ViewVC logotype

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log

Revision 172 - (hide annotations) (download)
Tue Jun 5 10:40:13 2007 UTC (7 years, 10 months ago) by ph10
File size: 129235 byte(s)
Drastically reduce workspace used for alternatives in groups; also some 
trailing space removals for a test release.

1 nigel 3 ChangeLog for PCRE
2     ------------------
4 ph10 172 Version 7.2 05-June-07
5 ph10 157 ---------------------
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.
9 ph10 159
10     2. Re-jig the pcregrep tests with different newline settings in an attempt
11     to make them independent of the local environment's newline setting.
12 ph10 157
13 ph10 160 3. Add code to configure.ac to remove -g from the CFLAGS default settings.
14 ph10 165
15 ph10 161 4. Some of the "internals" tests were previously cut out when the link size
16     was not 2, because the output contained actual offsets. The recent new
17 ph10 165 "Z" feature of pcretest means that these can be cut out, making the tests
18     usable with all link sizes.
20 ph10 164 5. Implemented Stan Switzer's goto replacement for longjmp() when not using
21     stack recursion. This gives a massive performance boost under BSD, but just
22 ph10 165 a small improvement under Linux. However, it saves one field in the frame
23 ph10 164 in all cases.
24 ph10 172
25 ph10 166 6. Added more features from the forthcoming Perl 5.10:
26 ph10 172
27 ph10 166 (a) (?-n) (where n is a string of digits) is a relative subroutine or
28     recursion call. It refers to the nth most recently opened parentheses.
29 ph10 172
30 ph10 166 (b) (?+n) is also a relative subroutine call; it refers to the nth next
31 ph10 172 to be opened parentheses.
33     (c) Conditions that refer to capturing parentheses can be specified
34 ph10 167 relatively, for example, (?(-2)... or (?(+3)...
35 ph10 172
36 ph10 168 (d) \K resets the start of the current match so that everything before
37 ph10 172 is not part of it.
39 ph10 171 (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
40 ph10 172
41 ph10 171 (f) \g{name} is another synonym - part of Perl 5.10's unification of
42 ph10 172 reference syntax.
44     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
47     8. A pattern such as (.*(.)?)* caused pcre_exec() to fail by either not
48     terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
49 ph10 170 for detecting groups that can match an empty string.
50 ph10 159
51 ph10 172 9. A pattern with a very large number of alternatives (more than several
52     hundred) was running out of internal workspace during the pre-compile
53     phase, where pcre_compile() figures out how much memory will be needed. A
54     bit of new cunning has reduced the workspace needed for groups with
55     alternatives. The 1000-alternative test pattern now uses 12 bytes of
56     workspace instead of running out of the 4096 that are available.
59 ph10 155 Version 7.1 24-Apr-07
60 ph10 98 ---------------------
62 ph10 111 1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
63 ph10 122 that is more "standard", making use of automake and other Autotools. There
64 ph10 99 is some re-arrangement of the files and adjustment of comments consequent
65     on this.
66 ph10 111
67     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
68     for recursive directory scanning broke on some systems because the files
69     are not scanned in any specific order and on different systems the order
70     was different. A call to "sort" has been inserted into RunGrepTest for the
71     approprate test as a short-term fix. In the longer term there may be an
72 ph10 100 alternative.
73 ph10 111
74 ph10 100 3. I had an email from Eric Raymond about problems translating some of PCRE's
75 ph10 111 man pages to HTML (despite the fact that I distribute HTML pages, some
76     people do their own conversions for various reasons). The problems
77     concerned the use of low-level troff macros .br and .in. I have therefore
78     removed all such uses from the man pages (some were redundant, some could
79 ph10 113 be replaced by .nf/.fi pairs). The 132html script that I use to generate
80     HTML has been updated to handle .nf/.fi and to complain if it encounters
81     .br or .in.
82 ph10 111
83 ph10 100 4. Updated comments in configure.ac that get placed in config.h.in and also
84 ph10 123 arranged for config.h to be included in the distribution, with the name
85 ph10 111 config.h.generic, for the benefit of those who have to compile without
86     Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
88     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
89     Weber: (1) pcre_internal.h was missing some function renames; (2) updated
90 ph10 127 makevp.bat for the current PCRE, using the additional files
91 ph10 135 makevp_c.txt, makevp_l.txt, and pcregexp.pas.
92 ph10 111
93     6. A Windows user reported a minor discrepancy with test 2, which turned out
94     to be caused by a trailing space on an input line that had got lost in his
95 ph10 102 copy. The trailing space was an accident, so I've just removed it.
96 ph10 111
97 ph10 104 7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
98 ph10 111 that is needed.
100 ph10 105 8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
101     as "const" (a) because they are and (b) because it helps the PHP
102     maintainers who have recently made a script to detect big data structures
103 ph10 111 in the php code that should be moved to the .rodata section. I remembered
104     to update Builducptable as well, so it won't revert if ucptable.h is ever
105 ph10 105 re-created.
106 ph10 111
107     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
108 ph10 107 pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
109 ph10 111 order to be able to cut out the UTF-8 tables in the latter when UTF-8
110     support is not required. This saves 1.5-2K of code, which is important in
111 ph10 107 some applications.
112 ph10 111
113 ph10 110 Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
114 ph10 111 so as not to refer to the tables, even though these functions will never be
115     called when UTF-8 support is disabled. Otherwise there are problems with a
116     shared library.
118 ph10 118 10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
120     (a) It was defining its arguments as char * instead of void *.
122 ph10 123 (b) It was assuming that all moves were upwards in memory; this was true
123     a long time ago when I wrote it, but is no longer the case.
125 ph10 118 The emulated memove() is provided for those environments that have neither
126 ph10 123 memmove() nor bcopy(). I didn't think anyone used it these days, but that
127 ph10 118 is clearly not the case, as these two bugs were recently reported.
128 ph10 123
129 ph10 111 11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
130 ph10 123 and Detrail to create the HTML documentation, the .txt form of the man
131     pages, and it removes trailing spaces from listed files. It also creates
132     pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
133     case, it wraps all the #defines with #ifndefs. This script should be run
134 ph10 111 before "make dist".
135 ph10 123
136 ph10 115 12. Fixed two fairly obscure bugs concerned with quantified caseless matching
137     with Unicode property support.
138 ph10 123
139     (a) For a maximizing quantifier, if the two different cases of the
140     character were of different lengths in their UTF-8 codings (there are
141     some cases like this - I found 11), and the matching function had to
142 ph10 115 back up over a mixture of the two cases, it incorrectly assumed they
143     were both the same length.
144 ph10 123
145     (b) When PCRE was configured to use the heap rather than the stack for
146     recursion during matching, it was not correctly preserving the data for
147     the other case of a UTF-8 character when checking ahead for a match
148     while processing a minimizing repeat. If the check also involved
149 ph10 115 matching a wide character, but failed, corruption could cause an
150     erroneous result when trying to check for a repeat of the original
151     character.
152 ph10 123
153 ph10 116 13. Some tidying changes to the testing mechanism:
154 ph10 98
155 ph10 116 (a) The RunTest script now detects the internal link size and whether there
156     is UTF-8 and UCP support by running ./pcretest -C instead of relying on
157 ph10 123 values substituted by "configure". (The RunGrepTest script already did
158     this for UTF-8.) The configure.ac script no longer substitutes the
159     relevant variables.
161 ph10 116 (b) The debugging options /B and /D in pcretest show the compiled bytecode
162     with length and offset values. This means that the output is different
163     for different internal link sizes. Test 2 is skipped for link sizes
164     other than 2 because of this, bypassing the problem. Unfortunately,
165     there was also a test in test 3 (the locale tests) that used /B and
166 ph10 123 failed for link sizes other than 2. Rather than cut the whole test out,
167     I have added a new /Z option to pcretest that replaces the length and
168     offset values with spaces. This is now used to make test 3 independent
169 ph10 122 of link size. (Test 2 will be tidied up later.)
170 ph10 123
171     14. If erroroffset was passed as NULL to pcre_compile, it provoked a
172 ph10 122 segmentation fault instead of returning the appropriate error message.
173 ph10 134
174 ph10 131 15. In multiline mode when the newline sequence was set to "any", the pattern
175 ph10 134 ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
176     This doesn't seem right; it now treats the CRLF combination as the line
177     ending, and so does not match in that case. It's only a pattern such as ^$
178     that would hit this one: something like ^ABC$ would have failed after \r
179     and then tried again after \r\n.
181 ph10 131 16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
182 ph10 134 in an attempt to make files that differ only in their line terminators
183     compare equal. This works on Linux.
184 ph10 142
185 ph10 141 17. Under certain error circumstances pcregrep might try to free random memory
186     as it exited. This is now fixed, thanks to valgrind.
187 ph10 142
188 ph10 141 19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
189 ph10 142 "abc\r\n\r\n", it found an unwanted second match after the second \r. This
190     was because its rules for how to advance for /g after matching an empty
191 ph10 143 string at the end of a line did not allow for this case. They now check for
192     it specially.
193 ph10 150
194     20. pcretest is supposed to handle patterns and data of any length, by
195     extending its buffers when necessary. It was getting this wrong when the
196 ph10 147 buffer for a data line had to be extended.
197 ph10 150
198 ph10 149 21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
199 ph10 150 CRLF as a newline sequence.
200 ph10 152
201 ph10 151 22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
202 ph10 152 out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
203     I have nevertheless tidied it up.
204 ph10 111
205 ph10 152 23. Added some casts to kill warnings from HP-UX ia64 compiler.
206 ph10 123
207 ph10 153 24. Added a man page for pcre-config.
208 ph10 152
209 ph10 153
210 nigel 93 Version 7.0 19-Dec-06
211     ---------------------
213     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
214     moving to gcc 4.1.1.
216     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
217     sys/time.h, which is documented as needed for this function. It doesn't
218     seem to matter on Linux, but it showed up on some releases of OS X.
220     3. It seems that there are systems where bytes whose values are greater than
221     127 match isprint() in the "C" locale. The "C" locale should be the
222     default when a C program starts up. In most systems, only ASCII printing
223     characters match isprint(). This difference caused the output from pcretest
224     to vary, making some of the tests fail. I have changed pcretest so that:
226     (a) When it is outputting text in the compiled version of a pattern, bytes
227     other than 32-126 are always shown as hex escapes.
229     (b) When it is outputting text that is a matched part of a subject string,
230     it does the same, unless a different locale has been set for the match
231     (using the /L modifier). In this case, it uses isprint() to decide.
233     4. Fixed a major bug that caused incorrect computation of the amount of memory
234     required for a compiled pattern when options that changed within the
235     pattern affected the logic of the preliminary scan that determines the
236     length. The relevant options are -x, and -i in UTF-8 mode. The result was
237     that the computed length was too small. The symptoms of this bug were
238     either the PCRE error "internal error: code overflow" from pcre_compile(),
239     or a glibc crash with a message such as "pcretest: free(): invalid next
240     size (fast)". Examples of patterns that provoked this bug (shown in
241     pcretest format) are:
243     /(?-x: )/x
244     /(?x)(?-x: \s*#\s*)/
245     /((?i)[\x{c0}])/8
246     /(?i:[\x{c0}])/8
248     HOWEVER: Change 17 below makes this fix obsolete as the memory computation
249     is now done differently.
251     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
252     wrapper classes; (b) implement a new function in the C++ scanner that is
253     more efficient than the old way of doing things because it avoids levels of
254     recursion in the regex matching; (c) add a paragraph to the documentation
255     for the FullMatch() function.
257     6. The escape sequence \n was being treated as whatever was defined as
258     "newline". Not only was this contrary to the documentation, which states
259     that \n is character 10 (hex 0A), but it also went horribly wrong when
260     "newline" was defined as CRLF. This has been fixed.
262     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
263     was being set to -1 for the "end of line" case (supposedly a value that no
264     character can have). Though this value is never used (the check for end of
265     line is "zero bytes in current character"), it caused compiler complaints.
266     I've changed it to 0xffffffff.
268     8. In pcre_version.c, the version string was being built by a sequence of
269     C macros that, in the event of PCRE_PRERELEASE being defined as an empty
270     string (as it is for production releases) called a macro with an empty
271     argument. The C standard says the result of this is undefined. The gcc
272     compiler treats it as an empty string (which was what was wanted) but it is
273     reported that Visual C gives an error. The source has been hacked around to
274     avoid this problem.
276     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
277     builds of pcretest, and changed the call to _setmode() to use _O_BINARY
278     instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
279     of them did).
281     10. Originally, pcretest opened its input and output without "b"; then I was
282     told that "b" was needed in some environments, so it was added for release
283     5.0 to both the input and output. (It makes no difference on Unix-like
284     systems.) Later I was told that it is wrong for the input on Windows. I've
285     now abstracted the modes into two macros, to make it easier to fiddle with
286     them, and removed "b" from the input mode under Windows.
288     11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
290     12. Added -help and --help to pcretest as an official way of being reminded
291     of the options.
293     13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
294     and pcrecpp.cc because they annoy compilers at high warning levels.
296     14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
298     15. Fixed an occurrence of == in configure.ac that should have been = (shell
299     scripts are not C programs :-) and which was not noticed because it works
300     on Linux.
302     16. pcretest is supposed to handle any length of pattern and data line (as one
303     line or as a continued sequence of lines) by extending its input buffer if
304     necessary. This feature was broken for very long pattern lines, leading to
305     a string of junk being passed to pcre_compile() if the pattern was longer
306     than about 50K.
308     17. I have done a major re-factoring of the way pcre_compile() computes the
309     amount of memory needed for a compiled pattern. Previously, there was code
310     that made a preliminary scan of the pattern in order to do this. That was
311     OK when PCRE was new, but as the facilities have expanded, it has become
312     harder and harder to keep it in step with the real compile phase, and there
313     have been a number of bugs (see for example, 4 above). I have now found a
314     cunning way of running the real compile function in a "fake" mode that
315     enables it to compute how much memory it would need, while actually only
316     ever using a few hundred bytes of working memory and without too many
317     tests of the mode. This should make future maintenance and development
318     easier. A side effect of this work is that the limit of 200 on the nesting
319     depth of parentheses has been removed (though this was never a serious
320     limitation, I suspect). However, there is a downside: pcre_compile() now
321     runs more slowly than before (30% or more, depending on the pattern). I
322     hope this isn't a big issue. There is no effect on runtime performance.
324     18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
325     newline (only possible for the last line of a file) and it was a
326     pattern that set a locale (followed by /Lsomething), pcretest crashed.
328     19. Added additional timing features to pcretest. (1) The -tm option now times
329     matching only, not compiling. (2) Both -t and -tm can be followed, as a
330     separate command line item, by a number that specifies the number of
331     repeats to use when timing. The default is 50000; this gives better
332     precision, but takes uncomfortably long for very large patterns.
334     20. Extended pcre_study() to be more clever in cases where a branch of a
335     subpattern has no definite first character. For example, (a*|b*)[cd] would
336     previously give no result from pcre_study(). Now it recognizes that the
337     first character must be a, b, c, or d.
339     21. There was an incorrect error "recursive call could loop indefinitely" if
340     a subpattern (or the entire pattern) that was being tested for matching an
341     empty string contained only one non-empty item after a nested subpattern.
342     For example, the pattern (?>\x{100}*)\d(?R) provoked this error
343     incorrectly, because the \d was being skipped in the check.
345     22. The pcretest program now has a new pattern option /B and a command line
346     option -b, which is equivalent to adding /B to every pattern. This causes
347     it to show the compiled bytecode, without the additional information that
348     -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
349     is the same as /B/I).
351     23. A new optimization is now able automatically to treat some sequences such
352     as a*b as a*+b. More specifically, if something simple (such as a character
353     or a simple class like \d) has an unlimited quantifier, and is followed by
354     something that cannot possibly match the quantified thing, the quantifier
355     is automatically "possessified".
357     24. A recursive reference to a subpattern whose number was greater than 39
358     went wrong under certain circumstances in UTF-8 mode. This bug could also
359     have affected the operation of pcre_study().
361     25. Realized that a little bit of performance could be had by replacing
362     (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
364     26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
366     27. Possessive quantifiers such as a++ were previously implemented by turning
367     them into atomic groups such as ($>a+). Now they have their own opcodes,
368     which improves performance. This includes the automatically created ones
369     from 23 above.
371     28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
372     lookahead was broken if it was not anchored. PCRE was mistakenly expecting
373     the first matched character to be a colon. This applied both to named and
374     numbered groups.
376     29. The ucpinternal.h header file was missing its idempotency #ifdef.
378     30. I was sent a "project" file called libpcre.a.dev which I understand makes
379     building PCRE on Windows easier, so I have included it in the distribution.
381     31. There is now a check in pcretest against a ridiculously large number being
382     returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
383     loop, the loop is abandoned.
385     32. Forward references to subpatterns in conditions such as (?(2)...) where
386     subpattern 2 is defined later cause pcre_compile() to search forwards in
387     the pattern for the relevant set of parentheses. This search went wrong
388     when there were unescaped parentheses in a character class, parentheses
389     escaped with \Q...\E, or parentheses in a #-comment in /x mode.
391     33. "Subroutine" calls and backreferences were previously restricted to
392     referencing subpatterns earlier in the regex. This restriction has now
393     been removed.
395     34. Added a number of extra features that are going to be in Perl 5.10. On the
396     whole, these are just syntactic alternatives for features that PCRE had
397     previously implemented using the Python syntax or my own invention. The
398     other formats are all retained for compatibility.
400     (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
401     as (?P<name>...). The new forms, as well as being in Perl 5.10, are
402     also .NET compatible.
404     (b) A recursion or subroutine call to a named group can now be defined as
405     (?&name) as well as (?P>name).
407     (c) A backreference to a named group can now be defined as \k<name> or
408     \k'name' as well as (?P=name). The new forms, as well as being in Perl
409     5.10, are also .NET compatible.
411     (d) A conditional reference to a named group can now use the syntax
412     (?(<name>) or (?('name') as well as (?(name).
414     (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
415     groups (named and numbered) that are never evaluated inline, but can be
416     called as "subroutines" from elsewhere. In effect, the DEFINE condition
417     is always false. There may be only one alternative in such a group.
419     (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
420     as the simple (?(R). The condition is true only if the most recent
421     recursion is that of the given number or name. It does not search out
422     through the entire recursion stack.
424     (g) The escape \gN or \g{N} has been added, where N is a positive or
425     negative number, specifying an absolute or relative reference.
427     35. Tidied to get rid of some further signed/unsigned compiler warnings and
428     some "unreachable code" warnings.
430     36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
431     things, this adds five new scripts.
433     37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
434     There were also incompatibilities regarding the handling of \Q..\E inside
435     character classes, for example with patterns like [\Qa\E-\Qz\E] where the
436     hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
438     38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
439     matches an empty string, and forcibly breaks the loop. There were bugs in
440     this code in non-simple cases. For a pattern such as ^(a()*)* matched
441     against aaaa the result was just "a" rather than "aaaa", for example. Two
442     separate and independent bugs (that affected different cases) have been
443     fixed.
445     39. Refactored the code to abolish the use of different opcodes for small
446     capturing bracket numbers. This is a tidy that I avoided doing when I
447     removed the limit on the number of capturing brackets for 3.5 back in 2001.
448     The new approach is not only tidier, it makes it possible to reduce the
449     memory needed to fix the previous bug (38).
451     40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
452     sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
453     processing dot, circumflex, or dollar metacharacters, or #-comments in /x
454     mode.
456     41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
457     report.
459     42. Applied patch, originally from Ari Pollak, modified by Google, to allow
460     copy construction and assignment in the C++ wrapper.
462     43. Updated pcregrep to support "--newline=any". In the process, I fixed a
463     couple of bugs that could have given wrong results in the "--newline=crlf"
464     case.
466     44. Added a number of casts and did some reorganization of signed/unsigned int
467     variables following suggestions from Dair Grant. Also renamed the variable
468     "this" as "item" because it is a C++ keyword.
470     45. Arranged for dftables to add
472     #include "pcre_internal.h"
474     to pcre_chartables.c because without it, gcc 4.x may remove the array
475     definition from the final binary if PCRE is built into a static library and
476     dead code stripping is activated.
478     46. For an unanchored pattern, if a match attempt fails at the start of a
479     newline sequence, and the newline setting is CRLF or ANY, and the next two
480     characters are CRLF, advance by two characters instead of one.
483 nigel 91 Version 6.7 04-Jul-06
484     ---------------------
486     1. In order to handle tests when input lines are enormously long, pcretest has
487     been re-factored so that it automatically extends its buffers when
488     necessary. The code is crude, but this _is_ just a test program. The
489     default size has been increased from 32K to 50K.
491     2. The code in pcre_study() was using the value of the re argument before
492     testing it for NULL. (Of course, in any sensible call of the function, it
493     won't be NULL.)
495     3. The memmove() emulation function in pcre_internal.h, which is used on
496     systems that lack both memmove() and bcopy() - that is, hardly ever -
497     was missing a "static" storage class specifier.
499     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
500     containing an extended class (one that cannot be represented by a bitmap
501     because it contains high-valued characters or Unicode property items, e.g.
502     [\pZ]). Almost always one would set UTF-8 mode when processing such a
503     pattern, but PCRE should not loop if you do not (it no longer does).
504     [Detail: two cases were found: (a) a repeated subpattern containing an
505     extended class; (b) a recursive reference to a subpattern that followed a
506     previous extended class. It wasn't skipping over the extended class
507     correctly when UTF-8 mode was not set.]
509     5. A negated single-character class was not being recognized as fixed-length
510     in lookbehind assertions such as (?<=[^f]), leading to an incorrect
511     compile error "lookbehind assertion is not fixed length".
513     6. The RunPerlTest auxiliary script was showing an unexpected difference
514     between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
515     write a Perl script that can interpret lines of an input file either as
516     byte characters or as UTF-8, which is what "perltest" was being required to
517     do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
518     can't do is switch easily at run time between having the "use utf8;" pragma
519     or not. In the end, I fudged it by using the RunPerlTest script to insert
520     "use utf8;" explicitly for the UTF-8 tests.
522     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
523     the end of the subject string, contrary to the documentation and to what
524     Perl does. This was true of both matching functions. Now it matches only at
525     the start of the subject and immediately after *internal* newlines.
527     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
528     a pointer to an int instead of a pointer to an unsigned long int. This
529     caused problems on 64-bit systems.
531     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
532     instance of the 'standard' template library not being so standard".
534     10. There was no check on the number of named subpatterns nor the maximum
535     length of a subpattern name. The product of these values is used to compute
536     the size of the memory block for a compiled pattern. By supplying a very
537     long subpattern name and a large number of named subpatterns, the size
538     computation could be caused to overflow. This is now prevented by limiting
539     the length of names to 32 characters, and the number of named subpatterns
540     to 10,000.
542     11. Subpatterns that are repeated with specific counts have to be replicated in
543     the compiled pattern. The size of memory for this was computed from the
544     length of the subpattern and the repeat count. The latter is limited to
545     65535, but there was no limit on the former, meaning that integer overflow
546     could in principle occur. The compiled length of a repeated subpattern is
547     now limited to 30,000 bytes in order to prevent this.
549     12. Added the optional facility to have named substrings with the same name.
551     13. Added the ability to use a named substring as a condition, using the
552     Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
553     are numbers (not recommended). Forward references are permitted.
555     14. Added forward references in named backreferences (if you see what I mean).
557     15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
558     pattern could run off the end of the subject. For example, the pattern
559     "(?s)(.{1,5})"8 did this with the subject "ab".
561     16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
562     PCRE_CASELESS was set when matching characters that were quantified with ?
563     or *.
565     17. A character class other than a single negated character that had a minimum
566     but no maximum quantifier - for example [ab]{6,} - was not handled
567     correctly by pce_dfa_exec(). It would match only one character.
569     18. A valid (though odd) pattern that looked like a POSIX character
570     class but used an invalid character after [ (for example [[,abc,]]) caused
571     pcre_compile() to give the error "Failed: internal error: code overflow" or
572     in some cases to crash with a glibc free() error. This could even happen if
573     the pattern terminated after [[ but there just happened to be a sequence of
574     letters, a binary zero, and a closing ] in the memory that followed.
576     19. Perl's treatment of octal escapes in the range \400 to \777 has changed
577     over the years. Originally (before any Unicode support), just the bottom 8
578     bits were taken. Thus, for example, \500 really meant \100. Nowadays the
579     output from "man perlunicode" includes this:
581     The regular expression compiler produces polymorphic opcodes. That
582     is, the pattern adapts to the data and automatically switches to
583     the Unicode character scheme when presented with Unicode data--or
584     instead uses a traditional byte scheme when presented with byte
585     data.
587     Sadly, a wide octal escape does not cause a switch, and in a string with
588     no other multibyte characters, these octal escapes are treated as before.
589     Thus, in Perl, the pattern /\500/ actually matches \100 but the pattern
590     /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
591     Unicode string.
593     I have not perpetrated such confusion in PCRE. Up till now, it took just
594     the bottom 8 bits, as in old Perl. I have now made octal escapes with
595     values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
596     translate to the appropriate multibyte character.
598     29. Applied some refactoring to reduce the number of warnings from Microsoft
599     and Borland compilers. This has included removing the fudge introduced
600     seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
601     a warning about an unused variable.
603     21. PCRE has not included VT (character 0x0b) in the set of whitespace
604     characters since release 4.0, because Perl (from release 5.004) does not.
605     [Or at least, is documented not to: some releases seem to be in conflict
606     with the documentation.] However, when a pattern was studied with
607     pcre_study() and all its branches started with \s, PCRE still included VT
608     as a possible starting character. Of course, this did no harm; it just
609     caused an unnecessary match attempt.
611     22. Removed a now-redundant internal flag bit that recorded the fact that case
612     dependency changed within the pattern. This was once needed for "required
613     byte" processing, but is no longer used. This recovers a now-scarce options
614     bit. Also moved the least significant internal flag bit to the most-
615     significant bit of the word, which was not previously used (hangover from
616     the days when it was an int rather than a uint) to free up another bit for
617     the future.
619     23. Added support for CRLF line endings as well as CR and LF. As well as the
620     default being selectable at build time, it can now be changed at runtime
621     via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
622     specify that it is scanning data with non-default line endings.
624     24. Changed the definition of CXXLINK to make it agree with the definition of
625     LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
627     25. Applied Ian Taylor's patches to avoid using another stack frame for tail
628     recursions. This makes a big different to stack usage for some patterns.
630     26. If a subpattern containing a named recursion or subroutine reference such
631     as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
632     the space required for the compiled pattern went wrong and gave too small a
633     value. Depending on the environment, this could lead to "Failed: internal
634     error: code overflow at offset 49" or "glibc detected double free or
635     corruption" errors.
637     27. Applied patches from Google (a) to support the new newline modes and (b) to
638     advance over multibyte UTF-8 characters in GlobalReplace.
640     28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
641     difference for some implementation of PCRE in some Windows version.
643     29. Added some extra testing facilities to pcretest:
645     \q<number> in a data line sets the "match limit" value
646     \Q<number> in a data line sets the "match recursion limt" value
647     -S <number> sets the stack size, where <number> is in megabytes
649     The -S option isn't available for Windows.
652 nigel 89 Version 6.6 06-Feb-06
653     ---------------------
655     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
656     in pcreposix.h. I have copied the definition from pcre.h.
658     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
659     because pcre.h is no longer a built file.
661     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
662     not normally included in the compiled code.
665 nigel 87 Version 6.5 01-Feb-06
666     ---------------------
668     1. When using the partial match feature with pcre_dfa_exec(), it was not
669     anchoring the second and subsequent partial matches at the new starting
670     point. This could lead to incorrect results. For example, with the pattern
671     /1234/, partially matching against "123" and then "a4" gave a match.
673     2. Changes to pcregrep:
675     (a) All non-match returns from pcre_exec() were being treated as failures
676     to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
677     error message is output. Some extra information is given for the
679     probably the only errors that are likely to be caused by users (by
680     specifying a regex that has nested indefinite repeats, for instance).
681     If there are more than 20 of these errors, pcregrep is abandoned.
683     (b) A binary zero was treated as data while matching, but terminated the
684     output line if it was written out. This has been fixed: binary zeroes
685     are now no different to any other data bytes.
687     (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
688     used to set a locale for matching. The --locale=xxxx long option has
689     been added (no short equivalent) to specify a locale explicitly on the
690     pcregrep command, overriding the environment variables.
692     (d) When -B was used with -n, some line numbers in the output were one less
693     than they should have been.
695     (e) Added the -o (--only-matching) option.
697     (f) If -A or -C was used with -c (count only), some lines of context were
698     accidentally printed for the final match.
700     (g) Added the -H (--with-filename) option.
702     (h) The combination of options -rh failed to suppress file names for files
703     that were found from directory arguments.
705     (i) Added the -D (--devices) and -d (--directories) options.
707     (j) Added the -F (--fixed-strings) option.
709     (k) Allow "-" to be used as a file name for -f as well as for a data file.
711     (l) Added the --colo(u)r option.
713     (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
714     is not present by default.
716     3. A nasty bug was discovered in the handling of recursive patterns, that is,
717     items such as (?R) or (?1), when the recursion could match a number of
718     alternatives. If it matched one of the alternatives, but subsequently,
719     outside the recursion, there was a failure, the code tried to back up into
720     the recursion. However, because of the way PCRE is implemented, this is not
721     possible, and the result was an incorrect result from the match.
723     In order to prevent this happening, the specification of recursion has
724     been changed so that all such subpatterns are automatically treated as
725     atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
727     4. I had overlooked the fact that, in some locales, there are characters for
728     which isalpha() is true but neither isupper() nor islower() are true. In
729     the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
730     and ordfeminine) are like this. This affected the treatment of \w and \W
731     when they appeared in character classes, but not when they appeared outside
732     a character class. The bit map for "word" characters is now created
733     separately from the results of isalnum() instead of just taking it from the
734     upper, lower, and digit maps. (Plus the underscore character, of course.)
736     5. The above bug also affected the handling of POSIX character classes such as
737     [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
738     permanent tables. Instead, the bit maps for such a class were previously
739     created as the appropriate unions of the upper, lower, and digit bitmaps.
740     Now they are created by subtraction from the [[:word:]] class, which has
741     its own bitmap.
743     6. The [[:blank:]] character class matches horizontal, but not vertical space.
744     It is created by subtracting the vertical space characters (\x09, \x0a,
745     \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
746     subtraction was done in the overall bitmap for a character class, meaning
747     that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
748     be recognized. This bug has been fixed.
750     7. Patches from the folks at Google:
752     (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
753     real life, but is still worth protecting against".
755     (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
756     regular expressions".
758     (c) pcre_scanner.cc: avoid use of std::count() because not all systems
759     have it.
761     (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
762     "configure" and the latter not, in order to fix a problem somebody had
763     with compiling the Arg class on HP-UX.
765     (e) Improve the error-handling of the C++ wrapper a little bit.
767     (f) New tests for checking recursion limiting.
769     8. The pcre_memmove() function, which is used only if the environment does not
770     have a standard memmove() function (and is therefore rarely compiled),
771     contained two bugs: (a) use of int instead of size_t, and (b) it was not
772     returning a result (though PCRE never actually uses the result).
774     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
775     large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
776     returned instead of calling malloc() with an overflowing number that would
777     most likely cause subsequent chaos.
779     10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
781     11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
782     with this option is matched, the nmatch and pmatch options of regexec() are
783     ignored.
785     12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
786     provided in case anyone wants to the the POSIX interface with UTF-8
787     strings.
789     13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
790     C++ linking (needed for some HP-UX environments).
792     14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
793     (unused parameter) and in the pcre_printint() function (omitted "default"
794     switch label when the default is to do nothing).
796     15. Added some code to make it possible, when PCRE is compiled as a C++
797     library, to replace subject pointers for pcre_exec() with a smart pointer
798     class, thus making it possible to process discontinuous strings.
800     16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
801     much the same function. They were added by different people who were trying
802     to make PCRE easy to compile on non-Unix systems. It has been suggested
803     that PCRE_EXPORT be abolished now that there is more automatic apparatus
804     for compiling on Windows systems. I have therefore replaced it with
805     PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
806     defaults to "extern" for C or "extern C" for C++, which works fine on
807     Unix-like systems. It is now possible to override the value of PCRE_DATA_
808     SCOPE with something explicit in config.h. In addition:
810     (a) pcreposix.h still had just "extern" instead of either of these macros;
811     I have replaced it with PCRE_DATA_SCOPE.
813     (b) Functions such as _pcre_xclass(), which are internal to the library,
814     but external in the C sense, all had PCRE_EXPORT in their definitions.
815     This is apparently wrong for the Windows case, so I have removed it.
816     (It makes no difference on Unix-like systems.)
818     17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
819     of recursive calls to match(). This is different to MATCH_LIMIT because
820     that limits the total number of calls to match(), not all of which increase
821     the depth of recursion. Limiting the recursion depth limits the amount of
822     stack (or heap if NO_RECURSE is set) that is used. The default can be set
823     when PCRE is compiled, and changed at run time. A patch from Google adds
824     this functionality to the C++ interface.
826     18. Changes to the handling of Unicode character properties:
828     (a) Updated the table to Unicode 4.1.0.
830     (b) Recognize characters that are not in the table as "Cn" (undefined).
832     (c) I revised the way the table is implemented to a much improved format
833     which includes recognition of ranges. It now supports the ranges that
834     are defined in UnicodeData.txt, and it also amalgamates other
835     characters into ranges. This has reduced the number of entries in the
836     table from around 16,000 to around 3,000, thus reducing its size
837     considerably. I realized I did not need to use a tree structure after
838     all - a binary chop search is just as efficient. Having reduced the
839     number of entries, I extended their size from 6 bytes to 8 bytes to
840     allow for more data.
842     (d) Added support for Unicode script names via properties such as \p{Han}.
844     19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
845     matching that character.
847     20. When matching a repeated Unicode property with a minimum greater than zero,
848     (for example \pL{2,}), PCRE could look past the end of the subject if it
849     reached it while seeking the minimum number of characters. This could
850     happen only if some of the characters were more than one byte long, because
851     there is a check for at least the minimum number of bytes.
853     21. Refactored the implementation of \p and \P so as to be more general, to
854     allow for more different types of property in future. This has changed the
855     compiled form incompatibly. Anybody with saved compiled patterns that use
856     \p or \P will have to recompile them.
858     22. Added "Any" and "L&" to the supported property types.
860     23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
861     but give a compile time error if the value is greater than 0xff.
863     24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
864     accidentally not being installed or uninstalled.
866     25. The pcre.h file was built from pcre.h.in, but the only changes that were
867     made were to insert the current release number. This seemed silly, because
868     it made things harder for people building PCRE on systems that don't run
869     "configure". I have turned pcre.h into a distributed file, no longer built
870     by "configure", with the version identification directly included. There is
871     no longer a pcre.h.in file.
873     However, this change necessitated a change to the pcre-config script as
874     well. It is built from pcre-config.in, and one of the substitutions was the
875     release number. I have updated configure.ac so that ./configure now finds
876     the release number by grepping pcre.h.
878     26. Added the ability to run the tests under valgrind.
881 nigel 85 Version 6.4 05-Sep-05
882     ---------------------
884     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
885     "--" to be printed when multiple files were scanned, even when none of the
886     -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
887     consider it to be a bug, and have restored the previous behaviour.
889     2. A couple of code tidies to get rid of compiler warnings.
891     3. The pcretest program used to cheat by referring to symbols in the library
892     whose names begin with _pcre_. These are internal symbols that are not
893     really supposed to be visible externally, and in some environments it is
894     possible to suppress them. The cheating is now confined to including
895     certain files from the library's source, which is a bit cleaner.
897     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
898     file's purpose clearer.
900     5. Reorganized pcre_ucp_findchar().
903 nigel 83 Version 6.3 15-Aug-05
904     ---------------------
906     1. The file libpcre.pc.in did not have general read permission in the tarball.
908     2. There were some problems when building without C++ support:
910     (a) If C++ support was not built, "make install" and "make test" still
911     tried to test it.
913     (b) There were problems when the value of CXX was explicitly set. Some
914     changes have been made to try to fix these, and ...
916     (c) --disable-cpp can now be used to explicitly disable C++ support.
918     (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
919     backslash in a target when C++ was disabled. This confuses some
920     versions of "make", apparently. Using an intermediate variable solves
921     this. (Same for CPP_LOBJ.)
923     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
924     (non-Windows) now includes $(CFLAGS) because these flags are sometimes
925     necessary on certain architectures.
927     4. Added a setting of -export-symbols-regex to the link command to remove
928     those symbols that are exported in the C sense, but actually are local
929     within the library, and not documented. Their names all begin with
930     "_pcre_". This is not a perfect job, because (a) we have to except some
931     symbols that pcretest ("illegally") uses, and (b) the facility isn't always
932     available (and never for static libraries). I have made a note to try to
933     find a way round (a) in the future.
936 nigel 81 Version 6.2 01-Aug-05
937     ---------------------
939     1. There was no test for integer overflow of quantifier values. A construction
940     such as {1111111111111111} would give undefined results. What is worse, if
941     a minimum quantifier for a parenthesized subpattern overflowed and became
942     negative, the calculation of the memory size went wrong. This could have
943     led to memory overwriting.
945     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
947     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
948     operating environments where this matters.
950     4. Applied Giuseppe Maxia's patch to add additional features for controlling
951     PCRE options from within the C++ wrapper.
953     5. Named capturing subpatterns were not being correctly counted when a pattern
954     was compiled. This caused two problems: (a) If there were more than 100
955     such subpatterns, the calculation of the memory needed for the whole
956     compiled pattern went wrong, leading to an overflow error. (b) Numerical
957     back references of the form \12, where the number was greater than 9, were
958     not recognized as back references, even though there were sufficient
959     previous subpatterns.
961     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
962     versions of gcc, e.g. 2.95.4.
965 nigel 79 Version 6.1 21-Jun-05
966     ---------------------
968     1. There was one reference to the variable "posix" in pcretest.c that was not
969     surrounded by "#if !defined NOPOSIX".
971     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
972     the cross-check on the old pcre_info() function, for the benefit of the
973     cut-down version of PCRE that is currently imported into Exim.
975     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
976     allocation error. I've done the easy fix, which wastes 2 bytes for sensible
977     patterns that start (?i) but I don't think that matters. The use of (?i) is
978     just an example; this all applies to the other options as well.
980     4. Since libtool seems to echo the compile commands it is issuing, the output
981     from "make" can be reduced a bit by putting "@" in front of each libtool
982     compile command.
984     5. Patch from the folks at Google for configure.in to be a bit more thorough
985     in checking for a suitable C++ installation before trying to compile the
986     C++ stuff. This should fix a reported problem when a compiler was present,
987     but no suitable headers.
989     6. The man pages all had just "PCRE" as their title. I have changed them to
990     be the relevant file name. I have also arranged that these names are
991     retained in the file doc/pcre.txt, which is a concatenation in text format
992     of all the man pages except the little individual ones for each function.
994     7. The NON-UNIX-USE file had not been updated for the different set of source
995     files that come with release 6. I also added a few comments about the C++
996     wrapper.
999 nigel 77 Version 6.0 07-Jun-05
1000     ---------------------
1002     1. Some minor internal re-organization to help with my DFA experiments.
1004     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1005     didn't matter for the library itself when fully configured, but did matter
1006     when compiling without UCP support, or within Exim, where the ucp files are
1007     not imported.
1009     3. Refactoring of the library code to split up the various functions into
1010     different source modules. The addition of the new DFA matching code (see
1011     below) to a single monolithic source would have made it really too
1012     unwieldy, quite apart from causing all the code to be include in a
1013     statically linked application, when only some functions are used. This is
1014     relevant even without the DFA addition now that patterns can be compiled in
1015     one application and matched in another.
1017     The downside of splitting up is that there have to be some external
1018     functions and data tables that are used internally in different modules of
1019     the library but which are not part of the API. These have all had their
1020     names changed to start with "_pcre_" so that they are unlikely to clash
1021     with other external names.
1023     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1024     a different (DFA) algorithm. Although it is slower than the original
1025     function, it does have some advantages for certain types of matching
1026     problem.
1028     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1029     including restarting after a partial match.
1031     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1032     defined when compiling for Windows was sent to me. I have put it into the
1033     code, though I have no means of testing or verifying it.
1035     7. Added the pcre_refcount() auxiliary function.
1037     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1038     match before or at the first newline in the subject string. In pcretest,
1039     the /f option on a pattern can be used to set this.
1041     9. A repeated \w when used in UTF-8 mode with characters greater than 256
1042     would behave wrongly. This has been present in PCRE since release 4.0.
1044     10. A number of changes to the pcregrep command:
1046     (a) Refactored how -x works; insert ^(...)$ instead of setting
1047     PCRE_ANCHORED and checking the length, in preparation for adding
1048     something similar for -w.
1050     (b) Added the -w (match as a word) option.
1052     (c) Refactored the way lines are read and buffered so as to have more
1053     than one at a time available.
1055     (d) Implemented a pcregrep test script.
1057     (e) Added the -M (multiline match) option. This allows patterns to match
1058     over several lines of the subject. The buffering ensures that at least
1059     8K, or the rest of the document (whichever is the shorter) is available
1060     for matching (and similarly the previous 8K for lookbehind assertions).
1062     (f) Changed the --help output so that it now says
1064     -w, --word-regex(p)
1066     instead of two lines, one with "regex" and the other with "regexp"
1067     because that confused at least one person since the short forms are the
1068     same. (This required a bit of code, as the output is generated
1069     automatically from a table. It wasn't just a text change.)
1071     (g) -- can be used to terminate pcregrep options if the next thing isn't an
1072     option but starts with a hyphen. Could be a pattern or a path name
1073     starting with a hyphen, for instance.
1075     (h) "-" can be given as a file name to represent stdin.
1077     (i) When file names are being printed, "(standard input)" is used for
1078     the standard input, for compatibility with GNU grep. Previously
1079     "<stdin>" was used.
1081     (j) The option --label=xxx can be used to supply a name to be used for
1082     stdin when file names are being printed. There is no short form.
1084     (k) Re-factored the options decoding logic because we are going to add
1085     two more options that take data. Such options can now be given in four
1086     different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1088     (l) Added the -A, -B, and -C options for requesting that lines of context
1089     around matches be printed.
1091     (m) Added the -L option to print the names of files that do not contain
1092     any matching lines, that is, the complement of -l.
1094     (n) The return code is 2 if any file cannot be opened, but pcregrep does
1095     continue to scan other files.
1097     (o) The -s option was incorrectly implemented. For compatibility with other
1098     greps, it now suppresses the error message for a non-existent or non-
1099     accessible file (but not the return code). There is a new option called
1100     -q that suppresses the output of matching lines, which was what -s was
1101     previously doing.
1103     (p) Added --include and --exclude options to specify files for inclusion
1104     and exclusion when recursing.
1106     11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1107     Hopefully, it now does.
1109     12. Missing cast in pcre_study().
1111     13. Added an "uninstall" target to the makefile.
1113     14. Replaced "extern" in the function prototypes in Makefile.in with
1114     "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1115     world, but is set differently for Windows.
1117     15. Added a second compiling function called pcre_compile2(). The only
1118     difference is that it has an extra argument, which is a pointer to an
1119     integer error code. When there is a compile-time failure, this is set
1120     non-zero, in addition to the error test pointer being set to point to an
1121     error message. The new argument may be NULL if no error number is required
1122     (but then you may as well call pcre_compile(), which is now just a
1123     wrapper). This facility is provided because some applications need a
1124     numeric error indication, but it has also enabled me to tidy up the way
1125     compile-time errors are handled in the POSIX wrapper.
1127     16. Added VPATH=.libs to the makefile; this should help when building with one
1128     prefix path and installing with another. (Or so I'm told by someone who
1129     knows more about this stuff than I do.)
1131     17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1132     passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1133     match everything, including newlines. This is not POSIX-compatible, but
1134     somebody wanted the feature. From pcretest it can be activated by using
1135     both the P and the s flags.
1137     18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1139     19. libpcre.pc was being incorrectly installed as executable.
1141     20. A couple of places in pcretest check for end-of-line by looking for '\n';
1142     it now also looks for '\r' so that it will work unmodified on Windows.
1144     21. Added Google's contributed C++ wrapper to the distribution.
1146     22. Added some untidy missing memory free() calls in pcretest, to keep
1147     Electric Fence happy when testing.
1151 nigel 75 Version 5.0 13-Sep-04
1152     ---------------------
1154     1. Internal change: literal characters are no longer packed up into items
1155     containing multiple characters in a single byte-string. Each character
1156     is now matched using a separate opcode. However, there may be more than one
1157     byte in the character in UTF-8 mode.
1159     2. The pcre_callout_block structure has two new fields: pattern_position and
1160     next_item_length. These contain the offset in the pattern to the next match
1161     item, and its length, respectively.
1163     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1164     insertion of callouts before each pattern item. Added the /C option to
1165     pcretest to make use of this.
1167     4. On the advice of a Windows user, the lines
1169     #if defined(_WIN32) || defined(WIN32)
1170     _setmode( _fileno( stdout ), 0x8000 );
1171     #endif /* defined(_WIN32) || defined(WIN32) */
1173     have been added to the source of pcretest. This apparently does useful
1174     magic in relation to line terminators.
1176     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1177     for the benefit of those environments where the "b" makes a difference.
1179     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1180     to know about it. I have put a hack into configure.in that adds in code
1181     to set GCC=yes if CC=icc. This seems to end up at a point in the
1182     generated configure script that is early enough to affect the setting of
1183     compiler options, which is what is needed, but I have no means of testing
1184     whether it really works. (The user who reported this had patched the
1185     generated configure script, which of course I cannot do.)
1187     LATER: After change 22 below (new libtool files), the configure script
1188     seems to know about icc (and also ecc). Therefore, I have commented out
1189     this hack in configure.in.
1191     7. Added support for pkg-config (2 patches were sent in).
1193     8. Negated POSIX character classes that used a combination of internal tables
1194     were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1195     [[:^ascii]]. Typically, they would match almost any characters. The other
1196     POSIX classes were not broken in this way.
1198     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1199     to find the match, as PCRE was deluded into thinking that the match had to
1200     start at the start point or following a newline. The same bug applied to
1201     patterns with negative forward assertions or any backward assertions
1202     preceding ".*" at the start, unless the pattern required a fixed first
1203     character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1205     10. In UTF-8 mode, when moving forwards in the subject after a failed match
1206     starting at the last subject character, bytes beyond the end of the subject
1207     string were read.
1209     11. Renamed the variable "class" as "classbits" to make life easier for C++
1210     users. (Previously there was a macro definition, but it apparently wasn't
1211     enough.)
1213     12. Added the new field "tables" to the extra data so that tables can be passed
1214     in at exec time, or the internal tables can be re-selected. This allows
1215     a compiled regex to be saved and re-used at a later time by a different
1216     program that might have everything at different addresses.
1218     13. Modified the pcre-config script so that, when run on Solaris, it shows a
1219     -R library as well as a -L library.
1221     14. The debugging options of pcretest (-d on the command line or D on a
1222     pattern) showed incorrect output for anything following an extended class
1223     that contained multibyte characters and which was followed by a quantifier.
1225     15. Added optional support for general category Unicode character properties
1226     via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1227     support. It adds about 90K to the size of the library. The meanings of the
1228     inbuilt class escapes such as \d and \s have NOT been changed.
1230     16. Updated pcredemo.c to include calls to free() to release the memory for the
1231     compiled pattern.
1233     17. The generated file chartables.c was being created in the source directory
1234     instead of in the building directory. This caused the build to fail if the
1235     source directory was different from the building directory, and was
1236     read-only.
1238     18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1239     file. No doubt somebody will tell me if they don't make sense... Also added
1240     Dan Mooney's comments about building on OpenVMS.
1242     19. Added support for partial matching via the PCRE_PARTIAL option for
1243     pcre_exec() and the \P data escape in pcretest.
1245     20. Extended pcretest with 3 new pattern features:
1247     (i) A pattern option of the form ">rest-of-line" causes pcretest to
1248     write the compiled pattern to the file whose name is "rest-of-line".
1249     This is a straight binary dump of the data, with the saved pointer to
1250     the character tables forced to be NULL. The study data, if any, is
1251     written too. After writing, pcretest reads a new pattern.
1253     (ii) If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1254     compiled pattern from the given file. There must not be any
1255     occurrences of "<" in the file name (pretty unlikely); if there are,
1256     pcretest will instead treat the initial "<" as a pattern delimiter.
1257     After reading in the pattern, pcretest goes on to read data lines as
1258     usual.
1260     (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1261     and 16-bit fields in a compiled pattern, to simulate a pattern that
1262     was compiled on a host of opposite endianness.
1264     21. The pcre-exec() function can now cope with patterns that were compiled on
1265     hosts of opposite endianness, with this restriction:
1267     As for any compiled expression that is saved and used later, the tables
1268     pointer field cannot be preserved; the extra_data field in the arguments
1269     to pcre_exec() should be used to pass in a tables address if a value
1270     other than the default internal tables were used at compile time.
1272     22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1273     now diagnosed as an error. Previously, most of the time, a negative number
1274     would have been treated as zero, but if in addition "ovector" was passed as
1275     NULL, a crash could occur.
1277     23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1278     new versions from the libtool 1.5 distribution (the last one is a copy of
1279     a file called libtool.m4). This seems to have fixed the need to patch
1280     "configure" to support Darwin 1.3 (which I used to do). However, I still
1281     had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1282     workstation).
1284     24. Changed the PCRE licence to be the more standard "BSD" licence.
1287 nigel 73 Version 4.5 01-Dec-03
1288     ---------------------
1290     1. There has been some re-arrangement of the code for the match() function so
1291     that it can be compiled in a version that does not call itself recursively.
1292     Instead, it keeps those local variables that need separate instances for
1293     each "recursion" in a frame on the heap, and gets/frees frames whenever it
1294     needs to "recurse". Keeping track of where control must go is done by means
1295     of setjmp/longjmp. The whole thing is implemented by a set of macros that
1296     hide most of the details from the main code, and operates only if
1297     NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1298     "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1299     operating.
1301     To make it easier for callers to provide specially tailored get/free
1302     functions for this usage, two new functions, pcre_stack_malloc, and
1303     pcre_stack_free, are used. They are always called in strict stacking order,
1304     and the size of block requested is always the same.
1306     The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1307     PCRE has been compiled to use the stack or the heap for recursion. The
1308     -C option of pcretest uses this to show which version is compiled.
1310     A new data escape \S, is added to pcretest; it causes the amounts of store
1311     obtained and freed by both kinds of malloc/free at match time to be added
1312     to the output.
1314     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1315     what's available on my current Linux desktop machine.
1317     3. When matching a UTF-8 string, the test for a valid string at the start has
1318     been extended. If start_offset is not zero, PCRE now checks that it points
1319     to a byte that is the start of a UTF-8 character. If not, it returns
1320     PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1321     this is necessary because there may be backward assertions in the pattern.
1322     When matching the same subject several times, it may save resources to use
1323     PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1325     4. The code for checking the validity of UTF-8 strings has been tightened so
1326     that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1327     containing "overlong sequences".
1329     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1330     I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1331     should have been "&", but it just so happened that all the cases this let
1332     through by mistake were picked up later in the function.
1334     6. I had used a variable called "isblank" - this is a C99 function, causing
1335     some compilers to warn. To avoid this, I renamed it (as "blankclass").
1337     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1338     prompting; (b) run "./pcretest /dev/null" at the start of the test script
1339     so the version is shown; (c) stop "make test" echoing "./RunTest".
1341     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1343     9. The prototype for memmove() for systems that don't have it was using
1344     size_t, but the inclusion of the header that defines size_t was later. I've
1345     moved the #includes for the C headers earlier to avoid this.
1347     10. Added some adjustments to the code to make it easier to compiler on certain
1348     special systems:
1350     (a) Some "const" qualifiers were missing.
1351     (b) Added the macro EXPORT before all exported functions; by default this
1352     is defined to be empty.
1353     (c) Changed the dftables auxiliary program (that builds chartables.c) so
1354     that it reads its output file name as an argument instead of writing
1355     to the standard output and assuming this can be redirected.
1357     11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
1358     class containing characters with values greater than 255, PCRE compilation
1359     went into a loop.
1361     12. A recursive reference to a subpattern that was within another subpattern
1362     that had a minimum quantifier of zero caused PCRE to crash. For example,
1363     (x(y(?2))z)? provoked this bug with a subject that got as far as the
1364     recursion. If the recursively-called subpattern itself had a zero repeat,
1365     that was OK.
1367     13. In pcretest, the buffer for reading a data line was set at 30K, but the
1368     buffer into which it was copied (for escape processing) was still set at
1369     1024, so long lines caused crashes.
1371     14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
1372     "internal error: code overflow...". This applied to any character class
1373     that was followed by a possessive quantifier.
1375     15. Modified the Makefile to add libpcre.la as a prerequisite for
1376     libpcreposix.la because I was told this is needed for a parallel build to
1377     work.
1379     16. If a pattern that contained .* following optional items at the start was
1380     studied, the wrong optimizing data was generated, leading to matching
1381     errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
1382     matching string must start with a or b or c. The correct conclusion for
1383     this pattern is that a match can start with any character.
1386 nigel 71 Version 4.4 13-Aug-03
1387     ---------------------
1389     1. In UTF-8 mode, a character class containing characters with values between
1390     127 and 255 was not handled correctly if the compiled pattern was studied.
1391     In fixing this, I have also improved the studying algorithm for such
1392     classes (slightly).
1394     2. Three internal functions had redundant arguments passed to them. Removal
1395     might give a very teeny performance improvement.
1397     3. Documentation bug: the value of the capture_top field in a callout is *one
1398     more than* the number of the hightest numbered captured substring.
1400     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
1401     in incorrectly linking with a previously installed version. They now link
1402     explicitly with libpcre.la.
1404     5. configure.in no longer needs to recognize Cygwin specially.
1406     6. A problem in pcre.in for Windows platforms is fixed.
1408     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
1409     pcretest, it used to include the size of the study block as part of its
1410     output. Unfortunately, the structure contains a field that has a different
1411     size on different hardware architectures. This meant that the tests that
1412     showed this size failed. As the block is currently always of a fixed size,
1413     this information isn't actually particularly useful in pcretest output, so
1414     I have just removed it.
1416     8. Three pre-processor statements accidentally did not start in column 1.
1417     Sadly, there are *still* compilers around that complain, even though
1418     standard C has not required this for well over a decade. Sigh.
1420     9. In pcretest, the code for checking callouts passed small integers in the
1421     callout_data field, which is a void * field. However, some picky compilers
1422     complained about the casts involved for this on 64-bit systems. Now
1423     pcretest passes the address of the small integer instead, which should get
1424     rid of the warnings.
1426     10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
1427     both compile and run time, and gives an error if an invalid UTF-8 sequence
1428     is found. There is a option for disabling this check in cases where the
1429     string is known to be correct and/or the maximum performance is wanted.
1431     11. In response to a bug report, I changed one line in Makefile.in from
1433     -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
1434     to
1435     -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
1437     to look similar to other lines, but I have no way of telling whether this
1438     is the right thing to do, as I do not use Windows. No doubt I'll get told
1439     if it's wrong...
1442 nigel 69 Version 4.3 21-May-03
1443     ---------------------
1445     1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
1446     Makefile.
1448     2. Some refactoring to improve the quality of the code:
1450     (i) The utf8_table... variables are now declared "const".
1452     (ii) The code for \cx, which used the "case flipping" table to upper case
1453     lower case letters, now just substracts 32. This is ASCII-specific,
1454     but the whole concept of \cx is ASCII-specific, so it seems
1455     reasonable.
1457     (iii) PCRE was using its character types table to recognize decimal and
1458     hexadecimal digits in the pattern. This is silly, because it handles
1459     only 0-9, a-f, and A-F, but the character types table is locale-
1460     specific, which means strange things might happen. A private
1461     table is now used for this - though it costs 256 bytes, a table is
1462     much faster than multiple explicit tests. Of course, the standard
1463     character types table is still used for matching digits in subject
1464     strings against \d.
1466     (iv) Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
1467     ending in _t are). So I've renamed it as ESC_tee.
1469     3. The first argument for regexec() in the POSIX wrapper should have been
1470     defined as "const".
1472     4. Changed pcretest to use malloc() for its buffers so that they can be
1473     Electric Fenced for debugging.
1475     5. There were several places in the code where, in UTF-8 mode, PCRE would try
1476     to read one or more bytes before the start of the subject string. Often this
1477     had no effect on PCRE's behaviour, but in some circumstances it could
1478     provoke a segmentation fault.
1480     6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
1481     to try to read one or more bytes before the start of the subject string.
1483     7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
1484     UTF-8 support could misbehave in various ways if the subject string
1485     contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
1486     area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
1487     back over UTF-8 characters.)
1490 nigel 67 Version 4.2 14-Apr-03
1491     ---------------------
1493     1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
1495     2. Changes to the building process, supplied by Ronald Landheer-Cieslak
1496     [ON_WINDOWS]: new variable, "#" on non-Windows platforms
1497     [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
1498     [WIN_PREFIX]: new variable, "cyg" for Cygwin
1499     * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
1500     and BUILD_EXEEXT
1501     Note: automatic setting of the BUILD variables is not yet working
1502     set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
1503     compile-time but not at link-time
1504     [LINK]: use for linking executables only
1505     make different versions for Windows and non-Windows
1506     [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
1507     libraries
1508     [LINK_FOR_BUILD]: new variable
1509     [OBJEXT]: use throughout
1510     [EXEEXT]: use throughout
1511     <winshared>: new target
1512     <wininstall>: new target
1513     <dftables.o>: use native compiler
1514     <dftables>: use native linker
1515     <install>: handle Windows platform correctly
1516     <clean>: ditto
1517     <check>: ditto
1518     copy DLL to top builddir before testing
1520     As part of these changes, -no-undefined was removed again. This was reported
1521     to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
1522     in any case.
1524     3. Some tidies to get rid of compiler warnings:
1526     . In the match_data structure, match_limit was an unsigned long int, whereas
1527     match_call_count was an int. I've made them both unsigned long ints.
1529     . In pcretest the fact that a const uschar * doesn't automatically cast to
1530     a void * provoked a warning.
1532     . Turning on some more compiler warnings threw up some "shadow" variables
1533     and a few more missing casts.
1535     4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
1536     option, a class that contained a single character with a value between 128
1537     and 255 (e.g. /[\xFF]/) caused PCRE to crash.
1539     5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
1540     option, a class that contained several characters, but with at least one
1541     whose value was between 128 and 255 caused PCRE to crash.
1544 nigel 65 Version 4.1 12-Mar-03
1545     ---------------------
1546 nigel 63
1547 nigel 65 1. Compiling with gcc -pedantic found a couple of places where casts were
1548     needed, and a string in dftables.c that was longer than standard compilers are
1549     required to support.
1551     2. Compiling with Sun's compiler found a few more places where the code could
1552     be tidied up in order to avoid warnings.
1554     3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
1555     first of these names is deprecated in the latest Autoconf in favour of the name
1556     CC_FOR_BUILD, because "host" is typically used to mean the system on which the
1557     compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
1558     analogy I have changed it to CFLAGS_FOR_BUILD.
1560     4. Added -no-undefined to the linking command in the Makefile, because this is
1561     apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
1562     linking step for the pcreposix library.
1564     5. PCRE was failing to diagnose the case of two named groups with the same
1565     name.
1567     6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
1568     literal character that is needed in the subject for a match, and scans along to
1569     ensure that it is present before embarking on the full matching process. This
1570     saves time in cases of nested unlimited repeats that are never going to match.
1571     Problem: the scan can take a lot of time if the subject is very long (e.g.
1572     megabytes), thus penalizing straightforward matches. It is now done only if the
1573     amount of subject to be scanned is less than 1000 bytes.
1575     7. A lesser problem with the same optimization is that it was recording the
1576     first character of an anchored pattern as "needed", thus provoking a search
1577     right along the subject, even when the first match of the pattern was going to
1578     fail. The "needed" character is now not set for anchored patterns, unless it
1579     follows something in the pattern that is of non-fixed length. Thus, it still
1580     fulfils its original purpose of finding quick non-matches in cases of nested
1581     unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
1584     Version 4.0 17-Feb-03
1585     ---------------------
1587 nigel 63 1. If a comment in an extended regex that started immediately after a meta-item
1588     extended to the end of string, PCRE compiled incorrect data. This could lead to
1589     all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
1591     2. Moved to autoconf 2.53 and libtool 1.4.2.
1593     3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
1594     the special perltest8 script is no longer needed - all the tests can be run
1595     from a single perltest script.
1597     4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
1598     by \s. It has now been removed in PCRE. This means it isn't recognized as
1599     whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
1600     class [:space:] *does* include VT, thereby creating a mess.
1602     5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
1603     space and tab.
1605     6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
1606     its new features into the main test script, reducing the number of scripts.
1608     7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
1609     were backward compatible, and made the (?i) apply to the whole pattern, as if
1610     /i were given. Now it behaves more logically, and applies the option setting
1611     only to what follows. PCRE has been changed to follow suit. However, if it
1612     finds options settings right at the start of the pattern, it extracts them into
1613     the global options, as before. Thus, they show up in the info data.
1615     8. Added support for the \Q...\E escape sequence. Characters in between are
1616     treated as literals. This is slightly different from Perl in that $ and @ are
1617     also handled as literals inside the quotes. In Perl, they will cause variable
1618     interpolation. Note the following examples:
1620     Pattern PCRE matches Perl matches
1622     \Qabc$xyz\E abc$xyz abc followed by the contents of $xyz
1623     \Qabc\$xyz\E abc\$xyz abc\$xyz
1624     \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
1626     For compatibility with Perl, \Q...\E sequences are recognized inside character
1627     classes as well as outside them.
1629     9. Re-organized 3 code statements in pcretest to avoid "overflow in
1630     floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
1631     (size_t) cast to one statement in pcretest and one in pcreposix to avoid
1632     signed/unsigned warnings.
1634     10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
1635     option for pcretest, so I've replaced it by a simple function that does just
1636     that job.
1638     11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
1639     "pcregrep -".
1641     12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
1642     Java package. This provides some syntactic sugar for simple cases of what my
1643     documentation calls "once-only subpatterns". A pattern such as x*+ is the same
1644     as (?>x*). In other words, if what is inside (?>...) is just a single repeated
1645     item, you can use this simplified notation. Note that only makes sense with
1646     greedy quantifiers. Consequently, the use of the possessive quantifier forces
1647     greediness, whatever the setting of the PCRE_UNGREEDY option.
1649     13. A change of greediness default within a pattern was not taking effect at
1650     the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
1651     subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
1652     was abstracted outside.
1654     14. PCRE now supports the \G assertion. It is true when the current matching
1655     position is at the start point of the match. This differs from \A when the
1656     starting offset is non-zero. Used with the /g option of pcretest (or similar
1657     code), it works in the same way as it does for Perl's /g option. If all
1658     alternatives of a regex begin with \G, the expression is anchored to the start
1659     match position, and the "anchored" flag is set in the compiled expression.
1661     15. Some bugs concerning the handling of certain option changes within patterns
1662     have been fixed. These applied to options other than (?ims). For example,
1663     "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
1664     been the other way round. Some of this was related to change 7 above.
1666     16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
1667     features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
1668     and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
1669     POSIX classes only within a class (e.g. /[[:alpha:]]/).
1671     17. Added support for Perl's \C escape. This matches one byte, even in UTF8
1672     mode. Unlike ".", it always matches newline, whatever the setting of
1673     PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
1674     assertions. Perl allows it, but it doesn't (in general) work because it can't
1675     calculate the length of the lookbehind. At least, that's the case for Perl
1676     5.8.0 - I've been told they are going to document that it doesn't work in
1677     future.
1679     18. Added an error diagnosis for escapes that PCRE does not support: these are
1680     \L, \l, \N, \P, \p, \U, \u, and \X.
1682     19. Although correctly diagnosing a missing ']' in a character class, PCRE was
1683     reading past the end of the pattern in cases such as /[abcd/.
1685     20. PCRE was getting more memory than necessary for patterns with classes that
1686     contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
1688     21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
1689     compiling PCRE for use with Virtual Pascal.
1691     22. Small fix to the Makefile to make it work properly if the build is done
1692     outside the source tree.
1694     23. Added a new extension: a condition to go with recursion. If a conditional
1695     subpattern starts with (?(R) the "true" branch is used if recursion has
1696     happened, whereas the "false" branch is used only at the top level.
1698     24. When there was a very long string of literal characters (over 255 bytes
1699     without UTF support, over 250 bytes with UTF support), the computation of how
1700     much memory was required could be incorrect, leading to segfaults or other
1701     strange effects.
1703     25. PCRE was incorrectly assuming anchoring (either to start of subject or to
1704     start of line for a non-DOTALL pattern) when a pattern started with (.*) and
1705     there was a subsequent back reference to those brackets. This meant that, for
1706     example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
1707     possible to check for precisely this case. All we can do is abandon the
1708     optimization if .* occurs inside capturing brackets when there are any back
1709     references whatsoever. (See below for a better fix that came later.)
1711     26. The handling of the optimization for finding the first character of a
1712     non-anchored pattern, and for finding a character that is required later in the
1713     match were failing in some cases. This didn't break the matching; it just
1714     failed to optimize when it could. The way this is done has been re-implemented.
1716     27. Fixed typo in error message for invalid (?R item (it said "(?p").
1718     28. Added a new feature that provides some of the functionality that Perl
1719     provides with (?{...}). The facility is termed a "callout". The way it is done
1720     in PCRE is for the caller to provide an optional function, by setting
1721     pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
1722     global variable. By default it is unset, which disables all calling out. To get
1723     the function called, the regex must include (?C) at appropriate points. This
1724     is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
1725     This provides a means of identifying different callout points. When PCRE
1726     reaches such a point in the regex, if pcre_callout has been set, the external
1727     function is called. It is provided with data in a structure called
1728     pcre_callout_block, which is defined in pcre.h. If the function returns 0,
1729     matching continues; if it returns a non-zero value, the match at the current
1730     point fails. However, backtracking will occur if possible. [This was changed
1731     later and other features added - see item 49 below.]
1733     29. pcretest is upgraded to test the callout functionality. It provides a
1734     callout function that displays information. By default, it shows the start of
1735     the match and the current position in the text. There are some new data escapes
1736     to vary what happens:
1738     \C+ in addition, show current contents of captured substrings
1739     \C- do not supply a callout function
1740     \C!n return 1 when callout number n is reached
1741     \C!n!m return 1 when callout number n is reached for the mth time
1743     30. If pcregrep was called with the -l option and just a single file name, it
1744     output "<stdin>" if a match was found, instead of the file name.
1746     31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
1747     slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
1748     pcre_exec(). This saves a malloc/free per call. The default value of
1749     POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
1750     when configuring.
1752     32. The default maximum size of a compiled pattern is 64K. There have been a
1753     few cases of people hitting this limit. The code now uses macros to handle the
1754     storing of links as offsets within the compiled pattern. It defaults to 2-byte
1755     links, but this can be changed to 3 or 4 bytes by --with-link-size when
1756     configuring. Tests 2 and 5 work only with 2-byte links because they output
1757     debugging information about compiled patterns.
1759     33. Internal code re-arrangements:
1761     (a) Moved the debugging function for printing out a compiled regex into
1762     its own source file (printint.c) and used #include to pull it into
1763     pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
1764     separate copies.
1766     (b) Defined the list of op-code names for debugging as a macro in
1767     internal.h so that it is next to the definition of the opcodes.
1769     (c) Defined a table of op-code lengths for simpler skipping along compiled
1770     code. This is again a macro in internal.h so that it is next to the
1771     definition of the opcodes.
1773     34. Added support for recursive calls to individual subpatterns, along the
1774     lines of Robin Houston's patch (but implemented somewhat differently).
1776     35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
1777     allow it to read and process whole directories in Win32. This code was
1778     contributed by Lionel Fourquaux; it has not been tested by me.
1780     36. Added support for named subpatterns. The Python syntax (?P<name>...) is
1781     used to name a group. Names consist of alphanumerics and underscores, and must
1782     be unique. Back references use the syntax (?P=name) and recursive calls use
1783     (?P>name) which is a PCRE extension to the Python extension. Groups still have
1784     numbers. The function pcre_fullinfo() can be used after compilation to extract
1785     a name/number map. There are three relevant calls:
1787     PCRE_INFO_NAMEENTRYSIZE yields the size of each entry in the map
1788     PCRE_INFO_NAMECOUNT yields the number of entries
1789     PCRE_INFO_NAMETABLE yields a pointer to the map.
1791     The map is a vector of fixed-size entries. The size of each entry depends on
1792     the length of the longest name used. The first two bytes of each entry are the
1793     group number, most significant byte first. There follows the corresponding
1794     name, zero terminated. The names are in alphabetical order.
1796     37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
1797     case instead of 255. Making it the same both with and without UTF-8 support
1798     means that the same test output works with both.
1800     38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
1801     calling malloc() with a zero argument.
1803     39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
1804     optimization. I've improved things by keeping a bitmap of backreferences with
1805     numbers 1-31 so that if .* occurs inside capturing brackets that are not in
1806     fact referenced, the optimization can be applied. It is unlikely that a
1807     relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
1808     the match to follow \n) will appear inside brackets with a number greater than
1809     31, but if it does, any back reference > 31 suppresses the optimization.
1811     40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
1812     of disabling numbered capturing parentheses. Any opening parenthesis that is
1813     not followed by ? behaves as if it were followed by ?: but named parentheses
1814     can still be used for capturing (and they will acquire numbers in the usual
1815     way).
1817     41. Redesigned the return codes from the match() function into yes/no/error so
1818     that errors can be passed back from deep inside the nested calls. A malloc
1819     failure while inside a recursive subpattern call now causes the
1820     PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
1822     42. It is now possible to set a limit on the number of times the match()
1823     function is called in a call to pcre_exec(). This facility makes it possible to
1824     limit the amount of recursion and backtracking, though not in a directly
1825     obvious way, because the match() function is used in a number of different
1826     circumstances. The count starts from zero for each position in the subject
1827     string (for non-anchored patterns). The default limit is, for compatibility, a
1828     large number, namely 10 000 000. You can change this in two ways:
1830     (a) When configuring PCRE before making, you can use --with-match-limit=n
1831     to set a default value for the compiled library.
1833     (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
1834     a different value is set. See 45 below.
1836     If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
1838     43. Added a new function pcre_config(int, void *) to enable run-time extraction
1839     of things that can be changed at compile time. The first argument specifies
1840     what is wanted and the second points to where the information is to be placed.
1841     The current list of available information is:
1845     The output is an integer that is set to one if UTF-8 support is available;
1846     otherwise it is set to zero.
1850     The output is an integer that it set to the value of the code that is used for
1851     newline. It is either LF (10) or CR (13).
1855     The output is an integer that contains the number of bytes used for internal
1856     linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
1860     The output is an integer that contains the threshold above which the POSIX
1861     interface uses malloc() for output vectors. See item 31 above.
1865     The output is an unsigned integer that contains the default limit of the number
1866     of match() calls in a pcre_exec() execution. See 42 above.
1868     44. pcretest has been upgraded by the addition of the -C option. This causes it
1869     to extract all the available output from the new pcre_config() function, and to
1870     output it. The program then exits immediately.
1872     45. A need has arisen to pass over additional data with calls to pcre_exec() in
1873     order to support additional features. One way would have been to define
1874     pcre_exec2() (for example) with extra arguments, but this would not have been
1875     extensible, and would also have required all calls to the original function to
1876     be mapped to the new one. Instead, I have chosen to extend the mechanism that
1877     is used for passing in "extra" data from pcre_study().
1879     The pcre_extra structure is now exposed and defined in pcre.h. It currently
1880     contains the following fields:
1882     flags a bitmap indicating which of the following fields are set
1883     study_data opaque data from pcre_study()
1884     match_limit a way of specifying a limit on match() calls for a specific
1885     call to pcre_exec()
1886     callout_data data for callouts (see 49 below)
1888     The flag bits are also defined in pcre.h, and are
1894     The pcre_study() function now returns one of these new pcre_extra blocks, with
1895     the actual study data pointed to by the study_data field, and the
1896     PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
1897     before. That is, this change is entirely upwards-compatible and requires no
1898     change to existing code.
1900     If you want to pass in additional data to pcre_exec(), you can either place it
1901     in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
1902     block.
1904     46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
1905     data string contains the escape sequence \M, pcretest calls pcre_exec() several
1906     times with different match limits, until it finds the minimum value needed for
1907     pcre_exec() to complete. The value is then output. This can be instructive; for
1908     most simple matches the number is quite small, but for pathological cases it
1909     gets very large very quickly.
1911     47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
1912     returns the size of the data block pointed to by the study_data field in a
1913     pcre_extra block, that is, the value that was passed as the argument to
1914     pcre_malloc() when PCRE was getting memory in which to place the information
1915     created by pcre_study(). The fourth argument should point to a size_t variable.
1916     pcretest has been extended so that this information is shown after a successful
1917     pcre_study() call when information about the compiled regex is being displayed.
1919     48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
1920     because what follows is always an absolute path. (Later: it turns out that this
1921     is more than cosmetic for MinGW, because it doesn't like empty path
1922     components.)
1924     49. Some changes have been made to the callout feature (see 28 above):
1926     (i) A callout function now has three choices for what it returns:
1928     0 => success, carry on matching
1929     > 0 => failure at this point, but backtrack if possible
1930     < 0 => serious error, return this value from pcre_exec()
1932     Negative values should normally be chosen from the set of PCRE_ERROR_xxx
1933     values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
1934     "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
1935     use by callout functions. It will never be used by PCRE itself.
1937     (ii) The pcre_extra structure (see 45 above) has a void * field called
1938     callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
1939     pcre_callout_block structure has a field of the same name. The contents of
1940     the field passed in the pcre_extra structure are passed to the callout
1941     function in the corresponding field in the callout block. This makes it
1942     easier to use the same callout-containing regex from multiple threads. For
1943     testing, the pcretest program has a new data escape
1945     \C*n pass the number n (may be negative) as callout_data
1947     If the callout function in pcretest receives a non-zero value as
1948     callout_data, it returns that value.
1950     50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
1951     there were some redundant $(CFLAGS) in commands that are now specified as
1952     $(LINK), which already includes $(CFLAGS).
1954     51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
1955     has been compiled with UTF-8 support *and* pcre_compile() has been compiled
1956     with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
1957     one-byte characters throughout. Note that case-insensitive matching applies
1958     only to characters whose values are less than 256. PCRE doesn't support the
1959     notion of cases for higher-valued characters.
1961     (i) A character class whose characters are all within 0-255 is handled as
1962     a bit map, and the map is inverted for negative classes. Previously, a
1963     character > 255 always failed to match such a class; however it should
1964     match if the class was a negative one (e.g. [^ab]). This has been fixed.
1966     (ii) A negated character class with a single character < 255 is coded as
1967     "not this character" (OP_NOT). This wasn't working properly when the test
1968     character was multibyte, either singly or repeated.
1970     (iii) Repeats of multibyte characters are now handled correctly in UTF-8
1971     mode, for example: \x{100}{2,3}.
1973     (iv) The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
1974     singly or repeated) now correctly test multibyte characters. However,
1975     PCRE doesn't recognize any characters with values greater than 255 as
1976     digits, spaces, or word characters. Such characters always match \D, \S,
1977     and \W, and never match \d, \s, or \w.
1979     (v) Classes may now contain characters and character ranges with values
1980     greater than 255. For example: [ab\x{100}-\x{400}].
1982     (vi) pcregrep now has a --utf-8 option (synonym -u) which makes it call
1983     PCRE in UTF-8 mode.
1985     52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
1986     PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
1987     retained for backwards compatibility. (Note that LASTLITERAL is also a byte
1988     value.)
1990     53. The single man page has become too large. I have therefore split it up into
1991     a number of separate man pages. These also give rise to individual HTML pages;
1992     these are now put in a separate directory, and there is an index.html page that
1993     lists them all. Some hyperlinking between the pages has been installed.
1995     54. Added convenience functions for handling named capturing parentheses.
1997     55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
1998     aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
1999     true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2000     are faulted.
2002     56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2003     calling configure. These values are used when compiling the dftables.c program
2004     which is run to generate the source of the default character tables. They
2005     default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2006     you will need to set these values.
2008     57. Updated the building process for Windows DLL, as provided by Fred Cox.
2011     Version 3.9 02-Jan-02
2012 nigel 61 ---------------------
2014     1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2016     2. If --disable-static was given, the building process failed when trying to
2017     build pcretest and pcregrep. (For some reason it was using libtool to compile
2018     them, which is not right, as they aren't part of the library.)
2021 nigel 59 Version 3.8 18-Dec-01
2022     ---------------------
2024     1. The experimental UTF-8 code was completely screwed up. It was packing the
2025     bytes in the wrong order. How dumb can you get?
2028 nigel 57 Version 3.7 29-Oct-01
2029     ---------------------
2031     1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2032     This caused pcretest, when used on the test data, to segfault. Unfortunately,
2033     this didn't happen under Solaris 8, where I normally test things.
2035 nigel 59 2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2036     doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2037     isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2038     this fix an hour or so after the initial 3.7 release.)
2039 nigel 57
2040 nigel 59
2041 nigel 55 Version 3.6 23-Oct-01
2042     ---------------------
2044     1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2045     offsets passed as NULL with zero offset count.
2047     2. The config.guess and config.sub files had not been updated when I moved to
2048     the latest autoconf.
2051 nigel 53 Version 3.5 15-Aug-01
2052     ---------------------
2053 nigel 5
2054 nigel 53 1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2055     had been forgotten.
2057     2. By using declared but undefined structures, we can avoid using "void"
2058     definitions in pcre.h while keeping the internal definitions of the structures
2059     private.
2061     3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2062     user point of view, this means that both static and shared libraries are built
2063     by default, but this can be individually controlled. More of the work of
2064     handling this static/shared cases is now inside libtool instead of PCRE's make
2065     file.
2067     4. The pcretest utility is now installed along with pcregrep because it is
2068     useful for users (to test regexs) and by doing this, it automatically gets
2069     relinked by libtool. The documentation has been turned into a man page, so
2070     there are now .1, .txt, and .html versions in /doc.
2072     5. Upgrades to pcregrep:
2073     (i) Added long-form option names like gnu grep.
2074     (ii) Added --help to list all options with an explanatory phrase.
2075     (iii) Added -r, --recursive to recurse into sub-directories.
2076     (iv) Added -f, --file to read patterns from a file.
2078     6. pcre_exec() was referring to its "code" argument before testing that
2079     argument for NULL (and giving an error if it was NULL).
2081     7. Upgraded Makefile.in to allow for compiling in a different directory from
2082     the source directory.
2084     8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2085     options bits, the pointer it was passed was to an int instead of to an unsigned
2086     long int. This mattered only on 64-bit systems.
2088     9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2089     generated) instead of pcre.in, which it its source. Also made the same change
2090     in several of the .c files.
2092     10. A new release of gcc defines printf() as a macro, which broke pcretest
2093     because it had an ifdef in the middle of a string argument for printf(). Fixed
2094     by using separate calls to printf().
2096     11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2097     script, to force use of CR or LF instead of \n in the source. On non-Unix
2098     systems, the value can be set in config.h.
2100     12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2101     absolute limit. Changed the text of the error message to make this clear, and
2102     likewise updated the man page.
2104     13. The limit of 99 on the number of capturing subpatterns has been removed.
2105     The new limit is 65535, which I hope will not be a "real" limit.
2108 nigel 51 Version 3.4 22-Aug-00
2109     ---------------------
2111     1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2113     2. Diagnose condition (?(0) as an error instead of crashing on matching.
2116 nigel 49 Version 3.3 01-Aug-00
2117     ---------------------
2119     1. If an octal character was given, but the value was greater than \377, it
2120     was not getting masked to the least significant bits, as documented. This could
2121     lead to crashes in some systems.
2123     2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
2124     the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
2126     3. Added the functions pcre_free_substring() and pcre_free_substring_list().
2127     These just pass their arguments on to (pcre_free)(), but they are provided
2128     because some uses of PCRE bind it to non-C systems that can call its functions,
2129     but cannot call free() or pcre_free() directly.
2131     4. Add "make test" as a synonym for "make check". Corrected some comments in
2132     the Makefile.
2134     5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
2135     Makefile.
2137     6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
2138     command called pgrep for grepping around the active processes.
2140     7. Added the beginnings of support for UTF-8 character strings.
2142     8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
2143     RANLIB to ./ltconfig so that they are used by libtool. I think these are all
2144     the relevant ones. (AR is not passed because ./ltconfig does its own figuring
2145     out for the ar command.)
2148 nigel 47 Version 3.2 12-May-00
2149     ---------------------
2151     This is purely a bug fixing release.
2153     1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
2154     of ZA. This was just one example of several cases that could provoke this bug,
2155     which was introduced by change 9 of version 2.00. The code for breaking
2156     infinite loops after an iteration that matches an empty string was't working
2157     correctly.
2159     2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
2160     when matched against abbab (for example). After matching an empty string, it
2161     wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
2162     caused it to match further down the string than it should.
2164     3. The code contained an inclusion of sys/types.h. It isn't clear why this
2165     was there because it doesn't seem to be needed, and it causes trouble on some
2166     systems, as it is not a Standard C header. It has been removed.
2168     4. Made 4 silly changes to the source to avoid stupid compiler warnings that
2169     were reported on the Macintosh. The changes were from
2171     while ((c = *(++ptr)) != 0 && c != '\n');
2172     to
2173     while ((c = *(++ptr)) != 0 && c != '\n') ;
2175     Totally extraordinary, but if that's what it takes...
2177     5. PCRE is being used in one environment where neither memmove() nor bcopy() is
2178     available. Added HAVE_BCOPY and an autoconf test for it; if neither
2179     HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
2180     assumes the way PCRE uses memmove() (always moving upwards).
2182     6. PCRE is being used in one environment where strchr() is not available. There
2183     was only one use in pcre.c, and writing it out to avoid strchr() probably gives
2184     faster code anyway.
2187 nigel 45 Version 3.1 09-Feb-00
2188     ---------------------
2190     The only change in this release is the fixing of some bugs in Makefile.in for
2191     the "install" target:
2193     (1) It was failing to install pcreposix.h.
2195     (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
2198 nigel 43 Version 3.0 01-Feb-00
2199     ---------------------
2200 nigel 41
2201     1. Add support for the /+ modifier to perltest (to output $` like it does in
2202     pcretest).
2204     2. Add support for the /g modifier to perltest.
2206     3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
2207     matches null strings.
2209     4. Fix perltest so that it doesn't do unwanted things when fed an empty
2210     pattern. Perl treats empty patterns specially - it reuses the most recent
2211     pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
2212     effect.
2214     5. The POSIX interface was broken in that it was just handing over the POSIX
2215     captured string vector to pcre_exec(), but (since release 2.00) PCRE has
2216     required a bigger vector, with some working space on the end. This means that
2217     the POSIX wrapper now has to get and free some memory, and copy the results.
2219 nigel 43 6. Added some simple autoconf support, placing the test data and the
2220     documentation in separate directories, re-organizing some of the
2221     information files, and making it build pcre-config (a GNU standard). Also added
2222     libtool support for building PCRE as a shared library, which is now the
2223     default.
2224 nigel 41
2225 nigel 43 7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
2226     09 are not valid octal constants. Single digits will be used for minor values
2227     less than 10.
2229     8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
2230     existing programs that set these in the POSIX interface can use PCRE without
2231     modification.
2233     9. Added a new function, pcre_fullinfo() with an extensible interface. It can
2234     return all that pcre_info() returns, plus additional data. The pcre_info()
2235     function is retained for compatibility, but is considered to be obsolete.
2237     10. Added experimental recursion feature (?R) to handle one common case that
2238     Perl 5.6 will be able to do with (?p{...}).
2240     11. Added support for POSIX character classes like [:alpha:], which Perl is
2241     adopting.
2244 nigel 39 Version 2.08 31-Aug-99
2245     ----------------------
2247     1. When startoffset was not zero and the pattern began with ".*", PCRE was not
2248     trying to match at the startoffset position, but instead was moving forward to
2249     the next newline as if a previous match had failed.
2251     2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
2252     and could get into a loop if a null string was matched other than at the start
2253     of the subject.
2255     3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
2256     be distinguished at compile time, and for completeness also added PCRE_DATE.
2258     5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
2259     in GnuWin32 environments.
2262 nigel 37 Version 2.07 29-Jul-99
2263     ----------------------
2265     1. The documentation is now supplied in plain text form and HTML as well as in
2266     the form of man page sources.
2268     2. C++ compilers don't like assigning (void *) values to other pointer types.
2269     In particular this affects malloc(). Although there is no problem in Standard
2270     C, I've put in casts to keep C++ compilers happy.
2272     3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
2273     should be (const char *).
2275     4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
2276     be useful for non-Unix systems who don't want to bother with the POSIX stuff.
2277     However, I haven't made this a standard facility. The documentation doesn't
2278     mention it, and the Makefile doesn't support it.
2280     5. The Makefile now contains an "install" target, with editable destinations at
2281     the top of the file. The pcretest program is not installed.
2283     6. pgrep -V now gives the PCRE version number and date.
2285     7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
2286     causing the entire string to be ignored, instead of just the last character.
2288     8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
2289     non-matching string, it can take a very, very long time, even for strings of
2290     quite modest length, because of the nested recursion. PCRE now does better in
2291     some of these cases. It does this by remembering the last required literal
2292     character in the pattern, and pre-searching the subject to ensure it is present
2293     before running the real match. In other words, it applies a heuristic to detect
2294     some types of certain failure quickly, and in the above example, if presented
2295     with a string that has no trailing " it gives "no match" very quickly.
2297     9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
2298     other alternatives are tried instead.
2301 nigel 35 Version 2.06 09-Jun-99
2302     ----------------------
2304     1. Change pcretest's output for amount of store used to show just the code
2305     space, because the remainder (the data block) varies in size between 32-bit and
2306     64-bit systems.
2308     2. Added an extra argument to pcre_exec() to supply an offset in the subject to
2309     start matching at. This allows lookbehinds to work when searching for multiple
2310     occurrences in a string.
2312     3. Added additional options to pcretest for testing multiple occurrences:
2314     /+ outputs the rest of the string that follows a match
2315     /g loops for multiple occurrences, using the new startoffset argument
2316     /G loops for multiple occurrences by passing an incremented pointer
2318     4. PCRE wasn't doing the "first character" optimization for patterns starting
2319     with \b or \B, though it was doing it for other lookbehind assertions. That is,
2320     it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
2321     the letter 'x'. On long subject strings, this gives a significant speed-up.
2324 nigel 33 Version 2.05 21-Apr-99
2325     ----------------------
2327     1. Changed the type of magic_number from int to long int so that it works
2328     properly on 16-bit systems.
2330     2. Fixed a bug which caused patterns starting with .* not to work correctly
2331     when the subject string contained newline characters. PCRE was assuming
2332     anchoring for such patterns in all cases, which is not correct because .* will
2333     not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
2334     DOTALL is set at top level; otherwise it knows that patterns starting with .*
2335     must be retried after every newline in the subject.
2338 nigel 31 Version 2.04 18-Feb-99
2339     ----------------------
2341     1. For parenthesized subpatterns with repeats whose minimum was zero, the
2342     computation of the store needed to hold the pattern was incorrect (too large).
2343     If such patterns were nested a few deep, this could multiply and become a real
2344     problem.
2346     2. Added /M option to pcretest to show the memory requirement of a specific
2347     pattern. Made -m a synonym of -s (which does this globally) for compatibility.
2349     3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
2350     compiled in such a way that the backtracking after subsequent failure was
2351     pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
2352     ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
2355 nigel 29 Version 2.03 02-Feb-99
2356     ----------------------
2358     1. Fixed typo and small mistake in man page.
2360 nigel 31 2. Added 4th condition (GPL supersedes if conflict) and created separate
2361     LICENCE file containing the conditions.
2362 nigel 29
2363     3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
2364     Perl, that is the internal \ allows the delimiter to be included in the
2365     pattern. Locked out the use of \ as a delimiter. If \ immediately follows
2366     the final delimiter, add \ to the end of the pattern (to test the error).
2368     4. Added the convenience functions for extracting substrings after a successful
2369     match. Updated pcretest to make it able to test these functions.
2372 nigel 27 Version 2.02 14-Jan-99
2373     ----------------------
2375     1. Initialized the working variables associated with each extraction so that
2376     their saving and restoring doesn't refer to uninitialized store.
2378     2. Put dummy code into study.c in order to trick the optimizer of the IBM C
2379     compiler for OS/2 into generating correct code. Apparently IBM isn't going to
2380     fix the problem.
2382     3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
2383     calls, and wasn't printing the correct value for compiling calls. Increased the
2384     default value of LOOPREPEAT, and the number of significant figures in the
2385     times.
2387     4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
2389     5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
2390     a building problem on Windows NT with a FAT file system.
2393 nigel 25 Version 2.01 21-Oct-98
2394     ----------------------
2396     1. Changed the API for pcre_compile() to allow for the provision of a pointer
2397     to character tables built by pcre_maketables() in the current locale. If NULL
2398     is passed, the default tables are used.
2401 nigel 23 Version 2.00 24-Sep-98
2402 nigel 21 ----------------------
2404 nigel 23 1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
2405     it any more.
2407     2. Allow quantification of (?>) groups, and make it work correctly.
2409     3. The first character computation wasn't working for (?>) groups.
2411     4. Correct the implementation of \Z (it is permitted to match on the \n at the
2412     end of the subject) and add 5.005's \z, which really does match only at the
2413     very end of the subject.
2415     5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
2417     6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
2418     DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
2419     localized options. All options to pcre_study() were also removed.
2421     7. Add other new features from 5.005:
2423     $(?<= positive lookbehind
2424     $(?<! negative lookbehind
2425     (?imsx-imsx) added the unsetting capability
2426     such a setting is global if at outer level; local otherwise
2427     (?imsx-imsx:) non-capturing groups with option setting
2428     (?(cond)re|re) conditional pattern matching
2430     A backreference to itself in a repeated group matches the previous
2431     captured string.
2433     8. General tidying up of studying (both automatic and via "study")
2434     consequential on the addition of new assertions.
2436     9. As in 5.005, unlimited repeated groups that could match an empty substring
2437     are no longer faulted at compile time. Instead, the loop is forcibly broken at
2438     runtime if any iteration does actually match an empty substring.
2440     10. Include the RunTest script in the distribution.
2442     11. Added tests from the Perl 5.005_02 distribution. This showed up a few
2443     discrepancies, some of which were old and were also with respect to 5.004. They
2444     have now been fixed.
2447     Version 1.09 28-Apr-98
2448     ----------------------
2450 nigel 21 1. A negated single character class followed by a quantifier with a minimum
2451     value of one (e.g. [^x]{1,6} ) was not compiled correctly. This could lead to
2452     program crashes, or just wrong answers. This did not apply to negated classes
2453     containing more than one character, or to minima other than one.
2456 nigel 19 Version 1.08 27-Mar-98
2457     ----------------------
2459     1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
2461     2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
2462     latter must appear before anything that relies on it in the pattern.
2465 nigel 17 Version 1.07 16-Feb-98
2466     ----------------------
2468     1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
2469     repeat of a potentially empty string).
2472 nigel 15 Version 1.06 23-Jan-98
2473     ----------------------
2475     1. Added Markus Oberhumer's little patches for C++.
2477     2. Literal strings longer than 255 characters were broken.
2480 nigel 13 Version 1.05 23-Dec-97
2481     ----------------------
2483     1. Negated character classes containing more than one character were failing if
2484     PCRE_CASELESS was set at run time.
2487 nigel 11 Version 1.04 19-Dec-97
2488     ----------------------
2490     1. Corrected the man page, where some "const" qualifiers had been omitted.
2492     2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
2493     input syntax.
2495     3. Fixed memory leak which occurred when a regex with back references was
2496     matched with an offsets vector that wasn't big enough. The temporary memory
2497     that is used in this case wasn't being freed if the match failed.
2499     4. Tidied pcretest to ensure it frees memory that it gets.
2501     5. Temporary memory was being obtained in the case where the passed offsets
2502     vector was exactly big enough.
2504     6. Corrected definition of offsetof() from change 5 below.
2506     7. I had screwed up change 6 below and broken the rules for the use of
2507     setjmp(). Now fixed.
2510 nigel 9 Version 1.03 18-Dec-97
2511     ----------------------
2513     1. A erroneous regex with a missing opening parenthesis was correctly
2514     diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
2515     on some systems.
2517     2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
2518     it was reported that one broken compiler failed on the former because "code" is
2519     also an independent variable.
2521     3. The erroneous regex a[]b caused an array overrun reference.
2523     4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
2524     fail on data ending with that character. (It was going on too far, and checking
2525     the next character, typically a binary zero.) This was specific to the
2526     optimized code for single-character negative classes.
2528     5. Added a contributed patch from the TIN world which does the following:
2530     + Add an undef for memmove, in case the the system defines a macro for it.
2532     + Add a definition of offsetof(), in case there isn't one. (I don't know
2533     the reason behind this - offsetof() is part of the ANSI standard - but
2534     it does no harm).
2536     + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
2537     most of the places where whitespace preceded '#'. I have given up and
2538     allowed the remaining 2 cases to be at the margin.
2540     + Rename some variables in pcre to eliminate shadowing. This seems very
2541     pedantic, but does no harm, of course.
2543     6. Moved the call to setjmp() into its own function, to get rid of warnings
2544     from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
2546     7. Constructs such as \d{8,} were compiling into the equivalent of
2547 nigel 11 \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
2548 nigel 9 outcome, but in this particular case used more store than had been allocated,
2549     which caused the bug to be discovered because it threw up an internal error.
2551     8. The debugging code in both pcre and pcretest for outputting the compiled
2552     form of a regex was going wrong in the case of back references followed by
2553     curly-bracketed repeats.
2556 nigel 7 Version 1.02 12-Dec-97
2557     ----------------------
2559     1. Typos in pcre.3 and comments in the source fixed.
2561     2. Applied a contributed patch to get rid of places where it used to remove
2562     'const' from variables, and fixed some signed/unsigned and uninitialized
2563     variable warnings.
2565     3. Added the "runtest" target to Makefile.
2567     4. Set default compiler flag to -O2 rather than just -O.
2570 nigel 5 Version 1.01 19-Nov-97
2571     ----------------------
2573     1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
2574     like /([ab]*)*/, that is, for classes with more than one character in them.
2576     2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
2577     as /((?>a*))*/ (a PCRE_EXTRA facility).
2580     Version 1.00 18-Nov-97
2581     ----------------------
2583     1. Added compile-time macros to support systems such as SunOS4 which don't have
2584     memmove() or strerror() but have other things that can be used instead.
2586     2. Arranged that "make clean" removes the executables.
2589 nigel 3 Version 0.99 27-Oct-97
2590     ----------------------
2592     1. Fixed bug in code for optimizing classes with only one character. It was
2593     initializing a 32-byte map regardless, which could cause it to run off the end
2594     of the memory it had got.
2596     2. Added, conditional on PCRE_EXTRA, the proposed (?>REGEX) construction.
2599     Version 0.98 22-Oct-97
2600     ----------------------
2602     1. Fixed bug in code for handling temporary memory usage when there are more
2603     back references than supplied space in the ovector. This could cause segfaults.
2606     Version 0.97 21-Oct-97
2607     ----------------------
2609     1. Added the \X "cut" facility, conditional on PCRE_EXTRA.
2611     2. Optimized negated single characters not to use a bit map.
2613     3. Brought error texts together as macro definitions; clarified some of them;
2614     fixed one that was wrong - it said "range out of order" when it meant "invalid
2615     escape sequence".
2617     4. Changed some char * arguments to const char *.
2619     5. Added PCRE_NOTBOL and PCRE_NOTEOL (from POSIX).
2621     6. Added the POSIX-style API wrapper in pcreposix.a and testing facilities in
2622     pcretest.
2625     Version 0.96 16-Oct-97
2626     ----------------------
2628     1. Added a simple "pgrep" utility to the distribution.
2630     2. Fixed an incompatibility with Perl: "{" is now treated as a normal character
2631     unless it appears in one of the precise forms "{ddd}", "{ddd,}", or "{ddd,ddd}"
2632     where "ddd" means "one or more decimal digits".
2634     3. Fixed serious bug. If a pattern had a back reference, but the call to
2635     pcre_exec() didn't supply a large enough ovector to record the related
2636     identifying subpattern, the match always failed. PCRE now remembers the number
2637     of the largest back reference, and gets some temporary memory in which to save
2638     the offsets during matching if necessary, in order to ensure that
2639     backreferences always work.
2641     4. Increased the compatibility with Perl in a number of ways:
2643     (a) . no longer matches \n by default; an option PCRE_DOTALL is provided
2644     to request this handling. The option can be set at compile or exec time.
2646     (b) $ matches before a terminating newline by default; an option
2647     PCRE_DOLLAR_ENDONLY is provided to override this (but not in multiline
2648     mode). The option can be set at compile or exec time.
2650     (c) The handling of \ followed by a digit other than 0 is now supposed to be
2651     the same as Perl's. If the decimal number it represents is less than 10
2652     or there aren't that many previous left capturing parentheses, an octal
2653     escape is read. Inside a character class, it's always an octal escape,
2654     even if it is a single digit.
2656     (d) An escaped but undefined alphabetic character is taken as a literal,
2657     unless PCRE_EXTRA is set. Currently this just reserves the remaining
2658     escapes.
2660     (e) {0} is now permitted. (The previous item is removed from the compiled
2661     pattern).
2663     5. Changed all the names of code files so that the basic parts are no longer
2664     than 10 characters, and abolished the teeny "globals.c" file.
2666     6. Changed the handling of character classes; they are now done with a 32-byte
2667     bit map always.
2669     7. Added the -d and /D options to pcretest to make it possible to look at the
2670     internals of compilation without having to recompile pcre.
2673     Version 0.95 23-Sep-97
2674     ----------------------
2676     1. Fixed bug in pre-pass concerning escaped "normal" characters such as \x5c or
2677     \x20 at the start of a run of normal characters. These were being treated as
2678     real characters, instead of the source characters being re-checked.
2681     Version 0.94 18-Sep-97
2682     ----------------------
2684     1. The functions are now thread-safe, with the caveat that the global variables
2685     containing pointers to malloc() and free() or alternative functions are the
2686     same for all threads.
2688     2. Get pcre_study() to generate a bitmap of initial characters for non-
2689     anchored patterns when this is possible, and use it if passed to pcre_exec().
2692     Version 0.93 15-Sep-97
2693     ----------------------
2695     1. /(b)|(:+)/ was computing an incorrect first character.
2697     2. Add pcre_study() to the API and the passing of pcre_extra to pcre_exec(),
2698     but not actually doing anything yet.
2700     3. Treat "-" characters in classes that cannot be part of ranges as literals,
2701     as Perl does (e.g. [-az] or [az-]).
2703     4. Set the anchored flag if a branch starts with .* or .*? because that tests
2704     all possible positions.
2706     5. Split up into different modules to avoid including unneeded functions in a
2707     compiled binary. However, compile and exec are still in one module. The "study"
2708     function is split off.
2710     6. The character tables are now in a separate module whose source is generated
2711     by an auxiliary program - but can then be edited by hand if required. There are
2712     now no calls to isalnum(), isspace(), isdigit(), isxdigit(), tolower() or
2713     toupper() in the code.
2715     7. Turn the malloc/free funtions variables into pcre_malloc and pcre_free and
2716     make them global. Abolish the function for setting them, as the caller can now
2717     set them directly.
2720     Version 0.92 11-Sep-97
2721     ----------------------
2723     1. A repeat with a fixed maximum and a minimum of 1 for an ordinary character
2724     (e.g. /a{1,3}/) was broken (I mis-optimized it).
2726     2. Caseless matching was not working in character classes if the characters in
2727     the pattern were in upper case.
2729     3. Make ranges like [W-c] work in the same way as Perl for caseless matching.
2731     4. Make PCRE_ANCHORED public and accept as a compile option.
2733     5. Add an options word to pcre_exec() and accept PCRE_ANCHORED and
2734     PCRE_CASELESS at run time. Add escapes \A and \I to pcretest to cause it to
2735     pass them.
2737     6. Give an error if bad option bits passed at compile or run time.
2739     7. Add PCRE_MULTILINE at compile and exec time, and (?m) as well. Add \M to
2740     pcretest to cause it to pass that flag.
2742     8. Add pcre_info(), to get the number of identifying subpatterns, the stored
2743     options, and the first character, if set.
2745     9. Recognize C+ or C{n,m} where n >= 1 as providing a fixed starting character.
2748     Version 0.91 10-Sep-97
2749     ----------------------
2751     1. PCRE was failing to diagnose unlimited repeats of subpatterns that could
2752     match the empty string as in /(a*)*/. It was looping and ultimately crashing.
2754     2. PCRE was looping on encountering an indefinitely repeated back reference to
2755     a subpattern that had matched an empty string, e.g. /(a|)\1*/. It now does what
2756     Perl does - treats the match as successful.
2758     ****


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

ViewVC Help
Powered by ViewVC 1.1.12