xref: /vim-8.2.3635/runtime/doc/quickfix.txt (revision d2ea7cf1)
1*quickfix.txt*  For Vim version 8.2.  Last change: 2021 May 22
2
3
4		  VIM REFERENCE MANUAL    by Bram Moolenaar
5
6
7This subject is introduced in section |30.1| of the user manual.
8
91. Using QuickFix commands		|quickfix|
102. The error window			|quickfix-window|
113. Using more than one list of errors	|quickfix-error-lists|
124. Using :make				|:make_makeprg|
135. Using :grep				|grep|
146. Selecting a compiler			|compiler-select|
157. The error format			|error-file-format|
168. The directory stack			|quickfix-directory-stack|
179. Specific error file formats		|errorformats|
1810. Customizing the quickfix window	|quickfix-window-function|
19
20The quickfix commands are not available when the |+quickfix| feature was
21disabled at compile time.
22
23=============================================================================
241. Using QuickFix commands			*quickfix* *Quickfix* *E42*
25
26Vim has a special mode to speedup the edit-compile-edit cycle.  This is
27inspired by the quickfix option of the Manx's Aztec C compiler on the Amiga.
28The idea is to save the error messages from the compiler in a file and use Vim
29to jump to the errors one by one.  You can examine each problem and fix it,
30without having to remember all the error messages.
31
32In Vim the quickfix commands are used more generally to find a list of
33positions in files.  For example, |:vimgrep| finds pattern matches.  You can
34use the positions in a script with the |getqflist()| function.  Thus you can
35do a lot more than the edit/compile/fix cycle!
36
37If you have the error messages in a file you can start Vim with: >
38	vim -q filename
39
40From inside Vim an easy way to run a command and handle the output is with the
41|:make| command (see below).
42
43The 'errorformat' option should be set to match the error messages from your
44compiler (see |errorformat| below).
45
46							*quickfix-ID*
47Each quickfix list has a unique identifier called the quickfix ID and this
48number will not change within a Vim session. The |getqflist()| function can be
49used to get the identifier assigned to a list. There is also a quickfix list
50number which may change whenever more than ten lists are added to a quickfix
51stack.
52
53						*location-list* *E776*
54A location list is a window-local quickfix list. You get one after commands
55like `:lvimgrep`, `:lgrep`, `:lhelpgrep`, `:lmake`, etc., which create a
56location list instead of a quickfix list as the corresponding `:vimgrep`,
57`:grep`, `:helpgrep`, `:make` do.
58						*location-list-file-window*
59A location list is associated with a window and each window can have a
60separate location list.  A location list can be associated with only one
61window.  The location list is independent of the quickfix list.
62
63When a window with a location list is split, the new window gets a copy of the
64location list.  When there are no longer any references to a location list,
65the location list is destroyed.
66
67						*quickfix-changedtick*
68Every quickfix and location list has a read-only changedtick variable that
69tracks the total number of changes made to the list.  Every time the quickfix
70list is modified, this count is incremented. This can be used to perform an
71action only when the list has changed.  The |getqflist()| and |getloclist()|
72functions can be used to query the current value of changedtick.  You cannot
73change the changedtick variable.
74
75The following quickfix commands can be used.  The location list commands are
76similar to the quickfix commands, replacing the 'c' prefix in the quickfix
77command with 'l'.
78
79							*E924*
80If the current window was closed by an |autocommand| while processing a
81location list command, it will be aborted.
82
83							*E925* *E926*
84If the current quickfix or location list was changed by an |autocommand| while
85processing a quickfix or location list command, it will be aborted.
86
87							*:cc*
88:cc[!] [nr]		Display error [nr].  If [nr] is omitted, the same
89:[nr]cc[!]		error is displayed again.  Without [!] this doesn't
90			work when jumping to another buffer, the current buffer
91			has been changed, there is the only window for the
92			buffer and both 'hidden' and 'autowrite' are off.
93			When jumping to another buffer with [!] any changes to
94			the current buffer are lost, unless 'hidden' is set or
95			there is another window for this buffer.
96			The 'switchbuf' settings are respected when jumping
97			to a buffer.
98			When used in the quickfix window the line number can
99			be used, including "." for the current line and "$"
100			for the last line.
101
102							*:ll*
103:ll[!] [nr]		Same as ":cc", except the location list for the
104:[nr]ll[!]		current window is used instead of the quickfix list.
105
106						*:cn* *:cne* *:cnext* *E553*
107:[count]cn[ext][!]	Display the [count] next error in the list that
108			includes a file name.  If there are no file names at
109			all, go to the [count] next error.  See |:cc| for
110			[!] and 'switchbuf'.
111
112							*:lne* *:lnext*
113:[count]lne[xt][!]	Same as ":cnext", except the location list for the
114			current window is used instead of the quickfix list.
115
116:[count]cN[ext][!]		*:cp* *:cprevious*  *:cprev* *:cN* *:cNext*
117:[count]cp[revious][!]	Display the [count] previous error in the list that
118			includes a file name.  If there are no file names at
119			all, go to the [count] previous error.  See |:cc| for
120			[!] and 'switchbuf'.
121
122
123:[count]lN[ext][!]		*:lp* *:lprevious* *:lprev* *:lN* *:lNext*
124:[count]lp[revious][!]	Same as ":cNext" and ":cprevious", except the location
125			list for the current window is used instead of the
126			quickfix list.
127
128							*:cabo* *:cabove*
129:[count]cabo[ve]	Go to the [count] error above the current line in the
130			current buffer.  If [count] is omitted, then 1 is
131			used.  If there are no errors, then an error message
132			is displayed.  Assumes that the entries in a quickfix
133			list are sorted by their buffer number and line
134			number. If there are multiple errors on the same line,
135			then only the first entry is used.  If [count] exceeds
136			the number of entries above the current line, then the
137			first error in the file is selected.
138
139							*:lab* *:labove*
140:[count]lab[ove]	Same as ":cabove", except the location list for the
141			current window is used instead of the quickfix list.
142
143							*:cbel* *:cbelow*
144:[count]cbel[ow]	Go to the [count] error below the current line in the
145			current buffer.  If [count] is omitted, then 1 is
146			used.  If there are no errors, then an error message
147			is displayed.  Assumes that the entries in a quickfix
148			list are sorted by their buffer number and line
149			number.  If there are multiple errors on the same
150			line, then only the first entry is used.  If [count]
151			exceeds the number of entries below the current line,
152			then the last error in the file is selected.
153
154							*:lbel* *:lbelow*
155:[count]lbel[ow]	Same as ":cbelow", except the location list for the
156			current window is used instead of the quickfix list.
157
158							*:cbe* *:cbefore*
159:[count]cbe[fore]	Go to the [count] error before the current cursor
160			position in the current buffer.  If [count] is
161			omitted, then 1 is used.  If there are no errors, then
162			an error message is displayed.  Assumes that the
163			entries in a quickfix list are sorted by their buffer,
164			line and column numbers.  If [count] exceeds the
165			number of entries before the current position, then
166			the first error in the file is selected.
167
168							*:lbe* *:lbefore*
169:[count]lbe[fore]	Same as ":cbefore", except the location list for the
170			current window is used instead of the quickfix list.
171
172							*:caf* *:cafter*
173:[count]caf[ter]	Go to the [count] error after the current cursor
174			position in the current buffer.  If [count] is
175			omitted, then 1 is used.  If there are no errors, then
176			an error message is displayed.  Assumes that the
177			entries in a quickfix list are sorted by their buffer,
178			line and column numbers.  If [count] exceeds the
179			number of entries after the current position, then
180			the last error in the file is selected.
181
182							*:laf* *:lafter*
183:[count]laf[ter]	Same as ":cafter", except the location list for the
184			current window is used instead of the quickfix list.
185
186							*:cnf* *:cnfile*
187:[count]cnf[ile][!]	Display the first error in the [count] next file in
188			the list that includes a file name.  If there are no
189			file names at all or if there is no next file, go to
190			the [count] next error.  See |:cc| for [!] and
191			'switchbuf'.
192
193							*:lnf* *:lnfile*
194:[count]lnf[ile][!]	Same as ":cnfile", except the location list for the
195			current window is used instead of the quickfix list.
196
197:[count]cNf[ile][!]			*:cpf* *:cpfile* *:cNf* *:cNfile*
198:[count]cpf[ile][!]	Display the last error in the [count] previous file in
199			the list that includes a file name.  If there are no
200			file names at all or if there is no next file, go to
201			the [count] previous error.  See |:cc| for [!] and
202			'switchbuf'.
203
204
205:[count]lNf[ile][!]			*:lpf* *:lpfile* *:lNf* *:lNfile*
206:[count]lpf[ile][!]	Same as ":cNfile" and ":cpfile", except the location
207			list for the current window is used instead of the
208			quickfix list.
209
210							*:crewind* *:cr*
211:cr[ewind][!] [nr]	Display error [nr].  If [nr] is omitted, the FIRST
212			error is displayed.  See |:cc|.
213
214							*:lrewind* *:lr*
215:lr[ewind][!] [nr]	Same as ":crewind", except the location list for the
216			current window is used instead of the quickfix list.
217
218							*:cfirst* *:cfir*
219:cfir[st][!] [nr]	Same as ":crewind".
220
221							*:lfirst* *:lfir*
222:lfir[st][!] [nr]	Same as ":lrewind".
223
224							*:clast* *:cla*
225:cla[st][!] [nr]	Display error [nr].  If [nr] is omitted, the LAST
226			error is displayed.  See |:cc|.
227
228							*:llast* *:lla*
229:lla[st][!] [nr]	Same as ":clast", except the location list for the
230			current window is used instead of the quickfix list.
231
232							*:cq* *:cquit*
233:cq[uit][!]
234:{N}cq[uit][!]
235:cq[uit][!] {N}		Quit Vim with error code {N}.  {N} defaults to one.
236			Useful when Vim is called from another program:
237			e.g., a compiler will not compile the same file again,
238			`git commit` will abort the committing process, `fc`
239			(built-in for shells like bash and zsh) will not
240			execute the command, etc.
241			{N} can also be zero, in which case Vim exits
242			normally.
243			WARNING: All changes in files are lost!  Also when the
244			[!] is not used.  It works like ":qall!" |:qall|,
245			except that Vim returns a non-zero exit code.
246
247							*:cf* *:cfile*
248:cf[ile][!] [errorfile]	Read the error file and jump to the first error.
249			This is done automatically when Vim is started with
250			the -q option.  You can use this command when you
251			keep Vim running while compiling.  If you give the
252			name of the errorfile, the 'errorfile' option will
253			be set to [errorfile].  See |:cc| for [!].
254			If the encoding of the error file differs from the
255			'encoding' option, you can use the 'makeencoding'
256			option to specify the encoding.
257
258							*:lf* *:lfi* *:lfile*
259:lf[ile][!] [errorfile]	Same as ":cfile", except the location list for the
260			current window is used instead of the quickfix list.
261			You can not use the -q command-line option to set
262			the location list.
263
264
265:cg[etfile] [errorfile]					*:cg* *:cgetfile*
266			Read the error file.  Just like ":cfile" but don't
267			jump to the first error.
268			If the encoding of the error file differs from the
269			'encoding' option, you can use the 'makeencoding'
270			option to specify the encoding.
271
272
273:lg[etfile] [errorfile]					*:lg* *:lge* *:lgetfile*
274			Same as ":cgetfile", except the location list for the
275			current window is used instead of the quickfix list.
276
277							*:caddf* *:caddfile*
278:caddf[ile] [errorfile]	Read the error file and add the errors from the
279			errorfile to the current quickfix list. If a quickfix
280			list is not present, then a new list is created.
281			If the encoding of the error file differs from the
282			'encoding' option, you can use the 'makeencoding'
283			option to specify the encoding.
284
285							*:laddf* *:laddfile*
286:laddf[ile] [errorfile]	Same as ":caddfile", except the location list for the
287			current window is used instead of the quickfix list.
288
289						*:cb* *:cbuffer* *E681*
290:cb[uffer][!] [bufnr]	Read the error list from the current buffer.
291			When [bufnr] is given it must be the number of a
292			loaded buffer.  That buffer will then be used instead
293			of the current buffer.
294			A range can be specified for the lines to be used.
295			Otherwise all lines in the buffer are used.
296			See |:cc| for [!].
297
298						*:lb* *:lbuffer*
299:lb[uffer][!] [bufnr]	Same as ":cbuffer", except the location list for the
300			current window is used instead of the quickfix list.
301
302						*:cgetb* *:cgetbuffer*
303:cgetb[uffer] [bufnr]	Read the error list from the current buffer.  Just
304			like ":cbuffer" but don't jump to the first error.
305
306						*:lgetb* *:lgetbuffer*
307:lgetb[uffer] [bufnr]	Same as ":cgetbuffer", except the location list for
308			the current window is used instead of the quickfix
309			list.
310
311						*:cad* *:cadd* *:caddbuffer*
312:cad[dbuffer] [bufnr]	Read the error list from the current buffer and add
313			the errors to the current quickfix list.  If a
314			quickfix list is not present, then a new list is
315			created. Otherwise, same as ":cbuffer".
316
317							*:laddb* *:laddbuffer*
318:laddb[uffer] [bufnr]	Same as ":caddbuffer", except the location list for
319			the current window is used instead of the quickfix
320			list.
321
322							*:cex* *:cexpr* *E777*
323:cex[pr][!] {expr}	Create a quickfix list using the result of {expr} and
324			jump to the first error.
325			If {expr} is a String, then each newline terminated
326			line in the String is processed using the global value
327			of 'errorformat' and the result is added to the
328			quickfix list.
329			If {expr} is a List, then each String item in the list
330			is processed and added to the quickfix list.  Non
331			String items in the List are ignored.
332			See |:cc| for [!].
333			Examples: >
334				:cexpr system('grep -n xyz *')
335				:cexpr getline(1, '$')
336<
337							*:lex* *:lexpr*
338:lex[pr][!] {expr}	Same as |:cexpr|, except the location list for the
339			current window is used instead of the quickfix list.
340
341							*:cgete* *:cgetexpr*
342:cgete[xpr] {expr}	Create a quickfix list using the result of {expr}.
343			Just like |:cexpr|, but don't jump to the first error.
344
345							*:lgete* *:lgetexpr*
346:lgete[xpr] {expr}	Same as |:cgetexpr|, except the location list for the
347			current window is used instead of the quickfix list.
348
349							*:cadde* *:caddexpr*
350:cadde[xpr] {expr}	Evaluate {expr} and add the resulting lines to the
351			current quickfix list. If a quickfix list is not
352			present, then a new list is created. The current
353			cursor position will not be changed. See |:cexpr| for
354			more information.
355			Example: >
356    :g/mypattern/caddexpr expand("%") . ":" . line(".") .  ":" . getline(".")
357<
358						*:lad* *:addd* *:laddexpr*
359:lad[dexpr] {expr}	Same as ":caddexpr", except the location list for the
360			current window is used instead of the quickfix list.
361
362							*:cl* *:clist*
363:cl[ist] [from] [, [to]]
364			List all errors that are valid |quickfix-valid|.
365			If numbers [from] and/or [to] are given, the respective
366			range of errors is listed.  A negative number counts
367			from the last error backwards, -1 being the last error.
368			The 'switchbuf' settings are respected when jumping
369			to a buffer.
370			The |:filter| command can be used to display only the
371			quickfix entries matching a supplied pattern. The
372			pattern is matched against the filename, module name,
373			pattern and text of the entry.
374
375:cl[ist] +{count}	List the current and next {count} valid errors.  This
376			is similar to ":clist from from+count", where "from"
377			is the current error position.
378
379:cl[ist]! [from] [, [to]]
380			List all errors.
381
382:cl[ist]! +{count}	List the current and next {count} error lines.  This
383                        is useful to see unrecognized lines after the current
384			one.  For example, if ":clist" shows:
385        8384 testje.java:252: error: cannot find symbol ~
386                        Then using ":cl! +3" shows the reason:
387        8384 testje.java:252: error: cannot find symbol ~
388        8385:   ZexitCode = Fmainx(); ~
389        8386:               ^ ~
390        8387:   symbol:   method Fmainx() ~
391
392:lli[st] [from] [, [to]]				*:lli* *:llist*
393			Same as ":clist", except the location list for the
394			current window is used instead of the quickfix list.
395
396:lli[st]! [from] [, [to]]
397			List all the entries in the location list for the
398			current window.
399
400If you insert or delete lines, mostly the correct error location is still
401found because hidden marks are used.  Sometimes, when the mark has been
402deleted for some reason, the message "line changed" is shown to warn you that
403the error location may not be correct.  If you quit Vim and start again the
404marks are lost and the error locations may not be correct anymore.
405
406Two autocommands are available for running commands before and after a
407quickfix command (':make', ':grep' and so on) is executed. See
408|QuickFixCmdPre| and |QuickFixCmdPost| for details.
409
410						*QuickFixCmdPost-example*
411When 'encoding' differs from the locale, the error messages may have a
412different encoding from what Vim is using.  To convert the messages you can
413use this code: >
414	function QfMakeConv()
415	   let qflist = getqflist()
416	   for i in qflist
417	      let i.text = iconv(i.text, "cp936", "utf-8")
418	   endfor
419	   call setqflist(qflist)
420	endfunction
421
422	au QuickfixCmdPost make call QfMakeConv()
423Another option is using 'makeencoding'.
424
425							*quickfix-title*
426Every quickfix and location list has a title. By default the title is set to
427the command that created the list. The |getqflist()| and |getloclist()|
428functions can be used to get the title of a quickfix and a location list
429respectively. The |setqflist()| and |setloclist()| functions can be used to
430modify the title of a quickfix and location list respectively. Examples: >
431	call setqflist([], 'a', {'title' : 'Cmd output'})
432	echo getqflist({'title' : 1})
433	call setloclist(3, [], 'a', {'title' : 'Cmd output'})
434	echo getloclist(3, {'title' : 1})
435<
436							*quickfix-index*
437When you jump to a quickfix/location list entry using any of the quickfix
438commands (e.g. |:cc|, |:cnext|, |:cprev|, etc.), that entry becomes the
439currently selected entry. The index of the currently selected entry in a
440quickfix/location list can be obtained using the getqflist()/getloclist()
441functions. Examples: >
442	echo getqflist({'idx' : 0}).idx
443	echo getqflist({'id' : qfid, 'idx' : 0}).idx
444	echo getloclist(2, {'idx' : 0}).idx
445<
446For a new quickfix list, the first entry is selected and the index is 1.  Any
447entry in any quickfix/location list can be set as the currently selected entry
448using the setqflist() function. Examples: >
449	call setqflist([], 'a', {'idx' : 12})
450	call setqflist([], 'a', {'id' : qfid, 'idx' : 7})
451	call setloclist(1, [], 'a', {'idx' : 7})
452<
453							*quickfix-size*
454You can get the number of entries (size) in a quickfix and a location list
455using the |getqflist()| and |getloclist()| functions respectively. Examples: >
456	echo getqflist({'size' : 1})
457	echo getloclist(5, {'size' : 1})
458<
459							*quickfix-context*
460Any Vim type can be associated as a context with a quickfix or location list.
461The |setqflist()| and the |setloclist()| functions can be used to associate a
462context with a quickfix and a location list respectively. The |getqflist()|
463and the |getloclist()| functions can be used to retrieve the context of a
464quickfix and a location list respectively. This is useful for a Vim plugin
465dealing with multiple quickfix/location lists.
466Examples: >
467
468	let somectx = {'name' : 'Vim', 'type' : 'Editor'}
469	call setqflist([], 'a', {'context' : somectx})
470	echo getqflist({'context' : 1})
471
472	let newctx = ['red', 'green', 'blue']
473	call setloclist(2, [], 'a', {'id' : qfid, 'context' : newctx})
474	echo getloclist(2, {'id' : qfid, 'context' : 1})
475<
476							*quickfix-parse*
477You can parse a list of lines using 'errorformat' without creating or
478modifying a quickfix list using the |getqflist()| function. Examples: >
479	echo getqflist({'lines' : ["F1:10:Line10", "F2:20:Line20"]})
480	echo getqflist({'lines' : systemlist('grep -Hn quickfix *')})
481This returns a dictionary where the 'items' key contains the list of quickfix
482entries parsed from lines. The following shows how to use a custom
483'errorformat' to parse the lines without modifying the 'errorformat' option: >
484	echo getqflist({'efm' : '%f#%l#%m', 'lines' : ['F1#10#Line']})
485<
486
487EXECUTE A COMMAND IN ALL THE BUFFERS IN QUICKFIX OR LOCATION LIST:
488							*:cdo*
489:cdo[!] {cmd}		Execute {cmd} in each valid entry in the quickfix list.
490			It works like doing this: >
491				:cfirst
492				:{cmd}
493				:cnext
494				:{cmd}
495				etc.
496<			When the current file can't be |abandon|ed and the [!]
497			is not present, the command fails.
498			When going to the next entry fails execution stops.
499			The last buffer (or where an error occurred) becomes
500			the current buffer.
501			{cmd} can contain '|' to concatenate several commands.
502
503			Only valid entries in the quickfix list are used.
504			A range can be used to select entries, e.g.: >
505				:10,$cdo cmd
506<			To skip entries 1 to 9.
507
508			Note: While this command is executing, the Syntax
509			autocommand event is disabled by adding it to
510			'eventignore'.  This considerably speeds up editing
511			each buffer.
512			Also see |:bufdo|, |:tabdo|, |:argdo|, |:windo|,
513			|:ldo|, |:cfdo| and |:lfdo|.
514
515							*:cfdo*
516:cfdo[!] {cmd}		Execute {cmd} in each file in the quickfix list.
517			It works like doing this: >
518				:cfirst
519				:{cmd}
520				:cnfile
521				:{cmd}
522				etc.
523<			Otherwise it works the same as `:cdo`.
524
525							*:ldo*
526:ld[o][!] {cmd}		Execute {cmd} in each valid entry in the location list
527			for the current window.
528			It works like doing this: >
529				:lfirst
530				:{cmd}
531				:lnext
532				:{cmd}
533				etc.
534<			Only valid entries in the location list are used.
535			Otherwise it works the same as `:cdo`.
536
537							*:lfdo*
538:lfdo[!] {cmd}		Execute {cmd} in each file in the location list for
539			the current window.
540			It works like doing this: >
541				:lfirst
542				:{cmd}
543				:lnfile
544				:{cmd}
545				etc.
546<			Otherwise it works the same as `:ldo`.
547
548FILTERING A QUICKFIX OR LOCATION LIST:
549				    *cfilter-plugin* *:Cfilter* *:Lfilter*
550If you have too many entries in a quickfix list, you can use the cfilter
551plugin to reduce the number of entries.  Load the plugin with: >
552
553    packadd cfilter
554
555Then you can use the following commands to filter a quickfix/location list: >
556
557    :Cfilter[!] /{pat}/
558    :Lfilter[!] /{pat}/
559
560The |:Cfilter| command creates a new quickfix list from the entries matching
561{pat} in the current quickfix list. {pat} is a Vim |regular-expression|
562pattern. Both the file name and the text of the entries are matched against
563{pat}. If the optional ! is supplied, then the entries not matching {pat} are
564used. The pattern can be optionally enclosed using one of the following
565characters: ', ", /. If the pattern is empty, then the last used search
566pattern is used.
567
568The |:Lfilter| command does the same as |:Cfilter| but operates on the current
569location list.
570
571=============================================================================
5722. The error window					*quickfix-window*
573
574					    *:cope* *:copen* *w:quickfix_title*
575:cope[n] [height]	Open a window to show the current list of errors.
576
577			When [height] is given, the window becomes that high
578			(if there is room).  When [height] is omitted the
579			window is made ten lines high.
580
581			If there already is a quickfix window, it will be made
582			the current window.  It is not possible to open a
583			second quickfix window.  If [height] is given the
584			existing window will be resized to it.
585
586							*quickfix-buffer*
587			The window will contain a special buffer, with
588			'buftype' equal to "quickfix".  Don't change this!
589			The window will have the w:quickfix_title variable set
590			which will indicate the command that produced the
591			quickfix list. This can be used to compose a custom
592			status line if the value of 'statusline' is adjusted
593			properly. Whenever this buffer is modified by a
594			quickfix command or function, the |b:changedtick|
595			variable is incremented.  You can get the number of
596			this buffer using the getqflist() and getloclist()
597			functions by passing the 'qfbufnr' item. For a
598			location list, this buffer is wiped out when the
599			location list is removed.
600
601							*:lop* *:lopen*
602:lop[en] [height]	Open a window to show the location list for the
603			current window. Works only when the location list for
604			the current window is present.  You can have more than
605			one location window opened at a time.  Otherwise, it
606			acts the same as ":copen".
607
608							*:ccl* *:cclose*
609:ccl[ose]		Close the quickfix window.
610
611							*:lcl* *:lclose*
612:lcl[ose]		Close the window showing the location list for the
613			current window.
614
615							*:cw* *:cwindow*
616:cw[indow] [height]	Open the quickfix window when there are recognized
617			errors.  If the window is already open and there are
618			no recognized errors, close the window.
619
620							*:lw* *:lwindow*
621:lw[indow] [height]	Same as ":cwindow", except use the window showing the
622			location list for the current window.
623
624							*:cbo* *:cbottom*
625:cbo[ttom]		Put the cursor in the last line of the quickfix window
626			and scroll to make it visible.  This is useful for
627			when errors are added by an asynchronous callback.
628			Only call it once in a while if there are many
629			updates to avoid a lot of redrawing.
630
631							*:lbo* *:lbottom*
632:lbo[ttom]		Same as ":cbottom", except use the window showing the
633			location list for the current window.
634
635Normally the quickfix window is at the bottom of the screen.  If there are
636vertical splits, it's at the bottom of the rightmost column of windows.  To
637make it always occupy the full width: >
638	:botright cwindow
639You can move the window around with |window-moving| commands.
640For example, to move it to the top: CTRL-W K
641The 'winfixheight' option will be set, which means that the window will mostly
642keep its height, ignoring 'winheight' and 'equalalways'.  You can change the
643height manually (e.g., by dragging the status line above it with the mouse).
644
645In the quickfix window, each line is one error.  The line number is equal to
646the error number.  The current entry is highlighted with the QuickFixLine
647highlighting.  You can change it to your liking, e.g.: >
648	:hi QuickFixLine ctermbg=Yellow guibg=Yellow
649
650You can use ":.cc" to jump to the error under the cursor.
651Hitting the <Enter> key or double-clicking the mouse on a line has the same
652effect.  The file containing the error is opened in the window above the
653quickfix window.  If there already is a window for that file, it is used
654instead.  If the buffer in the used window has changed, and the error is in
655another file, jumping to the error will fail.  You will first have to make
656sure the window contains a buffer which can be abandoned.
657					*CTRL-W_<Enter>* *CTRL-W_<CR>*
658You can use CTRL-W <Enter> to open a new window and jump to the error there.
659
660When the quickfix window has been filled, two autocommand events are
661triggered.  First the 'filetype' option is set to "qf", which triggers the
662FileType event (also see |qf.vim|).  Then the BufReadPost event is triggered,
663using "quickfix" for the buffer name.  This can be used to perform some action
664on the listed errors.  Example: >
665	au BufReadPost quickfix  setlocal modifiable
666		\ | silent exe 'g/^/s//\=line(".")." "/'
667		\ | setlocal nomodifiable
668This prepends the line number to each line.  Note the use of "\=" in the
669substitute string of the ":s" command, which is used to evaluate an
670expression.
671The BufWinEnter event is also triggered, again using "quickfix" for the buffer
672name.
673
674Note: When adding to an existing quickfix list the autocommand are not
675triggered.
676
677Note: Making changes in the quickfix window has no effect on the list of
678errors.  'modifiable' is off to avoid making changes.  If you delete or insert
679lines anyway, the relation between the text and the error number is messed up.
680If you really want to do this, you could write the contents of the quickfix
681window to a file and use ":cfile" to have it parsed and used as the new error
682list.
683
684						*location-list-window*
685The location list window displays the entries in a location list.  When you
686open a location list window, it is created below the current window and
687displays the location list for the current window.  The location list window
688is similar to the quickfix window, except that you can have more than one
689location list window open at a time. When you use a location list command in
690this window, the displayed location list is used.
691
692When you select a file from the location list window, the following steps are
693used to find a window to edit the file:
694
6951. If a window with the location list displayed in the location list window is
696   present, then the file is opened in that window.
6972. If the above step fails and if the file is already opened in another
698   window, then that window is used.
6993. If the above step fails then an existing window showing a buffer with
700   'buftype' not set is used.
7014. If the above step fails, then the file is edited in a new window.
702
703In all of the above cases, if the location list for the selected window is not
704yet set, then it is set to the location list displayed in the location list
705window.
706
707							*quickfix-window-ID*
708You can use the |getqflist()| and |getloclist()| functions to obtain the
709window ID of the quickfix window and location list window respectively (if
710present).  Examples: >
711	echo getqflist({'winid' : 1}).winid
712	echo getloclist(2, {'winid' : 1}).winid
713<
714							*getqflist-examples*
715The |getqflist()| and |getloclist()| functions can be used to get the various
716attributes of a quickfix and location list respectively. Some examples for
717using these functions are below:
718>
719    " get the title of the current quickfix list
720    :echo getqflist({'title' : 0}).title
721
722    " get the identifier of the current quickfix list
723    :let qfid = getqflist({'id' : 0}).id
724
725    " get the identifier of the fourth quickfix list in the stack
726    :let qfid = getqflist({'nr' : 4, 'id' : 0}).id
727
728    " check whether a quickfix list with a specific identifier exists
729    :if getqflist({'id' : qfid}).id == qfid
730
731    " get the index of the current quickfix list in the stack
732    :let qfnum = getqflist({'nr' : 0}).nr
733
734    " get the items of a quickfix list specified by an identifier
735    :echo getqflist({'id' : qfid, 'items' : 0}).items
736
737    " get the number of entries in a quickfix list specified by an id
738    :echo getqflist({'id' : qfid, 'size' : 0}).size
739
740    " get the context of the third quickfix list in the stack
741    :echo getqflist({'nr' : 3, 'context' : 0}).context
742
743    " get the number of quickfix lists in the stack
744    :echo getqflist({'nr' : '$'}).nr
745
746    " get the number of times the current quickfix list is changed
747    :echo getqflist({'changedtick' : 0}).changedtick
748
749    " get the current entry in a quickfix list specified by an identifier
750    :echo getqflist({'id' : qfid, 'idx' : 0}).idx
751
752    " get all the quickfix list attributes using an identifier
753    :echo getqflist({'id' : qfid, 'all' : 0})
754
755    " parse text from a List of lines and return a quickfix list
756    :let myList = ["a.java:10:L10", "b.java:20:L20"]
757    :echo getqflist({'lines' : myList}).items
758
759    " parse text using a custom 'efm' and return a quickfix list
760    :echo getqflist({'lines' : ['a.c#10#Line 10'], 'efm':'%f#%l#%m'}).items
761
762    " get the quickfix list window id
763    :echo getqflist({'winid' : 0}).winid
764
765    " get the quickfix list window buffer number
766    :echo getqflist({'qfbufnr' : 0}).qfbufnr
767
768    " get the context of the current location list
769    :echo getloclist(0, {'context' : 0}).context
770
771    " get the location list window id of the third window
772    :echo getloclist(3, {'winid' : 0}).winid
773
774    " get the location list window buffer number of the third window
775    :echo getloclist(3, {'qfbufnr' : 0}).qfbufnr
776
777    " get the file window id of a location list window (winnr: 4)
778    :echo getloclist(4, {'filewinid' : 0}).filewinid
779<
780							*setqflist-examples*
781The |setqflist()| and |setloclist()| functions can be used to set the various
782attributes of a quickfix and location list respectively. Some examples for
783using these functions are below:
784>
785    " create an empty quickfix list with a title and a context
786    :let t = 'Search results'
787    :let c = {'cmd' : 'grep'}
788    :call setqflist([], ' ', {'title' : t, 'context' : c})
789
790    " set the title of the current quickfix list
791    :call setqflist([], 'a', {'title' : 'Mytitle'})
792
793    " change the current entry in the list specified by an identifier
794    :call setqflist([], 'a', {'id' : qfid, 'idx' : 10})
795
796    " set the context of a quickfix list specified by an identifier
797    :call setqflist([], 'a', {'id' : qfid, 'context' : {'val' : 100}})
798
799    " create a new quickfix list from a command output
800    :call setqflist([], ' ', {'lines' : systemlist('grep -Hn main *.c')})
801
802    " parse text using a custom efm and add to a particular quickfix list
803    :call setqflist([], 'a', {'id' : qfid,
804		\ 'lines' : ["a.c#10#L10", "b.c#20#L20"], 'efm':'%f#%l#%m'})
805
806    " add items to the quickfix list specified by an identifier
807    :let newItems = [{'filename' : 'a.txt', 'lnum' : 10, 'text' : "Apple"},
808		    \ {'filename' : 'b.txt', 'lnum' : 20, 'text' : "Orange"}]
809    :call setqflist([], 'a', {'id' : qfid, 'items' : newItems})
810
811    " empty a quickfix list specified by an identifier
812    :call setqflist([], 'r', {'id' : qfid, 'items' : []})
813
814    " free all the quickfix lists in the stack
815    :call setqflist([], 'f')
816
817    " set the title of the fourth quickfix list
818    :call setqflist([], 'a', {'nr' : 4, 'title' : 'SomeTitle'})
819
820    " create a new quickfix list at the end of the stack
821    :call setqflist([], ' ', {'nr' : '$',
822			\ 'lines' : systemlist('grep -Hn class *.java')})
823
824    " create a new location list from a command output
825    :call setloclist(0, [], ' ', {'lines' : systemlist('grep -Hn main *.c')})
826
827    " replace the location list entries for the third window
828    :call setloclist(3, [], 'r', {'items' : newItems})
829<
830=============================================================================
8313. Using more than one list of errors			*quickfix-error-lists*
832
833So far has been assumed that there is only one list of errors.  Actually the
834ten last used lists are remembered.  When starting a new list, the previous
835ones are automatically kept.  Two commands can be used to access older error
836lists.  They set one of the existing error lists as the current one.
837
838						*:colder* *:col* *E380*
839:col[der] [count]	Go to older error list.  When [count] is given, do
840			this [count] times.  When already at the oldest error
841			list, an error message is given.
842
843						*:lolder* *:lol*
844:lol[der] [count]	Same as `:colder`, except use the location list for
845			the current window instead of the quickfix list.
846
847						*:cnewer* *:cnew* *E381*
848:cnew[er] [count]	Go to newer error list.  When [count] is given, do
849			this [count] times.  When already at the newest error
850			list, an error message is given.
851
852						*:lnewer* *:lnew*
853:lnew[er] [count]	Same as `:cnewer`, except use the location list for
854			the current window instead of the quickfix list.
855
856						*:chistory* *:chi*
857:[count]chi[story]	Show the list of error lists.  The current list is
858			marked with ">".  The output looks like:
859				  error list 1 of 3; 43 errors ~
860				> error list 2 of 3; 0 errors ~
861				  error list 3 of 3; 15 errors ~
862
863			When [count] is given, then the count'th quickfix
864			list is made the current list. Example: >
865				" Make the 4th quickfix list current
866				:4chistory
867<
868						*:lhistory* *:lhi*
869:[count]lhi[story]	Show the list of location lists, otherwise like
870			`:chistory`.
871
872When adding a new error list, it becomes the current list.
873
874When ":colder" has been used and ":make" or ":grep" is used to add a new error
875list, one newer list is overwritten.  This is especially useful if you are
876browsing with ":grep" |grep|.  If you want to keep the more recent error
877lists, use ":cnewer 99" first.
878
879To get the number of lists in the quickfix and location list stack, you can
880use the |getqflist()| and |getloclist()| functions respectively with the list
881number set to the special value '$'. Examples: >
882	echo getqflist({'nr' : '$'}).nr
883	echo getloclist(3, {'nr' : '$'}).nr
884To get the number of the current list in the stack: >
885	echo getqflist({'nr' : 0}).nr
886<
887=============================================================================
8884. Using :make						*:make_makeprg*
889
890							*:mak* *:make*
891:mak[e][!] [arguments]	1. All relevant |QuickFixCmdPre| autocommands are
892			   executed.
893			2. If the 'autowrite' option is on, write any changed
894			   buffers
895			3. An errorfile name is made from 'makeef'.  If
896			   'makeef' doesn't contain "##", and a file with this
897			   name already exists, it is deleted.
898			4. The program given with the 'makeprg' option is
899			   started (default "make") with the optional
900			   [arguments] and the output is saved in the
901			   errorfile (for Unix it is also echoed on the
902			   screen).
903			5. The errorfile is read using 'errorformat'.
904			6. All relevant |QuickFixCmdPost| autocommands are
905			   executed.  See example below.
906			7. If [!] is not given the first error is jumped to.
907			8. The errorfile is deleted.
908			9. You can now move through the errors with commands
909			   like |:cnext| and |:cprevious|, see above.
910			This command does not accept a comment, any "
911			characters are considered part of the arguments.
912			If the encoding of the program output differs from the
913			'encoding' option, you can use the 'makeencoding'
914			option to specify the encoding.
915
916							*:lmak* *:lmake*
917:lmak[e][!] [arguments]
918			Same as ":make", except the location list for the
919			current window is used instead of the quickfix list.
920
921The ":make" command executes the command given with the 'makeprg' option.
922This is done by passing the command to the shell given with the 'shell'
923option.  This works almost like typing
924
925	":!{makeprg} [arguments] {shellpipe} {errorfile}".
926
927{makeprg} is the string given with the 'makeprg' option.  Any command can be
928used, not just "make".  Characters '%' and '#' are expanded as usual on a
929command-line.  You can use "%<" to insert the current file name without
930extension, or "#<" to insert the alternate file name without extension, for
931example: >
932   :set makeprg=make\ #<.o
933
934[arguments] is anything that is typed after ":make".
935{shellpipe} is the 'shellpipe' option.
936{errorfile} is the 'makeef' option, with ## replaced to make it unique.
937
938The placeholder "$*" can be used for the argument list in {makeprg} if the
939command needs some additional characters after its arguments.  The $* is
940replaced then by all arguments.  Example: >
941   :set makeprg=latex\ \\\\nonstopmode\ \\\\input\\{$*}
942or simpler >
943   :let &mp = 'latex \\nonstopmode \\input\{$*}'
944"$*" can be given multiple times, for example: >
945   :set makeprg=gcc\ -o\ $*\ $*
946
947The 'shellpipe' option defaults to ">" for the Amiga and ">%s 2>&1" for Win32.
948This means that the output of the compiler is saved in a file and not shown on
949the screen directly.  For Unix "| tee" is used.  The compiler output is shown
950on the screen and saved in a file the same time.  Depending on the shell used
951"|& tee" or "2>&1| tee" is the default, so stderr output will be included.
952
953If 'shellpipe' is empty, the {errorfile} part will be omitted.  This is useful
954for compilers that write to an errorfile themselves (e.g., Manx's Amiga C).
955
956
957Using QuickFixCmdPost to fix the encoding ~
958
959It may be that 'encoding' is set to an encoding that differs from the messages
960your build program produces.  This example shows how to fix this after Vim has
961read the error messages: >
962
963	function QfMakeConv()
964	   let qflist = getqflist()
965	   for i in qflist
966	      let i.text = iconv(i.text, "cp936", "utf-8")
967	   endfor
968	   call setqflist(qflist)
969	endfunction
970
971	au QuickfixCmdPost make call QfMakeConv()
972
973(Example by Faque Cheng)
974Another option is using 'makeencoding'.
975
976==============================================================================
9775. Using :vimgrep and :grep				*grep* *lid*
978
979Vim has two ways to find matches for a pattern: Internal and external.  The
980advantage of the internal grep is that it works on all systems and uses the
981powerful Vim search patterns.  An external grep program can be used when the
982Vim grep does not do what you want.
983
984The internal method will be slower, because files are read into memory.  The
985advantages are:
986- Line separators and encoding are automatically recognized, as if a file is
987  being edited.
988- Uses Vim search patterns.  Multi-line patterns can be used.
989- When plugins are enabled: compressed and remote files can be searched.
990	|gzip| |netrw|
991
992To be able to do this Vim loads each file as if it is being edited.  When
993there is no match in the file the associated buffer is wiped out again.  The
994'hidden' option is ignored here to avoid running out of memory or file
995descriptors when searching many files.  However, when the |:hide| command
996modifier is used the buffers are kept loaded.  This makes following searches
997in the same files a lot faster.
998
999Note that |:copen| (or |:lopen| for |:lgrep|) may be used to open a buffer
1000containing the search results in linked form.  The |:silent| command may be
1001used to suppress the default full screen grep output.  The ":grep!" form of
1002the |:grep| command doesn't jump to the first match automatically.  These
1003commands can be combined to create a NewGrep command: >
1004
1005        command! -nargs=+ NewGrep execute 'silent grep! <args>' | copen 42
1006
1007
10085.1 using Vim's internal grep
1009
1010					*:vim* *:vimgrep* *E682* *E683*
1011:vim[grep][!] /{pattern}/[g][j][f] {file} ...
1012			Search for {pattern} in the files {file} ... and set
1013			the error list to the matches.  Files matching
1014			'wildignore' are ignored; files in 'suffixes' are
1015			searched last.
1016
1017			{pattern} is a Vim search pattern.  Instead of
1018			enclosing it in / any non-ID character (see
1019			|'isident'|) can be used, so long as it does not
1020			appear in {pattern}.
1021			'ignorecase' applies.  To overrule it put |/\c| in the
1022			pattern to ignore case or |/\C| to match case.
1023			'smartcase' is not used.
1024			If {pattern} is empty (e.g. // is specified), the last
1025			used search pattern is used. |last-pattern|
1026
1027			Flags:
1028			'g'  Without the 'g' flag each line is added only
1029			     once.  With 'g' every match is added.
1030
1031			'j'  Without the 'j' flag Vim jumps to the first
1032			     match.  With 'j' only the quickfix list is
1033			     updated.  With the [!] any changes in the current
1034			     buffer are abandoned.
1035
1036			'f'  When the 'f' flag is specified, fuzzy string
1037			     matching is used to find matching lines. In this
1038			     case, {pattern} is treated as a literal string
1039			     instead of a regular expression.  See
1040			     |fuzzy-match| for more information about fuzzy
1041			     matching strings.
1042
1043			|QuickFixCmdPre| and |QuickFixCmdPost| are triggered.
1044			A file that is opened for matching may use a buffer
1045			number, but it is reused if possible to avoid
1046			consuming buffer numbers.
1047
1048:{count}vim[grep] ...
1049			When a number is put before the command this is used
1050			as the maximum number of matches to find.  Use
1051			":1vimgrep pattern file" to find only the first.
1052			Useful if you only want to check if there is a match
1053			and quit quickly when it's found.
1054
1055			Every second or so the searched file name is displayed
1056			to give you an idea of the progress made.
1057			Examples: >
1058				:vimgrep /an error/ *.c
1059				:vimgrep /\<FileName\>/ *.h include/*
1060				:vimgrep /myfunc/ **/*.c
1061<			For the use of "**" see |starstar-wildcard|.
1062
1063:vim[grep][!] {pattern} {file} ...
1064			Like above, but instead of enclosing the pattern in a
1065			non-ID character use a white-separated pattern.  The
1066			pattern must start with an ID character.
1067			Example: >
1068				:vimgrep Error *.c
1069<
1070							*:lv* *:lvimgrep*
1071:lv[imgrep][!] /{pattern}/[g][j][f] {file} ...
1072:lv[imgrep][!] {pattern} {file} ...
1073			Same as ":vimgrep", except the location list for the
1074			current window is used instead of the quickfix list.
1075
1076						*:vimgrepa* *:vimgrepadd*
1077:vimgrepa[dd][!] /{pattern}/[g][j][f] {file} ...
1078:vimgrepa[dd][!] {pattern} {file} ...
1079			Just like ":vimgrep", but instead of making a new list
1080			of errors the matches are appended to the current
1081			list.
1082
1083						*:lvimgrepa* *:lvimgrepadd*
1084:lvimgrepa[dd][!] /{pattern}/[g][j][f] {file} ...
1085:lvimgrepa[dd][!] {pattern} {file} ...
1086			Same as ":vimgrepadd", except the location list for
1087			the current window is used instead of the quickfix
1088			list.
1089
10905.2 External grep
1091
1092Vim can interface with "grep" and grep-like programs (such as the GNU
1093id-utils) in a similar way to its compiler integration (see |:make| above).
1094
1095[Unix trivia: The name for the Unix "grep" command comes from ":g/re/p", where
1096"re" stands for Regular Expression.]
1097
1098							    *:gr* *:grep*
1099:gr[ep][!] [arguments]	Just like ":make", but use 'grepprg' instead of
1100			'makeprg' and 'grepformat' instead of 'errorformat'.
1101			When 'grepprg' is "internal" this works like
1102			|:vimgrep|.  Note that the pattern needs to be
1103			enclosed in separator characters then.
1104			If the encoding of the program output differs from the
1105			'encoding' option, you can use the 'makeencoding'
1106			option to specify the encoding.
1107
1108							    *:lgr* *:lgrep*
1109:lgr[ep][!] [arguments]	Same as ":grep", except the location list for the
1110			current window is used instead of the quickfix list.
1111
1112							*:grepa* *:grepadd*
1113:grepa[dd][!] [arguments]
1114			Just like ":grep", but instead of making a new list of
1115			errors the matches are appended to the current list.
1116			Example: >
1117				:call setqflist([])
1118				:bufdo grepadd! something %
1119<			The first command makes a new error list which is
1120			empty.  The second command executes "grepadd" for each
1121			listed buffer.  Note the use of ! to avoid that
1122			":grepadd" jumps to the first error, which is not
1123			allowed with |:bufdo|.
1124			An example that uses the argument list and avoids
1125			errors for files without matches: >
1126				:silent argdo try
1127				  \ | grepadd! something %
1128				  \ | catch /E480:/
1129				  \ | endtry"
1130<
1131			If the encoding of the program output differs from the
1132			'encoding' option, you can use the 'makeencoding'
1133			option to specify the encoding.
1134
1135							*:lgrepa* *:lgrepadd*
1136:lgrepa[dd][!] [arguments]
1137			Same as ":grepadd", except the location list for the
1138			current window is used instead of the quickfix list.
1139
11405.3 Setting up external grep
1141
1142If you have a standard "grep" program installed, the :grep command may work
1143well with the defaults.  The syntax is very similar to the standard command: >
1144
1145	:grep foo *.c
1146
1147Will search all files with the .c extension for the substring "foo".  The
1148arguments to :grep are passed straight to the "grep" program, so you can use
1149whatever options your "grep" supports.
1150
1151By default, :grep invokes grep with the -n option (show file and line
1152numbers).  You can change this with the 'grepprg' option.  You will need to set
1153'grepprg' if:
1154
1155a)	You are using a program that isn't called "grep"
1156b)	You have to call grep with a full path
1157c)	You want to pass other options automatically (e.g. case insensitive
1158	search.)
1159
1160Once "grep" has executed, Vim parses the results using the 'grepformat'
1161option.  This option works in the same way as the 'errorformat' option - see
1162that for details.  You may need to change 'grepformat' from the default if
1163your grep outputs in a non-standard format, or you are using some other
1164program with a special format.
1165
1166Once the results are parsed, Vim loads the first file containing a match and
1167jumps to the appropriate line, in the same way that it jumps to a compiler
1168error in |quickfix| mode.  You can then use the |:cnext|, |:clist|, etc.
1169commands to see the other matches.
1170
1171
11725.4 Using :grep with id-utils
1173
1174You can set up :grep to work with the GNU id-utils like this: >
1175
1176	:set grepprg=lid\ -Rgrep\ -s
1177	:set grepformat=%f:%l:%m
1178
1179then >
1180	:grep (regexp)
1181
1182works just as you'd expect.
1183(provided you remembered to mkid first :)
1184
1185
11865.5 Browsing source code with :vimgrep or :grep
1187
1188Using the stack of error lists that Vim keeps, you can browse your files to
1189look for functions and the functions they call.  For example, suppose that you
1190have to add an argument to the read_file() function.  You enter this command: >
1191
1192	:vimgrep /\<read_file\>/ *.c
1193
1194You use ":cn" to go along the list of matches and add the argument.  At one
1195place you have to get the new argument from a higher level function msg(), and
1196need to change that one too.  Thus you use: >
1197
1198	:vimgrep /\<msg\>/ *.c
1199
1200While changing the msg() functions, you find another function that needs to
1201get the argument from a higher level.  You can again use ":vimgrep" to find
1202these functions.  Once you are finished with one function, you can use >
1203
1204	:colder
1205
1206to go back to the previous one.
1207
1208This works like browsing a tree: ":vimgrep" goes one level deeper, creating a
1209list of branches.  ":colder" goes back to the previous level.  You can mix
1210this use of ":vimgrep" and "colder" to browse all the locations in a tree-like
1211way.  If you do this consistently, you will find all locations without the
1212need to write down a "todo" list.
1213
1214=============================================================================
12156. Selecting a compiler					*compiler-select*
1216
1217						*:comp* *:compiler* *E666*
1218:comp[iler][!] {name}		Set options to work with compiler {name}.
1219				Without the "!" options are set for the
1220				current buffer.  With "!" global options are
1221				set.
1222				If you use ":compiler foo" in "file.foo" and
1223				then ":compiler! bar" in another buffer, Vim
1224				will keep on using "foo" in "file.foo".
1225				{not available when compiled without the
1226				|+eval| feature}
1227
1228
1229The Vim plugins in the "compiler" directory will set options to use the
1230selected compiler.  For `:compiler` local options are set, for `:compiler!`
1231global options.
1232							*current_compiler*
1233To support older Vim versions, the plugins always use "current_compiler" and
1234not "b:current_compiler".  What the command actually does is the following:
1235
1236- Delete the "current_compiler" and "b:current_compiler" variables.
1237- Define the "CompilerSet" user command.  With "!" it does ":set", without "!"
1238  it does ":setlocal".
1239- Execute ":runtime! compiler/{name}.vim".  The plugins are expected to set
1240  options with "CompilerSet" and set the "current_compiler" variable to the
1241  name of the compiler.
1242- Delete the "CompilerSet" user command.
1243- Set "b:current_compiler" to the value of "current_compiler".
1244- Without "!" the old value of "current_compiler" is restored.
1245
1246
1247For writing a compiler plugin, see |write-compiler-plugin|.
1248
1249
1250GCC					*quickfix-gcc*	*compiler-gcc*
1251
1252There's one variable you can set for the GCC compiler:
1253
1254g:compiler_gcc_ignore_unmatched_lines
1255				Ignore lines that don't match any patterns
1256				defined for GCC.  Useful if output from
1257				commands run from make are generating false
1258				positives.
1259
1260
1261MANX AZTEC C				*quickfix-manx* *compiler-manx*
1262
1263To use Vim with Manx's Aztec C compiler on the Amiga you should do the
1264following:
1265- Set the CCEDIT environment variable with the command: >
1266	mset "CCEDIT=vim -q"
1267- Compile with the -qf option.  If the compiler finds any errors, Vim is
1268  started and the cursor is positioned on the first error.  The error message
1269  will be displayed on the last line.  You can go to other errors with the
1270  commands mentioned above.  You can fix the errors and write the file(s).
1271- If you exit Vim normally the compiler will re-compile the same file.  If you
1272  exit with the :cq command, the compiler will terminate.  Do this if you
1273  cannot fix the error, or if another file needs to be compiled first.
1274
1275There are some restrictions to the Quickfix mode on the Amiga.  The
1276compiler only writes the first 25 errors to the errorfile (Manx's
1277documentation does not say how to get more).  If you want to find the others,
1278you will have to fix a few errors and exit the editor.  After recompiling,
1279up to 25 remaining errors will be found.
1280
1281If Vim was started from the compiler, the :sh and some :!  commands will not
1282work, because Vim is then running in the same process as the compiler and
1283stdin (standard input) will not be interactive.
1284
1285
1286PERL					*quickfix-perl* *compiler-perl*
1287
1288The Perl compiler plugin doesn't actually compile, but invokes Perl's internal
1289syntax checking feature and parses the output for possible errors so you can
1290correct them in quick-fix mode.
1291
1292Warnings are forced regardless of "no warnings" or "$^W = 0" within the file
1293being checked.  To disable this set g:perl_compiler_force_warnings to a zero
1294value.  For example: >
1295	let g:perl_compiler_force_warnings = 0
1296
1297
1298PYUNIT COMPILER						*compiler-pyunit*
1299
1300This is not actually a compiler, but a unit testing framework for the
1301Python language.  It is included into standard Python distribution
1302starting from version 2.0.  For older versions, you can get it from
1303http://pyunit.sourceforge.net.
1304
1305When you run your tests with the help of the framework, possible errors
1306are parsed by Vim and presented for you in quick-fix mode.
1307
1308Unfortunately, there is no standard way to run the tests.
1309The alltests.py script seems to be used quite often, that's all.
1310Useful values for the 'makeprg' options therefore are:
1311 setlocal makeprg=./alltests.py " Run a testsuite
1312 setlocal makeprg=python\ %:S   " Run a single testcase
1313
1314Also see http://vim.sourceforge.net/tip_view.php?tip_id=280.
1315
1316
1317TEX COMPILER						*compiler-tex*
1318
1319Included in the distribution compiler for TeX ($VIMRUNTIME/compiler/tex.vim)
1320uses make command if possible.  If the compiler finds a file named "Makefile"
1321or "makefile" in the current directory, it supposes that you want to process
1322your *TeX files with make, and the makefile does the right work.  In this case
1323compiler sets 'errorformat' for *TeX output and leaves 'makeprg' untouched.  If
1324neither "Makefile" nor "makefile" is found, the compiler will not use make.
1325You can force the compiler to ignore makefiles by defining
1326b:tex_ignore_makefile or g:tex_ignore_makefile variable (they are checked for
1327existence only).
1328
1329If the compiler chose not to use make, it needs to choose a right program for
1330processing your input.  If b:tex_flavor or g:tex_flavor (in this precedence)
1331variable exists, it defines TeX flavor for :make (actually, this is the name
1332of executed command), and if both variables do not exist, it defaults to
1333"latex".  For example, while editing chapter2.tex \input-ed from mypaper.tex
1334written in AMS-TeX: >
1335
1336	:let b:tex_flavor = 'amstex'
1337	:compiler tex
1338<	[editing...] >
1339	:make mypaper
1340
1341Note that you must specify a name of the file to process as an argument (to
1342process the right file when editing \input-ed or \include-ed file; portable
1343solution for substituting % for no arguments is welcome).  This is not in the
1344semantics of make, where you specify a target, not source, but you may specify
1345filename without extension ".tex" and mean this as "make filename.dvi or
1346filename.pdf or filename.some_result_extension according to compiler".
1347
1348Note: tex command line syntax is set to usable both for MikTeX (suggestion
1349by Srinath Avadhanula) and teTeX (checked by Artem Chuprina).  Suggestion
1350from |errorformat-LaTeX| is too complex to keep it working for different
1351shells and OSes and also does not allow to use other available TeX options,
1352if any.  If your TeX doesn't support "-interaction=nonstopmode", please
1353report it with different means to express \nonstopmode from the command line.
1354
1355=============================================================================
13567. The error format					*error-file-format*
1357
1358					*errorformat* *E372* *E373* *E374*
1359						*E375* *E376* *E377* *E378*
1360The 'errorformat' option specifies a list of formats that are recognized.  The
1361first format that matches with an error message is used.  You can add several
1362formats for different messages your compiler produces, or even entries for
1363multiple compilers.  See |efm-entries|.
1364
1365Each entry in 'errorformat' is a scanf-like string that describes the format.
1366First, you need to know how scanf works.  Look in the documentation of your
1367C compiler.  Below you find the % items that Vim understands.  Others are
1368invalid.
1369
1370Special characters in 'errorformat' are comma and backslash.  See
1371|efm-entries| for how to deal with them.  Note that a literal "%" is matched
1372by "%%", thus it is not escaped with a backslash.
1373Keep in mind that in the `:make` and `:grep` output all NUL characters are
1374replaced with SOH (0x01).
1375
1376Note: By default the difference between upper and lowercase is ignored.  If
1377you want to match case, add "\C" to the pattern |/\C|.
1378
1379Vim will read lines of any length, but only the first 4095 bytes are used, the
1380rest is ignored.  Items can only be 1023 bytes long.
1381
1382
1383Basic items
1384
1385	%f		file name (finds a string)
1386	%o		module name (finds a string)
1387	%l		line number (finds a number)
1388	%c		column number (finds a number representing character
1389			column of the error, byte index, a <tab> is 1
1390			character column)
1391	%v		virtual column number (finds a number representing
1392			screen column of the error (1 <tab> == 8 screen
1393			columns))
1394	%t		error type (finds a single character):
1395			    e - error message
1396			    w - warning message
1397			    i - info message
1398			    n - note message
1399	%n		error number (finds a number)
1400	%m		error message (finds a string)
1401	%r		matches the "rest" of a single-line file message %O/P/Q
1402	%p		pointer line (finds a sequence of '-', '.', ' ' or
1403			tabs and uses the length for the column number)
1404	%*{conv}	any scanf non-assignable conversion
1405	%%		the single '%' character
1406	%s		search text (finds a string)
1407
1408The "%f" conversion may depend on the current 'isfname' setting.  "~/" is
1409expanded to the home directory and environment variables are expanded.
1410
1411The "%f" and "%m" conversions have to detect the end of the string.  This
1412normally happens by matching following characters and items.  When nothing is
1413following the rest of the line is matched.  If "%f" is followed by a '%' or a
1414backslash, it will look for a sequence of 'isfname' characters.
1415
1416On MS-Windows a leading "C:" will be included in "%f", even when using "%f:".
1417This means that a file name which is a single alphabetical letter will not be
1418detected.
1419
1420The "%p" conversion is normally followed by a "^".  It's used for compilers
1421that output a line like: >
1422	    ^
1423or >
1424   ---------^
1425to indicate the column of the error.  This is to be used in a multi-line error
1426message.  See |errorformat-javac| for a  useful example.
1427
1428The "%s" conversion specifies the text to search for, to locate the error line.
1429The text is used as a literal string.  The anchors "^" and "$" are added to
1430the text to locate the error line exactly matching the search text and the
1431text is prefixed with the "\V" atom to make it "very nomagic".  The "%s"
1432conversion can be used to locate lines without a line number in the error
1433output.  Like the output of the "grep" shell command.
1434When the pattern is present the line number will not be used.
1435
1436The "%o" conversion specifies the module name in quickfix entry.  If present
1437it will be used in quickfix error window instead of the filename.  The module
1438name is used only for displaying purposes, the file name is used when jumping
1439to the file.
1440
1441Changing directory
1442
1443The following uppercase conversion characters specify the type of special
1444format strings.  At most one of them may be given as a prefix at the beginning
1445of a single comma-separated format pattern.
1446Some compilers produce messages that consist of directory names that have to
1447be prepended to each file name read by %f (example: GNU make).  The following
1448codes can be used to scan these directory names; they will be stored in an
1449internal directory stack.					*E379*
1450	%D		"enter directory" format string; expects a following
1451			  %f that finds the directory name
1452	%X		"leave directory" format string; expects following %f
1453
1454When defining an "enter directory" or "leave directory" format, the "%D" or
1455"%X" has to be given at the start of that substring.  Vim tracks the directory
1456changes and prepends the current directory to each erroneous file found with a
1457relative path.  See |quickfix-directory-stack| for details, tips and
1458limitations.
1459
1460
1461Multi-line messages				*errorformat-multi-line*
1462
1463It is possible to read the output of programs that produce multi-line
1464messages, i.e. error strings that consume more than one line.  Possible
1465prefixes are:
1466	%E		start of a multi-line error message
1467	%W		start of a multi-line warning message
1468	%I		start of a multi-line informational message
1469	%N		start of a multi-line note message
1470	%A		start of a multi-line message (unspecified type)
1471	%>		for next line start with current pattern again |efm-%>|
1472	%C		continuation of a multi-line message
1473	%Z		end of a multi-line message
1474These can be used with '+' and '-', see |efm-ignore| below.
1475
1476Using "\n" in the pattern won't work to match multi-line messages.
1477
1478Example: Your compiler happens to write out errors in the following format
1479(leading line numbers not being part of the actual output):
1480
1481     1	Error 275 ~
1482     2	line 42 ~
1483     3	column 3 ~
1484     4	' ' expected after '--' ~
1485
1486The appropriate error format string has to look like this: >
1487   :set efm=%EError\ %n,%Cline\ %l,%Ccolumn\ %c,%Z%m
1488
1489And the |:clist| error message generated for this error is:
1490
1491 1:42 col 3 error 275:  ' ' expected after '--'
1492
1493Another example: Think of a Python interpreter that produces the following
1494error message (line numbers are not part of the actual output):
1495
1496     1	==============================================================
1497     2	FAIL: testGetTypeIdCachesResult (dbfacadeTest.DjsDBFacadeTest)
1498     3	--------------------------------------------------------------
1499     4	Traceback (most recent call last):
1500     5	  File "unittests/dbfacadeTest.py", line 89, in testFoo
1501     6	    self.assertEquals(34, dtid)
1502     7	  File "/usr/lib/python2.2/unittest.py", line 286, in
1503     8	 failUnlessEqual
1504     9	    raise self.failureException, \
1505    10	AssertionError: 34 != 33
1506    11
1507    12	--------------------------------------------------------------
1508    13	Ran 27 tests in 0.063s
1509
1510Say you want |:clist| write the relevant information of this message only,
1511namely:
1512 5 unittests/dbfacadeTest.py:89:  AssertionError: 34 != 33
1513
1514Then the error format string could be defined as follows: >
1515  :set efm=%C\ %.%#,%A\ \ File\ \"%f\"\\,\ line\ %l%.%#,%Z%[%^\ ]%\\@=%m
1516
1517Note that the %C string is given before the %A here: since the expression
1518' %.%#' (which stands for the regular expression ' .*') matches every line
1519starting with a space, followed by any characters to the end of the line,
1520it also hides line 7 which would trigger a separate error message otherwise.
1521Error format strings are always parsed pattern by pattern until the first
1522match occurs.
1523							*efm-%>*
1524The %> item can be used to avoid trying patterns that appear earlier in
1525'errorformat'.  This is useful for patterns that match just about anything.
1526For example, if the error looks like this:
1527
1528	Error in line 123 of foo.c: ~
1529	unknown variable "i" ~
1530
1531This can be found with: >
1532	:set efm=xxx,%E%>Error in line %l of %f:,%Z%m
1533Where "xxx" has a pattern that would also match the second line.
1534
1535Important: There is no memory of what part of the errorformat matched before;
1536every line in the error file gets a complete new run through the error format
1537lines.  For example, if one has: >
1538  setlocal efm=aa,bb,cc,dd,ee
1539Where aa, bb, etc. are error format strings.  Each line of the error file will
1540be matched to the pattern aa, then bb, then cc, etc.  Just because cc matched
1541the previous error line does _not_ mean that dd will be tried first on the
1542current line, even if cc and dd are multi-line errorformat strings.
1543
1544
1545
1546Separate file name			*errorformat-separate-filename*
1547
1548These prefixes are useful if the file name is given once and multiple messages
1549follow that refer to this file name.
1550	%O		single-line file message: overread the matched part
1551	%P		single-line file message: push file %f onto the stack
1552	%Q		single-line file message: pop the last file from stack
1553
1554Example: Given a compiler that produces the following error logfile (without
1555leading line numbers):
1556
1557     1	[a1.tt]
1558     2	(1,17)  error: ';' missing
1559     3	(21,2)  warning: variable 'z' not defined
1560     4	(67,3)  error: end of file found before string ended
1561     5
1562     6	[a2.tt]
1563     7
1564     8	[a3.tt]
1565     9	NEW compiler v1.1
1566    10	(2,2)   warning: variable 'x' not defined
1567    11	(67,3)  warning: 's' already defined
1568
1569This logfile lists several messages for each file enclosed in [...] which are
1570properly parsed by an error format like this: >
1571  :set efm=%+P[%f],(%l\\,%c)%*[\ ]%t%*[^:]:\ %m,%-Q
1572
1573A call of |:clist| writes them accordingly with their correct filenames:
1574
1575  2 a1.tt:1 col 17 error: ';' missing
1576  3 a1.tt:21 col 2 warning: variable 'z' not defined
1577  4 a1.tt:67 col 3 error: end of file found before string ended
1578  8 a3.tt:2 col 2 warning: variable 'x' not defined
1579  9 a3.tt:67 col 3 warning: 's' already defined
1580
1581Unlike the other prefixes that all match against whole lines, %P, %Q and %O
1582can be used to match several patterns in the same line.  Thus it is possible
1583to parse even nested files like in the following line:
1584  {"file1" {"file2" error1} error2 {"file3" error3 {"file4" error4 error5}}}
1585The %O then parses over strings that do not contain any push/pop file name
1586information.  See |errorformat-LaTeX| for an extended example.
1587
1588
1589Ignoring and using whole messages			*efm-ignore*
1590
1591The codes '+' or '-' can be combined with the uppercase codes above; in that
1592case they have to precede the letter, e.g. '%+A' or '%-G':
1593	%-		do not include the matching multi-line in any output
1594	%+		include the whole matching line in the %m error string
1595
1596One prefix is only useful in combination with '+' or '-', namely %G.  It parses
1597over lines containing general information like compiler version strings or
1598other headers that can be skipped.
1599	%-G		ignore this message
1600	%+G		general message
1601
1602
1603Pattern matching
1604
1605The scanf()-like "%*[]" notation is supported for backward-compatibility
1606with previous versions of Vim.  However, it is also possible to specify
1607(nearly) any Vim supported regular expression in format strings.
1608Since meta characters of the regular expression language can be part of
1609ordinary matching strings or file names (and therefore internally have to
1610be escaped), meta symbols have to be written with leading '%':
1611	%\		The single '\' character.  Note that this has to be
1612			escaped ("%\\") in ":set errorformat=" definitions.
1613	%.		The single '.' character.
1614	%#		The single '*'(!) character.
1615	%^		The single '^' character.  Note that this is not
1616			useful, the pattern already matches start of line.
1617	%$		The single '$' character.  Note that this is not
1618			useful, the pattern already matches end of line.
1619	%[		The single '[' character for a [] character range.
1620	%~		The single '~' character.
1621When using character classes in expressions (see |/\i| for an overview),
1622terms containing the "\+" quantifier can be written in the scanf() "%*"
1623notation.  Example: "%\\d%\\+" ("\d\+", "any number") is equivalent to "%*\\d".
1624Important note: The \(...\) grouping of sub-matches can not be used in format
1625specifications because it is reserved for internal conversions.
1626
1627
1628Multiple entries in 'errorformat'			*efm-entries*
1629
1630To be able to detect output from several compilers, several format patterns
1631may be put in 'errorformat', separated by commas (note: blanks after the comma
1632are ignored).  The first pattern that has a complete match is used.  If no
1633match is found, matching parts from the last one will be used, although the
1634file name is removed and the error message is set to the whole message.  If
1635there is a pattern that may match output from several compilers (but not in a
1636right way), put it after one that is more restrictive.
1637
1638To include a comma in a pattern precede it with a backslash (you have to type
1639two in a ":set" command).  To include a backslash itself give two backslashes
1640(you have to type four in a ":set" command).  You also need to put a backslash
1641before a space for ":set".
1642
1643
1644Valid matches						*quickfix-valid*
1645
1646If a line does not completely match one of the entries in 'errorformat', the
1647whole line is put in the error message and the entry is marked "not valid"
1648These lines are skipped with the ":cn" and ":cp" commands (unless there is
1649no valid line at all).  You can use ":cl!" to display all the error messages.
1650
1651If the error format does not contain a file name Vim cannot switch to the
1652correct file.  You will have to do this by hand.
1653
1654
1655Examples
1656
1657The format of the file from the Amiga Aztec compiler is:
1658
1659	filename>linenumber:columnnumber:errortype:errornumber:errormessage
1660
1661	filename	name of the file in which the error was detected
1662	linenumber	line number where the error was detected
1663	columnnumber	column number where the error was detected
1664	errortype	type of the error, normally a single 'E' or 'W'
1665	errornumber	number of the error (for lookup in the manual)
1666	errormessage	description of the error
1667
1668This can be matched with this 'errorformat' entry:
1669	%f>%l:%c:%t:%n:%m
1670
1671Some examples for C compilers that produce single-line error outputs:
1672%f:%l:\ %t%*[^0123456789]%n:\ %m	for Manx/Aztec C error messages
1673					(scanf() doesn't understand [0-9])
1674%f\ %l\ %t%*[^0-9]%n:\ %m		for SAS C
1675\"%f\"\\,%*[^0-9]%l:\ %m		for generic C compilers
1676%f:%l:\ %m				for GCC
1677%f:%l:\ %m,%Dgmake[%*\\d]:\ Entering\ directory\ `%f',
1678%Dgmake[%*\\d]:\ Leaving\ directory\ `%f'
1679					for GCC with gmake (concat the lines!)
1680%f(%l)\ :\ %*[^:]:\ %m			old SCO C compiler (pre-OS5)
1681%f(%l)\ :\ %t%*[^0-9]%n:\ %m		idem, with error type and number
1682%f:%l:\ %m,In\ file\ included\ from\ %f:%l:,\^I\^Ifrom\ %f:%l%m
1683					for GCC, with some extras
1684
1685Extended examples for the handling of multi-line messages are given below,
1686see |errorformat-Jikes| and |errorformat-LaTeX|.
1687
1688Note the backslash in front of a space and double quote.  It is required for
1689the :set command.  There are two backslashes in front of a comma, one for the
1690:set command and one to avoid recognizing the comma as a separator of error
1691formats.
1692
1693
1694Filtering messages
1695
1696If you have a compiler that produces error messages that do not fit in the
1697format string, you could write a program that translates the error messages
1698into this format.  You can use this program with the ":make" command by
1699changing the 'makeprg' option.  For example: >
1700   :set mp=make\ \\\|&\ error_filter
1701The backslashes before the pipe character are required to avoid it to be
1702recognized as a command separator.  The backslash before each space is
1703required for the set command.
1704
1705=============================================================================
17068. The directory stack				*quickfix-directory-stack*
1707
1708Quickfix maintains a stack for saving all used directories parsed from the
1709make output.  For GNU-make this is rather simple, as it always prints the
1710absolute path of all directories it enters and leaves.  Regardless if this is
1711done via a 'cd' command in the makefile or with the parameter "-C dir" (change
1712to directory before reading the makefile).  It may be useful to use the switch
1713"-w" to force GNU-make to print out the working directory before and after
1714processing.
1715
1716Maintaining the correct directory is more complicated if you don't use
1717GNU-make.  AIX-make for example doesn't print any information about its
1718working directory.  Then you need to enhance the makefile.  In the makefile of
1719LessTif there is a command which echoes "Making {target} in {dir}".  The
1720special problem here is that it doesn't print information on leaving the
1721directory and that it doesn't print the absolute path.
1722
1723To solve the problem with relative paths and missing "leave directory"
1724messages Vim uses the following algorithm:
1725
17261) Check if the given directory is a subdirectory of the current directory.
1727   If this is true, store it as the current directory.
17282) If it is not a subdir of the current directory, try if this is a
1729   subdirectory of one of the upper directories.
17303) If the directory still isn't found, it is assumed to be a subdirectory
1731   of Vim's current directory.
1732
1733Additionally it is checked for every file, if it really exists in the
1734identified directory.  If not, it is searched in all other directories of the
1735directory stack (NOT the directory subtree!).  If it is still not found, it is
1736assumed that it is in Vim's current directory.
1737
1738There are limitations in this algorithm.  These examples assume that make just
1739prints information about entering a directory in the form "Making all in dir".
1740
17411) Assume you have following directories and files:
1742   ./dir1
1743   ./dir1/file1.c
1744   ./file1.c
1745
1746   If make processes the directory "./dir1" before the current directory and
1747   there is an error in the file "./file1.c", you will end up with the file
1748   "./dir1/file.c" loaded by Vim.
1749
1750   This can only be solved with a "leave directory" message.
1751
17522) Assume you have following directories and files:
1753   ./dir1
1754   ./dir1/dir2
1755   ./dir2
1756
1757   You get the following:
1758
1759   Make output			  Directory interpreted by Vim
1760   ------------------------	  ----------------------------
1761   Making all in dir1		  ./dir1
1762   Making all in dir2		  ./dir1/dir2
1763   Making all in dir2		  ./dir1/dir2
1764
1765   This can be solved by printing absolute directories in the "enter directory"
1766   message or by printing "leave directory" messages.
1767
1768To avoid this problem, ensure to print absolute directory names and "leave
1769directory" messages.
1770
1771Examples for Makefiles:
1772
1773Unix:
1774    libs:
1775	    for dn in $(LIBDIRS); do				\
1776		(cd $$dn; echo "Entering dir '$$(pwd)'"; make); \
1777		echo "Leaving dir";				\
1778	    done
1779
1780Add
1781    %DEntering\ dir\ '%f',%XLeaving\ dir
1782to your 'errorformat' to handle the above output.
1783
1784Note that Vim doesn't check if the directory name in a "leave directory"
1785messages is the current directory.  This is why you could just use the message
1786"Leaving dir".
1787
1788=============================================================================
17899. Specific error file formats			*errorformats*
1790
1791						*errorformat-Jikes*
1792Jikes(TM), a source-to-bytecode Java compiler published by IBM Research,
1793produces simple multi-line error messages.
1794
1795An 'errorformat' string matching the produced messages is shown below.
1796The following lines can be placed in the user's |vimrc| to overwrite Vim's
1797recognized default formats, or see |:set+=| how to install this format
1798additionally to the default. >
1799
1800  :set efm=%A%f:%l:%c:%*\\d:%*\\d:,
1801	\%C%*\\s%trror:%m,
1802	\%+C%*[^:]%trror:%m,
1803	\%C%*\\s%tarning:%m,
1804	\%C%m
1805<
1806Jikes(TM) produces a single-line error message when invoked with the option
1807"+E", and can be matched with the following: >
1808
1809  :setl efm=%f:%l:%v:%*\\d:%*\\d:%*\\s%m
1810<
1811						*errorformat-javac*
1812This 'errorformat' has been reported to work well for javac, which outputs a
1813line with "^" to indicate the column of the error: >
1814  :setl efm=%A%f:%l:\ %m,%-Z%p^,%-C%.%#
1815or: >
1816  :setl efm=%A%f:%l:\ %m,%+Z%p^,%+C%.%#,%-G%.%#
1817<
1818Here is an alternative from Michael F. Lamb for Unix that filters the errors
1819first: >
1820  :setl errorformat=%Z%f:%l:\ %m,%A%p^,%-G%*[^sl]%.%#
1821  :setl makeprg=javac\ %:S\ 2>&1\ \\\|\ vim-javac-filter
1822
1823You need to put the following in "vim-javac-filter" somewhere in your path
1824(e.g., in ~/bin) and make it executable: >
1825   #!/bin/sed -f
1826   /\^$/s/\t/\ /g;/:[0-9]\+:/{h;d};/^[ \t]*\^/G;
1827
1828In English, that sed script:
1829- Changes single tabs to single spaces and
1830- Moves the line with the filename, line number, error message to just after
1831  the pointer line. That way, the unused error text between doesn't break
1832  vim's notion of a "multi-line message" and also doesn't force us to include
1833  it as a "continuation of a multi-line message."
1834
1835						*errorformat-ant*
1836For ant (http://jakarta.apache.org/) the above errorformat has to be modified
1837to honour the leading [javac] in front of each javac output line: >
1838  :set efm=%A\ %#[javac]\ %f:%l:\ %m,%-Z\ %#[javac]\ %p^,%-C%.%#
1839
1840The 'errorformat' can also be configured to handle ant together with either
1841javac or jikes.  If you're using jikes, you should tell ant to use jikes' +E
1842command line switch which forces jikes to generate one-line error messages.
1843This is what the second line (of a build.xml file) below does: >
1844  <property name = "build.compiler"       value = "jikes"/>
1845  <property name = "build.compiler.emacs" value = "true"/>
1846
1847The 'errorformat' which handles ant with both javac and jikes is: >
1848  :set efm=\ %#[javac]\ %#%f:%l:%c:%*\\d:%*\\d:\ %t%[%^:]%#:%m,
1849	   \%A\ %#[javac]\ %f:%l:\ %m,%-Z\ %#[javac]\ %p^,%-C%.%#
1850<
1851						*errorformat-jade*
1852parsing jade (see http://www.jclark.com/) errors is simple: >
1853  :set efm=jade:%f:%l:%c:%t:%m
1854<
1855						*errorformat-LaTeX*
1856The following is an example how an 'errorformat' string can be specified
1857for the (La)TeX typesetting system which displays error messages over
1858multiple lines.  The output of ":clist" and ":cc" etc. commands displays
1859multi-lines in a single line, leading white space is removed.
1860It should be easy to adopt the above LaTeX errorformat to any compiler output
1861consisting of multi-line errors.
1862
1863The commands can be placed in a |vimrc| file or some other Vim script file,
1864e.g. a script containing LaTeX related stuff which is loaded only when editing
1865LaTeX sources.
1866Make sure to copy all lines of the example (in the given order), afterwards
1867remove the comment lines.  For the '\' notation at the start of some lines see
1868|line-continuation|.
1869
1870		First prepare 'makeprg' such that LaTeX will report multiple
1871		errors; do not stop when the first error has occurred: >
1872 :set makeprg=latex\ \\\\nonstopmode\ \\\\input\\{$*}
1873<
1874		Start of multi-line error messages: >
1875 :set efm=%E!\ LaTeX\ %trror:\ %m,
1876	\%E!\ %m,
1877<		Start of multi-line warning messages; the first two also
1878		include the line number.  Meaning of some regular expressions:
1879		  - "%.%#"  (".*")   matches a (possibly empty) string
1880		  - "%*\\d" ("\d\+") matches a number >
1881	\%+WLaTeX\ %.%#Warning:\ %.%#line\ %l%.%#,
1882	\%+W%.%#\ at\ lines\ %l--%*\\d,
1883	\%WLaTeX\ %.%#Warning:\ %m,
1884<		Possible continuations of error/warning messages; the first
1885		one also includes the line number: >
1886	\%Cl.%l\ %m,
1887	\%+C\ \ %m.,
1888	\%+C%.%#-%.%#,
1889	\%+C%.%#[]%.%#,
1890	\%+C[]%.%#,
1891	\%+C%.%#%[{}\\]%.%#,
1892	\%+C<%.%#>%.%#,
1893	\%C\ \ %m,
1894<		Lines that match the following patterns do not contain any
1895		important information; do not include them in messages: >
1896	\%-GSee\ the\ LaTeX%m,
1897	\%-GType\ \ H\ <return>%m,
1898	\%-G\ ...%.%#,
1899	\%-G%.%#\ (C)\ %.%#,
1900	\%-G(see\ the\ transcript%.%#),
1901<		Generally exclude any empty or whitespace-only line from
1902		being displayed: >
1903	\%-G\\s%#,
1904<		The LaTeX output log does not specify the names of erroneous
1905		source files per line; rather they are given globally,
1906		enclosed in parentheses.
1907		The following patterns try to match these names and store
1908		them in an internal stack.  The patterns possibly scan over
1909		the same input line (one after another), the trailing "%r"
1910		conversion indicates the "rest" of the line that will be
1911		parsed in the next go until the end of line is reached.
1912
1913		Overread a file name enclosed in '('...')'; do not push it
1914		on a stack since the file apparently does not contain any
1915		error: >
1916	\%+O(%f)%r,
1917<		Push a file name onto the stack.  The name is given after '(': >
1918	\%+P(%f%r,
1919	\%+P\ %\\=(%f%r,
1920	\%+P%*[^()](%f%r,
1921	\%+P[%\\d%[^()]%#(%f%r,
1922<		Pop the last stored file name when a ')' is scanned: >
1923	\%+Q)%r,
1924	\%+Q%*[^()])%r,
1925	\%+Q[%\\d%*[^()])%r
1926
1927Note that in some cases file names in the LaTeX output log cannot be parsed
1928properly.  The parser might have been messed up by unbalanced parentheses
1929then.  The above example tries to catch the most relevant cases only.
1930You can customize the given setting to suit your own purposes, for example,
1931all the annoying "Overfull ..." warnings could be excluded from being
1932recognized as an error.
1933Alternatively to filtering the LaTeX compiler output, it is also possible
1934to directly read the *.log file that is produced by the [La]TeX compiler.
1935This contains even more useful information about possible error causes.
1936However, to properly parse such a complex file, an external filter should
1937be used.  See the description further above how to make such a filter known
1938by Vim.
1939
1940						*errorformat-Perl*
1941In $VIMRUNTIME/tools you can find the efm_perl.pl script, which filters Perl
1942error messages into a format that quickfix mode will understand.  See the
1943start of the file about how to use it.  (This script is deprecated, see
1944|compiler-perl|.)
1945
1946=============================================================================
194710. Customizing the quickfix window		*quickfix-window-function*
1948
1949The default format for the lines displayed in the quickfix window and location
1950list window is:
1951
1952    <filename>|<lnum> col <col>|<text>
1953
1954The values displayed in each line correspond to the "bufnr", "lnum", "col" and
1955"text" fields returned by the |getqflist()| function.
1956
1957For some quickfix/location lists, the displayed text needs to be customized.
1958For example, if only the filename is present for a quickfix entry, then the
1959two "|" field separator characters after the filename are not needed.  Another
1960use case is to customize the path displayed for a filename. By default, the
1961complete path (which may be too long) is displayed for files which are not
1962under the current directory tree. The file path may need to be simplified to a
1963common parent directory.
1964
1965The displayed text can be customized by setting the 'quickfixtextfunc' option
1966to a Vim function.  This function will be called with a dict argument and
1967should return a List of strings to be displayed in the quickfix or location
1968list window. The dict argument will have the following fields:
1969
1970    quickfix	set to 1 when called for a quickfix list and 0 when called for
1971		a location list.
1972    winid	for a location list, set to the id of the window with the
1973		location list. For a quickfix list, set to 0. Can be used in
1974		getloclist() to get the location list entry.
1975    id		quickfix or location list identifier
1976    start_idx	index of the first entry for which text should be returned
1977    end_idx	index of the last entry for which text should be returned
1978
1979The function should return a single line of text to display in the quickfix
1980window for each entry from start_idx to end_idx. The function can obtain
1981information about the entries using the |getqflist()| function and specifying
1982the quickfix list identifier "id". For a location list, getloclist() function
1983can be used with the 'winid' argument. If an empty list is returned, then the
1984default format is used to display all the entries. If an item in the returned
1985list is an empty string, then the default format is used to display the
1986corresponding entry.
1987
1988If a quickfix or location list specific customization is needed, then the
1989'quickfixtextfunc' attribute of the list can be set using the |setqflist()| or
1990|setloclist()| function. This overrides the global 'quickfixtextfunc' option.
1991
1992The example below displays the list of old files (|v:oldfiles|) in a quickfix
1993window. As there is no line, column number and error text information
1994associated with each entry, the 'quickfixtextfunc' function returns only the
1995filename.
1996Example: >
1997    " create a quickfix list from v:oldfiles
1998    call setqflist([], ' ', {'lines' : v:oldfiles, 'efm' : '%f',
1999					\ 'quickfixtextfunc' : 'QfOldFiles'})
2000    func QfOldFiles(info)
2001	" get information about a range of quickfix entries
2002	let items = getqflist({'id' : a:info.id, 'items' : 1}).items
2003	let l = []
2004	for idx in range(a:info.start_idx - 1, a:info.end_idx - 1)
2005	    " use the simplified file name
2006	  call add(l, fnamemodify(bufname(items[idx].bufnr), ':p:.'))
2007	endfor
2008	return l
2009    endfunc
2010<
2011
2012 vim:tw=78:ts=8:noet:ft=help:norl:
2013