1PCREGREP(1) General Commands Manual PCREGREP(1)
2
6 pcregrep - a grep with Perl-compatible regular expressions.
7
9 pcregrep [options] [long options] [pattern] [path1 path2 ...]
10
12 pcregrep searches files for character patterns, in the same way as
14 other grep commands do, but it uses the PCRE regular expression library
15 to support patterns that are compatible with the regular expressions of
16 Perl 5. See pcrepattern(3) for a full description of syntax and seman‐
17 tics of the regular expressions that PCRE supports.
18 Patterns, whether supplied on the command line or in a separate file,
20 are given without delimiters. For example:
21 pcregrep Thursday /etc/motd
23 If you attempt to use delimiters (for example, by surrounding a pattern
25 with slashes, as is common in Perl scripts), they are interpreted as
26 part of the pattern. Quotes can of course be used to delimit patterns
27 on the command line because they are interpreted by the shell, and
28 indeed they are required if a pattern contains white space or shell
29 metacharacters.
30 The first argument that follows any option settings is treated as the
32 single pattern to be matched when neither -e nor -f is present. Con‐
33 versely, when one or both of these options are used to specify pat‐
34 terns, all arguments are treated as path names. At least one of -e, -f,
35 or an argument pattern must be provided.
36 If no files are specified, pcregrep reads the standard input. The stan‐
38 dard input can also be referenced by a name consisting of a single
39 hyphen. For example:
40 pcregrep some-pattern /file1 - /file3
42 By default, each line that matches a pattern is copied to the standard
44 output, and if there is more than one file, the file name is output at
45 the start of each line, followed by a colon. However, there are options
46 that can change how pcregrep behaves. In particular, the -M option
47 makes it possible to search for patterns that span line boundaries.
48 What defines a line boundary is controlled by the -N (--newline)
49 option.
50 Patterns are limited to 8K or BUFSIZ characters, whichever is the
52 greater. BUFSIZ is defined in <stdio.h>. When there is more than one
53 pattern (specified by the use of -e and/or -f), each pattern is applied
54 to each line in the order in which they are defined, except that all
55 the -e patterns are tried before the -f patterns.
56 By default, as soon as one pattern matches (or fails to match when -v
58 is used), no further patterns are considered. However, if --colour (or
59 --color) is used to colour the matching substrings, or if --only-match‐
60 ing, --file-offsets, or --line-offsets is used to output only the part
61 of the line that matched (either shown literally, or as an offset),
62 scanning resumes immediately following the match, so that further
63 matches on the same line can be found. If there are multiple patterns,
64 they are all tried on the remainder of the line, but patterns that fol‐
65 low the one that matched are not tried on the earlier part of the line.
66 This is the same behaviour as GNU grep, but it does mean that the order
68 in which multiple patterns are specified can affect the output when one
69 of the above options is used.
70 Patterns that can match an empty string are accepted, but empty string
72 matches are never recognized. An example is the pattern
73 "(super)?(man)?", in which all components are optional. This pattern
74 finds all occurrences of both "super" and "man"; the output differs
75 from matching with "super|man" when only the matching substrings are
76 being shown.
77 If the LC_ALL or LC_CTYPE environment variable is set, pcregrep uses
79 the value to set a locale when calling the PCRE library. The --locale
80 option can be used to override this.
81
83 It is possible to compile pcregrep so that it uses libz or libbz2 to
85 read files whose names end in .gz or .bz2, respectively. You can find
86 out whether your binary has support for one or both of these file types
87 by running it with the --help option. If the appropriate support is not
88 present, files are treated as plain text. The standard input is always
89 so treated.
90
92 The order in which some of the options appear can affect the output.
94 For example, both the -h and -l options affect the printing of file
95 names. Whichever comes later in the command line will be the one that
96 takes effect.
97 -- This terminate the list of options. It is useful if the next
99 item on the command line starts with a hyphen but is not an
100 option. This allows for the processing of patterns and file‐
101 names that start with hyphens.
102 -A number, --after-context=number
104 Output number lines of context after each matching line. If
105 filenames and/or line numbers are being output, a hyphen sep‐
106 arator is used instead of a colon for the context lines. A
107 line containing "--" is output between each group of lines,
108 unless they are in fact contiguous in the input file. The
109 value of number is expected to be relatively small. However,
110 pcregrep guarantees to have up to 8K of following text avail‐
111 able for context output.
112 -B number, --before-context=number
114 Output number lines of context before each matching line. If
115 filenames and/or line numbers are being output, a hyphen sep‐
116 arator is used instead of a colon for the context lines. A
117 line containing "--" is output between each group of lines,
118 unless they are in fact contiguous in the input file. The
119 value of number is expected to be relatively small. However,
120 pcregrep guarantees to have up to 8K of preceding text avail‐
121 able for context output.
122 -C number, --context=number
124 Output number lines of context both before and after each
125 matching line. This is equivalent to setting both -A and -B
126 to the same value.
127 -c, --count
129 Do not output individual lines from the files that are being
130 scanned; instead output the number of lines that would other‐
131 wise have been shown. If no lines are selected, the number
132 zero is output. If several files are are being scanned, a
133 count is output for each of them. However, if the --files-
134 with-matches option is also used, only those files whose
135 counts are greater than zero are listed. When -c is used, the
136 -A, -B, and -C options are ignored.
137 --colour, --color
139 If this option is given without any data, it is equivalent to
140 "--colour=auto". If data is required, it must be given in
141 the same shell item, separated by an equals sign.
142 --colour=value, --color=value
144 This option specifies under what circumstances the parts of a
145 line that matched a pattern should be coloured in the output.
146 By default, the output is not coloured. The value (which is
147 optional, see above) may be "never", "always", or "auto". In
148 the latter case, colouring happens only if the standard out‐
149 put is connected to a terminal. More resources are used when
150 colouring is enabled, because pcregrep has to search for all
151 possible matches in a line, not just one, in order to colour
152 them all.
153 The colour that is used can be specified by setting the envi‐
155 ronment variable PCREGREP_COLOUR or PCREGREP_COLOR. The value
156 of this variable should be a string of two numbers, separated
157 by a semicolon. They are copied directly into the control
158 string for setting colour on a terminal, so it is your
159 responsibility to ensure that they make sense. If neither of
160 the environment variables is set, the default is "1;31",
161 which gives red.
162 -D action, --devices=action
164 If an input path is not a regular file or a directory,
165 "action" specifies how it is to be processed. Valid values
166 are "read" (the default) or "skip" (silently skip the path).
167 -d action, --directories=action
169 If an input path is a directory, "action" specifies how it is
170 to be processed. Valid values are "read" (the default),
171 "recurse" (equivalent to the -r option), or "skip" (silently
172 skip the path). In the default case, directories are read as
173 if they were ordinary files. In some operating systems the
174 effect of reading a directory like this is an immediate end-
175 of-file.
176 -e pattern, --regex=pattern, --regexp=pattern
178 Specify a pattern to be matched. This option can be used mul‐
179 tiple times in order to specify several patterns. It can also
180 be used as a way of specifying a single pattern that starts
181 with a hyphen. When -e is used, no argument pattern is taken
182 from the command line; all arguments are treated as file
183 names. There is an overall maximum of 100 patterns. They are
184 applied to each line in the order in which they are defined
185 until one matches (or fails to match if -v is used). If -f is
186 used with -e, the command line patterns are matched first,
187 followed by the patterns from the file, independent of the
188 order in which these options are specified. Note that multi‐
189 ple use of -e is not the same as a single pattern with alter‐
190 natives. For example, X|Y finds the first character in a line
191 that is X or Y, whereas if the two patterns are given sepa‐
192 rately, pcregrep finds X if it is present, even if it follows
193 Y in the line. It finds Y only if there is no X in the line.
194 This really matters only if you are using -o to show the
195 part(s) of the line that matched.
196 --exclude=pattern
198 When pcregrep is searching the files in a directory as a con‐
199 sequence of the -r (recursive search) option, any regular
200 files whose names match the pattern are excluded. Subdirecto‐
201 ries are not excluded by this option; they are searched
202 recursively, subject to the --exclude_dir and --include_dir
203 options. The pattern is a PCRE regular expression, and is
204 matched against the final component of the file name (not the
205 entire path). If a file name matches both --include and
206 --exclude, it is excluded. There is no short form for this
207 option.
208 --exclude_dir=pattern
210 When pcregrep is searching the contents of a directory as a
211 consequence of the -r (recursive search) option, any subdi‐
212 rectories whose names match the pattern are excluded. (Note
213 that the --exclude option does not affect subdirectories.)
214 The pattern is a PCRE regular expression, and is matched
215 against the final component of the name (not the entire
216 path). If a subdirectory name matches both --include_dir and
217 --exclude_dir, it is excluded. There is no short form for
218 this option.
219 -F, --fixed-strings
221 Interpret each pattern as a list of fixed strings, separated
222 by newlines, instead of as a regular expression. The -w
223 (match as a word) and -x (match whole line) options can be
224 used with -F. They apply to each of the fixed strings. A line
225 is selected if any of the fixed strings are found in it (sub‐
226 ject to -w or -x, if present).
227 -f filename, --file=filename
229 Read a number of patterns from the file, one per line, and
230 match them against each line of input. A data line is output
231 if any of the patterns match it. The filename can be given as
232 "-" to refer to the standard input. When -f is used, patterns
233 specified on the command line using -e may also be present;
234 they are tested before the file's patterns. However, no other
235 pattern is taken from the command line; all arguments are
236 treated as file names. There is an overall maximum of 100
237 patterns. Trailing white space is removed from each line, and
238 blank lines are ignored. An empty file contains no patterns
239 and therefore matches nothing. See also the comments about
240 multiple patterns versus a single pattern with alternatives
241 in the description of -e above.
242 --file-offsets
244 Instead of showing lines or parts of lines that match, show
245 each match as an offset from the start of the file and a
246 length, separated by a comma. In this mode, no context is
247 shown. That is, the -A, -B, and -C options are ignored. If
248 there is more than one match in a line, each of them is shown
249 separately. This option is mutually exclusive with --line-
250 offsets and --only-matching.
251 -H, --with-filename
253 Force the inclusion of the filename at the start of output
254 lines when searching a single file. By default, the filename
255 is not shown in this case. For matching lines, the filename
256 is followed by a colon; for context lines, a hyphen separator
257 is used. If a line number is also being output, it follows
258 the file name.
259 -h, --no-filename
261 Suppress the output filenames when searching multiple files.
262 By default, filenames are shown when multiple files are
263 searched. For matching lines, the filename is followed by a
264 colon; for context lines, a hyphen separator is used. If a
265 line number is also being output, it follows the file name.
266 --help Output a help message, giving brief details of the command
268 options and file type support, and then exit.
269 -i, --ignore-case
271 Ignore upper/lower case distinctions during comparisons.
272 --include=pattern
274 When pcregrep is searching the files in a directory as a con‐
275 sequence of the -r (recursive search) option, only those reg‐
276 ular files whose names match the pattern are included. Subdi‐
277 rectories are always included and searched recursively, sub‐
278 ject to the --include_dir and --exclude_dir options. The pat‐
279 tern is a PCRE regular expression, and is matched against the
280 final component of the file name (not the entire path). If a
281 file name matches both --include and --exclude, it is
282 excluded. There is no short form for this option.
283 --include_dir=pattern
285 When pcregrep is searching the contents of a directory as a
286 consequence of the -r (recursive search) option, only those
287 subdirectories whose names match the pattern are included.
288 (Note that the --include option does not affect subdirecto‐
289 ries.) The pattern is a PCRE regular expression, and is
290 matched against the final component of the name (not the
291 entire path). If a subdirectory name matches both
292 --include_dir and --exclude_dir, it is excluded. There is no
293 short form for this option.
294 -L, --files-without-match
296 Instead of outputting lines from the files, just output the
297 names of the files that do not contain any lines that would
298 have been output. Each file name is output once, on a sepa‐
299 rate line.
300 -l, --files-with-matches
302 Instead of outputting lines from the files, just output the
303 names of the files containing lines that would have been out‐
304 put. Each file name is output once, on a separate line.
305 Searching normally stops as soon as a matching line is found
306 in a file. However, if the -c (count) option is also used,
307 matching continues in order to obtain the correct count, and
308 those files that have at least one match are listed along
309 with their counts. Using this option with -c is a way of sup‐
310 pressing the listing of files with no matches.
311 --label=name
313 This option supplies a name to be used for the standard input
314 when file names are being output. If not supplied, "(standard
315 input)" is used. There is no short form for this option.
316 --line-buffered
318 When this option is given, input is read and processed line
319 by line, and the output is flushed after each write. By
320 default, input is read in large chunks, unless pcregrep can
321 determine that it is reading from a terminal (which is cur‐
322 rently possible only in Unix environments). Output to termi‐
323 nal is normally automatically flushed by the operating sys‐
324 tem. This option can be useful when the input or output is
325 attached to a pipe and you do not want pcregrep to buffer up
326 large amounts of data. However, its use will affect perfor‐
327 mance, and the -M (multiline) option ceases to work.
328 --line-offsets
330 Instead of showing lines or parts of lines that match, show
331 each match as a line number, the offset from the start of the
332 line, and a length. The line number is terminated by a colon
333 (as usual; see the -n option), and the offset and length are
334 separated by a comma. In this mode, no context is shown.
335 That is, the -A, -B, and -C options are ignored. If there is
336 more than one match in a line, each of them is shown sepa‐
337 rately. This option is mutually exclusive with --file-offsets
338 and --only-matching.
339 --locale=locale-name
341 This option specifies a locale to be used for pattern match‐
342 ing. It overrides the value in the LC_ALL or LC_CTYPE envi‐
343 ronment variables. If no locale is specified, the PCRE
344 library's default (usually the "C" locale) is used. There is
345 no short form for this option.
346 -M, --multiline
348 Allow patterns to match more than one line. When this option
349 is given, patterns may usefully contain literal newline char‐
350 acters and internal occurrences of ^ and $ characters. The
351 output for any one match may consist of more than one line.
352 When this option is set, the PCRE library is called in "mul‐
353 tiline" mode. There is a limit to the number of lines that
354 can be matched, imposed by the way that pcregrep buffers the
355 input file as it scans it. However, pcregrep ensures that at
356 least 8K characters or the rest of the document (whichever is
357 the shorter) are available for forward matching, and simi‐
358 larly the previous 8K characters (or all the previous charac‐
359 ters, if fewer than 8K) are guaranteed to be available for
360 lookbehind assertions. This option does not work when input
361 is read line by line (see --line-buffered.)
362 -N newline-type, --newline=newline-type
364 The PCRE library supports five different conventions for
365 indicating the ends of lines. They are the single-character
366 sequences CR (carriage return) and LF (linefeed), the two-
367 character sequence CRLF, an "anycrlf" convention, which rec‐
368 ognizes any of the preceding three types, and an "any" con‐
369 vention, in which any Unicode line ending sequence is assumed
370 to end a line. The Unicode sequences are the three just men‐
371 tioned, plus VT (vertical tab, U+000B), FF (formfeed,
372 U+000C), NEL (next line, U+0085), LS (line separator,
373 U+2028), and PS (paragraph separator, U+2029).
374 When the PCRE library is built, a default line-ending
376 sequence is specified. This is normally the standard
377 sequence for the operating system. Unless otherwise specified
378 by this option, pcregrep uses the library's default. The
379 possible values for this option are CR, LF, CRLF, ANYCRLF, or
380 ANY. This makes it possible to use pcregrep on files that
381 have come from other environments without having to modify
382 their line endings. If the data that is being scanned does
383 not agree with the convention set by this option, pcregrep
384 may behave in strange ways.
385 -n, --line-number
387 Precede each output line by its line number in the file, fol‐
388 lowed by a colon for matching lines or a hyphen for context
389 lines. If the filename is also being output, it precedes the
390 line number. This option is forced if --line-offsets is used.
391 -o, --only-matching
393 Show only the part of the line that matched a pattern. In
394 this mode, no context is shown. That is, the -A, -B, and -C
395 options are ignored. If there is more than one match in a
396 line, each of them is shown separately. If -o is combined
397 with -v (invert the sense of the match to find non-matching
398 lines), no output is generated, but the return code is set
399 appropriately. This option is mutually exclusive with --file-
400 offsets and --line-offsets.
401 -q, --quiet
403 Work quietly, that is, display nothing except error messages.
404 The exit status indicates whether or not any matches were
405 found.
406 -r, --recursive
408 If any given path is a directory, recursively scan the files
409 it contains, taking note of any --include and --exclude set‐
410 tings. By default, a directory is read as a normal file; in
411 some operating systems this gives an immediate end-of-file.
412 This option is a shorthand for setting the -d option to
413 "recurse".
414 -s, --no-messages
416 Suppress error messages about non-existent or unreadable
417 files. Such files are quietly skipped. However, the return
418 code is still 2, even if matches were found in other files.
419 -u, --utf-8
421 Operate in UTF-8 mode. This option is available only if PCRE
422 has been compiled with UTF-8 support. Both patterns and sub‐
423 ject lines must be valid strings of UTF-8 characters.
424 -V, --version
426 Write the version numbers of pcregrep and the PCRE library
427 that is being used to the standard error stream.
428 -v, --invert-match
430 Invert the sense of the match, so that lines which do not
431 match any of the patterns are the ones that are found.
432 -w, --word-regex, --word-regexp
434 Force the patterns to match only whole words. This is equiva‐
435 lent to having \b at the start and end of the pattern.
436 -x, --line-regex, --line-regexp
438 Force the patterns to be anchored (each must start matching
439 at the beginning of a line) and in addition, require them to
440 match entire lines. This is equivalent to having ^ and $
441 characters at the start and end of each alternative branch in
442 every pattern.
443
445 The environment variables LC_ALL and LC_CTYPE are examined, in that
447 order, for a locale. The first one that is set is used. This can be
448 overridden by the --locale option. If no locale is set, the PCRE
449 library's default (usually the "C" locale) is used.
450
452 The -N (--newline) option allows pcregrep to scan files with different
454 newline conventions from the default. However, the setting of this
455 option does not affect the way in which pcregrep writes information to
456 the standard error and output streams. It uses the string "\n" in C
457 printf() calls to indicate newlines, relying on the C I/O library to
458 convert this to an appropriate sequence if the output is sent to a
459 file.
460
462 The majority of short and long forms of pcregrep's options are the same
464 as in the GNU grep program. Any long option of the form --xxx-regexp
465 (GNU terminology) is also available as --xxx-regex (PCRE terminology).
466 However, the --locale, -M, --multiline, -u, and --utf-8 options are
467 specific to pcregrep. If both the -c and -l options are given, GNU grep
468 lists only file names, without counts, but pcregrep gives the counts.
469
471 There are four different ways in which an option with data can be spec‐
473 ified. If a short form option is used, the data may follow immedi‐
474 ately, or in the next command line item. For example:
475 -f/some/file
477 -f /some/file
478 If a long form option is used, the data may appear in the same command
480 line item, separated by an equals character, or (with one exception) it
481 may appear in the next command line item. For example:
482 --file=/some/file
484 --file /some/file
485 Note, however, that if you want to supply a file name beginning with ~
487 as data in a shell command, and have the shell expand ~ to a home
488 directory, you must separate the file name from the option, because the
489 shell does not treat ~ specially unless it is at the start of an item.
490 The exception to the above is the --colour (or --color) option, for
492 which the data is optional. If this option does have data, it must be
493 given in the first form, using an equals character. Otherwise it will
494 be assumed that it has no data.
495
497 It is possible to supply a regular expression that takes a very long
499 time to fail to match certain lines. Such patterns normally involve
500 nested indefinite repeats, for example: (a+)*\d when matched against a
501 line of a's with no final digit. The PCRE matching function has a
502 resource limit that causes it to abort in these circumstances. If this
503 happens, pcregrep outputs an error message and the line that caused the
504 problem to the standard error stream. If there are more than 20 such
505 errors, pcregrep gives up.
506
508 Exit status is 0 if any matches were found, 1 if no matches were found,
510 and 2 for syntax errors and non-existent or inacessible files (even if
511 matches were found in other files) or too many matching errors. Using
512 the -s option to suppress error messages about inaccessble files does
513 not affect the return code.
514
516 pcrepattern(3), pcretest(1).
518
520 Philip Hazel
522 University Computing Service
523 Cambridge CB2 3QH, England.
524
526 Last updated: 21 May 2010
528 Copyright (c) 1997-2010 University of Cambridge.
529 PCREGREP(1)