1<html> 2<head> 3<title>pcre2grep specification</title> 4</head> 5<body bgcolor="#FFFFFF" text="#00005A" link="#0066FF" alink="#3399FF" vlink="#2222BB"> 6<h1>pcre2grep man page</h1> 7<p> 8Return to the <a href="index.html">PCRE2 index page</a>. 9</p> 10<p> 11This page is part of the PCRE2 HTML documentation. It was generated 12automatically from the original man page. If there is any nonsense in it, 13please consult the man page, in case the conversion went wrong. 14<br> 15<ul> 16<li><a name="TOC1" href="#SEC1">SYNOPSIS</a> 17<li><a name="TOC2" href="#SEC2">DESCRIPTION</a> 18<li><a name="TOC3" href="#SEC3">SUPPORT FOR COMPRESSED FILES</a> 19<li><a name="TOC4" href="#SEC4">BINARY FILES</a> 20<li><a name="TOC5" href="#SEC5">BINARY ZEROS IN PATTERNS</a> 21<li><a name="TOC6" href="#SEC6">OPTIONS</a> 22<li><a name="TOC7" href="#SEC7">ENVIRONMENT VARIABLES</a> 23<li><a name="TOC8" href="#SEC8">NEWLINES</a> 24<li><a name="TOC9" href="#SEC9">OPTIONS COMPATIBILITY</a> 25<li><a name="TOC10" href="#SEC10">OPTIONS WITH DATA</a> 26<li><a name="TOC11" href="#SEC11">USING PCRE2'S CALLOUT FACILITY</a> 27<li><a name="TOC12" href="#SEC12">MATCHING ERRORS</a> 28<li><a name="TOC13" href="#SEC13">DIAGNOSTICS</a> 29<li><a name="TOC14" href="#SEC14">SEE ALSO</a> 30<li><a name="TOC15" href="#SEC15">AUTHOR</a> 31<li><a name="TOC16" href="#SEC16">REVISION</a> 32</ul> 33<br><a name="SEC1" href="#TOC1">SYNOPSIS</a><br> 34<P> 35<b>pcre2grep [options] [long options] [pattern] [path1 path2 ...]</b> 36</P> 37<br><a name="SEC2" href="#TOC1">DESCRIPTION</a><br> 38<P> 39<b>pcre2grep</b> searches files for character patterns, in the same way as other 40grep commands do, but it uses the PCRE2 regular expression library to support 41patterns that are compatible with the regular expressions of Perl 5. See 42<a href="pcre2syntax.html"><b>pcre2syntax</b>(3)</a> 43for a quick-reference summary of pattern syntax, or 44<a href="pcre2pattern.html"><b>pcre2pattern</b>(3)</a> 45for a full description of the syntax and semantics of the regular expressions 46that PCRE2 supports. 47</P> 48<P> 49Patterns, whether supplied on the command line or in a separate file, are given 50without delimiters. For example: 51<pre> 52 pcre2grep Thursday /etc/motd 53</pre> 54If you attempt to use delimiters (for example, by surrounding a pattern with 55slashes, as is common in Perl scripts), they are interpreted as part of the 56pattern. Quotes can of course be used to delimit patterns on the command line 57because they are interpreted by the shell, and indeed quotes are required if a 58pattern contains white space or shell metacharacters. 59</P> 60<P> 61The first argument that follows any option settings is treated as the single 62pattern to be matched when neither <b>-e</b> nor <b>-f</b> is present. 63Conversely, when one or both of these options are used to specify patterns, all 64arguments are treated as path names. At least one of <b>-e</b>, <b>-f</b>, or an 65argument pattern must be provided. 66</P> 67<P> 68If no files are specified, <b>pcre2grep</b> reads the standard input. The 69standard input can also be referenced by a name consisting of a single hyphen. 70For example: 71<pre> 72 pcre2grep some-pattern file1 - file3 73</pre> 74By default, input files are searched line by line. Each line that matches a 75pattern is copied to the standard output, and if there is more than one file, 76the file name is output at the start of each line, followed by a colon. 77However, there are options that can change how <b>pcre2grep</b> behaves. For 78example, the <b>-M</b> option makes it possible to search for strings that span 79line boundaries. What defines a line boundary is controlled by the <b>-N</b> 80(<b>--newline</b>) option. The <b>-h</b> and <b>-H</b> options control whether or 81not file names are shown, and the <b>-Z</b> option changes the file name 82terminator to a zero byte. 83</P> 84<P> 85The amount of memory used for buffering files that are being scanned is 86controlled by parameters that can be set by the <b>--buffer-size</b> and 87<b>--max-buffer-size</b> options. The first of these sets the size of buffer 88that is obtained at the start of processing. If an input file contains very 89long lines, a larger buffer may be needed; this is handled by automatically 90extending the buffer, up to the limit specified by <b>--max-buffer-size</b>. The 91default values for these parameters can be set when <b>pcre2grep</b> is 92built; if nothing is specified, the defaults are set to 20KiB and 1MiB 93respectively. An error occurs if a line is too long and the buffer can no 94longer be expanded. 95</P> 96<P> 97The block of memory that is actually used is three times the "buffer size", to 98allow for buffering "before" and "after" lines. If the buffer size is too 99small, fewer than requested "before" and "after" lines may be output. 100</P> 101<P> 102Patterns can be no longer than 8KiB or BUFSIZ bytes, whichever is the greater. 103BUFSIZ is defined in <b><stdio.h></b>. When there is more than one pattern 104(specified by the use of <b>-e</b> and/or <b>-f</b>), each pattern is applied to 105each line in the order in which they are defined, except that all the <b>-e</b> 106patterns are tried before the <b>-f</b> patterns. 107</P> 108<P> 109By default, as soon as one pattern matches a line, no further patterns are 110considered. However, if <b>--colour</b> (or <b>--color</b>) is used to colour the 111matching substrings, or if <b>--only-matching</b>, <b>--file-offsets</b>, 112<b>--line-offsets</b>, or <b>--output</b> is used to output only the part of the 113line that matched (either shown literally, or as an offset), the behaviour is 114different. In this situation, all the patterns are applied to the line. If 115there is more than one match, the one that begins nearest to the start of the 116subject is processed; if there is more than one match at that position, the one 117with the longest matching substring is processed; if the matching substrings 118are equal, the first match found is processed. 119</P> 120<P> 121Scanning with all the patterns resumes immediately following the match, so that 122later matches on the same line can be found. Note, however, that an overlapping 123match that starts in the middle of another match will not be processed. 124</P> 125<P> 126The above behaviour was changed at release 10.41 to be more compatible with GNU 127grep. In earlier releases, <b>pcre2grep</b> did not recognize matches from 128later patterns that were earlier in the subject. 129</P> 130<P> 131Patterns that can match an empty string are accepted, but empty string 132matches are never recognized. An example is the pattern "(super)?(man)?", in 133which all components are optional. This pattern finds all occurrences of both 134"super" and "man"; the output differs from matching with "super|man" when only 135the matching substrings are being shown. 136</P> 137<P> 138If the <b>LC_ALL</b> or <b>LC_CTYPE</b> environment variable is set, 139<b>pcre2grep</b> uses the value to set a locale when calling the PCRE2 library. 140The <b>--locale</b> option can be used to override this. 141</P> 142<br><a name="SEC3" href="#TOC1">SUPPORT FOR COMPRESSED FILES</a><br> 143<P> 144Compile-time options for <b>pcre2grep</b> can set it up to use <b>libz</b> or 145<b>libbz2</b> for reading compressed files whose names end in <b>.gz</b> or 146<b>.bz2</b>, respectively. You can find out whether your <b>pcre2grep</b> binary 147has support for one or both of these file types by running it with the 148<b>--help</b> option. If the appropriate support is not present, all files are 149treated as plain text. The standard input is always so treated. If a file with 150a <b>.gz</b> or <b>.bz2</b> extension is not in fact compressed, it is read as a 151plain text file. When input is from a compressed .gz or .bz2 file, the 152<b>--line-buffered</b> option is ignored. 153</P> 154<br><a name="SEC4" href="#TOC1">BINARY FILES</a><br> 155<P> 156By default, a file that contains a binary zero byte within the first 1024 bytes 157is identified as a binary file, and is processed specially. However, if the 158newline type is specified as NUL, that is, the line terminator is a binary 159zero, the test for a binary file is not applied. See the <b>--binary-files</b> 160option for a means of changing the way binary files are handled. 161</P> 162<br><a name="SEC5" href="#TOC1">BINARY ZEROS IN PATTERNS</a><br> 163<P> 164Patterns passed from the command line are strings that are terminated by a 165binary zero, so cannot contain internal zeros. However, patterns that are read 166from a file via the <b>-f</b> option may contain binary zeros. 167</P> 168<br><a name="SEC6" href="#TOC1">OPTIONS</a><br> 169<P> 170The order in which some of the options appear can affect the output. For 171example, both the <b>-H</b> and <b>-l</b> options affect the printing of file 172names. Whichever comes later in the command line will be the one that takes 173effect. Similarly, except where noted below, if an option is given twice, the 174later setting is used. Numerical values for options may be followed by K or M, 175to signify multiplication by 1024 or 1024*1024 respectively. 176</P> 177<P> 178<b>--</b> 179This terminates the list of options. It is useful if the next item on the 180command line starts with a hyphen but is not an option. This allows for the 181processing of patterns and file names that start with hyphens. 182</P> 183<P> 184<b>-A</b> <i>number</i>, <b>--after-context=</b><i>number</i> 185Output up to <i>number</i> lines of context after each matching line. Fewer 186lines are output if the next match or the end of the file is reached, or if the 187processing buffer size has been set too small. If file names and/or line 188numbers are being output, a hyphen separator is used instead of a colon for the 189context lines (the <b>-Z</b> option can be used to change the file name 190terminator to a zero byte). A line containing "--" is output between each group 191of lines, unless they are in fact contiguous in the input file. The value of 192<i>number</i> is expected to be relatively small. When <b>-c</b> is used, 193<b>-A</b> is ignored. 194</P> 195<P> 196<b>-a</b>, <b>--text</b> 197Treat binary files as text. This is equivalent to 198<b>--binary-files</b>=<i>text</i>. 199</P> 200<P> 201<b>--allow-lookaround-bsk</b> 202PCRE2 now forbids the use of \K in lookarounds by default, in line with Perl. 203This option causes <b>pcre2grep</b> to set the PCRE2_EXTRA_ALLOW_LOOKAROUND_BSK 204option, which enables this somewhat dangerous usage. 205</P> 206<P> 207<b>-B</b> <i>number</i>, <b>--before-context=</b><i>number</i> 208Output up to <i>number</i> lines of context before each matching line. Fewer 209lines are output if the previous match or the start of the file is within 210<i>number</i> lines, or if the processing buffer size has been set too small. If 211file names and/or line numbers are being output, a hyphen separator is used 212instead of a colon for the context lines (the <b>-Z</b> option can be used to 213change the file name terminator to a zero byte). A line containing "--" is 214output between each group of lines, unless they are in fact contiguous in the 215input file. The value of <i>number</i> is expected to be relatively small. When 216<b>-c</b> is used, <b>-B</b> is ignored. 217</P> 218<P> 219<b>--binary-files=</b><i>word</i> 220Specify how binary files are to be processed. If the word is "binary" (the 221default), pattern matching is performed on binary files, but the only output is 222"Binary file <name> matches" when a match succeeds. If the word is "text", 223which is equivalent to the <b>-a</b> or <b>--text</b> option, binary files are 224processed in the same way as any other file. In this case, when a match 225succeeds, the output may be binary garbage, which can have nasty effects if 226sent to a terminal. If the word is "without-match", which is equivalent to the 227<b>-I</b> option, binary files are not processed at all; they are assumed not to 228be of interest and are skipped without causing any output or affecting the 229return code. 230</P> 231<P> 232<b>--buffer-size=</b><i>number</i> 233Set the parameter that controls how much memory is obtained at the start of 234processing for buffering files that are being scanned. See also 235<b>--max-buffer-size</b> below. 236</P> 237<P> 238<b>-C</b> <i>number</i>, <b>--context=</b><i>number</i> 239Output <i>number</i> lines of context both before and after each matching line. 240This is equivalent to setting both <b>-A</b> and <b>-B</b> to the same value. 241</P> 242<P> 243<b>-c</b>, <b>--count</b> 244Do not output lines from the files that are being scanned; instead output the 245number of lines that would have been shown, either because they matched, or, if 246<b>-v</b> is set, because they failed to match. By default, this count is 247exactly the same as the number of lines that would have been output, but if the 248<b>-M</b> (multiline) option is used (without <b>-v</b>), there may be more 249suppressed lines than the count (that is, the number of matches). 250<br> 251<br> 252If no lines are selected, the number zero is output. If several files are are 253being scanned, a count is output for each of them and the <b>-t</b> option can 254be used to cause a total to be output at the end. However, if the 255<b>--files-with-matches</b> option is also used, only those files whose counts 256are greater than zero are listed. When <b>-c</b> is used, the <b>-A</b>, 257<b>-B</b>, and <b>-C</b> options are ignored. 258</P> 259<P> 260<b>--colour</b>, <b>--color</b> 261If this option is given without any data, it is equivalent to "--colour=auto". 262If data is required, it must be given in the same shell item, separated by an 263equals sign. 264</P> 265<P> 266<b>--colour=</b><i>value</i>, <b>--color=</b><i>value</i> 267This option specifies under what circumstances the parts of a line that matched 268a pattern should be coloured in the output. It is ignored if 269<b>--file-offsets</b>, <b>--line-offsets</b>, or <b>--output</b> is set. By 270default, output is not coloured. The value for the <b>--colour</b> option (which 271is optional, see above) may be "never", "always", or "auto". In the latter 272case, colouring happens only if the standard output is connected to a terminal. 273More resources are used when colouring is enabled, because <b>pcre2grep</b> has 274to search for all possible matches in a line, not just one, in order to colour 275them all. 276<br> 277<br> 278The colour that is used can be specified by setting one of the environment 279variables PCRE2GREP_COLOUR, PCRE2GREP_COLOR, PCREGREP_COLOUR, or 280PCREGREP_COLOR, which are checked in that order. If none of these are set, 281<b>pcre2grep</b> looks for GREP_COLORS or GREP_COLOR (in that order). The value 282of the variable should be a string of two numbers, separated by a semicolon, 283except in the case of GREP_COLORS, which must start with "ms=" or "mt=" 284followed by two semicolon-separated colours, terminated by the end of the 285string or by a colon. If GREP_COLORS does not start with "ms=" or "mt=" it is 286ignored, and GREP_COLOR is checked. 287<br> 288<br> 289If the string obtained from one of the above variables contains any characters 290other than semicolon or digits, the setting is ignored and the default colour 291is used. The string is copied directly into the control string for setting 292colour on a terminal, so it is your responsibility to ensure that the values 293make sense. If no relevant environment variable is set, the default is "1;31", 294which gives red. 295</P> 296<P> 297<b>-D</b> <i>action</i>, <b>--devices=</b><i>action</i> 298If an input path is not a regular file or a directory, "action" specifies how 299it is to be processed. Valid values are "read" (the default) or "skip" 300(silently skip the path). 301</P> 302<P> 303<b>-d</b> <i>action</i>, <b>--directories=</b><i>action</i> 304If an input path is a directory, "action" specifies how it is to be processed. 305Valid values are "read" (the default in non-Windows environments, for 306compatibility with GNU grep), "recurse" (equivalent to the <b>-r</b> option), or 307"skip" (silently skip the path, the default in Windows environments). In the 308"read" case, directories are read as if they were ordinary files. In some 309operating systems the effect of reading a directory like this is an immediate 310end-of-file; in others it may provoke an error. 311</P> 312<P> 313<b>--depth-limit</b>=<i>number</i> 314See <b>--match-limit</b> below. 315</P> 316<P> 317<b>-e</b> <i>pattern</i>, <b>--regex=</b><i>pattern</i>, <b>--regexp=</b><i>pattern</i> 318Specify a pattern to be matched. This option can be used multiple times in 319order to specify several patterns. It can also be used as a way of specifying a 320single pattern that starts with a hyphen. When <b>-e</b> is used, no argument 321pattern is taken from the command line; all arguments are treated as file 322names. There is no limit to the number of patterns. They are applied to each 323line in the order in which they are defined. 324<br> 325<br> 326If <b>-f</b> is used with <b>-e</b>, the command line patterns are matched first, 327followed by the patterns from the file(s), independent of the order in which 328these options are specified. 329</P> 330<P> 331<b>--exclude</b>=<i>pattern</i> 332Files (but not directories) whose names match the pattern are skipped without 333being processed. This applies to all files, whether listed on the command line, 334obtained from <b>--file-list</b>, or by scanning a directory. The pattern is a 335PCRE2 regular expression, and is matched against the final component of the 336file name, not the entire path. The <b>-F</b>, <b>-w</b>, and <b>-x</b> options do 337not apply to this pattern. The option may be given any number of times in order 338to specify multiple patterns. If a file name matches both an <b>--include</b> 339and an <b>--exclude</b> pattern, it is excluded. There is no short form for this 340option. 341</P> 342<P> 343<b>--exclude-from=</b><i>filename</i> 344Treat each non-empty line of the file as the data for an <b>--exclude</b> 345option. What constitutes a newline when reading the file is the operating 346system's default. The <b>--newline</b> option has no effect on this option. This 347option may be given more than once in order to specify a number of files to 348read. 349</P> 350<P> 351<b>--exclude-dir</b>=<i>pattern</i> 352Directories whose names match the pattern are skipped without being processed, 353whatever the setting of the <b>--recursive</b> option. This applies to all 354directories, whether listed on the command line, obtained from 355<b>--file-list</b>, or by scanning a parent directory. The pattern is a PCRE2 356regular expression, and is matched against the final component of the directory 357name, not the entire path. The <b>-F</b>, <b>-w</b>, and <b>-x</b> options do not 358apply to this pattern. The option may be given any number of times in order to 359specify more than one pattern. If a directory matches both <b>--include-dir</b> 360and <b>--exclude-dir</b>, it is excluded. There is no short form for this 361option. 362</P> 363<P> 364<b>-F</b>, <b>--fixed-strings</b> 365Interpret each data-matching pattern as a list of fixed strings, separated by 366newlines, instead of as a regular expression. What constitutes a newline for 367this purpose is controlled by the <b>--newline</b> option. The <b>-w</b> (match 368as a word) and <b>-x</b> (match whole line) options can be used with <b>-F</b>. 369They apply to each of the fixed strings. A line is selected if any of the fixed 370strings are found in it (subject to <b>-w</b> or <b>-x</b>, if present). This 371option applies only to the patterns that are matched against the contents of 372files; it does not apply to patterns specified by any of the <b>--include</b> or 373<b>--exclude</b> options. 374</P> 375<P> 376<b>-f</b> <i>filename</i>, <b>--file=</b><i>filename</i> 377Read patterns from the file, one per line. As is the case with patterns on the 378command line, no delimiters should be used. What constitutes a newline when 379reading the file is the operating system's default interpretation of \n. The 380<b>--newline</b> option has no effect on this option. Trailing white space is 381removed from each line, and blank lines are ignored. An empty file contains no 382patterns and therefore matches nothing. Patterns read from a file in this way 383may contain binary zeros, which are treated as ordinary data characters. 384<br> 385<br> 386If this option is given more than once, all the specified files are read. A 387data line is output if any of the patterns match it. A file name can be given 388as "-" to refer to the standard input. When <b>-f</b> is used, patterns 389specified on the command line using <b>-e</b> may also be present; they are 390matched before the file's patterns. However, no pattern is taken from the 391command line; all arguments are treated as the names of paths to be searched. 392</P> 393<P> 394<b>--file-list</b>=<i>filename</i> 395Read a list of files and/or directories that are to be scanned from the given 396file, one per line. What constitutes a newline when reading the file is the 397operating system's default. Trailing white space is removed from each line, and 398blank lines are ignored. These paths are processed before any that are listed 399on the command line. The file name can be given as "-" to refer to the standard 400input. If <b>--file</b> and <b>--file-list</b> are both specified as "-", 401patterns are read first. This is useful only when the standard input is a 402terminal, from which further lines (the list of files) can be read after an 403end-of-file indication. If this option is given more than once, all the 404specified files are read. 405</P> 406<P> 407<b>--file-offsets</b> 408Instead of showing lines or parts of lines that match, show each match as an 409offset from the start of the file and a length, separated by a comma. In this 410mode, <b>--colour</b> has no effect, and no context is shown. That is, the 411<b>-A</b>, <b>-B</b>, and <b>-C</b> options are ignored. If there is more than one 412match in a line, each of them is shown separately. This option is mutually 413exclusive with <b>--output</b>, <b>--line-offsets</b>, and <b>--only-matching</b>. 414</P> 415<P> 416<b>-H</b>, <b>--with-filename</b> 417Force the inclusion of the file name at the start of output lines when 418searching a single file. The file name is not normally shown in this case. 419By default, for matching lines, the file name is followed by a colon; for 420context lines, a hyphen separator is used. The <b>-Z</b> option can be used to 421change the terminator to a zero byte. If a line number is also being output, 422it follows the file name. When the <b>-M</b> option causes a pattern to match 423more than one line, only the first is preceded by the file name. This option 424overrides any previous <b>-h</b>, <b>-l</b>, or <b>-L</b> options. 425</P> 426<P> 427<b>-h</b>, <b>--no-filename</b> 428Suppress the output file names when searching multiple files. File names are 429normally shown when multiple files are searched. By default, for matching 430lines, the file name is followed by a colon; for context lines, a hyphen 431separator is used. The <b>-Z</b> option can be used to change the terminator to 432a zero byte. If a line number is also being output, it follows the file name. 433This option overrides any previous <b>-H</b>, <b>-L</b>, or <b>-l</b> options. 434</P> 435<P> 436<b>--heap-limit</b>=<i>number</i> 437See <b>--match-limit</b> below. 438</P> 439<P> 440<b>--help</b> 441Output a help message, giving brief details of the command options and file 442type support, and then exit. Anything else on the command line is 443ignored. 444</P> 445<P> 446<b>-I</b> 447Ignore binary files. This is equivalent to 448<b>--binary-files</b>=<i>without-match</i>. 449</P> 450<P> 451<b>-i</b>, <b>--ignore-case</b> 452Ignore upper/lower case distinctions during comparisons. 453</P> 454<P> 455<b>--include</b>=<i>pattern</i> 456If any <b>--include</b> patterns are specified, the only files that are 457processed are those whose names match one of the patterns and do not match an 458<b>--exclude</b> pattern. This option does not affect directories, but it 459applies to all files, whether listed on the command line, obtained from 460<b>--file-list</b>, or by scanning a directory. The pattern is a PCRE2 regular 461expression, and is matched against the final component of the file name, not 462the entire path. The <b>-F</b>, <b>-w</b>, and <b>-x</b> options do not apply to 463this pattern. The option may be given any number of times. If a file name 464matches both an <b>--include</b> and an <b>--exclude</b> pattern, it is excluded. 465There is no short form for this option. 466</P> 467<P> 468<b>--include-from=</b><i>filename</i> 469Treat each non-empty line of the file as the data for an <b>--include</b> 470option. What constitutes a newline for this purpose is the operating system's 471default. The <b>--newline</b> option has no effect on this option. This option 472may be given any number of times; all the files are read. 473</P> 474<P> 475<b>--include-dir</b>=<i>pattern</i> 476If any <b>--include-dir</b> patterns are specified, the only directories that 477are processed are those whose names match one of the patterns and do not match 478an <b>--exclude-dir</b> pattern. This applies to all directories, whether listed 479on the command line, obtained from <b>--file-list</b>, or by scanning a parent 480directory. The pattern is a PCRE2 regular expression, and is matched against 481the final component of the directory name, not the entire path. The <b>-F</b>, 482<b>-w</b>, and <b>-x</b> options do not apply to this pattern. The option may be 483given any number of times. If a directory matches both <b>--include-dir</b> and 484<b>--exclude-dir</b>, it is excluded. There is no short form for this option. 485</P> 486<P> 487<b>-L</b>, <b>--files-without-match</b> 488Instead of outputting lines from the files, just output the names of the files 489that do not contain any lines that would have been output. Each file name is 490output once, on a separate line by default, but if the <b>-Z</b> option is set, 491they are separated by zero bytes instead of newlines. This option overrides any 492previous <b>-H</b>, <b>-h</b>, or <b>-l</b> options. 493</P> 494<P> 495<b>-l</b>, <b>--files-with-matches</b> 496Instead of outputting lines from the files, just output the names of the files 497containing lines that would have been output. Each file name is output once, on 498a separate line, but if the <b>-Z</b> option is set, they are separated by zero 499bytes instead of newlines. Searching normally stops as soon as a matching line 500is found in a file. However, if the <b>-c</b> (count) option is also used, 501matching continues in order to obtain the correct count, and those files that 502have at least one match are listed along with their counts. Using this option 503with <b>-c</b> is a way of suppressing the listing of files with no matches that 504occurs with <b>-c</b> on its own. This option overrides any previous <b>-H</b>, 505<b>-h</b>, or <b>-L</b> options. 506</P> 507<P> 508<b>--label</b>=<i>name</i> 509This option supplies a name to be used for the standard input when file names 510are being output. If not supplied, "(standard input)" is used. There is no 511short form for this option. 512</P> 513<P> 514<b>--line-buffered</b> 515When this option is given, non-compressed input is read and processed line by 516line, and the output is flushed after each write. By default, input is read in 517large chunks, unless <b>pcre2grep</b> can determine that it is reading from a 518terminal, which is currently possible only in Unix-like environments or 519Windows. Output to terminal is normally automatically flushed by the operating 520system. This option can be useful when the input or output is attached to a 521pipe and you do not want <b>pcre2grep</b> to buffer up large amounts of data. 522However, its use will affect performance, and the <b>-M</b> (multiline) option 523ceases to work. When input is from a compressed .gz or .bz2 file, 524<b>--line-buffered</b> is ignored. 525</P> 526<P> 527<b>--line-offsets</b> 528Instead of showing lines or parts of lines that match, show each match as a 529line number, the offset from the start of the line, and a length. The line 530number is terminated by a colon (as usual; see the <b>-n</b> option), and the 531offset and length are separated by a comma. In this mode, <b>--colour</b> has no 532effect, and no context is shown. That is, the <b>-A</b>, <b>-B</b>, and <b>-C</b> 533options are ignored. If there is more than one match in a line, each of them is 534shown separately. This option is mutually exclusive with <b>--output</b>, 535<b>--file-offsets</b>, and <b>--only-matching</b>. 536</P> 537<P> 538<b>--locale</b>=<i>locale-name</i> 539This option specifies a locale to be used for pattern matching. It overrides 540the value in the <b>LC_ALL</b> or <b>LC_CTYPE</b> environment variables. If no 541locale is specified, the PCRE2 library's default (usually the "C" locale) is 542used. There is no short form for this option. 543</P> 544<P> 545<b>-M</b>, <b>--multiline</b> 546Allow patterns to match more than one line. When this option is set, the PCRE2 547library is called in "multiline" mode. This allows a matched string to extend 548past the end of a line and continue on one or more subsequent lines. Patterns 549used with <b>-M</b> may usefully contain literal newline characters and internal 550occurrences of ^ and $ characters. The output for a successful match may 551consist of more than one line. The first line is the line in which the match 552started, and the last line is the line in which the match ended. If the matched 553string ends with a newline sequence, the output ends at the end of that line. 554If <b>-v</b> is set, none of the lines in a multi-line match are output. Once a 555match has been handled, scanning restarts at the beginning of the line after 556the one in which the match ended. 557<br> 558<br> 559The newline sequence that separates multiple lines must be matched as part of 560the pattern. For example, to find the phrase "regular expression" in a file 561where "regular" might be at the end of a line and "expression" at the start of 562the next line, you could use this command: 563<pre> 564 pcre2grep -M 'regular\s+expression' <file> 565</pre> 566The \s escape sequence matches any white space character, including newlines, 567and is followed by + so as to match trailing white space on the first line as 568well as possibly handling a two-character newline sequence. 569<br> 570<br> 571There is a limit to the number of lines that can be matched, imposed by the way 572that <b>pcre2grep</b> buffers the input file as it scans it. With a sufficiently 573large processing buffer, this should not be a problem, but the <b>-M</b> option 574does not work when input is read line by line (see <b>--line-buffered</b>.) 575</P> 576<P> 577<b>-m</b> <i>number</i>, <b>--max-count</b>=<i>number</i> 578Stop processing after finding <i>number</i> matching lines, or non-matching 579lines if <b>-v</b> is also set. Any trailing context lines are output after the 580final match. In multiline mode, each multiline match counts as just one line 581for this purpose. If this limit is reached when reading the standard input from 582a regular file, the file is left positioned just after the last matching line. 583If <b>-c</b> is also set, the count that is output is never greater than 584<i>number</i>. This option has no effect if used with <b>-L</b>, <b>-l</b>, or 585<b>-q</b>, or when just checking for a match in a binary file. 586</P> 587<P> 588<b>--match-limit</b>=<i>number</i> 589Processing some regular expression patterns may take a very long time to search 590for all possible matching strings. Others may require a very large amount of 591memory. There are three options that set resource limits for matching. 592<br> 593<br> 594The <b>--match-limit</b> option provides a means of limiting computing resource 595usage when processing patterns that are not going to match, but which have a 596very large number of possibilities in their search trees. The classic example 597is a pattern that uses nested unlimited repeats. Internally, PCRE2 has a 598counter that is incremented each time around its main processing loop. If the 599value set by <b>--match-limit</b> is reached, an error occurs. 600<br> 601<br> 602The <b>--heap-limit</b> option specifies, as a number of kibibytes (units of 6031024 bytes), the maximum amount of heap memory that may be used for matching. 604<br> 605<br> 606The <b>--depth-limit</b> option limits the depth of nested backtracking points, 607which indirectly limits the amount of memory that is used. The amount of memory 608needed for each backtracking point depends on the number of capturing 609parentheses in the pattern, so the amount of memory that is used before this 610limit acts varies from pattern to pattern. This limit is of use only if it is 611set smaller than <b>--match-limit</b>. 612<br> 613<br> 614There are no short forms for these options. The default limits can be set 615when the PCRE2 library is compiled; if they are not specified, the defaults 616are very large and so effectively unlimited. 617</P> 618<P> 619<b>--max-buffer-size</b>=<i>number</i> 620This limits the expansion of the processing buffer, whose initial size can be 621set by <b>--buffer-size</b>. The maximum buffer size is silently forced to be no 622smaller than the starting buffer size. 623</P> 624<P> 625<b>-N</b> <i>newline-type</i>, <b>--newline</b>=<i>newline-type</i> 626Six different conventions for indicating the ends of lines in scanned files are 627supported. For example: 628<pre> 629 pcre2grep -N CRLF 'some pattern' <file> 630</pre> 631The newline type may be specified in upper, lower, or mixed case. If the 632newline type is NUL, lines are separated by binary zero characters. The other 633types are the single-character sequences CR (carriage return) and LF 634(linefeed), the two-character sequence CRLF, an "anycrlf" type, which 635recognizes any of the preceding three types, and an "any" type, for which any 636Unicode line ending sequence is assumed to end a line. The Unicode sequences 637are the three just mentioned, plus VT (vertical tab, U+000B), FF (form feed, 638U+000C), NEL (next line, U+0085), LS (line separator, U+2028), and PS 639(paragraph separator, U+2029). 640<br> 641<br> 642When the PCRE2 library is built, a default line-ending sequence is specified. 643This is normally the standard sequence for the operating system. Unless 644otherwise specified by this option, <b>pcre2grep</b> uses the library's default. 645<br> 646<br> 647This option makes it possible to use <b>pcre2grep</b> to scan files that have 648come from other environments without having to modify their line endings. If 649the data that is being scanned does not agree with the convention set by this 650option, <b>pcre2grep</b> may behave in strange ways. Note that this option does 651not apply to files specified by the <b>-f</b>, <b>--exclude-from</b>, or 652<b>--include-from</b> options, which are expected to use the operating system's 653standard newline sequence. 654</P> 655<P> 656<b>-n</b>, <b>--line-number</b> 657Precede each output line by its line number in the file, followed by a colon 658for matching lines or a hyphen for context lines. If the file name is also 659being output, it precedes the line number. When the <b>-M</b> option causes a 660pattern to match more than one line, only the first is preceded by its line 661number. This option is forced if <b>--line-offsets</b> is used. 662</P> 663<P> 664<b>--no-jit</b> 665If the PCRE2 library is built with support for just-in-time compiling (which 666speeds up matching), <b>pcre2grep</b> automatically makes use of this, unless it 667was explicitly disabled at build time. This option can be used to disable the 668use of JIT at run time. It is provided for testing and working round problems. 669It should never be needed in normal use. 670</P> 671<P> 672<b>-O</b> <i>text</i>, <b>--output</b>=<i>text</i> 673When there is a match, instead of outputting the line that matched, output just 674the text specified in this option, followed by an operating-system standard 675newline. In this mode, <b>--colour</b> has no effect, and no context is shown. 676That is, the <b>-A</b>, <b>-B</b>, and <b>-C</b> options are ignored. The 677<b>--newline</b> option has no effect on this option, which is mutually 678exclusive with <b>--only-matching</b>, <b>--file-offsets</b>, and 679<b>--line-offsets</b>. However, like <b>--only-matching</b>, if there is more 680than one match in a line, each of them causes a line of output. 681<br> 682<br> 683Escape sequences starting with a dollar character may be used to insert the 684contents of the matched part of the line and/or captured substrings into the 685text. 686<br> 687<br> 688$<digits> or ${<digits>} is replaced by the captured substring of the given 689decimal number; zero substitutes the whole match. If the number is greater than 690the number of capturing substrings, or if the capture is unset, the replacement 691is empty. 692<br> 693<br> 694$a is replaced by bell; $b by backspace; $e by escape; $f by form feed; $n by 695newline; $r by carriage return; $t by tab; $v by vertical tab. 696<br> 697<br> 698$o<digits> or $o{<digits>} is replaced by the character whose code point is the 699given octal number. In the first form, up to three octal digits are processed. 700When more digits are needed in Unicode mode to specify a wide character, the 701second form must be used. 702<br> 703<br> 704$x<digits> or $x{<digits>} is replaced by the character represented by the 705given hexadecimal number. In the first form, up to two hexadecimal digits are 706processed. When more digits are needed in Unicode mode to specify a wide 707character, the second form must be used. 708<br> 709<br> 710Any other character is substituted by itself. In particular, $$ is replaced by 711a single dollar. 712</P> 713<P> 714<b>-o</b>, <b>--only-matching</b> 715Show only the part of the line that matched a pattern instead of the whole 716line. In this mode, no context is shown. That is, the <b>-A</b>, <b>-B</b>, and 717<b>-C</b> options are ignored. If there is more than one match in a line, each 718of them is shown separately, on a separate line of output. If <b>-o</b> is 719combined with <b>-v</b> (invert the sense of the match to find non-matching 720lines), no output is generated, but the return code is set appropriately. If 721the matched portion of the line is empty, nothing is output unless the file 722name or line number are being printed, in which case they are shown on an 723otherwise empty line. This option is mutually exclusive with <b>--output</b>, 724<b>--file-offsets</b> and <b>--line-offsets</b>. 725</P> 726<P> 727<b>-o</b><i>number</i>, <b>--only-matching</b>=<i>number</i> 728Show only the part of the line that matched the capturing parentheses of the 729given number. Up to 50 capturing parentheses are supported by default. This 730limit can be changed via the <b>--om-capture</b> option. A pattern may contain 731any number of capturing parentheses, but only those whose number is within the 732limit can be accessed by <b>-o</b>. An error occurs if the number specified by 733<b>-o</b> is greater than the limit. 734<br> 735<br> 736-o0 is the same as <b>-o</b> without a number. Because these options can be 737given without an argument (see above), if an argument is present, it must be 738given in the same shell item, for example, -o3 or --only-matching=2. The 739comments given for the non-argument case above also apply to this option. If 740the specified capturing parentheses do not exist in the pattern, or were not 741set in the match, nothing is output unless the file name or line number are 742being output. 743<br> 744<br> 745If this option is given multiple times, multiple substrings are output for each 746match, in the order the options are given, and all on one line. For example, 747-o3 -o1 -o3 causes the substrings matched by capturing parentheses 3 and 1 and 748then 3 again to be output. By default, there is no separator (but see the next 749but one option). 750</P> 751<P> 752<b>--om-capture</b>=<i>number</i> 753Set the number of capturing parentheses that can be accessed by <b>-o</b>. The 754default is 50. 755</P> 756<P> 757<b>--om-separator</b>=<i>text</i> 758Specify a separating string for multiple occurrences of <b>-o</b>. The default 759is an empty string. Separating strings are never coloured. 760</P> 761<P> 762<b>-q</b>, <b>--quiet</b> 763Work quietly, that is, display nothing except error messages. The exit 764status indicates whether or not any matches were found. 765</P> 766<P> 767<b>-r</b>, <b>--recursive</b> 768If any given path is a directory, recursively scan the files it contains, 769taking note of any <b>--include</b> and <b>--exclude</b> settings. By default, a 770directory is read as a normal file; in some operating systems this gives an 771immediate end-of-file. This option is a shorthand for setting the <b>-d</b> 772option to "recurse". 773</P> 774<P> 775<b>--recursion-limit</b>=<i>number</i> 776This is an obsolete synonym for <b>--depth-limit</b>. See <b>--match-limit</b> 777above for details. 778</P> 779<P> 780<b>-s</b>, <b>--no-messages</b> 781Suppress error messages about non-existent or unreadable files. Such files are 782quietly skipped. However, the return code is still 2, even if matches were 783found in other files. 784</P> 785<P> 786<b>-t</b>, <b>--total-count</b> 787This option is useful when scanning more than one file. If used on its own, 788<b>-t</b> suppresses all output except for a grand total number of matching 789lines (or non-matching lines if <b>-v</b> is used) in all the files. If <b>-t</b> 790is used with <b>-c</b>, a grand total is output except when the previous output 791is just one line. In other words, it is not output when just one file's count 792is listed. If file names are being output, the grand total is preceded by 793"TOTAL:". Otherwise, it appears as just another number. The <b>-t</b> option is 794ignored when used with <b>-L</b> (list files without matches), because the grand 795total would always be zero. 796</P> 797<P> 798<b>-u</b>, <b>--utf</b> 799Operate in UTF-8 mode. This option is available only if PCRE2 has been compiled 800with UTF-8 support. All patterns (including those for any <b>--exclude</b> and 801<b>--include</b> options) and all lines that are scanned must be valid strings 802of UTF-8 characters. If an invalid UTF-8 string is encountered, an error 803occurs. 804</P> 805<P> 806<b>-U</b>, <b>--utf-allow-invalid</b> 807As <b>--utf</b>, but in addition subject lines may contain invalid UTF-8 code 808unit sequences. These can never form part of any pattern match. Patterns 809themselves, however, must still be valid UTF-8 strings. This facility allows 810valid UTF-8 strings to be sought within arbitrary byte sequences in executable 811or other binary files. For more details about matching in non-valid UTF-8 812strings, see the 813<a href="pcre2unicode.html"><b>pcre2unicode</b>(3)</a> 814documentation. 815</P> 816<P> 817<b>-V</b>, <b>--version</b> 818Write the version numbers of <b>pcre2grep</b> and the PCRE2 library to the 819standard output and then exit. Anything else on the command line is 820ignored. 821</P> 822<P> 823<b>-v</b>, <b>--invert-match</b> 824Invert the sense of the match, so that lines which do <i>not</i> match any of 825the patterns are the ones that are found. When this option is set, options such 826as <b>--only-matching</b> and <b>--output</b>, which specify parts of a match 827that are to be output, are ignored. 828</P> 829<P> 830<b>-w</b>, <b>--word-regex</b>, <b>--word-regexp</b> 831Force the patterns only to match "words". That is, there must be a word 832boundary at the start and end of each matched string. This is equivalent to 833having "\b(?:" at the start of each pattern, and ")\b" at the end. This 834option applies only to the patterns that are matched against the contents of 835files; it does not apply to patterns specified by any of the <b>--include</b> or 836<b>--exclude</b> options. 837</P> 838<P> 839<b>-x</b>, <b>--line-regex</b>, <b>--line-regexp</b> 840Force the patterns to start matching only at the beginnings of lines, and in 841addition, require them to match entire lines. In multiline mode the match may 842be more than one line. This is equivalent to having "^(?:" at the start of each 843pattern and ")$" at the end. This option applies only to the patterns that are 844matched against the contents of files; it does not apply to patterns specified 845by any of the <b>--include</b> or <b>--exclude</b> options. 846</P> 847<P> 848<b>-Z</b>, <b>--null</b> 849Terminate files names in the regular output with a zero byte (the NUL 850character) instead of what would normally appear. This is useful when file 851names contain unusual characters such as colons, hyphens, or even newlines. The 852option does not apply to file names in error messages. 853</P> 854<br><a name="SEC7" href="#TOC1">ENVIRONMENT VARIABLES</a><br> 855<P> 856The environment variables <b>LC_ALL</b> and <b>LC_CTYPE</b> are examined, in that 857order, for a locale. The first one that is set is used. This can be overridden 858by the <b>--locale</b> option. If no locale is set, the PCRE2 library's default 859(usually the "C" locale) is used. 860</P> 861<br><a name="SEC8" href="#TOC1">NEWLINES</a><br> 862<P> 863The <b>-N</b> (<b>--newline</b>) option allows <b>pcre2grep</b> to scan files with 864newline conventions that differ from the default. This option affects only the 865way scanned files are processed. It does not affect the interpretation of files 866specified by the <b>-f</b>, <b>--file-list</b>, <b>--exclude-from</b>, or 867<b>--include-from</b> options. 868</P> 869<P> 870Any parts of the scanned input files that are written to the standard output 871are copied with whatever newline sequences they have in the input. However, if 872the final line of a file is output, and it does not end with a newline 873sequence, a newline sequence is added. If the newline setting is CR, LF, CRLF 874or NUL, that line ending is output; for the other settings (ANYCRLF or ANY) a 875single NL is used. 876</P> 877<P> 878The newline setting does not affect the way in which <b>pcre2grep</b> writes 879newlines in informational messages to the standard output and error streams. 880Under Windows, the standard output is set to be binary, so that "\r\n" at the 881ends of output lines that are copied from the input is not converted to 882"\r\r\n" by the C I/O library. This means that any messages written to the 883standard output must end with "\r\n". For all other operating systems, and 884for all messages to the standard error stream, "\n" is used. 885</P> 886<br><a name="SEC9" href="#TOC1">OPTIONS COMPATIBILITY</a><br> 887<P> 888Many of the short and long forms of <b>pcre2grep</b>'s options are the same 889as in the GNU <b>grep</b> program. Any long option of the form 890<b>--xxx-regexp</b> (GNU terminology) is also available as <b>--xxx-regex</b> 891(PCRE2 terminology). However, the <b>--depth-limit</b>, <b>--file-list</b>, 892<b>--file-offsets</b>, <b>--heap-limit</b>, <b>--include-dir</b>, 893<b>--line-offsets</b>, <b>--locale</b>, <b>--match-limit</b>, <b>-M</b>, 894<b>--multiline</b>, <b>-N</b>, <b>--newline</b>, <b>--om-separator</b>, 895<b>--output</b>, <b>-u</b>, <b>--utf</b>, <b>-U</b>, and <b>--utf-allow-invalid</b> 896options are specific to <b>pcre2grep</b>, as is the use of the 897<b>--only-matching</b> option with a capturing parentheses number. 898</P> 899<P> 900Although most of the common options work the same way, a few are different in 901<b>pcre2grep</b>. For example, the <b>--include</b> option's argument is a glob 902for GNU <b>grep</b>, but a regular expression for <b>pcre2grep</b>. If both the 903<b>-c</b> and <b>-l</b> options are given, GNU grep lists only file names, 904without counts, but <b>pcre2grep</b> gives the counts as well. 905</P> 906<br><a name="SEC10" href="#TOC1">OPTIONS WITH DATA</a><br> 907<P> 908There are four different ways in which an option with data can be specified. 909If a short form option is used, the data may follow immediately, or (with one 910exception) in the next command line item. For example: 911<pre> 912 -f/some/file 913 -f /some/file 914</pre> 915The exception is the <b>-o</b> option, which may appear with or without data. 916Because of this, if data is present, it must follow immediately in the same 917item, for example -o3. 918</P> 919<P> 920If a long form option is used, the data may appear in the same command line 921item, separated by an equals character, or (with two exceptions) it may appear 922in the next command line item. For example: 923<pre> 924 --file=/some/file 925 --file /some/file 926</pre> 927Note, however, that if you want to supply a file name beginning with ~ as data 928in a shell command, and have the shell expand ~ to a home directory, you must 929separate the file name from the option, because the shell does not treat ~ 930specially unless it is at the start of an item. 931</P> 932<P> 933The exceptions to the above are the <b>--colour</b> (or <b>--color</b>) and 934<b>--only-matching</b> options, for which the data is optional. If one of these 935options does have data, it must be given in the first form, using an equals 936character. Otherwise <b>pcre2grep</b> will assume that it has no data. 937</P> 938<br><a name="SEC11" href="#TOC1">USING PCRE2'S CALLOUT FACILITY</a><br> 939<P> 940<b>pcre2grep</b> has, by default, support for calling external programs or 941scripts or echoing specific strings during matching by making use of PCRE2's 942callout facility. However, this support can be completely or partially disabled 943when <b>pcre2grep</b> is built. You can find out whether your binary has support 944for callouts by running it with the <b>--help</b> option. If callout support is 945completely disabled, all callouts in patterns are ignored by <b>pcre2grep</b>. 946If the facility is partially disabled, calling external programs is not 947supported, and callouts that request it are ignored. 948</P> 949<P> 950A callout in a PCRE2 pattern is of the form (?C<arg>) where the argument is 951either a number or a quoted string (see the 952<a href="pcre2callout.html"><b>pcre2callout</b></a> 953documentation for details). Numbered callouts are ignored by <b>pcre2grep</b>; 954only callouts with string arguments are useful. 955</P> 956<br><b> 957Echoing a specific string 958</b><br> 959<P> 960Starting the callout string with a pipe character invokes an echoing facility 961that avoids calling an external program or script. This facility is always 962available, provided that callouts were not completely disabled when 963<b>pcre2grep</b> was built. The rest of the callout string is processed as a 964zero-terminated string, which means it should not contain any internal binary 965zeros. It is written to the output, having first been passed through the same 966escape processing as text from the <b>--output</b> (<b>-O</b>) option (see 967above). However, $0 cannot be used to insert a matched substring because the 968match is still in progress. Instead, the single character '0' is inserted. Any 969syntax errors in the string (for example, a dollar not followed by another 970character) causes the callout to be ignored. No terminator is added to the 971output string, so if you want a newline, you must include it explicitly using 972the escape $n. For example: 973<pre> 974 pcre2grep '(.)(..(.))(?C"|[$1] [$2] [$3]$n")' <some file> 975</pre> 976Matching continues normally after the string is output. If you want to see only 977the callout output but not any output from an actual match, you should end the 978pattern with (*FAIL). 979</P> 980<br><b> 981Calling external programs or scripts 982</b><br> 983<P> 984This facility can be independently disabled when <b>pcre2grep</b> is built. It 985is supported for Windows, where a call to <b>_spawnvp()</b> is used, for VMS, 986where <b>lib$spawn()</b> is used, and for any Unix-like environment where 987<b>fork()</b> and <b>execv()</b> are available. 988</P> 989<P> 990If the callout string does not start with a pipe (vertical bar) character, it 991is parsed into a list of substrings separated by pipe characters. The first 992substring must be an executable name, with the following substrings specifying 993arguments: 994<pre> 995 executable_name|arg1|arg2|... 996</pre> 997Any substring (including the executable name) may contain escape sequences 998started by a dollar character. These are the same as for the <b>--output</b> 999(<b>-O</b>) option documented above, except that $0 cannot insert the matched 1000string because the match is still in progress. Instead, the character '0' 1001is inserted. If you need a literal dollar or pipe character in any 1002substring, use $$ or $| respectively. Here is an example: 1003<pre> 1004 echo -e "abcde\n12345" | pcre2grep \ 1005 '(?x)(.)(..(.)) 1006 (?C"/bin/echo|Arg1: [$1] [$2] [$3]|Arg2: $|${1}$| ($4)")()' - 1007 1008 Output: 1009 1010 Arg1: [a] [bcd] [d] Arg2: |a| () 1011 abcde 1012 Arg1: [1] [234] [4] Arg2: |1| () 1013 12345 1014</pre> 1015The parameters for the system call that is used to run the program or script 1016are zero-terminated strings. This means that binary zero characters in the 1017callout argument will cause premature termination of their substrings, and 1018therefore should not be present. Any syntax errors in the string (for example, 1019a dollar not followed by another character) causes the callout to be ignored. 1020If running the program fails for any reason (including the non-existence of the 1021executable), a local matching failure occurs and the matcher backtracks in the 1022normal way. 1023</P> 1024<br><a name="SEC12" href="#TOC1">MATCHING ERRORS</a><br> 1025<P> 1026It is possible to supply a regular expression that takes a very long time to 1027fail to match certain lines. Such patterns normally involve nested indefinite 1028repeats, for example: (a+)*\d when matched against a line of a's with no final 1029digit. The PCRE2 matching function has a resource limit that causes it to abort 1030in these circumstances. If this happens, <b>pcre2grep</b> outputs an error 1031message and the line that caused the problem to the standard error stream. If 1032there are more than 20 such errors, <b>pcre2grep</b> gives up. 1033</P> 1034<P> 1035The <b>--match-limit</b> option of <b>pcre2grep</b> can be used to set the 1036overall resource limit. There are also other limits that affect the amount of 1037memory used during matching; see the discussion of <b>--heap-limit</b> and 1038<b>--depth-limit</b> above. 1039</P> 1040<br><a name="SEC13" href="#TOC1">DIAGNOSTICS</a><br> 1041<P> 1042Exit status is 0 if any matches were found, 1 if no matches were found, and 2 1043for syntax errors, overlong lines, non-existent or inaccessible files (even if 1044matches were found in other files) or too many matching errors. Using the 1045<b>-s</b> option to suppress error messages about inaccessible files does not 1046affect the return code. 1047</P> 1048<P> 1049When run under VMS, the return code is placed in the symbol PCRE2GREP_RC 1050because VMS does not distinguish between exit(0) and exit(1). 1051</P> 1052<br><a name="SEC14" href="#TOC1">SEE ALSO</a><br> 1053<P> 1054<b>pcre2pattern</b>(3), <b>pcre2syntax</b>(3), <b>pcre2callout</b>(3), 1055<b>pcre2unicode</b>(3). 1056</P> 1057<br><a name="SEC15" href="#TOC1">AUTHOR</a><br> 1058<P> 1059Philip Hazel 1060<br> 1061Retired from University Computing Service 1062<br> 1063Cambridge, England. 1064<br> 1065</P> 1066<br><a name="SEC16" href="#TOC1">REVISION</a><br> 1067<P> 1068Last updated: 21 November 2022 1069<br> 1070Copyright © 1997-2022 University of Cambridge. 1071<br> 1072<p> 1073Return to the <a href="index.html">PCRE2 index page</a>. 1074</p> 1075