ViewVC logotype

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log

Revision 157 - (hide annotations) (download)
Tue May 1 09:19:56 2007 UTC (7 years, 11 months ago) by ph10
File size: 127019 byte(s)
Try "french" locale if "fr_FR" not present. Update NEWS re Win32 libs.

1 nigel 3 ChangeLog for PCRE
2     ------------------
4 ph10 157 Version 7.2 01-May-07
5     ---------------------
7     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
8     which is apparently normally available under Windows.
11 ph10 155 Version 7.1 24-Apr-07
12 ph10 98 ---------------------
14 ph10 111 1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
15 ph10 122 that is more "standard", making use of automake and other Autotools. There
16 ph10 99 is some re-arrangement of the files and adjustment of comments consequent
17     on this.
18 ph10 111
19     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
20     for recursive directory scanning broke on some systems because the files
21     are not scanned in any specific order and on different systems the order
22     was different. A call to "sort" has been inserted into RunGrepTest for the
23     approprate test as a short-term fix. In the longer term there may be an
24 ph10 100 alternative.
25 ph10 111
26 ph10 100 3. I had an email from Eric Raymond about problems translating some of PCRE's
27 ph10 111 man pages to HTML (despite the fact that I distribute HTML pages, some
28     people do their own conversions for various reasons). The problems
29     concerned the use of low-level troff macros .br and .in. I have therefore
30     removed all such uses from the man pages (some were redundant, some could
31 ph10 113 be replaced by .nf/.fi pairs). The 132html script that I use to generate
32     HTML has been updated to handle .nf/.fi and to complain if it encounters
33     .br or .in.
34 ph10 111
35 ph10 100 4. Updated comments in configure.ac that get placed in config.h.in and also
36 ph10 123 arranged for config.h to be included in the distribution, with the name
37 ph10 111 config.h.generic, for the benefit of those who have to compile without
38     Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
40     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
41     Weber: (1) pcre_internal.h was missing some function renames; (2) updated
42 ph10 127 makevp.bat for the current PCRE, using the additional files
43 ph10 135 makevp_c.txt, makevp_l.txt, and pcregexp.pas.
44 ph10 111
45     6. A Windows user reported a minor discrepancy with test 2, which turned out
46     to be caused by a trailing space on an input line that had got lost in his
47 ph10 102 copy. The trailing space was an accident, so I've just removed it.
48 ph10 111
49 ph10 104 7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
50 ph10 111 that is needed.
52 ph10 105 8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
53     as "const" (a) because they are and (b) because it helps the PHP
54     maintainers who have recently made a script to detect big data structures
55 ph10 111 in the php code that should be moved to the .rodata section. I remembered
56     to update Builducptable as well, so it won't revert if ucptable.h is ever
57 ph10 105 re-created.
58 ph10 111
59     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
60 ph10 107 pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
61 ph10 111 order to be able to cut out the UTF-8 tables in the latter when UTF-8
62     support is not required. This saves 1.5-2K of code, which is important in
63 ph10 107 some applications.
64 ph10 111
65 ph10 110 Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
66 ph10 111 so as not to refer to the tables, even though these functions will never be
67     called when UTF-8 support is disabled. Otherwise there are problems with a
68     shared library.
70 ph10 118 10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
72     (a) It was defining its arguments as char * instead of void *.
74 ph10 123 (b) It was assuming that all moves were upwards in memory; this was true
75     a long time ago when I wrote it, but is no longer the case.
77 ph10 118 The emulated memove() is provided for those environments that have neither
78 ph10 123 memmove() nor bcopy(). I didn't think anyone used it these days, but that
79 ph10 118 is clearly not the case, as these two bugs were recently reported.
80 ph10 123
81 ph10 111 11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
82 ph10 123 and Detrail to create the HTML documentation, the .txt form of the man
83     pages, and it removes trailing spaces from listed files. It also creates
84     pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
85     case, it wraps all the #defines with #ifndefs. This script should be run
86 ph10 111 before "make dist".
87 ph10 123
88 ph10 115 12. Fixed two fairly obscure bugs concerned with quantified caseless matching
89     with Unicode property support.
90 ph10 123
91     (a) For a maximizing quantifier, if the two different cases of the
92     character were of different lengths in their UTF-8 codings (there are
93     some cases like this - I found 11), and the matching function had to
94 ph10 115 back up over a mixture of the two cases, it incorrectly assumed they
95     were both the same length.
96 ph10 123
97     (b) When PCRE was configured to use the heap rather than the stack for
98     recursion during matching, it was not correctly preserving the data for
99     the other case of a UTF-8 character when checking ahead for a match
100     while processing a minimizing repeat. If the check also involved
101 ph10 115 matching a wide character, but failed, corruption could cause an
102     erroneous result when trying to check for a repeat of the original
103     character.
104 ph10 123
105 ph10 116 13. Some tidying changes to the testing mechanism:
106 ph10 98
107 ph10 116 (a) The RunTest script now detects the internal link size and whether there
108     is UTF-8 and UCP support by running ./pcretest -C instead of relying on
109 ph10 123 values substituted by "configure". (The RunGrepTest script already did
110     this for UTF-8.) The configure.ac script no longer substitutes the
111     relevant variables.
113 ph10 116 (b) The debugging options /B and /D in pcretest show the compiled bytecode
114     with length and offset values. This means that the output is different
115     for different internal link sizes. Test 2 is skipped for link sizes
116     other than 2 because of this, bypassing the problem. Unfortunately,
117     there was also a test in test 3 (the locale tests) that used /B and
118 ph10 123 failed for link sizes other than 2. Rather than cut the whole test out,
119     I have added a new /Z option to pcretest that replaces the length and
120     offset values with spaces. This is now used to make test 3 independent
121 ph10 122 of link size. (Test 2 will be tidied up later.)
122 ph10 123
123     14. If erroroffset was passed as NULL to pcre_compile, it provoked a
124 ph10 122 segmentation fault instead of returning the appropriate error message.
125 ph10 134
126 ph10 131 15. In multiline mode when the newline sequence was set to "any", the pattern
127 ph10 134 ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
128     This doesn't seem right; it now treats the CRLF combination as the line
129     ending, and so does not match in that case. It's only a pattern such as ^$
130     that would hit this one: something like ^ABC$ would have failed after \r
131     and then tried again after \r\n.
133 ph10 131 16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
134 ph10 134 in an attempt to make files that differ only in their line terminators
135     compare equal. This works on Linux.
136 ph10 142
137 ph10 141 17. Under certain error circumstances pcregrep might try to free random memory
138     as it exited. This is now fixed, thanks to valgrind.
139 ph10 142
140 ph10 141 19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
141 ph10 142 "abc\r\n\r\n", it found an unwanted second match after the second \r. This
142     was because its rules for how to advance for /g after matching an empty
143 ph10 143 string at the end of a line did not allow for this case. They now check for
144     it specially.
145 ph10 150
146     20. pcretest is supposed to handle patterns and data of any length, by
147     extending its buffers when necessary. It was getting this wrong when the
148 ph10 147 buffer for a data line had to be extended.
149 ph10 150
150 ph10 149 21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
151 ph10 150 CRLF as a newline sequence.
152 ph10 152
153 ph10 151 22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
154 ph10 152 out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
155     I have nevertheless tidied it up.
156 ph10 111
157 ph10 152 23. Added some casts to kill warnings from HP-UX ia64 compiler.
158 ph10 123
159 ph10 153 24. Added a man page for pcre-config.
160 ph10 152
161 ph10 153
162 nigel 93 Version 7.0 19-Dec-06
163     ---------------------
165     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
166     moving to gcc 4.1.1.
168     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
169     sys/time.h, which is documented as needed for this function. It doesn't
170     seem to matter on Linux, but it showed up on some releases of OS X.
172     3. It seems that there are systems where bytes whose values are greater than
173     127 match isprint() in the "C" locale. The "C" locale should be the
174     default when a C program starts up. In most systems, only ASCII printing
175     characters match isprint(). This difference caused the output from pcretest
176     to vary, making some of the tests fail. I have changed pcretest so that:
178     (a) When it is outputting text in the compiled version of a pattern, bytes
179     other than 32-126 are always shown as hex escapes.
181     (b) When it is outputting text that is a matched part of a subject string,
182     it does the same, unless a different locale has been set for the match
183     (using the /L modifier). In this case, it uses isprint() to decide.
185     4. Fixed a major bug that caused incorrect computation of the amount of memory
186     required for a compiled pattern when options that changed within the
187     pattern affected the logic of the preliminary scan that determines the
188     length. The relevant options are -x, and -i in UTF-8 mode. The result was
189     that the computed length was too small. The symptoms of this bug were
190     either the PCRE error "internal error: code overflow" from pcre_compile(),
191     or a glibc crash with a message such as "pcretest: free(): invalid next
192     size (fast)". Examples of patterns that provoked this bug (shown in
193     pcretest format) are:
195     /(?-x: )/x
196     /(?x)(?-x: \s*#\s*)/
197     /((?i)[\x{c0}])/8
198     /(?i:[\x{c0}])/8
200     HOWEVER: Change 17 below makes this fix obsolete as the memory computation
201     is now done differently.
203     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
204     wrapper classes; (b) implement a new function in the C++ scanner that is
205     more efficient than the old way of doing things because it avoids levels of
206     recursion in the regex matching; (c) add a paragraph to the documentation
207     for the FullMatch() function.
209     6. The escape sequence \n was being treated as whatever was defined as
210     "newline". Not only was this contrary to the documentation, which states
211     that \n is character 10 (hex 0A), but it also went horribly wrong when
212     "newline" was defined as CRLF. This has been fixed.
214     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
215     was being set to -1 for the "end of line" case (supposedly a value that no
216     character can have). Though this value is never used (the check for end of
217     line is "zero bytes in current character"), it caused compiler complaints.
218     I've changed it to 0xffffffff.
220     8. In pcre_version.c, the version string was being built by a sequence of
221     C macros that, in the event of PCRE_PRERELEASE being defined as an empty
222     string (as it is for production releases) called a macro with an empty
223     argument. The C standard says the result of this is undefined. The gcc
224     compiler treats it as an empty string (which was what was wanted) but it is
225     reported that Visual C gives an error. The source has been hacked around to
226     avoid this problem.
228     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
229     builds of pcretest, and changed the call to _setmode() to use _O_BINARY
230     instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
231     of them did).
233     10. Originally, pcretest opened its input and output without "b"; then I was
234     told that "b" was needed in some environments, so it was added for release
235     5.0 to both the input and output. (It makes no difference on Unix-like
236     systems.) Later I was told that it is wrong for the input on Windows. I've
237     now abstracted the modes into two macros, to make it easier to fiddle with
238     them, and removed "b" from the input mode under Windows.
240     11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
242     12. Added -help and --help to pcretest as an official way of being reminded
243     of the options.
245     13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
246     and pcrecpp.cc because they annoy compilers at high warning levels.
248     14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
250     15. Fixed an occurrence of == in configure.ac that should have been = (shell
251     scripts are not C programs :-) and which was not noticed because it works
252     on Linux.
254     16. pcretest is supposed to handle any length of pattern and data line (as one
255     line or as a continued sequence of lines) by extending its input buffer if
256     necessary. This feature was broken for very long pattern lines, leading to
257     a string of junk being passed to pcre_compile() if the pattern was longer
258     than about 50K.
260     17. I have done a major re-factoring of the way pcre_compile() computes the
261     amount of memory needed for a compiled pattern. Previously, there was code
262     that made a preliminary scan of the pattern in order to do this. That was
263     OK when PCRE was new, but as the facilities have expanded, it has become
264     harder and harder to keep it in step with the real compile phase, and there
265     have been a number of bugs (see for example, 4 above). I have now found a
266     cunning way of running the real compile function in a "fake" mode that
267     enables it to compute how much memory it would need, while actually only
268     ever using a few hundred bytes of working memory and without too many
269     tests of the mode. This should make future maintenance and development
270     easier. A side effect of this work is that the limit of 200 on the nesting
271     depth of parentheses has been removed (though this was never a serious
272     limitation, I suspect). However, there is a downside: pcre_compile() now
273     runs more slowly than before (30% or more, depending on the pattern). I
274     hope this isn't a big issue. There is no effect on runtime performance.
276     18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
277     newline (only possible for the last line of a file) and it was a
278     pattern that set a locale (followed by /Lsomething), pcretest crashed.
280     19. Added additional timing features to pcretest. (1) The -tm option now times
281     matching only, not compiling. (2) Both -t and -tm can be followed, as a
282     separate command line item, by a number that specifies the number of
283     repeats to use when timing. The default is 50000; this gives better
284     precision, but takes uncomfortably long for very large patterns.
286     20. Extended pcre_study() to be more clever in cases where a branch of a
287     subpattern has no definite first character. For example, (a*|b*)[cd] would
288     previously give no result from pcre_study(). Now it recognizes that the
289     first character must be a, b, c, or d.
291     21. There was an incorrect error "recursive call could loop indefinitely" if
292     a subpattern (or the entire pattern) that was being tested for matching an
293     empty string contained only one non-empty item after a nested subpattern.
294     For example, the pattern (?>\x{100}*)\d(?R) provoked this error
295     incorrectly, because the \d was being skipped in the check.
297     22. The pcretest program now has a new pattern option /B and a command line
298     option -b, which is equivalent to adding /B to every pattern. This causes
299     it to show the compiled bytecode, without the additional information that
300     -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
301     is the same as /B/I).
303     23. A new optimization is now able automatically to treat some sequences such
304     as a*b as a*+b. More specifically, if something simple (such as a character
305     or a simple class like \d) has an unlimited quantifier, and is followed by
306     something that cannot possibly match the quantified thing, the quantifier
307     is automatically "possessified".
309     24. A recursive reference to a subpattern whose number was greater than 39
310     went wrong under certain circumstances in UTF-8 mode. This bug could also
311     have affected the operation of pcre_study().
313     25. Realized that a little bit of performance could be had by replacing
314     (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
316     26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
318     27. Possessive quantifiers such as a++ were previously implemented by turning
319     them into atomic groups such as ($>a+). Now they have their own opcodes,
320     which improves performance. This includes the automatically created ones
321     from 23 above.
323     28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
324     lookahead was broken if it was not anchored. PCRE was mistakenly expecting
325     the first matched character to be a colon. This applied both to named and
326     numbered groups.
328     29. The ucpinternal.h header file was missing its idempotency #ifdef.
330     30. I was sent a "project" file called libpcre.a.dev which I understand makes
331     building PCRE on Windows easier, so I have included it in the distribution.
333     31. There is now a check in pcretest against a ridiculously large number being
334     returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
335     loop, the loop is abandoned.
337     32. Forward references to subpatterns in conditions such as (?(2)...) where
338     subpattern 2 is defined later cause pcre_compile() to search forwards in
339     the pattern for the relevant set of parentheses. This search went wrong
340     when there were unescaped parentheses in a character class, parentheses
341     escaped with \Q...\E, or parentheses in a #-comment in /x mode.
343     33. "Subroutine" calls and backreferences were previously restricted to
344     referencing subpatterns earlier in the regex. This restriction has now
345     been removed.
347     34. Added a number of extra features that are going to be in Perl 5.10. On the
348     whole, these are just syntactic alternatives for features that PCRE had
349     previously implemented using the Python syntax or my own invention. The
350     other formats are all retained for compatibility.
352     (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
353     as (?P<name>...). The new forms, as well as being in Perl 5.10, are
354     also .NET compatible.
356     (b) A recursion or subroutine call to a named group can now be defined as
357     (?&name) as well as (?P>name).
359     (c) A backreference to a named group can now be defined as \k<name> or
360     \k'name' as well as (?P=name). The new forms, as well as being in Perl
361     5.10, are also .NET compatible.
363     (d) A conditional reference to a named group can now use the syntax
364     (?(<name>) or (?('name') as well as (?(name).
366     (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
367     groups (named and numbered) that are never evaluated inline, but can be
368     called as "subroutines" from elsewhere. In effect, the DEFINE condition
369     is always false. There may be only one alternative in such a group.
371     (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
372     as the simple (?(R). The condition is true only if the most recent
373     recursion is that of the given number or name. It does not search out
374     through the entire recursion stack.
376     (g) The escape \gN or \g{N} has been added, where N is a positive or
377     negative number, specifying an absolute or relative reference.
379     35. Tidied to get rid of some further signed/unsigned compiler warnings and
380     some "unreachable code" warnings.
382     36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
383     things, this adds five new scripts.
385     37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
386     There were also incompatibilities regarding the handling of \Q..\E inside
387     character classes, for example with patterns like [\Qa\E-\Qz\E] where the
388     hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
390     38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
391     matches an empty string, and forcibly breaks the loop. There were bugs in
392     this code in non-simple cases. For a pattern such as ^(a()*)* matched
393     against aaaa the result was just "a" rather than "aaaa", for example. Two
394     separate and independent bugs (that affected different cases) have been
395     fixed.
397     39. Refactored the code to abolish the use of different opcodes for small
398     capturing bracket numbers. This is a tidy that I avoided doing when I
399     removed the limit on the number of capturing brackets for 3.5 back in 2001.
400     The new approach is not only tidier, it makes it possible to reduce the
401     memory needed to fix the previous bug (38).
403     40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
404     sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
405     processing dot, circumflex, or dollar metacharacters, or #-comments in /x
406     mode.
408     41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
409     report.
411     42. Applied patch, originally from Ari Pollak, modified by Google, to allow
412     copy construction and assignment in the C++ wrapper.
414     43. Updated pcregrep to support "--newline=any". In the process, I fixed a
415     couple of bugs that could have given wrong results in the "--newline=crlf"
416     case.
418     44. Added a number of casts and did some reorganization of signed/unsigned int
419     variables following suggestions from Dair Grant. Also renamed the variable
420     "this" as "item" because it is a C++ keyword.
422     45. Arranged for dftables to add
424     #include "pcre_internal.h"
426     to pcre_chartables.c because without it, gcc 4.x may remove the array
427     definition from the final binary if PCRE is built into a static library and
428     dead code stripping is activated.
430     46. For an unanchored pattern, if a match attempt fails at the start of a
431     newline sequence, and the newline setting is CRLF or ANY, and the next two
432     characters are CRLF, advance by two characters instead of one.
435 nigel 91 Version 6.7 04-Jul-06
436     ---------------------
438     1. In order to handle tests when input lines are enormously long, pcretest has
439     been re-factored so that it automatically extends its buffers when
440     necessary. The code is crude, but this _is_ just a test program. The
441     default size has been increased from 32K to 50K.
443     2. The code in pcre_study() was using the value of the re argument before
444     testing it for NULL. (Of course, in any sensible call of the function, it
445     won't be NULL.)
447     3. The memmove() emulation function in pcre_internal.h, which is used on
448     systems that lack both memmove() and bcopy() - that is, hardly ever -
449     was missing a "static" storage class specifier.
451     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
452     containing an extended class (one that cannot be represented by a bitmap
453     because it contains high-valued characters or Unicode property items, e.g.
454     [\pZ]). Almost always one would set UTF-8 mode when processing such a
455     pattern, but PCRE should not loop if you do not (it no longer does).
456     [Detail: two cases were found: (a) a repeated subpattern containing an
457     extended class; (b) a recursive reference to a subpattern that followed a
458     previous extended class. It wasn't skipping over the extended class
459     correctly when UTF-8 mode was not set.]
461     5. A negated single-character class was not being recognized as fixed-length
462     in lookbehind assertions such as (?<=[^f]), leading to an incorrect
463     compile error "lookbehind assertion is not fixed length".
465     6. The RunPerlTest auxiliary script was showing an unexpected difference
466     between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
467     write a Perl script that can interpret lines of an input file either as
468     byte characters or as UTF-8, which is what "perltest" was being required to
469     do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
470     can't do is switch easily at run time between having the "use utf8;" pragma
471     or not. In the end, I fudged it by using the RunPerlTest script to insert
472     "use utf8;" explicitly for the UTF-8 tests.
474     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
475     the end of the subject string, contrary to the documentation and to what
476     Perl does. This was true of both matching functions. Now it matches only at
477     the start of the subject and immediately after *internal* newlines.
479     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
480     a pointer to an int instead of a pointer to an unsigned long int. This
481     caused problems on 64-bit systems.
483     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
484     instance of the 'standard' template library not being so standard".
486     10. There was no check on the number of named subpatterns nor the maximum
487     length of a subpattern name. The product of these values is used to compute
488     the size of the memory block for a compiled pattern. By supplying a very
489     long subpattern name and a large number of named subpatterns, the size
490     computation could be caused to overflow. This is now prevented by limiting
491     the length of names to 32 characters, and the number of named subpatterns
492     to 10,000.
494     11. Subpatterns that are repeated with specific counts have to be replicated in
495     the compiled pattern. The size of memory for this was computed from the
496     length of the subpattern and the repeat count. The latter is limited to
497     65535, but there was no limit on the former, meaning that integer overflow
498     could in principle occur. The compiled length of a repeated subpattern is
499     now limited to 30,000 bytes in order to prevent this.
501     12. Added the optional facility to have named substrings with the same name.
503     13. Added the ability to use a named substring as a condition, using the
504     Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
505     are numbers (not recommended). Forward references are permitted.
507     14. Added forward references in named backreferences (if you see what I mean).
509     15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
510     pattern could run off the end of the subject. For example, the pattern
511     "(?s)(.{1,5})"8 did this with the subject "ab".
513     16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
514     PCRE_CASELESS was set when matching characters that were quantified with ?
515     or *.
517     17. A character class other than a single negated character that had a minimum
518     but no maximum quantifier - for example [ab]{6,} - was not handled
519     correctly by pce_dfa_exec(). It would match only one character.
521     18. A valid (though odd) pattern that looked like a POSIX character
522     class but used an invalid character after [ (for example [[,abc,]]) caused
523     pcre_compile() to give the error "Failed: internal error: code overflow" or
524     in some cases to crash with a glibc free() error. This could even happen if
525     the pattern terminated after [[ but there just happened to be a sequence of
526     letters, a binary zero, and a closing ] in the memory that followed.
528     19. Perl's treatment of octal escapes in the range \400 to \777 has changed
529     over the years. Originally (before any Unicode support), just the bottom 8
530     bits were taken. Thus, for example, \500 really meant \100. Nowadays the
531     output from "man perlunicode" includes this:
533     The regular expression compiler produces polymorphic opcodes. That
534     is, the pattern adapts to the data and automatically switches to
535     the Unicode character scheme when presented with Unicode data--or
536     instead uses a traditional byte scheme when presented with byte
537     data.
539     Sadly, a wide octal escape does not cause a switch, and in a string with
540     no other multibyte characters, these octal escapes are treated as before.
541     Thus, in Perl, the pattern /\500/ actually matches \100 but the pattern
542     /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
543     Unicode string.
545     I have not perpetrated such confusion in PCRE. Up till now, it took just
546     the bottom 8 bits, as in old Perl. I have now made octal escapes with
547     values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
548     translate to the appropriate multibyte character.
550     29. Applied some refactoring to reduce the number of warnings from Microsoft
551     and Borland compilers. This has included removing the fudge introduced
552     seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
553     a warning about an unused variable.
555     21. PCRE has not included VT (character 0x0b) in the set of whitespace
556     characters since release 4.0, because Perl (from release 5.004) does not.
557     [Or at least, is documented not to: some releases seem to be in conflict
558     with the documentation.] However, when a pattern was studied with
559     pcre_study() and all its branches started with \s, PCRE still included VT
560     as a possible starting character. Of course, this did no harm; it just
561     caused an unnecessary match attempt.
563     22. Removed a now-redundant internal flag bit that recorded the fact that case
564     dependency changed within the pattern. This was once needed for "required
565     byte" processing, but is no longer used. This recovers a now-scarce options
566     bit. Also moved the least significant internal flag bit to the most-
567     significant bit of the word, which was not previously used (hangover from
568     the days when it was an int rather than a uint) to free up another bit for
569     the future.
571     23. Added support for CRLF line endings as well as CR and LF. As well as the
572     default being selectable at build time, it can now be changed at runtime
573     via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
574     specify that it is scanning data with non-default line endings.
576     24. Changed the definition of CXXLINK to make it agree with the definition of
577     LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
579     25. Applied Ian Taylor's patches to avoid using another stack frame for tail
580     recursions. This makes a big different to stack usage for some patterns.
582     26. If a subpattern containing a named recursion or subroutine reference such
583     as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
584     the space required for the compiled pattern went wrong and gave too small a
585     value. Depending on the environment, this could lead to "Failed: internal
586     error: code overflow at offset 49" or "glibc detected double free or
587     corruption" errors.
589     27. Applied patches from Google (a) to support the new newline modes and (b) to
590     advance over multibyte UTF-8 characters in GlobalReplace.
592     28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
593     difference for some implementation of PCRE in some Windows version.
595     29. Added some extra testing facilities to pcretest:
597     \q<number> in a data line sets the "match limit" value
598     \Q<number> in a data line sets the "match recursion limt" value
599     -S <number> sets the stack size, where <number> is in megabytes
601     The -S option isn't available for Windows.
604 nigel 89 Version 6.6 06-Feb-06
605     ---------------------
607     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
608     in pcreposix.h. I have copied the definition from pcre.h.
610     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
611     because pcre.h is no longer a built file.
613     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
614     not normally included in the compiled code.
617 nigel 87 Version 6.5 01-Feb-06
618     ---------------------
620     1. When using the partial match feature with pcre_dfa_exec(), it was not
621     anchoring the second and subsequent partial matches at the new starting
622     point. This could lead to incorrect results. For example, with the pattern
623     /1234/, partially matching against "123" and then "a4" gave a match.
625     2. Changes to pcregrep:
627     (a) All non-match returns from pcre_exec() were being treated as failures
628     to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
629     error message is output. Some extra information is given for the
631     probably the only errors that are likely to be caused by users (by
632     specifying a regex that has nested indefinite repeats, for instance).
633     If there are more than 20 of these errors, pcregrep is abandoned.
635     (b) A binary zero was treated as data while matching, but terminated the
636     output line if it was written out. This has been fixed: binary zeroes
637     are now no different to any other data bytes.
639     (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
640     used to set a locale for matching. The --locale=xxxx long option has
641     been added (no short equivalent) to specify a locale explicitly on the
642     pcregrep command, overriding the environment variables.
644     (d) When -B was used with -n, some line numbers in the output were one less
645     than they should have been.
647     (e) Added the -o (--only-matching) option.
649     (f) If -A or -C was used with -c (count only), some lines of context were
650     accidentally printed for the final match.
652     (g) Added the -H (--with-filename) option.
654     (h) The combination of options -rh failed to suppress file names for files
655     that were found from directory arguments.
657     (i) Added the -D (--devices) and -d (--directories) options.
659     (j) Added the -F (--fixed-strings) option.
661     (k) Allow "-" to be used as a file name for -f as well as for a data file.
663     (l) Added the --colo(u)r option.
665     (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
666     is not present by default.
668     3. A nasty bug was discovered in the handling of recursive patterns, that is,
669     items such as (?R) or (?1), when the recursion could match a number of
670     alternatives. If it matched one of the alternatives, but subsequently,
671     outside the recursion, there was a failure, the code tried to back up into
672     the recursion. However, because of the way PCRE is implemented, this is not
673     possible, and the result was an incorrect result from the match.
675     In order to prevent this happening, the specification of recursion has
676     been changed so that all such subpatterns are automatically treated as
677     atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
679     4. I had overlooked the fact that, in some locales, there are characters for
680     which isalpha() is true but neither isupper() nor islower() are true. In
681     the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
682     and ordfeminine) are like this. This affected the treatment of \w and \W
683     when they appeared in character classes, but not when they appeared outside
684     a character class. The bit map for "word" characters is now created
685     separately from the results of isalnum() instead of just taking it from the
686     upper, lower, and digit maps. (Plus the underscore character, of course.)
688     5. The above bug also affected the handling of POSIX character classes such as
689     [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
690     permanent tables. Instead, the bit maps for such a class were previously
691     created as the appropriate unions of the upper, lower, and digit bitmaps.
692     Now they are created by subtraction from the [[:word:]] class, which has
693     its own bitmap.
695     6. The [[:blank:]] character class matches horizontal, but not vertical space.
696     It is created by subtracting the vertical space characters (\x09, \x0a,
697     \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
698     subtraction was done in the overall bitmap for a character class, meaning
699     that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
700     be recognized. This bug has been fixed.
702     7. Patches from the folks at Google:
704     (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
705     real life, but is still worth protecting against".
707     (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
708     regular expressions".
710     (c) pcre_scanner.cc: avoid use of std::count() because not all systems
711     have it.
713     (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
714     "configure" and the latter not, in order to fix a problem somebody had
715     with compiling the Arg class on HP-UX.
717     (e) Improve the error-handling of the C++ wrapper a little bit.
719     (f) New tests for checking recursion limiting.
721     8. The pcre_memmove() function, which is used only if the environment does not
722     have a standard memmove() function (and is therefore rarely compiled),
723     contained two bugs: (a) use of int instead of size_t, and (b) it was not
724     returning a result (though PCRE never actually uses the result).
726     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
727     large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
728     returned instead of calling malloc() with an overflowing number that would
729     most likely cause subsequent chaos.
731     10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
733     11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
734     with this option is matched, the nmatch and pmatch options of regexec() are
735     ignored.
737     12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
738     provided in case anyone wants to the the POSIX interface with UTF-8
739     strings.
741     13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
742     C++ linking (needed for some HP-UX environments).
744     14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
745     (unused parameter) and in the pcre_printint() function (omitted "default"
746     switch label when the default is to do nothing).
748     15. Added some code to make it possible, when PCRE is compiled as a C++
749     library, to replace subject pointers for pcre_exec() with a smart pointer
750     class, thus making it possible to process discontinuous strings.
752     16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
753     much the same function. They were added by different people who were trying
754     to make PCRE easy to compile on non-Unix systems. It has been suggested
755     that PCRE_EXPORT be abolished now that there is more automatic apparatus
756     for compiling on Windows systems. I have therefore replaced it with
757     PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
758     defaults to "extern" for C or "extern C" for C++, which works fine on
759     Unix-like systems. It is now possible to override the value of PCRE_DATA_
760     SCOPE with something explicit in config.h. In addition:
762     (a) pcreposix.h still had just "extern" instead of either of these macros;
763     I have replaced it with PCRE_DATA_SCOPE.
765     (b) Functions such as _pcre_xclass(), which are internal to the library,
766     but external in the C sense, all had PCRE_EXPORT in their definitions.
767     This is apparently wrong for the Windows case, so I have removed it.
768     (It makes no difference on Unix-like systems.)
770     17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
771     of recursive calls to match(). This is different to MATCH_LIMIT because
772     that limits the total number of calls to match(), not all of which increase
773     the depth of recursion. Limiting the recursion depth limits the amount of
774     stack (or heap if NO_RECURSE is set) that is used. The default can be set
775     when PCRE is compiled, and changed at run time. A patch from Google adds
776     this functionality to the C++ interface.
778     18. Changes to the handling of Unicode character properties:
780     (a) Updated the table to Unicode 4.1.0.
782     (b) Recognize characters that are not in the table as "Cn" (undefined).
784     (c) I revised the way the table is implemented to a much improved format
785     which includes recognition of ranges. It now supports the ranges that
786     are defined in UnicodeData.txt, and it also amalgamates other
787     characters into ranges. This has reduced the number of entries in the
788     table from around 16,000 to around 3,000, thus reducing its size
789     considerably. I realized I did not need to use a tree structure after
790     all - a binary chop search is just as efficient. Having reduced the
791     number of entries, I extended their size from 6 bytes to 8 bytes to
792     allow for more data.
794     (d) Added support for Unicode script names via properties such as \p{Han}.
796     19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
797     matching that character.
799     20. When matching a repeated Unicode property with a minimum greater than zero,
800     (for example \pL{2,}), PCRE could look past the end of the subject if it
801     reached it while seeking the minimum number of characters. This could
802     happen only if some of the characters were more than one byte long, because
803     there is a check for at least the minimum number of bytes.
805     21. Refactored the implementation of \p and \P so as to be more general, to
806     allow for more different types of property in future. This has changed the
807     compiled form incompatibly. Anybody with saved compiled patterns that use
808     \p or \P will have to recompile them.
810     22. Added "Any" and "L&" to the supported property types.
812     23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
813     but give a compile time error if the value is greater than 0xff.
815     24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
816     accidentally not being installed or uninstalled.
818     25. The pcre.h file was built from pcre.h.in, but the only changes that were
819     made were to insert the current release number. This seemed silly, because
820     it made things harder for people building PCRE on systems that don't run
821     "configure". I have turned pcre.h into a distributed file, no longer built
822     by "configure", with the version identification directly included. There is
823     no longer a pcre.h.in file.
825     However, this change necessitated a change to the pcre-config script as
826     well. It is built from pcre-config.in, and one of the substitutions was the
827     release number. I have updated configure.ac so that ./configure now finds
828     the release number by grepping pcre.h.
830     26. Added the ability to run the tests under valgrind.
833 nigel 85 Version 6.4 05-Sep-05
834     ---------------------
836     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
837     "--" to be printed when multiple files were scanned, even when none of the
838     -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
839     consider it to be a bug, and have restored the previous behaviour.
841     2. A couple of code tidies to get rid of compiler warnings.
843     3. The pcretest program used to cheat by referring to symbols in the library
844     whose names begin with _pcre_. These are internal symbols that are not
845     really supposed to be visible externally, and in some environments it is
846     possible to suppress them. The cheating is now confined to including
847     certain files from the library's source, which is a bit cleaner.
849     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
850     file's purpose clearer.
852     5. Reorganized pcre_ucp_findchar().
855 nigel 83 Version 6.3 15-Aug-05
856     ---------------------
858     1. The file libpcre.pc.in did not have general read permission in the tarball.
860     2. There were some problems when building without C++ support:
862     (a) If C++ support was not built, "make install" and "make test" still
863     tried to test it.
865     (b) There were problems when the value of CXX was explicitly set. Some
866     changes have been made to try to fix these, and ...
868     (c) --disable-cpp can now be used to explicitly disable C++ support.
870     (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
871     backslash in a target when C++ was disabled. This confuses some
872     versions of "make", apparently. Using an intermediate variable solves
873     this. (Same for CPP_LOBJ.)
875     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
876     (non-Windows) now includes $(CFLAGS) because these flags are sometimes
877     necessary on certain architectures.
879     4. Added a setting of -export-symbols-regex to the link command to remove
880     those symbols that are exported in the C sense, but actually are local
881     within the library, and not documented. Their names all begin with
882     "_pcre_". This is not a perfect job, because (a) we have to except some
883     symbols that pcretest ("illegally") uses, and (b) the facility isn't always
884     available (and never for static libraries). I have made a note to try to
885     find a way round (a) in the future.
888 nigel 81 Version 6.2 01-Aug-05
889     ---------------------
891     1. There was no test for integer overflow of quantifier values. A construction
892     such as {1111111111111111} would give undefined results. What is worse, if
893     a minimum quantifier for a parenthesized subpattern overflowed and became
894     negative, the calculation of the memory size went wrong. This could have
895     led to memory overwriting.
897     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
899     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
900     operating environments where this matters.
902     4. Applied Giuseppe Maxia's patch to add additional features for controlling
903     PCRE options from within the C++ wrapper.
905     5. Named capturing subpatterns were not being correctly counted when a pattern
906     was compiled. This caused two problems: (a) If there were more than 100
907     such subpatterns, the calculation of the memory needed for the whole
908     compiled pattern went wrong, leading to an overflow error. (b) Numerical
909     back references of the form \12, where the number was greater than 9, were
910     not recognized as back references, even though there were sufficient
911     previous subpatterns.
913     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
914     versions of gcc, e.g. 2.95.4.
917 nigel 79 Version 6.1 21-Jun-05
918     ---------------------
920     1. There was one reference to the variable "posix" in pcretest.c that was not
921     surrounded by "#if !defined NOPOSIX".
923     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
924     the cross-check on the old pcre_info() function, for the benefit of the
925     cut-down version of PCRE that is currently imported into Exim.
927     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
928     allocation error. I've done the easy fix, which wastes 2 bytes for sensible
929     patterns that start (?i) but I don't think that matters. The use of (?i) is
930     just an example; this all applies to the other options as well.
932     4. Since libtool seems to echo the compile commands it is issuing, the output
933     from "make" can be reduced a bit by putting "@" in front of each libtool
934     compile command.
936     5. Patch from the folks at Google for configure.in to be a bit more thorough
937     in checking for a suitable C++ installation before trying to compile the
938     C++ stuff. This should fix a reported problem when a compiler was present,
939     but no suitable headers.
941     6. The man pages all had just "PCRE" as their title. I have changed them to
942     be the relevant file name. I have also arranged that these names are
943     retained in the file doc/pcre.txt, which is a concatenation in text format
944     of all the man pages except the little individual ones for each function.
946     7. The NON-UNIX-USE file had not been updated for the different set of source
947     files that come with release 6. I also added a few comments about the C++
948     wrapper.
951 nigel 77 Version 6.0 07-Jun-05
952     ---------------------
954     1. Some minor internal re-organization to help with my DFA experiments.
956     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
957     didn't matter for the library itself when fully configured, but did matter
958     when compiling without UCP support, or within Exim, where the ucp files are
959     not imported.
961     3. Refactoring of the library code to split up the various functions into
962     different source modules. The addition of the new DFA matching code (see
963     below) to a single monolithic source would have made it really too
964     unwieldy, quite apart from causing all the code to be include in a
965     statically linked application, when only some functions are used. This is
966     relevant even without the DFA addition now that patterns can be compiled in
967     one application and matched in another.
969     The downside of splitting up is that there have to be some external
970     functions and data tables that are used internally in different modules of
971     the library but which are not part of the API. These have all had their
972     names changed to start with "_pcre_" so that they are unlikely to clash
973     with other external names.
975     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
976     a different (DFA) algorithm. Although it is slower than the original
977     function, it does have some advantages for certain types of matching
978     problem.
980     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
981     including restarting after a partial match.
983     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
984     defined when compiling for Windows was sent to me. I have put it into the
985     code, though I have no means of testing or verifying it.
987     7. Added the pcre_refcount() auxiliary function.
989     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
990     match before or at the first newline in the subject string. In pcretest,
991     the /f option on a pattern can be used to set this.
993     9. A repeated \w when used in UTF-8 mode with characters greater than 256
994     would behave wrongly. This has been present in PCRE since release 4.0.
996     10. A number of changes to the pcregrep command:
998     (a) Refactored how -x works; insert ^(...)$ instead of setting
999     PCRE_ANCHORED and checking the length, in preparation for adding
1000     something similar for -w.
1002     (b) Added the -w (match as a word) option.
1004     (c) Refactored the way lines are read and buffered so as to have more
1005     than one at a time available.
1007     (d) Implemented a pcregrep test script.
1009     (e) Added the -M (multiline match) option. This allows patterns to match
1010     over several lines of the subject. The buffering ensures that at least
1011     8K, or the rest of the document (whichever is the shorter) is available
1012     for matching (and similarly the previous 8K for lookbehind assertions).
1014     (f) Changed the --help output so that it now says
1016     -w, --word-regex(p)
1018     instead of two lines, one with "regex" and the other with "regexp"
1019     because that confused at least one person since the short forms are the
1020     same. (This required a bit of code, as the output is generated
1021     automatically from a table. It wasn't just a text change.)
1023     (g) -- can be used to terminate pcregrep options if the next thing isn't an
1024     option but starts with a hyphen. Could be a pattern or a path name
1025     starting with a hyphen, for instance.
1027     (h) "-" can be given as a file name to represent stdin.
1029     (i) When file names are being printed, "(standard input)" is used for
1030     the standard input, for compatibility with GNU grep. Previously
1031     "<stdin>" was used.
1033     (j) The option --label=xxx can be used to supply a name to be used for
1034     stdin when file names are being printed. There is no short form.
1036     (k) Re-factored the options decoding logic because we are going to add
1037     two more options that take data. Such options can now be given in four
1038     different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1040     (l) Added the -A, -B, and -C options for requesting that lines of context
1041     around matches be printed.
1043     (m) Added the -L option to print the names of files that do not contain
1044     any matching lines, that is, the complement of -l.
1046     (n) The return code is 2 if any file cannot be opened, but pcregrep does
1047     continue to scan other files.
1049     (o) The -s option was incorrectly implemented. For compatibility with other
1050     greps, it now suppresses the error message for a non-existent or non-
1051     accessible file (but not the return code). There is a new option called
1052     -q that suppresses the output of matching lines, which was what -s was
1053     previously doing.
1055     (p) Added --include and --exclude options to specify files for inclusion
1056     and exclusion when recursing.
1058     11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1059     Hopefully, it now does.
1061     12. Missing cast in pcre_study().
1063     13. Added an "uninstall" target to the makefile.
1065     14. Replaced "extern" in the function prototypes in Makefile.in with
1066     "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1067     world, but is set differently for Windows.
1069     15. Added a second compiling function called pcre_compile2(). The only
1070     difference is that it has an extra argument, which is a pointer to an
1071     integer error code. When there is a compile-time failure, this is set
1072     non-zero, in addition to the error test pointer being set to point to an
1073     error message. The new argument may be NULL if no error number is required
1074     (but then you may as well call pcre_compile(), which is now just a
1075     wrapper). This facility is provided because some applications need a
1076     numeric error indication, but it has also enabled me to tidy up the way
1077     compile-time errors are handled in the POSIX wrapper.
1079     16. Added VPATH=.libs to the makefile; this should help when building with one
1080     prefix path and installing with another. (Or so I'm told by someone who
1081     knows more about this stuff than I do.)
1083     17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1084     passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1085     match everything, including newlines. This is not POSIX-compatible, but
1086     somebody wanted the feature. From pcretest it can be activated by using
1087     both the P and the s flags.
1089     18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1091     19. libpcre.pc was being incorrectly installed as executable.
1093     20. A couple of places in pcretest check for end-of-line by looking for '\n';
1094     it now also looks for '\r' so that it will work unmodified on Windows.
1096     21. Added Google's contributed C++ wrapper to the distribution.
1098     22. Added some untidy missing memory free() calls in pcretest, to keep
1099     Electric Fence happy when testing.
1103 nigel 75 Version 5.0 13-Sep-04
1104     ---------------------
1106     1. Internal change: literal characters are no longer packed up into items
1107     containing multiple characters in a single byte-string. Each character
1108     is now matched using a separate opcode. However, there may be more than one
1109     byte in the character in UTF-8 mode.
1111     2. The pcre_callout_block structure has two new fields: pattern_position and
1112     next_item_length. These contain the offset in the pattern to the next match
1113     item, and its length, respectively.
1115     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1116     insertion of callouts before each pattern item. Added the /C option to
1117     pcretest to make use of this.
1119     4. On the advice of a Windows user, the lines
1121     #if defined(_WIN32) || defined(WIN32)
1122     _setmode( _fileno( stdout ), 0x8000 );
1123     #endif /* defined(_WIN32) || defined(WIN32) */
1125     have been added to the source of pcretest. This apparently does useful
1126     magic in relation to line terminators.
1128     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1129     for the benefit of those environments where the "b" makes a difference.
1131     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1132     to know about it. I have put a hack into configure.in that adds in code
1133     to set GCC=yes if CC=icc. This seems to end up at a point in the
1134     generated configure script that is early enough to affect the setting of
1135     compiler options, which is what is needed, but I have no means of testing
1136     whether it really works. (The user who reported this had patched the
1137     generated configure script, which of course I cannot do.)
1139     LATER: After change 22 below (new libtool files), the configure script
1140     seems to know about icc (and also ecc). Therefore, I have commented out
1141     this hack in configure.in.
1143     7. Added support for pkg-config (2 patches were sent in).
1145     8. Negated POSIX character classes that used a combination of internal tables
1146     were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1147     [[:^ascii]]. Typically, they would match almost any characters. The other
1148     POSIX classes were not broken in this way.
1150     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1151     to find the match, as PCRE was deluded into thinking that the match had to
1152     start at the start point or following a newline. The same bug applied to
1153     patterns with negative forward assertions or any backward assertions
1154     preceding ".*" at the start, unless the pattern required a fixed first
1155     character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1157     10. In UTF-8 mode, when moving forwards in the subject after a failed match
1158     starting at the last subject character, bytes beyond the end of the subject
1159     string were read.
1161     11. Renamed the variable "class" as "classbits" to make life easier for C++
1162     users. (Previously there was a macro definition, but it apparently wasn't
1163     enough.)
1165     12. Added the new field "tables" to the extra data so that tables can be passed
1166     in at exec time, or the internal tables can be re-selected. This allows
1167     a compiled regex to be saved and re-used at a later time by a different
1168     program that might have everything at different addresses.
1170     13. Modified the pcre-config script so that, when run on Solaris, it shows a
1171     -R library as well as a -L library.
1173     14. The debugging options of pcretest (-d on the command line or D on a
1174     pattern) showed incorrect output for anything following an extended class
1175     that contained multibyte characters and which was followed by a quantifier.
1177     15. Added optional support for general category Unicode character properties
1178     via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1179     support. It adds about 90K to the size of the library. The meanings of the
1180     inbuilt class escapes such as \d and \s have NOT been changed.
1182     16. Updated pcredemo.c to include calls to free() to release the memory for the
1183     compiled pattern.
1185     17. The generated file chartables.c was being created in the source directory
1186     instead of in the building directory. This caused the build to fail if the
1187     source directory was different from the building directory, and was
1188     read-only.
1190     18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1191     file. No doubt somebody will tell me if they don't make sense... Also added
1192     Dan Mooney's comments about building on OpenVMS.
1194     19. Added support for partial matching via the PCRE_PARTIAL option for
1195     pcre_exec() and the \P data escape in pcretest.
1197     20. Extended pcretest with 3 new pattern features:
1199     (i) A pattern option of the form ">rest-of-line" causes pcretest to
1200     write the compiled pattern to the file whose name is "rest-of-line".
1201     This is a straight binary dump of the data, with the saved pointer to
1202     the character tables forced to be NULL. The study data, if any, is
1203     written too. After writing, pcretest reads a new pattern.
1205     (ii) If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1206     compiled pattern from the given file. There must not be any
1207     occurrences of "<" in the file name (pretty unlikely); if there are,
1208     pcretest will instead treat the initial "<" as a pattern delimiter.
1209     After reading in the pattern, pcretest goes on to read data lines as
1210     usual.
1212     (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1213     and 16-bit fields in a compiled pattern, to simulate a pattern that
1214     was compiled on a host of opposite endianness.
1216     21. The pcre-exec() function can now cope with patterns that were compiled on
1217     hosts of opposite endianness, with this restriction:
1219     As for any compiled expression that is saved and used later, the tables
1220     pointer field cannot be preserved; the extra_data field in the arguments
1221     to pcre_exec() should be used to pass in a tables address if a value
1222     other than the default internal tables were used at compile time.
1224     22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1225     now diagnosed as an error. Previously, most of the time, a negative number
1226     would have been treated as zero, but if in addition "ovector" was passed as
1227     NULL, a crash could occur.
1229     23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1230     new versions from the libtool 1.5 distribution (the last one is a copy of
1231     a file called libtool.m4). This seems to have fixed the need to patch
1232     "configure" to support Darwin 1.3 (which I used to do). However, I still
1233     had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1234     workstation).
1236     24. Changed the PCRE licence to be the more standard "BSD" licence.
1239 nigel 73 Version 4.5 01-Dec-03
1240     ---------------------
1242     1. There has been some re-arrangement of the code for the match() function so
1243     that it can be compiled in a version that does not call itself recursively.
1244     Instead, it keeps those local variables that need separate instances for
1245     each "recursion" in a frame on the heap, and gets/frees frames whenever it
1246     needs to "recurse". Keeping track of where control must go is done by means
1247     of setjmp/longjmp. The whole thing is implemented by a set of macros that
1248     hide most of the details from the main code, and operates only if
1249     NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1250     "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1251     operating.
1253     To make it easier for callers to provide specially tailored get/free
1254     functions for this usage, two new functions, pcre_stack_malloc, and
1255     pcre_stack_free, are used. They are always called in strict stacking order,
1256     and the size of block requested is always the same.
1258     The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1259     PCRE has been compiled to use the stack or the heap for recursion. The
1260     -C option of pcretest uses this to show which version is compiled.
1262     A new data escape \S, is added to pcretest; it causes the amounts of store
1263     obtained and freed by both kinds of malloc/free at match time to be added
1264     to the output.
1266     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1267     what's available on my current Linux desktop machine.
1269     3. When matching a UTF-8 string, the test for a valid string at the start has
1270     been extended. If start_offset is not zero, PCRE now checks that it points
1271     to a byte that is the start of a UTF-8 character. If not, it returns
1272     PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1273     this is necessary because there may be backward assertions in the pattern.
1274     When matching the same subject several times, it may save resources to use
1275     PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1277     4. The code for checking the validity of UTF-8 strings has been tightened so
1278     that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1279     containing "overlong sequences".
1281     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1282     I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1283     should have been "&", but it just so happened that all the cases this let
1284     through by mistake were picked up later in the function.
1286     6. I had used a variable called "isblank" - this is a C99 function, causing
1287     some compilers to warn. To avoid this, I renamed it (as "blankclass").
1289     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1290     prompting; (b) run "./pcretest /dev/null" at the start of the test script
1291     so the version is shown; (c) stop "make test" echoing "./RunTest".
1293     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1295     9. The prototype for memmove() for systems that don't have it was using
1296     size_t, but the inclusion of the header that defines size_t was later. I've
1297     moved the #includes for the C headers earlier to avoid this.
1299     10. Added some adjustments to the code to make it easier to compiler on certain
1300     special systems:
1302     (a) Some "const" qualifiers were missing.
1303     (b) Added the macro EXPORT before all exported functions; by default this
1304     is defined to be empty.
1305     (c) Changed the dftables auxiliary program (that builds chartables.c) so
1306     that it reads its output file name as an argument instead of writing
1307     to the standard output and assuming this can be redirected.
1309     11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
1310     class containing characters with values greater than 255, PCRE compilation
1311     went into a loop.
1313     12. A recursive reference to a subpattern that was within another subpattern
1314     that had a minimum quantifier of zero caused PCRE to crash. For example,
1315     (x(y(?2))z)? provoked this bug with a subject that got as far as the
1316     recursion. If the recursively-called subpattern itself had a zero repeat,
1317     that was OK.
1319     13. In pcretest, the buffer for reading a data line was set at 30K, but the
1320     buffer into which it was copied (for escape processing) was still set at
1321     1024, so long lines caused crashes.
1323     14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
1324     "internal error: code overflow...". This applied to any character class
1325     that was followed by a possessive quantifier.
1327     15. Modified the Makefile to add libpcre.la as a prerequisite for
1328     libpcreposix.la because I was told this is needed for a parallel build to
1329     work.
1331     16. If a pattern that contained .* following optional items at the start was
1332     studied, the wrong optimizing data was generated, leading to matching
1333     errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
1334     matching string must start with a or b or c. The correct conclusion for
1335     this pattern is that a match can start with any character.
1338 nigel 71 Version 4.4 13-Aug-03
1339     ---------------------
1341     1. In UTF-8 mode, a character class containing characters with values between
1342     127 and 255 was not handled correctly if the compiled pattern was studied.
1343     In fixing this, I have also improved the studying algorithm for such
1344     classes (slightly).
1346     2. Three internal functions had redundant arguments passed to them. Removal
1347     might give a very teeny performance improvement.
1349     3. Documentation bug: the value of the capture_top field in a callout is *one
1350     more than* the number of the hightest numbered captured substring.
1352     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
1353     in incorrectly linking with a previously installed version. They now link
1354     explicitly with libpcre.la.
1356     5. configure.in no longer needs to recognize Cygwin specially.
1358     6. A problem in pcre.in for Windows platforms is fixed.
1360     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
1361     pcretest, it used to include the size of the study block as part of its
1362     output. Unfortunately, the structure contains a field that has a different
1363     size on different hardware architectures. This meant that the tests that
1364     showed this size failed. As the block is currently always of a fixed size,
1365     this information isn't actually particularly useful in pcretest output, so
1366     I have just removed it.
1368     8. Three pre-processor statements accidentally did not start in column 1.
1369     Sadly, there are *still* compilers around that complain, even though
1370     standard C has not required this for well over a decade. Sigh.
1372     9. In pcretest, the code for checking callouts passed small integers in the
1373     callout_data field, which is a void * field. However, some picky compilers
1374     complained about the casts involved for this on 64-bit systems. Now
1375     pcretest passes the address of the small integer instead, which should get
1376     rid of the warnings.
1378     10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
1379     both compile and run time, and gives an error if an invalid UTF-8 sequence
1380     is found. There is a option for disabling this check in cases where the
1381     string is known to be correct and/or the maximum performance is wanted.
1383     11. In response to a bug report, I changed one line in Makefile.in from
1385     -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
1386     to
1387     -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
1389     to look similar to other lines, but I have no way of telling whether this
1390     is the right thing to do, as I do not use Windows. No doubt I'll get told
1391     if it's wrong...
1394 nigel 69 Version 4.3 21-May-03
1395     ---------------------
1397     1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
1398     Makefile.
1400     2. Some refactoring to improve the quality of the code:
1402     (i) The utf8_table... variables are now declared "const".
1404     (ii) The code for \cx, which used the "case flipping" table to upper case
1405     lower case letters, now just substracts 32. This is ASCII-specific,
1406     but the whole concept of \cx is ASCII-specific, so it seems
1407     reasonable.
1409     (iii) PCRE was using its character types table to recognize decimal and
1410     hexadecimal digits in the pattern. This is silly, because it handles
1411     only 0-9, a-f, and A-F, but the character types table is locale-
1412     specific, which means strange things might happen. A private
1413     table is now used for this - though it costs 256 bytes, a table is
1414     much faster than multiple explicit tests. Of course, the standard
1415     character types table is still used for matching digits in subject
1416     strings against \d.
1418     (iv) Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
1419     ending in _t are). So I've renamed it as ESC_tee.
1421     3. The first argument for regexec() in the POSIX wrapper should have been
1422     defined as "const".
1424     4. Changed pcretest to use malloc() for its buffers so that they can be
1425     Electric Fenced for debugging.
1427     5. There were several places in the code where, in UTF-8 mode, PCRE would try
1428     to read one or more bytes before the start of the subject string. Often this
1429     had no effect on PCRE's behaviour, but in some circumstances it could
1430     provoke a segmentation fault.
1432     6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
1433     to try to read one or more bytes before the start of the subject string.
1435     7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
1436     UTF-8 support could misbehave in various ways if the subject string
1437     contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
1438     area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
1439     back over UTF-8 characters.)
1442 nigel 67 Version 4.2 14-Apr-03
1443     ---------------------
1445     1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
1447     2. Changes to the building process, supplied by Ronald Landheer-Cieslak
1448     [ON_WINDOWS]: new variable, "#" on non-Windows platforms
1449     [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
1450     [WIN_PREFIX]: new variable, "cyg" for Cygwin
1451     * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
1452     and BUILD_EXEEXT
1453     Note: automatic setting of the BUILD variables is not yet working
1454     set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
1455     compile-time but not at link-time
1456     [LINK]: use for linking executables only
1457     make different versions for Windows and non-Windows
1458     [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
1459     libraries
1460     [LINK_FOR_BUILD]: new variable
1461     [OBJEXT]: use throughout
1462     [EXEEXT]: use throughout
1463     <winshared>: new target
1464     <wininstall>: new target
1465     <dftables.o>: use native compiler
1466     <dftables>: use native linker
1467     <install>: handle Windows platform correctly
1468     <clean>: ditto
1469     <check>: ditto
1470     copy DLL to top builddir before testing
1472     As part of these changes, -no-undefined was removed again. This was reported
1473     to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
1474     in any case.
1476     3. Some tidies to get rid of compiler warnings:
1478     . In the match_data structure, match_limit was an unsigned long int, whereas
1479     match_call_count was an int. I've made them both unsigned long ints.
1481     . In pcretest the fact that a const uschar * doesn't automatically cast to
1482     a void * provoked a warning.
1484     . Turning on some more compiler warnings threw up some "shadow" variables
1485     and a few more missing casts.
1487     4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
1488     option, a class that contained a single character with a value between 128
1489     and 255 (e.g. /[\xFF]/) caused PCRE to crash.
1491     5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
1492     option, a class that contained several characters, but with at least one
1493     whose value was between 128 and 255 caused PCRE to crash.
1496 nigel 65 Version 4.1 12-Mar-03
1497     ---------------------
1498 nigel 63
1499 nigel 65 1. Compiling with gcc -pedantic found a couple of places where casts were
1500     needed, and a string in dftables.c that was longer than standard compilers are
1501     required to support.
1503     2. Compiling with Sun's compiler found a few more places where the code could
1504     be tidied up in order to avoid warnings.
1506     3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
1507     first of these names is deprecated in the latest Autoconf in favour of the name
1508     CC_FOR_BUILD, because "host" is typically used to mean the system on which the
1509     compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
1510     analogy I have changed it to CFLAGS_FOR_BUILD.
1512     4. Added -no-undefined to the linking command in the Makefile, because this is
1513     apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
1514     linking step for the pcreposix library.
1516     5. PCRE was failing to diagnose the case of two named groups with the same
1517     name.
1519     6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
1520     literal character that is needed in the subject for a match, and scans along to
1521     ensure that it is present before embarking on the full matching process. This
1522     saves time in cases of nested unlimited repeats that are never going to match.
1523     Problem: the scan can take a lot of time if the subject is very long (e.g.
1524     megabytes), thus penalizing straightforward matches. It is now done only if the
1525     amount of subject to be scanned is less than 1000 bytes.
1527     7. A lesser problem with the same optimization is that it was recording the
1528     first character of an anchored pattern as "needed", thus provoking a search
1529     right along the subject, even when the first match of the pattern was going to
1530     fail. The "needed" character is now not set for anchored patterns, unless it
1531     follows something in the pattern that is of non-fixed length. Thus, it still
1532     fulfils its original purpose of finding quick non-matches in cases of nested
1533     unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
1536     Version 4.0 17-Feb-03
1537     ---------------------
1539 nigel 63 1. If a comment in an extended regex that started immediately after a meta-item
1540     extended to the end of string, PCRE compiled incorrect data. This could lead to
1541     all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
1543     2. Moved to autoconf 2.53 and libtool 1.4.2.
1545     3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
1546     the special perltest8 script is no longer needed - all the tests can be run
1547     from a single perltest script.
1549     4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
1550     by \s. It has now been removed in PCRE. This means it isn't recognized as
1551     whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
1552     class [:space:] *does* include VT, thereby creating a mess.
1554     5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
1555     space and tab.
1557     6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
1558     its new features into the main test script, reducing the number of scripts.
1560     7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
1561     were backward compatible, and made the (?i) apply to the whole pattern, as if
1562     /i were given. Now it behaves more logically, and applies the option setting
1563     only to what follows. PCRE has been changed to follow suit. However, if it
1564     finds options settings right at the start of the pattern, it extracts them into
1565     the global options, as before. Thus, they show up in the info data.
1567     8. Added support for the \Q...\E escape sequence. Characters in between are
1568     treated as literals. This is slightly different from Perl in that $ and @ are
1569     also handled as literals inside the quotes. In Perl, they will cause variable
1570     interpolation. Note the following examples:
1572     Pattern PCRE matches Perl matches
1574     \Qabc$xyz\E abc$xyz abc followed by the contents of $xyz
1575     \Qabc\$xyz\E abc\$xyz abc\$xyz
1576     \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
1578     For compatibility with Perl, \Q...\E sequences are recognized inside character
1579     classes as well as outside them.
1581     9. Re-organized 3 code statements in pcretest to avoid "overflow in
1582     floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
1583     (size_t) cast to one statement in pcretest and one in pcreposix to avoid
1584     signed/unsigned warnings.
1586     10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
1587     option for pcretest, so I've replaced it by a simple function that does just
1588     that job.
1590     11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
1591     "pcregrep -".
1593     12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
1594     Java package. This provides some syntactic sugar for simple cases of what my
1595     documentation calls "once-only subpatterns". A pattern such as x*+ is the same
1596     as (?>x*). In other words, if what is inside (?>...) is just a single repeated
1597     item, you can use this simplified notation. Note that only makes sense with
1598     greedy quantifiers. Consequently, the use of the possessive quantifier forces
1599     greediness, whatever the setting of the PCRE_UNGREEDY option.
1601     13. A change of greediness default within a pattern was not taking effect at
1602     the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
1603     subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
1604     was abstracted outside.
1606     14. PCRE now supports the \G assertion. It is true when the current matching
1607     position is at the start point of the match. This differs from \A when the
1608     starting offset is non-zero. Used with the /g option of pcretest (or similar
1609     code), it works in the same way as it does for Perl's /g option. If all
1610     alternatives of a regex begin with \G, the expression is anchored to the start
1611     match position, and the "anchored" flag is set in the compiled expression.
1613     15. Some bugs concerning the handling of certain option changes within patterns
1614     have been fixed. These applied to options other than (?ims). For example,
1615     "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
1616     been the other way round. Some of this was related to change 7 above.
1618     16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
1619     features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
1620     and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
1621     POSIX classes only within a class (e.g. /[[:alpha:]]/).
1623     17. Added support for Perl's \C escape. This matches one byte, even in UTF8
1624     mode. Unlike ".", it always matches newline, whatever the setting of
1625     PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
1626     assertions. Perl allows it, but it doesn't (in general) work because it can't
1627     calculate the length of the lookbehind. At least, that's the case for Perl
1628     5.8.0 - I've been told they are going to document that it doesn't work in
1629     future.
1631     18. Added an error diagnosis for escapes that PCRE does not support: these are
1632     \L, \l, \N, \P, \p, \U, \u, and \X.
1634     19. Although correctly diagnosing a missing ']' in a character class, PCRE was
1635     reading past the end of the pattern in cases such as /[abcd/.
1637     20. PCRE was getting more memory than necessary for patterns with classes that
1638     contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
1640     21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
1641     compiling PCRE for use with Virtual Pascal.
1643     22. Small fix to the Makefile to make it work properly if the build is done
1644     outside the source tree.
1646     23. Added a new extension: a condition to go with recursion. If a conditional
1647     subpattern starts with (?(R) the "true" branch is used if recursion has
1648     happened, whereas the "false" branch is used only at the top level.
1650     24. When there was a very long string of literal characters (over 255 bytes
1651     without UTF support, over 250 bytes with UTF support), the computation of how
1652     much memory was required could be incorrect, leading to segfaults or other
1653     strange effects.
1655     25. PCRE was incorrectly assuming anchoring (either to start of subject or to
1656     start of line for a non-DOTALL pattern) when a pattern started with (.*) and
1657     there was a subsequent back reference to those brackets. This meant that, for
1658     example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
1659     possible to check for precisely this case. All we can do is abandon the
1660     optimization if .* occurs inside capturing brackets when there are any back
1661     references whatsoever. (See below for a better fix that came later.)
1663     26. The handling of the optimization for finding the first character of a
1664     non-anchored pattern, and for finding a character that is required later in the
1665     match were failing in some cases. This didn't break the matching; it just
1666     failed to optimize when it could. The way this is done has been re-implemented.
1668     27. Fixed typo in error message for invalid (?R item (it said "(?p").
1670     28. Added a new feature that provides some of the functionality that Perl
1671     provides with (?{...}). The facility is termed a "callout". The way it is done
1672     in PCRE is for the caller to provide an optional function, by setting
1673     pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
1674     global variable. By default it is unset, which disables all calling out. To get
1675     the function called, the regex must include (?C) at appropriate points. This
1676     is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
1677     This provides a means of identifying different callout points. When PCRE
1678     reaches such a point in the regex, if pcre_callout has been set, the external
1679     function is called. It is provided with data in a structure called
1680     pcre_callout_block, which is defined in pcre.h. If the function returns 0,
1681     matching continues; if it returns a non-zero value, the match at the current
1682     point fails. However, backtracking will occur if possible. [This was changed
1683     later and other features added - see item 49 below.]
1685     29. pcretest is upgraded to test the callout functionality. It provides a
1686     callout function that displays information. By default, it shows the start of
1687     the match and the current position in the text. There are some new data escapes
1688     to vary what happens:
1690     \C+ in addition, show current contents of captured substrings
1691     \C- do not supply a callout function
1692     \C!n return 1 when callout number n is reached
1693     \C!n!m return 1 when callout number n is reached for the mth time
1695     30. If pcregrep was called with the -l option and just a single file name, it
1696     output "<stdin>" if a match was found, instead of the file name.
1698     31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
1699     slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
1700     pcre_exec(). This saves a malloc/free per call. The default value of
1701     POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
1702     when configuring.
1704     32. The default maximum size of a compiled pattern is 64K. There have been a
1705     few cases of people hitting this limit. The code now uses macros to handle the
1706     storing of links as offsets within the compiled pattern. It defaults to 2-byte
1707     links, but this can be changed to 3 or 4 bytes by --with-link-size when
1708     configuring. Tests 2 and 5 work only with 2-byte links because they output
1709     debugging information about compiled patterns.
1711     33. Internal code re-arrangements:
1713     (a) Moved the debugging function for printing out a compiled regex into
1714     its own source file (printint.c) and used #include to pull it into
1715     pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
1716     separate copies.
1718     (b) Defined the list of op-code names for debugging as a macro in
1719     internal.h so that it is next to the definition of the opcodes.
1721     (c) Defined a table of op-code lengths for simpler skipping along compiled
1722     code. This is again a macro in internal.h so that it is next to the
1723     definition of the opcodes.
1725     34. Added support for recursive calls to individual subpatterns, along the
1726     lines of Robin Houston's patch (but implemented somewhat differently).
1728     35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
1729     allow it to read and process whole directories in Win32. This code was
1730     contributed by Lionel Fourquaux; it has not been tested by me.
1732     36. Added support for named subpatterns. The Python syntax (?P<name>...) is
1733     used to name a group. Names consist of alphanumerics and underscores, and must
1734     be unique. Back references use the syntax (?P=name) and recursive calls use
1735     (?P>name) which is a PCRE extension to the Python extension. Groups still have
1736     numbers. The function pcre_fullinfo() can be used after compilation to extract
1737     a name/number map. There are three relevant calls:
1739     PCRE_INFO_NAMEENTRYSIZE yields the size of each entry in the map
1740     PCRE_INFO_NAMECOUNT yields the number of entries
1741     PCRE_INFO_NAMETABLE yields a pointer to the map.
1743     The map is a vector of fixed-size entries. The size of each entry depends on
1744     the length of the longest name used. The first two bytes of each entry are the
1745     group number, most significant byte first. There follows the corresponding
1746     name, zero terminated. The names are in alphabetical order.
1748     37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
1749     case instead of 255. Making it the same both with and without UTF-8 support
1750     means that the same test output works with both.
1752     38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
1753     calling malloc() with a zero argument.
1755     39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
1756     optimization. I've improved things by keeping a bitmap of backreferences with
1757     numbers 1-31 so that if .* occurs inside capturing brackets that are not in
1758     fact referenced, the optimization can be applied. It is unlikely that a
1759     relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
1760     the match to follow \n) will appear inside brackets with a number greater than
1761     31, but if it does, any back reference > 31 suppresses the optimization.
1763     40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
1764     of disabling numbered capturing parentheses. Any opening parenthesis that is
1765     not followed by ? behaves as if it were followed by ?: but named parentheses
1766     can still be used for capturing (and they will acquire numbers in the usual
1767     way).
1769     41. Redesigned the return codes from the match() function into yes/no/error so
1770     that errors can be passed back from deep inside the nested calls. A malloc
1771     failure while inside a recursive subpattern call now causes the
1772     PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
1774     42. It is now possible to set a limit on the number of times the match()
1775     function is called in a call to pcre_exec(). This facility makes it possible to
1776     limit the amount of recursion and backtracking, though not in a directly
1777     obvious way, because the match() function is used in a number of different
1778     circumstances. The count starts from zero for each position in the subject
1779     string (for non-anchored patterns). The default limit is, for compatibility, a
1780     large number, namely 10 000 000. You can change this in two ways:
1782     (a) When configuring PCRE before making, you can use --with-match-limit=n
1783     to set a default value for the compiled library.
1785     (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
1786     a different value is set. See 45 below.
1788     If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
1790     43. Added a new function pcre_config(int, void *) to enable run-time extraction
1791     of things that can be changed at compile time. The first argument specifies
1792     what is wanted and the second points to where the information is to be placed.
1793     The current list of available information is:
1797     The output is an integer that is set to one if UTF-8 support is available;
1798     otherwise it is set to zero.
1802     The output is an integer that it set to the value of the code that is used for
1803     newline. It is either LF (10) or CR (13).
1807     The output is an integer that contains the number of bytes used for internal
1808     linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
1812     The output is an integer that contains the threshold above which the POSIX
1813     interface uses malloc() for output vectors. See item 31 above.
1817     The output is an unsigned integer that contains the default limit of the number
1818     of match() calls in a pcre_exec() execution. See 42 above.
1820     44. pcretest has been upgraded by the addition of the -C option. This causes it
1821     to extract all the available output from the new pcre_config() function, and to
1822     output it. The program then exits immediately.
1824     45. A need has arisen to pass over additional data with calls to pcre_exec() in
1825     order to support additional features. One way would have been to define
1826     pcre_exec2() (for example) with extra arguments, but this would not have been
1827     extensible, and would also have required all calls to the original function to
1828     be mapped to the new one. Instead, I have chosen to extend the mechanism that
1829     is used for passing in "extra" data from pcre_study().
1831     The pcre_extra structure is now exposed and defined in pcre.h. It currently
1832     contains the following fields:
1834     flags a bitmap indicating which of the following fields are set
1835     study_data opaque data from pcre_study()
1836     match_limit a way of specifying a limit on match() calls for a specific
1837     call to pcre_exec()
1838     callout_data data for callouts (see 49 below)
1840     The flag bits are also defined in pcre.h, and are
1846     The pcre_study() function now returns one of these new pcre_extra blocks, with
1847     the actual study data pointed to by the study_data field, and the
1848     PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
1849     before. That is, this change is entirely upwards-compatible and requires no
1850     change to existing code.
1852     If you want to pass in additional data to pcre_exec(), you can either place it
1853     in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
1854     block.
1856     46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
1857     data string contains the escape sequence \M, pcretest calls pcre_exec() several
1858     times with different match limits, until it finds the minimum value needed for
1859     pcre_exec() to complete. The value is then output. This can be instructive; for
1860     most simple matches the number is quite small, but for pathological cases it
1861     gets very large very quickly.
1863     47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
1864     returns the size of the data block pointed to by the study_data field in a
1865     pcre_extra block, that is, the value that was passed as the argument to
1866     pcre_malloc() when PCRE was getting memory in which to place the information
1867     created by pcre_study(). The fourth argument should point to a size_t variable.
1868     pcretest has been extended so that this information is shown after a successful
1869     pcre_study() call when information about the compiled regex is being displayed.
1871     48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
1872     because what follows is always an absolute path. (Later: it turns out that this
1873     is more than cosmetic for MinGW, because it doesn't like empty path
1874     components.)
1876     49. Some changes have been made to the callout feature (see 28 above):
1878     (i) A callout function now has three choices for what it returns:
1880     0 => success, carry on matching
1881     > 0 => failure at this point, but backtrack if possible
1882     < 0 => serious error, return this value from pcre_exec()
1884     Negative values should normally be chosen from the set of PCRE_ERROR_xxx
1885     values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
1886     "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
1887     use by callout functions. It will never be used by PCRE itself.
1889     (ii) The pcre_extra structure (see 45 above) has a void * field called
1890     callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
1891     pcre_callout_block structure has a field of the same name. The contents of
1892     the field passed in the pcre_extra structure are passed to the callout
1893     function in the corresponding field in the callout block. This makes it
1894     easier to use the same callout-containing regex from multiple threads. For
1895     testing, the pcretest program has a new data escape
1897     \C*n pass the number n (may be negative) as callout_data
1899     If the callout function in pcretest receives a non-zero value as
1900     callout_data, it returns that value.
1902     50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
1903     there were some redundant $(CFLAGS) in commands that are now specified as
1904     $(LINK), which already includes $(CFLAGS).
1906     51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
1907     has been compiled with UTF-8 support *and* pcre_compile() has been compiled
1908     with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
1909     one-byte characters throughout. Note that case-insensitive matching applies
1910     only to characters whose values are less than 256. PCRE doesn't support the
1911     notion of cases for higher-valued characters.
1913     (i) A character class whose characters are all within 0-255 is handled as
1914     a bit map, and the map is inverted for negative classes. Previously, a
1915     character > 255 always failed to match such a class; however it should
1916     match if the class was a negative one (e.g. [^ab]). This has been fixed.
1918     (ii) A negated character class with a single character < 255 is coded as
1919     "not this character" (OP_NOT). This wasn't working properly when the test
1920     character was multibyte, either singly or repeated.
1922     (iii) Repeats of multibyte characters are now handled correctly in UTF-8
1923     mode, for example: \x{100}{2,3}.
1925     (iv) The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
1926     singly or repeated) now correctly test multibyte characters. However,
1927     PCRE doesn't recognize any characters with values greater than 255 as
1928     digits, spaces, or word characters. Such characters always match \D, \S,
1929     and \W, and never match \d, \s, or \w.
1931     (v) Classes may now contain characters and character ranges with values
1932     greater than 255. For example: [ab\x{100}-\x{400}].
1934     (vi) pcregrep now has a --utf-8 option (synonym -u) which makes it call
1935     PCRE in UTF-8 mode.
1937     52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
1938     PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
1939     retained for backwards compatibility. (Note that LASTLITERAL is also a byte
1940     value.)
1942     53. The single man page has become too large. I have therefore split it up into
1943     a number of separate man pages. These also give rise to individual HTML pages;
1944     these are now put in a separate directory, and there is an index.html page that
1945     lists them all. Some hyperlinking between the pages has been installed.
1947     54. Added convenience functions for handling named capturing parentheses.
1949     55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
1950     aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
1951     true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
1952     are faulted.
1954     56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
1955     calling configure. These values are used when compiling the dftables.c program
1956     which is run to generate the source of the default character tables. They
1957     default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
1958     you will need to set these values.
1960     57. Updated the building process for Windows DLL, as provided by Fred Cox.
1963     Version 3.9 02-Jan-02
1964 nigel 61 ---------------------
1966     1. A bit of extraneous text had somehow crept into the pcregrep documentation.
1968     2. If --disable-static was given, the building process failed when trying to
1969     build pcretest and pcregrep. (For some reason it was using libtool to compile
1970     them, which is not right, as they aren't part of the library.)
1973 nigel 59 Version 3.8 18-Dec-01
1974     ---------------------
1976     1. The experimental UTF-8 code was completely screwed up. It was packing the
1977     bytes in the wrong order. How dumb can you get?
1980 nigel 57 Version 3.7 29-Oct-01
1981     ---------------------
1983     1. In updating pcretest to check change 1 of version 3.6, I screwed up.
1984     This caused pcretest, when used on the test data, to segfault. Unfortunately,
1985     this didn't happen under Solaris 8, where I normally test things.
1987 nigel 59 2. The Makefile had to be changed to make it work on BSD systems, where 'make'
1988     doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
1989     isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
1990     this fix an hour or so after the initial 3.7 release.)
1991 nigel 57
1992 nigel 59
1993 nigel 55 Version 3.6 23-Oct-01
1994     ---------------------
1996     1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
1997     offsets passed as NULL with zero offset count.
1999     2. The config.guess and config.sub files had not been updated when I moved to
2000     the latest autoconf.
2003 nigel 53 Version 3.5 15-Aug-01
2004     ---------------------
2005 nigel 5
2006 nigel 53 1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2007     had been forgotten.
2009     2. By using declared but undefined structures, we can avoid using "void"
2010     definitions in pcre.h while keeping the internal definitions of the structures
2011     private.
2013     3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2014     user point of view, this means that both static and shared libraries are built
2015     by default, but this can be individually controlled. More of the work of
2016     handling this static/shared cases is now inside libtool instead of PCRE's make
2017     file.
2019     4. The pcretest utility is now installed along with pcregrep because it is
2020     useful for users (to test regexs) and by doing this, it automatically gets
2021     relinked by libtool. The documentation has been turned into a man page, so
2022     there are now .1, .txt, and .html versions in /doc.
2024     5. Upgrades to pcregrep:
2025     (i) Added long-form option names like gnu grep.
2026     (ii) Added --help to list all options with an explanatory phrase.
2027     (iii) Added -r, --recursive to recurse into sub-directories.
2028     (iv) Added -f, --file to read patterns from a file.
2030     6. pcre_exec() was referring to its "code" argument before testing that
2031     argument for NULL (and giving an error if it was NULL).
2033     7. Upgraded Makefile.in to allow for compiling in a different directory from
2034     the source directory.
2036     8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2037     options bits, the pointer it was passed was to an int instead of to an unsigned
2038     long int. This mattered only on 64-bit systems.
2040     9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2041     generated) instead of pcre.in, which it its source. Also made the same change
2042     in several of the .c files.
2044     10. A new release of gcc defines printf() as a macro, which broke pcretest
2045     because it had an ifdef in the middle of a string argument for printf(). Fixed
2046     by using separate calls to printf().
2048     11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2049     script, to force use of CR or LF instead of \n in the source. On non-Unix
2050     systems, the value can be set in config.h.
2052     12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2053     absolute limit. Changed the text of the error message to make this clear, and
2054     likewise updated the man page.
2056     13. The limit of 99 on the number of capturing subpatterns has been removed.
2057     The new limit is 65535, which I hope will not be a "real" limit.
2060 nigel 51 Version 3.4 22-Aug-00
2061     ---------------------
2063     1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2065     2. Diagnose condition (?(0) as an error instead of crashing on matching.
2068 nigel 49 Version 3.3 01-Aug-00
2069     ---------------------
2071     1. If an octal character was given, but the value was greater than \377, it
2072     was not getting masked to the least significant bits, as documented. This could
2073     lead to crashes in some systems.
2075     2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
2076     the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
2078     3. Added the functions pcre_free_substring() and pcre_free_substring_list().
2079     These just pass their arguments on to (pcre_free)(), but they are provided
2080     because some uses of PCRE bind it to non-C systems that can call its functions,
2081     but cannot call free() or pcre_free() directly.
2083     4. Add "make test" as a synonym for "make check". Corrected some comments in
2084     the Makefile.
2086     5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
2087     Makefile.
2089     6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
2090     command called pgrep for grepping around the active processes.
2092     7. Added the beginnings of support for UTF-8 character strings.
2094     8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
2095     RANLIB to ./ltconfig so that they are used by libtool. I think these are all
2096     the relevant ones. (AR is not passed because ./ltconfig does its own figuring
2097     out for the ar command.)
2100 nigel 47 Version 3.2 12-May-00
2101     ---------------------
2103     This is purely a bug fixing release.
2105     1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
2106     of ZA. This was just one example of several cases that could provoke this bug,
2107     which was introduced by change 9 of version 2.00. The code for breaking
2108     infinite loops after an iteration that matches an empty string was't working
2109     correctly.
2111     2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
2112     when matched against abbab (for example). After matching an empty string, it
2113     wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
2114     caused it to match further down the string than it should.
2116     3. The code contained an inclusion of sys/types.h. It isn't clear why this
2117     was there because it doesn't seem to be needed, and it causes trouble on some
2118     systems, as it is not a Standard C header. It has been removed.
2120     4. Made 4 silly changes to the source to avoid stupid compiler warnings that
2121     were reported on the Macintosh. The changes were from
2123     while ((c = *(++ptr)) != 0 && c != '\n');
2124     to
2125     while ((c = *(++ptr)) != 0 && c != '\n') ;
2127     Totally extraordinary, but if that's what it takes...
2129     5. PCRE is being used in one environment where neither memmove() nor bcopy() is
2130     available. Added HAVE_BCOPY and an autoconf test for it; if neither
2131     HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
2132     assumes the way PCRE uses memmove() (always moving upwards).
2134     6. PCRE is being used in one environment where strchr() is not available. There
2135     was only one use in pcre.c, and writing it out to avoid strchr() probably gives
2136     faster code anyway.
2139 nigel 45 Version 3.1 09-Feb-00
2140     ---------------------
2142     The only change in this release is the fixing of some bugs in Makefile.in for
2143     the "install" target:
2145     (1) It was failing to install pcreposix.h.
2147     (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
2150 nigel 43 Version 3.0 01-Feb-00
2151     ---------------------
2152 nigel 41
2153     1. Add support for the /+ modifier to perltest (to output $` like it does in
2154     pcretest).
2156     2. Add support for the /g modifier to perltest.
2158     3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
2159     matches null strings.
2161     4. Fix perltest so that it doesn't do unwanted things when fed an empty
2162     pattern. Perl treats empty patterns specially - it reuses the most recent
2163     pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
2164     effect.
2166     5. The POSIX interface was broken in that it was just handing over the POSIX
2167     captured string vector to pcre_exec(), but (since release 2.00) PCRE has
2168     required a bigger vector, with some working space on the end. This means that
2169     the POSIX wrapper now has to get and free some memory, and copy the results.
2171 nigel 43 6. Added some simple autoconf support, placing the test data and the
2172     documentation in separate directories, re-organizing some of the
2173     information files, and making it build pcre-config (a GNU standard). Also added
2174     libtool support for building PCRE as a shared library, which is now the
2175     default.
2176 nigel 41
2177 nigel 43 7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
2178     09 are not valid octal constants. Single digits will be used for minor values
2179     less than 10.
2181     8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
2182     existing programs that set these in the POSIX interface can use PCRE without
2183     modification.
2185     9. Added a new function, pcre_fullinfo() with an extensible interface. It can
2186     return all that pcre_info() returns, plus additional data. The pcre_info()
2187     function is retained for compatibility, but is considered to be obsolete.
2189     10. Added experimental recursion feature (?R) to handle one common case that
2190     Perl 5.6 will be able to do with (?p{...}).
2192     11. Added support for POSIX character classes like [:alpha:], which Perl is
2193     adopting.
2196 nigel 39 Version 2.08 31-Aug-99
2197     ----------------------
2199     1. When startoffset was not zero and the pattern began with ".*", PCRE was not
2200     trying to match at the startoffset position, but instead was moving forward to
2201     the next newline as if a previous match had failed.
2203     2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
2204     and could get into a loop if a null string was matched other than at the start
2205     of the subject.
2207     3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
2208     be distinguished at compile time, and for completeness also added PCRE_DATE.
2210     5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
2211     in GnuWin32 environments.
2214 nigel 37 Version 2.07 29-Jul-99
2215     ----------------------
2217     1. The documentation is now supplied in plain text form and HTML as well as in
2218     the form of man page sources.
2220     2. C++ compilers don't like assigning (void *) values to other pointer types.
2221     In particular this affects malloc(). Although there is no problem in Standard
2222     C, I've put in casts to keep C++ compilers happy.
2224     3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
2225     should be (const char *).
2227     4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
2228     be useful for non-Unix systems who don't want to bother with the POSIX stuff.
2229     However, I haven't made this a standard facility. The documentation doesn't
2230     mention it, and the Makefile doesn't support it.
2232     5. The Makefile now contains an "install" target, with editable destinations at
2233     the top of the file. The pcretest program is not installed.
2235     6. pgrep -V now gives the PCRE version number and date.
2237     7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
2238     causing the entire string to be ignored, instead of just the last character.
2240     8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
2241     non-matching string, it can take a very, very long time, even for strings of
2242     quite modest length, because of the nested recursion. PCRE now does better in
2243     some of these cases. It does this by remembering the last required literal
2244     character in the pattern, and pre-searching the subject to ensure it is present
2245     before running the real match. In other words, it applies a heuristic to detect
2246     some types of certain failure quickly, and in the above example, if presented
2247     with a string that has no trailing " it gives "no match" very quickly.
2249     9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
2250     other alternatives are tried instead.
2253 nigel 35 Version 2.06 09-Jun-99
2254     ----------------------
2256     1. Change pcretest's output for amount of store used to show just the code
2257     space, because the remainder (the data block) varies in size between 32-bit and
2258     64-bit systems.
2260     2. Added an extra argument to pcre_exec() to supply an offset in the subject to
2261     start matching at. This allows lookbehinds to work when searching for multiple
2262     occurrences in a string.
2264     3. Added additional options to pcretest for testing multiple occurrences:
2266     /+ outputs the rest of the string that follows a match
2267     /g loops for multiple occurrences, using the new startoffset argument
2268     /G loops for multiple occurrences by passing an incremented pointer
2270     4. PCRE wasn't doing the "first character" optimization for patterns starting
2271     with \b or \B, though it was doing it for other lookbehind assertions. That is,
2272     it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
2273     the letter 'x'. On long subject strings, this gives a significant speed-up.
2276 nigel 33 Version 2.05 21-Apr-99
2277     ----------------------
2279     1. Changed the type of magic_number from int to long int so that it works
2280     properly on 16-bit systems.
2282     2. Fixed a bug which caused patterns starting with .* not to work correctly
2283     when the subject string contained newline characters. PCRE was assuming
2284     anchoring for such patterns in all cases, which is not correct because .* will
2285     not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
2286     DOTALL is set at top level; otherwise it knows that patterns starting with .*
2287     must be retried after every newline in the subject.
2290 nigel 31 Version 2.04 18-Feb-99
2291     ----------------------
2293     1. For parenthesized subpatterns with repeats whose minimum was zero, the
2294     computation of the store needed to hold the pattern was incorrect (too large).
2295     If such patterns were nested a few deep, this could multiply and become a real
2296     problem.
2298     2. Added /M option to pcretest to show the memory requirement of a specific
2299     pattern. Made -m a synonym of -s (which does this globally) for compatibility.
2301     3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
2302     compiled in such a way that the backtracking after subsequent failure was
2303     pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
2304     ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
2307 nigel 29 Version 2.03 02-Feb-99
2308     ----------------------
2310     1. Fixed typo and small mistake in man page.
2312 nigel 31 2. Added 4th condition (GPL supersedes if conflict) and created separate
2313     LICENCE file containing the conditions.
2314 nigel 29
2315     3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
2316     Perl, that is the internal \ allows the delimiter to be included in the
2317     pattern. Locked out the use of \ as a delimiter. If \ immediately follows
2318     the final delimiter, add \ to the end of the pattern (to test the error).
2320     4. Added the convenience functions for extracting substrings after a successful
2321     match. Updated pcretest to make it able to test these functions.
2324 nigel 27 Version 2.02 14-Jan-99
2325     ----------------------
2327     1. Initialized the working variables associated with each extraction so that
2328     their saving and restoring doesn't refer to uninitialized store.
2330     2. Put dummy code into study.c in order to trick the optimizer of the IBM C
2331     compiler for OS/2 into generating correct code. Apparently IBM isn't going to
2332     fix the problem.
2334     3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
2335     calls, and wasn't printing the correct value for compiling calls. Increased the
2336     default value of LOOPREPEAT, and the number of significant figures in the
2337     times.
2339     4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
2341     5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
2342     a building problem on Windows NT with a FAT file system.
2345 nigel 25 Version 2.01 21-Oct-98
2346     ----------------------
2348     1. Changed the API for pcre_compile() to allow for the provision of a pointer
2349     to character tables built by pcre_maketables() in the current locale. If NULL
2350     is passed, the default tables are used.
2353 nigel 23 Version 2.00 24-Sep-98
2354 nigel 21 ----------------------
2356 nigel 23 1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
2357     it any more.
2359     2. Allow quantification of (?>) groups, and make it work correctly.
2361     3. The first character computation wasn't working for (?>) groups.
2363     4. Correct the implementation of \Z (it is permitted to match on the \n at the
2364     end of the subject) and add 5.005's \z, which really does match only at the
2365     very end of the subject.
2367     5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
2369     6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
2370     DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
2371     localized options. All options to pcre_study() were also removed.
2373     7. Add other new features from 5.005:
2375     $(?<= positive lookbehind
2376     $(?<! negative lookbehind
2377     (?imsx-imsx) added the unsetting capability
2378     such a setting is global if at outer level; local otherwise
2379     (?imsx-imsx:) non-capturing groups with option setting
2380     (?(cond)re|re) conditional pattern matching
2382     A backreference to itself in a repeated group matches the previous
2383     captured string.
2385     8. General tidying up of studying (both automatic and via "study")
2386     consequential on the addition of new assertions.
2388     9. As in 5.005, unlimited repeated groups that could match an empty substring
2389     are no longer faulted at compile time. Instead, the loop is forcibly broken at
2390     runtime if any iteration does actually match an empty substring.
2392     10. Include the RunTest script in the distribution.
2394     11. Added tests from the Perl 5.005_02 distribution. This showed up a few
2395     discrepancies, some of which were old and were also with respect to 5.004. They
2396     have now been fixed.
2399     Version 1.09 28-Apr-98
2400     ----------------------
2402 nigel 21 1. A negated single character class followed by a quantifier with a minimum
2403     value of one (e.g. [^x]{1,6} ) was not compiled correctly. This could lead to
2404     program crashes, or just wrong answers. This did not apply to negated classes
2405     containing more than one character, or to minima other than one.
2408 nigel 19 Version 1.08 27-Mar-98
2409     ----------------------
2411     1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
2413     2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
2414     latter must appear before anything that relies on it in the pattern.
2417 nigel 17 Version 1.07 16-Feb-98
2418     ----------------------
2420     1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
2421     repeat of a potentially empty string).
2424 nigel 15 Version 1.06 23-Jan-98
2425     ----------------------
2427     1. Added Markus Oberhumer's little patches for C++.
2429     2. Literal strings longer than 255 characters were broken.
2432 nigel 13 Version 1.05 23-Dec-97
2433     ----------------------
2435     1. Negated character classes containing more than one character were failing if
2436     PCRE_CASELESS was set at run time.
2439 nigel 11 Version 1.04 19-Dec-97
2440     ----------------------
2442     1. Corrected the man page, where some "const" qualifiers had been omitted.
2444     2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
2445     input syntax.
2447     3. Fixed memory leak which occurred when a regex with back references was
2448     matched with an offsets vector that wasn't big enough. The temporary memory
2449     that is used in this case wasn't being freed if the match failed.
2451     4. Tidied pcretest to ensure it frees memory that it gets.
2453     5. Temporary memory was being obtained in the case where the passed offsets
2454     vector was exactly big enough.
2456     6. Corrected definition of offsetof() from change 5 below.
2458     7. I had screwed up change 6 below and broken the rules for the use of
2459     setjmp(). Now fixed.
2462 nigel 9 Version 1.03 18-Dec-97
2463     ----------------------
2465     1. A erroneous regex with a missing opening parenthesis was correctly
2466     diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
2467     on some systems.
2469     2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
2470     it was reported that one broken compiler failed on the former because "code" is
2471     also an independent variable.
2473     3. The erroneous regex a[]b caused an array overrun reference.
2475     4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
2476     fail on data ending with that character. (It was going on too far, and checking
2477     the next character, typically a binary zero.) This was specific to the
2478     optimized code for single-character negative classes.
2480     5. Added a contributed patch from the TIN world which does the following:
2482     + Add an undef for memmove, in case the the system defines a macro for it.
2484     + Add a definition of offsetof(), in case there isn't one. (I don't know
2485     the reason behind this - offsetof() is part of the ANSI standard - but
2486     it does no harm).
2488     + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
2489     most of the places where whitespace preceded '#'. I have given up and
2490     allowed the remaining 2 cases to be at the margin.
2492     + Rename some variables in pcre to eliminate shadowing. This seems very
2493     pedantic, but does no harm, of course.
2495     6. Moved the call to setjmp() into its own function, to get rid of warnings
2496     from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
2498     7. Constructs such as \d{8,} were compiling into the equivalent of
2499 nigel 11 \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
2500 nigel 9 outcome, but in this particular case used more store than had been allocated,
2501     which caused the bug to be discovered because it threw up an internal error.
2503     8. The debugging code in both pcre and pcretest for outputting the compiled
2504     form of a regex was going wrong in the case of back references followed by
2505     curly-bracketed repeats.
2508 nigel 7 Version 1.02 12-Dec-97
2509     ----------------------
2511     1. Typos in pcre.3 and comments in the source fixed.
2513     2. Applied a contributed patch to get rid of places where it used to remove
2514     'const' from variables, and fixed some signed/unsigned and uninitialized
2515     variable warnings.
2517     3. Added the "runtest" target to Makefile.
2519     4. Set default compiler flag to -O2 rather than just -O.
2522 nigel 5 Version 1.01 19-Nov-97
2523     ----------------------
2525     1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
2526     like /([ab]*)*/, that is, for classes with more than one character in them.
2528     2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
2529     as /((?>a*))*/ (a PCRE_EXTRA facility).
2532     Version 1.00 18-Nov-97
2533     ----------------------
2535     1. Added compile-time macros to support systems such as SunOS4 which don't have
2536     memmove() or strerror() but have other things that can be used instead.
2538     2. Arranged that "make clean" removes the executables.
2541 nigel 3 Version 0.99 27-Oct-97
2542     ----------------------
2544     1. Fixed bug in code for optimizing classes with only one character. It was
2545     initializing a 32-byte map regardless, which could cause it to run off the end
2546     of the memory it had got.
2548     2. Added, conditional on PCRE_EXTRA, the proposed (?>REGEX) construction.
2551     Version 0.98 22-Oct-97
2552     ----------------------
2554     1. Fixed bug in code for handling temporary memory usage when there are more
2555     back references than supplied space in the ovector. This could cause segfaults.
2558     Version 0.97 21-Oct-97
2559     ----------------------
2561     1. Added the \X "cut" facility, conditional on PCRE_EXTRA.
2563     2. Optimized negated single characters not to use a bit map.
2565     3. Brought error texts together as macro definitions; clarified some of them;
2566     fixed one that was wrong - it said "range out of order" when it meant "invalid
2567     escape sequence".
2569     4. Changed some char * arguments to const char *.
2571     5. Added PCRE_NOTBOL and PCRE_NOTEOL (from POSIX).
2573     6. Added the POSIX-style API wrapper in pcreposix.a and testing facilities in
2574     pcretest.
2577     Version 0.96 16-Oct-97
2578     ----------------------
2580     1. Added a simple "pgrep" utility to the distribution.
2582     2. Fixed an incompatibility with Perl: "{" is now treated as a normal character
2583     unless it appears in one of the precise forms "{ddd}", "{ddd,}", or "{ddd,ddd}"
2584     where "ddd" means "one or more decimal digits".
2586     3. Fixed serious bug. If a pattern had a back reference, but the call to
2587     pcre_exec() didn't supply a large enough ovector to record the related
2588     identifying subpattern, the match always failed. PCRE now remembers the number
2589     of the largest back reference, and gets some temporary memory in which to save
2590     the offsets during matching if necessary, in order to ensure that
2591     backreferences always work.
2593     4. Increased the compatibility with Perl in a number of ways:
2595     (a) . no longer matches \n by default; an option PCRE_DOTALL is provided
2596     to request this handling. The option can be set at compile or exec time.
2598     (b) $ matches before a terminating newline by default; an option
2599     PCRE_DOLLAR_ENDONLY is provided to override this (but not in multiline
2600     mode). The option can be set at compile or exec time.
2602     (c) The handling of \ followed by a digit other than 0 is now supposed to be
2603     the same as Perl's. If the decimal number it represents is less than 10
2604     or there aren't that many previous left capturing parentheses, an octal
2605     escape is read. Inside a character class, it's always an octal escape,
2606     even if it is a single digit.
2608     (d) An escaped but undefined alphabetic character is taken as a literal,
2609     unless PCRE_EXTRA is set. Currently this just reserves the remaining
2610     escapes.
2612     (e) {0} is now permitted. (The previous item is removed from the compiled
2613     pattern).
2615     5. Changed all the names of code files so that the basic parts are no longer
2616     than 10 characters, and abolished the teeny "globals.c" file.
2618     6. Changed the handling of character classes; they are now done with a 32-byte
2619     bit map always.
2621     7. Added the -d and /D options to pcretest to make it possible to look at the
2622     internals of compilation without having to recompile pcre.
2625     Version 0.95 23-Sep-97
2626     ----------------------
2628     1. Fixed bug in pre-pass concerning escaped "normal" characters such as \x5c or
2629     \x20 at the start of a run of normal characters. These were being treated as
2630     real characters, instead of the source characters being re-checked.
2633     Version 0.94 18-Sep-97
2634     ----------------------
2636     1. The functions are now thread-safe, with the caveat that the global variables
2637     containing pointers to malloc() and free() or alternative functions are the
2638     same for all threads.
2640     2. Get pcre_study() to generate a bitmap of initial characters for non-
2641     anchored patterns when this is possible, and use it if passed to pcre_exec().
2644     Version 0.93 15-Sep-97
2645     ----------------------
2647     1. /(b)|(:+)/ was computing an incorrect first character.
2649     2. Add pcre_study() to the API and the passing of pcre_extra to pcre_exec(),
2650     but not actually doing anything yet.
2652     3. Treat "-" characters in classes that cannot be part of ranges as literals,
2653     as Perl does (e.g. [-az] or [az-]).
2655     4. Set the anchored flag if a branch starts with .* or .*? because that tests
2656     all possible positions.
2658     5. Split up into different modules to avoid including unneeded functions in a
2659     compiled binary. However, compile and exec are still in one module. The "study"
2660     function is split off.
2662     6. The character tables are now in a separate module whose source is generated
2663     by an auxiliary program - but can then be edited by hand if required. There are
2664     now no calls to isalnum(), isspace(), isdigit(), isxdigit(), tolower() or
2665     toupper() in the code.
2667     7. Turn the malloc/free funtions variables into pcre_malloc and pcre_free and
2668     make them global. Abolish the function for setting them, as the caller can now
2669     set them directly.
2672     Version 0.92 11-Sep-97
2673     ----------------------
2675     1. A repeat with a fixed maximum and a minimum of 1 for an ordinary character
2676     (e.g. /a{1,3}/) was broken (I mis-optimized it).
2678     2. Caseless matching was not working in character classes if the characters in
2679     the pattern were in upper case.
2681     3. Make ranges like [W-c] work in the same way as Perl for caseless matching.
2683     4. Make PCRE_ANCHORED public and accept as a compile option.
2685     5. Add an options word to pcre_exec() and accept PCRE_ANCHORED and
2686     PCRE_CASELESS at run time. Add escapes \A and \I to pcretest to cause it to
2687     pass them.
2689     6. Give an error if bad option bits passed at compile or run time.
2691     7. Add PCRE_MULTILINE at compile and exec time, and (?m) as well. Add \M to
2692     pcretest to cause it to pass that flag.
2694     8. Add pcre_info(), to get the number of identifying subpatterns, the stored
2695     options, and the first character, if set.
2697     9. Recognize C+ or C{n,m} where n >= 1 as providing a fixed starting character.
2700     Version 0.91 10-Sep-97
2701     ----------------------
2703     1. PCRE was failing to diagnose unlimited repeats of subpatterns that could
2704     match the empty string as in /(a*)*/. It was looping and ultimately crashing.
2706     2. PCRE was looping on encountering an indefinitely repeated back reference to
2707     a subpattern that had matched an empty string, e.g. /(a|)\1*/. It now does what
2708     Perl does - treats the match as successful.
2710     ****


Name Value
svn:eol-style native
svn:keywords "Author Date Id Revision Url"

ViewVC Help
Powered by ViewVC 1.1.12