/[pcre]/code/trunk/README
ViewVC logotype

Diff of /code/trunk/README

Parent Directory Parent Directory | Revision Log Revision Log | View Patch Patch

revision 3 by nigel, Sat Feb 24 21:38:01 2007 UTC revision 35 by nigel, Sat Feb 24 21:39:05 2007 UTC
# Line 1  Line 1 
1  README file for PCRE (Perl-compatible regular expressions)  README file for PCRE (Perl-compatible regular expressions)
2  ----------------------------------------------------------  ----------------------------------------------------------
3    
4    *******************************************************************************
5    *           IMPORTANT FOR THOSE UPGRADING FROM VERSIONS BEFORE 2.00           *
6    *                                                                             *
7    * Please note that there has been a change in the API such that a larger      *
8    * ovector is required at matching time, to provide some additional workspace. *
9    * The new man page has details. This change was necessary in order to support *
10    * some of the new functionality in Perl 5.005.                                *
11    *                                                                             *
12    *           IMPORTANT FOR THOSE UPGRADING FROM VERSION 2.00                   *
13    *                                                                             *
14    * Another (I hope this is the last!) change has been made to the API for the  *
15    * pcre_compile() function. An additional argument has been added to make it   *
16    * possible to pass over a pointer to character tables built in the current    *
17    * locale by pcre_maketables(). To use the default tables, this new arguement  *
18    * should be passed as NULL.                                                   *
19    *                                                                             *
20    *           IMPORTANT FOR THOSE UPGRADING FROM VERSION 2.05                   *
21    *                                                                             *
22    * Yet another (and again I hope this really is the last) change has been made *
23    * to the API for the pcre_exec() function. An additional argument has been    *
24    * added to make it possible to start the match other than at the start of the *
25    * subject string. This is important if there are lookbehinds. The new man     *
26    * page has the details, but you just want to convert existing programs, all   *
27    * you need to do is to stick in a new fifth argument to pcre_exec(), with a   *
28    * value of zero. For example, change                                          *
29    *                                                                             *
30    *   pcre_exec(pattern, extra, subject, length, options, ovec, ovecsize)       *
31    * to                                                                          *
32    *   pcre_exec(pattern, extra, subject, length, 0, options, ovec, ovecsize)    *
33    *******************************************************************************
34    
35    
36  The distribution should contain the following files:  The distribution should contain the following files:
37    
38    ChangeLog         log of changes to the code    ChangeLog         log of changes to the code
39      LICENCE           conditions for the use of PCRE
40    Makefile          for building PCRE    Makefile          for building PCRE
   Performance       notes on performance  
41    README            this file    README            this file
42      RunTest           a shell script for running tests
43    Tech.Notes        notes on the encoding    Tech.Notes        notes on the encoding
44    pcre.3            man page for the functions    pcre.3            man page for the functions
45    pcreposix.3       man page for the POSIX wrapper API    pcreposix.3       man page for the POSIX wrapper API
46    maketables.c      auxiliary program for building chartables.c    dftables.c        auxiliary program for building chartables.c
47      get.c             )
48      maketables.c      )
49    study.c           ) source of    study.c           ) source of
50    pcre.c            )   the functions    pcre.c            )   the functions
51    pcreposix.c       )    pcreposix.c       )
# Line 21  The distribution should contain the foll Line 56  The distribution should contain the foll
56    pgrep.1           man page for pgrep    pgrep.1           man page for pgrep
57    pgrep.c           source of a grep utility that uses PCRE    pgrep.c           source of a grep utility that uses PCRE
58    perltest          Perl test program    perltest          Perl test program
59    testinput         test data, compatible with Perl    testinput1        test data, compatible with Perl 5.004 and 5.005
60    testinput2        test data for error messages and non-Perl things    testinput2        test data for error messages and non-Perl things
61    testoutput        test results corresponding to testinput    testinput3        test data, compatible with Perl 5.005
62      testinput4        test data for locale-specific tests
63      testoutput1       test results corresponding to testinput1
64    testoutput2       test results corresponding to testinput2    testoutput2       test results corresponding to testinput2
65      testoutput3       test results corresponding to testinput3
66      testoutput4       test results corresponding to testinput4
67    
68    To build PCRE, edit Makefile for your system (it is a fairly simple make file,
69    and there are some comments at the top) and then run it. It builds two
70    libraries called libpcre.a and libpcreposix.a, a test program called pcretest,
71    and the pgrep command.
72    
73    To test PCRE, run the RunTest script in the pcre directory. This runs pcretest
74    on each of the testinput files in turn, and compares the output with the
75    contents of the corresponding testoutput file. A file called testtry is used to
76    hold the output from pcretest (which is documented below).
77    
78    To run pcretest on just one of the test files, give its number as an argument
79    to RunTest, for example:
80    
81      RunTest 3
82    
83    The first and third test files can also be fed directly into the perltest
84    program to check that Perl gives the same results. The third file requires the
85    additional features of release 5.005, which is why it is kept separate from the
86    main test input, which needs only Perl 5.004. In the long run, when 5.005 is
87    widespread, these two test files may get amalgamated.
88    
89    The second set of tests check pcre_info(), pcre_study(), pcre_copy_substring(),
90    pcre_get_substring(), pcre_get_substring_list(), error detection and run-time
91    flags that are specific to PCRE, as well as the POSIX wrapper API.
92    
93    The fourth set of tests checks pcre_maketables(), the facility for building a
94    set of character tables for a specific locale and using them instead of the
95    default tables. The tests make use of the "fr" (French) locale. Before running
96    the test, the script checks for the presence of this locale by running the
97    "locale" command. If that command fails, or if it doesn't include "fr" in the
98    list of available locales, the fourth test cannot be run, and a comment is
99    output to say why. If running this test produces instances of the error
100    
101      ** Failed to set locale "fr"
102    
103  To build PCRE, edit Makefile for your system (it is a fairly simple make file)  in the comparison output, it means that locale is not available on your system,
104  and then run it. It builds a two libraries called libpcre.a and libpcreposix.a,  despite being listed by "locale". This does not mean that PCRE is broken.
 a test program called pcretest, and the pgrep command.  
   
 To test PCRE, run pcretest on the file testinput, and compare the output with  
 the contents of testoutput. There should be no differences. For example:  
   
   pcretest testinput /tmp/anything  
   diff /tmp/anything testoutput  
   
 Do the same with testinput2, comparing the output with testoutput2, but this  
 time using the -i flag for pcretest, i.e.  
   
   pcretest -i testinput2 /tmp/anything  
   diff /tmp/anything testoutput2  
   
 There are two sets of tests because the first set can also be fed directly into  
 the perltest program to check that Perl gives the same results. The second set  
 of tests check pcre_info(), pcre_study(), error detection and run-time flags  
 that are specific to PCRE, as well as the POSIX wrapper API.  
105    
106  To install PCRE, copy libpcre.a to any suitable library directory (e.g.  To install PCRE, copy libpcre.a to any suitable library directory (e.g.
107  /usr/local/lib), pcre.h to any suitable include directory (e.g.  /usr/local/lib), pcre.h to any suitable include directory (e.g.
# Line 63  themselves still follow Perl syntax and Line 119  themselves still follow Perl syntax and
119  for the POSIX-style functions is called pcreposix.h. The official POSIX name is  for the POSIX-style functions is called pcreposix.h. The official POSIX name is
120  regex.h, but I didn't want to risk possible problems with existing files of  regex.h, but I didn't want to risk possible problems with existing files of
121  that name by distributing it that way. To use it with an existing program that  that name by distributing it that way. To use it with an existing program that
122  uses the POSIX API it will have to be renamed or pointed at by a link.  uses the POSIX API, it will have to be renamed or pointed at by a link.
123    
124    
125  Character tables  Character tables
126  ----------------  ----------------
127    
128  PCRE uses four tables for manipulating and identifying characters. These are  PCRE uses four tables for manipulating and identifying characters. The final
129  compiled from a source file called chartables.c. This is not supplied in  argument of the pcre_compile() function is a pointer to a block of memory
130  the distribution, but is built by the program maketables (compiled from  containing the concatenated tables. A call to pcre_maketables() can be used to
131  maketables.c), which uses the ANSI C character handling functions such as  generate a set of tables in the current locale. If the final argument for
132  isalnum(), isalpha(), isupper(), islower(), etc. to build the table sources.  pcre_compile() is passed as NULL, a set of default tables that is built into
133  This means that the default C locale set in your system may affect the contents  the binary is used.
134  of the tables. You can change the tables by editing chartables.c and then  
135  re-building PCRE. If you do this, you should probably also edit Makefile to  The source file called chartables.c contains the default set of tables. This is
136  ensure that the file doesn't ever get re-generated.  not supplied in the distribution, but is built by the program dftables
137    (compiled from dftables.c), which uses the ANSI C character handling functions
138  The first two tables pcre_lcc[] and pcre_fcc[] provide lower casing and a  such as isalnum(), isalpha(), isupper(), islower(), etc. to build the table
139  case flipping functions, respectively. The pcre_cbits[] table consists of four  sources. This means that the default C locale which is set for your system will
140  32-byte bit maps which identify digits, letters, "word" characters, and white  control the contents of these default tables. You can change the default tables
141  space, respectively. These are used when building 32-byte bit maps that  by editing chartables.c and then re-building PCRE. If you do this, you should
142  represent character classes.  probably also edit Makefile to ensure that the file doesn't ever get
143    re-generated.
144    
145    The first two 256-byte tables provide lower casing and case flipping functions,
146    respectively. The next table consists of three 32-byte bit maps which identify
147    digits, "word" characters, and white space, respectively. These are used when
148    building 32-byte bit maps that represent character classes.
149    
150  The pcre_ctypes[] table has bits indicating various character types, as  The final 256-byte table has bits indicating various character types, as
151  follows:  follows:
152    
153      1   white space character      1   white space character
# Line 114  The program handles any number of sets o Line 176  The program handles any number of sets o
176  set starts with a regular expression, and continues with any number of data  set starts with a regular expression, and continues with any number of data
177  lines to be matched against the pattern. An empty line signals the end of the  lines to be matched against the pattern. An empty line signals the end of the
178  set. The regular expressions are given enclosed in any non-alphameric  set. The regular expressions are given enclosed in any non-alphameric
179  delimiters, for example  delimiters other than backslash, for example
180    
181    /(a|bc)x+yz/    /(a|bc)x+yz/
182    
183  and may be followed by i, m, s, or x to set the PCRE_CASELESS, PCRE_MULTILINE,  White space before the initial delimiter is ignored. A regular expression may
184  PCRE_DOTALL, or PCRE_EXTENDED options, respectively. These options have the  be continued over several input lines, in which case the newline characters are
185  same effect as they do in Perl.  included within it. See the testinput files for many examples. It is possible
186    to include the delimiter within the pattern by escaping it, for example
187  There are also some upper case options that do not match Perl options: /A, /E,  
188  and /X set PCRE_ANCHORED, PCRE_DOLLAR_ENDONLY, and PCRE_EXTRA respectively.    /abc\/def/
189  The /D option is a PCRE debugging feature. It causes the internal form of  
190  compiled regular expressions to be output after compilation. The /S option  If you do so, the escape and the delimiter form part of the pattern, but since
191  causes pcre_study() to be called after the expression has been compiled, and  delimiters are always non-alphameric, this does not affect its interpretation.
192  the results used when the expression is matched. If /I is present as well as  If the terminating delimiter is immediately followed by a backslash, for
193  /S, then pcre_study() is called with the PCRE_CASELESS option.  example,
194    
195  Finally, the /P option causes pcretest to call PCRE via the POSIX wrapper API    /abc/\
196  rather than its native API. When this is done, all other options except /i and  
197  /m are ignored. REG_ICASE is set if /i is present, and REG_NEWLINE is set if /m  then a backslash is added to the end of the pattern. This is done to provide a
198  is present. The wrapper functions force PCRE_DOLLAR_ENDONLY always, and  way of testing the error condition that arises if a pattern finishes with a
199  PCRE_DOTALL unless REG_NEWLINE is set.  backslash, because
200    
201  A regular expression can extend over several lines of input; the newlines are    /abc\/
202  included in it. See the testinput file for many examples.  
203    is interpreted as the first line of a pattern that starts with "abc/", causing
204    pcretest to read the next line as a continuation of the regular expression.
205    
206    The pattern may be followed by i, m, s, or x to set the PCRE_CASELESS,
207    PCRE_MULTILINE, PCRE_DOTALL, or PCRE_EXTENDED options, respectively. For
208    example:
209    
210      /caseless/i
211    
212    These modifier letters have the same effect as they do in Perl. There are
213    others which set PCRE options that do not correspond to anything in Perl: /A,
214    /E, and /X set PCRE_ANCHORED, PCRE_DOLLAR_ENDONLY, and PCRE_EXTRA respectively.
215    
216    Searching for all possible matches within each subject string can be requested
217    by the /g or /G modifier. The /g modifier behaves similarly to the way it does
218    in Perl. After finding a match, PCRE is called again to search the remainder of
219    the subject string. The difference between /g and /G is that the former uses
220    the start_offset argument to pcre_exec() to start searching at a new point
221    within the entire string, whereas the latter passes over a shortened substring.
222    This makes a difference to the matching process if the pattern begins with a
223    lookbehind assertion (including \b or \B).
224    
225    There are a number of other modifiers for controlling the way pcretest
226    operates.
227    
228    The /+ modifier requests that as well as outputting the substring that matched
229    the entire pattern, pcretest should in addition output the remainder of the
230    subject string. This is useful for tests where the subject contains multiple
231    copies of the same substring.
232    
233    The /L modifier must be followed directly by the name of a locale, for example,
234    
235      /pattern/Lfr
236    
237    For this reason, it must be the last modifier letter. The given locale is set,
238    pcre_maketables() is called to build a set of character tables for the locale,
239    and this is then passed to pcre_compile() when compiling the regular
240    expression. Without an /L modifier, NULL is passed as the tables pointer; that
241    is, /L applies only to the expression on which it appears.
242    
243    The /I modifier requests that pcretest output information about the compiled
244    expression (whether it is anchored, has a fixed first character, and so on). It
245    does this by calling pcre_info() after compiling an expression, and outputting
246    the information it gets back. If the pattern is studied, the results of that
247    are also output.
248    
249    The /D modifier is a PCRE debugging feature, which also assumes /I. It causes
250    the internal form of compiled regular expressions to be output after
251    compilation.
252    
253    The /S modifier causes pcre_study() to be called after the expression has been
254    compiled, and the results used when the expression is matched.
255    
256    The /M modifier causes information about the size of memory block used to hold
257    the compile pattern to be output.
258    
259    Finally, the /P modifier causes pcretest to call PCRE via the POSIX wrapper API
260    rather than its native API. When this is done, all other modifiers except /i,
261    /m, and /+ are ignored. REG_ICASE is set if /i is present, and REG_NEWLINE is
262    set if /m is present. The wrapper functions force PCRE_DOLLAR_ENDONLY always,
263    and PCRE_DOTALL unless REG_NEWLINE is set.
264    
265  Before each data line is passed to pcre_exec(), leading and trailing whitespace  Before each data line is passed to pcre_exec(), leading and trailing whitespace
266  is removed, and it is then scanned for \ escapes. The following are recognized:  is removed, and it is then scanned for \ escapes. The following are recognized:
# Line 155  is removed, and it is then scanned for \ Line 278  is removed, and it is then scanned for \
278    
279    \A     pass the PCRE_ANCHORED option to pcre_exec()    \A     pass the PCRE_ANCHORED option to pcre_exec()
280    \B     pass the PCRE_NOTBOL option to pcre_exec()    \B     pass the PCRE_NOTBOL option to pcre_exec()
281    \E     pass the PCRE_DOLLAR_ENDONLY option to pcre_exec()    \Cdd   call pcre_copy_substring() for substring dd after a successful match
282    \I     pass the PCRE_CASELESS option to pcre_exec()             (any decimal number less than 32)
283    \M     pass the PCRE_MULTILINE option to pcre_exec()    \Gdd   call pcre_get_substring() for substring dd after a successful match
284    \S     pass the PCRE_DOTALL option to pcre_exec()             (any decimal number less than 32)
285      \L     call pcre_get_substringlist() after a successful match
286    \Odd   set the size of the output vector passed to pcre_exec() to dd    \Odd   set the size of the output vector passed to pcre_exec() to dd
287             (any number of decimal digits)             (any number of decimal digits)
288    \Z     pass the PCRE_NOTEOL option to pcre_exec()    \Z     pass the PCRE_NOTEOL option to pcre_exec()
# Line 171  If /P was present on the regex, causing Line 295  If /P was present on the regex, causing
295  \B, and \Z have any effect, causing REG_NOTBOL and REG_NOTEOL to be passed to  \B, and \Z have any effect, causing REG_NOTBOL and REG_NOTEOL to be passed to
296  regexec() respectively.  regexec() respectively.
297    
298  When a match succeeds, pcretest outputs the list of identified substrings that  When a match succeeds, pcretest outputs the list of captured substrings that
299  pcre_exec() returns, starting with number 0 for the string that matched the  pcre_exec() returns, starting with number 0 for the string that matched the
300  whole pattern. Here is an example of an interactive pcretest run.  whole pattern. Here is an example of an interactive pcretest run.
301    
302    $ pcretest    $ pcretest
303    Testing Perl-Compatible Regular Expressions    PCRE version 2.06 08-Jun-1999
   PCRE version 0.90 08-Sep-1997  
304    
305        re> /^abc(\d+)/      re> /^abc(\d+)/
306      data> abc123    data> abc123
307     0: abc123     0: abc123
308     1: 123     1: 123
309      data> xyz    data> xyz
310    No match    No match
311    
312    If the strings contain any non-printing characters, they are output as \0x
313    escapes. If the pattern has the /+ modifier, then the output for substring 0 is
314    followed by the the rest of the subject string, identified by "0+" like this:
315    
316        re> /cat/+
317      data> cataract
318       0: cat
319       0+ aract
320    
321    If the pattern has the /g or /G modifier, the results of successive matching
322    attempts are output in sequence, like this:
323    
324        re> /\Bi(\w\w)/g
325      data> Mississippi
326       0: iss
327       1: ss
328       0: iss
329       1: ss
330       0: ipp
331       1: pp
332    
333    "No match" is output only if the first match attempt fails.
334    
335    If any of \C, \G, or \L are present in a data line that is successfully
336    matched, the substrings extracted by the convenience functions are output with
337    C, G, or L after the string number instead of a colon. This is in addition to
338    the normal full list. The string length (that is, the return from the
339    extraction function) is given in parentheses after each string for \C and \G.
340    
341  Note that while patterns can be continued over several lines (a plain ">"  Note that while patterns can be continued over several lines (a plain ">"
342  prompt is used for continuations), data lines may not. However newlines can be  prompt is used for continuations), data lines may not. However newlines can be
343  included in data by means of the \n escape.  included in data by means of the \n escape.
# Line 197  following flags has any effect in this c Line 349  following flags has any effect in this c
349  If the option -d is given to pcretest, it is equivalent to adding /D to each  If the option -d is given to pcretest, it is equivalent to adding /D to each
350  regular expression: the internal form is output after compilation.  regular expression: the internal form is output after compilation.
351    
352  If the option -i (for "information") is given to pcretest, it calls pcre_info()  If the option -i is given to pcretest, it is equivalent to adding /I to each
353  after compiling an expression, and outputs the information it gets back. If the  regular expression: information about the compiled pattern is given after
354  pattern is studied, the results of that are also output.  compilation.
355    
356  If the option -s is given to pcretest, it outputs the size of each compiled  If the option -m is given to pcretest, it outputs the size of each compiled
357  pattern after it has been compiled.  pattern after it has been compiled. It is equivalent to adding /M to each
358    regular expression. For compatibility with earlier versions of pcretest, -s is
359    a synonym for -m.
360    
361  If the -t option is given, each compile, study, and match is run 2000 times  If the -t option is given, each compile, study, and match is run 20000 times
362  while being timed, and the resulting time per compile or match is output in  while being timed, and the resulting time per compile or match is output in
363  milliseconds. Do not set -t with -s, because you will then get the size output  milliseconds. Do not set -t with -s, because you will then get the size output
364  2000 times and the timing will be distorted.  20000 times and the timing will be distorted. If you want to change the number
365    of repetitions used for timing, edit the definition of LOOPREPEAT at the top of
366    pcretest.c
367    
368    
369    
# Line 216  The perltest program Line 372  The perltest program
372    
373  The perltest program tests Perl's regular expressions; it has the same  The perltest program tests Perl's regular expressions; it has the same
374  specification as pcretest, and so can be given identical input, except that  specification as pcretest, and so can be given identical input, except that
375  input patterns can be followed only by Perl's lower case options.  input patterns can be followed only by Perl's lower case modifiers. The
376    contents of testinput1 and testinput3 meet this condition.
377    
378  The data lines are processed as Perl strings, so if they contain $ or @  The data lines are processed as Perl strings, so if they contain $ or @
379  characters, these have to be escaped. For this reason, all such characters in  characters, these have to be escaped. For this reason, all such characters in
380  the testinput file are escaped so that it can be used for perltest as well as  testinput1 and testinput3 are escaped so that they can be used for perltest as
381  for pcretest, and the special upper case options such as /A that pcretest  well as for pcretest, and the special upper case modifiers such as /A that
382  recognizes are not used in this file. The output should be identical, apart  pcretest recognizes are not used in these files. The output should be
383  from the initial identifying banner.  identical, apart from the initial identifying banner.
384    
385  The testinput2 file is not suitable for feeding to Perltest, since it does  The testinput2 and testinput4 files are not suitable for feeding to perltest,
386  make use of the special upper case options and escapes that pcretest uses to  since they do make use of the special upper case modifiers and escapes that
387  test additional features of PCRE.  pcretest uses to test some features of PCRE. The first of these files also
388    contains malformed regular expressions, in order to check that PCRE diagnoses
389    them correctly.
390    
391  Philip Hazel <ph10@cam.ac.uk>  Philip Hazel <ph10@cam.ac.uk>
392  October 1997  June 1999

Legend:
Removed from v.3  
changed lines
  Added in v.35

  ViewVC Help
Powered by ViewVC 1.1.5