1CTAGS-FAQ(7) Universal Ctags CTAGS-FAQ(7)
2
6 ctags-faq - Universal Ctags FAQ
7 This is the Universal Ctags FAQ (Frequently-Asked Questions). It is
9 based on Exuberant Ctags FAQ
10 Contents
12 • DESCRIPTION
13 • What is the difference between Universal Ctags and Exuberant Ctags?
15 • How can I avoid having to specify my favorite option every time?
17 • What are these strange bits of text beginning with ;" which follow
19 many of the lines in the tag file?
20 • Why can't I jump to class::member?
22 • Why do I end up on the wrong line when I jump to a tag?
24 • How do I jump to the tag I want instead of the wrong one by the
26 same name?
27 • How can I locate all references to a specific function or variable?
29 • Why does appending tags to a tag file tag so long?
31 • How should I set up tag files for a multi-level directory hierar‐
33 chy?
34 • Does Universal Ctags support Unicode file names?
36 • Why does zsh cause "zsh: no matches found" error?
38 • SEE ALSO
40 • AUTHOR
42
44 What is the difference between Universal Ctags and Exuberant Ctags?
45 Universal Ctags is an unofficial fork of Exuberant Ctags. The differ‐
46 ences are summarized in ctags-incompatibilities(7) man page.
47 The most notable one is that Universal Ctags doesn't read ~/.ctags
49 file. Instead, it reads *.ctags under ~/.ctags.d directory.
50 How can I avoid having to specify my favorite option every time?
52 Either by setting the environment variable CTAGS to your custom op‐
53 tions, or putting them into a ~/.ctags.d/anyname.ctags file in your
54 home directory.
55 What are these strange bits of text beginning with ;" which follow many of
57 the lines in the tag file?
58 These are extension flags. They are added in order to provide extra in‐
59 formation about the tag that may be utilized by the editor in order to
60 more intelligently handle tags. They are appended to the EX command
61 part of the tag line in a manner that provides backwards compatibility
62 with existing implementations of the Vi editor. The semicolon is an EX
63 command separator and the double quote begins an EX comment. Thus, the
64 extension flags appear as an EX comment and should be ignored by the
65 editor when it processes the EX command.
66 Some non-vi editors, however, implement only the bare minimum of EX
68 commands in order to process the search command or line number in the
69 third field of the tag file. If you encounter this problem, use the op‐
70 tion --format=1 to generate a tag file without these extensions (remem‐
71 ber that you can set the CTAGS environment variable to any default ar‐
72 guments you wish to supply). Then ask the supplier of your editor to
73 implement handling of this feature of EX commands.
74 Why can't I jump to class::member?
76 Because, by default, ctags only generates tags for the separate identi‐
77 fiers found in the source files. If you specify the --extra=+q option,
78 then ctags will also generate a second, class-qualified tag for each
79 class member (data and function/method) in the form class::member for
80 C++, and in the form class.method for Eiffel and Java.
81 Why do I end up on the wrong line when I jump to a tag?
83 By default, ctags encodes the line number in the file where macro (#de‐
84 fine) tags are found. This was done to remain compatible with the orig‐
85 inal UNIX version of ctags. If you change the file containing the tag
86 without rebuilding the tag file, the location of tag in the tag file
87 may no longer match the current location.
88 In order to avoid this problem, you can specify the option --excmd=p,
90 which causes ctags to use a search pattern to locate macro tags. I have
91 never uncovered the reason why the original UNIX ctags used line num‐
92 bers exclusively for macro tags, but have so far resisted changing the
93 default behavior of Exuberant (and Universal) Ctags to behave differ‐
94 ently.
95 How do I jump to the tag I want instead of the wrong one by the same name?
97 A tag file is simple a list of tag names and where to find them. If
98 there are duplicate entries, you often end up going to the wrong one
99 because the tag file is sorted and your editor locates the first one in
100 the tag file.
101 Standard Vi provides no facilities to alter this behavior. However, Vim
103 has some nice features to minimize this problem, primarily by examining
104 all matches and choosing the best one under the circumstances. Vim also
105 provides commands which allow for selection of the desired matching
106 tag.
107 How can I locate all references to a specific function or variable?
109 There are several packages already available which provide this capa‐
110 bility. Namely, these are: GLOBAL source code tag system, GNU
111 id-utils, cscope, and cflow. As of this writing, they can be found in
112 the following locations:
113 • GLOBAL: http://www.gnu.org/software/global
115 • id-utils: http://www.gnu.org/software/idutils/idutils.html
117 • cscope: http://cscope.sourceforge.net
119 • cflow: ftp://www.ibiblio.org/pub/Linux/devel/lang/c
121 Why does appending tags to a tag file tag so long?
123 Sometimes, in an attempt to build a global tag file for all source
124 files in a large source tree of many directories, someone will make an
125 attempt to run ctags in append (-a) mode on every directory in the hi‐
126 erarchy. Each time ctags is invoked, its default behavior is to sort
127 the tag file once the tags for that execution have been added. As the
128 cumulative tag file grows, the sort time increases arithmetically.
129 The best way to avoid this problem (and the most efficient) is to make
131 use of the --recurse (or -R) option of ctags by executing the following
132 command in the root of the directory hierarchy (thus running ctags only
133 once):
134 ctags -R
136 If you really insist on running ctags separately on each directory, you
138 can avoid the sort pass each time by specifying the option --sort=no.
139 Once the tag file is completely built, use the sort command to manually
140 sort the final tag file, or let the final invocation of ctags sort the
141 file.
142 How should I set up tag files for a multi-level directory hierarchy?
144 There are a few ways of approaching this:
145 1. A local tag file in each directory containing only the tags for
147 source files in that directory.
148 2. One single big, global tag file present in the root directory of
150 your hierarchy, containing all tags present in all source files in
151 the hierarchy.
152 3. A local tag file in each directory containing only the tags for
154 source files in that directory, in addition to one single global tag
155 file present in the root directory of your hierarchy, containing all
156 non-static tags present in all source files in the hierarchy.
157 4. A local tag file in each directory of the hierarchy, each one con‐
159 taining all tags present in source files in that directory and all
160 non-static tags in every directory below it (note that this implies
161 also having one big tag file in the root directory of the hierar‐
162 chy).
163 Each of these approaches has its own set of advantages and disadvan‐
165 tages, depending upon your particular conditions. Which approach is
166 deemed best depends upon the following factors:
167 A. The ability of your editor to use multiple tag files.
169 If your editor cannot make use of multiple tag files (original vi
171 implementations could not), then one large tag file is the only way
172 to go if you ever desire to jump to tags located in other directo‐
173 ries. If you never need to jump to tags in another directory (i.e.
174 the source in each directory is entirely self-contained), then a lo‐
175 cal tag file in each directory will fit your needs.
176 B. The time is takes for your editor to look up a tag in the tag file.
178 The significance of this factor depends upon the size of your source
180 tree and on whether the source files are located on a local or re‐
181 mote file system. For source and tag files located on a local file
182 system, looking up a tag is not as big a hit as one might first
183 imagine, since vi implementations typically perform a binary search
184 on a sorted tag file. This may or may not be true for the editor you
185 use. For files located on a remote file system, reading a large file
186 is an expensive operation.
187 C. Whether or not you expect the source code to change and the time it
189 takes to rebuild a tag file to account for changes to the source
190 code.
191 While Universal Ctags is particularly fast in scanning source code
193 (around 1-2 MB/sec), a large project may still result in objection‐
194 able delays if one wishes to keep their tag file(s) up to date on a
195 frequent basis, or if the files are located on a remote file system.
196 D. The presence of duplicate tags in the source code and the ability to
198 handle them.
199 The impact of this factor is influenced by the following three is‐
201 sues:
202 1. How common are duplicate tags in your project?
204 2. Does your editor provide any facilities for dealing with dupli‐
206 cate tags?
207 While standard vi does not, many modern vi implementations, such
209 as Vim have good facilities for selecting the desired match from
210 the list of duplicates. If your editor does not support duplicate
211 tags, then it will typically send you to only one of them,
212 whether or not that is the one you wanted (and not even notifying
213 you that there are other potential matches).
214 3. What is the significance of duplicate tags?
216 For example, if you have two tags of the same name from entirely
218 isolated software components, jumping first to the match found in
219 component B while working in component A may be entirely mislead‐
220 ing, distracting or inconvenient (to keep having to choose which
221 one if your editor provides you with a list of matches). How‐
222 ever, if you have two tags of the same name for parallel builds
223 (say two initialization routines for different hosts), you may
224 always want to specify which one you want.
225 Of the approaches listed above, I tend to favor Approach 3. My editor
227 of choice is Vim, which provides a rich set of features for handling
228 multiple tag files, which partly influences my choice. If you are work‐
229 ing with source files on a remote file system, then I would recommend
230 either Approach 3 or Approach 4, depending upon the hit when reading
231 the global tag file.
232 The advantages of Approach 3 are many (assuming that your editor has
234 the ability to support both multiple tag files and duplicate tags). All
235 lookups of tag located in the current directory are fast and the local
236 tag file can be quickly and easily regenerated in one second or less (I
237 have even mapped a keystroke to do this easily). A lookup of a (neces‐
238 sarily non-static) tag found in another directory fails a lookup in the
239 local tag file, but is found in the global tag file, which satisfies
240 all cross-directory lookups. The global tag file can be automatically
241 regenerated periodically with a cron job (and perhaps the local tag
242 files also).
243 Now I give an example of how you would implement Approach 3. Means of
245 implementing the other approaches can be performed in a similar manner.
246 Here is a visual representation of an example directory hierarchy:
248 project
250 `-----misccomp
251 | `...
252 `-----sysint
253 `-----client
254 | `-----hdrs
255 | `-----lib
256 | `-----src
257 | `-----test
258 `-----common
259 | `-----hdrs
260 | `-----lib
261 | `-----src
262 | `-----test
263 `-----server
264 `-----hdrs
265 `-----lib
266 `-----src
267 `-----test
268 Here is a recommended solution (conceptually) to build the tag files:
270 1. Within each of the leaf nodes (i.e. hdrs, lib, src, test) build a
272 tag file using "ctags *.[ch]". This can be easily be done for the
273 whole hierarchy by making a shell script, call it dirtags, contain‐
274 ing the following lines:
275 #!/bin/sh
277 cd $1
278 ctags *
279 Now execute the following command:
281 find * -type d -exec dirtags {} \;
283 These tag files are trivial (and extremely quick) to rebuild while
285 making changes within a directory. The following Vim key mapping is
286 quite useful to rebuild the tag file in the directory of the current
287 source file:
288 :nmap ,t :!(cd %:p:h;ctags *.[ch])&<CR><CR>
290 2. Build the global tag file:
292 cd ~/project
294 ctags --file-scope=no -R
295 thus constructing a tag file containing only non-static tags for all
297 source files in all descendent directories.
298 3. Configure your editor to read the local tag file first, then consult
300 the global tag file when not found in the local tag file. In Vim,
301 this is done as follows:
302 :set tags=./tags,tags,~/project/tags
304 If you wish to implement Approach 4, you would need to replace the
306 dirtags script of step 1 with the following:
307 #!/bin/sh
309 cd $1
310 ctags *
311 # Now append the non-static tags from descendent directories
312 find * -type d -prune -print | ctags -aR --file-scope=no -L-
313 And replace the configuration of step 3 with this:
315 :set tags=./tags;$HOME,tags
317 As a caveat, it should be noted that step 2 builds a global tag file
319 whose file names will be relative to the directory in which the global
320 tag file is being built. This takes advantage of the Vim tagrelative
321 option, which causes the path to be interpreted a relative to the loca‐
322 tion of the tag file instead of the current directory. For standard vi,
323 which always interprets the paths as relative to the current directory,
324 we need to build the global tag file with absolute path names. This can
325 be accomplished by replacing step 2 with the following:
326 cd ~/project
328 ctags --file-scope=no -R `pwd`
329 Does Universal Ctags support Unicode file names?
331 Yes, Unicode file names are supported on unix-like platforms (Linux,
332 macOS, Cygwin, etc.).
333 However, on Windows, you need to use Windows 10 version 1903 or later
335 to use Unicode file names. (This is an experimental feature, though.)
336 On older versions on Windows, Universal Ctags only support file names
337 represented in the current code page. If you still want to use Unicode
338 file names on them, use Cygwin or MSYS2 version of Universal Ctags as a
339 workaround.
340 Why does zsh cause zsh: no matches found error?
342 zsh causes error on the following cases;
343 ctags --extra=+* ...
345 ctags --exclude=foo/* ...
346 This is the 2nd most significant incompatibility feature of zsh.
348 Cited from "Z-Shell Frequently-Asked Questions", "2.1: Differences from
350 sh and ksh";
351 ... The next most classic difference is that unmatched glob patterns
352 cause the command to abort; set NO_NOMATCH for those.
353 You may add "setopt nonomatch" on your ~/.zshrc. Or you can escape glob
355 patterns with backslash;
356 ctags --extra=+\* ...
358 ctags --exclude=foo/\* ...
359 Or quote them;
361 ctags '--extra=+*' ...
363 ctags '--exclude=foo/*' ...
364
366 The official Universal Ctags web site at:
367 https://ctags.io/
369 ctags(1), tags(5)
371
373 This FAQ is based on Exuberant Ctags FAQ by Darren Hiebert and
374 vberthoux@users.sourceforge.net
375 Universal Ctags project: https://ctags.io/
3776.0.0 CTAGS-FAQ(7)