DESCRIPTION

pcretest was written as a test program for the PCRE regular expression library itself, but it can also be used for experimenting with regular expressions. This document describes the features of the test program; for details of the regular expressions themselves, see the pcrepattern(3) documentation. For details of the PCRE library function calls and their options, see the pcreapi(3) , pcre16(3) and pcre32(3) documentation.

The input for pcretest is a sequence of regular expression patterns and strings to be matched, as described below. The output shows the result of each match. Options on the command line and the patterns control PCRE options and exactly what is output.

As PCRE has evolved, it has acquired many different features, and as a result, pcretest now has rather a lot of obscure options for testing every possible feature. Some of these options are specifically designed for use in conjunction with the test script and data files that are distributed as part of PCRE, and are unlikely to be of use otherwise. They are all documented here, but without much justification.

INPUT DATA FORMAT

Input to pcretest is processed line by line, either by calling the C library's fgets() function, or via the libreadline library (see below). In Unix-like environments, fgets() treats any bytes other than newline as data characters. However, in some Windows environments character 26 (hex 1A) causes an immediate end of file, and no further data is read. For maximum portability, therefore, it is safest to use only ASCII characters in pcretest input files.

PCRE's 8-BIT, 16-BIT AND 32-BIT LIBRARIES

From release 8.30, two separate PCRE libraries can be built. The original one supports 8-bit character strings, whereas the newer 16-bit library supports character strings encoded in 16-bit units. From release 8.32, a third library can be built, supporting character strings encoded in 32-bit units. The pcretest program can be used to test all three libraries. However, it is itself still an 8-bit program, reading 8-bit input and writing 8-bit output. When testing the 16-bit or 32-bit library, the patterns and data strings are converted to 16- or 32-bit format before being passed to the PCRE library functions. Results are converted to 8-bit for output.

References to functions and structures of the form pcre[16|32]_xx below mean "pcre_xx when using the 8-bit library, pcre16_xx when using the 16-bit library, or pcre32_xx when using the 32-bit library".

COMMAND LINE OPTIONS

DESCRIPTION

If pcretest is given two filename arguments, it reads from the first and writes to the second. If it is given only one filename argument, it reads from that file and writes to stdout. Otherwise, it reads from stdin and writes to stdout, and prompts for each line of input, using "re>" to prompt for regular expressions, and "data>" to prompt for data lines.

When pcretest is built, a configuration option can specify that it should be linked with the libreadline library. When this is done, if the input is from a terminal, it is read using the readline() function. This provides line-editing and history facilities. The output from the −help option states whether or not readline() will be used.

The program handles any number of sets of input on a single input file. Each set starts with a regular expression, and continues with any number of data lines to be matched against the pattern.

Each data line is matched separately and independently. If you want to do multi-line matches, you have to use the \n escape sequence (or \r or \r\n, etc., depending on the newline setting) in a single line of input to encode the newline sequences. There is no limit on the length of data lines; the input buffer is automatically extended if it is too small.

An empty line signals the end of the data lines, at which point a new regular expression is read. The regular expressions are given enclosed in any non-alphanumeric delimiters other than backslash, for example:

/(a|bc)x+yz/

White space before the initial delimiter is ignored. A regular expression may be continued over several input lines, in which case the newline characters are included within it. It is possible to include the delimiter within the pattern by escaping it, for example

/abc\/def/

If you do so, the escape and the delimiter form part of the pattern, but since delimiters are always non-alphanumeric, this does not affect its interpretation. If the terminating delimiter is immediately followed by a backslash, for example,

/abc/\

then a backslash is added to the end of the pattern. This is done to provide a way of testing the error condition that arises if a pattern finishes with a backslash, because

/abc\/

is interpreted as the first line of a pattern that starts with "abc/", causing pcretest to read the next line as a continuation of the regular expression.

PATTERN MODIFIERS

A pattern may be followed by any number of modifiers, which are mostly single characters, though some of these can be qualified by further characters. Following Perl usage, these are referred to below as, for example, "the /i modifier", even though the delimiter of the pattern need not always be a slash, and no slash is used when writing modifiers. White space may appear between the final pattern delimiter and the first modifier, and between the modifiers themselves. For reference, here is a complete list of modifiers. They fall into several groups that are described in detail in the following sections.

/caseless/i

/^abc/m<CRLF>

/pattern/Lfr_FR

 1  normal match only
 2  soft partial match only
 3  normal match and soft partial match
 4  hard partial match only
 6  soft and hard partial match
 7  all three modes (default)

 0   the default ASCII tables, as distributed in
       pcre_chartables.c.dist
 1   a set of tables defining ISO 8859 characters

 /i    REG_ICASE
 /m    REG_NEWLINE
 /N    REG_NOSUB
 /s    REG_DOTALL     )
 /U    REG_UNGREEDY   ) These options are not part of
 /W    REG_UCP        )   the POSIX standard
 /8    REG_UTF8       )

DATA LINES

Before each data line is passed to pcre[16|32]_exec(), leading and trailing white space is removed, and it is then scanned for \ escapes. Some of these are pretty esoteric features, intended for checking out some of the more complicated features of PCRE. If you are just testing "ordinary" regular expressions, you probably don't need any of these. The following escapes are recognized:

The use of \x{hh...} is not dependent on the use of the /8 modifier on the pattern. It is recognized always. There may be any number of hexadecimal digits inside the braces; invalid values provoke error messages.

Note that \xhh specifies one byte rather than one character in UTF-8 mode; this makes it possible to construct invalid UTF-8 sequences for testing purposes. On the other hand, \x{hh} is interpreted as a UTF-8 character in UTF-8 mode, generating more than one byte if the value is greater than 127. When testing the 8-bit library not in UTF-8 mode, \x{hh} generates one byte for values less than 256, and causes an error for greater values.

In UTF-16 mode, all 4-digit \x{hhhh} values are accepted. This makes it possible to construct invalid UTF-16 sequences for testing purposes.

In UTF-32 mode, all 4- to 8-digit \x{...} values are accepted. This makes it possible to construct invalid UTF-32 sequences for testing purposes.

The escapes that specify line ending sequences are literal strings, exactly as shown. No more than one newline setting should be present in any data line.

A backslash followed by anything else just escapes the anything else. If the very last character is a backslash, it is ignored. This gives a way of passing an empty line as data, since a real empty line terminates the data input.

The \J escape provides a way of setting the maximum stack size that is used by the just-in-time optimization code. It is ignored if JIT optimization is not being used. Providing a stack that is larger than the default 32K is necessary only for very complicated patterns.

If \M is present, pcretest calls pcre[16|32]_exec() several times, with different values in the match_limit and match_limit_recursion fields of the pcre[16|32]_extra data structure, until it finds the minimum numbers for each parameter that allow pcre[16|32]_exec() to complete without error. Because this is testing a specific feature of the normal interpretive pcre[16|32]_exec() execution, the use of any JIT optimization that might have been set up by the /S+ qualifier of −s+ option is disabled.

The match_limit number is a measure of the amount of backtracking that takes place, and checking it out can be instructive. For most simple matches, the number is quite small, but for patterns with very large numbers of matching possibilities, it can become large very quickly with increasing length of subject string. The match_limit_recursion number is a measure of how much stack (or, if PCRE is compiled with NO_RECURSE, how much heap) memory is needed to complete the match attempt.

When \O is used, the value specified may be higher or lower than the size set by the −O command line option (or defaulted to 45); \O applies only to the call of pcre[16|32]_exec() for the line in which it appears.

If the /P modifier was present on the pattern, causing the POSIX wrapper API to be used, the only option-setting sequences that have any effect are \B, \N, and \Z, causing REG_NOTBOL, REG_NOTEMPTY, and REG_NOTEOL, respectively, to be passed to regexec().

THE ALTERNATIVE MATCHING FUNCTION

By default, pcretest uses the standard PCRE matching function, pcre[16|32]_exec() to match each data line. PCRE also supports an alternative matching function, pcre[16|32]_dfa_test(), which operates in a different way, and has some restrictions. The differences between the two functions are described in the pcrematching(3) documentation.

If a data line contains the \D escape sequence, or if the command line contains the −dfa option, the alternative matching function is used. This function finds all possible matches at a given point. If, however, the \F escape sequence is present in the data line, it stops after the first match is found. This is always the shortest possible match.

DEFAULT OUTPUT FROM PCRETEST

This section describes the output when the normal matching function, pcre[16|32]_exec(), is being used.

When a match succeeds, pcretest outputs the list of captured substrings that pcre[16|32]_exec() returns, starting with number 0 for the string that matched the whole pattern. Otherwise, it outputs "No match" when the return is PCRE_ERROR_NOMATCH, and "Partial match:" followed by the partially matching substring when pcre[16|32]_exec() returns PCRE_ERROR_PARTIAL. (Note that this is the entire substring that was inspected during the partial match; it may include characters before the actual match start if a lookbehind assertion, \K, \b, or \B was involved.) For any other return, pcretest outputs the PCRE negative error number and a short descriptive phrase. If the error is a failed UTF string check, the offset of the start of the failing character and the reason code are also output, provided that the size of the output vector is at least two. Here is an example of an interactive pcretest run.

 $ pcretest
 PCRE version 8.13 2011-04-30

   re> /^abc(\d+)/
 data> abc123
  0: abc123
  1: 123
 data> xyz
 No match

Unset capturing substrings that are not followed by one that is set are not returned by pcre[16|32]_exec(), and are not shown by pcretest. In the following example, there are two capturing substrings, but when the first data line is matched, the second, unset substring is not shown. An "internal" unset substring is shown as "<unset>", as for the second data line.

   re> /(a)|(b)/
 data> a
  0: a
  1: a
 data> b
  0: b
  1: <unset>
  2: b

If the strings contain any non-printing characters, they are output as \xhh escapes if the value is less than 256 and UTF mode is not set. Otherwise they are output as \x{hh...} escapes. See below for the definition of non-printing characters. If the pattern has the /+ modifier, the output for substring 0 is followed by the the rest of the subject string, identified by "0+" like this:

   re> /cat/+
 data> cataract
  0: cat
  0+ aract

If the pattern has the /g or /G modifier, the results of successive matching attempts are output in sequence, like this:

   re> /\Bi(\w\w)/g
 data> Mississippi
  0: iss
  1: ss
  0: iss
  1: ss
  0: ipp
  1: pp

"No match" is output only if the first match attempt fails. Here is an example of a failure message (the offset 4 that is specified by \>4 is past the end of the subject string):

   re> /xyz/
 data> xyz\>4
 Error -24 (bad offset value)

If any of the sequences \C, \G, or \L are present in a data line that is successfully matched, the substrings extracted by the convenience functions are output with C, G, or L after the string number instead of a colon. This is in addition to the normal full list. The string length (that is, the return from the extraction function) is given in parentheses after each string for \C and \G.

Note that whereas patterns can be continued over several lines (a plain ">" prompt is used for continuations), data lines may not. However newlines can be included in data by means of the \n escape (or \r, \r\n, etc., depending on the newline sequence setting).

OUTPUT FROM THE ALTERNATIVE MATCHING FUNCTION

When the alternative matching function, pcre[16|32]_dfa_exec(), is used (by means of the \D escape sequence or the −dfa command line option), the output consists of a list of all the matches that start at the first point in the subject where there is at least one match. For example:

   re> /(tang|tangerine|tan)/
 data> yellow tangerine\D
  0: tangerine
  1: tang
  2: tan

(Using the normal matching function on this data finds only "tang".) The longest matching string is always given first (and numbered zero). After a PCRE_ERROR_PARTIAL return, the output is "Partial match:", followed by the partially matching substring. (Note that this is the entire substring that was inspected during the partial match; it may include characters before the actual match start if a lookbehind assertion, \K, \b, or \B was involved.)

If /g is present on the pattern, the search for further matches resumes at the end of the longest match. For example:

   re> /(tang|tangerine|tan)/g
 data> yellow tangerine and tangy sultana\D
  0: tangerine
  1: tang
  2: tan
  0: tang
  1: tan
  0: tan

Since the matching function does not support substring capture, the escape sequences that are concerned with captured substrings are not relevant.

RESTARTING AFTER A PARTIAL MATCH

When the alternative matching function has given the PCRE_ERROR_PARTIAL return, indicating that the subject partially matched the pattern, you can restart the match with additional subject data by means of the \R escape sequence. For example:

   re> /^\d?\d(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\d\d$/
 data> 23ja\P\D
 Partial match: 23ja
 data> n05\R\D
  0: n05

For further information about partial matching, see the pcrepartial(3) documentation.

CALLOUTS

If the pattern contains any callout requests, pcretest's callout function is called during matching. This works with both matching functions. By default, the called function displays the callout number, the start and current positions in the text at the callout time, and the next pattern item to be tested. For example:

 --->pqrabcdef
   0    ^  ^     \d

This output indicates that callout number 0 occurred for a match attempt starting at the fourth character of the subject string, when the pointer was at the seventh character of the data, and when the next pattern item was \d. Just one circumflex is output if the start and current positions are the same.

Callouts numbered 255 are assumed to be automatic callouts, inserted as a result of the /C pattern modifier. In this case, instead of showing the callout number, the offset in the pattern, preceded by a plus, is output. For example:

   re> /\d?[A-E]\*/C
 data> E*
 --->E*
  +0 ^      \d?
  +3 ^      [A-E]
  +8 ^^     \*
 +10 ^ ^
  0: E*

If a pattern contains (*MARK) items, an additional line is output whenever a change of latest mark is passed to the callout function. For example:

   re> /a(*MARK:X)bc/C
 data> abc
 --->abc
  +0 ^       a
  +1 ^^      (*MARK:X)
 +10 ^^      b
 Latest Mark: X
 +11 ^ ^     c
 +12 ^  ^
  0: abc

The mark changes between matching "a" and "b", but stays the same for the rest of the match, so nothing more is output. If, as a result of backtracking, the mark reverts to being unset, the text "<unset>" is output.

The callout function in pcretest returns zero (carry on matching) by default, but you can use a \C item in a data line (as described above) to change this and other parameters of the callout.

Inserting callouts can be helpful when using pcretest to check complicated regular expressions. For further information about callouts, see the pcrecallout(3) documentation.

NON-PRINTING CHARACTERS

When pcretest is outputting text in the compiled version of a pattern, bytes other than 32-126 are always treated as non-printing characters are are therefore shown as hex escapes.

When pcretest is outputting text that is a matched part of a subject string, it behaves in the same way, unless a different locale has been set for the pattern (using the /L modifier). In this case, the isprint() function to distinguish printing and non-printing characters.

SAVING AND RELOADING COMPILED PATTERNS

The facilities described in this section are not available when the POSIX interface to PCRE is being used, that is, when the /P pattern modifier is specified.

When the POSIX interface is not in use, you can cause pcretest to write a compiled pattern to a file, by following the modifiers with > and a file name. For example:

/pattern/im >/some/file

See the pcreprecompile(3) documentation for a discussion about saving and re-using compiled patterns. Note that if the pattern was successfully studied with JIT optimization, the JIT data cannot be saved.

The data that is written is binary. The first eight bytes are the length of the compiled pattern data followed by the length of the optional study data, each written as four bytes in big-endian order (most significant byte first). If there is no study data (either the pattern was not studied, or studying did not return any data), the second length is zero. The lengths are followed by an exact copy of the compiled pattern. If there is additional study data, this (excluding any JIT data) follows immediately after the compiled pattern. After writing the file, pcretest expects to read a new pattern.

A saved pattern can be reloaded into pcretest by specifying < and a file name instead of a pattern. The name of the file must not contain a < character, as otherwise pcretest will interpret the line as a pattern delimited by < characters. For example:

  re> </some/file
 Compiled pattern loaded from /some/file
 No study data

If the pattern was previously studied with the JIT optimization, the JIT information cannot be saved and restored, and so is lost. When the pattern has been loaded, pcretest proceeds to read data lines in the usual way.

You can copy a file written by pcretest to a different host and reload it there, even if the new host has opposite endianness to the one on which the pattern was compiled. For example, you can compile on an i86 machine and run on a SPARC machine. When a pattern is reloaded on a host with different endianness, the confirmation message is changed to:

Compiled pattern (byte-inverted) loaded from /some/file

The test suite contains some saved pre-compiled patterns with different endianness. These are reloaded using "<!" instead of just "<". This suppresses the "(byte-inverted)" text so that the output is the same on all hosts. It also forces debugging output once the pattern has been reloaded.

File names for saving and reloading can be absolute or relative, but note that the shell facility of expanding a file name that starts with a tilde (~) is not available.

The ability to save and reload files in pcretest is intended for testing and experimentation. It is not intended for production use because only a single pattern can be written to a file. Furthermore, there is no facility for supplying custom character tables for use with a reloaded pattern. If the original pattern was compiled with custom tables, an attempt to match a subject string using a reloaded pattern is likely to cause pcretest to crash. Finally, if you attempt to load a file that is not in the correct format, the result is undefined.

SEE ALSO

pcre(3), pcre16(3), pcre32(3), pcreapi(3), pcrecallout(3), pcrejit, pcrematching(3), pcrepartial(d), pcrepattern(3), pcreprecompile(3).

AUTHOR

Philip Hazel
University Computing Service
Cambridge CB2 3QH, England.

REVISION

Last updated: 26 April 2013
Copyright (c) 1997-2013 University of Cambridge.