ViewVC Help

View File | Revision Log | Show Annotations | Download File

/cvs/rxvt-unicode/doc/rxvt.7.man.in

Comparing rxvt-unicode/doc/rxvt.7.man.in (file contents):
Revision 1.57 by root, Tue Jan 31 00:53:49 2006 UTC vs.
Revision 1.65 by root, Tue Jan 31 21:00:26 2006 UTC

…		…
153	This document contains the \s-1FAQ\s0, the \s-1RXVT\s0 \s-1TECHNICAL\s0 \s-1REFERENCE\s0 documenting	153	This document contains the \s-1FAQ\s0, the \s-1RXVT\s0 \s-1TECHNICAL\s0 \s-1REFERENCE\s0 documenting
154	all escape sequences, and other background information.	154	all escape sequences, and other background information.
155	.PP	155	.PP
156	The newest version of this document is also available on the World Wide Web at	156	The newest version of this document is also available on the World Wide Web at
157	<http://cvs.schmorp.de/browse/*checkout*/rxvt\-unicode/doc/rxvt.7.html>.	157	<http://cvs.schmorp.de/browse/*checkout*/rxvt\-unicode/doc/rxvt.7.html>.
158	.SH "FREQUENTLY ASKED QUESTIONS"	158	.SH "RXVT\-UNICODE/URXVT FREQUENTLY ASKED QUESTIONS"
159	.IX Header "FREQUENTLY ASKED QUESTIONS"	159	.IX Header "RXVT-UNICODE/URXVT FREQUENTLY ASKED QUESTIONS"
160	.IP "The new selection selects pieces that are too big, how can I select single words?" 4	160	.Sh "Meta, Features & Commandline Issues"
161	.IX Item "The new selection selects pieces that are too big, how can I select single words?"	161	.IX Subsection "Meta, Features & Commandline Issues"
162	Yes. For example, if you want to select alphanumeric words, you can use	162	\fIMy question isn't answered here, can I ask a human?\fR
163	the following resource:	163	.IX Subsection "My question isn't answered here, can I ask a human?"
164	.Sp	164	.PP
		165	Before sending me mail, you could go to \s-1IRC:\s0 \f(CW\*(C`irc.freenode.net\*(C'\fR,
		166	channel \f(CW\*(C`#rxvt\-unicode\*(C'\fR has some rxvt-unicode enthusiasts that might be
		167	interested in learning about new and exciting problems (but not FAQs :).
		168	.PP
		169	\fIDoes it support tabs, can I have a tabbed rxvt\-unicode?\fR
		170	.IX Subsection "Does it support tabs, can I have a tabbed rxvt-unicode?"
		171	.PP
		172	Beginning with version 7.3, there is a perl extension that implements a
		173	simple tabbed terminal. It is installed by default, so any of these should
		174	give you tabs:
		175	.PP
165	.Vb 1	176	.Vb 1
166	\& URxvt.selection.pattern-0: ([[:word:]]+)	177	\& @@URXVT_NAME@@ -pe tabbed
167	.Ve	178	.Ve
168	.Sp	179	.PP
169	If you click more than twice, the selection will be extended
170	more and more.
171	.Sp
172	To get a selection that is very similar to the old code, try this pattern:
173	.Sp
174	.Vb 1	180	.Vb 1
175	\& URxvt.selection.pattern-0: ([^"&'()*,;<=>?@[\e\e\e\e]^`{|})]+)	181	\& URxvt.perl-ext-common: default,tabbed
176	.Ve	182	.Ve
177	.Sp	183	.PP
178	Please also note that the \fILeftClick Shift-LeftClik\fR combination also	184	It will also work fine with tabbing functionality of many window managers
179	selects words like the old code.	185	or similar tabbing programs, and its embedding-features allow it to be
180	.IP "I don't like the new selection/popups/hotkeys/perl, how do I change/disable it?" 4	186	embedded into other programs, as witnessed by \fIdoc/rxvt\-tabbed\fR or
181	.IX Item "I don't like the new selection/popups/hotkeys/perl, how do I change/disable it?"	187	the upcoming \f(CW\*(C`Gtk2::URxvt\*(C'\fR perl module, which features a tabbed urxvt
182	You can disable the perl extension completely by setting the	188	(murxvt) terminal as an example embedding application.
183	\&\fBperl-ext-common\fR resource to the empty string, which also keeps	189	.PP
184	rxvt-unicode from initialising perl, saving memory.	190	\fIHow do I know which rxvt-unicode version I'm using?\fR
185	.Sp	191	.IX Subsection "How do I know which rxvt-unicode version I'm using?"
186	If you only want to disable specific features, you first have to	192	.PP
187	identify which perl extension is responsible. For this, read the section	193	The version number is displayed with the usage (\-h). Also the escape
188	\&\fB\s-1PREPACKAGED\s0 \s-1EXTENSIONS\s0\fR in the @@RXVT_NAME@@\fIperl\fR\|(3) manpage. For	194	sequence \f(CW\*(C`ESC [ 8 n\*(C'\fR sets the window title to the version number. When
189	example, to disable the \fBselection-popup\fR and \fBoption-popup\fR, specify	195	using the @@URXVT_NAME@@c client, the version displayed is that of the
190	this \fBperl-ext-common\fR resource:	196	daemon.
191	.Sp	197	.PP
		198	\fIRxvt-unicode uses gobs of memory, how can I reduce that?\fR
		199	.IX Subsection "Rxvt-unicode uses gobs of memory, how can I reduce that?"
		200	.PP
		201	Rxvt-unicode tries to obey the rule of not charging you for something you
		202	don't use. One thing you should try is to configure out all settings that
		203	you don't need, for example, Xft support is a resource hog by design,
		204	when used. Compiling it out ensures that no Xft font will be loaded
		205	accidentally when rxvt-unicode tries to find a font for your characters.
		206	.PP
		207	Also, many people (me included) like large windows and even larger
		208	scrollback buffers: Without \f(CW\*(C`\-\-enable\-unicode3\*(C'\fR, rxvt-unicode will use
		209	6 bytes per screen cell. For a 160x?? window this amounts to almost a
		210	kilobyte per line. A scrollback buffer of 10000 lines will then (if full)
		211	use 10 Megabytes of memory. With \f(CW\*(C`\-\-enable\-unicode3\*(C'\fR it gets worse, as
		212	rxvt-unicode then uses 8 bytes per screen cell.
		213	.PP
		214	\fIHow can I start @@URXVT_NAME@@d in a race-free way?\fR
		215	.IX Subsection "How can I start @@URXVT_NAME@@d in a race-free way?"
		216	.PP
		217	Try \f(CW\*(C`@@URXVT_NAME@@d \-f \-o\*(C'\fR, which tells @@URXVT_NAME@@d to open the
		218	display, create the listening socket and then fork.
		219	.PP
		220	\fIHow do I distinguish wether I'm running rxvt-unicode or a regular xterm? I need this to decide about setting colors etc.\fR
		221	.IX Subsection "How do I distinguish wether I'm running rxvt-unicode or a regular xterm? I need this to decide about setting colors etc."
		222	.PP
		223	The original rxvt and rxvt-unicode always export the variable \*(L"\s-1COLORTERM\s0\*(R",
		224	so you can check and see if that is set. Note that several programs, \s-1JED\s0,
		225	slrn, Midnight Commander automatically check this variable to decide
		226	whether or not to use color.
		227	.PP
		228	\fIHow do I set the correct, full \s-1IP\s0 address for the \s-1DISPLAY\s0 variable?\fR
		229	.IX Subsection "How do I set the correct, full IP address for the DISPLAY variable?"
		230	.PP
		231	If you've compiled rxvt-unicode with \s-1DISPLAY_IS_IP\s0 and have enabled
		232	insecure mode then it is possible to use the following shell script
		233	snippets to correctly set the display. If your version of rxvt-unicode
		234	wasn't also compiled with \s-1ESCZ_ANSWER\s0 (as assumed in these snippets) then
		235	the \s-1COLORTERM\s0 variable can be used to distinguish rxvt-unicode from a
		236	regular xterm.
		237	.PP
		238	Courtesy of Chuck Blake <cblake@BBN.COM> with the following shell script
		239	snippets:
		240	.PP
192	.Vb 1	241	.Vb 12
193	\& URxvt.perl-ext-common: default,-selection-popup,-option-popup	242	\& # Bourne/Korn/POSIX family of shells:
		243	\& [ ${TERM:-foo} = foo ] && TERM=xterm # assume an xterm if we don't know
		244	\& if [ ${TERM:-foo} = xterm ]; then
		245	\& stty -icanon -echo min 0 time 15 # see if enhanced rxvt or not
		246	\& echo -n '^[Z'
		247	\& read term_id
		248	\& stty icanon echo
		249	\& if [ ""${term_id} = '^[[?1;2C' -a ${DISPLAY:-foo} = foo ]; then
		250	\& echo -n '^[[7n' # query the rxvt we are in for the DISPLAY string
		251	\& read DISPLAY # set it in our local shell
		252	\& fi
		253	\& fi
194	.Ve	254	.Ve
195	.Sp	255	.PP
196	This will keep the default extensions, but disable the two popup	256	\fIHow do I compile the manual pages on my own?\fR
197	extensions. Some extensions can also be configured, for example,	257	.IX Subsection "How do I compile the manual pages on my own?"
198	scrollback search mode is triggered by \fBM\-s\fR. You can move it to any	258	.PP
199	other combination either by setting the \fBsearchable-scrollback\fR resource:	259	You need to have a recent version of perl installed as \fI/usr/bin/perl\fR,
200	.Sp	260	one that comes with \fIpod2man\fR, \fIpod2text\fR and \fIpod2html\fR. Then go to
201	.Vb 1	261	the doc subdirectory and enter \f(CW\*(C`make alldoc\*(C'\fR.
202	\& URxvt.searchable-scrollback: CM-s	262	.PP
203	.Ve
204	.IP "The cursor moves when selecting text in the current input line, how do I switch this off?" 4
205	.IX Item "The cursor moves when selecting text in the current input line, how do I switch this off?"
206	.PD 0
207	.IP "During rlogin/ssh/telnet/etc. sessions, clicking near the cursor outputs strange escape sequences, how do I fix this?" 4
208	.IX Item "During rlogin/ssh/telnet/etc. sessions, clicking near the cursor outputs strange escape sequences, how do I fix this?"
209	.PD
210	These are caused by the \f(CW\*(C`readline\*(C'\fR perl extension. Under normal
211	circumstances, it will move your cursor around when you click into the
212	line that contains it. It tries hard not to do this at the wrong moment,
213	but when running a program that doesn't parse cursor movements or in some
214	cases during rlogin sessions, it fails to detect this properly.
215	.Sp
216	You can permamently switch this feature off by disabling the \f(CW\*(C`readline\*(C'\fR
217	extension:
218	.Sp
219	.Vb 1
220	\& URxvt.perl-ext-common: default,-readline
221	.Ve
222	.IP "Why doesn't rxvt-unicode read my resources?" 4
223	.IX Item "Why doesn't rxvt-unicode read my resources?"
224	Well, why, indeed? It does, in a way very similar to other X
225	applications. Most importantly, this means that if you or your \s-1OS\s0 loads
226	resources into the X display (the right way to do it), rxvt-unicode will
227	ignore any resource files in your home directory. It will only read
228	\&\fI$HOME/.Xdefaults\fR when no resources are attached to the display.
229	.Sp
230	If you have or use an \fI$HOME/.Xresources\fR file, chances are that
231	resources are loaded into your X\-server. In this case, you have to
232	re-login after every change (or run \fIxrdb \-merge \f(CI$HOME\fI/.Xresources\fR).
233	.Sp
234	Also consider the form resources have to use:
235	.Sp
236	.Vb 1
237	\& URxvt.resource: value
238	.Ve
239	.Sp
240	If you want to use another form (there are lots of different ways of
241	specifying resources), make sure you understand wether and why it
242	works. If unsure, use the form above.
243	.IP "I can't get transparency working, what am I doing wrong?" 4
244	.IX Item "I can't get transparency working, what am I doing wrong?"
245	First of all, transparency isn't officially supported in rxvt\-unicode, so
246	you are mostly on your own. Do not bug the author about it (but you may
247	bug everybody else). Also, if you can't get it working consider it a rite
248	of passage: ... and you failed.
249	.Sp
250	Here are four ways to get transparency. \fBDo\fR read the manpage and option
251	descriptions for the programs mentioned and rxvt\-unicode. Really, do it!
252	.Sp
253	1. Use inheritPixmap:
254	.Sp
255	.Vb 2
256	\& Esetroot wallpaper.jpg
257	\& @@RXVT_NAME@@ -ip -tint red -sh 40
258	.Ve
259	.Sp
260	That works. If you think it doesn't, you lack transparency and tinting
261	support, or you are unable to read.
262	.Sp
263	2. Use a simple pixmap and emulate pseudo\-transparency. This enables you
264	to use effects other than tinting and shading: Just shade/tint/whatever
265	your picture with gimp:
266	.Sp
267	.Vb 2
268	\& convert wallpaper.jpg -blur 20x20 -modulate 30 background.xpm
269	\& @@RXVT_NAME@@ -pixmap background.xpm -pe automove-background
270	.Ve
271	.Sp
272	That works. If you think it doesn't, you lack \s-1XPM\s0 and Perl support, or you
273	are unable to read.
274	.Sp
275	3. Use an \s-1ARGB\s0 visual:
276	.Sp
277	.Vb 1
278	\& @@RXVT_NAME@@ -depth 32 -fg grey90 -bg rgba:0000/0000/4444/cccc
279	.Ve
280	.Sp
281	This requires \s-1XFT\s0 support, and the support of your X\-server. If that
282	doesn't work for you, blame Xorg and Keith Packard. \s-1ARGB\s0 visuals aren't
283	there yet, no matter what they claim. Rxvt-Unicode contains the neccessary
284	bugfixes and workarounds for Xft and Xlib to make it work, but that
285	doesn't mean that your \s-1WM\s0 has the required kludges in place.
286	.Sp
287	4. Use xcompmgr and let it do the job:
288	.Sp
289	.Vb 2
290	\& xprop -frame -f _NET_WM_WINDOW_OPACITY 32c \e
291	\& -set _NET_WM_WINDOW_OPACITY 0xc0000000
292	.Ve
293	.Sp
294	Then click on a window you want to make transparent. Replace \f(CW0xc0000000\fR
295	by other values to change the degree of opacity. If it doesn't work and
296	your server crashes, you got to keep the pieces.
297	.IP "Isn't rxvt supposed to be small? Don't all those features bloat?" 4	263	\fIIsn't rxvt-unicode supposed to be small? Don't all those features bloat?\fR
298	.IX Item "Isn't rxvt supposed to be small? Don't all those features bloat?"	264	.IX Subsection "Isn't rxvt-unicode supposed to be small? Don't all those features bloat?"
		265	.PP
299	I often get asked about this, and I think, no, they didn't cause extra	266	I often get asked about this, and I think, no, they didn't cause extra
300	bloat. If you compare a minimal rxvt and a minimal urxvt, you can see	267	bloat. If you compare a minimal rxvt and a minimal urxvt, you can see
301	that the urxvt binary is larger (due to some encoding tables always being	268	that the urxvt binary is larger (due to some encoding tables always being
302	compiled in), but it actually uses less memory (\s-1RSS\s0) after startup. Even	269	compiled in), but it actually uses less memory (\s-1RSS\s0) after startup. Even
303	with \f(CW\*(C`\-\-disable\-everything\*(C'\fR, this comparison is a bit unfair, as many	270	with \f(CW\*(C`\-\-disable\-everything\*(C'\fR, this comparison is a bit unfair, as many
304	features unique to urxvt (locale, encoding conversion, iso14755 etc.) are	271	features unique to urxvt (locale, encoding conversion, iso14755 etc.) are
305	already in use in this mode.	272	already in use in this mode.
306	.Sp	273	.PP
307	.Vb 3	274	.Vb 3
308	\& text data bss drs rss filename	275	\& text data bss drs rss filename
309	\& 98398 1664 24 15695 1824 rxvt --disable-everything	276	\& 98398 1664 24 15695 1824 rxvt --disable-everything
310	\& 188985 9048 66616 18222 1788 urxvt --disable-everything	277	\& 188985 9048 66616 18222 1788 urxvt --disable-everything
311	.Ve	278	.Ve
312	.Sp	279	.PP
313	When you \f(CW\*(C`\-\-enable\-everything\*(C'\fR (which _is_ unfair, as this involves xft	280	When you \f(CW\*(C`\-\-enable\-everything\*(C'\fR (which _is_ unfair, as this involves xft
314	and full locale/XIM support which are quite bloaty inside libX11 and my	281	and full locale/XIM support which are quite bloaty inside libX11 and my
315	libc), the two diverge, but not unreasnobaly so.	282	libc), the two diverge, but not unreasnobaly so.
316	.Sp	283	.PP
317	.Vb 3	284	.Vb 3
318	\& text data bss drs rss filename	285	\& text data bss drs rss filename
319	\& 163431 2152 24 20123 2060 rxvt --enable-everything	286	\& 163431 2152 24 20123 2060 rxvt --enable-everything
320	\& 1035683 49680 66648 29096 3680 urxvt --enable-everything	287	\& 1035683 49680 66648 29096 3680 urxvt --enable-everything
321	.Ve	288	.Ve
322	.Sp	289	.PP
323	The very large size of the text section is explained by the east-asian	290	The very large size of the text section is explained by the east-asian
324	encoding tables, which, if unused, take up disk space but nothing else	291	encoding tables, which, if unused, take up disk space but nothing else
325	and can be compiled out unless you rely on X11 core fonts that use those	292	and can be compiled out unless you rely on X11 core fonts that use those
326	encodings. The \s-1BSS\s0 size comes from the 64k emergency buffer that my c++	293	encodings. The \s-1BSS\s0 size comes from the 64k emergency buffer that my c++
327	compiler allocates (but of course doesn't use unless you are out of	294	compiler allocates (but of course doesn't use unless you are out of
328	memory). Also, using an xft font instead of a core font immediately adds a	295	memory). Also, using an xft font instead of a core font immediately adds a
329	few megabytes of \s-1RSS\s0. Xft indeed is responsible for a lot of \s-1RSS\s0 even when	296	few megabytes of \s-1RSS\s0. Xft indeed is responsible for a lot of \s-1RSS\s0 even when
330	not used.	297	not used.
331	.Sp	298	.PP
332	Of course, due to every character using two or four bytes instead of one,	299	Of course, due to every character using two or four bytes instead of one,
333	a large scrollback buffer will ultimately make rxvt-unicode use more	300	a large scrollback buffer will ultimately make rxvt-unicode use more
334	memory.	301	memory.
335	.Sp	302	.PP
336	Compared to e.g. Eterm (5112k), aterm (3132k) and xterm (4680k), this	303	Compared to e.g. Eterm (5112k), aterm (3132k) and xterm (4680k), this
337	still fares rather well. And compared to some monsters like gnome-terminal	304	still fares rather well. And compared to some monsters like gnome-terminal
338	(21152k + extra 4204k in separate processes) or konsole (22200k + extra	305	(21152k + extra 4204k in separate processes) or konsole (22200k + extra
339	43180k in daemons that stay around after exit, plus half a minute of	306	43180k in daemons that stay around after exit, plus half a minute of
340	startup time, including the hundreds of warnings it spits out), it fares	307	startup time, including the hundreds of warnings it spits out), it fares
341	extremely well *g*.	308	extremely well *g*.
		309	.PP
342	.IP "Why \*(C+, isn't that unportable/bloated/uncool?" 4	310	\fIWhy \*(C+, isn't that unportable/bloated/uncool?\fR
343	.IX Item "Why , isn't that unportable/bloated/uncool?"	311	.IX Subsection "Why , isn't that unportable/bloated/uncool?"
		312	.PP
344	Is this a question? :) It comes up very often. The simple answer is: I had	313	Is this a question? :) It comes up very often. The simple answer is: I had
345	to write it, and \*(C+ allowed me to write and maintain it in a fraction	314	to write it, and \*(C+ allowed me to write and maintain it in a fraction
346	of the time and effort (which is a scarce resource for me). Put even	315	of the time and effort (which is a scarce resource for me). Put even
347	shorter: It simply wouldn't exist without \*(C+.	316	shorter: It simply wouldn't exist without \*(C+.
348	.Sp	317	.PP
349	My personal stance on this is that \*(C+ is less portable than C, but in	318	My personal stance on this is that \*(C+ is less portable than C, but in
350	the case of rxvt-unicode this hardly matters, as its portability limits	319	the case of rxvt-unicode this hardly matters, as its portability limits
351	are defined by things like X11, pseudo terminals, locale support and unix	320	are defined by things like X11, pseudo terminals, locale support and unix
352	domain sockets, which are all less portable than \*(C+ itself.	321	domain sockets, which are all less portable than \*(C+ itself.
353	.Sp	322	.PP
354	Regarding the bloat, see the above question: It's easy to write programs	323	Regarding the bloat, see the above question: It's easy to write programs
355	in C that use gobs of memory, an certainly possible to write programs in	324	in C that use gobs of memory, an certainly possible to write programs in
356	\&\*(C+ that don't. \*(C+ also often comes with large libraries, but this is	325	\&\*(C+ that don't. \*(C+ also often comes with large libraries, but this is
357	not necessarily the case with \s-1GCC\s0. Here is what rxvt links against on my	326	not necessarily the case with \s-1GCC\s0. Here is what rxvt links against on my
358	system with a minimal config:	327	system with a minimal config:
359	.Sp	328	.PP
360	.Vb 4	329	.Vb 4
361	\& libX11.so.6 => /usr/X11R6/lib/libX11.so.6 (0x00002aaaaabc3000)	330	\& libX11.so.6 => /usr/X11R6/lib/libX11.so.6 (0x00002aaaaabc3000)
362	\& libc.so.6 => /lib/libc.so.6 (0x00002aaaaadde000)	331	\& libc.so.6 => /lib/libc.so.6 (0x00002aaaaadde000)
363	\& libdl.so.2 => /lib/libdl.so.2 (0x00002aaaab01d000)	332	\& libdl.so.2 => /lib/libdl.so.2 (0x00002aaaab01d000)
364	\& /lib64/ld-linux-x86-64.so.2 (0x00002aaaaaaab000)	333	\& /lib64/ld-linux-x86-64.so.2 (0x00002aaaaaaab000)
365	.Ve	334	.Ve
366	.Sp	335	.PP
367	And here is rxvt\-unicode:	336	And here is rxvt\-unicode:
368	.Sp	337	.PP
369	.Vb 5	338	.Vb 5
370	\& libX11.so.6 => /usr/X11R6/lib/libX11.so.6 (0x00002aaaaabc3000)	339	\& libX11.so.6 => /usr/X11R6/lib/libX11.so.6 (0x00002aaaaabc3000)
371	\& libgcc_s.so.1 => /lib/libgcc_s.so.1 (0x00002aaaaada2000)	340	\& libgcc_s.so.1 => /lib/libgcc_s.so.1 (0x00002aaaaada2000)
372	\& libc.so.6 => /lib/libc.so.6 (0x00002aaaaaeb0000)	341	\& libc.so.6 => /lib/libc.so.6 (0x00002aaaaaeb0000)
373	\& libdl.so.2 => /lib/libdl.so.2 (0x00002aaaab0ee000)	342	\& libdl.so.2 => /lib/libdl.so.2 (0x00002aaaab0ee000)
374	\& /lib64/ld-linux-x86-64.so.2 (0x00002aaaaaaab000)	343	\& /lib64/ld-linux-x86-64.so.2 (0x00002aaaaaaab000)
375	.Ve	344	.Ve
376	.Sp	345	.PP
377	No large bloated libraries (of course, none were linked in statically),	346	No large bloated libraries (of course, none were linked in statically),
378	except maybe libX11 :)	347	except maybe libX11 :)
379	.IP "Does it support tabs, can I have a tabbed rxvt\-unicode?" 4	348	.Sh "Rendering, Font & Look and Feel Issues"
380	.IX Item "Does it support tabs, can I have a tabbed rxvt-unicode?"	349	.IX Subsection "Rendering, Font & Look and Feel Issues"
381	Beginning with version 7.3, there is a perl extension that implements a	350	\fII can't get transparency working, what am I doing wrong?\fR
382	simple tabbed terminal. It is installed by default, so any of these should	351	.IX Subsection "I can't get transparency working, what am I doing wrong?"
383	give you tabs:	352	.PP
384	.Sp	353	First of all, transparency isn't officially supported in rxvt\-unicode, so
		354	you are mostly on your own. Do not bug the author about it (but you may
		355	bug everybody else). Also, if you can't get it working consider it a rite
		356	of passage: ... and you failed.
		357	.PP
		358	Here are four ways to get transparency. \fBDo\fR read the manpage and option
		359	descriptions for the programs mentioned and rxvt\-unicode. Really, do it!
		360	.PP
		361	1. Use inheritPixmap:
		362	.PP
		363	.Vb 2
		364	\& Esetroot wallpaper.jpg
		365	\& @@URXVT_NAME@@ -ip -tint red -sh 40
		366	.Ve
		367	.PP
		368	That works. If you think it doesn't, you lack transparency and tinting
		369	support, or you are unable to read.
		370	.PP
		371	2. Use a simple pixmap and emulate pseudo\-transparency. This enables you
		372	to use effects other than tinting and shading: Just shade/tint/whatever
		373	your picture with gimp or any other tool:
		374	.PP
		375	.Vb 2
		376	\& convert wallpaper.jpg -blur 20x20 -modulate 30 background.xpm
		377	\& @@URXVT_NAME@@ -pixmap background.xpm -pe automove-background
		378	.Ve
		379	.PP
		380	That works. If you think it doesn't, you lack \s-1XPM\s0 and Perl support, or you
		381	are unable to read.
		382	.PP
		383	3. Use an \s-1ARGB\s0 visual:
		384	.PP
385	.Vb 1	385	.Vb 1
386	\& @@RXVT_NAME@@ -pe tabbed	386	\& @@URXVT_NAME@@ -depth 32 -fg grey90 -bg rgba:0000/0000/4444/cccc
387	.Ve	387	.Ve
388	.Sp	388	.PP
		389	This requires \s-1XFT\s0 support, and the support of your X\-server. If that
		390	doesn't work for you, blame Xorg and Keith Packard. \s-1ARGB\s0 visuals aren't
		391	there yet, no matter what they claim. Rxvt-Unicode contains the neccessary
		392	bugfixes and workarounds for Xft and Xlib to make it work, but that
		393	doesn't mean that your \s-1WM\s0 has the required kludges in place.
		394	.PP
		395	4. Use xcompmgr and let it do the job:
		396	.PP
		397	.Vb 2
		398	\& xprop -frame -f _NET_WM_WINDOW_OPACITY 32c \e
		399	\& -set _NET_WM_WINDOW_OPACITY 0xc0000000
		400	.Ve
		401	.PP
		402	Then click on a window you want to make transparent. Replace \f(CW0xc0000000\fR
		403	by other values to change the degree of opacity. If it doesn't work and
		404	your server crashes, you got to keep the pieces.
		405	.PP
		406	\fIWhy do some chinese characters look so different than others?\fR
		407	.IX Subsection "Why do some chinese characters look so different than others?"
		408	.PP
		409	This is because there is a difference between script and language \*(--
		410	rxvt-unicode does not know which language the text that is output is,
		411	as it only knows the unicode character codes. If rxvt-unicode first
		412	sees a japanese/chinese character, it might choose a japanese font for
		413	display. Subsequent japanese characters will use that font. Now, many
		414	chinese characters aren't represented in japanese fonts, so when the first
		415	non-japanese character comes up, rxvt-unicode will look for a chinese font
		416	\&\*(-- unfortunately at this point, it will still use the japanese font for
		417	chinese characters that are also in the japanese font.
		418	.PP
		419	The workaround is easy: just tag a chinese font at the end of your font
		420	list (see the previous question). The key is to view the font list as
		421	a preference list: If you expect more japanese, list a japanese font
		422	first. If you expect more chinese, put a chinese font first.
		423	.PP
		424	In the future it might be possible to switch language preferences at
		425	runtime (the internal data structure has no problem with using different
		426	fonts for the same character at the same time, but no interface for this
		427	has been designed yet).
		428	.PP
		429	Until then, you might get away with switching fonts at runtime (see \*(L"Can I switch the fonts at runtime?\*(R" later in this document).
		430	.PP
		431	\fIWhy does rxvt-unicode sometimes leave pixel droppings?\fR
		432	.IX Subsection "Why does rxvt-unicode sometimes leave pixel droppings?"
		433	.PP
		434	Most fonts were not designed for terminal use, which means that character
		435	size varies a lot. A font that is otherwise fine for terminal use might
		436	contain some characters that are simply too wide. Rxvt-unicode will avoid
		437	these characters. For characters that are just \*(L"a bit\*(R" too wide a special
		438	\&\*(L"careful\*(R" rendering mode is used that redraws adjacent characters.
		439	.PP
		440	All of this requires that fonts do not lie about character sizes,
		441	however: Xft fonts often draw glyphs larger than their acclaimed bounding
		442	box, and rxvt-unicode has no way of detecting this (the correct way is to
		443	ask for the character bounding box, which unfortunately is wrong in these
		444	cases).
		445	.PP
		446	It's not clear (to me at least), wether this is a bug in Xft, freetype,
		447	or the respective font. If you encounter this problem you might try using
		448	the \f(CW\*(C`\-lsp\*(C'\fR option to give the font more height. If that doesn't work, you
		449	might be forced to use a different font.
		450	.PP
		451	All of this is not a problem when using X11 core fonts, as their bounding
		452	box data is correct.
		453	.PP
		454	\fIHow can I keep rxvt-unicode from using reverse video so much?\fR
		455	.IX Subsection "How can I keep rxvt-unicode from using reverse video so much?"
		456	.PP
		457	First of all, make sure you are running with the right terminal settings
		458	(\f(CW\*(C`TERM=rxvt\-unicode\*(C'\fR), which will get rid of most of these effects. Then
		459	make sure you have specified colours for italic and bold, as otherwise
		460	rxvt-unicode might use reverse video to simulate the effect:
		461	.PP
		462	.Vb 2
		463	\& URxvt.colorBD: white
		464	\& URxvt.colorIT: green
		465	.Ve
		466	.PP
		467	\fISome programs assume totally weird colours (red instead of blue), how can I fix that?\fR
		468	.IX Subsection "Some programs assume totally weird colours (red instead of blue), how can I fix that?"
		469	.PP
		470	For some unexplainable reason, some rare programs assume a very weird
		471	colour palette when confronted with a terminal with more than the standard
		472	8 colours (rxvt\-unicode supports 88). The right fix is, of course, to fix
		473	these programs not to assume non-ISO colours without very good reasons.
		474	.PP
		475	In the meantime, you can either edit your \f(CW\*(C`rxvt\-unicode\*(C'\fR terminfo
		476	definition to only claim 8 colour support or use \f(CW\*(C`TERM=rxvt\*(C'\fR, which will
		477	fix colours but keep you from using other rxvt-unicode features.
		478	.PP
		479	\fICan I switch the fonts at runtime?\fR
		480	.IX Subsection "Can I switch the fonts at runtime?"
		481	.PP
		482	Yes, using an escape sequence. Try something like this, which has the same
		483	effect as using the \f(CW\*(C`\-fn\*(C'\fR switch, and takes effect immediately:
		484	.PP
389	.Vb 1	485	.Vb 1
		486	\& printf '\ee]50;%s\e007' "9x15bold,xft:Kochi Gothic"
		487	.Ve
		488	.PP
		489	This is useful if you e.g. work primarily with japanese (and prefer a
		490	japanese font), but you have to switch to chinese temporarily, where
		491	japanese fonts would only be in your way.
		492	.PP
		493	You can think of this as a kind of manual \s-1ISO\-2022\s0 switching.
		494	.PP
		495	\fIWhy do italic characters look as if clipped?\fR
		496	.IX Subsection "Why do italic characters look as if clipped?"
		497	.PP
		498	Many fonts have difficulties with italic characters and hinting. For
		499	example, the otherwise very nicely hinted font \f(CW\*(C`xft:Bitstream Vera Sans
		500	Mono\*(C'\fR completely fails in it's italic face. A workaround might be to
		501	enable freetype autohinting, i.e. like this:
		502	.PP
		503	.Vb 2
		504	\& URxvt.italicFont: xft:Bitstream Vera Sans Mono:italic:autohint=true
		505	\& URxvt.boldItalicFont: xft:Bitstream Vera Sans Mono:bold:italic:autohint=true
		506	.Ve
		507	.PP
		508	\fICan I speed up Xft rendering somehow?\fR
		509	.IX Subsection "Can I speed up Xft rendering somehow?"
		510	.PP
		511	Yes, the most obvious way to speed it up is to avoid Xft entirely, as
		512	it is simply slow. If you still want Xft fonts you might try to disable
		513	antialiasing (by appending \f(CW\*(C`:antialias=false\*(C'\fR), which saves lots of
		514	memory and also speeds up rendering considerably.
		515	.PP
		516	\fIRxvt-unicode doesn't seem to anti-alias its fonts, what is wrong?\fR
		517	.IX Subsection "Rxvt-unicode doesn't seem to anti-alias its fonts, what is wrong?"
		518	.PP
		519	Rxvt-unicode will use whatever you specify as a font. If it needs to
		520	fall back to it's default font search list it will prefer X11 core
		521	fonts, because they are small and fast, and then use Xft fonts. It has
		522	antialiasing disabled for most of them, because the author thinks they
		523	look best that way.
		524	.PP
		525	If you want antialiasing, you have to specify the fonts manually.
		526	.PP
		527	\fIWhat's with this bold/blink stuff?\fR
		528	.IX Subsection "What's with this bold/blink stuff?"
		529	.PP
		530	If no bold colour is set via \f(CW\*(C`colorBD:\*(C'\fR, bold will invert text using the
		531	standard foreground colour.
		532	.PP
		533	For the standard background colour, blinking will actually make the
		534	text blink when compiled with \f(CW\*(C`\-\-enable\-blinking\*(C'\fR. with standard
		535	colours. Without \f(CW\*(C`\-\-enable\-blinking\*(C'\fR, the blink attribute will be
		536	ignored.
		537	.PP
		538	On \s-1ANSI\s0 colours, bold/blink attributes are used to set high-intensity
		539	foreground/background colors.
		540	.PP
		541	color0\-7 are the low-intensity colors.
		542	.PP
		543	color8\-15 are the corresponding high-intensity colors.
		544	.PP
		545	\fII don't like the screen colors. How do I change them?\fR
		546	.IX Subsection "I don't like the screen colors. How do I change them?"
		547	.PP
		548	You can change the screen colors at run-time using \fI~/.Xdefaults\fR
		549	resources (or as long\-options).
		550	.PP
		551	Here are values that are supposed to resemble a \s-1VGA\s0 screen,
		552	including the murky brown that passes for low-intensity yellow:
		553	.PP
		554	.Vb 8
		555	\& URxvt.color0: #000000
		556	\& URxvt.color1: #A80000
		557	\& URxvt.color2: #00A800
		558	\& URxvt.color3: #A8A800
		559	\& URxvt.color4: #0000A8
		560	\& URxvt.color5: #A800A8
		561	\& URxvt.color6: #00A8A8
		562	\& URxvt.color7: #A8A8A8
		563	.Ve
		564	.PP
		565	.Vb 8
		566	\& URxvt.color8: #000054
		567	\& URxvt.color9: #FF0054
		568	\& URxvt.color10: #00FF54
		569	\& URxvt.color11: #FFFF54
		570	\& URxvt.color12: #0000FF
		571	\& URxvt.color13: #FF00FF
		572	\& URxvt.color14: #00FFFF
		573	\& URxvt.color15: #FFFFFF
		574	.Ve
		575	.PP
		576	And here is a more complete set of non-standard colors described (not by
		577	me) as \*(L"pretty girly\*(R".
		578	.PP
		579	.Vb 18
		580	\& URxvt.cursorColor: #dc74d1
		581	\& URxvt.pointerColor: #dc74d1
		582	\& URxvt.background: #0e0e0e
		583	\& URxvt.foreground: #4ad5e1
		584	\& URxvt.color0: #000000
		585	\& URxvt.color8: #8b8f93
		586	\& URxvt.color1: #dc74d1
		587	\& URxvt.color9: #dc74d1
		588	\& URxvt.color2: #0eb8c7
		589	\& URxvt.color10: #0eb8c7
		590	\& URxvt.color3: #dfe37e
		591	\& URxvt.color11: #dfe37e
		592	\& URxvt.color5: #9e88f0
		593	\& URxvt.color13: #9e88f0
		594	\& URxvt.color6: #73f7ff
		595	\& URxvt.color14: #73f7ff
		596	\& URxvt.color7: #e1dddd
		597	\& URxvt.color15: #e1dddd
		598	.Ve
		599	.PP
		600	\fIWhy do some characters look so much different than others?\fR
		601	.IX Subsection "Why do some characters look so much different than others?"
		602	.PP
		603	See next entry.
		604	.PP
		605	\fIHow does rxvt-unicode choose fonts?\fR
		606	.IX Subsection "How does rxvt-unicode choose fonts?"
		607	.PP
		608	Most fonts do not contain the full range of Unicode, which is
		609	fine. Chances are that the font you (or the admin/package maintainer of
		610	your system/os) have specified does not cover all the characters you want
		611	to display.
		612	.PP
		613	\&\fBrxvt-unicode\fR makes a best-effort try at finding a replacement
		614	font. Often the result is fine, but sometimes the chosen font looks
		615	bad/ugly/wrong. Some fonts have totally strange characters that don't
		616	resemble the correct glyph at all, and rxvt-unicode lacks the artificial
		617	intelligence to detect that a specific glyph is wrong: it has to believe
		618	the font that the characters it claims to contain indeed look correct.
		619	.PP
		620	In that case, select a font of your taste and add it to the font list,
		621	e.g.:
		622	.PP
		623	.Vb 1
		624	\& @@URXVT_NAME@@ -fn basefont,font2,font3...
		625	.Ve
		626	.PP
		627	When rxvt-unicode sees a character, it will first look at the base
		628	font. If the base font does not contain the character, it will go to the
		629	next font, and so on. Specifying your own fonts will also speed up this
		630	search and use less resources within rxvt-unicode and the X\-server.
		631	.PP
		632	The only limitation is that none of the fonts may be larger than the base
		633	font, as the base font defines the terminal character cell size, which
		634	must be the same due to the way terminals work.
		635	.Sh "Keyboard, Mouse & User Interaction"
		636	.IX Subsection "Keyboard, Mouse & User Interaction"
		637	\fIThe new selection selects pieces that are too big, how can I select single words?\fR
		638	.IX Subsection "The new selection selects pieces that are too big, how can I select single words?"
		639	.PP
		640	If you want to select e.g. alphanumeric words, you can use the following
		641	setting:
		642	.PP
		643	.Vb 1
		644	\& URxvt.selection.pattern-0: ([[:word:]]+)
		645	.Ve
		646	.PP
		647	If you click more than twice, the selection will be extended
		648	more and more.
		649	.PP
		650	To get a selection that is very similar to the old code, try this pattern:
		651	.PP
		652	.Vb 1
		653	\& URxvt.selection.pattern-0: ([^"&'()*,;<=>?@[\e\e\e\e]^`{|})]+)
		654	.Ve
		655	.PP
		656	Please also note that the \fILeftClick Shift-LeftClik\fR combination also
		657	selects words like the old code.
		658	.PP
		659	\fII don't like the new selection/popups/hotkeys/perl, how do I change/disable it?\fR
		660	.IX Subsection "I don't like the new selection/popups/hotkeys/perl, how do I change/disable it?"
		661	.PP
		662	You can disable the perl extension completely by setting the
		663	\&\fBperl-ext-common\fR resource to the empty string, which also keeps
		664	rxvt-unicode from initialising perl, saving memory.
		665	.PP
		666	If you only want to disable specific features, you first have to
		667	identify which perl extension is responsible. For this, read the section
		668	\&\fB\s-1PREPACKAGED\s0 \s-1EXTENSIONS\s0\fR in the @@URXVT_NAME@@\fIperl\fR\|(3) manpage. For
		669	example, to disable the \fBselection-popup\fR and \fBoption-popup\fR, specify
		670	this \fBperl-ext-common\fR resource:
		671	.PP
		672	.Vb 1
		673	\& URxvt.perl-ext-common: default,-selection-popup,-option-popup
		674	.Ve
		675	.PP
		676	This will keep the default extensions, but disable the two popup
		677	extensions. Some extensions can also be configured, for example,
		678	scrollback search mode is triggered by \fBM\-s\fR. You can move it to any
		679	other combination either by setting the \fBsearchable-scrollback\fR resource:
		680	.PP
		681	.Vb 1
		682	\& URxvt.searchable-scrollback: CM-s
		683	.Ve
		684	.PP
		685	\fIThe cursor moves when selecting text in the current input line, how do I switch this off?\fR
		686	.IX Subsection "The cursor moves when selecting text in the current input line, how do I switch this off?"
		687	.PP
		688	See next entry.
		689	.PP
		690	\fIDuring rlogin/ssh/telnet/etc. sessions, clicking near the cursor outputs strange escape sequences, how do I fix this?\fR
		691	.IX Subsection "During rlogin/ssh/telnet/etc. sessions, clicking near the cursor outputs strange escape sequences, how do I fix this?"
		692	.PP
		693	These are caused by the \f(CW\*(C`readline\*(C'\fR perl extension. Under normal
		694	circumstances, it will move your cursor around when you click into the
		695	line that contains it. It tries hard not to do this at the wrong moment,
		696	but when running a program that doesn't parse cursor movements or in some
		697	cases during rlogin sessions, it fails to detect this properly.
		698	.PP
		699	You can permamently switch this feature off by disabling the \f(CW\*(C`readline\*(C'\fR
		700	extension:
		701	.PP
		702	.Vb 1
390	\& URxvt.perl-ext-common: default,tabbed	703	\& URxvt.perl-ext-common: default,-readline
391	.Ve	704	.Ve
392	.Sp	705	.PP
393	It will also work fine with tabbing functionality of many window managers	706	\fIMy numerical keypad acts weird and generates differing output?\fR
394	or similar tabbing programs, and its embedding-features allow it to be	707	.IX Subsection "My numerical keypad acts weird and generates differing output?"
395	embedded into other programs, as witnessed by \fIdoc/rxvt\-tabbed\fR or	708	.PP
396	the upcoming \f(CW\*(C`Gtk2::URxvt\*(C'\fR perl module, which features a tabbed urxvt	709	Some Debian GNUL/Linux users seem to have this problem, although no
397	(murxvt) terminal as an example embedding application.	710	specific details were reported so far. It is possible that this is caused
398	.IP "How do I know which rxvt-unicode version I'm using?" 4	711	by the wrong \f(CW\*(C`TERM\*(C'\fR setting, although the details of wether and how
399	.IX Item "How do I know which rxvt-unicode version I'm using?"	712	this can happen are unknown, as \f(CW\*(C`TERM=rxvt\*(C'\fR should offer a compatible
400	The version number is displayed with the usage (\-h). Also the escape	713	keymap. See the answer to the previous question, and please report if that
401	sequence \f(CW\*(C`ESC [ 8 n\*(C'\fR sets the window title to the version number. When	714	helped.
402	using the @@RXVT_NAME@@c client, the version displayed is that of the	715	.PP
403	daemon.	716	\fIMy Compose (Multi_key) key is no longer working.\fR
404	.IP "I am using Debian GNU/Linux and have a problem..." 4	717	.IX Subsection "My Compose (Multi_key) key is no longer working."
405	.IX Item "I am using Debian GNU/Linux and have a problem..."	718	.PP
406	The Debian GNU/Linux package of rxvt-unicode in sarge contains large	719	The most common causes for this are that either your locale is not set
407	patches that considerably change the behaviour of rxvt-unicode (but	720	correctly, or you specified a \fBpreeditStyle\fR that is not supported by
408	unfortunately this notice has been removed). Before reporting a bug to	721	your input method. For example, if you specified \fBOverTheSpot\fR and
409	the original rxvt-unicode author please download and install the genuine	722	your input method (e.g. the default input method handling Compose keys)
410	version (<http://software.schmorp.de#rxvt\-unicode>) and try to reproduce	723	does not support this (for instance because it is not visual), then
411	the problem. If you cannot, chances are that the problems are specific to	724	rxvt-unicode will continue without an input method.
412	Debian GNU/Linux, in which case it should be reported via the Debian Bug	725	.PP
413	Tracking System (use \f(CW\*(C`reportbug\*(C'\fR to report the bug).	726	In this case either do not specify a \fBpreeditStyle\fR or specify more than
414	.Sp	727	one pre-edit style, such as \fBOverTheSpot,Root,None\fR.
415	For other problems that also affect the Debian package, you can and	728	.PP
416	probably should use the Debian \s-1BTS\s0, too, because, after all, it's also a	729	\fII cannot type \f(CI\*(C`Ctrl\-Shift\-2\*(C'\fI to get an \s-1ASCII\s0 \s-1NUL\s0 character due to \s-1ISO\s0 14755\fR
417	bug in the Debian version and it serves as a reminder for other users that	730	.IX Subsection "I cannot type Ctrl-Shift-2 to get an ASCII NUL character due to ISO 14755"
418	might encounter the same issue.	731	.PP
419	.IP "I am maintaining rxvt-unicode for distribution/OS \s-1XXX\s0, any recommendation?" 4	732	Either try \f(CW\*(C`Ctrl\-2\*(C'\fR alone (it often is mapped to \s-1ASCII\s0 \s-1NUL\s0 even on
420	.IX Item "I am maintaining rxvt-unicode for distribution/OS XXX, any recommendation?"	733	international keyboards) or simply use \s-1ISO\s0 14755 support to your
421	You should build one binary with the default options. \fIconfigure\fR	734	advantage, typing <Ctrl\-Shift\-0> to get a \s-1ASCII\s0 \s-1NUL\s0. This works for other
422	now enables most useful options, and the trend goes to making them	735	codes, too, such as \f(CW\*(C`Ctrl\-Shift\-1\-d\*(C'\fR to type the default telnet escape
423	runtime\-switchable, too, so there is usually no drawback to enbaling them,	736	character and so on.
424	except higher disk and possibly memory usage. The perl interpreter should	737	.PP
425	be enabled, as important functionality (menus, selection, likely more in	738	\fIMouse cut/paste suddenly no longer works.\fR
426	the future) depends on it.	739	.IX Subsection "Mouse cut/paste suddenly no longer works."
427	.Sp	740	.PP
428	You should not overwrite the \f(CW\*(C`perl\-ext\-common\*(C'\fR snd \f(CW\*(C`perl\-ext\*(C'\fR resources	741	Make sure that mouse reporting is actually turned off since killing
429	system-wide (except maybe with \f(CW\*(C`defaults\*(C'\fR). This will result in useful	742	some editors prematurely may leave the mouse in mouse report mode. I've
430	behaviour. If your distribution aims at low memory, add an empty	743	heard that tcsh may use mouse reporting unless it otherwise specified. A
431	\&\f(CW\*(C`perl\-ext\-common\*(C'\fR resource to the app-defaults file. This will keep the	744	quick check is to see if cut/paste works when the Alt or Shift keys are
432	perl interpreter disabled until the user enables it.	745	depressed.
433	.Sp	746	.PP
434	If you can/want build more binaries, I recommend building a minimal	747	\fIWhat's with the strange Backspace/Delete key behaviour?\fR
435	one with \f(CW\*(C`\-\-disable\-everything\*(C'\fR (very useful) and a maximal one with	748	.IX Subsection "What's with the strange Backspace/Delete key behaviour?"
436	\&\f(CW\*(C`\-\-enable\-everything\*(C'\fR (less useful, it will be very big due to a lot of	749	.PP
437	encodings built-in that increase download times and are rarely used).	750	Assuming that the physical Backspace key corresponds to the
438	.IP "I need to make it setuid/setgid to support utmp/ptys on my \s-1OS\s0, is this safe?" 4	751	BackSpace keysym (not likely for Linux ... see the following
439	.IX Item "I need to make it setuid/setgid to support utmp/ptys on my OS, is this safe?"	752	question) there are two standard values that can be used for
440	It should be, starting with release 7.1. You are encouraged to properly	753	Backspace: \f(CW\*(C`^H\*(C'\fR and \f(CW\*(C`^?\*(C'\fR.
441	install urxvt with privileges necessary for your \s-1OS\s0 now.	754	.PP
442	.Sp	755	Historically, either value is correct, but rxvt-unicode adopts the debian
443	When rxvt-unicode detects that it runs setuid or setgid, it will fork	756	policy of using \f(CW\*(C`^?\*(C'\fR when unsure, because it's the one only only correct
444	into a helper process for privileged operations (pty handling on some	757	choice :).
445	systems, utmp/wtmp/lastlog handling on others) and drop privileges	758	.PP
446	immediately. This is much safer than most other terminals that keep	759	Rxvt-unicode tries to inherit the current stty settings and uses the value
447	privileges while running (but is more relevant to urxvt, as it contains	760	of `erase' to guess the value for backspace. If rxvt-unicode wasn't
448	things as perl interpreters, which might be \*(L"helpful\*(R" to attackers).	761	started from a terminal (say, from a menu or by remote shell), then the
449	.Sp	762	system value of `erase', which corresponds to \s-1CERASE\s0 in <termios.h>, will
450	This forking is done as the very first within \fImain()\fR, which is very early	763	be used (which may not be the same as your stty setting).
451	and reduces possible bugs to initialisation code run before \fImain()\fR, or	764	.PP
452	things like the dynamic loader of your system, which should result in very	765	For starting a new rxvt\-unicode:
453	little risk.	766	.PP
		767	.Vb 3
		768	\& # use Backspace = ^H
		769	\& $ stty erase ^H
		770	\& $ @@URXVT_NAME@@
		771	.Ve
		772	.PP
		773	.Vb 3
		774	\& # use Backspace = ^?
		775	\& $ stty erase ^?
		776	\& $ @@URXVT_NAME@@
		777	.Ve
		778	.PP
		779	Toggle with \f(CW\*(C`ESC [ 36 h\*(C'\fR / \f(CW\*(C`ESC [ 36 l\*(C'\fR.
		780	.PP
		781	For an existing rxvt\-unicode:
		782	.PP
		783	.Vb 3
		784	\& # use Backspace = ^H
		785	\& $ stty erase ^H
		786	\& $ echo -n "^[[36h"
		787	.Ve
		788	.PP
		789	.Vb 3
		790	\& # use Backspace = ^?
		791	\& $ stty erase ^?
		792	\& $ echo -n "^[[36l"
		793	.Ve
		794	.PP
		795	This helps satisfy some of the Backspace discrepancies that occur, but
		796	if you use Backspace = \f(CW\*(C`^H\*(C'\fR, make sure that the termcap/terminfo value
		797	properly reflects that.
		798	.PP
		799	The Delete key is a another casualty of the ill-defined Backspace problem.
		800	To avoid confusion between the Backspace and Delete keys, the Delete
		801	key has been assigned an escape sequence to match the vt100 for Execute
		802	(\f(CW\*(C`ESC [ 3 ~\*(C'\fR) and is in the supplied termcap/terminfo.
		803	.PP
		804	Some other Backspace problems:
		805	.PP
		806	some editors use termcap/terminfo,
		807	some editors (vim I'm told) expect Backspace = ^H,
		808	\&\s-1GNU\s0 Emacs (and Emacs-like editors) use ^H for help.
		809	.PP
		810	Perhaps someday this will all be resolved in a consistent manner.
		811	.PP
		812	\fII don't like the key\-bindings. How do I change them?\fR
		813	.IX Subsection "I don't like the key-bindings. How do I change them?"
		814	.PP
		815	There are some compile-time selections available via configure. Unless
		816	you have run \*(L"configure\*(R" with the \f(CW\*(C`\-\-disable\-resources\*(C'\fR option you can
		817	use the `keysym' resource to alter the keystrings associated with keysyms.
		818	.PP
		819	Here's an example for a URxvt session started using \f(CW\*(C`@@URXVT_NAME@@ \-name URxvt\*(C'\fR
		820	.PP
		821	.Vb 20
		822	\& URxvt.keysym.Home: \e033[1~
		823	\& URxvt.keysym.End: \e033[4~
		824	\& URxvt.keysym.C-apostrophe: \e033<C-'>
		825	\& URxvt.keysym.C-slash: \e033<C-/>
		826	\& URxvt.keysym.C-semicolon: \e033<C-;>
		827	\& URxvt.keysym.C-grave: \e033<C-`>
		828	\& URxvt.keysym.C-comma: \e033<C-,>
		829	\& URxvt.keysym.C-period: \e033<C-.>
		830	\& URxvt.keysym.C-0x60: \e033<C-`>
		831	\& URxvt.keysym.C-Tab: \e033<C-Tab>
		832	\& URxvt.keysym.C-Return: \e033<C-Return>
		833	\& URxvt.keysym.S-Return: \e033<S-Return>
		834	\& URxvt.keysym.S-space: \e033<S-Space>
		835	\& URxvt.keysym.M-Up: \e033<M-Up>
		836	\& URxvt.keysym.M-Down: \e033<M-Down>
		837	\& URxvt.keysym.M-Left: \e033<M-Left>
		838	\& URxvt.keysym.M-Right: \e033<M-Right>
		839	\& URxvt.keysym.M-C-0: list \e033<M-C- 0123456789 >
		840	\& URxvt.keysym.M-C-a: list \e033<M-C- abcdefghijklmnopqrstuvwxyz >
		841	\& URxvt.keysym.F12: command:\e033]701;zh_CN.GBK\e007
		842	.Ve
		843	.PP
		844	See some more examples in the documentation for the \fBkeysym\fR resource.
		845	.PP
		846	\fII'm using keyboard model \s-1XXX\s0 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\fR
		847	.IX Subsection "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"
		848	.PP
		849	.Vb 6
		850	\& KP_Insert == Insert
		851	\& F22 == Print
		852	\& F27 == Home
		853	\& F29 == Prior
		854	\& F33 == End
		855	\& F35 == Next
		856	.Ve
		857	.PP
		858	Rather than have rxvt-unicode try to accommodate all the various possible
		859	keyboard mappings, it is better to use `xmodmap' to remap the keys as
		860	required for your particular machine.
		861	.Sh "Terminal Configuration"
		862	.IX Subsection "Terminal Configuration"
		863	\fIWhy doesn't rxvt-unicode read my resources?\fR
		864	.IX Subsection "Why doesn't rxvt-unicode read my resources?"
		865	.PP
		866	Well, why, indeed? It does, in a way very similar to other X
		867	applications. Most importantly, this means that if you or your \s-1OS\s0 loads
		868	resources into the X display (the right way to do it), rxvt-unicode will
		869	ignore any resource files in your home directory. It will only read
		870	\&\fI$HOME/.Xdefaults\fR when no resources are attached to the display.
		871	.PP
		872	If you have or use an \fI$HOME/.Xresources\fR file, chances are that
		873	resources are loaded into your X\-server. In this case, you have to
		874	re-login after every change (or run \fIxrdb \-merge \f(CI$HOME\fI/.Xresources\fR).
		875	.PP
		876	Also consider the form resources have to use:
		877	.PP
		878	.Vb 1
		879	\& URxvt.resource: value
		880	.Ve
		881	.PP
		882	If you want to use another form (there are lots of different ways of
		883	specifying resources), make sure you understand wether and why it
		884	works. If unsure, use the form above.
		885	.PP
454	.IP "When I log-in to another system it tells me about missing terminfo data?" 4	886	\fIWhen I log-in to another system it tells me about missing terminfo data?\fR
455	.IX Item "When I log-in to another system it tells me about missing terminfo data?"	887	.IX Subsection "When I log-in to another system it tells me about missing terminfo data?"
		888	.PP
456	The terminal description used by rxvt-unicode is not as widely available	889	The terminal description used by rxvt-unicode is not as widely available
457	as that for xterm, or even rxvt (for which the same problem often arises).	890	as that for xterm, or even rxvt (for which the same problem often arises).
458	.Sp	891	.PP
459	The correct solution for this problem is to install the terminfo, this can	892	The correct solution for this problem is to install the terminfo, this can
460	be done like this (with ncurses' infocmp):	893	be done like this (with ncurses' infocmp):
461	.Sp	894	.PP
462	.Vb 2	895	.Vb 2
463	\& REMOTE=remotesystem.domain	896	\& REMOTE=remotesystem.domain
464	\& infocmp rxvt-unicode | ssh $REMOTE "cat >/tmp/ti && tic /tmp/ti"	897	\& infocmp rxvt-unicode | ssh $REMOTE "cat >/tmp/ti && tic /tmp/ti"
465	.Ve	898	.Ve
466	.Sp	899	.PP
467	\&... or by installing rxvt-unicode normally on the remote system,	900	\&... or by installing rxvt-unicode normally on the remote system,
468	.Sp	901	.PP
469	If you cannot or do not want to do this, then you can simply set	902	If you cannot or do not want to do this, then you can simply set
470	\&\f(CW\*(C`TERM=rxvt\*(C'\fR or even \f(CW\*(C`TERM=xterm\*(C'\fR, and live with the small number of	903	\&\f(CW\*(C`TERM=rxvt\*(C'\fR or even \f(CW\*(C`TERM=xterm\*(C'\fR, and live with the small number of
471	problems arising, which includes wrong keymapping, less and different	904	problems arising, which includes wrong keymapping, less and different
472	colours and some refresh errors in fullscreen applications. It's a nice	905	colours and some refresh errors in fullscreen applications. It's a nice
473	quick-and-dirty workaround for rare cases, though.	906	quick-and-dirty workaround for rare cases, though.
474	.Sp	907	.PP
475	If you always want to do this (and are fine with the consequences) you	908	If you always want to do this (and are fine with the consequences) you
476	can either recompile rxvt-unicode with the desired \s-1TERM\s0 value or use a	909	can either recompile rxvt-unicode with the desired \s-1TERM\s0 value or use a
477	resource to set it:	910	resource to set it:
478	.Sp	911	.PP
479	.Vb 1	912	.Vb 1
480	\& URxvt.termName: rxvt	913	\& URxvt.termName: rxvt
481	.Ve	914	.Ve
482	.Sp	915	.PP
483	If you don't plan to use \fBrxvt\fR (quite common...) you could also replace	916	If you don't plan to use \fBrxvt\fR (quite common...) you could also replace
484	the rxvt terminfo file with the rxvt-unicode one.	917	the rxvt terminfo file with the rxvt-unicode one and use \f(CW\*(C`TERM=rxvt\*(C'\fR.
485	.ie n .IP """tic"" outputs some error when compiling the terminfo entry." 4	918	.PP
486	.el .IP "\f(CWtic\fR outputs some error when compiling the terminfo entry." 4	919	\fI\f(CI\*(C`tic\*(C'\fI outputs some error when compiling the terminfo entry.\fR
487	.IX Item "tic outputs some error when compiling the terminfo entry."	920	.IX Subsection "tic outputs some error when compiling the terminfo entry."
		921	.PP
488	Most likely it's the empty definition for \f(CW\*(C`enacs=\*(C'\fR. Just replace it by	922	Most likely it's the empty definition for \f(CW\*(C`enacs=\*(C'\fR. Just replace it by
489	\&\f(CW\*(C`enacs=\eE[0@\*(C'\fR and try again.	923	\&\f(CW\*(C`enacs=\eE[0@\*(C'\fR and try again.
490	.ie n .IP """bash""'s readline does not work correctly under @@RXVT_NAME@@." 4	924	.PP
491	.el .IP "\f(CWbash\fR's readline does not work correctly under @@RXVT_NAME@@." 4	925	\fI\f(CI\*(C`bash\*(C'\fI's readline does not work correctly under @@URXVT_NAME@@.\fR
492	.IX Item "bash's readline does not work correctly under @@RXVT_NAME@@."	926	.IX Subsection "bash's readline does not work correctly under @@URXVT_NAME@@."
493	.PD 0	927	.PP
		928	See next entry.
		929	.PP
494	.IP "I need a termcap file entry." 4	930	\fII need a termcap file entry.\fR
495	.IX Item "I need a termcap file entry."	931	.IX Subsection "I need a termcap file entry."
496	.PD	932	.PP
497	One reason you might want this is that some distributions or operating	933	One reason you might want this is that some distributions or operating
498	systems still compile some programs using the long-obsoleted termcap	934	systems still compile some programs using the long-obsoleted termcap
499	library (Fedora Core's bash is one example) and rely on a termcap entry	935	library (Fedora Core's bash is one example) and rely on a termcap entry
500	for \f(CW\*(C`rxvt\-unicode\*(C'\fR.	936	for \f(CW\*(C`rxvt\-unicode\*(C'\fR.
501	.Sp	937	.PP
502	You could use rxvt's termcap entry with resonable results in many cases.	938	You could use rxvt's termcap entry with resonable results in many cases.
503	You can also create a termcap entry by using terminfo's infocmp program	939	You can also create a termcap entry by using terminfo's infocmp program
504	like this:	940	like this:
505	.Sp	941	.PP
506	.Vb 1	942	.Vb 1
507	\& infocmp -C rxvt-unicode	943	\& infocmp -C rxvt-unicode
508	.Ve	944	.Ve
509	.Sp	945	.PP
510	Or you could use this termcap entry, generated by the command above:	946	Or you could use this termcap entry, generated by the command above:
511	.Sp	947	.PP
512	.Vb 20	948	.Vb 20
513	\& rxvt-unicode|rxvt-unicode terminal (X Window System):\e	949	\& rxvt-unicode|rxvt-unicode terminal (X Window System):\e
514	\& :am:bw:eo:km:mi:ms:xn:xo:\e	950	\& :am:bw:eo:km:mi:ms:xn:xo:\e
515	\& :co#80:it#8:li#24:lm#0:\e	951	\& :co#80:it#8:li#24:lm#0:\e
516	\& :AL=\eE[%dL:DC=\eE[%dP:DL=\eE[%dM:DO=\eE[%dB:IC=\eE[%d@:\e	952	\& :AL=\eE[%dL:DC=\eE[%dP:DL=\eE[%dM:DO=\eE[%dB:IC=\eE[%d@:\e
…		…
529	\& :sc=\eE7:se=\eE[27m:sf=^J:so=\eE[7m:sr=\eEM:st=\eEH:ta=^I:\e	965	\& :sc=\eE7:se=\eE[27m:sf=^J:so=\eE[7m:sr=\eEM:st=\eEH:ta=^I:\e
530	\& :te=\eE[r\eE[?1049l:ti=\eE[?1049h:ue=\eE[24m:up=\eE[A:\e	966	\& :te=\eE[r\eE[?1049l:ti=\eE[?1049h:ue=\eE[24m:up=\eE[A:\e
531	\& :us=\eE[4m:vb=\eE[?5h\eE[?5l:ve=\eE[?25h:vi=\eE[?25l:\e	967	\& :us=\eE[4m:vb=\eE[?5h\eE[?5l:ve=\eE[?25h:vi=\eE[?25l:\e
532	\& :vs=\eE[?25h:	968	\& :vs=\eE[?25h:
533	.Ve	969	.Ve
534	.ie n .IP "Why does ""ls"" no longer have coloured output?" 4	970	.PP
535	.el .IP "Why does \f(CWls\fR no longer have coloured output?" 4	971	\fIWhy does \f(CI\*(C`ls\*(C'\fI no longer have coloured output?\fR
536	.IX Item "Why does ls no longer have coloured output?"	972	.IX Subsection "Why does ls no longer have coloured output?"
		973	.PP
537	The \f(CW\*(C`ls\*(C'\fR in the \s-1GNU\s0 coreutils unfortunately doesn't use terminfo to	974	The \f(CW\*(C`ls\*(C'\fR in the \s-1GNU\s0 coreutils unfortunately doesn't use terminfo to
538	decide wether a terminal has colour, but uses it's own configuration	975	decide wether a terminal has colour, but uses it's own configuration
539	file. Needless to say, \f(CW\*(C`rxvt\-unicode\*(C'\fR is not in it's default file (among	976	file. Needless to say, \f(CW\*(C`rxvt\-unicode\*(C'\fR is not in it's default file (among
540	with most other terminals supporting colour). Either add:	977	with most other terminals supporting colour). Either add:
541	.Sp	978	.PP
542	.Vb 1	979	.Vb 1
543	\& TERM rxvt-unicode	980	\& TERM rxvt-unicode
544	.Ve	981	.Ve
545	.Sp	982	.PP
546	to \f(CW\*(C`/etc/DIR_COLORS\*(C'\fR or simply add:	983	to \f(CW\*(C`/etc/DIR_COLORS\*(C'\fR or simply add:
547	.Sp	984	.PP
548	.Vb 1	985	.Vb 1
549	\& alias ls='ls --color=auto'	986	\& alias ls='ls --color=auto'
550	.Ve	987	.Ve
551	.Sp	988	.PP
552	to your \f(CW\*(C`.profile\*(C'\fR or \f(CW\*(C`.bashrc\*(C'\fR.	989	to your \f(CW\*(C`.profile\*(C'\fR or \f(CW\*(C`.bashrc\*(C'\fR.
		990	.PP
553	.IP "Why doesn't vim/emacs etc. use the 88 colour mode?" 4	991	\fIWhy doesn't vim/emacs etc. use the 88 colour mode?\fR
554	.IX Item "Why doesn't vim/emacs etc. use the 88 colour mode?"	992	.IX Subsection "Why doesn't vim/emacs etc. use the 88 colour mode?"
555	.PD 0	993	.PP
		994	See next entry.
		995	.PP
556	.IP "Why doesn't vim/emacs etc. make use of italic?" 4	996	\fIWhy doesn't vim/emacs etc. make use of italic?\fR
557	.IX Item "Why doesn't vim/emacs etc. make use of italic?"	997	.IX Subsection "Why doesn't vim/emacs etc. make use of italic?"
		998	.PP
		999	See next entry.
		1000	.PP
558	.IP "Why are the secondary screen-related options not working properly?" 4	1001	\fIWhy are the secondary screen-related options not working properly?\fR
559	.IX Item "Why are the secondary screen-related options not working properly?"	1002	.IX Subsection "Why are the secondary screen-related options not working properly?"
560	.PD	1003	.PP
561	Make sure you are using \f(CW\*(C`TERM=rxvt\-unicode\*(C'\fR. Some pre-packaged	1004	Make sure you are using \f(CW\*(C`TERM=rxvt\-unicode\*(C'\fR. Some pre-packaged
562	distributions (most notably Debian GNU/Linux) break rxvt-unicode	1005	distributions (most notably Debian GNU/Linux) break rxvt-unicode
563	by setting \f(CW\*(C`TERM\*(C'\fR to \f(CW\*(C`rxvt\*(C'\fR, which doesn't have these extra	1006	by setting \f(CW\*(C`TERM\*(C'\fR to \f(CW\*(C`rxvt\*(C'\fR, which doesn't have these extra
564	features. Unfortunately, some of these (most notably, again, Debian	1007	features. Unfortunately, some of these (most notably, again, Debian
565	GNU/Linux) furthermore fail to even install the \f(CW\*(C`rxvt\-unicode\*(C'\fR terminfo	1008	GNU/Linux) furthermore fail to even install the \f(CW\*(C`rxvt\-unicode\*(C'\fR terminfo
566	file, so you will need to install it on your own (See the question \fBWhen	1009	file, so you will need to install it on your own (See the question \fBWhen
567	I log-in to another system it tells me about missing terminfo data?\fR on	1010	I log-in to another system it tells me about missing terminfo data?\fR on
568	how to do this).	1011	how to do this).
569	.IP "My numerical keypad acts weird and generates differing output?" 4	1012	.Sh "Encoding / Locale / Input Method Issues"
570	.IX Item "My numerical keypad acts weird and generates differing output?"	1013	.IX Subsection "Encoding / Locale / Input Method Issues"
571	Some Debian GNUL/Linux users seem to have this problem, although no
572	specific details were reported so far. It is possible that this is caused
573	by the wrong \f(CW\*(C`TERM\*(C'\fR setting, although the details of wether and how
574	this can happen are unknown, as \f(CW\*(C`TERM=rxvt\*(C'\fR should offer a compatible
575	keymap. See the answer to the previous question, and please report if that
576	helped.
577	.IP "Rxvt-unicode does not seem to understand the selected encoding?" 4	1014	\fIRxvt-unicode does not seem to understand the selected encoding?\fR
578	.IX Item "Rxvt-unicode does not seem to understand the selected encoding?"	1015	.IX Subsection "Rxvt-unicode does not seem to understand the selected encoding?"
579	.PD 0	1016	.PP
		1017	See next entry.
		1018	.PP
580	.IP "Unicode does not seem to work?" 4	1019	\fIUnicode does not seem to work?\fR
581	.IX Item "Unicode does not seem to work?"	1020	.IX Subsection "Unicode does not seem to work?"
582	.PD	1021	.PP
583	If you encounter strange problems like typing an accented character but	1022	If you encounter strange problems like typing an accented character but
584	getting two unrelated other characters or similar, or if program output is	1023	getting two unrelated other characters or similar, or if program output is
585	subtly garbled, then you should check your locale settings.	1024	subtly garbled, then you should check your locale settings.
586	.Sp	1025	.PP
587	Rxvt-unicode must be started with the same \f(CW\*(C`LC_CTYPE\*(C'\fR setting as the	1026	Rxvt-unicode must be started with the same \f(CW\*(C`LC_CTYPE\*(C'\fR setting as the
588	programs. Often rxvt-unicode is started in the \f(CW\*(C`C\*(C'\fR locale, while the	1027	programs. Often rxvt-unicode is started in the \f(CW\*(C`C\*(C'\fR locale, while the
589	login script running within the rxvt-unicode window changes the locale to	1028	login script running within the rxvt-unicode window changes the locale to
590	something else, e.g. \f(CW\*(C`en_GB.UTF\-8\*(C'\fR. Needless to say, this is not going to work.	1029	something else, e.g. \f(CW\*(C`en_GB.UTF\-8\*(C'\fR. Needless to say, this is not going to work.
591	.Sp	1030	.PP
592	The best thing is to fix your startup environment, as you will likely run	1031	The best thing is to fix your startup environment, as you will likely run
593	into other problems. If nothing works you can try this in your .profile.	1032	into other problems. If nothing works you can try this in your .profile.
594	.Sp	1033	.PP
595	.Vb 1	1034	.Vb 1
596	\& printf '\ee]701;%s\e007' "$LC_CTYPE"	1035	\& printf '\ee]701;%s\e007' "$LC_CTYPE"
597	.Ve	1036	.Ve
598	.Sp	1037	.PP
599	If this doesn't work, then maybe you use a \f(CW\*(C`LC_CTYPE\*(C'\fR specification not	1038	If this doesn't work, then maybe you use a \f(CW\*(C`LC_CTYPE\*(C'\fR specification not
600	supported on your systems. Some systems have a \f(CW\*(C`locale\*(C'\fR command which	1039	supported on your systems. Some systems have a \f(CW\*(C`locale\*(C'\fR command which
601	displays this (also, \f(CW\*(C`perl \-e0\*(C'\fR can be used to check locale settings, as	1040	displays this (also, \f(CW\*(C`perl \-e0\*(C'\fR can be used to check locale settings, as
602	it will complain loudly if it cannot set the locale). If it displays something	1041	it will complain loudly if it cannot set the locale). If it displays something
603	like:	1042	like:
604	.Sp	1043	.PP
605	.Vb 1	1044	.Vb 1
606	\& locale: Cannot set LC_CTYPE to default locale: ...	1045	\& locale: Cannot set LC_CTYPE to default locale: ...
607	.Ve	1046	.Ve
608	.Sp	1047	.PP
609	Then the locale you specified is not supported on your system.	1048	Then the locale you specified is not supported on your system.
610	.Sp	1049	.PP
611	If nothing works and you are sure that everything is set correctly then	1050	If nothing works and you are sure that everything is set correctly then
612	you will need to remember a little known fact: Some programs just don't	1051	you will need to remember a little known fact: Some programs just don't
613	support locales :(	1052	support locales :(
614	.IP "Why do some characters look so much different than others?" 4	1053	.PP
615	.IX Item "Why do some characters look so much different than others?"	1054	\fIHow does rxvt-unicode determine the encoding to use?\fR
616	.PD 0	1055	.IX Subsection "How does rxvt-unicode determine the encoding to use?"
617	.IP "How does rxvt-unicode choose fonts?" 4	1056	.PP
618	.IX Item "How does rxvt-unicode choose fonts?"	1057	See next entry.
619	.PD	1058	.PP
620	Most fonts do not contain the full range of Unicode, which is	1059	\fIIs there an option to switch encodings?\fR
621	fine. Chances are that the font you (or the admin/package maintainer of	1060	.IX Subsection "Is there an option to switch encodings?"
622	your system/os) have specified does not cover all the characters you want	1061	.PP
623	to display.	1062	Unlike some other terminals, rxvt-unicode has no encoding switch, and no
624	.Sp	1063	specific \*(L"utf\-8\*(R" mode, such as xterm. In fact, it doesn't even know about
625	\&\fBrxvt-unicode\fR makes a best-effort try at finding a replacement	1064	\&\s-1UTF\-8\s0 or any other encodings with respect to terminal I/O.
626	font. Often the result is fine, but sometimes the chosen font looks	1065	.PP
627	bad/ugly/wrong. Some fonts have totally strange characters that don't	1066	The reasons is that there exists a perfectly fine mechanism for selecting
628	resemble the correct glyph at all, and rxvt-unicode lacks the artificial	1067	the encoding, doing I/O and (most important) communicating this to all
629	intelligence to detect that a specific glyph is wrong: it has to believe	1068	applications so everybody agrees on character properties such as width
630	the font that the characters it claims to contain indeed look correct.	1069	and code number. This mechanism is the \fIlocale\fR. Applications not using
631	.Sp	1070	that info will have problems (for example, \f(CW\*(C`xterm\*(C'\fR gets the width of
632	In that case, select a font of your taste and add it to the font list,	1071	characters wrong as it uses it's own, locale-independent table under all
633	e.g.:	1072	locales).
634	.Sp	1073	.PP
		1074	Rxvt-unicode uses the \f(CW\*(C`LC_CTYPE\*(C'\fR locale category to select encoding. All
		1075	programs doing the same (that is, most) will automatically agree in the
		1076	interpretation of characters.
		1077	.PP
		1078	Unfortunately, there is no system-independent way to select locales, nor
		1079	is there a standard on how locale specifiers will look like.
		1080	.PP
		1081	On most systems, the content of the \f(CW\*(C`LC_CTYPE\*(C'\fR environment variable
		1082	contains an arbitrary string which corresponds to an already-installed
		1083	locale. Common names for locales are \f(CW\*(C`en_US.UTF\-8\*(C'\fR, \f(CW\*(C`de_DE.ISO\-8859\-15\*(C'\fR,
		1084	\&\f(CW\*(C`ja_JP.EUC\-JP\*(C'\fR, i.e. \f(CW\*(C`language_country.encoding\*(C'\fR, but other forms
		1085	(i.e. \f(CW\*(C`de\*(C'\fR or \f(CW\*(C`german\*(C'\fR) are also common.
		1086	.PP
		1087	Rxvt-unicode ignores all other locale categories, and except for
		1088	the encoding, ignores country or language-specific settings,
		1089	i.e. \f(CW\*(C`de_DE.UTF\-8\*(C'\fR and \f(CW\*(C`ja_JP.UTF\-8\*(C'\fR are the normally same to
		1090	rxvt\-unicode.
		1091	.PP
		1092	If you want to use a specific encoding you have to make sure you start
		1093	rxvt-unicode with the correct \f(CW\*(C`LC_CTYPE\*(C'\fR category.
		1094	.PP
		1095	\fICan I switch locales at runtime?\fR
		1096	.IX Subsection "Can I switch locales at runtime?"
		1097	.PP
		1098	Yes, using an escape sequence. Try something like this, which sets
		1099	rxvt\-unicode's idea of \f(CW\*(C`LC_CTYPE\*(C'\fR.
		1100	.PP
635	.Vb 1	1101	.Vb 1
636	\& @@RXVT_NAME@@ -fn basefont,font2,font3...	1102	\& printf '\ee]701;%s\e007' ja_JP.SJIS
637	.Ve	1103	.Ve
638	.Sp	1104	.PP
639	When rxvt-unicode sees a character, it will first look at the base	1105	See also the previous answer.
640	font. If the base font does not contain the character, it will go to the	1106	.PP
641	next font, and so on. Specifying your own fonts will also speed up this	1107	Sometimes this capability is rather handy when you want to work in
642	search and use less resources within rxvt-unicode and the X\-server.	1108	one locale (e.g. \f(CW\*(C`de_DE.UTF\-8\*(C'\fR) but some programs don't support it
643	.Sp	1109	(e.g. \s-1UTF\-8\s0). For example, I use this script to start \f(CW\*(C`xjdic\*(C'\fR, which
644	The only limitation is that none of the fonts may be larger than the base	1110	first switches to a locale supported by xjdic and back later:
645	font, as the base font defines the terminal character cell size, which	1111	.PP
646	must be the same due to the way terminals work.	1112	.Vb 3
647	.IP "Why do some chinese characters look so different than others?" 4	1113	\& printf '\ee]701;%s\e007' ja_JP.SJIS
648	.IX Item "Why do some chinese characters look so different than others?"	1114	\& xjdic -js
649	This is because there is a difference between script and language \*(--	1115	\& printf '\ee]701;%s\e007' de_DE.UTF-8
650	rxvt-unicode does not know which language the text that is output is,	1116	.Ve
651	as it only knows the unicode character codes. If rxvt-unicode first	1117	.PP
652	sees a japanese/chinese character, it might choose a japanese font for	1118	You can also use xterm's \f(CW\*(C`luit\*(C'\fR program, which usually works fine, except
653	display. Subsequent japanese characters will use that font. Now, many	1119	for some locales where character width differs between program\- and
654	chinese characters aren't represented in japanese fonts, so when the first	1120	rxvt\-unicode\-locales.
655	non-japanese character comes up, rxvt-unicode will look for a chinese font	1121	.PP
656	\&\*(-- unfortunately at this point, it will still use the japanese font for	1122	\fIMy input method wants <some encoding> but I want \s-1UTF\-8\s0, what can I do?\fR
657	chinese characters that are also in the japanese font.	1123	.IX Subsection "My input method wants <some encoding> but I want UTF-8, what can I do?"
658	.Sp	1124	.PP
659	The workaround is easy: just tag a chinese font at the end of your font	1125	You can specify separate locales for the input method and the rest of the
660	list (see the previous question). The key is to view the font list as	1126	terminal, using the resource \f(CW\*(C`imlocale\*(C'\fR:
661	a preference list: If you expect more japanese, list a japanese font	1127	.PP
662	first. If you expect more chinese, put a chinese font first.	1128	.Vb 1
663	.Sp	1129	\& URxvt.imlocale: ja_JP.EUC-JP
664	In the future it might be possible to switch language preferences at	1130	.Ve
665	runtime (the internal data structure has no problem with using different	1131	.PP
666	fonts for the same character at the same time, but no interface for this	1132	Now you can start your terminal with \f(CW\*(C`LC_CTYPE=ja_JP.UTF\-8\*(C'\fR and still
667	has been designed yet).	1133	use your input method. Please note, however, that you will not be able to
668	.Sp	1134	input characters outside \f(CW\*(C`EUC\-JP\*(C'\fR in a normal way then, as your input
669	Until then, you might get away with switching fonts at runtime (see \*(L"Can I switch the fonts at runtime?\*(R" later in this document).	1135	method limits you.
670	.IP "Why does rxvt-unicode sometimes leave pixel droppings?" 4	1136	.PP
671	.IX Item "Why does rxvt-unicode sometimes leave pixel droppings?"	1137	\fIRxvt-unicode crashes when the X Input Method changes or exits.\fR
672	Most fonts were not designed for terminal use, which means that character	1138	.IX Subsection "Rxvt-unicode crashes when the X Input Method changes or exits."
673	size varies a lot. A font that is otherwise fine for terminal use might	1139	.PP
674	contain some characters that are simply too wide. Rxvt-unicode will avoid	1140	Unfortunately, this is unavoidable, as the \s-1XIM\s0 protocol is racy by
675	these characters. For characters that are just \*(L"a bit\*(R" too wide a special	1141	design. Applications can avoid some crashes at the expense of memory
676	\&\*(L"careful\*(R" rendering mode is used that redraws adjacent characters.	1142	leaks, and Input Methods can avoid some crashes by careful ordering at
677	.Sp	1143	exit time. \fBkinput2\fR (and derived input methods) generally succeeds,
678	All of this requires that fonts do not lie about character sizes,	1144	while \fB\s-1SCIM\s0\fR (or similar input methods) fails. In the end, however,
679	however: Xft fonts often draw glyphs larger than their acclaimed bounding	1145	crashes cannot be completely avoided even if both sides cooperate.
680	box, and rxvt-unicode has no way of detecting this (the correct way is to	1146	.PP
681	ask for the character bounding box, which unfortunately is wrong in these	1147	So the only workaround is not to kill your Input Method Servers.
682	cases).	1148	.Sh "Operating Systems / Package Maintaining"
683	.Sp	1149	.IX Subsection "Operating Systems / Package Maintaining"
684	It's not clear (to me at least), wether this is a bug in Xft, freetype,	1150	\fII am using Debian GNU/Linux and have a problem...\fR
685	or the respective font. If you encounter this problem you might try using	1151	.IX Subsection "I am using Debian GNU/Linux and have a problem..."
686	the \f(CW\*(C`\-lsp\*(C'\fR option to give the font more height. If that doesn't work, you	1152	.PP
687	might be forced to use a different font.	1153	The Debian GNU/Linux package of rxvt-unicode in sarge contains large
688	.Sp	1154	patches that considerably change the behaviour of rxvt-unicode (but
689	All of this is not a problem when using X11 core fonts, as their bounding	1155	unfortunately this notice has been removed). Before reporting a bug to
690	box data is correct.	1156	the original rxvt-unicode author please download and install the genuine
		1157	version (<http://software.schmorp.de#rxvt\-unicode>) and try to reproduce
		1158	the problem. If you cannot, chances are that the problems are specific to
		1159	Debian GNU/Linux, in which case it should be reported via the Debian Bug
		1160	Tracking System (use \f(CW\*(C`reportbug\*(C'\fR to report the bug).
		1161	.PP
		1162	For other problems that also affect the Debian package, you can and
		1163	probably should use the Debian \s-1BTS\s0, too, because, after all, it's also a
		1164	bug in the Debian version and it serves as a reminder for other users that
		1165	might encounter the same issue.
		1166	.PP
		1167	\fII am maintaining rxvt-unicode for distribution/OS \s-1XXX\s0, any recommendation?\fR
		1168	.IX Subsection "I am maintaining rxvt-unicode for distribution/OS XXX, any recommendation?"
		1169	.PP
		1170	You should build one binary with the default options. \fIconfigure\fR
		1171	now enables most useful options, and the trend goes to making them
		1172	runtime\-switchable, too, so there is usually no drawback to enbaling them,
		1173	except higher disk and possibly memory usage. The perl interpreter should
		1174	be enabled, as important functionality (menus, selection, likely more in
		1175	the future) depends on it.
		1176	.PP
		1177	You should not overwrite the \f(CW\*(C`perl\-ext\-common\*(C'\fR snd \f(CW\*(C`perl\-ext\*(C'\fR resources
		1178	system-wide (except maybe with \f(CW\*(C`defaults\*(C'\fR). This will result in useful
		1179	behaviour. If your distribution aims at low memory, add an empty
		1180	\&\f(CW\*(C`perl\-ext\-common\*(C'\fR resource to the app-defaults file. This will keep the
		1181	perl interpreter disabled until the user enables it.
		1182	.PP
		1183	If you can/want build more binaries, I recommend building a minimal
		1184	one with \f(CW\*(C`\-\-disable\-everything\*(C'\fR (very useful) and a maximal one with
		1185	\&\f(CW\*(C`\-\-enable\-everything\*(C'\fR (less useful, it will be very big due to a lot of
		1186	encodings built-in that increase download times and are rarely used).
		1187	.PP
		1188	\fII need to make it setuid/setgid to support utmp/ptys on my \s-1OS\s0, is this safe?\fR
		1189	.IX Subsection "I need to make it setuid/setgid to support utmp/ptys on my OS, is this safe?"
		1190	.PP
		1191	It should be, starting with release 7.1. You are encouraged to properly
		1192	install urxvt with privileges necessary for your \s-1OS\s0 now.
		1193	.PP
		1194	When rxvt-unicode detects that it runs setuid or setgid, it will fork
		1195	into a helper process for privileged operations (pty handling on some
		1196	systems, utmp/wtmp/lastlog handling on others) and drop privileges
		1197	immediately. This is much safer than most other terminals that keep
		1198	privileges while running (but is more relevant to urxvt, as it contains
		1199	things as perl interpreters, which might be \*(L"helpful\*(R" to attackers).
		1200	.PP
		1201	This forking is done as the very first within \fImain()\fR, which is very early
		1202	and reduces possible bugs to initialisation code run before \fImain()\fR, or
		1203	things like the dynamic loader of your system, which should result in very
		1204	little risk.
		1205	.PP
691	.IP "On Solaris 9, many line-drawing characters are too wide." 4	1206	\fIOn Solaris 9, many line-drawing characters are too wide.\fR
692	.IX Item "On Solaris 9, many line-drawing characters are too wide."	1207	.IX Subsection "On Solaris 9, many line-drawing characters are too wide."
		1208	.PP
693	Seems to be a known bug, read	1209	Seems to be a known bug, read
694	<http://nixdoc.net/files/forum/about34198.html>. Some people use the	1210	<http://nixdoc.net/files/forum/about34198.html>. Some people use the
695	following ugly workaround to get non-double-wide-characters working:	1211	following ugly workaround to get non-double-wide-characters working:
696	.Sp	1212	.PP
697	.Vb 1	1213	.Vb 1
698	\& #define wcwidth(x) wcwidth(x) > 1 ? 1 : wcwidth(x)	1214	\& #define wcwidth(x) wcwidth(x) > 1 ? 1 : wcwidth(x)
699	.Ve	1215	.Ve
700	.IP "My Compose (Multi_key) key is no longer working." 4	1216	.PP
701	.IX Item "My Compose (Multi_key) key is no longer working."
702	The most common causes for this are that either your locale is not set
703	correctly, or you specified a \fBpreeditStyle\fR that is not supported by
704	your input method. For example, if you specified \fBOverTheSpot\fR and
705	your input method (e.g. the default input method handling Compose keys)
706	does not support this (for instance because it is not visual), then
707	rxvt-unicode will continue without an input method.
708	.Sp
709	In this case either do not specify a \fBpreeditStyle\fR or specify more than
710	one pre-edit style, such as \fBOverTheSpot,Root,None\fR.
711	.ie n .IP "I cannot type ""Ctrl\-Shift\-2"" to get an \s-1ASCII\s0 \s-1NUL\s0 character due to \s-1ISO\s0 14755" 4
712	.el .IP "I cannot type \f(CWCtrl\-Shift\-2\fR to get an \s-1ASCII\s0 \s-1NUL\s0 character due to \s-1ISO\s0 14755" 4
713	.IX Item "I cannot type Ctrl-Shift-2 to get an ASCII NUL character due to ISO 14755"
714	Either try \f(CW\*(C`Ctrl\-2\*(C'\fR alone (it often is mapped to \s-1ASCII\s0 \s-1NUL\s0 even on
715	international keyboards) or simply use \s-1ISO\s0 14755 support to your
716	advantage, typing <Ctrl\-Shift\-0> to get a \s-1ASCII\s0 \s-1NUL\s0. This works for other
717	codes, too, such as \f(CW\*(C`Ctrl\-Shift\-1\-d\*(C'\fR to type the default telnet escape
718	character and so on.
719	.IP "How can I keep rxvt-unicode from using reverse video so much?" 4
720	.IX Item "How can I keep rxvt-unicode from using reverse video so much?"
721	First of all, make sure you are running with the right terminal settings
722	(\f(CW\*(C`TERM=rxvt\-unicode\*(C'\fR), which will get rid of most of these effects. Then
723	make sure you have specified colours for italic and bold, as otherwise
724	rxvt-unicode might use reverse video to simulate the effect:
725	.Sp
726	.Vb 2
727	\& URxvt.colorBD: white
728	\& URxvt.colorIT: green
729	.Ve
730	.IP "Some programs assume totally weird colours (red instead of blue), how can I fix that?" 4
731	.IX Item "Some programs assume totally weird colours (red instead of blue), how can I fix that?"
732	For some unexplainable reason, some rare programs assume a very weird
733	colour palette when confronted with a terminal with more than the standard
734	8 colours (rxvt\-unicode supports 88). The right fix is, of course, to fix
735	these programs not to assume non-ISO colours without very good reasons.
736	.Sp
737	In the meantime, you can either edit your \f(CW\*(C`rxvt\-unicode\*(C'\fR terminfo
738	definition to only claim 8 colour support or use \f(CW\*(C`TERM=rxvt\*(C'\fR, which will
739	fix colours but keep you from using other rxvt-unicode features.
740	.IP "I am on FreeBSD and rxvt-unicode does not seem to work at all." 4	1217	\fII am on FreeBSD and rxvt-unicode does not seem to work at all.\fR
741	.IX Item "I am on FreeBSD and rxvt-unicode does not seem to work at all."	1218	.IX Subsection "I am on FreeBSD and rxvt-unicode does not seem to work at all."
		1219	.PP
742	Rxvt-unicode requires the symbol \f(CW\*(C`_\|_STDC_ISO_10646_\|_\*(C'\fR to be defined	1220	Rxvt-unicode requires the symbol \f(CW\*(C`_\|_STDC_ISO_10646_\|_\*(C'\fR to be defined
743	in your compile environment, or an implementation that implements it,	1221	in your compile environment, or an implementation that implements it,
744	wether it defines the symbol or not. \f(CW\*(C`_\|_STDC_ISO_10646_\|_\*(C'\fR requires that	1222	wether it defines the symbol or not. \f(CW\*(C`_\|_STDC_ISO_10646_\|_\*(C'\fR requires that
745	\&\fBwchar_t\fR is represented as unicode.	1223	\&\fBwchar_t\fR is represented as unicode.
746	.Sp	1224	.PP
747	As you might have guessed, FreeBSD does neither define this symobl nor	1225	As you might have guessed, FreeBSD does neither define this symobl nor
748	does it support it. Instead, it uses it's own internal representation of	1226	does it support it. Instead, it uses it's own internal representation of
749	\&\fBwchar_t\fR. This is, of course, completely fine with respect to standards.	1227	\&\fBwchar_t\fR. This is, of course, completely fine with respect to standards.
750	.Sp	1228	.PP
751	However, that means rxvt-unicode only works in \f(CW\*(C`POSIX\*(C'\fR, \f(CW\*(C`ISO\-8859\-1\*(C'\fR and	1229	However, that means rxvt-unicode only works in \f(CW\*(C`POSIX\*(C'\fR, \f(CW\*(C`ISO\-8859\-1\*(C'\fR and
752	\&\f(CW\*(C`UTF\-8\*(C'\fR locales under FreeBSD (which all use Unicode as \fBwchar_t\fR.	1230	\&\f(CW\*(C`UTF\-8\*(C'\fR locales under FreeBSD (which all use Unicode as \fBwchar_t\fR.
753	.Sp	1231	.PP
754	\&\f(CW\*(C`_\|_STDC_ISO_10646_\|_\*(C'\fR is the only sane way to support multi-language	1232	\&\f(CW\*(C`_\|_STDC_ISO_10646_\|_\*(C'\fR is the only sane way to support multi-language
755	apps in an \s-1OS\s0, as using a locale-dependent (and non\-standardized)	1233	apps in an \s-1OS\s0, as using a locale-dependent (and non\-standardized)
756	representation of \fBwchar_t\fR makes it impossible to convert between	1234	representation of \fBwchar_t\fR makes it impossible to convert between
757	\&\fBwchar_t\fR (as used by X11 and your applications) and any other encoding	1235	\&\fBwchar_t\fR (as used by X11 and your applications) and any other encoding
758	without implementing OS-specific-wrappers for each and every locale. There	1236	without implementing OS-specific-wrappers for each and every locale. There
759	simply are no APIs to convert \fBwchar_t\fR into anything except the current	1237	simply are no APIs to convert \fBwchar_t\fR into anything except the current
760	locale encoding.	1238	locale encoding.
761	.Sp	1239	.PP
762	Some applications (such as the formidable \fBmlterm\fR) work around this	1240	Some applications (such as the formidable \fBmlterm\fR) work around this
763	by carrying their own replacement functions for character set handling	1241	by carrying their own replacement functions for character set handling
764	with them, and either implementing OS-dependent hacks or doing multiple	1242	with them, and either implementing OS-dependent hacks or doing multiple
765	conversions (which is slow and unreliable in case the \s-1OS\s0 implements	1243	conversions (which is slow and unreliable in case the \s-1OS\s0 implements
766	encodings slightly different than the terminal emulator).	1244	encodings slightly different than the terminal emulator).
767	.Sp	1245	.PP
768	The rxvt-unicode author insists that the right way to fix this is in the	1246	The rxvt-unicode author insists that the right way to fix this is in the
769	system libraries once and for all, instead of forcing every app to carry	1247	system libraries once and for all, instead of forcing every app to carry
770	complete replacements for them :)	1248	complete replacements for them :)
		1249	.PP
771	.IP "I use Solaris 9 and it doesn't compile/work/etc." 4	1250	\fII use Solaris 9 and it doesn't compile/work/etc.\fR
772	.IX Item "I use Solaris 9 and it doesn't compile/work/etc."	1251	.IX Subsection "I use Solaris 9 and it doesn't compile/work/etc."
		1252	.PP
773	Try the diff in \fIdoc/solaris9.patch\fR as a base. It fixes the worst	1253	Try the diff in \fIdoc/solaris9.patch\fR as a base. It fixes the worst
774	problems with \f(CW\*(C`wcwidth\*(C'\fR and a compile problem.	1254	problems with \f(CW\*(C`wcwidth\*(C'\fR and a compile problem.
		1255	.PP
775	.IP "How can I use rxvt-unicode under cygwin?" 4	1256	\fIHow can I use rxvt-unicode under cygwin?\fR
776	.IX Item "How can I use rxvt-unicode under cygwin?"	1257	.IX Subsection "How can I use rxvt-unicode under cygwin?"
		1258	.PP
777	rxvt-unicode should compile and run out of the box on cygwin, using	1259	rxvt-unicode should compile and run out of the box on cygwin, using
778	the X11 libraries that come with cygwin. libW11 emulation is no	1260	the X11 libraries that come with cygwin. libW11 emulation is no
779	longer supported (and makes no sense, either, as it only supported a	1261	longer supported (and makes no sense, either, as it only supported a
780	single font). I recommend starting the X\-server in \f(CW\*(C`\-multiwindow\*(C'\fR or	1262	single font). I recommend starting the X\-server in \f(CW\*(C`\-multiwindow\*(C'\fR or
781	\&\f(CW\*(C`\-rootless\*(C'\fR mode instead, which will result in similar look&feel as the	1263	\&\f(CW\*(C`\-rootless\*(C'\fR mode instead, which will result in similar look&feel as the
782	old libW11 emulation.	1264	old libW11 emulation.
783	.Sp	1265	.PP
784	At the time of this writing, cygwin didn't seem to support any multi-byte	1266	At the time of this writing, cygwin didn't seem to support any multi-byte
785	encodings (you might try \f(CW\*(C`LC_CTYPE=C\-UTF\-8\*(C'\fR), so you are likely limited	1267	encodings (you might try \f(CW\*(C`LC_CTYPE=C\-UTF\-8\*(C'\fR), so you are likely limited
786	to 8\-bit encodings.	1268	to 8\-bit encodings.
787	.IP "How does rxvt-unicode determine the encoding to use?" 4
788	.IX Item "How does rxvt-unicode determine the encoding to use?"
789	.PD 0
790	.IP "Is there an option to switch encodings?" 4
791	.IX Item "Is there an option to switch encodings?"
792	.PD
793	Unlike some other terminals, rxvt-unicode has no encoding switch, and no
794	specific \*(L"utf\-8\*(R" mode, such as xterm. In fact, it doesn't even know about
795	\&\s-1UTF\-8\s0 or any other encodings with respect to terminal I/O.
796	.Sp
797	The reasons is that there exists a perfectly fine mechanism for selecting
798	the encoding, doing I/O and (most important) communicating this to all
799	applications so everybody agrees on character properties such as width
800	and code number. This mechanism is the \fIlocale\fR. Applications not using
801	that info will have problems (for example, \f(CW\*(C`xterm\*(C'\fR gets the width of
802	characters wrong as it uses it's own, locale-independent table under all
803	locales).
804	.Sp
805	Rxvt-unicode uses the \f(CW\*(C`LC_CTYPE\*(C'\fR locale category to select encoding. All
806	programs doing the same (that is, most) will automatically agree in the
807	interpretation of characters.
808	.Sp
809	Unfortunately, there is no system-independent way to select locales, nor
810	is there a standard on how locale specifiers will look like.
811	.Sp
812	On most systems, the content of the \f(CW\*(C`LC_CTYPE\*(C'\fR environment variable
813	contains an arbitrary string which corresponds to an already-installed
814	locale. Common names for locales are \f(CW\*(C`en_US.UTF\-8\*(C'\fR, \f(CW\*(C`de_DE.ISO\-8859\-15\*(C'\fR,
815	\&\f(CW\*(C`ja_JP.EUC\-JP\*(C'\fR, i.e. \f(CW\*(C`language_country.encoding\*(C'\fR, but other forms
816	(i.e. \f(CW\*(C`de\*(C'\fR or \f(CW\*(C`german\*(C'\fR) are also common.
817	.Sp
818	Rxvt-unicode ignores all other locale categories, and except for
819	the encoding, ignores country or language-specific settings,
820	i.e. \f(CW\*(C`de_DE.UTF\-8\*(C'\fR and \f(CW\*(C`ja_JP.UTF\-8\*(C'\fR are the normally same to
821	rxvt\-unicode.
822	.Sp
823	If you want to use a specific encoding you have to make sure you start
824	rxvt-unicode with the correct \f(CW\*(C`LC_CTYPE\*(C'\fR category.
825	.IP "Can I switch locales at runtime?" 4
826	.IX Item "Can I switch locales at runtime?"
827	Yes, using an escape sequence. Try something like this, which sets
828	rxvt\-unicode's idea of \f(CW\*(C`LC_CTYPE\*(C'\fR.
829	.Sp
830	.Vb 1
831	\& printf '\ee]701;%s\e007' ja_JP.SJIS
832	.Ve
833	.Sp
834	See also the previous answer.
835	.Sp
836	Sometimes this capability is rather handy when you want to work in
837	one locale (e.g. \f(CW\*(C`de_DE.UTF\-8\*(C'\fR) but some programs don't support it
838	(e.g. \s-1UTF\-8\s0). For example, I use this script to start \f(CW\*(C`xjdic\*(C'\fR, which
839	first switches to a locale supported by xjdic and back later:
840	.Sp
841	.Vb 3
842	\& printf '\ee]701;%s\e007' ja_JP.SJIS
843	\& xjdic -js
844	\& printf '\ee]701;%s\e007' de_DE.UTF-8
845	.Ve
846	.Sp
847	You can also use xterm's \f(CW\*(C`luit\*(C'\fR program, which usually works fine, except
848	for some locales where character width differs between program\- and
849	rxvt\-unicode\-locales.
850	.IP "Can I switch the fonts at runtime?" 4
851	.IX Item "Can I switch the fonts at runtime?"
852	Yes, using an escape sequence. Try something like this, which has the same
853	effect as using the \f(CW\*(C`\-fn\*(C'\fR switch, and takes effect immediately:
854	.Sp
855	.Vb 1
856	\& printf '\ee]50;%s\e007' "9x15bold,xft:Kochi Gothic"
857	.Ve
858	.Sp
859	This is useful if you e.g. work primarily with japanese (and prefer a
860	japanese font), but you have to switch to chinese temporarily, where
861	japanese fonts would only be in your way.
862	.Sp
863	You can think of this as a kind of manual \s-1ISO\-2022\s0 switching.
864	.IP "Why do italic characters look as if clipped?" 4
865	.IX Item "Why do italic characters look as if clipped?"
866	Many fonts have difficulties with italic characters and hinting. For
867	example, the otherwise very nicely hinted font \f(CW\*(C`xft:Bitstream Vera Sans
868	Mono\*(C'\fR completely fails in it's italic face. A workaround might be to
869	enable freetype autohinting, i.e. like this:
870	.Sp
871	.Vb 2
872	\& URxvt.italicFont: xft:Bitstream Vera Sans Mono:italic:autohint=true
873	\& URxvt.boldItalicFont: xft:Bitstream Vera Sans Mono:bold:italic:autohint=true
874	.Ve
875	.IP "My input method wants <some encoding> but I want \s-1UTF\-8\s0, what can I do?" 4
876	.IX Item "My input method wants <some encoding> but I want UTF-8, what can I do?"
877	You can specify separate locales for the input method and the rest of the
878	terminal, using the resource \f(CW\*(C`imlocale\*(C'\fR:
879	.Sp
880	.Vb 1
881	\& URxvt.imlocale: ja_JP.EUC-JP
882	.Ve
883	.Sp
884	Now you can start your terminal with \f(CW\*(C`LC_CTYPE=ja_JP.UTF\-8\*(C'\fR and still
885	use your input method. Please note, however, that you will not be able to
886	input characters outside \f(CW\*(C`EUC\-JP\*(C'\fR in a normal way then, as your input
887	method limits you.
888	.IP "Rxvt-unicode crashes when the X Input Method changes or exits." 4
889	.IX Item "Rxvt-unicode crashes when the X Input Method changes or exits."
890	Unfortunately, this is unavoidable, as the \s-1XIM\s0 protocol is racy by
891	design. Applications can avoid some crashes at the expense of memory
892	leaks, and Input Methods can avoid some crashes by careful ordering at
893	exit time. \fBkinput2\fR (and derived input methods) generally succeeds,
894	while \fB\s-1SCIM\s0\fR (or similar input methods) fails. In the end, however,
895	crashes cannot be completely avoided even if both sides cooperate.
896	.Sp
897	So the only workaround is not to kill your Input Method Servers.
898	.IP "Rxvt-unicode uses gobs of memory, how can I reduce that?" 4
899	.IX Item "Rxvt-unicode uses gobs of memory, how can I reduce that?"
900	Rxvt-unicode tries to obey the rule of not charging you for something you
901	don't use. One thing you should try is to configure out all settings that
902	you don't need, for example, Xft support is a resource hog by design,
903	when used. Compiling it out ensures that no Xft font will be loaded
904	accidentally when rxvt-unicode tries to find a font for your characters.
905	.Sp
906	Also, many people (me included) like large windows and even larger
907	scrollback buffers: Without \f(CW\*(C`\-\-enable\-unicode3\*(C'\fR, rxvt-unicode will use
908	6 bytes per screen cell. For a 160x?? window this amounts to almost a
909	kilobyte per line. A scrollback buffer of 10000 lines will then (if full)
910	use 10 Megabytes of memory. With \f(CW\*(C`\-\-enable\-unicode3\*(C'\fR it gets worse, as
911	rxvt-unicode then uses 8 bytes per screen cell.
912	.IP "Can I speed up Xft rendering somehow?" 4
913	.IX Item "Can I speed up Xft rendering somehow?"
914	Yes, the most obvious way to speed it up is to avoid Xft entirely, as
915	it is simply slow. If you still want Xft fonts you might try to disable
916	antialiasing (by appending \f(CW\*(C`:antialias=false\*(C'\fR), which saves lots of
917	memory and also speeds up rendering considerably.
918	.IP "Rxvt-unicode doesn't seem to anti-alias its fonts, what is wrong?" 4
919	.IX Item "Rxvt-unicode doesn't seem to anti-alias its fonts, what is wrong?"
920	Rxvt-unicode will use whatever you specify as a font. If it needs to
921	fall back to it's default font search list it will prefer X11 core
922	fonts, because they are small and fast, and then use Xft fonts. It has
923	antialiasing disabled for most of them, because the author thinks they
924	look best that way.
925	.Sp
926	If you want antialiasing, you have to specify the fonts manually.
927	.IP "Mouse cut/paste suddenly no longer works." 4
928	.IX Item "Mouse cut/paste suddenly no longer works."
929	Make sure that mouse reporting is actually turned off since killing
930	some editors prematurely may leave the mouse in mouse report mode. I've
931	heard that tcsh may use mouse reporting unless it otherwise specified. A
932	quick check is to see if cut/paste works when the Alt or Shift keys are
933	depressed.
934	.IP "What's with this bold/blink stuff?" 4
935	.IX Item "What's with this bold/blink stuff?"
936	If no bold colour is set via \f(CW\*(C`colorBD:\*(C'\fR, bold will invert text using the
937	standard foreground colour.
938	.Sp
939	For the standard background colour, blinking will actually make the
940	text blink when compiled with \f(CW\*(C`\-\-enable\-blinking\*(C'\fR. with standard
941	colours. Without \f(CW\*(C`\-\-enable\-blinking\*(C'\fR, the blink attribute will be
942	ignored.
943	.Sp
944	On \s-1ANSI\s0 colours, bold/blink attributes are used to set high-intensity
945	foreground/background colors.
946	.Sp
947	color0\-7 are the low-intensity colors.
948	.Sp
949	color8\-15 are the corresponding high-intensity colors.
950	.IP "I don't like the screen colors. How do I change them?" 4
951	.IX Item "I don't like the screen colors. How do I change them?"
952	You can change the screen colors at run-time using \fI~/.Xdefaults\fR
953	resources (or as long\-options).
954	.Sp
955	Here are values that are supposed to resemble a \s-1VGA\s0 screen,
956	including the murky brown that passes for low-intensity yellow:
957	.Sp
958	.Vb 8
959	\& URxvt.color0: #000000
960	\& URxvt.color1: #A80000
961	\& URxvt.color2: #00A800
962	\& URxvt.color3: #A8A800
963	\& URxvt.color4: #0000A8
964	\& URxvt.color5: #A800A8
965	\& URxvt.color6: #00A8A8
966	\& URxvt.color7: #A8A8A8
967	.Ve
968	.Sp
969	.Vb 8
970	\& URxvt.color8: #000054
971	\& URxvt.color9: #FF0054
972	\& URxvt.color10: #00FF54
973	\& URxvt.color11: #FFFF54
974	\& URxvt.color12: #0000FF
975	\& URxvt.color13: #FF00FF
976	\& URxvt.color14: #00FFFF
977	\& URxvt.color15: #FFFFFF
978	.Ve
979	.Sp
980	And here is a more complete set of non-standard colors described (not by
981	me) as \*(L"pretty girly\*(R".
982	.Sp
983	.Vb 18
984	\& URxvt.cursorColor: #dc74d1
985	\& URxvt.pointerColor: #dc74d1
986	\& URxvt.background: #0e0e0e
987	\& URxvt.foreground: #4ad5e1
988	\& URxvt.color0: #000000
989	\& URxvt.color8: #8b8f93
990	\& URxvt.color1: #dc74d1
991	\& URxvt.color9: #dc74d1
992	\& URxvt.color2: #0eb8c7
993	\& URxvt.color10: #0eb8c7
994	\& URxvt.color3: #dfe37e
995	\& URxvt.color11: #dfe37e
996	\& URxvt.color5: #9e88f0
997	\& URxvt.color13: #9e88f0
998	\& URxvt.color6: #73f7ff
999	\& URxvt.color14: #73f7ff
1000	\& URxvt.color7: #e1dddd
1001	\& URxvt.color15: #e1dddd
1002	.Ve
1003	.IP "How can I start @@RXVT_NAME@@d in a race-free way?" 4
1004	.IX Item "How can I start @@RXVT_NAME@@d in a race-free way?"
1005	Try \f(CW\*(C`@@RXVT_NAME@@d \-f \-o\*(C'\fR, which tells @@RXVT_NAME@@d to open the
1006	display, create the listening socket and then fork.
1007	.IP "What's with the strange Backspace/Delete key behaviour?" 4
1008	.IX Item "What's with the strange Backspace/Delete key behaviour?"
1009	Assuming that the physical Backspace key corresponds to the
1010	BackSpace keysym (not likely for Linux ... see the following
1011	question) there are two standard values that can be used for
1012	Backspace: \f(CW\*(C`^H\*(C'\fR and \f(CW\*(C`^?\*(C'\fR.
1013	.Sp
1014	Historically, either value is correct, but rxvt-unicode adopts the debian
1015	policy of using \f(CW\*(C`^?\*(C'\fR when unsure, because it's the one only only correct
1016	choice :).
1017	.Sp
1018	Rxvt-unicode tries to inherit the current stty settings and uses the value
1019	of `erase' to guess the value for backspace. If rxvt-unicode wasn't
1020	started from a terminal (say, from a menu or by remote shell), then the
1021	system value of `erase', which corresponds to \s-1CERASE\s0 in <termios.h>, will
1022	be used (which may not be the same as your stty setting).
1023	.Sp
1024	For starting a new rxvt\-unicode:
1025	.Sp
1026	.Vb 3
1027	\& # use Backspace = ^H
1028	\& $ stty erase ^H
1029	\& $ @@RXVT_NAME@@
1030	.Ve
1031	.Sp
1032	.Vb 3
1033	\& # use Backspace = ^?
1034	\& $ stty erase ^?
1035	\& $ @@RXVT_NAME@@
1036	.Ve
1037	.Sp
1038	Toggle with \f(CW\*(C`ESC [ 36 h\*(C'\fR / \f(CW\*(C`ESC [ 36 l\*(C'\fR.
1039	.Sp
1040	For an existing rxvt\-unicode:
1041	.Sp
1042	.Vb 3
1043	\& # use Backspace = ^H
1044	\& $ stty erase ^H
1045	\& $ echo -n "^[[36h"
1046	.Ve
1047	.Sp
1048	.Vb 3
1049	\& # use Backspace = ^?
1050	\& $ stty erase ^?
1051	\& $ echo -n "^[[36l"
1052	.Ve
1053	.Sp
1054	This helps satisfy some of the Backspace discrepancies that occur, but
1055	if you use Backspace = \f(CW\*(C`^H\*(C'\fR, make sure that the termcap/terminfo value
1056	properly reflects that.
1057	.Sp
1058	The Delete key is a another casualty of the ill-defined Backspace problem.
1059	To avoid confusion between the Backspace and Delete keys, the Delete
1060	key has been assigned an escape sequence to match the vt100 for Execute
1061	(\f(CW\*(C`ESC [ 3 ~\*(C'\fR) and is in the supplied termcap/terminfo.
1062	.Sp
1063	Some other Backspace problems:
1064	.Sp
1065	some editors use termcap/terminfo,
1066	some editors (vim I'm told) expect Backspace = ^H,
1067	\&\s-1GNU\s0 Emacs (and Emacs-like editors) use ^H for help.
1068	.Sp
1069	Perhaps someday this will all be resolved in a consistent manner.
1070	.IP "I don't like the key\-bindings. How do I change them?" 4
1071	.IX Item "I don't like the key-bindings. How do I change them?"
1072	There are some compile-time selections available via configure. Unless
1073	you have run \*(L"configure\*(R" with the \f(CW\*(C`\-\-disable\-resources\*(C'\fR option you can
1074	use the `keysym' resource to alter the keystrings associated with keysyms.
1075	.Sp
1076	Here's an example for a URxvt session started using \f(CW\*(C`@@RXVT_NAME@@ \-name URxvt\*(C'\fR
1077	.Sp
1078	.Vb 20
1079	\& URxvt.keysym.Home: \e033[1~
1080	\& URxvt.keysym.End: \e033[4~
1081	\& URxvt.keysym.C-apostrophe: \e033<C-'>
1082	\& URxvt.keysym.C-slash: \e033<C-/>
1083	\& URxvt.keysym.C-semicolon: \e033<C-;>
1084	\& URxvt.keysym.C-grave: \e033<C-`>
1085	\& URxvt.keysym.C-comma: \e033<C-,>
1086	\& URxvt.keysym.C-period: \e033<C-.>
1087	\& URxvt.keysym.C-0x60: \e033<C-`>
1088	\& URxvt.keysym.C-Tab: \e033<C-Tab>
1089	\& URxvt.keysym.C-Return: \e033<C-Return>
1090	\& URxvt.keysym.S-Return: \e033<S-Return>
1091	\& URxvt.keysym.S-space: \e033<S-Space>
1092	\& URxvt.keysym.M-Up: \e033<M-Up>
1093	\& URxvt.keysym.M-Down: \e033<M-Down>
1094	\& URxvt.keysym.M-Left: \e033<M-Left>
1095	\& URxvt.keysym.M-Right: \e033<M-Right>
1096	\& URxvt.keysym.M-C-0: list \e033<M-C- 0123456789 >
1097	\& URxvt.keysym.M-C-a: list \e033<M-C- abcdefghijklmnopqrstuvwxyz >
1098	\& URxvt.keysym.F12: command:\e033]701;zh_CN.GBK\e007
1099	.Ve
1100	.Sp
1101	See some more examples in the documentation for the \fBkeysym\fR resource.
1102	.IP "I'm using keyboard model \s-1XXX\s0 that has extra Prior/Next/Insert keys. How do I make use of them? For example, the Sun Keyboard type 4 has the following mappings that rxvt-unicode doesn't recognize." 4
1103	.IX Item "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 mappings that rxvt-unicode doesn't recognize."
1104	.Vb 6
1105	\& KP_Insert == Insert
1106	\& F22 == Print
1107	\& F27 == Home
1108	\& F29 == Prior
1109	\& F33 == End
1110	\& F35 == Next
1111	.Ve
1112	.Sp
1113	Rather than have rxvt-unicode try to accommodate all the various possible
1114	keyboard mappings, it is better to use `xmodmap' to remap the keys as
1115	required for your particular machine.
1116	.IP "How do I distinguish wether I'm running rxvt-unicode or a regular xterm? I need this to decide about setting colors etc." 4
1117	.IX Item "How do I distinguish wether I'm running rxvt-unicode or a regular xterm? I need this to decide about setting colors etc."
1118	rxvt and rxvt-unicode always export the variable \*(L"\s-1COLORTERM\s0\*(R", so you can
1119	check and see if that is set. Note that several programs, \s-1JED\s0, slrn,
1120	Midnight Commander automatically check this variable to decide whether or
1121	not to use color.
1122	.IP "How do I set the correct, full \s-1IP\s0 address for the \s-1DISPLAY\s0 variable?" 4
1123	.IX Item "How do I set the correct, full IP address for the DISPLAY variable?"
1124	If you've compiled rxvt-unicode with \s-1DISPLAY_IS_IP\s0 and have enabled
1125	insecure mode then it is possible to use the following shell script
1126	snippets to correctly set the display. If your version of rxvt-unicode
1127	wasn't also compiled with \s-1ESCZ_ANSWER\s0 (as assumed in these snippets) then
1128	the \s-1COLORTERM\s0 variable can be used to distinguish rxvt-unicode from a
1129	regular xterm.
1130	.Sp
1131	Courtesy of Chuck Blake <cblake@BBN.COM> with the following shell script
1132	snippets:
1133	.Sp
1134	.Vb 12
1135	\& # Bourne/Korn/POSIX family of shells:
1136	\& [ ${TERM:-foo} = foo ] && TERM=xterm # assume an xterm if we don't know
1137	\& if [ ${TERM:-foo} = xterm ]; then
1138	\& stty -icanon -echo min 0 time 15 # see if enhanced rxvt or not
1139	\& echo -n '^[Z'
1140	\& read term_id
1141	\& stty icanon echo
1142	\& if [ ""${term_id} = '^[[?1;2C' -a ${DISPLAY:-foo} = foo ]; then
1143	\& echo -n '^[[7n' # query the rxvt we are in for the DISPLAY string
1144	\& read DISPLAY # set it in our local shell
1145	\& fi
1146	\& fi
1147	.Ve
1148	.IP "How do I compile the manual pages for myself?" 4
1149	.IX Item "How do I compile the manual pages for myself?"
1150	You need to have a recent version of perl installed as \fI/usr/bin/perl\fR,
1151	one that comes with \fIpod2man\fR, \fIpod2text\fR and \fIpod2html\fR. Then go to
1152	the doc subdirectory and enter \f(CW\*(C`make alldoc\*(C'\fR.
1153	.IP "My question isn't answered here, can I ask a human?" 4
1154	.IX Item "My question isn't answered here, can I ask a human?"
1155	Before sending me mail, you could go to \s-1IRC:\s0 \f(CW\*(C`irc.freenode.net\*(C'\fR,
1156	channel \f(CW\*(C`#rxvt\-unicode\*(C'\fR has some rxvt-unicode enthusiasts that might be
1157	interested in learning about new and exciting problems (but not FAQs :).
1158	.SH "RXVT TECHNICAL REFERENCE"	1269	.SH "RXVT-UNICODE TECHNICAL REFERENCE"
1159	.IX Header "RXVT TECHNICAL REFERENCE"	1270	.IX Header "RXVT-UNICODE TECHNICAL REFERENCE"
1160	.SH "DESCRIPTION"	1271	.SH "DESCRIPTION"
1161	.IX Header "DESCRIPTION"	1272	.IX Header "DESCRIPTION"
1162	The rest of this document describes various technical aspects of	1273	The rest of this document describes various technical aspects of
1163	\&\fBrxvt-unicode\fR. First the description of supported command sequences,	1274	\&\fBrxvt-unicode\fR. First the description of supported command sequences,
1164	followed by pixmap support and last by a description of all features	1275	followed by pixmap support and last by a description of all features
…		…
2015	.TS	2126	.TS
2016	l l .	2127	l l .
2017	4 Shift	2128	4 Shift
2018	8 Meta	2129	8 Meta
2019	16 Control	2130	16 Control
2020	32 Double Click (Rxvt extension)	2131	32 Double Click (rxvt extension)
2021	.TE	2132	.TE
2022		2133
2023	Col = \fB\f(CB\*(C`<x> \- SPACE\*(C'\fB\fR	2134	Col = \fB\f(CB\*(C`<x> \- SPACE\*(C'\fB\fR
2024	.Sp	2135	.Sp
2025	Row = \fB\f(CB\*(C`<y> \- SPACE\*(C'\fB\fR	2136	Row = \fB\f(CB\*(C`<y> \- SPACE\*(C'\fB\fR

Diff Legend

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