1GREP(1) User Commands GREP(1)
2
6 grep - print lines that match patterns
7
9 grep [OPTION...] PATTERNS [FILE...]
10 grep [OPTION...] -e PATTERNS ... [FILE...]
11 grep [OPTION...] -f PATTERN_FILE ... [FILE...]
12
14 grep searches for PATTERNS in each FILE. PATTERNS is one or more
15 patterns separated by newline characters, and grep prints each line
16 that matches a pattern. Typically PATTERNS should be quoted when grep
17 is used in a shell command.
18 A FILE of “-” stands for standard input. If no FILE is given,
20 recursive searches examine the working directory, and nonrecursive
21 searches read standard input.
22
24 Generic Program Information
25 --help Output a usage message and exit.
26 -V, --version
28 Output the version number of grep and exit.
29 Pattern Syntax
31 -E, --extended-regexp
32 Interpret PATTERNS as extended regular expressions (EREs, see
33 below).
34 -F, --fixed-strings
36 Interpret PATTERNS as fixed strings, not regular expressions.
37 -G, --basic-regexp
39 Interpret PATTERNS as basic regular expressions (BREs, see
40 below). This is the default.
41 -P, --perl-regexp
43 Interpret PATTERNS as Perl-compatible regular expressions
44 (PCREs). This option is experimental when combined with the -z
45 (--null-data) option, and grep -P may warn of unimplemented
46 features.
47 Matching Control
49 -e PATTERNS, --regexp=PATTERNS
50 Use PATTERNS as the patterns. If this option is used multiple
51 times or is combined with the -f (--file) option, search for all
52 patterns given. This option can be used to protect a pattern
53 beginning with “-”.
54 -f FILE, --file=FILE
56 Obtain patterns from FILE, one per line. If this option is used
57 multiple times or is combined with the -e (--regexp) option,
58 search for all patterns given. The empty file contains zero
59 patterns, and therefore matches nothing. If FILE is - , read
60 patterns from standard input.
61 -i, --ignore-case
63 Ignore case distinctions in patterns and input data, so that
64 characters that differ only in case match each other.
65 --no-ignore-case
67 Do not ignore case distinctions in patterns and input data.
68 This is the default. This option is useful for passing to shell
69 scripts that already use -i, to cancel its effects because the
70 two options override each other.
71 -v, --invert-match
73 Invert the sense of matching, to select non-matching lines.
74 -w, --word-regexp
76 Select only those lines containing matches that form whole
77 words. The test is that the matching substring must either be
78 at the beginning of the line, or preceded by a non-word
79 constituent character. Similarly, it must be either at the end
80 of the line or followed by a non-word constituent character.
81 Word-constituent characters are letters, digits, and the
82 underscore. This option has no effect if -x is also specified.
83 -x, --line-regexp
85 Select only those matches that exactly match the whole line.
86 For a regular expression pattern, this is like parenthesizing
87 the pattern and then surrounding it with ^ and $.
88 General Output Control
90 -c, --count
91 Suppress normal output; instead print a count of matching lines
92 for each input file. With the -v, --invert-match option (see
93 above), count non-matching lines.
94 --color[=WHEN], --colour[=WHEN]
96 Surround the matched (non-empty) strings, matching lines,
97 context lines, file names, line numbers, byte offsets, and
98 separators (for fields and groups of context lines) with escape
99 sequences to display them in color on the terminal. The colors
100 are defined by the environment variable GREP_COLORS. WHEN is
101 never, always, or auto.
102 -L, --files-without-match
104 Suppress normal output; instead print the name of each input
105 file from which no output would normally have been printed.
106 -l, --files-with-matches
108 Suppress normal output; instead print the name of each input
109 file from which output would normally have been printed.
110 Scanning each input file stops upon first match.
111 -m NUM, --max-count=NUM
113 Stop reading a file after NUM matching lines. If NUM is zero,
114 grep stops right away without reading input. A NUM of -1 is
115 treated as infinity and grep does not stop; this is the default.
116 If the input is standard input from a regular file, and NUM
117 matching lines are output, grep ensures that the standard input
118 is positioned to just after the last matching line before
119 exiting, regardless of the presence of trailing context lines.
120 This enables a calling process to resume a search. When grep
121 stops after NUM matching lines, it outputs any trailing context
122 lines. When the -c or --count option is also used, grep does
123 not output a count greater than NUM. When the -v or
124 --invert-match option is also used, grep stops after outputting
125 NUM non-matching lines.
126 -o, --only-matching
128 Print only the matched (non-empty) parts of a matching line,
129 with each such part on a separate output line.
130 -q, --quiet, --silent
132 Quiet; do not write anything to standard output. Exit
133 immediately with zero status if any match is found, even if an
134 error was detected. Also see the -s or --no-messages option.
135 -s, --no-messages
137 Suppress error messages about nonexistent or unreadable files.
138 Output Line Prefix Control
140 -b, --byte-offset
141 Print the 0-based byte offset within the input file before each
142 line of output. If -o (--only-matching) is specified, print the
143 offset of the matching part itself.
144 -H, --with-filename
146 Print the file name for each match. This is the default when
147 there is more than one file to search. This is a GNU extension.
148 -h, --no-filename
150 Suppress the prefixing of file names on output. This is the
151 default when there is only one file (or only standard input) to
152 search.
153 --label=LABEL
155 Display input actually coming from standard input as input
156 coming from file LABEL. This can be useful for commands that
157 transform a file's contents before searching, e.g., gzip -cd
158 foo.gz | grep --label=foo -H 'some pattern'. See also the -H
159 option.
160 -n, --line-number
162 Prefix each line of output with the 1-based line number within
163 its input file.
164 -T, --initial-tab
166 Make sure that the first character of actual line content lies
167 on a tab stop, so that the alignment of tabs looks normal. This
168 is useful with options that prefix their output to the actual
169 content: -H,-n, and -b. In order to improve the probability
170 that lines from a single file will all start at the same column,
171 this also causes the line number and byte offset (if present) to
172 be printed in a minimum size field width.
173 -Z, --null
175 Output a zero byte (the ASCII NUL character) instead of the
176 character that normally follows a file name. For example, grep
177 -lZ outputs a zero byte after each file name instead of the
178 usual newline. This option makes the output unambiguous, even
179 in the presence of file names containing unusual characters like
180 newlines. This option can be used with commands like find
181 -print0, perl -0, sort -z, and xargs -0 to process arbitrary
182 file names, even those that contain newline characters.
183 Context Line Control
185 -A NUM, --after-context=NUM
186 Print NUM lines of trailing context after matching lines.
187 Places a line containing a group separator (--) between
188 contiguous groups of matches. With the -o or --only-matching
189 option, this has no effect and a warning is given.
190 -B NUM, --before-context=NUM
192 Print NUM lines of leading context before matching lines.
193 Places a line containing a group separator (--) between
194 contiguous groups of matches. With the -o or --only-matching
195 option, this has no effect and a warning is given.
196 -C NUM, -NUM, --context=NUM
198 Print NUM lines of output context. Places a line containing a
199 group separator (--) between contiguous groups of matches. With
200 the -o or --only-matching option, this has no effect and a
201 warning is given.
202 --group-separator=SEP
204 When -A, -B, or -C are in use, print SEP instead of -- between
205 groups of lines.
206 --no-group-separator
208 When -A, -B, or -C are in use, do not print a separator between
209 groups of lines.
210 File and Directory Selection
212 -a, --text
213 Process a binary file as if it were text; this is equivalent to
214 the --binary-files=text option.
215 --binary-files=TYPE
217 If a file's data or metadata indicate that the file contains
218 binary data, assume that the file is of type TYPE. Non-text
219 bytes indicate binary data; these are either output bytes that
220 are improperly encoded for the current locale, or null input
221 bytes when the -z option is not given.
222 By default, TYPE is binary, and grep suppresses output after
224 null input binary data is discovered, and suppresses output
225 lines that contain improperly encoded data. When some output is
226 suppressed, grep follows any output with a message to standard
227 error saying that a binary file matches.
228 If TYPE is without-match, when grep discovers null input binary
230 data it assumes that the rest of the file does not match; this
231 is equivalent to the -I option.
232 If TYPE is text, grep processes a binary file as if it were
234 text; this is equivalent to the -a option.
235 When type is binary, grep may treat non-text bytes as line
237 terminators even without the -z option. This means choosing
238 binary versus text can affect whether a pattern matches a file.
239 For example, when type is binary the pattern q$ might match q
240 immediately followed by a null byte, even though this is not
241 matched when type is text. Conversely, when type is binary the
242 pattern . (period) might not match a null byte.
243 Warning: The -a option might output binary garbage, which can
245 have nasty side effects if the output is a terminal and if the
246 terminal driver interprets some of it as commands. On the other
247 hand, when reading files whose text encodings are unknown, it
248 can be helpful to use -a or to set LC_ALL='C' in the
249 environment, in order to find more matches even if the matches
250 are unsafe for direct display.
251 -D ACTION, --devices=ACTION
253 If an input file is a device, FIFO or socket, use ACTION to
254 process it. By default, ACTION is read, which means that
255 devices are read just as if they were ordinary files. If ACTION
256 is skip, devices are silently skipped.
257 -d ACTION, --directories=ACTION
259 If an input file is a directory, use ACTION to process it. By
260 default, ACTION is read, i.e., read directories just as if they
261 were ordinary files. If ACTION is skip, silently skip
262 directories. If ACTION is recurse, read all files under each
263 directory, recursively, following symbolic links only if they
264 are on the command line. This is equivalent to the -r option.
265 --exclude=GLOB
267 Skip any command-line file with a name suffix that matches the
268 pattern GLOB, using wildcard matching; a name suffix is either
269 the whole name, or a trailing part that starts with a non-slash
270 character immediately after a slash (/) in the name. When
271 searching recursively, skip any subfile whose base name matches
272 GLOB; the base name is the part after the last slash. A pattern
273 can use *, ?, and [...] as wildcards, and \ to quote a wildcard
274 or backslash character literally.
275 --exclude-from=FILE
277 Skip files whose base name matches any of the file-name globs
278 read from FILE (using wildcard matching as described under
279 --exclude).
280 --exclude-dir=GLOB
282 Skip any command-line directory with a name suffix that matches
283 the pattern GLOB. When searching recursively, skip any
284 subdirectory whose base name matches GLOB. Ignore any redundant
285 trailing slashes in GLOB.
286 -I Process a binary file as if it did not contain matching data;
288 this is equivalent to the --binary-files=without-match option.
289 --include=GLOB
291 Search only files whose base name matches GLOB (using wildcard
292 matching as described under --exclude). If contradictory
293 --include and --exclude options are given, the last matching one
294 wins. If no --include or --exclude options match, a file is
295 included unless the first such option is --include.
296 -r, --recursive
298 Read all files under each directory, recursively, following
299 symbolic links only if they are on the command line. Note that
300 if no file operand is given, grep searches the working
301 directory. This is equivalent to the -d recurse option.
302 -R, --dereference-recursive
304 Read all files under each directory, recursively. Follow all
305 symbolic links, unlike -r.
306 Other Options
308 --line-buffered
309 Use line buffering on output. This can cause a performance
310 penalty.
311 -U, --binary
313 Treat the file(s) as binary. By default, under MS-DOS and MS-
314 Windows, grep guesses whether a file is text or binary as
315 described for the --binary-files option. If grep decides the
316 file is a text file, it strips the CR characters from the
317 original file contents (to make regular expressions with ^ and $
318 work correctly). Specifying -U overrules this guesswork,
319 causing all files to be read and passed to the matching
320 mechanism verbatim; if the file is a text file with CR/LF pairs
321 at the end of each line, this will cause some regular
322 expressions to fail. This option has no effect on platforms
323 other than MS-DOS and MS-Windows.
324 -z, --null-data
326 Treat input and output data as sequences of lines, each
327 terminated by a zero byte (the ASCII NUL character) instead of a
328 newline. Like the -Z or --null option, this option can be used
329 with commands like sort -z to process arbitrary file names.
330
332 A regular expression is a pattern that describes a set of strings.
333 Regular expressions are constructed analogously to arithmetic
334 expressions, by using various operators to combine smaller expressions.
335 grep understands three different versions of regular expression syntax:
337 “basic” (BRE), “extended” (ERE) and “perl” (PCRE). In GNU grep, basic
338 and extended regular expressions are merely different notations for the
339 same pattern-matching functionality. In other implementations, basic
340 regular expressions are ordinarily less powerful than extended, though
341 occasionally it is the other way around. The following description
342 applies to extended regular expressions; differences for basic regular
343 expressions are summarized afterwards. Perl-compatible regular
344 expressions have different functionality, and are documented in
345 pcre2syntax(3) and pcre2pattern(3), but work only if PCRE support is
346 enabled.
347 The fundamental building blocks are the regular expressions that match
349 a single character. Most characters, including all letters and digits,
350 are regular expressions that match themselves. Any meta-character with
351 special meaning may be quoted by preceding it with a backslash.
352 The period . matches any single character. It is unspecified whether
354 it matches an encoding error.
355 Character Classes and Bracket Expressions
357 A bracket expression is a list of characters enclosed by [ and ]. It
358 matches any single character in that list. If the first character of
359 the list is the caret ^ then it matches any character not in the list;
360 it is unspecified whether it matches an encoding error. For example,
361 the regular expression [0123456789] matches any single digit.
362 Within a bracket expression, a range expression consists of two
364 characters separated by a hyphen. It matches any single character that
365 sorts between the two characters, inclusive, using the locale's
366 collating sequence and character set. For example, in the default C
367 locale, [a-d] is equivalent to [abcd]. Many locales sort characters in
368 dictionary order, and in these locales [a-d] is typically not
369 equivalent to [abcd]; it might be equivalent to [aBbCcDd], for example.
370 To obtain the traditional interpretation of bracket expressions, you
371 can use the C locale by setting the LC_ALL environment variable to the
372 value C.
373 Finally, certain named classes of characters are predefined within
375 bracket expressions, as follows. Their names are self explanatory, and
376 they are [:alnum:], [:alpha:], [:blank:], [:cntrl:], [:digit:],
377 [:graph:], [:lower:], [:print:], [:punct:], [:space:], [:upper:], and
378 [:xdigit:]. For example, [[:alnum:]] means the character class of
379 numbers and letters in the current locale. In the C locale and ASCII
380 character set encoding, this is the same as [0-9A-Za-z]. (Note that
381 the brackets in these class names are part of the symbolic names, and
382 must be included in addition to the brackets delimiting the bracket
383 expression.) Most meta-characters lose their special meaning inside
384 bracket expressions. To include a literal ] place it first in the
385 list. Similarly, to include a literal ^ place it anywhere but first.
386 Finally, to include a literal - place it last.
387 Anchoring
389 The caret ^ and the dollar sign $ are meta-characters that respectively
390 match the empty string at the beginning and end of a line.
391 The Backslash Character and Special Expressions
393 The symbols \< and \> respectively match the empty string at the
394 beginning and end of a word. The symbol \b matches the empty string at
395 the edge of a word, and \B matches the empty string provided it's not
396 at the edge of a word. The symbol \w is a synonym for [_[:alnum:]] and
397 \W is a synonym for [^_[:alnum:]].
398 Repetition
400 A regular expression may be followed by one of several repetition
401 operators:
402 ? The preceding item is optional and matched at most once.
403 * The preceding item will be matched zero or more times.
404 + The preceding item will be matched one or more times.
405 {n} The preceding item is matched exactly n times.
406 {n,} The preceding item is matched n or more times.
407 {,m} The preceding item is matched at most m times. This is a GNU
408 extension.
409 {n,m} The preceding item is matched at least n times, but not more
410 than m times.
411 Concatenation
413 Two regular expressions may be concatenated; the resulting regular
414 expression matches any string formed by concatenating two substrings
415 that respectively match the concatenated expressions.
416 Alternation
418 Two regular expressions may be joined by the infix operator |; the
419 resulting regular expression matches any string matching either
420 alternate expression.
421 Precedence
423 Repetition takes precedence over concatenation, which in turn takes
424 precedence over alternation. A whole expression may be enclosed in
425 parentheses to override these precedence rules and form a
426 subexpression.
427 Back-references and Subexpressions
429 The back-reference \n, where n is a single digit, matches the substring
430 previously matched by the nth parenthesized subexpression of the
431 regular expression.
432 Basic vs Extended Regular Expressions
434 In basic regular expressions the meta-characters ?, +, {, |, (, and )
435 lose their special meaning; instead use the backslashed versions \?,
436 \+, \{, \|, \(, and \).
437
439 Normally the exit status is 0 if a line is selected, 1 if no lines were
440 selected, and 2 if an error occurred. However, if the -q or --quiet or
441 --silent is used and a line is selected, the exit status is 0 even if
442 an error occurred.
443
445 The behavior of grep is affected by the following environment
446 variables.
447 The locale for category LC_foo is specified by examining the three
449 environment variables LC_ALL, LC_foo, LANG, in that order. The first
450 of these variables that is set specifies the locale. For example, if
451 LC_ALL is not set, but LC_MESSAGES is set to pt_BR, then the Brazilian
452 Portuguese locale is used for the LC_MESSAGES category. The C locale
453 is used if none of these environment variables are set, if the locale
454 catalog is not installed, or if grep was not compiled with national
455 language support (NLS). The shell command locale -a lists locales that
456 are currently available.
457 GREP_COLORS
459 Controls how the --color option highlights output. Its value is
460 a colon-separated list of capabilities that defaults to
461 ms=01;31:mc=01;31:sl=:cx=:fn=35:ln=32:bn=32:se=36 with the rv
462 and ne boolean capabilities omitted (i.e., false). Supported
463 capabilities are as follows.
464 sl= SGR substring for whole selected lines (i.e., matching
466 lines when the -v command-line option is omitted, or non-
467 matching lines when -v is specified). If however the
468 boolean rv capability and the -v command-line option are
469 both specified, it applies to context matching lines
470 instead. The default is empty (i.e., the terminal's
471 default color pair).
472 cx= SGR substring for whole context lines (i.e., non-matching
474 lines when the -v command-line option is omitted, or
475 matching lines when -v is specified). If however the
476 boolean rv capability and the -v command-line option are
477 both specified, it applies to selected non-matching lines
478 instead. The default is empty (i.e., the terminal's
479 default color pair).
480 rv Boolean value that reverses (swaps) the meanings of the
482 sl= and cx= capabilities when the -v command-line option
483 is specified. The default is false (i.e., the capability
484 is omitted).
485 mt=01;31
487 SGR substring for matching non-empty text in any matching
488 line (i.e., a selected line when the -v command-line
489 option is omitted, or a context line when -v is
490 specified). Setting this is equivalent to setting both
491 ms= and mc= at once to the same value. The default is a
492 bold red text foreground over the current line
493 background.
494 ms=01;31
496 SGR substring for matching non-empty text in a selected
497 line. (This is only used when the -v command-line option
498 is omitted.) The effect of the sl= (or cx= if rv)
499 capability remains active when this kicks in. The
500 default is a bold red text foreground over the current
501 line background.
502 mc=01;31
504 SGR substring for matching non-empty text in a context
505 line. (This is only used when the -v command-line option
506 is specified.) The effect of the cx= (or sl= if rv)
507 capability remains active when this kicks in. The
508 default is a bold red text foreground over the current
509 line background.
510 fn=35 SGR substring for file names prefixing any content line.
512 The default is a magenta text foreground over the
513 terminal's default background.
514 ln=32 SGR substring for line numbers prefixing any content
516 line. The default is a green text foreground over the
517 terminal's default background.
518 bn=32 SGR substring for byte offsets prefixing any content
520 line. The default is a green text foreground over the
521 terminal's default background.
522 se=36 SGR substring for separators that are inserted between
524 selected line fields (:), between context line fields,
525 (-), and between groups of adjacent lines when nonzero
526 context is specified (--). The default is a cyan text
527 foreground over the terminal's default background.
528 ne Boolean value that prevents clearing to the end of line
530 using Erase in Line (EL) to Right (\33[K) each time a
531 colorized item ends. This is needed on terminals on
532 which EL is not supported. It is otherwise useful on
533 terminals for which the back_color_erase (bce) boolean
534 terminfo capability does not apply, when the chosen
535 highlight colors do not affect the background, or when EL
536 is too slow or causes too much flicker. The default is
537 false (i.e., the capability is omitted).
538 Note that boolean capabilities have no =... part. They are
540 omitted (i.e., false) by default and become true when specified.
541 See the Select Graphic Rendition (SGR) section in the
543 documentation of the text terminal that is used for permitted
544 values and their meaning as character attributes. These
545 substring values are integers in decimal representation and can
546 be concatenated with semicolons. grep takes care of assembling
547 the result into a complete SGR sequence (\33[...m). Common
548 values to concatenate include 1 for bold, 4 for underline, 5 for
549 blink, 7 for inverse, 39 for default foreground color, 30 to 37
550 for foreground colors, 90 to 97 for 16-color mode foreground
551 colors, 38;5;0 to 38;5;255 for 88-color and 256-color modes
552 foreground colors, 49 for default background color, 40 to 47 for
553 background colors, 100 to 107 for 16-color mode background
554 colors, and 48;5;0 to 48;5;255 for 88-color and 256-color modes
555 background colors.
556 LC_ALL, LC_COLLATE, LANG
558 These variables specify the locale for the LC_COLLATE category,
559 which determines the collating sequence used to interpret range
560 expressions like [a-z].
561 LC_ALL, LC_CTYPE, LANG
563 These variables specify the locale for the LC_CTYPE category,
564 which determines the type of characters, e.g., which characters
565 are whitespace. This category also determines the character
566 encoding, that is, whether text is encoded in UTF-8, ASCII, or
567 some other encoding. In the C or POSIX locale, all characters
568 are encoded as a single byte and every byte is a valid
569 character.
570 LC_ALL, LC_MESSAGES, LANG
572 These variables specify the locale for the LC_MESSAGES category,
573 which determines the language that grep uses for messages. The
574 default C locale uses American English messages.
575 POSIXLY_CORRECT
577 If set, grep behaves as POSIX requires; otherwise, grep behaves
578 more like other GNU programs. POSIX requires that options that
579 follow file names must be treated as file names; by default,
580 such options are permuted to the front of the operand list and
581 are treated as options. Also, POSIX requires that unrecognized
582 options be diagnosed as “illegal”, but since they are not really
583 against the law the default is to diagnose them as “invalid”.
584
586 This man page is maintained only fitfully; the full documentation is
587 often more up-to-date.
588
590 Copyright 1998-2000, 2002, 2005-2023 Free Software Foundation, Inc.
591 This is free software; see the source for copying conditions. There is
593 NO warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR
594 PURPOSE.
595
597 Reporting Bugs
598 Email bug reports to the bug-reporting address ⟨bug-grep@gnu.org⟩. An
599 email archive ⟨https://lists.gnu.org/mailman/listinfo/bug-grep⟩ and a
600 bug tracker ⟨https://debbugs.gnu.org/cgi/pkgreport.cgi?package=grep⟩
601 are available.
602 Known Bugs
604 Large repetition counts in the {n,m} construct may cause grep to use
605 lots of memory. In addition, certain other obscure regular expressions
606 require exponential time and space, and may cause grep to run out of
607 memory.
608 Back-references are very slow, and may require exponential time.
610
612 The following example outputs the location and contents of any line
613 containing “f” and ending in “.c”, within all files in the current di‐
614 rectory whose names contain “g” and end in “.h”. The -n option outputs
615 line numbers, the -- argument treats expansions of “*g*.h” starting
616 with “-” as file names not options, and the empty file /dev/null causes
617 file names to be output even if only one file name happens to be of the
618 form “*g*.h”.
619 $ grep -n -- 'f.*\.c$' *g*.h /dev/null
621 argmatch.h:1:/* definitions and prototypes for argmatch.c
622 The only line that matches is line 1 of argmatch.h. Note that the reg‐
624 ular expression syntax used in the pattern differs from the globbing
625 syntax that the shell uses to match file names.
626
628 Regular Manual Pages
629 awk(1), cmp(1), diff(1), find(1), perl(1), sed(1), sort(1), xargs(1),
630 read(2), pcre2(3), pcre2syntax(3), pcre2pattern(3), terminfo(5),
631 glob(7), regex(7)
632 Full Documentation
634 A complete manual ⟨https://www.gnu.org/software/grep/manual/⟩ is avail‐
635 able. If the info and grep programs are properly installed at your
636 site, the command
637 info grep
639 should give you access to the complete manual.
641GNU grep 3.11 2019-12-29 GREP(1)