ViewVC logotype

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log

Revision 87 - (hide annotations) (download)
Sat Feb 24 21:41:21 2007 UTC (8 years, 2 months ago) by nigel
File size: 95036 byte(s)
Load pcre-6.5 into code/trunk.

1 nigel 3 ChangeLog for PCRE
2     ------------------
4 nigel 87 Version 6.5 01-Feb-06
5     ---------------------
7     1. When using the partial match feature with pcre_dfa_exec(), it was not
8     anchoring the second and subsequent partial matches at the new starting
9     point. This could lead to incorrect results. For example, with the pattern
10     /1234/, partially matching against "123" and then "a4" gave a match.
12     2. Changes to pcregrep:
14     (a) All non-match returns from pcre_exec() were being treated as failures
15     to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
16     error message is output. Some extra information is given for the
18     probably the only errors that are likely to be caused by users (by
19     specifying a regex that has nested indefinite repeats, for instance).
20     If there are more than 20 of these errors, pcregrep is abandoned.
22     (b) A binary zero was treated as data while matching, but terminated the
23     output line if it was written out. This has been fixed: binary zeroes
24     are now no different to any other data bytes.
26     (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
27     used to set a locale for matching. The --locale=xxxx long option has
28     been added (no short equivalent) to specify a locale explicitly on the
29     pcregrep command, overriding the environment variables.
31     (d) When -B was used with -n, some line numbers in the output were one less
32     than they should have been.
34     (e) Added the -o (--only-matching) option.
36     (f) If -A or -C was used with -c (count only), some lines of context were
37     accidentally printed for the final match.
39     (g) Added the -H (--with-filename) option.
41     (h) The combination of options -rh failed to suppress file names for files
42     that were found from directory arguments.
44     (i) Added the -D (--devices) and -d (--directories) options.
46     (j) Added the -F (--fixed-strings) option.
48     (k) Allow "-" to be used as a file name for -f as well as for a data file.
50     (l) Added the --colo(u)r option.
52     (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
53     is not present by default.
55     3. A nasty bug was discovered in the handling of recursive patterns, that is,
56     items such as (?R) or (?1), when the recursion could match a number of
57     alternatives. If it matched one of the alternatives, but subsequently,
58     outside the recursion, there was a failure, the code tried to back up into
59     the recursion. However, because of the way PCRE is implemented, this is not
60     possible, and the result was an incorrect result from the match.
62     In order to prevent this happening, the specification of recursion has
63     been changed so that all such subpatterns are automatically treated as
64     atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
66     4. I had overlooked the fact that, in some locales, there are characters for
67     which isalpha() is true but neither isupper() nor islower() are true. In
68     the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
69     and ordfeminine) are like this. This affected the treatment of \w and \W
70     when they appeared in character classes, but not when they appeared outside
71     a character class. The bit map for "word" characters is now created
72     separately from the results of isalnum() instead of just taking it from the
73     upper, lower, and digit maps. (Plus the underscore character, of course.)
75     5. The above bug also affected the handling of POSIX character classes such as
76     [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
77     permanent tables. Instead, the bit maps for such a class were previously
78     created as the appropriate unions of the upper, lower, and digit bitmaps.
79     Now they are created by subtraction from the [[:word:]] class, which has
80     its own bitmap.
82     6. The [[:blank:]] character class matches horizontal, but not vertical space.
83     It is created by subtracting the vertical space characters (\x09, \x0a,
84     \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
85     subtraction was done in the overall bitmap for a character class, meaning
86     that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
87     be recognized. This bug has been fixed.
89     7. Patches from the folks at Google:
91     (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
92     real life, but is still worth protecting against".
94     (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
95     regular expressions".
97     (c) pcre_scanner.cc: avoid use of std::count() because not all systems
98     have it.
100     (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
101     "configure" and the latter not, in order to fix a problem somebody had
102     with compiling the Arg class on HP-UX.
104     (e) Improve the error-handling of the C++ wrapper a little bit.
106     (f) New tests for checking recursion limiting.
108     8. The pcre_memmove() function, which is used only if the environment does not
109     have a standard memmove() function (and is therefore rarely compiled),
110     contained two bugs: (a) use of int instead of size_t, and (b) it was not
111     returning a result (though PCRE never actually uses the result).
113     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
114     large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
115     returned instead of calling malloc() with an overflowing number that would
116     most likely cause subsequent chaos.
118     10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
120     11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
121     with this option is matched, the nmatch and pmatch options of regexec() are
122     ignored.
124     12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
125     provided in case anyone wants to the the POSIX interface with UTF-8
126     strings.
128     13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
129     C++ linking (needed for some HP-UX environments).
131     14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
132     (unused parameter) and in the pcre_printint() function (omitted "default"
133     switch label when the default is to do nothing).
135     15. Added some code to make it possible, when PCRE is compiled as a C++
136     library, to replace subject pointers for pcre_exec() with a smart pointer
137     class, thus making it possible to process discontinuous strings.
139     16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
140     much the same function. They were added by different people who were trying
141     to make PCRE easy to compile on non-Unix systems. It has been suggested
142     that PCRE_EXPORT be abolished now that there is more automatic apparatus
143     for compiling on Windows systems. I have therefore replaced it with
144     PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
145     defaults to "extern" for C or "extern C" for C++, which works fine on
146     Unix-like systems. It is now possible to override the value of PCRE_DATA_
147     SCOPE with something explicit in config.h. In addition:
149     (a) pcreposix.h still had just "extern" instead of either of these macros;
150     I have replaced it with PCRE_DATA_SCOPE.
152     (b) Functions such as _pcre_xclass(), which are internal to the library,
153     but external in the C sense, all had PCRE_EXPORT in their definitions.
154     This is apparently wrong for the Windows case, so I have removed it.
155     (It makes no difference on Unix-like systems.)
157     17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
158     of recursive calls to match(). This is different to MATCH_LIMIT because
159     that limits the total number of calls to match(), not all of which increase
160     the depth of recursion. Limiting the recursion depth limits the amount of
161     stack (or heap if NO_RECURSE is set) that is used. The default can be set
162     when PCRE is compiled, and changed at run time. A patch from Google adds
163     this functionality to the C++ interface.
165     18. Changes to the handling of Unicode character properties:
167     (a) Updated the table to Unicode 4.1.0.
169     (b) Recognize characters that are not in the table as "Cn" (undefined).
171     (c) I revised the way the table is implemented to a much improved format
172     which includes recognition of ranges. It now supports the ranges that
173     are defined in UnicodeData.txt, and it also amalgamates other
174     characters into ranges. This has reduced the number of entries in the
175     table from around 16,000 to around 3,000, thus reducing its size
176     considerably. I realized I did not need to use a tree structure after
177     all - a binary chop search is just as efficient. Having reduced the
178     number of entries, I extended their size from 6 bytes to 8 bytes to
179     allow for more data.
181     (d) Added support for Unicode script names via properties such as \p{Han}.
183     19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
184     matching that character.
186     20. When matching a repeated Unicode property with a minimum greater than zero,
187     (for example \pL{2,}), PCRE could look past the end of the subject if it
188     reached it while seeking the minimum number of characters. This could
189     happen only if some of the characters were more than one byte long, because
190     there is a check for at least the minimum number of bytes.
192     21. Refactored the implementation of \p and \P so as to be more general, to
193     allow for more different types of property in future. This has changed the
194     compiled form incompatibly. Anybody with saved compiled patterns that use
195     \p or \P will have to recompile them.
197     22. Added "Any" and "L&" to the supported property types.
199     23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
200     but give a compile time error if the value is greater than 0xff.
202     24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
203     accidentally not being installed or uninstalled.
205     25. The pcre.h file was built from pcre.h.in, but the only changes that were
206     made were to insert the current release number. This seemed silly, because
207     it made things harder for people building PCRE on systems that don't run
208     "configure". I have turned pcre.h into a distributed file, no longer built
209     by "configure", with the version identification directly included. There is
210     no longer a pcre.h.in file.
212     However, this change necessitated a change to the pcre-config script as
213     well. It is built from pcre-config.in, and one of the substitutions was the
214     release number. I have updated configure.ac so that ./configure now finds
215     the release number by grepping pcre.h.
217     26. Added the ability to run the tests under valgrind.
220 nigel 85 Version 6.4 05-Sep-05
221     ---------------------
223     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
224     "--" to be printed when multiple files were scanned, even when none of the
225     -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
226     consider it to be a bug, and have restored the previous behaviour.
228     2. A couple of code tidies to get rid of compiler warnings.
230     3. The pcretest program used to cheat by referring to symbols in the library
231     whose names begin with _pcre_. These are internal symbols that are not
232     really supposed to be visible externally, and in some environments it is
233     possible to suppress them. The cheating is now confined to including
234     certain files from the library's source, which is a bit cleaner.
236     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
237     file's purpose clearer.
239     5. Reorganized pcre_ucp_findchar().
242 nigel 83 Version 6.3 15-Aug-05
243     ---------------------
245     1. The file libpcre.pc.in did not have general read permission in the tarball.
247     2. There were some problems when building without C++ support:
249     (a) If C++ support was not built, "make install" and "make test" still
250     tried to test it.
252     (b) There were problems when the value of CXX was explicitly set. Some
253     changes have been made to try to fix these, and ...
255     (c) --disable-cpp can now be used to explicitly disable C++ support.
257     (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
258     backslash in a target when C++ was disabled. This confuses some
259     versions of "make", apparently. Using an intermediate variable solves
260     this. (Same for CPP_LOBJ.)
262     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
263     (non-Windows) now includes $(CFLAGS) because these flags are sometimes
264     necessary on certain architectures.
266     4. Added a setting of -export-symbols-regex to the link command to remove
267     those symbols that are exported in the C sense, but actually are local
268     within the library, and not documented. Their names all begin with
269     "_pcre_". This is not a perfect job, because (a) we have to except some
270     symbols that pcretest ("illegally") uses, and (b) the facility isn't always
271     available (and never for static libraries). I have made a note to try to
272     find a way round (a) in the future.
275 nigel 81 Version 6.2 01-Aug-05
276     ---------------------
278     1. There was no test for integer overflow of quantifier values. A construction
279     such as {1111111111111111} would give undefined results. What is worse, if
280     a minimum quantifier for a parenthesized subpattern overflowed and became
281     negative, the calculation of the memory size went wrong. This could have
282     led to memory overwriting.
284     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
286     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
287     operating environments where this matters.
289     4. Applied Giuseppe Maxia's patch to add additional features for controlling
290     PCRE options from within the C++ wrapper.
292     5. Named capturing subpatterns were not being correctly counted when a pattern
293     was compiled. This caused two problems: (a) If there were more than 100
294     such subpatterns, the calculation of the memory needed for the whole
295     compiled pattern went wrong, leading to an overflow error. (b) Numerical
296     back references of the form \12, where the number was greater than 9, were
297     not recognized as back references, even though there were sufficient
298     previous subpatterns.
300     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
301     versions of gcc, e.g. 2.95.4.
304 nigel 79 Version 6.1 21-Jun-05
305     ---------------------
307     1. There was one reference to the variable "posix" in pcretest.c that was not
308     surrounded by "#if !defined NOPOSIX".
310     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
311     the cross-check on the old pcre_info() function, for the benefit of the
312     cut-down version of PCRE that is currently imported into Exim.
314     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
315     allocation error. I've done the easy fix, which wastes 2 bytes for sensible
316     patterns that start (?i) but I don't think that matters. The use of (?i) is
317     just an example; this all applies to the other options as well.
319     4. Since libtool seems to echo the compile commands it is issuing, the output
320     from "make" can be reduced a bit by putting "@" in front of each libtool
321     compile command.
323     5. Patch from the folks at Google for configure.in to be a bit more thorough
324     in checking for a suitable C++ installation before trying to compile the
325     C++ stuff. This should fix a reported problem when a compiler was present,
326     but no suitable headers.
328     6. The man pages all had just "PCRE" as their title. I have changed them to
329     be the relevant file name. I have also arranged that these names are
330     retained in the file doc/pcre.txt, which is a concatenation in text format
331     of all the man pages except the little individual ones for each function.
333     7. The NON-UNIX-USE file had not been updated for the different set of source
334     files that come with release 6. I also added a few comments about the C++
335     wrapper.
338 nigel 77 Version 6.0 07-Jun-05
339     ---------------------
341     1. Some minor internal re-organization to help with my DFA experiments.
343     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
344     didn't matter for the library itself when fully configured, but did matter
345     when compiling without UCP support, or within Exim, where the ucp files are
346     not imported.
348     3. Refactoring of the library code to split up the various functions into
349     different source modules. The addition of the new DFA matching code (see
350     below) to a single monolithic source would have made it really too
351     unwieldy, quite apart from causing all the code to be include in a
352     statically linked application, when only some functions are used. This is
353     relevant even without the DFA addition now that patterns can be compiled in
354     one application and matched in another.
356     The downside of splitting up is that there have to be some external
357     functions and data tables that are used internally in different modules of
358     the library but which are not part of the API. These have all had their
359     names changed to start with "_pcre_" so that they are unlikely to clash
360     with other external names.
362     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
363     a different (DFA) algorithm. Although it is slower than the original
364     function, it does have some advantages for certain types of matching
365     problem.
367     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
368     including restarting after a partial match.
370     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
371     defined when compiling for Windows was sent to me. I have put it into the
372     code, though I have no means of testing or verifying it.
374     7. Added the pcre_refcount() auxiliary function.
376     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
377     match before or at the first newline in the subject string. In pcretest,
378     the /f option on a pattern can be used to set this.
380     9. A repeated \w when used in UTF-8 mode with characters greater than 256
381     would behave wrongly. This has been present in PCRE since release 4.0.
383     10. A number of changes to the pcregrep command:
385     (a) Refactored how -x works; insert ^(...)$ instead of setting
386     PCRE_ANCHORED and checking the length, in preparation for adding
387     something similar for -w.
389     (b) Added the -w (match as a word) option.
391     (c) Refactored the way lines are read and buffered so as to have more
392     than one at a time available.
394     (d) Implemented a pcregrep test script.
396     (e) Added the -M (multiline match) option. This allows patterns to match
397     over several lines of the subject. The buffering ensures that at least
398     8K, or the rest of the document (whichever is the shorter) is available
399     for matching (and similarly the previous 8K for lookbehind assertions).
401     (f) Changed the --help output so that it now says
403     -w, --word-regex(p)
405     instead of two lines, one with "regex" and the other with "regexp"
406     because that confused at least one person since the short forms are the
407     same. (This required a bit of code, as the output is generated
408     automatically from a table. It wasn't just a text change.)
410     (g) -- can be used to terminate pcregrep options if the next thing isn't an
411     option but starts with a hyphen. Could be a pattern or a path name
412     starting with a hyphen, for instance.
414     (h) "-" can be given as a file name to represent stdin.
416     (i) When file names are being printed, "(standard input)" is used for
417     the standard input, for compatibility with GNU grep. Previously
418     "<stdin>" was used.
420     (j) The option --label=xxx can be used to supply a name to be used for
421     stdin when file names are being printed. There is no short form.
423     (k) Re-factored the options decoding logic because we are going to add
424     two more options that take data. Such options can now be given in four
425     different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
427     (l) Added the -A, -B, and -C options for requesting that lines of context
428     around matches be printed.
430     (m) Added the -L option to print the names of files that do not contain
431     any matching lines, that is, the complement of -l.
433     (n) The return code is 2 if any file cannot be opened, but pcregrep does
434     continue to scan other files.
436     (o) The -s option was incorrectly implemented. For compatibility with other
437     greps, it now suppresses the error message for a non-existent or non-
438     accessible file (but not the return code). There is a new option called
439     -q that suppresses the output of matching lines, which was what -s was
440     previously doing.
442     (p) Added --include and --exclude options to specify files for inclusion
443     and exclusion when recursing.
445     11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
446     Hopefully, it now does.
448     12. Missing cast in pcre_study().
450     13. Added an "uninstall" target to the makefile.
452     14. Replaced "extern" in the function prototypes in Makefile.in with
453     "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
454     world, but is set differently for Windows.
456     15. Added a second compiling function called pcre_compile2(). The only
457     difference is that it has an extra argument, which is a pointer to an
458     integer error code. When there is a compile-time failure, this is set
459     non-zero, in addition to the error test pointer being set to point to an
460     error message. The new argument may be NULL if no error number is required
461     (but then you may as well call pcre_compile(), which is now just a
462     wrapper). This facility is provided because some applications need a
463     numeric error indication, but it has also enabled me to tidy up the way
464     compile-time errors are handled in the POSIX wrapper.
466     16. Added VPATH=.libs to the makefile; this should help when building with one
467     prefix path and installing with another. (Or so I'm told by someone who
468     knows more about this stuff than I do.)
470     17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
471     passes PCRE_DOTALL to the pcre_compile() function, making the "." character
472     match everything, including newlines. This is not POSIX-compatible, but
473     somebody wanted the feature. From pcretest it can be activated by using
474     both the P and the s flags.
476     18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
478     19. libpcre.pc was being incorrectly installed as executable.
480     20. A couple of places in pcretest check for end-of-line by looking for '\n';
481     it now also looks for '\r' so that it will work unmodified on Windows.
483     21. Added Google's contributed C++ wrapper to the distribution.
485     22. Added some untidy missing memory free() calls in pcretest, to keep
486     Electric Fence happy when testing.
490 nigel 75 Version 5.0 13-Sep-04
491     ---------------------
493     1. Internal change: literal characters are no longer packed up into items
494     containing multiple characters in a single byte-string. Each character
495     is now matched using a separate opcode. However, there may be more than one
496     byte in the character in UTF-8 mode.
498     2. The pcre_callout_block structure has two new fields: pattern_position and
499     next_item_length. These contain the offset in the pattern to the next match
500     item, and its length, respectively.
502     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
503     insertion of callouts before each pattern item. Added the /C option to
504     pcretest to make use of this.
506     4. On the advice of a Windows user, the lines
508     #if defined(_WIN32) || defined(WIN32)
509     _setmode( _fileno( stdout ), 0x8000 );
510     #endif /* defined(_WIN32) || defined(WIN32) */
512     have been added to the source of pcretest. This apparently does useful
513     magic in relation to line terminators.
515     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
516     for the benefit of those environments where the "b" makes a difference.
518     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
519     to know about it. I have put a hack into configure.in that adds in code
520     to set GCC=yes if CC=icc. This seems to end up at a point in the
521     generated configure script that is early enough to affect the setting of
522     compiler options, which is what is needed, but I have no means of testing
523     whether it really works. (The user who reported this had patched the
524     generated configure script, which of course I cannot do.)
526     LATER: After change 22 below (new libtool files), the configure script
527     seems to know about icc (and also ecc). Therefore, I have commented out
528     this hack in configure.in.
530     7. Added support for pkg-config (2 patches were sent in).
532     8. Negated POSIX character classes that used a combination of internal tables
533     were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
534     [[:^ascii]]. Typically, they would match almost any characters. The other
535     POSIX classes were not broken in this way.
537     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
538     to find the match, as PCRE was deluded into thinking that the match had to
539     start at the start point or following a newline. The same bug applied to
540     patterns with negative forward assertions or any backward assertions
541     preceding ".*" at the start, unless the pattern required a fixed first
542     character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
544     10. In UTF-8 mode, when moving forwards in the subject after a failed match
545     starting at the last subject character, bytes beyond the end of the subject
546     string were read.
548     11. Renamed the variable "class" as "classbits" to make life easier for C++
549     users. (Previously there was a macro definition, but it apparently wasn't
550     enough.)
552     12. Added the new field "tables" to the extra data so that tables can be passed
553     in at exec time, or the internal tables can be re-selected. This allows
554     a compiled regex to be saved and re-used at a later time by a different
555     program that might have everything at different addresses.
557     13. Modified the pcre-config script so that, when run on Solaris, it shows a
558     -R library as well as a -L library.
560     14. The debugging options of pcretest (-d on the command line or D on a
561     pattern) showed incorrect output for anything following an extended class
562     that contained multibyte characters and which was followed by a quantifier.
564     15. Added optional support for general category Unicode character properties
565     via the \p, \P, and \X escapes. Unicode property support implies UTF-8
566     support. It adds about 90K to the size of the library. The meanings of the
567     inbuilt class escapes such as \d and \s have NOT been changed.
569     16. Updated pcredemo.c to include calls to free() to release the memory for the
570     compiled pattern.
572     17. The generated file chartables.c was being created in the source directory
573     instead of in the building directory. This caused the build to fail if the
574     source directory was different from the building directory, and was
575     read-only.
577     18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
578     file. No doubt somebody will tell me if they don't make sense... Also added
579     Dan Mooney's comments about building on OpenVMS.
581     19. Added support for partial matching via the PCRE_PARTIAL option for
582     pcre_exec() and the \P data escape in pcretest.
584     20. Extended pcretest with 3 new pattern features:
586     (i) A pattern option of the form ">rest-of-line" causes pcretest to
587     write the compiled pattern to the file whose name is "rest-of-line".
588     This is a straight binary dump of the data, with the saved pointer to
589     the character tables forced to be NULL. The study data, if any, is
590     written too. After writing, pcretest reads a new pattern.
592     (ii) If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
593     compiled pattern from the given file. There must not be any
594     occurrences of "<" in the file name (pretty unlikely); if there are,
595     pcretest will instead treat the initial "<" as a pattern delimiter.
596     After reading in the pattern, pcretest goes on to read data lines as
597     usual.
599     (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
600     and 16-bit fields in a compiled pattern, to simulate a pattern that
601     was compiled on a host of opposite endianness.
603     21. The pcre-exec() function can now cope with patterns that were compiled on
604     hosts of opposite endianness, with this restriction:
606     As for any compiled expression that is saved and used later, the tables
607     pointer field cannot be preserved; the extra_data field in the arguments
608     to pcre_exec() should be used to pass in a tables address if a value
609     other than the default internal tables were used at compile time.
611     22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
612     now diagnosed as an error. Previously, most of the time, a negative number
613     would have been treated as zero, but if in addition "ovector" was passed as
614     NULL, a crash could occur.
616     23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
617     new versions from the libtool 1.5 distribution (the last one is a copy of
618     a file called libtool.m4). This seems to have fixed the need to patch
619     "configure" to support Darwin 1.3 (which I used to do). However, I still
620     had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
621     workstation).
623     24. Changed the PCRE licence to be the more standard "BSD" licence.
626 nigel 73 Version 4.5 01-Dec-03
627     ---------------------
629     1. There has been some re-arrangement of the code for the match() function so
630     that it can be compiled in a version that does not call itself recursively.
631     Instead, it keeps those local variables that need separate instances for
632     each "recursion" in a frame on the heap, and gets/frees frames whenever it
633     needs to "recurse". Keeping track of where control must go is done by means
634     of setjmp/longjmp. The whole thing is implemented by a set of macros that
635     hide most of the details from the main code, and operates only if
636     NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
637     "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
638     operating.
640     To make it easier for callers to provide specially tailored get/free
641     functions for this usage, two new functions, pcre_stack_malloc, and
642     pcre_stack_free, are used. They are always called in strict stacking order,
643     and the size of block requested is always the same.
645     The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
646     PCRE has been compiled to use the stack or the heap for recursion. The
647     -C option of pcretest uses this to show which version is compiled.
649     A new data escape \S, is added to pcretest; it causes the amounts of store
650     obtained and freed by both kinds of malloc/free at match time to be added
651     to the output.
653     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
654     what's available on my current Linux desktop machine.
656     3. When matching a UTF-8 string, the test for a valid string at the start has
657     been extended. If start_offset is not zero, PCRE now checks that it points
658     to a byte that is the start of a UTF-8 character. If not, it returns
659     PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
660     this is necessary because there may be backward assertions in the pattern.
661     When matching the same subject several times, it may save resources to use
662     PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
664     4. The code for checking the validity of UTF-8 strings has been tightened so
665     that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
666     containing "overlong sequences".
668     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
669     I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
670     should have been "&", but it just so happened that all the cases this let
671     through by mistake were picked up later in the function.
673     6. I had used a variable called "isblank" - this is a C99 function, causing
674     some compilers to warn. To avoid this, I renamed it (as "blankclass").
676     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
677     prompting; (b) run "./pcretest /dev/null" at the start of the test script
678     so the version is shown; (c) stop "make test" echoing "./RunTest".
680     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
682     9. The prototype for memmove() for systems that don't have it was using
683     size_t, but the inclusion of the header that defines size_t was later. I've
684     moved the #includes for the C headers earlier to avoid this.
686     10. Added some adjustments to the code to make it easier to compiler on certain
687     special systems:
689     (a) Some "const" qualifiers were missing.
690     (b) Added the macro EXPORT before all exported functions; by default this
691     is defined to be empty.
692     (c) Changed the dftables auxiliary program (that builds chartables.c) so
693     that it reads its output file name as an argument instead of writing
694     to the standard output and assuming this can be redirected.
696     11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
697     class containing characters with values greater than 255, PCRE compilation
698     went into a loop.
700     12. A recursive reference to a subpattern that was within another subpattern
701     that had a minimum quantifier of zero caused PCRE to crash. For example,
702     (x(y(?2))z)? provoked this bug with a subject that got as far as the
703     recursion. If the recursively-called subpattern itself had a zero repeat,
704     that was OK.
706     13. In pcretest, the buffer for reading a data line was set at 30K, but the
707     buffer into which it was copied (for escape processing) was still set at
708     1024, so long lines caused crashes.
710     14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
711     "internal error: code overflow...". This applied to any character class
712     that was followed by a possessive quantifier.
714     15. Modified the Makefile to add libpcre.la as a prerequisite for
715     libpcreposix.la because I was told this is needed for a parallel build to
716     work.
718     16. If a pattern that contained .* following optional items at the start was
719     studied, the wrong optimizing data was generated, leading to matching
720     errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
721     matching string must start with a or b or c. The correct conclusion for
722     this pattern is that a match can start with any character.
725 nigel 71 Version 4.4 13-Aug-03
726     ---------------------
728     1. In UTF-8 mode, a character class containing characters with values between
729     127 and 255 was not handled correctly if the compiled pattern was studied.
730     In fixing this, I have also improved the studying algorithm for such
731     classes (slightly).
733     2. Three internal functions had redundant arguments passed to them. Removal
734     might give a very teeny performance improvement.
736     3. Documentation bug: the value of the capture_top field in a callout is *one
737     more than* the number of the hightest numbered captured substring.
739     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
740     in incorrectly linking with a previously installed version. They now link
741     explicitly with libpcre.la.
743     5. configure.in no longer needs to recognize Cygwin specially.
745     6. A problem in pcre.in for Windows platforms is fixed.
747     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
748     pcretest, it used to include the size of the study block as part of its
749     output. Unfortunately, the structure contains a field that has a different
750     size on different hardware architectures. This meant that the tests that
751     showed this size failed. As the block is currently always of a fixed size,
752     this information isn't actually particularly useful in pcretest output, so
753     I have just removed it.
755     8. Three pre-processor statements accidentally did not start in column 1.
756     Sadly, there are *still* compilers around that complain, even though
757     standard C has not required this for well over a decade. Sigh.
759     9. In pcretest, the code for checking callouts passed small integers in the
760     callout_data field, which is a void * field. However, some picky compilers
761     complained about the casts involved for this on 64-bit systems. Now
762     pcretest passes the address of the small integer instead, which should get
763     rid of the warnings.
765     10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
766     both compile and run time, and gives an error if an invalid UTF-8 sequence
767     is found. There is a option for disabling this check in cases where the
768     string is known to be correct and/or the maximum performance is wanted.
770     11. In response to a bug report, I changed one line in Makefile.in from
772     -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
773     to
774     -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
776     to look similar to other lines, but I have no way of telling whether this
777     is the right thing to do, as I do not use Windows. No doubt I'll get told
778     if it's wrong...
781 nigel 69 Version 4.3 21-May-03
782     ---------------------
784     1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
785     Makefile.
787     2. Some refactoring to improve the quality of the code:
789     (i) The utf8_table... variables are now declared "const".
791     (ii) The code for \cx, which used the "case flipping" table to upper case
792     lower case letters, now just substracts 32. This is ASCII-specific,
793     but the whole concept of \cx is ASCII-specific, so it seems
794     reasonable.
796     (iii) PCRE was using its character types table to recognize decimal and
797     hexadecimal digits in the pattern. This is silly, because it handles
798     only 0-9, a-f, and A-F, but the character types table is locale-
799     specific, which means strange things might happen. A private
800     table is now used for this - though it costs 256 bytes, a table is
801     much faster than multiple explicit tests. Of course, the standard
802     character types table is still used for matching digits in subject
803     strings against \d.
805     (iv) Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
806     ending in _t are). So I've renamed it as ESC_tee.
808     3. The first argument for regexec() in the POSIX wrapper should have been
809     defined as "const".
811     4. Changed pcretest to use malloc() for its buffers so that they can be
812     Electric Fenced for debugging.
814     5. There were several places in the code where, in UTF-8 mode, PCRE would try
815     to read one or more bytes before the start of the subject string. Often this
816     had no effect on PCRE's behaviour, but in some circumstances it could
817     provoke a segmentation fault.
819     6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
820     to try to read one or more bytes before the start of the subject string.
822     7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
823     UTF-8 support could misbehave in various ways if the subject string
824     contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
825     area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
826     back over UTF-8 characters.)
829 nigel 67 Version 4.2 14-Apr-03
830     ---------------------
832     1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
834     2. Changes to the building process, supplied by Ronald Landheer-Cieslak
835     [ON_WINDOWS]: new variable, "#" on non-Windows platforms
836     [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
837     [WIN_PREFIX]: new variable, "cyg" for Cygwin
838     * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
839     and BUILD_EXEEXT
840     Note: automatic setting of the BUILD variables is not yet working
841     set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
842     compile-time but not at link-time
843     [LINK]: use for linking executables only
844     make different versions for Windows and non-Windows
845     [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
846     libraries
847     [LINK_FOR_BUILD]: new variable
848     [OBJEXT]: use throughout
849     [EXEEXT]: use throughout
850     <winshared>: new target
851     <wininstall>: new target
852     <dftables.o>: use native compiler
853     <dftables>: use native linker
854     <install>: handle Windows platform correctly
855     <clean>: ditto
856     <check>: ditto
857     copy DLL to top builddir before testing
859     As part of these changes, -no-undefined was removed again. This was reported
860     to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
861     in any case.
863     3. Some tidies to get rid of compiler warnings:
865     . In the match_data structure, match_limit was an unsigned long int, whereas
866     match_call_count was an int. I've made them both unsigned long ints.
868     . In pcretest the fact that a const uschar * doesn't automatically cast to
869     a void * provoked a warning.
871     . Turning on some more compiler warnings threw up some "shadow" variables
872     and a few more missing casts.
874     4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
875     option, a class that contained a single character with a value between 128
876     and 255 (e.g. /[\xFF]/) caused PCRE to crash.
878     5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
879     option, a class that contained several characters, but with at least one
880     whose value was between 128 and 255 caused PCRE to crash.
883 nigel 65 Version 4.1 12-Mar-03
884     ---------------------
885 nigel 63
886 nigel 65 1. Compiling with gcc -pedantic found a couple of places where casts were
887     needed, and a string in dftables.c that was longer than standard compilers are
888     required to support.
890     2. Compiling with Sun's compiler found a few more places where the code could
891     be tidied up in order to avoid warnings.
893     3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
894     first of these names is deprecated in the latest Autoconf in favour of the name
895     CC_FOR_BUILD, because "host" is typically used to mean the system on which the
896     compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
897     analogy I have changed it to CFLAGS_FOR_BUILD.
899     4. Added -no-undefined to the linking command in the Makefile, because this is
900     apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
901     linking step for the pcreposix library.
903     5. PCRE was failing to diagnose the case of two named groups with the same
904     name.
906     6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
907     literal character that is needed in the subject for a match, and scans along to
908     ensure that it is present before embarking on the full matching process. This
909     saves time in cases of nested unlimited repeats that are never going to match.
910     Problem: the scan can take a lot of time if the subject is very long (e.g.
911     megabytes), thus penalizing straightforward matches. It is now done only if the
912     amount of subject to be scanned is less than 1000 bytes.
914     7. A lesser problem with the same optimization is that it was recording the
915     first character of an anchored pattern as "needed", thus provoking a search
916     right along the subject, even when the first match of the pattern was going to
917     fail. The "needed" character is now not set for anchored patterns, unless it
918     follows something in the pattern that is of non-fixed length. Thus, it still
919     fulfils its original purpose of finding quick non-matches in cases of nested
920     unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
923     Version 4.0 17-Feb-03
924     ---------------------
926 nigel 63 1. If a comment in an extended regex that started immediately after a meta-item
927     extended to the end of string, PCRE compiled incorrect data. This could lead to
928     all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
930     2. Moved to autoconf 2.53 and libtool 1.4.2.
932     3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
933     the special perltest8 script is no longer needed - all the tests can be run
934     from a single perltest script.
936     4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
937     by \s. It has now been removed in PCRE. This means it isn't recognized as
938     whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
939     class [:space:] *does* include VT, thereby creating a mess.
941     5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
942     space and tab.
944     6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
945     its new features into the main test script, reducing the number of scripts.
947     7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
948     were backward compatible, and made the (?i) apply to the whole pattern, as if
949     /i were given. Now it behaves more logically, and applies the option setting
950     only to what follows. PCRE has been changed to follow suit. However, if it
951     finds options settings right at the start of the pattern, it extracts them into
952     the global options, as before. Thus, they show up in the info data.
954     8. Added support for the \Q...\E escape sequence. Characters in between are
955     treated as literals. This is slightly different from Perl in that $ and @ are
956     also handled as literals inside the quotes. In Perl, they will cause variable
957     interpolation. Note the following examples:
959     Pattern PCRE matches Perl matches
961     \Qabc$xyz\E abc$xyz abc followed by the contents of $xyz
962     \Qabc\$xyz\E abc\$xyz abc\$xyz
963     \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
965     For compatibility with Perl, \Q...\E sequences are recognized inside character
966     classes as well as outside them.
968     9. Re-organized 3 code statements in pcretest to avoid "overflow in
969     floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
970     (size_t) cast to one statement in pcretest and one in pcreposix to avoid
971     signed/unsigned warnings.
973     10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
974     option for pcretest, so I've replaced it by a simple function that does just
975     that job.
977     11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
978     "pcregrep -".
980     12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
981     Java package. This provides some syntactic sugar for simple cases of what my
982     documentation calls "once-only subpatterns". A pattern such as x*+ is the same
983     as (?>x*). In other words, if what is inside (?>...) is just a single repeated
984     item, you can use this simplified notation. Note that only makes sense with
985     greedy quantifiers. Consequently, the use of the possessive quantifier forces
986     greediness, whatever the setting of the PCRE_UNGREEDY option.
988     13. A change of greediness default within a pattern was not taking effect at
989     the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
990     subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
991     was abstracted outside.
993     14. PCRE now supports the \G assertion. It is true when the current matching
994     position is at the start point of the match. This differs from \A when the
995     starting offset is non-zero. Used with the /g option of pcretest (or similar
996     code), it works in the same way as it does for Perl's /g option. If all
997     alternatives of a regex begin with \G, the expression is anchored to the start
998     match position, and the "anchored" flag is set in the compiled expression.
1000     15. Some bugs concerning the handling of certain option changes within patterns
1001     have been fixed. These applied to options other than (?ims). For example,
1002     "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
1003     been the other way round. Some of this was related to change 7 above.
1005     16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
1006     features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
1007     and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
1008     POSIX classes only within a class (e.g. /[[:alpha:]]/).
1010     17. Added support for Perl's \C escape. This matches one byte, even in UTF8
1011     mode. Unlike ".", it always matches newline, whatever the setting of
1012     PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
1013     assertions. Perl allows it, but it doesn't (in general) work because it can't
1014     calculate the length of the lookbehind. At least, that's the case for Perl
1015     5.8.0 - I've been told they are going to document that it doesn't work in
1016     future.
1018     18. Added an error diagnosis for escapes that PCRE does not support: these are
1019     \L, \l, \N, \P, \p, \U, \u, and \X.
1021     19. Although correctly diagnosing a missing ']' in a character class, PCRE was
1022     reading past the end of the pattern in cases such as /[abcd/.
1024     20. PCRE was getting more memory than necessary for patterns with classes that
1025     contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
1027     21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
1028     compiling PCRE for use with Virtual Pascal.
1030     22. Small fix to the Makefile to make it work properly if the build is done
1031     outside the source tree.
1033     23. Added a new extension: a condition to go with recursion. If a conditional
1034     subpattern starts with (?(R) the "true" branch is used if recursion has
1035     happened, whereas the "false" branch is used only at the top level.
1037     24. When there was a very long string of literal characters (over 255 bytes
1038     without UTF support, over 250 bytes with UTF support), the computation of how
1039     much memory was required could be incorrect, leading to segfaults or other
1040     strange effects.
1042     25. PCRE was incorrectly assuming anchoring (either to start of subject or to
1043     start of line for a non-DOTALL pattern) when a pattern started with (.*) and
1044     there was a subsequent back reference to those brackets. This meant that, for
1045     example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
1046     possible to check for precisely this case. All we can do is abandon the
1047     optimization if .* occurs inside capturing brackets when there are any back
1048     references whatsoever. (See below for a better fix that came later.)
1050     26. The handling of the optimization for finding the first character of a
1051     non-anchored pattern, and for finding a character that is required later in the
1052     match were failing in some cases. This didn't break the matching; it just
1053     failed to optimize when it could. The way this is done has been re-implemented.
1055     27. Fixed typo in error message for invalid (?R item (it said "(?p").
1057     28. Added a new feature that provides some of the functionality that Perl
1058     provides with (?{...}). The facility is termed a "callout". The way it is done
1059     in PCRE is for the caller to provide an optional function, by setting
1060     pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
1061     global variable. By default it is unset, which disables all calling out. To get
1062     the function called, the regex must include (?C) at appropriate points. This
1063     is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
1064     This provides a means of identifying different callout points. When PCRE
1065     reaches such a point in the regex, if pcre_callout has been set, the external
1066     function is called. It is provided with data in a structure called
1067     pcre_callout_block, which is defined in pcre.h. If the function returns 0,
1068     matching continues; if it returns a non-zero value, the match at the current
1069     point fails. However, backtracking will occur if possible. [This was changed
1070     later and other features added - see item 49 below.]
1072     29. pcretest is upgraded to test the callout functionality. It provides a
1073     callout function that displays information. By default, it shows the start of
1074     the match and the current position in the text. There are some new data escapes
1075     to vary what happens:
1077     \C+ in addition, show current contents of captured substrings
1078     \C- do not supply a callout function
1079     \C!n return 1 when callout number n is reached
1080     \C!n!m return 1 when callout number n is reached for the mth time
1082     30. If pcregrep was called with the -l option and just a single file name, it
1083     output "<stdin>" if a match was found, instead of the file name.
1085     31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
1086     slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
1087     pcre_exec(). This saves a malloc/free per call. The default value of
1088     POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
1089     when configuring.
1091     32. The default maximum size of a compiled pattern is 64K. There have been a
1092     few cases of people hitting this limit. The code now uses macros to handle the
1093     storing of links as offsets within the compiled pattern. It defaults to 2-byte
1094     links, but this can be changed to 3 or 4 bytes by --with-link-size when
1095     configuring. Tests 2 and 5 work only with 2-byte links because they output
1096     debugging information about compiled patterns.
1098     33. Internal code re-arrangements:
1100     (a) Moved the debugging function for printing out a compiled regex into
1101     its own source file (printint.c) and used #include to pull it into
1102     pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
1103     separate copies.
1105     (b) Defined the list of op-code names for debugging as a macro in
1106     internal.h so that it is next to the definition of the opcodes.
1108     (c) Defined a table of op-code lengths for simpler skipping along compiled
1109     code. This is again a macro in internal.h so that it is next to the
1110     definition of the opcodes.
1112     34. Added support for recursive calls to individual subpatterns, along the
1113     lines of Robin Houston's patch (but implemented somewhat differently).
1115     35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
1116     allow it to read and process whole directories in Win32. This code was
1117     contributed by Lionel Fourquaux; it has not been tested by me.
1119     36. Added support for named subpatterns. The Python syntax (?P<name>...) is
1120     used to name a group. Names consist of alphanumerics and underscores, and must
1121     be unique. Back references use the syntax (?P=name) and recursive calls use
1122     (?P>name) which is a PCRE extension to the Python extension. Groups still have
1123     numbers. The function pcre_fullinfo() can be used after compilation to extract
1124     a name/number map. There are three relevant calls:
1126     PCRE_INFO_NAMEENTRYSIZE yields the size of each entry in the map
1127     PCRE_INFO_NAMECOUNT yields the number of entries
1128     PCRE_INFO_NAMETABLE yields a pointer to the map.
1130     The map is a vector of fixed-size entries. The size of each entry depends on
1131     the length of the longest name used. The first two bytes of each entry are the
1132     group number, most significant byte first. There follows the corresponding
1133     name, zero terminated. The names are in alphabetical order.
1135     37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
1136     case instead of 255. Making it the same both with and without UTF-8 support
1137     means that the same test output works with both.
1139     38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
1140     calling malloc() with a zero argument.
1142     39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
1143     optimization. I've improved things by keeping a bitmap of backreferences with
1144     numbers 1-31 so that if .* occurs inside capturing brackets that are not in
1145     fact referenced, the optimization can be applied. It is unlikely that a
1146     relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
1147     the match to follow \n) will appear inside brackets with a number greater than
1148     31, but if it does, any back reference > 31 suppresses the optimization.
1150     40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
1151     of disabling numbered capturing parentheses. Any opening parenthesis that is
1152     not followed by ? behaves as if it were followed by ?: but named parentheses
1153     can still be used for capturing (and they will acquire numbers in the usual
1154     way).
1156     41. Redesigned the return codes from the match() function into yes/no/error so
1157     that errors can be passed back from deep inside the nested calls. A malloc
1158     failure while inside a recursive subpattern call now causes the
1159     PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
1161     42. It is now possible to set a limit on the number of times the match()
1162     function is called in a call to pcre_exec(). This facility makes it possible to
1163     limit the amount of recursion and backtracking, though not in a directly
1164     obvious way, because the match() function is used in a number of different
1165     circumstances. The count starts from zero for each position in the subject
1166     string (for non-anchored patterns). The default limit is, for compatibility, a
1167     large number, namely 10 000 000. You can change this in two ways:
1169     (a) When configuring PCRE before making, you can use --with-match-limit=n
1170     to set a default value for the compiled library.
1172     (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
1173     a different value is set. See 45 below.
1175     If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
1177     43. Added a new function pcre_config(int, void *) to enable run-time extraction
1178     of things that can be changed at compile time. The first argument specifies
1179     what is wanted and the second points to where the information is to be placed.
1180     The current list of available information is:
1184     The output is an integer that is set to one if UTF-8 support is available;
1185     otherwise it is set to zero.
1189     The output is an integer that it set to the value of the code that is used for
1190     newline. It is either LF (10) or CR (13).
1194     The output is an integer that contains the number of bytes used for internal
1195     linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
1199     The output is an integer that contains the threshold above which the POSIX
1200     interface uses malloc() for output vectors. See item 31 above.
1204     The output is an unsigned integer that contains the default limit of the number
1205     of match() calls in a pcre_exec() execution. See 42 above.
1207     44. pcretest has been upgraded by the addition of the -C option. This causes it
1208     to extract all the available output from the new pcre_config() function, and to
1209     output it. The program then exits immediately.
1211     45. A need has arisen to pass over additional data with calls to pcre_exec() in
1212     order to support additional features. One way would have been to define
1213     pcre_exec2() (for example) with extra arguments, but this would not have been
1214     extensible, and would also have required all calls to the original function to
1215     be mapped to the new one. Instead, I have chosen to extend the mechanism that
1216     is used for passing in "extra" data from pcre_study().
1218     The pcre_extra structure is now exposed and defined in pcre.h. It currently
1219     contains the following fields:
1221     flags a bitmap indicating which of the following fields are set
1222     study_data opaque data from pcre_study()
1223     match_limit a way of specifying a limit on match() calls for a specific
1224     call to pcre_exec()
1225     callout_data data for callouts (see 49 below)
1227     The flag bits are also defined in pcre.h, and are
1233     The pcre_study() function now returns one of these new pcre_extra blocks, with
1234     the actual study data pointed to by the study_data field, and the
1235     PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
1236     before. That is, this change is entirely upwards-compatible and requires no
1237     change to existing code.
1239     If you want to pass in additional data to pcre_exec(), you can either place it
1240     in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
1241     block.
1243     46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
1244     data string contains the escape sequence \M, pcretest calls pcre_exec() several
1245     times with different match limits, until it finds the minimum value needed for
1246     pcre_exec() to complete. The value is then output. This can be instructive; for
1247     most simple matches the number is quite small, but for pathological cases it
1248     gets very large very quickly.
1250     47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
1251     returns the size of the data block pointed to by the study_data field in a
1252     pcre_extra block, that is, the value that was passed as the argument to
1253     pcre_malloc() when PCRE was getting memory in which to place the information
1254     created by pcre_study(). The fourth argument should point to a size_t variable.
1255     pcretest has been extended so that this information is shown after a successful
1256     pcre_study() call when information about the compiled regex is being displayed.
1258     48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
1259     because what follows is always an absolute path. (Later: it turns out that this
1260     is more than cosmetic for MinGW, because it doesn't like empty path
1261     components.)
1263     49. Some changes have been made to the callout feature (see 28 above):
1265     (i) A callout function now has three choices for what it returns:
1267     0 => success, carry on matching
1268     > 0 => failure at this point, but backtrack if possible
1269     < 0 => serious error, return this value from pcre_exec()
1271     Negative values should normally be chosen from the set of PCRE_ERROR_xxx
1272     values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
1273     "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
1274     use by callout functions. It will never be used by PCRE itself.
1276     (ii) The pcre_extra structure (see 45 above) has a void * field called
1277     callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
1278     pcre_callout_block structure has a field of the same name. The contents of
1279     the field passed in the pcre_extra structure are passed to the callout
1280     function in the corresponding field in the callout block. This makes it
1281     easier to use the same callout-containing regex from multiple threads. For
1282     testing, the pcretest program has a new data escape
1284     \C*n pass the number n (may be negative) as callout_data
1286     If the callout function in pcretest receives a non-zero value as
1287     callout_data, it returns that value.
1289     50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
1290     there were some redundant $(CFLAGS) in commands that are now specified as
1291     $(LINK), which already includes $(CFLAGS).
1293     51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
1294     has been compiled with UTF-8 support *and* pcre_compile() has been compiled
1295     with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
1296     one-byte characters throughout. Note that case-insensitive matching applies
1297     only to characters whose values are less than 256. PCRE doesn't support the
1298     notion of cases for higher-valued characters.
1300     (i) A character class whose characters are all within 0-255 is handled as
1301     a bit map, and the map is inverted for negative classes. Previously, a
1302     character > 255 always failed to match such a class; however it should
1303     match if the class was a negative one (e.g. [^ab]). This has been fixed.
1305     (ii) A negated character class with a single character < 255 is coded as
1306     "not this character" (OP_NOT). This wasn't working properly when the test
1307     character was multibyte, either singly or repeated.
1309     (iii) Repeats of multibyte characters are now handled correctly in UTF-8
1310     mode, for example: \x{100}{2,3}.
1312     (iv) The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
1313     singly or repeated) now correctly test multibyte characters. However,
1314     PCRE doesn't recognize any characters with values greater than 255 as
1315     digits, spaces, or word characters. Such characters always match \D, \S,
1316     and \W, and never match \d, \s, or \w.
1318     (v) Classes may now contain characters and character ranges with values
1319     greater than 255. For example: [ab\x{100}-\x{400}].
1321     (vi) pcregrep now has a --utf-8 option (synonym -u) which makes it call
1322     PCRE in UTF-8 mode.
1324     52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
1325     PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
1326     retained for backwards compatibility. (Note that LASTLITERAL is also a byte
1327     value.)
1329     53. The single man page has become too large. I have therefore split it up into
1330     a number of separate man pages. These also give rise to individual HTML pages;
1331     these are now put in a separate directory, and there is an index.html page that
1332     lists them all. Some hyperlinking between the pages has been installed.
1334     54. Added convenience functions for handling named capturing parentheses.
1336     55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
1337     aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
1338     true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
1339     are faulted.
1341     56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
1342     calling configure. These values are used when compiling the dftables.c program
1343     which is run to generate the source of the default character tables. They
1344     default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
1345     you will need to set these values.
1347     57. Updated the building process for Windows DLL, as provided by Fred Cox.
1350     Version 3.9 02-Jan-02
1351 nigel 61 ---------------------
1353     1. A bit of extraneous text had somehow crept into the pcregrep documentation.
1355     2. If --disable-static was given, the building process failed when trying to
1356     build pcretest and pcregrep. (For some reason it was using libtool to compile
1357     them, which is not right, as they aren't part of the library.)
1360 nigel 59 Version 3.8 18-Dec-01
1361     ---------------------
1363     1. The experimental UTF-8 code was completely screwed up. It was packing the
1364     bytes in the wrong order. How dumb can you get?
1367 nigel 57 Version 3.7 29-Oct-01
1368     ---------------------
1370     1. In updating pcretest to check change 1 of version 3.6, I screwed up.
1371     This caused pcretest, when used on the test data, to segfault. Unfortunately,
1372     this didn't happen under Solaris 8, where I normally test things.
1374 nigel 59 2. The Makefile had to be changed to make it work on BSD systems, where 'make'
1375     doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
1376     isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
1377     this fix an hour or so after the initial 3.7 release.)
1378 nigel 57
1379 nigel 59
1380 nigel 55 Version 3.6 23-Oct-01
1381     ---------------------
1383     1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
1384     offsets passed as NULL with zero offset count.
1386     2. The config.guess and config.sub files had not been updated when I moved to
1387     the latest autoconf.
1390 nigel 53 Version 3.5 15-Aug-01
1391     ---------------------
1392 nigel 5
1393 nigel 53 1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
1394     had been forgotten.
1396     2. By using declared but undefined structures, we can avoid using "void"
1397     definitions in pcre.h while keeping the internal definitions of the structures
1398     private.
1400     3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
1401     user point of view, this means that both static and shared libraries are built
1402     by default, but this can be individually controlled. More of the work of
1403     handling this static/shared cases is now inside libtool instead of PCRE's make
1404     file.
1406     4. The pcretest utility is now installed along with pcregrep because it is
1407     useful for users (to test regexs) and by doing this, it automatically gets
1408     relinked by libtool. The documentation has been turned into a man page, so
1409     there are now .1, .txt, and .html versions in /doc.
1411     5. Upgrades to pcregrep:
1412     (i) Added long-form option names like gnu grep.
1413     (ii) Added --help to list all options with an explanatory phrase.
1414     (iii) Added -r, --recursive to recurse into sub-directories.
1415     (iv) Added -f, --file to read patterns from a file.
1417     6. pcre_exec() was referring to its "code" argument before testing that
1418     argument for NULL (and giving an error if it was NULL).
1420     7. Upgraded Makefile.in to allow for compiling in a different directory from
1421     the source directory.
1423     8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
1424     options bits, the pointer it was passed was to an int instead of to an unsigned
1425     long int. This mattered only on 64-bit systems.
1427     9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
1428     generated) instead of pcre.in, which it its source. Also made the same change
1429     in several of the .c files.
1431     10. A new release of gcc defines printf() as a macro, which broke pcretest
1432     because it had an ifdef in the middle of a string argument for printf(). Fixed
1433     by using separate calls to printf().
1435     11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
1436     script, to force use of CR or LF instead of \n in the source. On non-Unix
1437     systems, the value can be set in config.h.
1439     12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
1440     absolute limit. Changed the text of the error message to make this clear, and
1441     likewise updated the man page.
1443     13. The limit of 99 on the number of capturing subpatterns has been removed.
1444     The new limit is 65535, which I hope will not be a "real" limit.
1447 nigel 51 Version 3.4 22-Aug-00
1448     ---------------------
1450     1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
1452     2. Diagnose condition (?(0) as an error instead of crashing on matching.
1455 nigel 49 Version 3.3 01-Aug-00
1456     ---------------------
1458     1. If an octal character was given, but the value was greater than \377, it
1459     was not getting masked to the least significant bits, as documented. This could
1460     lead to crashes in some systems.
1462     2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
1463     the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
1465     3. Added the functions pcre_free_substring() and pcre_free_substring_list().
1466     These just pass their arguments on to (pcre_free)(), but they are provided
1467     because some uses of PCRE bind it to non-C systems that can call its functions,
1468     but cannot call free() or pcre_free() directly.
1470     4. Add "make test" as a synonym for "make check". Corrected some comments in
1471     the Makefile.
1473     5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
1474     Makefile.
1476     6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
1477     command called pgrep for grepping around the active processes.
1479     7. Added the beginnings of support for UTF-8 character strings.
1481     8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
1482     RANLIB to ./ltconfig so that they are used by libtool. I think these are all
1483     the relevant ones. (AR is not passed because ./ltconfig does its own figuring
1484     out for the ar command.)
1487 nigel 47 Version 3.2 12-May-00
1488     ---------------------
1490     This is purely a bug fixing release.
1492     1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
1493     of ZA. This was just one example of several cases that could provoke this bug,
1494     which was introduced by change 9 of version 2.00. The code for breaking
1495     infinite loops after an iteration that matches an empty string was't working
1496     correctly.
1498     2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
1499     when matched against abbab (for example). After matching an empty string, it
1500     wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
1501     caused it to match further down the string than it should.
1503     3. The code contained an inclusion of sys/types.h. It isn't clear why this
1504     was there because it doesn't seem to be needed, and it causes trouble on some
1505     systems, as it is not a Standard C header. It has been removed.
1507     4. Made 4 silly changes to the source to avoid stupid compiler warnings that
1508     were reported on the Macintosh. The changes were from
1510     while ((c = *(++ptr)) != 0 && c != '\n');
1511     to
1512     while ((c = *(++ptr)) != 0 && c != '\n') ;
1514     Totally extraordinary, but if that's what it takes...
1516     5. PCRE is being used in one environment where neither memmove() nor bcopy() is
1517     available. Added HAVE_BCOPY and an autoconf test for it; if neither
1518     HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
1519     assumes the way PCRE uses memmove() (always moving upwards).
1521     6. PCRE is being used in one environment where strchr() is not available. There
1522     was only one use in pcre.c, and writing it out to avoid strchr() probably gives
1523     faster code anyway.
1526 nigel 45 Version 3.1 09-Feb-00
1527     ---------------------
1529     The only change in this release is the fixing of some bugs in Makefile.in for
1530     the "install" target:
1532     (1) It was failing to install pcreposix.h.
1534     (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
1537 nigel 43 Version 3.0 01-Feb-00
1538     ---------------------
1539 nigel 41
1540     1. Add support for the /+ modifier to perltest (to output $` like it does in
1541     pcretest).
1543     2. Add support for the /g modifier to perltest.
1545     3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
1546     matches null strings.
1548     4. Fix perltest so that it doesn't do unwanted things when fed an empty
1549     pattern. Perl treats empty patterns specially - it reuses the most recent
1550     pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
1551     effect.
1553     5. The POSIX interface was broken in that it was just handing over the POSIX
1554     captured string vector to pcre_exec(), but (since release 2.00) PCRE has
1555     required a bigger vector, with some working space on the end. This means that
1556     the POSIX wrapper now has to get and free some memory, and copy the results.
1558 nigel 43 6. Added some simple autoconf support, placing the test data and the
1559     documentation in separate directories, re-organizing some of the
1560     information files, and making it build pcre-config (a GNU standard). Also added
1561     libtool support for building PCRE as a shared library, which is now the
1562     default.
1563 nigel 41
1564 nigel 43 7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
1565     09 are not valid octal constants. Single digits will be used for minor values
1566     less than 10.
1568     8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
1569     existing programs that set these in the POSIX interface can use PCRE without
1570     modification.
1572     9. Added a new function, pcre_fullinfo() with an extensible interface. It can
1573     return all that pcre_info() returns, plus additional data. The pcre_info()
1574     function is retained for compatibility, but is considered to be obsolete.
1576     10. Added experimental recursion feature (?R) to handle one common case that
1577     Perl 5.6 will be able to do with (?p{...}).
1579     11. Added support for POSIX character classes like [:alpha:], which Perl is
1580     adopting.
1583 nigel 39 Version 2.08 31-Aug-99
1584     ----------------------
1586     1. When startoffset was not zero and the pattern began with ".*", PCRE was not
1587     trying to match at the startoffset position, but instead was moving forward to
1588     the next newline as if a previous match had failed.
1590     2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
1591     and could get into a loop if a null string was matched other than at the start
1592     of the subject.
1594     3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
1595     be distinguished at compile time, and for completeness also added PCRE_DATE.
1597     5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
1598     in GnuWin32 environments.
1601 nigel 37 Version 2.07 29-Jul-99
1602     ----------------------
1604     1. The documentation is now supplied in plain text form and HTML as well as in
1605     the form of man page sources.
1607     2. C++ compilers don't like assigning (void *) values to other pointer types.
1608     In particular this affects malloc(). Although there is no problem in Standard
1609     C, I've put in casts to keep C++ compilers happy.
1611     3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
1612     should be (const char *).
1614     4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
1615     be useful for non-Unix systems who don't want to bother with the POSIX stuff.
1616     However, I haven't made this a standard facility. The documentation doesn't
1617     mention it, and the Makefile doesn't support it.
1619     5. The Makefile now contains an "install" target, with editable destinations at
1620     the top of the file. The pcretest program is not installed.
1622     6. pgrep -V now gives the PCRE version number and date.
1624     7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
1625     causing the entire string to be ignored, instead of just the last character.
1627     8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
1628     non-matching string, it can take a very, very long time, even for strings of
1629     quite modest length, because of the nested recursion. PCRE now does better in
1630     some of these cases. It does this by remembering the last required literal
1631     character in the pattern, and pre-searching the subject to ensure it is present
1632     before running the real match. In other words, it applies a heuristic to detect
1633     some types of certain failure quickly, and in the above example, if presented
1634     with a string that has no trailing " it gives "no match" very quickly.
1636     9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
1637     other alternatives are tried instead.
1640 nigel 35 Version 2.06 09-Jun-99
1641     ----------------------
1643     1. Change pcretest's output for amount of store used to show just the code
1644     space, because the remainder (the data block) varies in size between 32-bit and
1645     64-bit systems.
1647     2. Added an extra argument to pcre_exec() to supply an offset in the subject to
1648     start matching at. This allows lookbehinds to work when searching for multiple
1649     occurrences in a string.
1651     3. Added additional options to pcretest for testing multiple occurrences:
1653     /+ outputs the rest of the string that follows a match
1654     /g loops for multiple occurrences, using the new startoffset argument
1655     /G loops for multiple occurrences by passing an incremented pointer
1657     4. PCRE wasn't doing the "first character" optimization for patterns starting
1658     with \b or \B, though it was doing it for other lookbehind assertions. That is,
1659     it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
1660     the letter 'x'. On long subject strings, this gives a significant speed-up.
1663 nigel 33 Version 2.05 21-Apr-99
1664     ----------------------
1666     1. Changed the type of magic_number from int to long int so that it works
1667     properly on 16-bit systems.
1669     2. Fixed a bug which caused patterns starting with .* not to work correctly
1670     when the subject string contained newline characters. PCRE was assuming
1671     anchoring for such patterns in all cases, which is not correct because .* will
1672     not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
1673     DOTALL is set at top level; otherwise it knows that patterns starting with .*
1674     must be retried after every newline in the subject.
1677 nigel 31 Version 2.04 18-Feb-99
1678     ----------------------
1680     1. For parenthesized subpatterns with repeats whose minimum was zero, the
1681     computation of the store needed to hold the pattern was incorrect (too large).
1682     If such patterns were nested a few deep, this could multiply and become a real
1683     problem.
1685     2. Added /M option to pcretest to show the memory requirement of a specific
1686     pattern. Made -m a synonym of -s (which does this globally) for compatibility.
1688     3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
1689     compiled in such a way that the backtracking after subsequent failure was
1690     pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
1691     ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
1694 nigel 29 Version 2.03 02-Feb-99
1695     ----------------------
1697     1. Fixed typo and small mistake in man page.
1699 nigel 31 2. Added 4th condition (GPL supersedes if conflict) and created separate
1700     LICENCE file containing the conditions.
1701 nigel 29
1702     3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
1703     Perl, that is the internal \ allows the delimiter to be included in the
1704     pattern. Locked out the use of \ as a delimiter. If \ immediately follows
1705     the final delimiter, add \ to the end of the pattern (to test the error).
1707     4. Added the convenience functions for extracting substrings after a successful
1708     match. Updated pcretest to make it able to test these functions.
1711 nigel 27 Version 2.02 14-Jan-99
1712     ----------------------
1714     1. Initialized the working variables associated with each extraction so that
1715     their saving and restoring doesn't refer to uninitialized store.
1717     2. Put dummy code into study.c in order to trick the optimizer of the IBM C
1718     compiler for OS/2 into generating correct code. Apparently IBM isn't going to
1719     fix the problem.
1721     3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
1722     calls, and wasn't printing the correct value for compiling calls. Increased the
1723     default value of LOOPREPEAT, and the number of significant figures in the
1724     times.
1726     4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
1728     5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
1729     a building problem on Windows NT with a FAT file system.
1732 nigel 25 Version 2.01 21-Oct-98
1733     ----------------------
1735     1. Changed the API for pcre_compile() to allow for the provision of a pointer
1736     to character tables built by pcre_maketables() in the current locale. If NULL
1737     is passed, the default tables are used.
1740 nigel 23 Version 2.00 24-Sep-98
1741 nigel 21 ----------------------
1743 nigel 23 1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
1744     it any more.
1746     2. Allow quantification of (?>) groups, and make it work correctly.
1748     3. The first character computation wasn't working for (?>) groups.
1750     4. Correct the implementation of \Z (it is permitted to match on the \n at the
1751     end of the subject) and add 5.005's \z, which really does match only at the
1752     very end of the subject.
1754     5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
1756     6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
1757     DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
1758     localized options. All options to pcre_study() were also removed.
1760     7. Add other new features from 5.005:
1762     $(?<= positive lookbehind
1763     $(?<! negative lookbehind
1764     (?imsx-imsx) added the unsetting capability
1765     such a setting is global if at outer level; local otherwise
1766     (?imsx-imsx:) non-capturing groups with option setting
1767     (?(cond)re|re) conditional pattern matching
1769     A backreference to itself in a repeated group matches the previous
1770     captured string.
1772     8. General tidying up of studying (both automatic and via "study")
1773     consequential on the addition of new assertions.
1775     9. As in 5.005, unlimited repeated groups that could match an empty substring
1776     are no longer faulted at compile time. Instead, the loop is forcibly broken at
1777     runtime if any iteration does actually match an empty substring.
1779     10. Include the RunTest script in the distribution.
1781     11. Added tests from the Perl 5.005_02 distribution. This showed up a few
1782     discrepancies, some of which were old and were also with respect to 5.004. They
1783     have now been fixed.
1786     Version 1.09 28-Apr-98
1787     ----------------------
1789 nigel 21 1. A negated single character class followed by a quantifier with a minimum
1790     value of one (e.g. [^x]{1,6} ) was not compiled correctly. This could lead to
1791     program crashes, or just wrong answers. This did not apply to negated classes
1792     containing more than one character, or to minima other than one.
1795 nigel 19 Version 1.08 27-Mar-98
1796     ----------------------
1798     1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
1800     2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
1801     latter must appear before anything that relies on it in the pattern.
1804 nigel 17 Version 1.07 16-Feb-98
1805     ----------------------
1807     1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
1808     repeat of a potentially empty string).
1811 nigel 15 Version 1.06 23-Jan-98
1812     ----------------------
1814     1. Added Markus Oberhumer's little patches for C++.
1816     2. Literal strings longer than 255 characters were broken.
1819 nigel 13 Version 1.05 23-Dec-97
1820     ----------------------
1822     1. Negated character classes containing more than one character were failing if
1823     PCRE_CASELESS was set at run time.
1826 nigel 11 Version 1.04 19-Dec-97
1827     ----------------------
1829     1. Corrected the man page, where some "const" qualifiers had been omitted.
1831     2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
1832     input syntax.
1834     3. Fixed memory leak which occurred when a regex with back references was
1835     matched with an offsets vector that wasn't big enough. The temporary memory
1836     that is used in this case wasn't being freed if the match failed.
1838     4. Tidied pcretest to ensure it frees memory that it gets.
1840     5. Temporary memory was being obtained in the case where the passed offsets
1841     vector was exactly big enough.
1843     6. Corrected definition of offsetof() from change 5 below.
1845     7. I had screwed up change 6 below and broken the rules for the use of
1846     setjmp(). Now fixed.
1849 nigel 9 Version 1.03 18-Dec-97
1850     ----------------------
1852     1. A erroneous regex with a missing opening parenthesis was correctly
1853     diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
1854     on some systems.
1856     2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
1857     it was reported that one broken compiler failed on the former because "code" is
1858     also an independent variable.
1860     3. The erroneous regex a[]b caused an array overrun reference.
1862     4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
1863     fail on data ending with that character. (It was going on too far, and checking
1864     the next character, typically a binary zero.) This was specific to the
1865     optimized code for single-character negative classes.
1867     5. Added a contributed patch from the TIN world which does the following:
1869     + Add an undef for memmove, in case the the system defines a macro for it.
1871     + Add a definition of offsetof(), in case there isn't one. (I don't know
1872     the reason behind this - offsetof() is part of the ANSI standard - but
1873     it does no harm).
1875     + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
1876     most of the places where whitespace preceded '#'. I have given up and
1877     allowed the remaining 2 cases to be at the margin.
1879     + Rename some variables in pcre to eliminate shadowing. This seems very
1880     pedantic, but does no harm, of course.
1882     6. Moved the call to setjmp() into its own function, to get rid of warnings
1883     from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
1885     7. Constructs such as \d{8,} were compiling into the equivalent of
1886 nigel 11 \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
1887 nigel 9 outcome, but in this particular case used more store than had been allocated,
1888     which caused the bug to be discovered because it threw up an internal error.
1890     8. The debugging code in both pcre and pcretest for outputting the compiled
1891     form of a regex was going wrong in the case of back references followed by
1892     curly-bracketed repeats.
1895 nigel 7 Version 1.02 12-Dec-97
1896     ----------------------
1898     1. Typos in pcre.3 and comments in the source fixed.
1900     2. Applied a contributed patch to get rid of places where it used to remove
1901     'const' from variables, and fixed some signed/unsigned and uninitialized
1902     variable warnings.
1904     3. Added the "runtest" target to Makefile.
1906     4. Set default compiler flag to -O2 rather than just -O.
1909 nigel 5 Version 1.01 19-Nov-97
1910     ----------------------
1912     1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
1913     like /([ab]*)*/, that is, for classes with more than one character in them.
1915     2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
1916     as /((?>a*))*/ (a PCRE_EXTRA facility).
1919     Version 1.00 18-Nov-97
1920     ----------------------
1922     1. Added compile-time macros to support systems such as SunOS4 which don't have
1923     memmove() or strerror() but have other things that can be used instead.
1925     2. Arranged that "make clean" removes the executables.
1928 nigel 3 Version 0.99 27-Oct-97
1929     ----------------------
1931     1. Fixed bug in code for optimizing classes with only one character. It was
1932     initializing a 32-byte map regardless, which could cause it to run off the end
1933     of the memory it had got.
1935     2. Added, conditional on PCRE_EXTRA, the proposed (?>REGEX) construction.
1938     Version 0.98 22-Oct-97
1939     ----------------------
1941     1. Fixed bug in code for handling temporary memory usage when there are more
1942     back references than supplied space in the ovector. This could cause segfaults.
1945     Version 0.97 21-Oct-97
1946     ----------------------
1948     1. Added the \X "cut" facility, conditional on PCRE_EXTRA.
1950     2. Optimized negated single characters not to use a bit map.
1952     3. Brought error texts together as macro definitions; clarified some of them;
1953     fixed one that was wrong - it said "range out of order" when it meant "invalid
1954     escape sequence".
1956     4. Changed some char * arguments to const char *.
1958     5. Added PCRE_NOTBOL and PCRE_NOTEOL (from POSIX).
1960     6. Added the POSIX-style API wrapper in pcreposix.a and testing facilities in
1961     pcretest.
1964     Version 0.96 16-Oct-97
1965     ----------------------
1967     1. Added a simple "pgrep" utility to the distribution.
1969     2. Fixed an incompatibility with Perl: "{" is now treated as a normal character
1970     unless it appears in one of the precise forms "{ddd}", "{ddd,}", or "{ddd,ddd}"
1971     where "ddd" means "one or more decimal digits".
1973     3. Fixed serious bug. If a pattern had a back reference, but the call to
1974     pcre_exec() didn't supply a large enough ovector to record the related
1975     identifying subpattern, the match always failed. PCRE now remembers the number
1976     of the largest back reference, and gets some temporary memory in which to save
1977     the offsets during matching if necessary, in order to ensure that
1978     backreferences always work.
1980     4. Increased the compatibility with Perl in a number of ways:
1982     (a) . no longer matches \n by default; an option PCRE_DOTALL is provided
1983     to request this handling. The option can be set at compile or exec time.
1985     (b) $ matches before a terminating newline by default; an option
1986     PCRE_DOLLAR_ENDONLY is provided to override this (but not in multiline
1987     mode). The option can be set at compile or exec time.
1989     (c) The handling of \ followed by a digit other than 0 is now supposed to be
1990     the same as Perl's. If the decimal number it represents is less than 10
1991     or there aren't that many previous left capturing parentheses, an octal
1992     escape is read. Inside a character class, it's always an octal escape,
1993     even if it is a single digit.
1995     (d) An escaped but undefined alphabetic character is taken as a literal,
1996     unless PCRE_EXTRA is set. Currently this just reserves the remaining
1997     escapes.
1999     (e) {0} is now permitted. (The previous item is removed from the compiled
2000     pattern).
2002     5. Changed all the names of code files so that the basic parts are no longer
2003     than 10 characters, and abolished the teeny "globals.c" file.
2005     6. Changed the handling of character classes; they are now done with a 32-byte
2006     bit map always.
2008     7. Added the -d and /D options to pcretest to make it possible to look at the
2009     internals of compilation without having to recompile pcre.
2012     Version 0.95 23-Sep-97
2013     ----------------------
2015     1. Fixed bug in pre-pass concerning escaped "normal" characters such as \x5c or
2016     \x20 at the start of a run of normal characters. These were being treated as
2017     real characters, instead of the source characters being re-checked.
2020     Version 0.94 18-Sep-97
2021     ----------------------
2023     1. The functions are now thread-safe, with the caveat that the global variables
2024     containing pointers to malloc() and free() or alternative functions are the
2025     same for all threads.
2027     2. Get pcre_study() to generate a bitmap of initial characters for non-
2028     anchored patterns when this is possible, and use it if passed to pcre_exec().
2031     Version 0.93 15-Sep-97
2032     ----------------------
2034     1. /(b)|(:+)/ was computing an incorrect first character.
2036     2. Add pcre_study() to the API and the passing of pcre_extra to pcre_exec(),
2037     but not actually doing anything yet.
2039     3. Treat "-" characters in classes that cannot be part of ranges as literals,
2040     as Perl does (e.g. [-az] or [az-]).
2042     4. Set the anchored flag if a branch starts with .* or .*? because that tests
2043     all possible positions.
2045     5. Split up into different modules to avoid including unneeded functions in a
2046     compiled binary. However, compile and exec are still in one module. The "study"
2047     function is split off.
2049     6. The character tables are now in a separate module whose source is generated
2050     by an auxiliary program - but can then be edited by hand if required. There are
2051     now no calls to isalnum(), isspace(), isdigit(), isxdigit(), tolower() or
2052     toupper() in the code.
2054     7. Turn the malloc/free funtions variables into pcre_malloc and pcre_free and
2055     make them global. Abolish the function for setting them, as the caller can now
2056     set them directly.
2059     Version 0.92 11-Sep-97
2060     ----------------------
2062     1. A repeat with a fixed maximum and a minimum of 1 for an ordinary character
2063     (e.g. /a{1,3}/) was broken (I mis-optimized it).
2065     2. Caseless matching was not working in character classes if the characters in
2066     the pattern were in upper case.
2068     3. Make ranges like [W-c] work in the same way as Perl for caseless matching.
2070     4. Make PCRE_ANCHORED public and accept as a compile option.
2072     5. Add an options word to pcre_exec() and accept PCRE_ANCHORED and
2073     PCRE_CASELESS at run time. Add escapes \A and \I to pcretest to cause it to
2074     pass them.
2076     6. Give an error if bad option bits passed at compile or run time.
2078     7. Add PCRE_MULTILINE at compile and exec time, and (?m) as well. Add \M to
2079     pcretest to cause it to pass that flag.
2081     8. Add pcre_info(), to get the number of identifying subpatterns, the stored
2082     options, and the first character, if set.
2084     9. Recognize C+ or C{n,m} where n >= 1 as providing a fixed starting character.
2087     Version 0.91 10-Sep-97
2088     ----------------------
2090     1. PCRE was failing to diagnose unlimited repeats of subpatterns that could
2091     match the empty string as in /(a*)*/. It was looping and ultimately crashing.
2093     2. PCRE was looping on encountering an indefinitely repeated back reference to
2094     a subpattern that had matched an empty string, e.g. /(a|)\1*/. It now does what
2095     Perl does - treats the match as successful.
2097     ****

ViewVC Help
Powered by ViewVC 1.1.12