, the last value set
    by the AC_TRY_RUN loop was used, so only -lcurses was
  • configure.in: Added temporary setting of CPPFLAGS before
    AC_CHECK_HEADERS for X11 files. Previously
    <X11/DECkeysym.h> and <X11/Xpoll.h> were found only
    if they were in the compiler’s default include directories,
    even if `–x-includes=DIR’ had been used or AC_PATH_XTRA had
    found them elsewhere.

The problem on HP-UX was that we were linking with -lcurses
rather than -ltermcap due to the second bullet above, and
apparently something in HP-UX curses is broken. This seems very
strange, because the only thing used is tgetent, which should
affect any tty/pty configuration, but I lost interest in tracking
down the exact problem.

This is a patch mostly by Bjorn Helgaas
<helgaas@dhc.net> (I added the os2main.c change, and a
little of the documentation). From Bjorn’s description:

  • If you use the “-hc <color>” option or set the
    “highlightColor” resource, text is highlighted by changing only
    the background color, rather than using reverse video. I find
    this easier to read, especially when selecting multi-colored
    text, and it is similar to the way Netscape shows
  • Most of the code changes are under “#if
    OPT_HIGHLIGHT_COLOR”. The principal exception is in screen.c,
    where I added a couple calls to resetXtermGC(). This seems like
    it could be a bug even without the color highlighting changes
    (though I don’t pretend to understand all the logic).

This patch modifies the reset behavior of xterm slightly:

  • change the terminfo entry so that rs1 (one of the strings
    used by the ‘reset’ program) does a hard reset rather than
    switching character sets. This is more in accord with other
    terminal descriptions.
  • modifies the treatment of hard reset by the xterm program
    to reset the saved lines.
  • corrects hard reset by also resetting user-defined keys,
    i.e., DECUDK.

This fixes some minor bugs and adds new functionality:

  • add support for blinking text.
    • This does not actually cause the text to flash, but
      text with the blink attribute can be displayed in color,
      using new resources colorBL and colorBLMode.
    • If colors are not used, the blinking text will be
      displayed as before (just like bold). The main purpose of
      this is to make applications work properly when they assume
      the emulator supports blinking text.
    • I did this by moving the per-cell LINEWRAPPED flag to a
      per-line flag, to make room for the new BLINK flag. There
      were no per-line flags, so this changes a lot of
  • corrected missing save-cursor logic in the handling of SGR
    1048 (the new control sequence I added in patch #54,
    1997/10/17). Reported by Darren Hiebert.
  • flush the output of the transparent printing after each
    line Reported by Tomas Vanhala
  • correct the modes that are affected by save/restore cursor
    by adding WRAPAROUND and PROTECTED.
  • corrected placement of one of the XSync calls that I added
    in patch #51, 1997/9/15, which had the side-effect of writing
    on the window border when the xterm was resizing from 132 to 80
  • work around an incompatibility of the XKB definition used
    in xterm versus that symbol from IRIX 6.2’s imake definitions
    (by adjusting the standalone configure script).

My last patch has an off-by-one error in the comparison for
argc. Douglas Kosovic <douglask@dstc.edu.au> showed me
where (he got a core dump). Also, I think this explains Clint
Olsen’s problem, but the symptoms were more subtle (EINVAL for a
system call if the -display option is omitted).

This patch does the following:

  • implement logic to permit xterm to work with proportional

    • Thomas Wolff <Thomas.Wolff@sietec.de> requested
      this (but it isn’t exactly what he’s asking for – that’s a
      more involved task).
    • I chose to do this by rendering the characters on a
      fixed pitch, because it would not be useful for existing
      applications to display varying numbers of characters on
      each line.
    • Except that this forces the display to be wider, it
      works reasonably well. A couple of special cases (reverse +
      colorBD, for example) do not display with proper colors,
      since the inter-character gaps are painted with the
  • added a version number to the program (several people have
    requested this).
  • make the -version and -help options interpreted before the
    program attempts to open the display.
  • minor reorganization of the man-page (ordered the options,
    resources and translations alphabetically – and eliminated some
  • corrected a misspelled filename in Makefile.in, and added a
    lint rule.
  • updated the configure script to correct behavior when it
    cannot find imake, as well as to fix the IRIX+gcc build
    (conflict with /usr/include).
  • regenerated the configure script with a newer patch to
    autoconf that fixes a problem with environment space vs the
    configure –help message.

This patch is mostly concerned with the standalone configure
script; a few minor corrections are added:

  • add configure option –disable-tek4014, to allow xterm to
    be built without the tek4014 emulation.
  • add configure option –with-terminal-type, to allow xterm
    to be compiled with default $TERM value other than “xterm”
    (e.g., “xterm-16color”) — requested by Stephen Marley
  • fix a typo in the configure –help message — reported by
    Darren Hiebert <darren@hmi.com>.
  • review diffs between main.c and os2main.c, to make them
    more alike. (applies some minor bug-fixes to OS/2’s
  • add missing quotes in memmove/bcopy configure test

This patch is based on analysis by Arfst Ludwig
<arfst@luxor.IN-Berlin.DE>, who reported:

Setting the following resources xterm (all current versions)
receives a segmentation fault on <Btn2Up> after

          *XTerm*VT100*translations:    #override \
              ~Shift~Ctrl<Btn2Up>: insert-selection(PRIMARY, CUT_BUFFER0)\n\
              Shift~Ctrl<Btn2Up>:  insert-selection(CLIPBOARD, CUT_BUFFER1)\n\
              ~Shift<BtnUp>:       select-end(PRIMARY, CUT_BUFFER0)\n\
              Shift<BtnUp>:        select-end(CLIPBOARD, CUT_BUFFER1)

(The above resources intention is to be able to paste the
latest selection even if the xterm was cleared.)

And here is how it works (and a fix!): The widget given to
the action handler as first argument is not guaranteed to be a
XtermWidget (it can be the ScrollbarWidget). Instead of
accessing the widget’s member directly XtDisplay gives the
required pointer in a safe way.

I noticed that this was not the only instance (by reading the
code, and testing with his example), and extended the solution to
check the widget-class to ensure that it is indeed xterm’s widget
class before attempting to use it in the context of

This fixes the segmentation violation noted by Rogier Wolff
about a month ago. He’d set xterm to 400 (rows) by 150 columns,
which broke because there were limited buffers (200 rows) used
for juggling data when adding or deleting lines and for switching
between alternate and normal screens. I replaced this by an
allocated buffer.

The bug is simple to test if you set titeInhibit false.

This patch does the following:

  • correct a minor placement problem with the right
  • implement a new set of control sequences for switching
    between the normal and alternate VT100 screens. These work
    around the older sequences limitation that required
    modification of the runtime $TERMCAP to cooperate with the
    titeInhibit resource (that can’t work with terminfo). I do this
    by moving all of the functionality of the rmcur terminfo
    capability into the control sequences.
  • implement the alternate-screen menu entry

This patch adds a fix and implements a new feature (as well as
some minor typos):

  • JCHANDRA@Inf.COM (JCHANDRA) noted that there was still a
    problem with the wait call with the logging option. It hung
    when the logfile was opened as a command-line option. I fixed
    this by moving the StartLogging() call down past the place
    where I’d reset the setuid mode. So the logfile is opened as
    the real user, without having to fork.
  • improve Rohleder’s changes by renaming the command-line
    options and reducing the number of ifdef’s.
  • implement right-scrollbars for xterm (patch by Michael
    Rohleder <michael.rohleder@stadt-frankfurt.de>).

This patch addresses bugs and requests reported by

  • Bob Maynard <rmaynard@montana.com>,
  • Clint Olsen <olsenc@ichips.intel.com>,
  • Michael Schroeder
  • Pablo Ariel Kohan <pablo@memco.co.il>

Some of the changes are interrelated (it was an unusually busy

  • change the default resource value for colorMode to true,
    matching the Xterm.ad file.
  • correct behavior of ‘ech’ control, making the default and 0
    parameters erase one character rather than to the end of line
    (reported by Michael Schroeder).
  • add resource boldColors, command-line options +pc and -pc
    and configure-script option to specify behavior of xterm’s
    mapping bold colors 0 through 7 to colors 8 through 15.
    (request by Pablo Ariel Kohan).
  • add resource colorAttrMode to specify whether colorULMode
    and colorBDMode can override the ANSI colors (report by Clint
  • correct a conflict between colorULMode/colorBDMode versus
    ANSI colors, where exposure events would occasionally pick up
    the former (e.g., colorBD) rather than ANSI colors. Testing the
    colorAttrMode made this apparent, though it has been in the
    code since 3.2A (patch #35 in Jan
  • correct two problems with the optional logging support. On
    Linux at least, the waitpid call in creat_as hangs when the
    logging is toggled from the popup menu. Also, the mktemp
    template has the wrong number of X’s (since X11R5!). Fixed the
    waitpid problem by exploiting the fact that the setuid behavior
    is reset before the popup menus are available. (reported by
    Jayachandran C.).
  • add configure script options for building with the Xaw3d
    and neXtaw libraries.
  • correct CF_IMAKE_CFLAGS standalone configure script macro,
    so that it will pick up $(ALLDEFINES) rather than
    $(STD_DEFINES). This is needed to make scrollbars work on
    Linux, since that uses narrow prototypes. (reported by Bob
  • various minor updates to configure-script macros.

Most of this patch is related to the standalone configure
script, though there are fixes/enhancements as well:

  • add a new resource sunKeyboard, with associated
    command-line option and menu-toggle that allows using a normal
    Sun or PC keyboard to generated the complete DEC-style function
    keys and keypad.
  • correct a reversed foreground/background test in the
    control sequence that replies with the current SGR
  • correct, by invoking XSync, a display problem that caused
    the program to not properly update newly exposed areas when a
    font change or 80/132 resize request was not completely
  • restructured autoconf macros (I made a library of all of
    the macros across the complicated configure scripts I’m working
  • use the autoconf config.guess and config.sub scripts to
    better identify the host-os.
  • improve the configure script that uses ‘imake’ as a
    fallback for definitions.
  • correct several instances of unsigned/signed mixed

I’ve tested the configure script on Linux, SunOS 4.1.3,
Solaris 2.5.1, IRIX 5.2 and 6.2, AIX 3.2.5 and CLIX 3.1 (all but
the last run properly as well).

This is a collection of small fixes, and a couple of minor

  • plug a security hole in the implementation of Media Copy
    (print) by invoking setuid just before the main loop.
  • add an ifdef’d include for <sys/termio.h> for HP-UX,
    which allows the program to process SIGWINCH events (this is a
    bug in X11R6.3)
  • add state-table entries for VT52 emulation to enter/exit
    keypad application mode.
  • disable the popup-menu entry for 8-bit controls when the
    terminal-id is less than 200 (e.g., VT52 or VT100).
  • ensure that the popup-menu entry for 8-bit controls is
    updated when the application enables/disables this mode,
    including the response to a full-reset.
  • implement VT300 DECBKM feature: set interpretation of the
    backarrow key to either backspace or delete. The initial
    setting is via resource; it can also be modified in the main
    popup menu.
  • implement VTxxx KAM (ISO AM), which allows a keyboard to be
    locked (i.e., the terminal discards input).
  • implement VTxxx SRM, which is used to control local echoing
    of input on the terminal.
  • add terminfo and termcap entries for xterm-8bit, a
    variation of the xterm description that uses 8-bit control
  • add fallback definitions for Imakefile to allow it to work
    on some X11R5 systems that have no SpecialCObjectRule or
    ProgramTargetName macros.
  • add .c.o and .c.i rules to standalone Makefile.in
  • correct order of -lXmu and -lXext in standalone configure
  • add configure script options to allow selective disabling
    of active-icon, input-method and i18n code (mainly for users
    with X11R5 or an incomplete X11R6 configuration).
  • change menu-indices from #define’s to enum values, thereby
    making it work better with the ifdef’s for logging and
    active-icon (the X11R6.3 active-icon code is incorrectly
    ifdef’d; this corrects an error introduced by incorporating
    that code).
  • correct minor compile errors in the configuration where
    active-icon is not used.
  • add configure option to suppress echoing of long compiler
  • correct spelling of decTerminalID in configure script help
  • use gcc __attribute__((unused)) to quiet warnings about
    unused parameters when compiling with -W (to make it simpler to
    find the real problems).

This patch implements the VT100/VT220 Media Copy (i.e.,
print-screen) control sequences.

This patch does the following (all but the first affect only
the standalone configure script):

  • minor correction to positioning of underlines for small
    (e.g., 5×8) font size.

    The existing behavior allowed underlines to be drawn
    outside the character-cell, so they weren’t cleared properly
    under some circumstances.

  • adds more special-case tokens to the standalone configure
    script’s imake-option filter (e.g., “&&”, since a “make
    -n main.o” on my IRIX system uses that shell construct).
  • adds a ‘–enable-logging’ option for the standalone
    configure script.
  • adds check and ifdef’s for the standalone configure script
    to allow for building on platforms with X11R6, which lacks
    Xpoll.h (introduced in X11R6.1).

This patch does the following

  • corrects an indexing error in the doublesize character
    logic (button.c) that caused core dump (this was reported by J.
  • corrects the logic of ShowCursor when it is painting in a
    doublesize cell (charproc.c).
  • corrects, according to vttest, the behavior when switching
    to doublesize characters and back again (doublechr.c).
  • adds cbt (back_tab) to the terminfo description (this was
    something that I’d overlooked as applicable to curses
    optimization last summer).
  • corrects, for the standalone xmc test, the logic for
    disabling xmc.

This is a patch to provide test-support for some work I’m
doing on ncurses. It does not modify the normal configuration of
xterm; the code is compiled if the standalone configure option
“–enable-xmc-glitch” is specified.

This fixes the problem reported with xterm’s cursor color
versus the background (the second chunk in this patch) and also
removes some duplicate initialization of the cursor GC’s. If the
cursor color at startup is the same as the background, then xterm
will use the reverse GC, ignoring the setting of the colorMode

This implements the first part of the VT100 doublesize
characters for xterm, as well as fixing a handful of bugs:

  • the doublesize character support uses the normal font
    (using scaled fonts will be another patch) with blanks to
    simulate doublesize characters. This patch does most of the
    global changes that’ll be required. I’ve hidden most of the
    details in macros and ifdef’s so it’s easy to configure out
    (part of the patch is a configure option for that
  • corrected limits in DeleteChar() function — it’s always
    ignored the size of the left border and scrollbar. I noticed
    this when working on the doublesize characters since the glitch
    was doubled in size (i.e., it wrapped some garbage around the
    right margin).
  • corrected ‘memmove()’ logic, for standalone builds (it
    referenced a malloc wrapper from my development library).
  • add a check for HideCursor() to prevent repeated screen
    updates (which can cause a spurious cursor glitch to appear,
    e.g., during scrolling). I noticed this with the 3.2A version
    (but only a few weeks ago, when I started working on this

Here’s a fix for two problems:

  • modify handling of tgetent results in xterm and resize
    programs to make them tolerant of missing termcap file, or
    unknown terminal name. In this scheme, an explicit “-tn” option
    will succeed, overriding the fallback list.
  • a nit in the configure script (log extra information to
    help diagnose which case of the test-compile of tgetent was

Bram Moolenaar reported that the cursor color changed
unexpectedly while scrolling. The cause was that it used the same
GC’s as the logic that draws the ANSI colors. The bug only
appears if the cursorColor resource isn’t set, and has been
present since the initial implementation early last year. (The
same bug also appears in rxvt ;-). Here’s a fix.

Some nits found by Darren Hiebert (missing part of
install-rule, incorrect assignment for –enable-color-mode

Patch for the configure script’s logic for obtaining imake
predefined symbols.

This patch does the following:

  • integrate the 16-color change for ‘xterm’
  • minor fixes/clarification of tgetent in terminfo vs termcap
    to ‘resize’

(both changes also modify the configure script)

This implements a simple configuration script with autoconf
(to which I’ll add more options later). It does the

  • configures xterm to build with X11R5 (at least on SunOS
    4.1, Solaris 2.4, possibly IRIX – sorry network was down today,
    but I did test an earlier version yesterday).
  • enables/disables the configuration ifdef’s for ANSI color
    and VT52 emulation.

It does not make tests for the things that imake does (that’s
another project), instead it uses a hybrid of the autoconf tests
for libraries and adds imake’s compiler options (which are
necessary in some cases to get main.c to compile).

This corrects a minor, but annoying error in the vt220
emulation: the DECUDK is only supposed to be interpreted for
_shifted_ function keys.

This corrects something that I overlooked in patch #27
(21-aug-1996), which is that when trimming the region to be
repainted for the highlightSelection resource of xterm, I still
have to paint the background past the highlighted region. This
only happens when I first do a selection in a window that’s
partly off-screen, then move the window on-screen.

This patch does the following:

  • combines the coding for foreground and background colors
    into a single byte, reducing the memory required to store
    saved-lines in color. (I’ll take back that byte in a following
    patch to use to ensure the character-set, so there’s no
    long-term decrease in memory use).
  • modifies the PF1-PF4 coding in termcap/terminfo. Because
    xterm is still by default emulating vt100, the function key
    codes are vt100-compatible (I overlooked this in patch #31). I
    also reformatted the whole terminfo file into a single-column,
    for consistency.
  • adds an interim xterm-vt220 description to accommodate the
    old and new styles of function-keys (though probably it’d be
    better to drop the old-style altogether).

This patch does the following:

  • implement DECSTR (soft terminal reset). The biggest diff is
    due to adding another state table (note that there’s only one
    useful state here, but it’s only 256 bytes rather than 1k as it
    would have been before I reduced the size of state
  • some minor tidying up (e.g., signed/unsigned use bitcpy,
    MODE_DECCKM, resetColor, resetCharsets). More is done in
    patch #35.

I got the description of DECSTR from a vt420 user’s manual.
I’ll do some testing with vttest to ensure that there’s nothing
else to do than what was documented.

This adds to the reset-fix by Matthieu Herrb
<Mathieu.Herrb@mipnet.fr> a small change to make xterm able
to output 8-bit characters in VT100 mode. Applications that run
on real VT100’s don’t do that anyway, and this feature should be
removed sometime after finishing off the VT220 emulation (VT220’s
can do 8-bit characters). That would be a good time to change the
default terminal-id to 220.

This implements the REP (repeat) control for xterm. That isn’t
part of the DEC VTxxx series, but is defined in ISO 6429. (Note
that the base xterm terminal description is not changed
— I added a variant, “xterm-rep”).

This implements vt52 emulation in xterm (ifdef’d so it can be
removed). I’ve been using it for testing for the past month or

From bug-report by <auroux@clipper.ens.fr> (Denis
Auroux), missing reset to ground state. I checked through the
rest of that table and found another, in the
unimplemented MC (screen print).

This patch does the following:

  • corrects the restoration of color for bold/underline color
  • adds a resource ‘decTerminalID’ to control the reporting
    level of xterm (e.g., VT100, VT220).
  • uses the new resource to implement/correct the DA1, DA2 and
    DECRPTUI reports.
  • change valid-response code in DECRQSS from 0 to 1 (the
    manual says 0, but the VT420 terminal I’ve been testing on says

All of these changes are based on vttest 2.6

(Most of the volume in the patch is to add 2 more state tables
for parsing the 2nd/3rd device-attribute controls).

This patch corrects the following reported by Roland Rosenfeld

  • handle SGR 22, 24 and 25 in combination with colorUL and
    colorBD resources. Also noted & fixed reset of colored
    underline/bold with SGR 0.
  • a typo in the termcap (missing ‘[‘), from 3.1.2Dj (my

Roland also complained that he couldn’t use box characters


but that’s a known xterm limitation (the box characters must
be part of the font, in the first 32 locations).

This patch fixes one of my long-term gripes: xterm’s selection
doesn’t clearly show what’s being selected (as per David’s
request, it’s controlled by a resource, which defaults to the
older behavior).

Here’s a patch to fix a problem with xterm’s cut/paste and
another to modify the appearance of the highlighting while
selecting. (The changes are independent, so you can see if the
change to screen.c is desirable).

Here’s a correction for two minor bugs that I picked up in
testing, plus some lint (from Solaris 2.5) where NULL was used

  • make the second alternate font the same as the first
    (that’s what vt420 and dtterm do)
  • corrected DECSCL report when DECSCL hasn’t been set (i.e.,
    don’t return a ’60’).

This patch does several things. In effect, xterm can (I think)
do a reasonably good job of emulating vt220 and vt320 terminals
(as well as it was doing vt100, at any rate ;-).

It does NOT do:

  • soft fonts
  • rigel or sixel graphics

Anyway, I:

  • added ECH, CPL, CNL, SU, SD, CBT, CHT controls
  • added popup-menu for switching between DEC and Sun function
    keys. (corrected alignment err wrt logging entry at that
  • make xterm recognize both 8-bit and 7-bit controls
    (including popup menu for switching modes).
  • add user-definable function keys (aka DECUDK)
  • support concealed text
  • support protected text (both ISO compatible and DEC
    compatible – that’s not the same thing, btw), with SPA, EPA,
    DECSCA, DECSED, DECSEL controls.
  • implement DECSCL.

I’ll be continuing to test this patch for a while, but don’t
expect to add any new functionality (it passes all of the current
tests I’ve built in vttest, but I need to make more tests)..

This removes the blinking cursor I added last week (for
performance reasons). Time-permitting, I’ll revisit this after
3.2 is released (there will be more work after XFree86
3.2, I assume).

I looked more closely at my “double-negative” and realized
that I had been confused by the default color scheme (black on
white) in combination with reverse video. However, I did see that
the original_fg and original_bg data weren’t really used – so I
removed that logic.


  • during initialization, check if ANSI colors are set with
    non fg/bg values, disable color mode if not. This makes xterm
    tolerant of applications that allocate the whole color
  • implemented blinking cursor (default is off)

This patch does the following:

  • fixes some minor typography in the control-sequences
    documentation (it didn’t occur to me til I’d sent the last
    patch that I could use ghostview for previewing the troff
    output 😉
  • adjusts the shell’s background color in ReverseVideo so
    that flicker in resizing is reduced
  • adds an ifdef OPT_ISO_COLORS to allow configuring xterm
    without the ISO color support (saves a lot of memory)
  • used that ifdef to isolate/modify logic so that if the user
    doesn’t have the colorMode enabled, then ISO color support is
    disabled (saving memory).

    (If anyone needs numbers, I had savedLines set to 2000,
    and found a reduction from ~700k to ~400k of allocated
    memory, according to Purify).

This documents the changes in control sequences for window
operations that I added in my previous patch. I’m testing another
patch that allows the user to use less memory if colors aren’t

This patch does the following:

  • fixes the core dump that I reported on IRIX 5.2 (in main.c)

    (it’s worth noting that this bug exists in X11R6.1, so I’d
    like to assume that someone’s already submitted a fix to X

  • change the interpretation of zero rows or columns in a
    resize-window request to use the root window’s size (looking
    more carefully at dtterm, that seems to be what it does).
  • change some memmove calls to memcpy for slightly better
    performance. also, a couple of memset calls to bzero, since
    Quantify says bzero runs 20% faster, I assume because there’s
    one less argument.
  • interpret character sets 1 and 2 (so that vttest gives a
    reasonable result) Both rxvt and dtterm do something
  • fix a minor memory leak in the logic that retrieves the
    window or icon names (Purify found this for me while I ran

This implements the following:

  • escape sequences that act like the CDE dtterm’s window
    operations (though I have implemented the default width and
    height — I’ve seen a rather buggy dtterm running that seems to
    treat width=0 or height=0 literally — maybe that’s a feature,
    not a bug?)
  • minor tweak to the screen-repainting when resizing (I still
    cannot entirely get rid of flicker).
  • still more fixes to terminfo & termcap (I corrected my
    error for the hpa code and added some other stuff by comparing
    to ncurses’ description and rxvt’s).
  • a tweak to the changes by Michael Rohleder for the color
  • re-order attribute codes to allow later implementation of
    protected fields (dtterm supposedly does this; it’s probably
    more useful than blinking or invisible text — that uses up all
    of the available bits without changing the attribute scheme

This patch implements for xterm several minor features from
ISO 6429 which are useful for terminfo applications. The HPA and
VPA control sequences allow cursor movement along a row or
column, cutting down a little on the characters transmitted. The
other codes allow resetting specific graphic rendition attributes
without modifying the other attributes. (now if someone just had
time to implement blinking cursors…)

Adam Tla/lka <atlka@pg.gda.pl> told me a couple of weeks
ago that I’d missed some of the background coloring in xterm. I
investigated, and found that while I’d picked up on the
clear-to-bottom and clear-to-end-of-line operations, I’d
overlooked the insert/delete lines. Just so I wouldn’t overlook
any more of these, I updated a copy of vttest to test ISO colors
and bce (background color erase). This patch introduces a new
function, ClearCurBackground, whose calls replace the direct
XClearArea calls that I’d overlooked. (There’s also a few
compiler warnings fixed, etc 😉

This fixes the problem reported by David Dawes, by making the
50msec select timeout for the Xaw3d arrow scrollbar a resource.
(I made it a boolean for a variety of reasons — to make it a
number, you’d need an additional resource, to avoid breaking the

This patch brings the termcap and terminfo descriptions for
xterm up to date. I made the following changes:

  • reformatted the terminfo description in a single-column
    (this is ok for terminfo, and will simplify future patches —
    can’t do that for termcap, since it would impact buffer
    requirements on some systems).
  • omitted obsolete features in termcap to save a little space
    (bs, pt)
  • added color capabilities to termcap (ut, Co, NC, op, AB,
  • corrected some capabilities (vi, ve)
  • added ‘st’ (set tab)
  • reduced function keys in termcap for ‘xterm’ to 12 because
    color capabilities makes that description larger than 1023
  • created new termcap name ‘xtermm’ (monochrome) to match the
    terminfo list, and make that description have 20 function
  • added corresponding color capabilities to terminfo (bce,
    colors, pairs, op, ncv, setab, setaf)
  • corrected corresponding capabilities in terminfo (civis,
    cnorm, rmcup, smcup)
  • added capabilities (el1, hts)
  • in both, corrected home/end keys to match the code
    correction made by Thomas Mueller in 3.1.2Bk

    => (I’m still considering modifying the code &
    description to match the rxvt program).

This corrects my earlier changes for colors – the inner border
of the xterm was getting painted with the wrong color, since I’d
moved the call to set the background into the logic that tracks
SGR information.

This corrects a memory leak in xterm that happens whenever one
switches fonts.

This patch corrects the behavior of the ANSI colors in xterm
when reverse video is used, as well as some other lesser

  • button.c
    • (compiler warnings: shadowing of ‘time’, redundant
  • charproc.c
    • renamed screen.colors[] array to screen.Acolors[] to
      more easily distinguish the non-ANSI colors from the ANSI
    • moved logic of SGR_Save() into VTInitialize, getting
      rid of local private variables original_fg and
    • moved some logic into getXtermForeground and
      getXtermBackground from SGR_Foreground, SGR_Background,
    • corrected misleading ‘row’ to ‘col’ in case for CUF,
      CUB sequences.
  • ctlseqs.ms
  • xterm.man
    • (correct a misconception which I’d added that the
      color0 through color6 resource values apply to non-ANSI
  • ptyx.h
    • added original_fg, original_bg to TScreen
  • scrollbar.c
    • (compiler warnings: redundant cast)
  • util.c
    • new functions getXtermForeground and getXtermBackground
      replace the macros GET_FG and GET_BG, with the added
      functionality of checking for the reverse-video status of
    • in ReverseVideo, swap the SGR foreground and background
      colors also.

I observed an occasional glitch in the xterm’s color behavior;
a clear to end of line would get a color that had been used in a
program that supposedly reset colors. I traced this down to the
way xterm was modifying colors of GC’s on the fly; it didn’t
restore the original color of the GC, even though it would later
be used in functions (such as ClearRight) that assumed (my error)
that the GC would have the current foreground or background

I fixed this by resetting the GC’s colors with a new function
‘resetXtermGC()’, and direct calls on
SGR_Foreground/SGR_Background, as appropriate and using a new
function ‘updatedXtermGC()’ to encapsulate the logic that
modifies the GC’s color. (I also removed some commented-out code
that was trying to do this — the problem was a little more

This patch fixes the remaining problems that I had making
xterm run with x11r5, as well as a couple of other bugs. It
follows my patch from yesterday, that added ifdef’s for some of
the input-method resources.

  • corrected ifdef’s that suppress the input-method code
    (doesn’t exist in my x11r5, and xterm works adequately without
  • corrected fallback definition for ‘Select()’ macro (oops:
    I’d copied the wrong text…)
  • moved the declarations for the fd_set variables to
  • corrected an ifdef in resize.c (sunos 4.x doesn’t have
  • corrected (in main.c) some unused/orphaned variables.

This is mostly a documentation patch for xterm. It describes
the color control sequences in more detail, and documents some
other features of xterm that aren’t described elsewhere.

I’ve also added a couple of ifdef’s to fix (part of) the
problem that I’m working on (making the program work properly on
x11r5, where I’m doing memory testing — I have a “good” version
from mid-January, but my resync version doesn’t work properly on
x11r5). I’m not done with that yet.

I did a (clean) build of 3.1.2n on Linux 1.2.13 (ELF). I’ve
got an S3 card.

This fixes the following in the 3.1.2n xterm:

  • initialize cur_foreground, cur_background in charproc.c
    (Purify told me they weren’t initialized).
  • add interpretation of codes 39, 49, to reset background and
    foreground to default value (I’m told that ISO 6429 does this;
    but I don’t have a written reference — yet — can anyone help
    here?). Anyway, rxvt does it, and it’ll solve my remaining
    color management problems.
  • shadowing of ‘time’ in menu.c
  • ‘Cardinal’ vs ‘int’ in scrollbar.c
  • several changes to permit compile with X11R5 (the system
    that I’ve got Purify on won’t be upgraded to X11R6 for a long

I built this version (with a minor nit that I’ll patch soon)
on SunOS 4.1.3 so that I can test it some more with Purify.

btw: the changes made in Xpoll.h won’t work on some
older systems, because fd_set isn’t a defined type (I’ve got one
machine at least that this applies to).

This patch does all of the SGR foreground/background fixes
(i.e., clearing the screen after an SGR color is set causes that
color to be used in the foreground and/or background). If the
FG_COLOR and/or BG_COLOR flags aren’t set, then the xterm
foreground and background default to the window’s values. This
usage is consistent with various types of hardware (especially
the IBM PC), and is also used in rxvt.

  • charproc.c:
    • add/use new macros GET_FG, GET_BG – n/c.
    • add/use new functions SGR_Foreground() and
      SGR_Background() to set corresponding colors in GC’s, and
      to retain sense of “original” colors.

      => This makes redundant some of the corresponding
      logic in HideCursor to set the foreground and background,
      but I left it in since it may be fixing an
      unrelated requirement.

    • set GC’s in LoadNewFont() according to whether the SGR
      fg/bg colors are active.

      => This fixes some glitches in the accompanying
      resize, that leaves parts of the window in the original
      background color.

  • screen.c:
    • modified ClearBufRows() to use the SGR fg/bg colors if
      they’re set.
    • added function ScrnClearLines(), used this to replace
      portions of ScrnInsertLine() and ScrnDeleteLine(). The new
      function uses the SGR fg/bg colors if they’re set.

      => Otherwise, selection after an index or reverse
      index will paint the wrong colors.

    • modified ScrnDeleteChar() and ScrnInsertChar() to use
      SGR fg/bg colors.
  • util.c:
    • modified ClearRight() so that if either of the SGR
      fg/bg colors is set, we don’t bzero the attributes and
      color arrays, but instead fill them with the appropriate
    • modified ClearLeft to use SGR fg/bg colors.

This patch modifies the object code, by replacing indexing
expressions with temporary variables with the full indexing
expression. At first glance, this seems inefficient (it did to me
;-), until remembering comments made in the compilers newsgroups
that trying to “help” the compiler doesn’t really work that well.
A good optimizing compiler can do a better job than the
programmer can. (There’s a moral in the use of ‘register’
variables also, but I won’t fix those…).

Anyway, the revised code generates a smaller object…

  • charproc.c:
    • recode index expressions in ShowCursor() and
      HideCursor() using SCRN_BUF_xxxxS macros – changes
    • replace constant ‘4’ by MAX_PTRS – n/c.
  • ptyx.h:
    • defined the SCRN_BUF_xxxxS macros in terms of BUF_xxxxS
      macros, to pick up references to ScrnBuf data directly, and
      added MAX_PTRS symbol to pick up those ‘4’ constants strewn
      about the code – n/c.
  • screen.c:
    • recode index expressions in ScreenWrite() using
      SCRN_BUF_xxxxS macros – changes object.
    • replace constant ‘4’ by MAX_PTRS – n/c.
    • use macros BUF_CHARS, BUF_ATTRS – n/c.
    • cast calloc to ‘Char *’ to fix compiler warning on IRIX
      – n/c
  • scrollbar.c:
    • replace constant ‘4’ by MAX_PTRS – n/c.
    • cast calloc to ‘Char *’ to fix compiler warning on IRIX
      – n/c

When setting up for this phase, I saw that you’d corrected the
bug that I found in ClearLeft. I decided to make this series of
patches anyway, since readability never hurt (and there’s the
potential for finding another bug while reviewing this set).

  • button.c:
    • use SCRN_BUF_xxxxS macros – n/c
  • charproc.c:
    • use SCRN_BUF_xxxxS macros – n/c
  • ptyx.h:
    • added four macros: SCRN_BUF_CHARS, SCRN_BUF_ATTRS,
      SCRN_BUF_FORES, SCRN_BUF_BACKS to represent the four arrays
      that are derived from screen->buf.
  • screen.c:
    • use SCRN_BUF_xxxxS macros – n/c
  • util.c:
    • use SCRN_BUF_xxxxS macros – n/c

This is my third (and final cleanup) patch for xterm. It gets
rid of the unused stuff, and converts several functions to static
(thereby reducing their scope).

At this point, the only compile warnings I’ve got (on Linux)
are those about the select arguments (int vs fd_set type), and a
missing declaration for waitpid. Those both are hard to get right
without autoconfigure.

The next patches will address the functional changes…

  • Tekproc.c:
    • changed several functions to ‘static’ that aren’t used
      outside this module — changes object
  • charproc.c:
    • changed several functions to ‘static’ that aren’t used
      outside this module — changes object
    • deleted unused function unparsefputs — changes
  • main.c:
    • changed several functions to ‘static’ that aren’t used
      outside this module — changes object
    • ifdef’d out unused function ‘consolepr()’ — changes
    • removed unused variable ‘dummy_tio’ — changes
    • moved variable ‘discipline’ to quiet unused-warning —
      changes object
  • main.h:
    • deleted unused definition of DEFBORDERWIDTH – n/c
  • misc.c:
    • changed several functions to ‘static’ that aren’t used
      outside this module — changes object
    • provide dummy return statements for xerror and xioerror
      to quiet compiler warnings — changes object
  • ptyx.h:
    • change sbuf_address and abuf_address to ‘Char *’ –
  • resize.c:
    • changed several functions to ‘static’ that aren’t used
      outside this module — changes object
  • screen.c:
    • remove unnecessary ‘Char **’ casts – n/c
  • util.c:
    • changed several functions to ‘static’ that aren’t used
      outside this module — changes object

This is my second patch to xterm. It corrects most of the gcc
warnings (except for some that are due to X header files ;-). I
compared objects to keep track of the changes that don’t affect
the object code (n/c) versus those that do.

At this point, I’m compiling (fairly) clean with gcc

        -Wall -Wstrict-prototypes -Wmissing-prototypes -Wshadow -Wnested-externs

(I also compiled with -Wshadow, but while that found some
things that I wanted to find, there’s far too many warnings from
the X headers to be usable in this context).

The changes:

  • Tekproc.c:
    • parenthesized expression to avoid gcc warning —
    • corrected nested-extern declaration for Bool
      waiting_for_initial_map; ourTopLevelShellArgs, and
      number_ourTopLevelShellArgs – n/c
  • charproc.c:
    • corrected potentially-unintialized variables ‘scstype’,
      ‘xim’, and ‘input_style’ — changes object.
    • adjusted logic of VTparse so that gcc won’t warn about
      setjmp clobbering parsestate — changes object.
    • corrected initialization of ‘scstype’, which could have
      been clobbered by setjmp/longjmp – changes object.
    • corrected nested-extern declaration of ‘term’,
      ‘ProgramName’ – n/c
  • cursor.c:
    • corrected nested-extern declaration of ‘term’ —
    • renamed ‘term’ parameters to avoid gcc -Wshadow warning
      – n/c
  • input.c:
    • change interface of StringInput to assume size_t (i.e.,
      unsigned) nbytes — changes object.
    • change interface of funcvalue, and sunfuncvalue to use
      ‘KeySym’ type instead of ‘int’ – changes object.
  • main.c:
    • moved definitions of SIGNAL_T, SIGNAL_RETURN to proto.h
      – n/c
    • corrected missing params of ‘do_hangup()’ — changes
      object (note: the missing params were not used).
    • corrected missing param of ‘Error()’ — changes
    • corrected nested-extern ‘environ’ – n/c
    • adjusted assignments to ‘tty_got_hung’ and ‘no_dev_tty’
      so that gcc can see they won’t be clobbered by the longjmp
      – changes object.
    • use Size_t type – n/c.
  • menu.c:
    • removed redundant prototype for ‘do_hangup()’ —
    • renamed ‘time’ parameters to avoid gcc -Wshadow warning
      – n/c
  • menu.h:
    • renamed ‘time’ parameters to avoid gcc -Wshadow warning
      – n/c
  • misc.c:
    • corrected definition of ‘HandleFocusChange()’ —
      changes object
    • cast parameters in call to ‘TekExpose()’ — n/c
    • corrected nested-extern declarations of ‘term’,
      ‘toplevel’, ProgramName, and ‘environ’ — n/c.
    • use Size_t type – n/c.
  • proto.h:
    • moved definition of SIGNAL_T (and SIGNAL_RETURN) here
      from main.c, resize.c to allow use of this symbol in
      prototypes (mostly in xterm.h).
    • added definition ‘Size_t’ to use as corrected type for
      strncpy, malloc sizes – n/c.
  • resize.c:
    • moved SIGNAL_T definition to proto.h — n/c.
    • use Size_t type – n/c.
  • screen.c:
    • use Size_t type – n/c.
  • tabs.c:
    • corrected nested-extern declaration of ‘term’ —
  • util.c:
    • corrected/supplied parameters to ‘TekExpose()’ —
      changes object (note: ‘TekExpose()’ doesn’t use its
    • corrected nested-extern declaration of
      ‘waiting_for_initial_map’ — n/c.
    • renamed ‘term’ parameters to avoid gcc -Wshadow warning
      – n/c
  • xterm.h:
    • prototype ‘do_hangup()’, ‘HandleFocusChange()’,
      ‘TekExpose()’, ‘Error()’, ‘Exit()’ – forces changes in
      various places.
    • adjusted prototypes that pass ‘Boolean’ arguments to
      use ‘int’ (this is the “correct” ANSI approach to extended
      compiles; it’s worth mentioning that gcc doesn’t meet the
      ANSI spec here). I used gcc -Wconversion to find these, but
      there’s a lot of unrelated warnings that are due to setting
      NARROWPROTO in the config – n/c.
    • renamed ‘term’ parameters to avoid gcc -Wshadow warning
      – n/c

This is my first cleanup patch for xterm. It addresses all of
the gcc warnings for -Wall, -Wmissing-prototypes and
-Wstrict-prototypes that I can change without modifying the
object code. (I’m compiling this with gcc 2.7.0 for an aout
target, which makes it simple to compare objects. When I do
ELF-only, I’ve got a tool that compares that sort of thing as

Briefly, this patch adds (and uses) two header files in the
xterm directory:


I expect this to be the biggest patch by far. However (barring
a misplaced prototype), it shouldn’t break anything, since the
intent of the patch is to provide missing declarations.