! ! XTERM RESOURCES ! ! ! with some emphasis on 'Translation Tables' ! ! Documents & Commands: ! ! Thomas E. Dickey's "XFree86 xterm" ! , ! ! ( Dec 22 2000 : xterm-149) ! Both xterm and documentation/manpage are much better than ! the MIT version. It also comes with an FAQ. ! Note that it should compile as well on MIT-X11 systems. ! ! x-faq ! ! (also (old)) ! speedup-x-faq ! ! (also (old)) ! xt-faq ! ! ! Application defaults: ! /usr/X11[R6]/lib/X11/app-defaults/XTerm ! /usr/openwin/lib/app-defaults/XTerm (Solaris) ! ! KeySym names: ! /usr/X11[R6]/include/X11/keysymdef.h ! /usr/openwin/share/include/X11/keysymdef.h (Solaris) ! ! The tools xrdb(1) and xmodmap(1), on Solaris also xmakemap(1). ! ! appres(1) ! for debugging X resources--if you want to know about the actually ! used resources for a certain program. In particular it also pays ! attention to the resources loaded at start-up. ! Note: You have to give the exact Class- and instance name as argument ! to appres(1) (and optionally the instace name). ! ! Find it out with xprop(1) (2nd of "WM_CLASS(STRING)"). ! Althought the MIT proposed the first character uppercase ! (and if the first is an 'X', then also the second), ! the real values vary. Some examples: ! correct: XTerm, Dtterm, Fig ! weird: HPterm, XV, GV, Xev, Xmessage ! For xterm(1) this means "appres XTerm xterm" per default. ! ! xkeycaps(1): useful GUI for xmodmap ! ! with excellent documentation about the keyboard model in X11 ! ! »Using and Specifying X Resources« ! ! ! editres(1). Modify (and find out about) resources on the fly. ! ! About generating your 8bit chars with an american/english keyboard ! ) ! ) ! ! Recent XFree86 xterms know about unicode (UTF-8). ! See also for fonts. ! processing of this resource file additionally leasts at least ! about 50 ms on 'SunOS 5.5 sun4m sparc' Sparc5 [169MHz] ! (difference measured via 'time xterm -e /bin/true') ! Where to obtain resources from: ! See also: ! X(1), XtResolvePathname(3) ! $MIT_SRC/xc/doc/hardcopy/Xt/ or $MIT_SRC/xc/doc/specs/Xt/ (chpt 2, 11) ! ! Highest priority first: ! ! [0. coded properties when creating the widget, we can't influence this] ! ! 1. Command line options bound to resources through XtAppInitialize() ! e.g. -name, -background, -sb (*scrollbar resource) ! ! 2. Resources specified with the -xrm option. ! ! 3. $XENVIRONMENT (filename) or $HOME/.Xdefaults- ! usually too spezialized for all day use, but high priority ! if you need it. See uname() (or gethostname()) for . ! ! 4. Per-screen specifications stored in the SCREEN_RESOURCES, ! i.e. handled with "xrdb -screen ..." ! ! 5. RESOURCE_MANAGER property of the root window, handled with xrdb(1). ! If it is empty, then $HOME/.Xdefaults is used instead. ! ! 6. User resources searched by calling XtResolvePathname(): ! - $XUSERFILESEARCHPATH ! - otherwise XAPPLRESDIR/ (single directory) ! ($XAPPLRESDIR/%L/%N:$XAPPLRESDIR/%l/%N:$XAPPLRESDIR/%N:$HOME/%N) ! - otherwise $HOME/ ! ($HOME/%L/%N:$HOME/%l/%N:$HOME/%N) ! This is the usual mechanism for User stuff. ! ! 7. System resources searched by calling XtResolvePathname(): ! - $XFILESEARCHPATH ! - otherwise (typically) /usr/lib/X11/app-defaults/ ! ! 8. Fallback resources registered by the application ! with XtAppSetFallbackResources() ! ! Note that only the first resource file found in X[USER]FILESEARCHPATH ! is used, if the search path consists of several directories ! ! Usually XAPPLRESDIR might work fine, unless you want to use ! some of the following features: ! ! The format of X[USER]FILESEARCHPATH: ! The easiest way: something like "$HOME/lib/app-defaults/%N", ! then the filename is the class name ("XTerm"). ! See x-faq in particular, it's not really well documented ! anywhere else? Other % shortenings: ! %N The value of the filename parameter, or the ! application's class name. ! %T The value of the file "type". In this case, the ! literal string "app-defaults" ! %C customization resource (R5 only) ! %S Suffix. None for app-defaults. ! %L Language, locale, and codeset (e.g. "ja_JP.EUC") ! %l Language part of %L (e.g. "ja") ! %t The territory part of the display's language string ! %c The codeset part of the display's language string ! ! But be careful with some of these elements, particularly %T,%S in ! non-system PATHs, e.g. XUSERFILESEARCHPATH. ! ! Application defaults are convenient for application spcific ! resources. However, the DISPLAY specific stuff suits better ! in the XROOT property aks xrdb(1). ! ! Meanwhile i found an apparently rather complete summary ! of this (unfortunately german language) ! ! Preprocessor macros ("#ifdef COLOR") ! They are _not_ evaluated in application specific resource files ! (those files that are read at every startup--in contrast to those which are ! read in with xrdb(1) only once at session startup). ! ! #include ... ! #ifdef ! ... [#else] ! #endif ! etc, see cpp(1) ! ! See "xrdb -symbols" about the known preprocessor macros when processing your ! Resource File. Don't rely on the cpp-builtin-macros as xrdb(1) might ! explicitly undefine(!) them when calling cpp (Irix-5 e.g.). ! Usually one gets hardly any error messages. Try anyway... XTerm*StringConversionWarnings: on ! ! Translation Tables: ! ! In general: »Events« are mapped to »Actions« . ! These are often keyboard-events and string()-actions ! ! Syntax: ! * lines must end with '\' - except the last one ! (easier: empty line at end ) ! * single ressources need to be separated with '\n' ! reason: the translation table gets loaded as one string, ! and these '\n' are the delimiters for the single elements. ! * ...and thus comments are impossible in a translation. ! * string(0x0d) is 'return', string(0x1b) is 'escape', etc. ! The feature 'translation table' is provided by libXt, thus it's not ! really available in 'rxvt', because this terminal emulator intentionally ! doesn't link libXt for reasons of performance and size. ! The rxvt maintainer Geoff Wing via email on 'Fri, 7 May 1999 12:04:20 +1000' ! "It could be added as a compile-time define feature. I suspect it hasn't ! been because nobody who has wanted it has written it. It wouldn't have ! been originally added because back then rxvt didn't allow for so many ! compile-time defined features and ones that were perceived to consume ! lots of memory were very low on the list of features to add." ! Makes sense. ! See some hints about the following below. ! (because comments are hardly possible _in_ a translation) XTerm.VT100.translations: #override \ F12: keymap(other) \n\ F1: string("an email address") \n\ F2: string("another email address") \n\ NoneF8: string("LD_LIBRARY_PATH") \n\ MetaF8: string("LC_CTYPE") \n\ CtrlF8: string("XUSERFILESEARCHPATH") \n\ ShiftF8: string("DISPLAY") \n\ \ CtrlLeft: string(0x1b) string("b") \n\ CtrlRight: string(0x1b) string("f") \n\ \ :Metaa: string("ä") \n\ :Metao: string("ö") \n\ :Metau: string("ü") \n\ :MetaA: string("Ä") \n\ :MetaO: string("Ö") \n\ :MetaU: string("Ü") \n\ :Metas: string("ß") \n\ :Metacomma: string("«") \n\ :Metaperiod: string("»") \n\ :Metaslash: string("ø") \n\ :Metam: string("µ") \n\ :Metac: string("©") \n\ :MetaL: string("£") \n\ \ NoneF10: set-vt-font(1) \n\ NoneF11: set-vt-font(d) \n\ \ : \n\ : \n\ \ NonePrior: scroll-back(1,page) \n\ NoneNext: scroll-forw(1,page) \n\ MetaPrior: scroll-back(1) \n\ MetaNext: scroll-forw(1) \n\ ShiftPrior: scroll-back(3) \n\ ShiftNext: scroll-forw(3) \n\ \ \ F2: string("an email address") \n\ F3: string("slogin -l faw ") \n\ F4: string("test -x /amd/bin/bash && exec /amd/bin/bash \ || test -x /bin/bash && exec /bin/bash \ || test -x /usr/local/bin/bash && exec /usr/local/bin/bash -login \ || test -x /com/bin/bash && exec /com/bin/bash \ || echo 'no bash';") string(0x0d)\n\ NoneF9: string("TOP_INCLUDES = -I/usr/openwin/include") \n\ Meta: string("Button1") string(0x0d) \n\ ! keymap(other): more translation tables with more "keymaps" (see below) ! ! Strings: stuff often used, can span accounts - in contrast to ! the "rc" files. ..., email addresses, long strings,... ! if only the terminalemulator is yours. ! ! CtrlLeft/Right: ! For the emacs mode of the readline library (eg bash) ! ! Umlaute: (äöüß...) f.i. for my english Sun keyboard (type 5). ! ! It's a workaround to get »Meta« and Umlauts via ! the _same_ modifier. If you don't need that, better ! define Umlauts (etc) via xmodmap/xkeycaps! ! ! Keynames: Actually, you should use the KeySym names instead of ! the characters themeselves (',' vs 'comma' e.g.). ! See /usr/X11R6/include/X11/keysymdef.h ! /usr/openwin/share/include/X11/keysymdef.h (Solaris). ! The KeySyms in the headers are prefixed with 'XK_'. ! Take care of Caps inside: 'BackSpace' vs. 'Linefeed' ! ! Modifiers: ! '@Meta_Lx' : KeySym itself ! 'Metax' : shortcut for a modifier. possible: ! Ctrl, Shift, Lock, Meta, Hyper, Super, ! Alt, Mod1-5, Button1-5, None, Any. ! 'Nonex' : key must be pressed w/o any modifier ! ":" in front of a modifier : shift/lock get attention, ! "~" in front of a modifier : negation ("without this modifier") ! ! Which KeySyms are mapped on a certain key? ("F12, Meta_L, Control_L", ...) ! consult 'xev' or 'xkeycaps' , customize with 'xmodmap' or 'xkeycaps' ! ! If you want xev to also inform you about the currently pressed ! _modifier_, use the following source patch for xev: ! ! ! BackSpace vs Delete: ! This can become complicated... ! ! ! (and the two following paragraphs) ! ! There are two resources: ! XTerm.VT100.backarrowKey (default: true [backspace]) ! XTerm.backarrowKeyIsErase (default: false) ! ! See also: ! - 'stty erase [...]' ! - Locale settings, ~/.inputrc and /etc/inputrc for bash(1) ! - the bind builtin for pdksh(1) ! ! "Consistent BackSpace and Delete Configuration": ! ! Note that it's neither 'www.ibbnet.org' nor 'www.ibbnet.nl' ! (changed twice in the past). ! ! "Linux Backspace/Delete mini-HOWTO" ! ! ! To get defined but unflexible(!) behaviour, you could add ! "NoneBackSpace: string(0x08) \n\'" above. ! "NoneDelete: string(0x7f) \n\'" above. ! ! [try such with the second translation table below] ! ! Possible events (and actions) in xterm: See T.Dickey's XFree86 XTerm manual ! Another translation table: XTerm.VT100.otherKeymap.translations: #override \ NoneBackSpace: string(0x08) \n\ CtrlBackSpace: string(0x7F) \n\ NoneDelete: string(0x7f) \n\ F12: keymap(None)\n\ F1: string("different keymap! ") \n\ : set-vt-font(d) \n\ : set-vt-font(1) \n ! Do conversion of - ("rhomb" on Suns) to . ! Very useful for the emacs editing mode (f.i. bash/readline), ! e.g: Meta-b/-f. Use "false" then. XTerm*eightBitInput: false ! xfree86 xterm #122 - #127: ! XTerm*metaSendsEscape: true ! Which characters are considered to build a _single_ word ! when doing double-click-Cut&Paste? ! ! Every ascii-value gets mapped to another (arbitrary), already 'valid' ! character, with a colon. ! Here ascii 48 ("0") is the reference. See the manpage. ! ! I don't want these to separate 'words': ! # $ % + - . / @ ~ XTerm.VT100.charClass: 33:48,35-37:48,43:48,45-47:48,64:48,126:48 ! ! Reminder: ! ! " # $ % & ' ( ) * + , - . / ! 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, ! @ [ \ ] ^ _ ` { | } ~ ! 64, 91, 92, 93, 94, 48, 96, 123, 124, 125, 126, ! ! Wether this works also with _wrapped_ selections, depends on ! - don't be confused: MIT X11R5/6 and Suns openwin xterm ! will exclude the newline as well, but the logic for highlightening ! a selection was not updated. XFree86 xterm and rxvt in contrast ! also provide intuitive highlightening. ! - It only works if xterm is wrapping the line itself ! (not always really obvious for the user, though). ! - Among the different vi's, vim actually supports this with a ! clever and little hackish trick (see screen.c): ! ! But before: vim inspects the _name_ of the value of TERM. ! This must be similar to "xterm" (like "xterm-xfree86", which is ! better than "xterm-color", btw, see his FAQ). ! The terminfo entry _itself_ doesn't matter here ! (e.g.: 'xterm' and 'vs100' are the same entry, but with ! the latter it doesn't work). ! ! If vim has to wrap a word, it appends a space at the first part, ! this space will be wrapped by xterm. Going on with writing, vim ! in turn then positions the cursor again at the _beginning_ of this ! next line. Thus, the space is not visible. But xterm now believes ! that the two lines are actually a single one--as xterm _has_ done ! some wrapping also... ! Here comes a simple test for a 'wrapped_selection' (using an 80 chars wide terminal). ! Double click shall select the _whole_ word. ! For the FocusIn/Out-demo above and for ASCII-Art in particular: XTerm.VT100*font1: nil2 ! There shall be colours ! But first: ! ! If the following is not explicitly set, then "*Foreground" ! (with uppercase F, by accident e.g.) will override all other colors. ! To take care of such an accident, use/modify the following table. ! ! In other words: If you ever definitely wanted to avoid any colors, ! even those switched by raw escape sequences(!) then comment out ! the following and use the resource 'XTerm.VT100.Foreground'. ! ! The 'original' colors: XTerm.VT100*color0: black XTerm.VT100*color1: red3 XTerm.VT100*color2: green3 XTerm.VT100*color3: yellow3 XTerm.VT100*color4: blue3 XTerm.VT100*color5: magenta3 XTerm.VT100*color6: cyan3 XTerm.VT100*color7: gray90 XTerm.VT100*color8: gray30 XTerm.VT100*color9: red XTerm.VT100*color10: green XTerm.VT100*color11: yellow XTerm.VT100*color12: blue XTerm.VT100*color13: magenta XTerm.VT100*color14: cyan XTerm.VT100*color15: white XTerm*background: rgb:19/10/50 XTerm*foreground: rgb:ff/dd/70 XTerm*cursorColor: #ffdd70 ! different purposes, different colors, triggered by the instance name ! for the Class XTerm: 'xterm -name wega' wega*background: rgb:05/00/60 lyra*background: rgb:05/00/60 cellar*background: rgb:05/00/60 ultra15*background: black auspex2*background: black w3m*background: black ! local stuff: interesting graphics card problem, flickering (sun50) gk1*background: rgb:00/00/00 gk2*background: rgb:01/00/00 gk3*background: rgb:05/00/60 XTerm.VT100.geometry: 80x40 ! Xterm*geometry would also influence the window menu itself ! (via -