ViewVC Help

View File | Revision Log | Show Annotations | Download File

/cvs/rxvt-unicode/doc/rxvt.7.pod

Comparing rxvt-unicode/doc/rxvt.7.pod (file contents):
Revision 1.97 by root, Tue Jan 31 00:53:49 2006 UTC vs.
Revision 1.102 by root, Tue Jan 31 20:50:48 2006 UTC

…		…
21	The newest version of this document is also available on the World Wide Web at	21	The newest version of this document is also available on the World Wide Web at
22	L<http://cvs.schmorp.de/browse/*checkout*/rxvt-unicode/doc/rxvt.7.html>.	22	L<http://cvs.schmorp.de/browse/*checkout*/rxvt-unicode/doc/rxvt.7.html>.
23		23
24	=head1 FREQUENTLY ASKED QUESTIONS	24	=head1 FREQUENTLY ASKED QUESTIONS
25		25
26	=over 4
27		26
28	=item The new selection selects pieces that are too big, how can I select	27	=head2 Meta, Features & Commandline Issues
29	single words?
30		28
31	Yes. For example, if you want to select alphanumeric words, you can use	29	=head3 My question isn't answered here, can I ask a human?
32	the following resource:
33		30
34	URxvt.selection.pattern-0: ([[:word:]]+)	31	Before sending me mail, you could go to IRC: C<irc.freenode.net>,
		32	channel C<#rxvt-unicode> has some rxvt-unicode enthusiasts that might be
		33	interested in learning about new and exciting problems (but not FAQs :).
35		34
36	If you click more than twice, the selection will be extended	35	=head3 Does it support tabs, can I have a tabbed rxvt-unicode?
37	more and more.
38		36
39	To get a selection that is very similar to the old code, try this pattern:	37	Beginning with version 7.3, there is a perl extension that implements a
		38	simple tabbed terminal. It is installed by default, so any of these should
		39	give you tabs:
40		40
41	URxvt.selection.pattern-0: ([^"&'()*,;<=>?@[\\\\]^`{|})]+)	41	@@RXVT_NAME@@ -pe tabbed
42		42
43	Please also note that the I<LeftClick Shift-LeftClik> combination also
44	selects words like the old code.
45
46	=item I don't like the new selection/popups/hotkeys/perl, how do I
47	change/disable it?
48
49	You can disable the perl extension completely by setting the
50	B<perl-ext-common> resource to the empty string, which also keeps
51	rxvt-unicode from initialising perl, saving memory.
52
53	If you only want to disable specific features, you first have to
54	identify which perl extension is responsible. For this, read the section
55	B<PREPACKAGED EXTENSIONS> in the @@RXVT_NAME@@perl(3) manpage. For
56	example, to disable the B<selection-popup> and B<option-popup>, specify
57	this B<perl-ext-common> resource:
58
59	URxvt.perl-ext-common: default,-selection-popup,-option-popup
60
61	This will keep the default extensions, but disable the two popup
62	extensions. Some extensions can also be configured, for example,
63	scrollback search mode is triggered by B<M-s>. You can move it to any
64	other combination either by setting the B<searchable-scrollback> resource:
65
66	URxvt.searchable-scrollback: CM-s
67
68	=item The cursor moves when selecting text in the current input line, how
69	do I switch this off?
70
71	=item During rlogin/ssh/telnet/etc. sessions, clicking near the cursor
72	outputs strange escape sequences, how do I fix this?
73
74	These are caused by the C<readline> perl extension. Under normal
75	circumstances, it will move your cursor around when you click into the
76	line that contains it. It tries hard not to do this at the wrong moment,
77	but when running a program that doesn't parse cursor movements or in some
78	cases during rlogin sessions, it fails to detect this properly.
79
80	You can permamently switch this feature off by disabling the C<readline>
81	extension:
82
83	URxvt.perl-ext-common: default,-readline	43	URxvt.perl-ext-common: default,tabbed
84		44
85	=item Why doesn't rxvt-unicode read my resources?	45	It will also work fine with tabbing functionality of many window managers
		46	or similar tabbing programs, and its embedding-features allow it to be
		47	embedded into other programs, as witnessed by F<doc/rxvt-tabbed> or
		48	the upcoming C<Gtk2::URxvt> perl module, which features a tabbed urxvt
		49	(murxvt) terminal as an example embedding application.
86		50
87	Well, why, indeed? It does, in a way very similar to other X	51	=head3 How do I know which rxvt-unicode version I'm using?
88	applications. Most importantly, this means that if you or your OS loads
89	resources into the X display (the right way to do it), rxvt-unicode will
90	ignore any resource files in your home directory. It will only read
91	F<$HOME/.Xdefaults> when no resources are attached to the display.
92		52
93	If you have or use an F<$HOME/.Xresources> file, chances are that	53	The version number is displayed with the usage (-h). Also the escape
94	resources are loaded into your X-server. In this case, you have to	54	sequence C<ESC [ 8 n> sets the window title to the version number. When
95	re-login after every change (or run F<xrdb -merge $HOME/.Xresources>).	55	using the @@RXVT_NAME@@c client, the version displayed is that of the
		56	daemon.
96		57
97	Also consider the form resources have to use:	58	=head3 Rxvt-unicode uses gobs of memory, how can I reduce that?
98		59
99	URxvt.resource: value	60	Rxvt-unicode tries to obey the rule of not charging you for something you
		61	don't use. One thing you should try is to configure out all settings that
		62	you don't need, for example, Xft support is a resource hog by design,
		63	when used. Compiling it out ensures that no Xft font will be loaded
		64	accidentally when rxvt-unicode tries to find a font for your characters.
100		65
101	If you want to use another form (there are lots of different ways of	66	Also, many people (me included) like large windows and even larger
102	specifying resources), make sure you understand wether and why it	67	scrollback buffers: Without C<--enable-unicode3>, rxvt-unicode will use
103	works. If unsure, use the form above.	68	6 bytes per screen cell. For a 160x?? window this amounts to almost a
		69	kilobyte per line. A scrollback buffer of 10000 lines will then (if full)
		70	use 10 Megabytes of memory. With C<--enable-unicode3> it gets worse, as
		71	rxvt-unicode then uses 8 bytes per screen cell.
104		72
105	=item I can't get transparency working, what am I doing wrong?	73	=head3 How can I start @@RXVT_NAME@@d in a race-free way?
106		74
107	First of all, transparency isn't officially supported in rxvt-unicode, so	75	Try C<@@RXVT_NAME@@d -f -o>, which tells @@RXVT_NAME@@d to open the
108	you are mostly on your own. Do not bug the author about it (but you may	76	display, create the listening socket and then fork.
109	bug everybody else). Also, if you can't get it working consider it a rite
110	of passage: ... and you failed.
111		77
112	Here are four ways to get transparency. B<Do> read the manpage and option	78	=head3 How do I distinguish wether I'm running rxvt-unicode or a regular xterm? I need this to decide about setting colors etc.
113	descriptions for the programs mentioned and rxvt-unicode. Really, do it!
114		79
115	1. Use inheritPixmap:	80	rxvt and rxvt-unicode always export the variable "COLORTERM", so you can
		81	check and see if that is set. Note that several programs, JED, slrn,
		82	Midnight Commander automatically check this variable to decide whether or
		83	not to use color.
116		84
117	Esetroot wallpaper.jpg	85	=head3 How do I set the correct, full IP address for the DISPLAY variable?
118	@@RXVT_NAME@@ -ip -tint red -sh 40
119		86
120	That works. If you think it doesn't, you lack transparency and tinting	87	If you've compiled rxvt-unicode with DISPLAY_IS_IP and have enabled
121	support, or you are unable to read.	88	insecure mode then it is possible to use the following shell script
		89	snippets to correctly set the display. If your version of rxvt-unicode
		90	wasn't also compiled with ESCZ_ANSWER (as assumed in these snippets) then
		91	the COLORTERM variable can be used to distinguish rxvt-unicode from a
		92	regular xterm.
122		93
123	2. Use a simple pixmap and emulate pseudo-transparency. This enables you	94	Courtesy of Chuck Blake <cblake@BBN.COM> with the following shell script
124	to use effects other than tinting and shading: Just shade/tint/whatever	95	snippets:
125	your picture with gimp:
126		96
127	convert wallpaper.jpg -blur 20x20 -modulate 30 background.xpm	97	# Bourne/Korn/POSIX family of shells:
128	@@RXVT_NAME@@ -pixmap background.xpm -pe automove-background	98	[ ${TERM:-foo} = foo ] && TERM=xterm # assume an xterm if we don't know
		99	if [ ${TERM:-foo} = xterm ]; then
		100	stty -icanon -echo min 0 time 15 # see if enhanced rxvt or not
		101	echo -n '^[Z'
		102	read term_id
		103	stty icanon echo
		104	if [ ""${term_id} = '^[[?1;2C' -a ${DISPLAY:-foo} = foo ]; then
		105	echo -n '^[[7n' # query the rxvt we are in for the DISPLAY string
		106	read DISPLAY # set it in our local shell
		107	fi
		108	fi
129		109
130	That works. If you think it doesn't, you lack XPM and Perl support, or you	110	=head3 How do I compile the manual pages on my own?
131	are unable to read.
132		111
133	3. Use an ARGB visual:	112	You need to have a recent version of perl installed as F</usr/bin/perl>,
		113	one that comes with F<pod2man>, F<pod2text> and F<pod2html>. Then go to
		114	the doc subdirectory and enter C<make alldoc>.
134		115
135	@@RXVT_NAME@@ -depth 32 -fg grey90 -bg rgba:0000/0000/4444/cccc
136
137	This requires XFT support, and the support of your X-server. If that
138	doesn't work for you, blame Xorg and Keith Packard. ARGB visuals aren't
139	there yet, no matter what they claim. Rxvt-Unicode contains the neccessary
140	bugfixes and workarounds for Xft and Xlib to make it work, but that
141	doesn't mean that your WM has the required kludges in place.
142
143	4. Use xcompmgr and let it do the job:
144
145	xprop -frame -f _NET_WM_WINDOW_OPACITY 32c \
146	-set _NET_WM_WINDOW_OPACITY 0xc0000000
147
148	Then click on a window you want to make transparent. Replace C<0xc0000000>
149	by other values to change the degree of opacity. If it doesn't work and
150	your server crashes, you got to keep the pieces.
151
152	=item Isn't rxvt supposed to be small? Don't all those features bloat?	116	=head3 Isn't rxvt-unicode supposed to be small? Don't all those features bloat?
153		117
154	I often get asked about this, and I think, no, they didn't cause extra	118	I often get asked about this, and I think, no, they didn't cause extra
155	bloat. If you compare a minimal rxvt and a minimal urxvt, you can see	119	bloat. If you compare a minimal rxvt and a minimal urxvt, you can see
156	that the urxvt binary is larger (due to some encoding tables always being	120	that the urxvt binary is larger (due to some encoding tables always being
157	compiled in), but it actually uses less memory (RSS) after startup. Even	121	compiled in), but it actually uses less memory (RSS) after startup. Even
…		…
189	(21152k + extra 4204k in separate processes) or konsole (22200k + extra	153	(21152k + extra 4204k in separate processes) or konsole (22200k + extra
190	43180k in daemons that stay around after exit, plus half a minute of	154	43180k in daemons that stay around after exit, plus half a minute of
191	startup time, including the hundreds of warnings it spits out), it fares	155	startup time, including the hundreds of warnings it spits out), it fares
192	extremely well *g*.	156	extremely well *g*.
193		157
194	=item Why C++, isn't that unportable/bloated/uncool?	158	=head3 Why C++, isn't that unportable/bloated/uncool?
195		159
196	Is this a question? :) It comes up very often. The simple answer is: I had	160	Is this a question? :) It comes up very often. The simple answer is: I had
197	to write it, and C++ allowed me to write and maintain it in a fraction	161	to write it, and C++ allowed me to write and maintain it in a fraction
198	of the time and effort (which is a scarce resource for me). Put even	162	of the time and effort (which is a scarce resource for me). Put even
199	shorter: It simply wouldn't exist without C++.	163	shorter: It simply wouldn't exist without C++.
…		…
223	/lib64/ld-linux-x86-64.so.2 (0x00002aaaaaaab000)	187	/lib64/ld-linux-x86-64.so.2 (0x00002aaaaaaab000)
224		188
225	No large bloated libraries (of course, none were linked in statically),	189	No large bloated libraries (of course, none were linked in statically),
226	except maybe libX11 :)	190	except maybe libX11 :)
227		191
228	=item Does it support tabs, can I have a tabbed rxvt-unicode?
229		192
230	Beginning with version 7.3, there is a perl extension that implements a	193	=head2 Rendering, Font & Look and Feel Issues
231	simple tabbed terminal. It is installed by default, so any of these should
232	give you tabs:
233		194
234	@@RXVT_NAME@@ -pe tabbed	195	=head3 I can't get transparency working, what am I doing wrong?
235		196
		197	First of all, transparency isn't officially supported in rxvt-unicode, so
		198	you are mostly on your own. Do not bug the author about it (but you may
		199	bug everybody else). Also, if you can't get it working consider it a rite
		200	of passage: ... and you failed.
		201
		202	Here are four ways to get transparency. B<Do> read the manpage and option
		203	descriptions for the programs mentioned and rxvt-unicode. Really, do it!
		204
		205	1. Use inheritPixmap:
		206
		207	Esetroot wallpaper.jpg
		208	@@RXVT_NAME@@ -ip -tint red -sh 40
		209
		210	That works. If you think it doesn't, you lack transparency and tinting
		211	support, or you are unable to read.
		212
		213	2. Use a simple pixmap and emulate pseudo-transparency. This enables you
		214	to use effects other than tinting and shading: Just shade/tint/whatever
		215	your picture with gimp or any other tool:
		216
		217	convert wallpaper.jpg -blur 20x20 -modulate 30 background.xpm
		218	@@RXVT_NAME@@ -pixmap background.xpm -pe automove-background
		219
		220	That works. If you think it doesn't, you lack XPM and Perl support, or you
		221	are unable to read.
		222
		223	3. Use an ARGB visual:
		224
		225	@@RXVT_NAME@@ -depth 32 -fg grey90 -bg rgba:0000/0000/4444/cccc
		226
		227	This requires XFT support, and the support of your X-server. If that
		228	doesn't work for you, blame Xorg and Keith Packard. ARGB visuals aren't
		229	there yet, no matter what they claim. Rxvt-Unicode contains the neccessary
		230	bugfixes and workarounds for Xft and Xlib to make it work, but that
		231	doesn't mean that your WM has the required kludges in place.
		232
		233	4. Use xcompmgr and let it do the job:
		234
		235	xprop -frame -f _NET_WM_WINDOW_OPACITY 32c \
		236	-set _NET_WM_WINDOW_OPACITY 0xc0000000
		237
		238	Then click on a window you want to make transparent. Replace C<0xc0000000>
		239	by other values to change the degree of opacity. If it doesn't work and
		240	your server crashes, you got to keep the pieces.
		241
		242	=head3 Why do some chinese characters look so different than others?
		243
		244	This is because there is a difference between script and language --
		245	rxvt-unicode does not know which language the text that is output is,
		246	as it only knows the unicode character codes. If rxvt-unicode first
		247	sees a japanese/chinese character, it might choose a japanese font for
		248	display. Subsequent japanese characters will use that font. Now, many
		249	chinese characters aren't represented in japanese fonts, so when the first
		250	non-japanese character comes up, rxvt-unicode will look for a chinese font
		251	-- unfortunately at this point, it will still use the japanese font for
		252	chinese characters that are also in the japanese font.
		253
		254	The workaround is easy: just tag a chinese font at the end of your font
		255	list (see the previous question). The key is to view the font list as
		256	a preference list: If you expect more japanese, list a japanese font
		257	first. If you expect more chinese, put a chinese font first.
		258
		259	In the future it might be possible to switch language preferences at
		260	runtime (the internal data structure has no problem with using different
		261	fonts for the same character at the same time, but no interface for this
		262	has been designed yet).
		263
		264	Until then, you might get away with switching fonts at runtime (see L<Can
		265	I switch the fonts at runtime?> later in this document).
		266
		267	=head3 Why does rxvt-unicode sometimes leave pixel droppings?
		268
		269	Most fonts were not designed for terminal use, which means that character
		270	size varies a lot. A font that is otherwise fine for terminal use might
		271	contain some characters that are simply too wide. Rxvt-unicode will avoid
		272	these characters. For characters that are just "a bit" too wide a special
		273	"careful" rendering mode is used that redraws adjacent characters.
		274
		275	All of this requires that fonts do not lie about character sizes,
		276	however: Xft fonts often draw glyphs larger than their acclaimed bounding
		277	box, and rxvt-unicode has no way of detecting this (the correct way is to
		278	ask for the character bounding box, which unfortunately is wrong in these
		279	cases).
		280
		281	It's not clear (to me at least), wether this is a bug in Xft, freetype,
		282	or the respective font. If you encounter this problem you might try using
		283	the C<-lsp> option to give the font more height. If that doesn't work, you
		284	might be forced to use a different font.
		285
		286	All of this is not a problem when using X11 core fonts, as their bounding
		287	box data is correct.
		288
		289	=head3 How can I keep rxvt-unicode from using reverse video so much?
		290
		291	First of all, make sure you are running with the right terminal settings
		292	(C<TERM=rxvt-unicode>), which will get rid of most of these effects. Then
		293	make sure you have specified colours for italic and bold, as otherwise
		294	rxvt-unicode might use reverse video to simulate the effect:
		295
		296	URxvt.colorBD: white
		297	URxvt.colorIT: green
		298
		299	=head3 Some programs assume totally weird colours (red instead of blue), how can I fix that?
		300
		301	For some unexplainable reason, some rare programs assume a very weird
		302	colour palette when confronted with a terminal with more than the standard
		303	8 colours (rxvt-unicode supports 88). The right fix is, of course, to fix
		304	these programs not to assume non-ISO colours without very good reasons.
		305
		306	In the meantime, you can either edit your C<rxvt-unicode> terminfo
		307	definition to only claim 8 colour support or use C<TERM=rxvt>, which will
		308	fix colours but keep you from using other rxvt-unicode features.
		309
		310	=head3 Can I switch the fonts at runtime?
		311
		312	Yes, using an escape sequence. Try something like this, which has the same
		313	effect as using the C<-fn> switch, and takes effect immediately:
		314
		315	printf '\e]50;%s\007' "9x15bold,xft:Kochi Gothic"
		316
		317	This is useful if you e.g. work primarily with japanese (and prefer a
		318	japanese font), but you have to switch to chinese temporarily, where
		319	japanese fonts would only be in your way.
		320
		321	You can think of this as a kind of manual ISO-2022 switching.
		322
		323	=head3 Why do italic characters look as if clipped?
		324
		325	Many fonts have difficulties with italic characters and hinting. For
		326	example, the otherwise very nicely hinted font C<xft:Bitstream Vera Sans
		327	Mono> completely fails in it's italic face. A workaround might be to
		328	enable freetype autohinting, i.e. like this:
		329
		330	URxvt.italicFont: xft:Bitstream Vera Sans Mono:italic:autohint=true
		331	URxvt.boldItalicFont: xft:Bitstream Vera Sans Mono:bold:italic:autohint=true
		332
		333	=head3 Can I speed up Xft rendering somehow?
		334
		335	Yes, the most obvious way to speed it up is to avoid Xft entirely, as
		336	it is simply slow. If you still want Xft fonts you might try to disable
		337	antialiasing (by appending C<:antialias=false>), which saves lots of
		338	memory and also speeds up rendering considerably.
		339
		340	=head3 Rxvt-unicode doesn't seem to anti-alias its fonts, what is wrong?
		341
		342	Rxvt-unicode will use whatever you specify as a font. If it needs to
		343	fall back to it's default font search list it will prefer X11 core
		344	fonts, because they are small and fast, and then use Xft fonts. It has
		345	antialiasing disabled for most of them, because the author thinks they
		346	look best that way.
		347
		348	If you want antialiasing, you have to specify the fonts manually.
		349
		350	=head3 What's with this bold/blink stuff?
		351
		352	If no bold colour is set via C<colorBD:>, bold will invert text using the
		353	standard foreground colour.
		354
		355	For the standard background colour, blinking will actually make the
		356	text blink when compiled with C<--enable-blinking>. with standard
		357	colours. Without C<--enable-blinking>, the blink attribute will be
		358	ignored.
		359
		360	On ANSI colours, bold/blink attributes are used to set high-intensity
		361	foreground/background colors.
		362
		363	color0-7 are the low-intensity colors.
		364
		365	color8-15 are the corresponding high-intensity colors.
		366
		367	=head3 I don't like the screen colors. How do I change them?
		368
		369	You can change the screen colors at run-time using F<~/.Xdefaults>
		370	resources (or as long-options).
		371
		372	Here are values that are supposed to resemble a VGA screen,
		373	including the murky brown that passes for low-intensity yellow:
		374
		375	URxvt.color0: #000000
		376	URxvt.color1: #A80000
		377	URxvt.color2: #00A800
		378	URxvt.color3: #A8A800
		379	URxvt.color4: #0000A8
		380	URxvt.color5: #A800A8
		381	URxvt.color6: #00A8A8
		382	URxvt.color7: #A8A8A8
		383
		384	URxvt.color8: #000054
		385	URxvt.color9: #FF0054
		386	URxvt.color10: #00FF54
		387	URxvt.color11: #FFFF54
		388	URxvt.color12: #0000FF
		389	URxvt.color13: #FF00FF
		390	URxvt.color14: #00FFFF
		391	URxvt.color15: #FFFFFF
		392
		393	And here is a more complete set of non-standard colors described (not by
		394	me) as "pretty girly".
		395
		396	URxvt.cursorColor: #dc74d1
		397	URxvt.pointerColor: #dc74d1
		398	URxvt.background: #0e0e0e
		399	URxvt.foreground: #4ad5e1
		400	URxvt.color0: #000000
		401	URxvt.color8: #8b8f93
		402	URxvt.color1: #dc74d1
		403	URxvt.color9: #dc74d1
		404	URxvt.color2: #0eb8c7
		405	URxvt.color10: #0eb8c7
		406	URxvt.color3: #dfe37e
		407	URxvt.color11: #dfe37e
		408	URxvt.color5: #9e88f0
		409	URxvt.color13: #9e88f0
		410	URxvt.color6: #73f7ff
		411	URxvt.color14: #73f7ff
		412	URxvt.color7: #e1dddd
		413	URxvt.color15: #e1dddd
		414
		415	=head3 Why do some characters look so much different than others?
		416
		417	See next entry.
		418
		419	=head3 How does rxvt-unicode choose fonts?
		420
		421	Most fonts do not contain the full range of Unicode, which is
		422	fine. Chances are that the font you (or the admin/package maintainer of
		423	your system/os) have specified does not cover all the characters you want
		424	to display.
		425
		426	B<rxvt-unicode> makes a best-effort try at finding a replacement
		427	font. Often the result is fine, but sometimes the chosen font looks
		428	bad/ugly/wrong. Some fonts have totally strange characters that don't
		429	resemble the correct glyph at all, and rxvt-unicode lacks the artificial
		430	intelligence to detect that a specific glyph is wrong: it has to believe
		431	the font that the characters it claims to contain indeed look correct.
		432
		433	In that case, select a font of your taste and add it to the font list,
		434	e.g.:
		435
		436	@@RXVT_NAME@@ -fn basefont,font2,font3...
		437
		438	When rxvt-unicode sees a character, it will first look at the base
		439	font. If the base font does not contain the character, it will go to the
		440	next font, and so on. Specifying your own fonts will also speed up this
		441	search and use less resources within rxvt-unicode and the X-server.
		442
		443	The only limitation is that none of the fonts may be larger than the base
		444	font, as the base font defines the terminal character cell size, which
		445	must be the same due to the way terminals work.
		446
		447
		448	=head2 Keyboard, Mouse & User Interaction
		449
		450	=head3 The new selection selects pieces that are too big, how can I select single words?
		451
		452	If you want to select e.g. alphanumeric words, you can use the following
		453	setting:
		454
		455	URxvt.selection.pattern-0: ([[:word:]]+)
		456
		457	If you click more than twice, the selection will be extended
		458	more and more.
		459
		460	To get a selection that is very similar to the old code, try this pattern:
		461
		462	URxvt.selection.pattern-0: ([^"&'()*,;<=>?@[\\\\]^`{|})]+)
		463
		464	Please also note that the I<LeftClick Shift-LeftClik> combination also
		465	selects words like the old code.
		466
		467	=head3 I don't like the new selection/popups/hotkeys/perl, how do I change/disable it?
		468
		469	You can disable the perl extension completely by setting the
		470	B<perl-ext-common> resource to the empty string, which also keeps
		471	rxvt-unicode from initialising perl, saving memory.
		472
		473	If you only want to disable specific features, you first have to
		474	identify which perl extension is responsible. For this, read the section
		475	B<PREPACKAGED EXTENSIONS> in the @@RXVT_NAME@@perl(3) manpage. For
		476	example, to disable the B<selection-popup> and B<option-popup>, specify
		477	this B<perl-ext-common> resource:
		478
		479	URxvt.perl-ext-common: default,-selection-popup,-option-popup
		480
		481	This will keep the default extensions, but disable the two popup
		482	extensions. Some extensions can also be configured, for example,
		483	scrollback search mode is triggered by B<M-s>. You can move it to any
		484	other combination either by setting the B<searchable-scrollback> resource:
		485
		486	URxvt.searchable-scrollback: CM-s
		487
		488	=head3 The cursor moves when selecting text in the current input line, how do I switch this off?
		489
		490	See next entry.
		491
		492	=head3 During rlogin/ssh/telnet/etc. sessions, clicking near the cursor outputs strange escape sequences, how do I fix this?
		493
		494	These are caused by the C<readline> perl extension. Under normal
		495	circumstances, it will move your cursor around when you click into the
		496	line that contains it. It tries hard not to do this at the wrong moment,
		497	but when running a program that doesn't parse cursor movements or in some
		498	cases during rlogin sessions, it fails to detect this properly.
		499
		500	You can permamently switch this feature off by disabling the C<readline>
		501	extension:
		502
236	URxvt.perl-ext-common: default,tabbed	503	URxvt.perl-ext-common: default,-readline
237		504
238	It will also work fine with tabbing functionality of many window managers	505	=head3 My numerical keypad acts weird and generates differing output?
239	or similar tabbing programs, and its embedding-features allow it to be
240	embedded into other programs, as witnessed by F<doc/rxvt-tabbed> or
241	the upcoming C<Gtk2::URxvt> perl module, which features a tabbed urxvt
242	(murxvt) terminal as an example embedding application.
243		506
244	=item How do I know which rxvt-unicode version I'm using?	507	Some Debian GNUL/Linux users seem to have this problem, although no
		508	specific details were reported so far. It is possible that this is caused
		509	by the wrong C<TERM> setting, although the details of wether and how
		510	this can happen are unknown, as C<TERM=rxvt> should offer a compatible
		511	keymap. See the answer to the previous question, and please report if that
		512	helped.
245		513
246	The version number is displayed with the usage (-h). Also the escape	514	=head3 My Compose (Multi_key) key is no longer working.
247	sequence C<ESC [ 8 n> sets the window title to the version number. When
248	using the @@RXVT_NAME@@c client, the version displayed is that of the
249	daemon.
250		515
251	=item I am using Debian GNU/Linux and have a problem...	516	The most common causes for this are that either your locale is not set
		517	correctly, or you specified a B<preeditStyle> that is not supported by
		518	your input method. For example, if you specified B<OverTheSpot> and
		519	your input method (e.g. the default input method handling Compose keys)
		520	does not support this (for instance because it is not visual), then
		521	rxvt-unicode will continue without an input method.
252		522
253	The Debian GNU/Linux package of rxvt-unicode in sarge contains large	523	In this case either do not specify a B<preeditStyle> or specify more than
254	patches that considerably change the behaviour of rxvt-unicode (but	524	one pre-edit style, such as B<OverTheSpot,Root,None>.
255	unfortunately this notice has been removed). Before reporting a bug to
256	the original rxvt-unicode author please download and install the genuine
257	version (L<http://software.schmorp.de#rxvt-unicode>) and try to reproduce
258	the problem. If you cannot, chances are that the problems are specific to
259	Debian GNU/Linux, in which case it should be reported via the Debian Bug
260	Tracking System (use C<reportbug> to report the bug).
261		525
262	For other problems that also affect the Debian package, you can and	526	=head3 I cannot type C<Ctrl-Shift-2> to get an ASCII NUL character due to ISO 14755
263	probably should use the Debian BTS, too, because, after all, it's also a
264	bug in the Debian version and it serves as a reminder for other users that
265	might encounter the same issue.
266		527
267	=item I am maintaining rxvt-unicode for distribution/OS XXX, any	528	Either try C<Ctrl-2> alone (it often is mapped to ASCII NUL even on
268	recommendation?	529	international keyboards) or simply use ISO 14755 support to your
		530	advantage, typing <Ctrl-Shift-0> to get a ASCII NUL. This works for other
		531	codes, too, such as C<Ctrl-Shift-1-d> to type the default telnet escape
		532	character and so on.
269		533
270	You should build one binary with the default options. F<configure>	534	=head3 Mouse cut/paste suddenly no longer works.
271	now enables most useful options, and the trend goes to making them
272	runtime-switchable, too, so there is usually no drawback to enbaling them,
273	except higher disk and possibly memory usage. The perl interpreter should
274	be enabled, as important functionality (menus, selection, likely more in
275	the future) depends on it.
276		535
277	You should not overwrite the C<perl-ext-common> snd C<perl-ext> resources	536	Make sure that mouse reporting is actually turned off since killing
278	system-wide (except maybe with C<defaults>). This will result in useful	537	some editors prematurely may leave the mouse in mouse report mode. I've
279	behaviour. If your distribution aims at low memory, add an empty	538	heard that tcsh may use mouse reporting unless it otherwise specified. A
280	C<perl-ext-common> resource to the app-defaults file. This will keep the	539	quick check is to see if cut/paste works when the Alt or Shift keys are
281	perl interpreter disabled until the user enables it.	540	depressed.
282		541
283	If you can/want build more binaries, I recommend building a minimal	542	=head3 What's with the strange Backspace/Delete key behaviour?
284	one with C<--disable-everything> (very useful) and a maximal one with
285	C<--enable-everything> (less useful, it will be very big due to a lot of
286	encodings built-in that increase download times and are rarely used).
287		543
288	=item I need to make it setuid/setgid to support utmp/ptys on my OS, is this safe?	544	Assuming that the physical Backspace key corresponds to the
		545	BackSpace keysym (not likely for Linux ... see the following
		546	question) there are two standard values that can be used for
		547	Backspace: C<^H> and C<^?>.
289		548
290	It should be, starting with release 7.1. You are encouraged to properly	549	Historically, either value is correct, but rxvt-unicode adopts the debian
291	install urxvt with privileges necessary for your OS now.	550	policy of using C<^?> when unsure, because it's the one only only correct
		551	choice :).
292		552
293	When rxvt-unicode detects that it runs setuid or setgid, it will fork	553	Rxvt-unicode tries to inherit the current stty settings and uses the value
294	into a helper process for privileged operations (pty handling on some	554	of `erase' to guess the value for backspace. If rxvt-unicode wasn't
295	systems, utmp/wtmp/lastlog handling on others) and drop privileges	555	started from a terminal (say, from a menu or by remote shell), then the
296	immediately. This is much safer than most other terminals that keep	556	system value of `erase', which corresponds to CERASE in <termios.h>, will
297	privileges while running (but is more relevant to urxvt, as it contains	557	be used (which may not be the same as your stty setting).
298	things as perl interpreters, which might be "helpful" to attackers).
299		558
300	This forking is done as the very first within main(), which is very early	559	For starting a new rxvt-unicode:
301	and reduces possible bugs to initialisation code run before main(), or
302	things like the dynamic loader of your system, which should result in very
303	little risk.
304		560
		561	# use Backspace = ^H
		562	$ stty erase ^H
		563	$ @@RXVT_NAME@@
		564
		565	# use Backspace = ^?
		566	$ stty erase ^?
		567	$ @@RXVT_NAME@@
		568
		569	Toggle with C<ESC [ 36 h> / C<ESC [ 36 l>.
		570
		571	For an existing rxvt-unicode:
		572
		573	# use Backspace = ^H
		574	$ stty erase ^H
		575	$ echo -n "^[[36h"
		576
		577	# use Backspace = ^?
		578	$ stty erase ^?
		579	$ echo -n "^[[36l"
		580
		581	This helps satisfy some of the Backspace discrepancies that occur, but
		582	if you use Backspace = C<^H>, make sure that the termcap/terminfo value
		583	properly reflects that.
		584
		585	The Delete key is a another casualty of the ill-defined Backspace problem.
		586	To avoid confusion between the Backspace and Delete keys, the Delete
		587	key has been assigned an escape sequence to match the vt100 for Execute
		588	(C<ESC [ 3 ~>) and is in the supplied termcap/terminfo.
		589
		590	Some other Backspace problems:
		591
		592	some editors use termcap/terminfo,
		593	some editors (vim I'm told) expect Backspace = ^H,
		594	GNU Emacs (and Emacs-like editors) use ^H for help.
		595
		596	Perhaps someday this will all be resolved in a consistent manner.
		597
		598	=head3 I don't like the key-bindings. How do I change them?
		599
		600	There are some compile-time selections available via configure. Unless
		601	you have run "configure" with the C<--disable-resources> option you can
		602	use the `keysym' resource to alter the keystrings associated with keysyms.
		603
		604	Here's an example for a URxvt session started using C<@@RXVT_NAME@@ -name URxvt>
		605
		606	URxvt.keysym.Home: \033[1~
		607	URxvt.keysym.End: \033[4~
		608	URxvt.keysym.C-apostrophe: \033<C-'>
		609	URxvt.keysym.C-slash: \033<C-/>
		610	URxvt.keysym.C-semicolon: \033<C-;>
		611	URxvt.keysym.C-grave: \033<C-`>
		612	URxvt.keysym.C-comma: \033<C-,>
		613	URxvt.keysym.C-period: \033<C-.>
		614	URxvt.keysym.C-0x60: \033<C-`>
		615	URxvt.keysym.C-Tab: \033<C-Tab>
		616	URxvt.keysym.C-Return: \033<C-Return>
		617	URxvt.keysym.S-Return: \033<S-Return>
		618	URxvt.keysym.S-space: \033<S-Space>
		619	URxvt.keysym.M-Up: \033<M-Up>
		620	URxvt.keysym.M-Down: \033<M-Down>
		621	URxvt.keysym.M-Left: \033<M-Left>
		622	URxvt.keysym.M-Right: \033<M-Right>
		623	URxvt.keysym.M-C-0: list \033<M-C- 0123456789 >
		624	URxvt.keysym.M-C-a: list \033<M-C- abcdefghijklmnopqrstuvwxyz >
		625	URxvt.keysym.F12: command:\033]701;zh_CN.GBK\007
		626
		627	See some more examples in the documentation for the B<keysym> resource.
		628
		629	=head3 I'm using keyboard model XXX that has extra Prior/Next/Insert keys. How do I make use of them? For example, the Sun Keyboard type 4 has the following map
		630
		631	KP_Insert == Insert
		632	F22 == Print
		633	F27 == Home
		634	F29 == Prior
		635	F33 == End
		636	F35 == Next
		637
		638	Rather than have rxvt-unicode try to accommodate all the various possible
		639	keyboard mappings, it is better to use `xmodmap' to remap the keys as
		640	required for your particular machine.
		641
		642
		643
		644	=head2 Terminal Configuration
		645
		646	=head3 Why doesn't rxvt-unicode read my resources?
		647
		648	Well, why, indeed? It does, in a way very similar to other X
		649	applications. Most importantly, this means that if you or your OS loads
		650	resources into the X display (the right way to do it), rxvt-unicode will
		651	ignore any resource files in your home directory. It will only read
		652	F<$HOME/.Xdefaults> when no resources are attached to the display.
		653
		654	If you have or use an F<$HOME/.Xresources> file, chances are that
		655	resources are loaded into your X-server. In this case, you have to
		656	re-login after every change (or run F<xrdb -merge $HOME/.Xresources>).
		657
		658	Also consider the form resources have to use:
		659
		660	URxvt.resource: value
		661
		662	If you want to use another form (there are lots of different ways of
		663	specifying resources), make sure you understand wether and why it
		664	works. If unsure, use the form above.
		665
305	=item When I log-in to another system it tells me about missing terminfo data?	666	=head3 When I log-in to another system it tells me about missing terminfo data?
306		667
307	The terminal description used by rxvt-unicode is not as widely available	668	The terminal description used by rxvt-unicode is not as widely available
308	as that for xterm, or even rxvt (for which the same problem often arises).	669	as that for xterm, or even rxvt (for which the same problem often arises).
309		670
310	The correct solution for this problem is to install the terminfo, this can	671	The correct solution for this problem is to install the terminfo, this can
…		…
328	URxvt.termName: rxvt	689	URxvt.termName: rxvt
329		690
330	If you don't plan to use B<rxvt> (quite common...) you could also replace	691	If you don't plan to use B<rxvt> (quite common...) you could also replace
331	the rxvt terminfo file with the rxvt-unicode one.	692	the rxvt terminfo file with the rxvt-unicode one.
332		693
333	=item C<tic> outputs some error when compiling the terminfo entry.	694	=head3 C<tic> outputs some error when compiling the terminfo entry.
334		695
335	Most likely it's the empty definition for C<enacs=>. Just replace it by	696	Most likely it's the empty definition for C<enacs=>. Just replace it by
336	C<enacs=\E[0@> and try again.	697	C<enacs=\E[0@> and try again.
337		698
338	=item C<bash>'s readline does not work correctly under @@RXVT_NAME@@.	699	=head3 C<bash>'s readline does not work correctly under @@RXVT_NAME@@.
339		700
		701	See next entry.
		702
340	=item I need a termcap file entry.	703	=head3 I need a termcap file entry.
341		704
342	One reason you might want this is that some distributions or operating	705	One reason you might want this is that some distributions or operating
343	systems still compile some programs using the long-obsoleted termcap	706	systems still compile some programs using the long-obsoleted termcap
344	library (Fedora Core's bash is one example) and rely on a termcap entry	707	library (Fedora Core's bash is one example) and rely on a termcap entry
345	for C<rxvt-unicode>.	708	for C<rxvt-unicode>.
…		…
371	:sc=\E7:se=\E[27m:sf=^J:so=\E[7m:sr=\EM:st=\EH:ta=^I:\	734	:sc=\E7:se=\E[27m:sf=^J:so=\E[7m:sr=\EM:st=\EH:ta=^I:\
372	:te=\E[r\E[?1049l:ti=\E[?1049h:ue=\E[24m:up=\E[A:\	735	:te=\E[r\E[?1049l:ti=\E[?1049h:ue=\E[24m:up=\E[A:\
373	:us=\E[4m:vb=\E[?5h\E[?5l:ve=\E[?25h:vi=\E[?25l:\	736	:us=\E[4m:vb=\E[?5h\E[?5l:ve=\E[?25h:vi=\E[?25l:\
374	:vs=\E[?25h:	737	:vs=\E[?25h:
375		738
376	=item Why does C<ls> no longer have coloured output?	739	=head3 Why does C<ls> no longer have coloured output?
377		740
378	The C<ls> in the GNU coreutils unfortunately doesn't use terminfo to	741	The C<ls> in the GNU coreutils unfortunately doesn't use terminfo to
379	decide wether a terminal has colour, but uses it's own configuration	742	decide wether a terminal has colour, but uses it's own configuration
380	file. Needless to say, C<rxvt-unicode> is not in it's default file (among	743	file. Needless to say, C<rxvt-unicode> is not in it's default file (among
381	with most other terminals supporting colour). Either add:	744	with most other terminals supporting colour). Either add:
…		…
386		749
387	alias ls='ls --color=auto'	750	alias ls='ls --color=auto'
388		751
389	to your C<.profile> or C<.bashrc>.	752	to your C<.profile> or C<.bashrc>.
390		753
391	=item Why doesn't vim/emacs etc. use the 88 colour mode?	754	=head3 Why doesn't vim/emacs etc. use the 88 colour mode?
392		755
		756	See next entry.
		757
393	=item Why doesn't vim/emacs etc. make use of italic?	758	=head3 Why doesn't vim/emacs etc. make use of italic?
394		759
		760	See next entry.
		761
395	=item Why are the secondary screen-related options not working properly?	762	=head3 Why are the secondary screen-related options not working properly?
396		763
397	Make sure you are using C<TERM=rxvt-unicode>. Some pre-packaged	764	Make sure you are using C<TERM=rxvt-unicode>. Some pre-packaged
398	distributions (most notably Debian GNU/Linux) break rxvt-unicode	765	distributions (most notably Debian GNU/Linux) break rxvt-unicode
399	by setting C<TERM> to C<rxvt>, which doesn't have these extra	766	by setting C<TERM> to C<rxvt>, which doesn't have these extra
400	features. Unfortunately, some of these (most notably, again, Debian	767	features. Unfortunately, some of these (most notably, again, Debian
401	GNU/Linux) furthermore fail to even install the C<rxvt-unicode> terminfo	768	GNU/Linux) furthermore fail to even install the C<rxvt-unicode> terminfo
402	file, so you will need to install it on your own (See the question B<When	769	file, so you will need to install it on your own (See the question B<When
403	I log-in to another system it tells me about missing terminfo data?> on	770	I log-in to another system it tells me about missing terminfo data?> on
404	how to do this).	771	how to do this).
405		772
406	=item My numerical keypad acts weird and generates differing output?
407		773
408	Some Debian GNUL/Linux users seem to have this problem, although no	774	=head2 Encoding / Locale / Input Method Issues
409	specific details were reported so far. It is possible that this is caused
410	by the wrong C<TERM> setting, although the details of wether and how
411	this can happen are unknown, as C<TERM=rxvt> should offer a compatible
412	keymap. See the answer to the previous question, and please report if that
413	helped.
414		775
415	=item Rxvt-unicode does not seem to understand the selected encoding?	776	=head3 Rxvt-unicode does not seem to understand the selected encoding?
416		777
		778	See next entry.
		779
417	=item Unicode does not seem to work?	780	=head3 Unicode does not seem to work?
418		781
419	If you encounter strange problems like typing an accented character but	782	If you encounter strange problems like typing an accented character but
420	getting two unrelated other characters or similar, or if program output is	783	getting two unrelated other characters or similar, or if program output is
421	subtly garbled, then you should check your locale settings.	784	subtly garbled, then you should check your locale settings.
422		785
…		…
442		805
443	If nothing works and you are sure that everything is set correctly then	806	If nothing works and you are sure that everything is set correctly then
444	you will need to remember a little known fact: Some programs just don't	807	you will need to remember a little known fact: Some programs just don't
445	support locales :(	808	support locales :(
446		809
447	=item Why do some characters look so much different than others?	810	=head3 How does rxvt-unicode determine the encoding to use?
448		811
449	=item How does rxvt-unicode choose fonts?	812	See next entry.
450		813
451	Most fonts do not contain the full range of Unicode, which is	814	=head3 Is there an option to switch encodings?
452	fine. Chances are that the font you (or the admin/package maintainer of
453	your system/os) have specified does not cover all the characters you want
454	to display.
455		815
456	B<rxvt-unicode> makes a best-effort try at finding a replacement	816	Unlike some other terminals, rxvt-unicode has no encoding switch, and no
457	font. Often the result is fine, but sometimes the chosen font looks	817	specific "utf-8" mode, such as xterm. In fact, it doesn't even know about
458	bad/ugly/wrong. Some fonts have totally strange characters that don't	818	UTF-8 or any other encodings with respect to terminal I/O.
459	resemble the correct glyph at all, and rxvt-unicode lacks the artificial
460	intelligence to detect that a specific glyph is wrong: it has to believe
461	the font that the characters it claims to contain indeed look correct.
462		819
463	In that case, select a font of your taste and add it to the font list,	820	The reasons is that there exists a perfectly fine mechanism for selecting
464	e.g.:	821	the encoding, doing I/O and (most important) communicating this to all
465		822	applications so everybody agrees on character properties such as width
466	@@RXVT_NAME@@ -fn basefont,font2,font3...	823	and code number. This mechanism is the I<locale>. Applications not using
467		824	that info will have problems (for example, C<xterm> gets the width of
468	When rxvt-unicode sees a character, it will first look at the base	825	characters wrong as it uses it's own, locale-independent table under all
469	font. If the base font does not contain the character, it will go to the
470	next font, and so on. Specifying your own fonts will also speed up this
471	search and use less resources within rxvt-unicode and the X-server.
472
473	The only limitation is that none of the fonts may be larger than the base
474	font, as the base font defines the terminal character cell size, which
475	must be the same due to the way terminals work.
476
477	=item Why do some chinese characters look so different than others?
478
479	This is because there is a difference between script and language --
480	rxvt-unicode does not know which language the text that is output is,
481	as it only knows the unicode character codes. If rxvt-unicode first
482	sees a japanese/chinese character, it might choose a japanese font for
483	display. Subsequent japanese characters will use that font. Now, many
484	chinese characters aren't represented in japanese fonts, so when the first
485	non-japanese character comes up, rxvt-unicode will look for a chinese font
486	-- unfortunately at this point, it will still use the japanese font for
487	chinese characters that are also in the japanese font.
488
489	The workaround is easy: just tag a chinese font at the end of your font
490	list (see the previous question). The key is to view the font list as
491	a preference list: If you expect more japanese, list a japanese font
492	first. If you expect more chinese, put a chinese font first.
493
494	In the future it might be possible to switch language preferences at
495	runtime (the internal data structure has no problem with using different
496	fonts for the same character at the same time, but no interface for this
497	has been designed yet).
498
499	Until then, you might get away with switching fonts at runtime (see L<Can
500	I switch the fonts at runtime?> later in this document).
501
502	=item Why does rxvt-unicode sometimes leave pixel droppings?
503
504	Most fonts were not designed for terminal use, which means that character
505	size varies a lot. A font that is otherwise fine for terminal use might
506	contain some characters that are simply too wide. Rxvt-unicode will avoid
507	these characters. For characters that are just "a bit" too wide a special
508	"careful" rendering mode is used that redraws adjacent characters.
509
510	All of this requires that fonts do not lie about character sizes,
511	however: Xft fonts often draw glyphs larger than their acclaimed bounding
512	box, and rxvt-unicode has no way of detecting this (the correct way is to
513	ask for the character bounding box, which unfortunately is wrong in these
514	cases).	826	locales).
515		827
516	It's not clear (to me at least), wether this is a bug in Xft, freetype,	828	Rxvt-unicode uses the C<LC_CTYPE> locale category to select encoding. All
517	or the respective font. If you encounter this problem you might try using	829	programs doing the same (that is, most) will automatically agree in the
518	the C<-lsp> option to give the font more height. If that doesn't work, you	830	interpretation of characters.
519	might be forced to use a different font.
520		831
521	All of this is not a problem when using X11 core fonts, as their bounding	832	Unfortunately, there is no system-independent way to select locales, nor
522	box data is correct.	833	is there a standard on how locale specifiers will look like.
523		834
		835	On most systems, the content of the C<LC_CTYPE> environment variable
		836	contains an arbitrary string which corresponds to an already-installed
		837	locale. Common names for locales are C<en_US.UTF-8>, C<de_DE.ISO-8859-15>,
		838	C<ja_JP.EUC-JP>, i.e. C<language_country.encoding>, but other forms
		839	(i.e. C<de> or C<german>) are also common.
		840
		841	Rxvt-unicode ignores all other locale categories, and except for
		842	the encoding, ignores country or language-specific settings,
		843	i.e. C<de_DE.UTF-8> and C<ja_JP.UTF-8> are the normally same to
		844	rxvt-unicode.
		845
		846	If you want to use a specific encoding you have to make sure you start
		847	rxvt-unicode with the correct C<LC_CTYPE> category.
		848
		849	=head3 Can I switch locales at runtime?
		850
		851	Yes, using an escape sequence. Try something like this, which sets
		852	rxvt-unicode's idea of C<LC_CTYPE>.
		853
		854	printf '\e]701;%s\007' ja_JP.SJIS
		855
		856	See also the previous answer.
		857
		858	Sometimes this capability is rather handy when you want to work in
		859	one locale (e.g. C<de_DE.UTF-8>) but some programs don't support it
		860	(e.g. UTF-8). For example, I use this script to start C<xjdic>, which
		861	first switches to a locale supported by xjdic and back later:
		862
		863	printf '\e]701;%s\007' ja_JP.SJIS
		864	xjdic -js
		865	printf '\e]701;%s\007' de_DE.UTF-8
		866
		867	You can also use xterm's C<luit> program, which usually works fine, except
		868	for some locales where character width differs between program- and
		869	rxvt-unicode-locales.
		870
		871	=head3 My input method wants <some encoding> but I want UTF-8, what can I do?
		872
		873	You can specify separate locales for the input method and the rest of the
		874	terminal, using the resource C<imlocale>:
		875
		876	URxvt.imlocale: ja_JP.EUC-JP
		877
		878	Now you can start your terminal with C<LC_CTYPE=ja_JP.UTF-8> and still
		879	use your input method. Please note, however, that you will not be able to
		880	input characters outside C<EUC-JP> in a normal way then, as your input
		881	method limits you.
		882
		883	=head3 Rxvt-unicode crashes when the X Input Method changes or exits.
		884
		885	Unfortunately, this is unavoidable, as the XIM protocol is racy by
		886	design. Applications can avoid some crashes at the expense of memory
		887	leaks, and Input Methods can avoid some crashes by careful ordering at
		888	exit time. B<kinput2> (and derived input methods) generally succeeds,
		889	while B<SCIM> (or similar input methods) fails. In the end, however,
		890	crashes cannot be completely avoided even if both sides cooperate.
		891
		892	So the only workaround is not to kill your Input Method Servers.
		893
		894
		895	=head2 Operating Systems / Package Maintaining
		896
		897	=head3 I am using Debian GNU/Linux and have a problem...
		898
		899	The Debian GNU/Linux package of rxvt-unicode in sarge contains large
		900	patches that considerably change the behaviour of rxvt-unicode (but
		901	unfortunately this notice has been removed). Before reporting a bug to
		902	the original rxvt-unicode author please download and install the genuine
		903	version (L<http://software.schmorp.de#rxvt-unicode>) and try to reproduce
		904	the problem. If you cannot, chances are that the problems are specific to
		905	Debian GNU/Linux, in which case it should be reported via the Debian Bug
		906	Tracking System (use C<reportbug> to report the bug).
		907
		908	For other problems that also affect the Debian package, you can and
		909	probably should use the Debian BTS, too, because, after all, it's also a
		910	bug in the Debian version and it serves as a reminder for other users that
		911	might encounter the same issue.
		912
		913	=head3 I am maintaining rxvt-unicode for distribution/OS XXX, any recommendation?
		914
		915	You should build one binary with the default options. F<configure>
		916	now enables most useful options, and the trend goes to making them
		917	runtime-switchable, too, so there is usually no drawback to enbaling them,
		918	except higher disk and possibly memory usage. The perl interpreter should
		919	be enabled, as important functionality (menus, selection, likely more in
		920	the future) depends on it.
		921
		922	You should not overwrite the C<perl-ext-common> snd C<perl-ext> resources
		923	system-wide (except maybe with C<defaults>). This will result in useful
		924	behaviour. If your distribution aims at low memory, add an empty
		925	C<perl-ext-common> resource to the app-defaults file. This will keep the
		926	perl interpreter disabled until the user enables it.
		927
		928	If you can/want build more binaries, I recommend building a minimal
		929	one with C<--disable-everything> (very useful) and a maximal one with
		930	C<--enable-everything> (less useful, it will be very big due to a lot of
		931	encodings built-in that increase download times and are rarely used).
		932
		933	=head3 I need to make it setuid/setgid to support utmp/ptys on my OS, is this safe?
		934
		935	It should be, starting with release 7.1. You are encouraged to properly
		936	install urxvt with privileges necessary for your OS now.
		937
		938	When rxvt-unicode detects that it runs setuid or setgid, it will fork
		939	into a helper process for privileged operations (pty handling on some
		940	systems, utmp/wtmp/lastlog handling on others) and drop privileges
		941	immediately. This is much safer than most other terminals that keep
		942	privileges while running (but is more relevant to urxvt, as it contains
		943	things as perl interpreters, which might be "helpful" to attackers).
		944
		945	This forking is done as the very first within main(), which is very early
		946	and reduces possible bugs to initialisation code run before main(), or
		947	things like the dynamic loader of your system, which should result in very
		948	little risk.
		949
524	=item On Solaris 9, many line-drawing characters are too wide.	950	=head3 On Solaris 9, many line-drawing characters are too wide.
525		951
526	Seems to be a known bug, read	952	Seems to be a known bug, read
527	L<http://nixdoc.net/files/forum/about34198.html>. Some people use the	953	L<http://nixdoc.net/files/forum/about34198.html>. Some people use the
528	following ugly workaround to get non-double-wide-characters working:	954	following ugly workaround to get non-double-wide-characters working:
529		955
530	#define wcwidth(x) wcwidth(x) > 1 ? 1 : wcwidth(x)	956	#define wcwidth(x) wcwidth(x) > 1 ? 1 : wcwidth(x)
531		957
532	=item My Compose (Multi_key) key is no longer working.
533
534	The most common causes for this are that either your locale is not set
535	correctly, or you specified a B<preeditStyle> that is not supported by
536	your input method. For example, if you specified B<OverTheSpot> and
537	your input method (e.g. the default input method handling Compose keys)
538	does not support this (for instance because it is not visual), then
539	rxvt-unicode will continue without an input method.
540
541	In this case either do not specify a B<preeditStyle> or specify more than
542	one pre-edit style, such as B<OverTheSpot,Root,None>.
543
544	=item I cannot type C<Ctrl-Shift-2> to get an ASCII NUL character due to ISO 14755
545
546	Either try C<Ctrl-2> alone (it often is mapped to ASCII NUL even on
547	international keyboards) or simply use ISO 14755 support to your
548	advantage, typing <Ctrl-Shift-0> to get a ASCII NUL. This works for other
549	codes, too, such as C<Ctrl-Shift-1-d> to type the default telnet escape
550	character and so on.
551
552	=item How can I keep rxvt-unicode from using reverse video so much?
553
554	First of all, make sure you are running with the right terminal settings
555	(C<TERM=rxvt-unicode>), which will get rid of most of these effects. Then
556	make sure you have specified colours for italic and bold, as otherwise
557	rxvt-unicode might use reverse video to simulate the effect:
558
559	URxvt.colorBD: white
560	URxvt.colorIT: green
561
562	=item Some programs assume totally weird colours (red instead of blue), how can I fix that?
563
564	For some unexplainable reason, some rare programs assume a very weird
565	colour palette when confronted with a terminal with more than the standard
566	8 colours (rxvt-unicode supports 88). The right fix is, of course, to fix
567	these programs not to assume non-ISO colours without very good reasons.
568
569	In the meantime, you can either edit your C<rxvt-unicode> terminfo
570	definition to only claim 8 colour support or use C<TERM=rxvt>, which will
571	fix colours but keep you from using other rxvt-unicode features.
572
573	=item I am on FreeBSD and rxvt-unicode does not seem to work at all.	958	=head3 I am on FreeBSD and rxvt-unicode does not seem to work at all.
574		959
575	Rxvt-unicode requires the symbol C<__STDC_ISO_10646__> to be defined	960	Rxvt-unicode requires the symbol C<__STDC_ISO_10646__> to be defined
576	in your compile environment, or an implementation that implements it,	961	in your compile environment, or an implementation that implements it,
577	wether it defines the symbol or not. C<__STDC_ISO_10646__> requires that	962	wether it defines the symbol or not. C<__STDC_ISO_10646__> requires that
578	B<wchar_t> is represented as unicode.	963	B<wchar_t> is represented as unicode.
…		…
600		985
601	The rxvt-unicode author insists that the right way to fix this is in the	986	The rxvt-unicode author insists that the right way to fix this is in the
602	system libraries once and for all, instead of forcing every app to carry	987	system libraries once and for all, instead of forcing every app to carry
603	complete replacements for them :)	988	complete replacements for them :)
604		989
605	=item I use Solaris 9 and it doesn't compile/work/etc.	990	=head3 I use Solaris 9 and it doesn't compile/work/etc.
606		991
607	Try the diff in F<doc/solaris9.patch> as a base. It fixes the worst	992	Try the diff in F<doc/solaris9.patch> as a base. It fixes the worst
608	problems with C<wcwidth> and a compile problem.	993	problems with C<wcwidth> and a compile problem.
609		994
610	=item How can I use rxvt-unicode under cygwin?	995	=head3 How can I use rxvt-unicode under cygwin?
611		996
612	rxvt-unicode should compile and run out of the box on cygwin, using	997	rxvt-unicode should compile and run out of the box on cygwin, using
613	the X11 libraries that come with cygwin. libW11 emulation is no	998	the X11 libraries that come with cygwin. libW11 emulation is no
614	longer supported (and makes no sense, either, as it only supported a	999	longer supported (and makes no sense, either, as it only supported a
615	single font). I recommend starting the X-server in C<-multiwindow> or	1000	single font). I recommend starting the X-server in C<-multiwindow> or
…		…
617	old libW11 emulation.	1002	old libW11 emulation.
618		1003
619	At the time of this writing, cygwin didn't seem to support any multi-byte	1004	At the time of this writing, cygwin didn't seem to support any multi-byte
620	encodings (you might try C<LC_CTYPE=C-UTF-8>), so you are likely limited	1005	encodings (you might try C<LC_CTYPE=C-UTF-8>), so you are likely limited
621	to 8-bit encodings.	1006	to 8-bit encodings.
622
623	=item How does rxvt-unicode determine the encoding to use?
624
625	=item Is there an option to switch encodings?
626
627	Unlike some other terminals, rxvt-unicode has no encoding switch, and no
628	specific "utf-8" mode, such as xterm. In fact, it doesn't even know about
629	UTF-8 or any other encodings with respect to terminal I/O.
630
631	The reasons is that there exists a perfectly fine mechanism for selecting
632	the encoding, doing I/O and (most important) communicating this to all
633	applications so everybody agrees on character properties such as width
634	and code number. This mechanism is the I<locale>. Applications not using
635	that info will have problems (for example, C<xterm> gets the width of
636	characters wrong as it uses it's own, locale-independent table under all
637	locales).
638
639	Rxvt-unicode uses the C<LC_CTYPE> locale category to select encoding. All
640	programs doing the same (that is, most) will automatically agree in the
641	interpretation of characters.
642
643	Unfortunately, there is no system-independent way to select locales, nor
644	is there a standard on how locale specifiers will look like.
645
646	On most systems, the content of the C<LC_CTYPE> environment variable
647	contains an arbitrary string which corresponds to an already-installed
648	locale. Common names for locales are C<en_US.UTF-8>, C<de_DE.ISO-8859-15>,
649	C<ja_JP.EUC-JP>, i.e. C<language_country.encoding>, but other forms
650	(i.e. C<de> or C<german>) are also common.
651
652	Rxvt-unicode ignores all other locale categories, and except for
653	the encoding, ignores country or language-specific settings,
654	i.e. C<de_DE.UTF-8> and C<ja_JP.UTF-8> are the normally same to
655	rxvt-unicode.
656
657	If you want to use a specific encoding you have to make sure you start
658	rxvt-unicode with the correct C<LC_CTYPE> category.
659
660	=item Can I switch locales at runtime?
661
662	Yes, using an escape sequence. Try something like this, which sets
663	rxvt-unicode's idea of C<LC_CTYPE>.
664
665	printf '\e]701;%s\007' ja_JP.SJIS
666
667	See also the previous answer.
668
669	Sometimes this capability is rather handy when you want to work in
670	one locale (e.g. C<de_DE.UTF-8>) but some programs don't support it
671	(e.g. UTF-8). For example, I use this script to start C<xjdic>, which
672	first switches to a locale supported by xjdic and back later:
673
674	printf '\e]701;%s\007' ja_JP.SJIS
675	xjdic -js
676	printf '\e]701;%s\007' de_DE.UTF-8
677
678	You can also use xterm's C<luit> program, which usually works fine, except
679	for some locales where character width differs between program- and
680	rxvt-unicode-locales.
681
682	=item Can I switch the fonts at runtime?
683
684	Yes, using an escape sequence. Try something like this, which has the same
685	effect as using the C<-fn> switch, and takes effect immediately:
686
687	printf '\e]50;%s\007' "9x15bold,xft:Kochi Gothic"
688
689	This is useful if you e.g. work primarily with japanese (and prefer a
690	japanese font), but you have to switch to chinese temporarily, where
691	japanese fonts would only be in your way.
692
693	You can think of this as a kind of manual ISO-2022 switching.
694
695	=item Why do italic characters look as if clipped?
696
697	Many fonts have difficulties with italic characters and hinting. For
698	example, the otherwise very nicely hinted font C<xft:Bitstream Vera Sans
699	Mono> completely fails in it's italic face. A workaround might be to
700	enable freetype autohinting, i.e. like this:
701
702	URxvt.italicFont: xft:Bitstream Vera Sans Mono:italic:autohint=true
703	URxvt.boldItalicFont: xft:Bitstream Vera Sans Mono:bold:italic:autohint=true
704
705	=item My input method wants <some encoding> but I want UTF-8, what can I do?
706
707	You can specify separate locales for the input method and the rest of the
708	terminal, using the resource C<imlocale>:
709
710	URxvt.imlocale: ja_JP.EUC-JP
711
712	Now you can start your terminal with C<LC_CTYPE=ja_JP.UTF-8> and still
713	use your input method. Please note, however, that you will not be able to
714	input characters outside C<EUC-JP> in a normal way then, as your input
715	method limits you.
716
717	=item Rxvt-unicode crashes when the X Input Method changes or exits.
718
719	Unfortunately, this is unavoidable, as the XIM protocol is racy by
720	design. Applications can avoid some crashes at the expense of memory
721	leaks, and Input Methods can avoid some crashes by careful ordering at
722	exit time. B<kinput2> (and derived input methods) generally succeeds,
723	while B<SCIM> (or similar input methods) fails. In the end, however,
724	crashes cannot be completely avoided even if both sides cooperate.
725
726	So the only workaround is not to kill your Input Method Servers.
727
728	=item Rxvt-unicode uses gobs of memory, how can I reduce that?
729
730	Rxvt-unicode tries to obey the rule of not charging you for something you
731	don't use. One thing you should try is to configure out all settings that
732	you don't need, for example, Xft support is a resource hog by design,
733	when used. Compiling it out ensures that no Xft font will be loaded
734	accidentally when rxvt-unicode tries to find a font for your characters.
735
736	Also, many people (me included) like large windows and even larger
737	scrollback buffers: Without C<--enable-unicode3>, rxvt-unicode will use
738	6 bytes per screen cell. For a 160x?? window this amounts to almost a
739	kilobyte per line. A scrollback buffer of 10000 lines will then (if full)
740	use 10 Megabytes of memory. With C<--enable-unicode3> it gets worse, as
741	rxvt-unicode then uses 8 bytes per screen cell.
742
743	=item Can I speed up Xft rendering somehow?
744
745	Yes, the most obvious way to speed it up is to avoid Xft entirely, as
746	it is simply slow. If you still want Xft fonts you might try to disable
747	antialiasing (by appending C<:antialias=false>), which saves lots of
748	memory and also speeds up rendering considerably.
749
750	=item Rxvt-unicode doesn't seem to anti-alias its fonts, what is wrong?
751
752	Rxvt-unicode will use whatever you specify as a font. If it needs to
753	fall back to it's default font search list it will prefer X11 core
754	fonts, because they are small and fast, and then use Xft fonts. It has
755	antialiasing disabled for most of them, because the author thinks they
756	look best that way.
757
758	If you want antialiasing, you have to specify the fonts manually.
759
760	=item Mouse cut/paste suddenly no longer works.
761
762	Make sure that mouse reporting is actually turned off since killing
763	some editors prematurely may leave the mouse in mouse report mode. I've
764	heard that tcsh may use mouse reporting unless it otherwise specified. A
765	quick check is to see if cut/paste works when the Alt or Shift keys are
766	depressed.
767
768	=item What's with this bold/blink stuff?
769
770	If no bold colour is set via C<colorBD:>, bold will invert text using the
771	standard foreground colour.
772
773	For the standard background colour, blinking will actually make the
774	text blink when compiled with C<--enable-blinking>. with standard
775	colours. Without C<--enable-blinking>, the blink attribute will be
776	ignored.
777
778	On ANSI colours, bold/blink attributes are used to set high-intensity
779	foreground/background colors.
780
781	color0-7 are the low-intensity colors.
782
783	color8-15 are the corresponding high-intensity colors.
784
785	=item I don't like the screen colors. How do I change them?
786
787	You can change the screen colors at run-time using F<~/.Xdefaults>
788	resources (or as long-options).
789
790	Here are values that are supposed to resemble a VGA screen,
791	including the murky brown that passes for low-intensity yellow:
792
793	URxvt.color0: #000000
794	URxvt.color1: #A80000
795	URxvt.color2: #00A800
796	URxvt.color3: #A8A800
797	URxvt.color4: #0000A8
798	URxvt.color5: #A800A8
799	URxvt.color6: #00A8A8
800	URxvt.color7: #A8A8A8
801
802	URxvt.color8: #000054
803	URxvt.color9: #FF0054
804	URxvt.color10: #00FF54
805	URxvt.color11: #FFFF54
806	URxvt.color12: #0000FF
807	URxvt.color13: #FF00FF
808	URxvt.color14: #00FFFF
809	URxvt.color15: #FFFFFF
810
811	And here is a more complete set of non-standard colors described (not by
812	me) as "pretty girly".
813
814	URxvt.cursorColor: #dc74d1
815	URxvt.pointerColor: #dc74d1
816	URxvt.background: #0e0e0e
817	URxvt.foreground: #4ad5e1
818	URxvt.color0: #000000
819	URxvt.color8: #8b8f93
820	URxvt.color1: #dc74d1
821	URxvt.color9: #dc74d1
822	URxvt.color2: #0eb8c7
823	URxvt.color10: #0eb8c7
824	URxvt.color3: #dfe37e
825	URxvt.color11: #dfe37e
826	URxvt.color5: #9e88f0
827	URxvt.color13: #9e88f0
828	URxvt.color6: #73f7ff
829	URxvt.color14: #73f7ff
830	URxvt.color7: #e1dddd
831	URxvt.color15: #e1dddd
832
833	=item How can I start @@RXVT_NAME@@d in a race-free way?
834
835	Try C<@@RXVT_NAME@@d -f -o>, which tells @@RXVT_NAME@@d to open the
836	display, create the listening socket and then fork.
837
838	=item What's with the strange Backspace/Delete key behaviour?
839
840	Assuming that the physical Backspace key corresponds to the
841	BackSpace keysym (not likely for Linux ... see the following
842	question) there are two standard values that can be used for
843	Backspace: C<^H> and C<^?>.
844
845	Historically, either value is correct, but rxvt-unicode adopts the debian
846	policy of using C<^?> when unsure, because it's the one only only correct
847	choice :).
848
849	Rxvt-unicode tries to inherit the current stty settings and uses the value
850	of `erase' to guess the value for backspace. If rxvt-unicode wasn't
851	started from a terminal (say, from a menu or by remote shell), then the
852	system value of `erase', which corresponds to CERASE in <termios.h>, will
853	be used (which may not be the same as your stty setting).
854
855	For starting a new rxvt-unicode:
856
857	# use Backspace = ^H
858	$ stty erase ^H
859	$ @@RXVT_NAME@@
860
861	# use Backspace = ^?
862	$ stty erase ^?
863	$ @@RXVT_NAME@@
864
865	Toggle with C<ESC [ 36 h> / C<ESC [ 36 l>.
866
867	For an existing rxvt-unicode:
868
869	# use Backspace = ^H
870	$ stty erase ^H
871	$ echo -n "^[[36h"
872
873	# use Backspace = ^?
874	$ stty erase ^?
875	$ echo -n "^[[36l"
876
877	This helps satisfy some of the Backspace discrepancies that occur, but
878	if you use Backspace = C<^H>, make sure that the termcap/terminfo value
879	properly reflects that.
880
881	The Delete key is a another casualty of the ill-defined Backspace problem.
882	To avoid confusion between the Backspace and Delete keys, the Delete
883	key has been assigned an escape sequence to match the vt100 for Execute
884	(C<ESC [ 3 ~>) and is in the supplied termcap/terminfo.
885
886	Some other Backspace problems:
887
888	some editors use termcap/terminfo,
889	some editors (vim I'm told) expect Backspace = ^H,
890	GNU Emacs (and Emacs-like editors) use ^H for help.
891
892	Perhaps someday this will all be resolved in a consistent manner.
893
894	=item I don't like the key-bindings. How do I change them?
895
896	There are some compile-time selections available via configure. Unless
897	you have run "configure" with the C<--disable-resources> option you can
898	use the `keysym' resource to alter the keystrings associated with keysyms.
899
900	Here's an example for a URxvt session started using C<@@RXVT_NAME@@ -name URxvt>
901
902	URxvt.keysym.Home: \033[1~
903	URxvt.keysym.End: \033[4~
904	URxvt.keysym.C-apostrophe: \033<C-'>
905	URxvt.keysym.C-slash: \033<C-/>
906	URxvt.keysym.C-semicolon: \033<C-;>
907	URxvt.keysym.C-grave: \033<C-`>
908	URxvt.keysym.C-comma: \033<C-,>
909	URxvt.keysym.C-period: \033<C-.>
910	URxvt.keysym.C-0x60: \033<C-`>
911	URxvt.keysym.C-Tab: \033<C-Tab>
912	URxvt.keysym.C-Return: \033<C-Return>
913	URxvt.keysym.S-Return: \033<S-Return>
914	URxvt.keysym.S-space: \033<S-Space>
915	URxvt.keysym.M-Up: \033<M-Up>
916	URxvt.keysym.M-Down: \033<M-Down>
917	URxvt.keysym.M-Left: \033<M-Left>
918	URxvt.keysym.M-Right: \033<M-Right>
919	URxvt.keysym.M-C-0: list \033<M-C- 0123456789 >
920	URxvt.keysym.M-C-a: list \033<M-C- abcdefghijklmnopqrstuvwxyz >
921	URxvt.keysym.F12: command:\033]701;zh_CN.GBK\007
922
923	See some more examples in the documentation for the B<keysym> resource.
924
925	=item I'm using keyboard model XXX that has extra Prior/Next/Insert keys.
926	How do I make use of them? For example, the Sun Keyboard type 4
927	has the following mappings that rxvt-unicode doesn't recognize.
928
929	KP_Insert == Insert
930	F22 == Print
931	F27 == Home
932	F29 == Prior
933	F33 == End
934	F35 == Next
935
936	Rather than have rxvt-unicode try to accommodate all the various possible
937	keyboard mappings, it is better to use `xmodmap' to remap the keys as
938	required for your particular machine.
939
940	=item How do I distinguish wether I'm running rxvt-unicode or a regular xterm?
941	I need this to decide about setting colors etc.
942
943	rxvt and rxvt-unicode always export the variable "COLORTERM", so you can
944	check and see if that is set. Note that several programs, JED, slrn,
945	Midnight Commander automatically check this variable to decide whether or
946	not to use color.
947
948	=item How do I set the correct, full IP address for the DISPLAY variable?
949
950	If you've compiled rxvt-unicode with DISPLAY_IS_IP and have enabled
951	insecure mode then it is possible to use the following shell script
952	snippets to correctly set the display. If your version of rxvt-unicode
953	wasn't also compiled with ESCZ_ANSWER (as assumed in these snippets) then
954	the COLORTERM variable can be used to distinguish rxvt-unicode from a
955	regular xterm.
956
957	Courtesy of Chuck Blake <cblake@BBN.COM> with the following shell script
958	snippets:
959
960	# Bourne/Korn/POSIX family of shells:
961	[ ${TERM:-foo} = foo ] && TERM=xterm # assume an xterm if we don't know
962	if [ ${TERM:-foo} = xterm ]; then
963	stty -icanon -echo min 0 time 15 # see if enhanced rxvt or not
964	echo -n '^[Z'
965	read term_id
966	stty icanon echo
967	if [ ""${term_id} = '^[[?1;2C' -a ${DISPLAY:-foo} = foo ]; then
968	echo -n '^[[7n' # query the rxvt we are in for the DISPLAY string
969	read DISPLAY # set it in our local shell
970	fi
971	fi
972
973	=item How do I compile the manual pages for myself?
974
975	You need to have a recent version of perl installed as F</usr/bin/perl>,
976	one that comes with F<pod2man>, F<pod2text> and F<pod2html>. Then go to
977	the doc subdirectory and enter C<make alldoc>.
978
979	=item My question isn't answered here, can I ask a human?
980
981	Before sending me mail, you could go to IRC: C<irc.freenode.net>,
982	channel C<#rxvt-unicode> has some rxvt-unicode enthusiasts that might be
983	interested in learning about new and exciting problems (but not FAQs :).
984
985	=back
986		1007
987	=head1 RXVT TECHNICAL REFERENCE	1008	=head1 RXVT TECHNICAL REFERENCE
988		1009
989	=head1 DESCRIPTION	1010	=head1 DESCRIPTION
990		1011

Diff Legend

–	Removed lines
+	Added lines
<	Changed lines
>	Changed lines