ViewVC logotype

Diff of /code/trunk/ChangeLog

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

revision 27 by nigel, Sat Feb 24 21:38:49 2007 UTC revision 176 by ph10, Mon Jun 11 13:48:37 2007 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
4    Version 7.2 05-June-07
5    ---------------------
7     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
8        which is apparently normally available under Windows.
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.
13     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
15     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        "Z" feature of pcretest means that these can be cut out, making the tests
18        usable with all link sizes.
20     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        a small improvement under Linux. However, it saves one field in the frame
23        in all cases.
25     6. Added more features from the forthcoming Perl 5.10:
27        (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.
30        (b) (?+n) is also a relative subroutine call; it refers to the nth next
31            to be opened parentheses.
33        (c) Conditions that refer to capturing parentheses can be specified
34            relatively, for example, (?(-2)... or (?(+3)...
36        (d) \K resets the start of the current match so that everything before
37            is not part of it.
39        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
41        (f) \g{name} is another synonym - part of Perl 5.10's unification of
42            reference syntax.
44        (g) (?| introduces a group in which the numbering of parentheses in each
45            alternative starts with the same number.
47     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
50     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
51        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
52        for detecting groups that can match an empty string.
54     9. A pattern with a very large number of alternatives (more than several
55        hundred) was running out of internal workspace during the pre-compile
56        phase, where pcre_compile() figures out how much memory will be needed. A
57        bit of new cunning has reduced the workspace needed for groups with
58        alternatives. The 1000-alternative test pattern now uses 12 bytes of
59        workspace instead of running out of the 4096 that are available.
61    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
64    Version 7.1 24-Apr-07
65    ---------------------
67     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
68        that is more "standard", making use of automake and other Autotools. There
69        is some re-arrangement of the files and adjustment of comments consequent
70        on this.
72     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
73        for recursive directory scanning broke on some systems because the files
74        are not scanned in any specific order and on different systems the order
75        was different. A call to "sort" has been inserted into RunGrepTest for the
76        approprate test as a short-term fix. In the longer term there may be an
77        alternative.
79     3. I had an email from Eric Raymond about problems translating some of PCRE's
80        man pages to HTML (despite the fact that I distribute HTML pages, some
81        people do their own conversions for various reasons). The problems
82        concerned the use of low-level troff macros .br and .in. I have therefore
83        removed all such uses from the man pages (some were redundant, some could
84        be replaced by .nf/.fi pairs). The 132html script that I use to generate
85        HTML has been updated to handle .nf/.fi and to complain if it encounters
86        .br or .in.
88     4. Updated comments in configure.ac that get placed in config.h.in and also
89        arranged for config.h to be included in the distribution, with the name
90        config.h.generic, for the benefit of those who have to compile without
91        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
93     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
94        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
95        makevp.bat for the current PCRE, using the additional files
96        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
98     6. A Windows user reported a minor discrepancy with test 2, which turned out
99        to be caused by a trailing space on an input line that had got lost in his
100        copy. The trailing space was an accident, so I've just removed it.
102     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
103        that is needed.
105     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
106        as "const" (a) because they are and (b) because it helps the PHP
107        maintainers who have recently made a script to detect big data structures
108        in the php code that should be moved to the .rodata section. I remembered
109        to update Builducptable as well, so it won't revert if ucptable.h is ever
110        re-created.
112     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
113        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
114        order to be able to cut out the UTF-8 tables in the latter when UTF-8
115        support is not required. This saves 1.5-2K of code, which is important in
116        some applications.
118        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
119        so as not to refer to the tables, even though these functions will never be
120        called when UTF-8 support is disabled. Otherwise there are problems with a
121        shared library.
123    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
125        (a) It was defining its arguments as char * instead of void *.
127        (b) It was assuming that all moves were upwards in memory; this was true
128            a long time ago when I wrote it, but is no longer the case.
130        The emulated memove() is provided for those environments that have neither
131        memmove() nor bcopy(). I didn't think anyone used it these days, but that
132        is clearly not the case, as these two bugs were recently reported.
134    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
135        and Detrail to create the HTML documentation, the .txt form of the man
136        pages, and it removes trailing spaces from listed files. It also creates
137        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
138        case, it wraps all the #defines with #ifndefs. This script should be run
139        before "make dist".
141    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
142        with Unicode property support.
144        (a) For a maximizing quantifier, if the two different cases of the
145            character were of different lengths in their UTF-8 codings (there are
146            some cases like this - I found 11), and the matching function had to
147            back up over a mixture of the two cases, it incorrectly assumed they
148            were both the same length.
150        (b) When PCRE was configured to use the heap rather than the stack for
151            recursion during matching, it was not correctly preserving the data for
152            the other case of a UTF-8 character when checking ahead for a match
153            while processing a minimizing repeat. If the check also involved
154            matching a wide character, but failed, corruption could cause an
155            erroneous result when trying to check for a repeat of the original
156            character.
158    13. Some tidying changes to the testing mechanism:
160        (a) The RunTest script now detects the internal link size and whether there
161            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
162            values substituted by "configure". (The RunGrepTest script already did
163            this for UTF-8.) The configure.ac script no longer substitutes the
164            relevant variables.
166        (b) The debugging options /B and /D in pcretest show the compiled bytecode
167            with length and offset values. This means that the output is different
168            for different internal link sizes. Test 2 is skipped for link sizes
169            other than 2 because of this, bypassing the problem. Unfortunately,
170            there was also a test in test 3 (the locale tests) that used /B and
171            failed for link sizes other than 2. Rather than cut the whole test out,
172            I have added a new /Z option to pcretest that replaces the length and
173            offset values with spaces. This is now used to make test 3 independent
174            of link size. (Test 2 will be tidied up later.)
176    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
177        segmentation fault instead of returning the appropriate error message.
179    15. In multiline mode when the newline sequence was set to "any", the pattern
180        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
181        This doesn't seem right; it now treats the CRLF combination as the line
182        ending, and so does not match in that case. It's only a pattern such as ^$
183        that would hit this one: something like ^ABC$ would have failed after \r
184        and then tried again after \r\n.
186    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
187        in an attempt to make files that differ only in their line terminators
188        compare equal. This works on Linux.
190    17. Under certain error circumstances pcregrep might try to free random memory
191        as it exited. This is now fixed, thanks to valgrind.
193    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
194        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
195        was because its rules for how to advance for /g after matching an empty
196        string at the end of a line did not allow for this case. They now check for
197        it specially.
199    20. pcretest is supposed to handle patterns and data of any length, by
200        extending its buffers when necessary. It was getting this wrong when the
201        buffer for a data line had to be extended.
203    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
204        CRLF as a newline sequence.
206    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
207        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
208        I have nevertheless tidied it up.
210    23. Added some casts to kill warnings from HP-UX ia64 compiler.
212    24. Added a man page for pcre-config.
215    Version 7.0 19-Dec-06
216    ---------------------
218     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
219        moving to gcc 4.1.1.
221     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
222        sys/time.h, which is documented as needed for this function. It doesn't
223        seem to matter on Linux, but it showed up on some releases of OS X.
225     3. It seems that there are systems where bytes whose values are greater than
226        127 match isprint() in the "C" locale. The "C" locale should be the
227        default when a C program starts up. In most systems, only ASCII printing
228        characters match isprint(). This difference caused the output from pcretest
229        to vary, making some of the tests fail. I have changed pcretest so that:
231        (a) When it is outputting text in the compiled version of a pattern, bytes
232            other than 32-126 are always shown as hex escapes.
234        (b) When it is outputting text that is a matched part of a subject string,
235            it does the same, unless a different locale has been set for the match
236            (using the /L modifier). In this case, it uses isprint() to decide.
238     4. Fixed a major bug that caused incorrect computation of the amount of memory
239        required for a compiled pattern when options that changed within the
240        pattern affected the logic of the preliminary scan that determines the
241        length. The relevant options are -x, and -i in UTF-8 mode. The result was
242        that the computed length was too small. The symptoms of this bug were
243        either the PCRE error "internal error: code overflow" from pcre_compile(),
244        or a glibc crash with a message such as "pcretest: free(): invalid next
245        size (fast)". Examples of patterns that provoked this bug (shown in
246        pcretest format) are:
248          /(?-x: )/x
249          /(?x)(?-x: \s*#\s*)/
250          /((?i)[\x{c0}])/8
251          /(?i:[\x{c0}])/8
253        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
254        is now done differently.
256     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
257        wrapper classes; (b) implement a new function in the C++ scanner that is
258        more efficient than the old way of doing things because it avoids levels of
259        recursion in the regex matching; (c) add a paragraph to the documentation
260        for the FullMatch() function.
262     6. The escape sequence \n was being treated as whatever was defined as
263        "newline". Not only was this contrary to the documentation, which states
264        that \n is character 10 (hex 0A), but it also went horribly wrong when
265        "newline" was defined as CRLF. This has been fixed.
267     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
268        was being set to -1 for the "end of line" case (supposedly a value that no
269        character can have). Though this value is never used (the check for end of
270        line is "zero bytes in current character"), it caused compiler complaints.
271        I've changed it to 0xffffffff.
273     8. In pcre_version.c, the version string was being built by a sequence of
274        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
275        string (as it is for production releases) called a macro with an empty
276        argument. The C standard says the result of this is undefined. The gcc
277        compiler treats it as an empty string (which was what was wanted) but it is
278        reported that Visual C gives an error. The source has been hacked around to
279        avoid this problem.
281     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
282        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
283        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
284        of them did).
286    10. Originally, pcretest opened its input and output without "b"; then I was
287        told that "b" was needed in some environments, so it was added for release
288        5.0 to both the input and output. (It makes no difference on Unix-like
289        systems.) Later I was told that it is wrong for the input on Windows. I've
290        now abstracted the modes into two macros, to make it easier to fiddle with
291        them, and removed "b" from the input mode under Windows.
293    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
295    12. Added -help and --help to pcretest as an official way of being reminded
296        of the options.
298    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
299        and pcrecpp.cc because they annoy compilers at high warning levels.
301    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
303    15. Fixed an occurrence of == in configure.ac that should have been = (shell
304        scripts are not C programs :-) and which was not noticed because it works
305        on Linux.
307    16. pcretest is supposed to handle any length of pattern and data line (as one
308        line or as a continued sequence of lines) by extending its input buffer if
309        necessary. This feature was broken for very long pattern lines, leading to
310        a string of junk being passed to pcre_compile() if the pattern was longer
311        than about 50K.
313    17. I have done a major re-factoring of the way pcre_compile() computes the
314        amount of memory needed for a compiled pattern. Previously, there was code
315        that made a preliminary scan of the pattern in order to do this. That was
316        OK when PCRE was new, but as the facilities have expanded, it has become
317        harder and harder to keep it in step with the real compile phase, and there
318        have been a number of bugs (see for example, 4 above). I have now found a
319        cunning way of running the real compile function in a "fake" mode that
320        enables it to compute how much memory it would need, while actually only
321        ever using a few hundred bytes of working memory and without too many
322        tests of the mode. This should make future maintenance and development
323        easier. A side effect of this work is that the limit of 200 on the nesting
324        depth of parentheses has been removed (though this was never a serious
325        limitation, I suspect). However, there is a downside: pcre_compile() now
326        runs more slowly than before (30% or more, depending on the pattern). I
327        hope this isn't a big issue. There is no effect on runtime performance.
329    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
330        newline (only possible for the last line of a file) and it was a
331        pattern that set a locale (followed by /Lsomething), pcretest crashed.
333    19. Added additional timing features to pcretest. (1) The -tm option now times
334        matching only, not compiling. (2) Both -t and -tm can be followed, as a
335        separate command line item, by a number that specifies the number of
336        repeats to use when timing. The default is 50000; this gives better
337        precision, but takes uncomfortably long for very large patterns.
339    20. Extended pcre_study() to be more clever in cases where a branch of a
340        subpattern has no definite first character. For example, (a*|b*)[cd] would
341        previously give no result from pcre_study(). Now it recognizes that the
342        first character must be a, b, c, or d.
344    21. There was an incorrect error "recursive call could loop indefinitely" if
345        a subpattern (or the entire pattern) that was being tested for matching an
346        empty string contained only one non-empty item after a nested subpattern.
347        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
348        incorrectly, because the \d was being skipped in the check.
350    22. The pcretest program now has a new pattern option /B and a command line
351        option -b, which is equivalent to adding /B to every pattern. This causes
352        it to show the compiled bytecode, without the additional information that
353        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
354        is the same as /B/I).
356    23. A new optimization is now able automatically to treat some sequences such
357        as a*b as a*+b. More specifically, if something simple (such as a character
358        or a simple class like \d) has an unlimited quantifier, and is followed by
359        something that cannot possibly match the quantified thing, the quantifier
360        is automatically "possessified".
362    24. A recursive reference to a subpattern whose number was greater than 39
363        went wrong under certain circumstances in UTF-8 mode. This bug could also
364        have affected the operation of pcre_study().
366    25. Realized that a little bit of performance could be had by replacing
367        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
369    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
371    27. Possessive quantifiers such as a++ were previously implemented by turning
372        them into atomic groups such as ($>a+). Now they have their own opcodes,
373        which improves performance. This includes the automatically created ones
374        from 23 above.
376    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
377        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
378        the first matched character to be a colon. This applied both to named and
379        numbered groups.
381    29. The ucpinternal.h header file was missing its idempotency #ifdef.
383    30. I was sent a "project" file called libpcre.a.dev which I understand makes
384        building PCRE on Windows easier, so I have included it in the distribution.
386    31. There is now a check in pcretest against a ridiculously large number being
387        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
388        loop, the loop is abandoned.
390    32. Forward references to subpatterns in conditions such as (?(2)...) where
391        subpattern 2 is defined later cause pcre_compile() to search forwards in
392        the pattern for the relevant set of parentheses. This search went wrong
393        when there were unescaped parentheses in a character class, parentheses
394        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
396    33. "Subroutine" calls and backreferences were previously restricted to
397        referencing subpatterns earlier in the regex. This restriction has now
398        been removed.
400    34. Added a number of extra features that are going to be in Perl 5.10. On the
401        whole, these are just syntactic alternatives for features that PCRE had
402        previously implemented using the Python syntax or my own invention. The
403        other formats are all retained for compatibility.
405        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
406            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
407            also .NET compatible.
409        (b) A recursion or subroutine call to a named group can now be defined as
410            (?&name) as well as (?P>name).
412        (c) A backreference to a named group can now be defined as \k<name> or
413            \k'name' as well as (?P=name). The new forms, as well as being in Perl
414            5.10, are also .NET compatible.
416        (d) A conditional reference to a named group can now use the syntax
417            (?(<name>) or (?('name') as well as (?(name).
419        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
420            groups (named and numbered) that are never evaluated inline, but can be
421            called as "subroutines" from elsewhere. In effect, the DEFINE condition
422            is always false. There may be only one alternative in such a group.
424        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
425            as the simple (?(R). The condition is true only if the most recent
426            recursion is that of the given number or name. It does not search out
427            through the entire recursion stack.
429        (g) The escape \gN or \g{N} has been added, where N is a positive or
430            negative number, specifying an absolute or relative reference.
432    35. Tidied to get rid of some further signed/unsigned compiler warnings and
433        some "unreachable code" warnings.
435    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
436        things, this adds five new scripts.
438    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
439        There were also incompatibilities regarding the handling of \Q..\E inside
440        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
441        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
443    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
444        matches an empty string, and forcibly breaks the loop. There were bugs in
445        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
446        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
447        separate and independent bugs (that affected different cases) have been
448        fixed.
450    39. Refactored the code to abolish the use of different opcodes for small
451        capturing bracket numbers. This is a tidy that I avoided doing when I
452        removed the limit on the number of capturing brackets for 3.5 back in 2001.
453        The new approach is not only tidier, it makes it possible to reduce the
454        memory needed to fix the previous bug (38).
456    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
457        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
458        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
459        mode.
461    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
462        report.
464    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
465        copy construction and assignment in the C++ wrapper.
467    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
468        couple of bugs that could have given wrong results in the "--newline=crlf"
469        case.
471    44. Added a number of casts and did some reorganization of signed/unsigned int
472        variables following suggestions from Dair Grant. Also renamed the variable
473        "this" as "item" because it is a C++ keyword.
475    45. Arranged for dftables to add
477          #include "pcre_internal.h"
479        to pcre_chartables.c because without it, gcc 4.x may remove the array
480        definition from the final binary if PCRE is built into a static library and
481        dead code stripping is activated.
483    46. For an unanchored pattern, if a match attempt fails at the start of a
484        newline sequence, and the newline setting is CRLF or ANY, and the next two
485        characters are CRLF, advance by two characters instead of one.
488    Version 6.7 04-Jul-06
489    ---------------------
491     1. In order to handle tests when input lines are enormously long, pcretest has
492        been re-factored so that it automatically extends its buffers when
493        necessary. The code is crude, but this _is_ just a test program. The
494        default size has been increased from 32K to 50K.
496     2. The code in pcre_study() was using the value of the re argument before
497        testing it for NULL. (Of course, in any sensible call of the function, it
498        won't be NULL.)
500     3. The memmove() emulation function in pcre_internal.h, which is used on
501        systems that lack both memmove() and bcopy() - that is, hardly ever -
502        was missing a "static" storage class specifier.
504     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
505        containing an extended class (one that cannot be represented by a bitmap
506        because it contains high-valued characters or Unicode property items, e.g.
507        [\pZ]). Almost always one would set UTF-8 mode when processing such a
508        pattern, but PCRE should not loop if you do not (it no longer does).
509        [Detail: two cases were found: (a) a repeated subpattern containing an
510        extended class; (b) a recursive reference to a subpattern that followed a
511        previous extended class. It wasn't skipping over the extended class
512        correctly when UTF-8 mode was not set.]
514     5. A negated single-character class was not being recognized as fixed-length
515        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
516        compile error "lookbehind assertion is not fixed length".
518     6. The RunPerlTest auxiliary script was showing an unexpected difference
519        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
520        write a Perl script that can interpret lines of an input file either as
521        byte characters or as UTF-8, which is what "perltest" was being required to
522        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
523        can't do is switch easily at run time between having the "use utf8;" pragma
524        or not. In the end, I fudged it by using the RunPerlTest script to insert
525        "use utf8;" explicitly for the UTF-8 tests.
527     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
528        the end of the subject string, contrary to the documentation and to what
529        Perl does. This was true of both matching functions. Now it matches only at
530        the start of the subject and immediately after *internal* newlines.
532     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
533        a pointer to an int instead of a pointer to an unsigned long int. This
534        caused problems on 64-bit systems.
536     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
537        instance of the 'standard' template library not being so standard".
539    10. There was no check on the number of named subpatterns nor the maximum
540        length of a subpattern name. The product of these values is used to compute
541        the size of the memory block for a compiled pattern. By supplying a very
542        long subpattern name and a large number of named subpatterns, the size
543        computation could be caused to overflow. This is now prevented by limiting
544        the length of names to 32 characters, and the number of named subpatterns
545        to 10,000.
547    11. Subpatterns that are repeated with specific counts have to be replicated in
548        the compiled pattern. The size of memory for this was computed from the
549        length of the subpattern and the repeat count. The latter is limited to
550        65535, but there was no limit on the former, meaning that integer overflow
551        could in principle occur. The compiled length of a repeated subpattern is
552        now limited to 30,000 bytes in order to prevent this.
554    12. Added the optional facility to have named substrings with the same name.
556    13. Added the ability to use a named substring as a condition, using the
557        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
558        are numbers (not recommended). Forward references are permitted.
560    14. Added forward references in named backreferences (if you see what I mean).
562    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
563        pattern could run off the end of the subject. For example, the pattern
564        "(?s)(.{1,5})"8 did this with the subject "ab".
566    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
567        PCRE_CASELESS was set when matching characters that were quantified with ?
568        or *.
570    17. A character class other than a single negated character that had a minimum
571        but no maximum quantifier - for example [ab]{6,} - was not handled
572        correctly by pce_dfa_exec(). It would match only one character.
574    18. A valid (though odd) pattern that looked like a POSIX character
575        class but used an invalid character after [ (for example [[,abc,]]) caused
576        pcre_compile() to give the error "Failed: internal error: code overflow" or
577        in some cases to crash with a glibc free() error. This could even happen if
578        the pattern terminated after [[ but there just happened to be a sequence of
579        letters, a binary zero, and a closing ] in the memory that followed.
581    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
582        over the years. Originally (before any Unicode support), just the bottom 8
583        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
584        output from "man perlunicode" includes this:
586          The regular expression compiler produces polymorphic opcodes.  That
587          is, the pattern adapts to the data and automatically switches to
588          the Unicode character scheme when presented with Unicode data--or
589          instead uses a traditional byte scheme when presented with byte
590          data.
592        Sadly, a wide octal escape does not cause a switch, and in a string with
593        no other multibyte characters, these octal escapes are treated as before.
594        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
595        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
596        Unicode string.
598        I have not perpetrated such confusion in PCRE. Up till now, it took just
599        the bottom 8 bits, as in old Perl. I have now made octal escapes with
600        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
601        translate to the appropriate multibyte character.
603    29. Applied some refactoring to reduce the number of warnings from Microsoft
604        and Borland compilers. This has included removing the fudge introduced
605        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
606        a warning about an unused variable.
608    21. PCRE has not included VT (character 0x0b) in the set of whitespace
609        characters since release 4.0, because Perl (from release 5.004) does not.
610        [Or at least, is documented not to: some releases seem to be in conflict
611        with the documentation.] However, when a pattern was studied with
612        pcre_study() and all its branches started with \s, PCRE still included VT
613        as a possible starting character. Of course, this did no harm; it just
614        caused an unnecessary match attempt.
616    22. Removed a now-redundant internal flag bit that recorded the fact that case
617        dependency changed within the pattern. This was once needed for "required
618        byte" processing, but is no longer used. This recovers a now-scarce options
619        bit. Also moved the least significant internal flag bit to the most-
620        significant bit of the word, which was not previously used (hangover from
621        the days when it was an int rather than a uint) to free up another bit for
622        the future.
624    23. Added support for CRLF line endings as well as CR and LF. As well as the
625        default being selectable at build time, it can now be changed at runtime
626        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
627        specify that it is scanning data with non-default line endings.
629    24. Changed the definition of CXXLINK to make it agree with the definition of
630        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
632    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
633        recursions. This makes a big different to stack usage for some patterns.
635    26. If a subpattern containing a named recursion or subroutine reference such
636        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
637        the space required for the compiled pattern went wrong and gave too small a
638        value. Depending on the environment, this could lead to "Failed: internal
639        error: code overflow at offset 49" or "glibc detected double free or
640        corruption" errors.
642    27. Applied patches from Google (a) to support the new newline modes and (b) to
643        advance over multibyte UTF-8 characters in GlobalReplace.
645    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
646        difference for some implementation of PCRE in some Windows version.
648    29. Added some extra testing facilities to pcretest:
650        \q<number>   in a data line sets the "match limit" value
651        \Q<number>   in a data line sets the "match recursion limt" value
652        -S <number>  sets the stack size, where <number> is in megabytes
654        The -S option isn't available for Windows.
657    Version 6.6 06-Feb-06
658    ---------------------
660     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
661        in pcreposix.h. I have copied the definition from pcre.h.
663     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
664        because pcre.h is no longer a built file.
666     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
667        not normally included in the compiled code.
670    Version 6.5 01-Feb-06
671    ---------------------
673     1. When using the partial match feature with pcre_dfa_exec(), it was not
674        anchoring the second and subsequent partial matches at the new starting
675        point. This could lead to incorrect results. For example, with the pattern
676        /1234/, partially matching against "123" and then "a4" gave a match.
678     2. Changes to pcregrep:
680        (a) All non-match returns from pcre_exec() were being treated as failures
681            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
682            error message is output. Some extra information is given for the
683            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
684            probably the only errors that are likely to be caused by users (by
685            specifying a regex that has nested indefinite repeats, for instance).
686            If there are more than 20 of these errors, pcregrep is abandoned.
688        (b) A binary zero was treated as data while matching, but terminated the
689            output line if it was written out. This has been fixed: binary zeroes
690            are now no different to any other data bytes.
692        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
693            used to set a locale for matching. The --locale=xxxx long option has
694            been added (no short equivalent) to specify a locale explicitly on the
695            pcregrep command, overriding the environment variables.
697        (d) When -B was used with -n, some line numbers in the output were one less
698            than they should have been.
700        (e) Added the -o (--only-matching) option.
702        (f) If -A or -C was used with -c (count only), some lines of context were
703            accidentally printed for the final match.
705        (g) Added the -H (--with-filename) option.
707        (h) The combination of options -rh failed to suppress file names for files
708            that were found from directory arguments.
710        (i) Added the -D (--devices) and -d (--directories) options.
712        (j) Added the -F (--fixed-strings) option.
714        (k) Allow "-" to be used as a file name for -f as well as for a data file.
716        (l) Added the --colo(u)r option.
718        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
719            is not present by default.
721     3. A nasty bug was discovered in the handling of recursive patterns, that is,
722        items such as (?R) or (?1), when the recursion could match a number of
723        alternatives. If it matched one of the alternatives, but subsequently,
724        outside the recursion, there was a failure, the code tried to back up into
725        the recursion. However, because of the way PCRE is implemented, this is not
726        possible, and the result was an incorrect result from the match.
728        In order to prevent this happening, the specification of recursion has
729        been changed so that all such subpatterns are automatically treated as
730        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
732     4. I had overlooked the fact that, in some locales, there are characters for
733        which isalpha() is true but neither isupper() nor islower() are true. In
734        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
735        and ordfeminine) are like this. This affected the treatment of \w and \W
736        when they appeared in character classes, but not when they appeared outside
737        a character class. The bit map for "word" characters is now created
738        separately from the results of isalnum() instead of just taking it from the
739        upper, lower, and digit maps. (Plus the underscore character, of course.)
741     5. The above bug also affected the handling of POSIX character classes such as
742        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
743        permanent tables. Instead, the bit maps for such a class were previously
744        created as the appropriate unions of the upper, lower, and digit bitmaps.
745        Now they are created by subtraction from the [[:word:]] class, which has
746        its own bitmap.
748     6. The [[:blank:]] character class matches horizontal, but not vertical space.
749        It is created by subtracting the vertical space characters (\x09, \x0a,
750        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
751        subtraction was done in the overall bitmap for a character class, meaning
752        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
753        be recognized. This bug has been fixed.
755     7. Patches from the folks at Google:
757          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
758          real life, but is still worth protecting against".
760          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
761          regular expressions".
763          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
764          have it.
766          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
767          "configure" and the latter not, in order to fix a problem somebody had
768          with compiling the Arg class on HP-UX.
770          (e) Improve the error-handling of the C++ wrapper a little bit.
772          (f) New tests for checking recursion limiting.
774     8. The pcre_memmove() function, which is used only if the environment does not
775        have a standard memmove() function (and is therefore rarely compiled),
776        contained two bugs: (a) use of int instead of size_t, and (b) it was not
777        returning a result (though PCRE never actually uses the result).
779     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
780        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
781        returned instead of calling malloc() with an overflowing number that would
782        most likely cause subsequent chaos.
784    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
786    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
787        with this option is matched, the nmatch and pmatch options of regexec() are
788        ignored.
790    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
791        provided in case anyone wants to the the POSIX interface with UTF-8
792        strings.
794    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
795        C++ linking (needed for some HP-UX environments).
797    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
798        (unused parameter) and in the pcre_printint() function (omitted "default"
799        switch label when the default is to do nothing).
801    15. Added some code to make it possible, when PCRE is compiled as a C++
802        library, to replace subject pointers for pcre_exec() with a smart pointer
803        class, thus making it possible to process discontinuous strings.
805    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
806        much the same function. They were added by different people who were trying
807        to make PCRE easy to compile on non-Unix systems. It has been suggested
808        that PCRE_EXPORT be abolished now that there is more automatic apparatus
809        for compiling on Windows systems. I have therefore replaced it with
810        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
811        defaults to "extern" for C or "extern C" for C++, which works fine on
812        Unix-like systems. It is now possible to override the value of PCRE_DATA_
813        SCOPE with something explicit in config.h. In addition:
815        (a) pcreposix.h still had just "extern" instead of either of these macros;
816            I have replaced it with PCRE_DATA_SCOPE.
818        (b) Functions such as _pcre_xclass(), which are internal to the library,
819            but external in the C sense, all had PCRE_EXPORT in their definitions.
820            This is apparently wrong for the Windows case, so I have removed it.
821            (It makes no difference on Unix-like systems.)
823    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
824        of recursive calls to match(). This is different to MATCH_LIMIT because
825        that limits the total number of calls to match(), not all of which increase
826        the depth of recursion. Limiting the recursion depth limits the amount of
827        stack (or heap if NO_RECURSE is set) that is used. The default can be set
828        when PCRE is compiled, and changed at run time. A patch from Google adds
829        this functionality to the C++ interface.
831    18. Changes to the handling of Unicode character properties:
833        (a) Updated the table to Unicode 4.1.0.
835        (b) Recognize characters that are not in the table as "Cn" (undefined).
837        (c) I revised the way the table is implemented to a much improved format
838            which includes recognition of ranges. It now supports the ranges that
839            are defined in UnicodeData.txt, and it also amalgamates other
840            characters into ranges. This has reduced the number of entries in the
841            table from around 16,000 to around 3,000, thus reducing its size
842            considerably. I realized I did not need to use a tree structure after
843            all - a binary chop search is just as efficient. Having reduced the
844            number of entries, I extended their size from 6 bytes to 8 bytes to
845            allow for more data.
847        (d) Added support for Unicode script names via properties such as \p{Han}.
849    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
850        matching that character.
852    20. When matching a repeated Unicode property with a minimum greater than zero,
853        (for example \pL{2,}), PCRE could look past the end of the subject if it
854        reached it while seeking the minimum number of characters. This could
855        happen only if some of the characters were more than one byte long, because
856        there is a check for at least the minimum number of bytes.
858    21. Refactored the implementation of \p and \P so as to be more general, to
859        allow for more different types of property in future. This has changed the
860        compiled form incompatibly. Anybody with saved compiled patterns that use
861        \p or \P will have to recompile them.
863    22. Added "Any" and "L&" to the supported property types.
865    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
866        but give a compile time error if the value is greater than 0xff.
868    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
869        accidentally not being installed or uninstalled.
871    25. The pcre.h file was built from pcre.h.in, but the only changes that were
872        made were to insert the current release number. This seemed silly, because
873        it made things harder for people building PCRE on systems that don't run
874        "configure". I have turned pcre.h into a distributed file, no longer built
875        by "configure", with the version identification directly included. There is
876        no longer a pcre.h.in file.
878        However, this change necessitated a change to the pcre-config script as
879        well. It is built from pcre-config.in, and one of the substitutions was the
880        release number. I have updated configure.ac so that ./configure now finds
881        the release number by grepping pcre.h.
883    26. Added the ability to run the tests under valgrind.
886    Version 6.4 05-Sep-05
887    ---------------------
889     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
890        "--" to be printed when multiple files were scanned, even when none of the
891        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
892        consider it to be a bug, and have restored the previous behaviour.
894     2. A couple of code tidies to get rid of compiler warnings.
896     3. The pcretest program used to cheat by referring to symbols in the library
897        whose names begin with _pcre_. These are internal symbols that are not
898        really supposed to be visible externally, and in some environments it is
899        possible to suppress them. The cheating is now confined to including
900        certain files from the library's source, which is a bit cleaner.
902     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
903        file's purpose clearer.
905     5. Reorganized pcre_ucp_findchar().
908    Version 6.3 15-Aug-05
909    ---------------------
911     1. The file libpcre.pc.in did not have general read permission in the tarball.
913     2. There were some problems when building without C++ support:
915        (a) If C++ support was not built, "make install" and "make test" still
916            tried to test it.
918        (b) There were problems when the value of CXX was explicitly set. Some
919            changes have been made to try to fix these, and ...
921        (c) --disable-cpp can now be used to explicitly disable C++ support.
923        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
924            backslash in a target when C++ was disabled. This confuses some
925            versions of "make", apparently. Using an intermediate variable solves
926            this. (Same for CPP_LOBJ.)
928     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
929        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
930        necessary on certain architectures.
932     4. Added a setting of -export-symbols-regex to the link command to remove
933        those symbols that are exported in the C sense, but actually are local
934        within the library, and not documented. Their names all begin with
935        "_pcre_". This is not a perfect job, because (a) we have to except some
936        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
937        available (and never for static libraries). I have made a note to try to
938        find a way round (a) in the future.
941    Version 6.2 01-Aug-05
942    ---------------------
944     1. There was no test for integer overflow of quantifier values. A construction
945        such as {1111111111111111} would give undefined results. What is worse, if
946        a minimum quantifier for a parenthesized subpattern overflowed and became
947        negative, the calculation of the memory size went wrong. This could have
948        led to memory overwriting.
950     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
952     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
953        operating environments where this matters.
955     4. Applied Giuseppe Maxia's patch to add additional features for controlling
956        PCRE options from within the C++ wrapper.
958     5. Named capturing subpatterns were not being correctly counted when a pattern
959        was compiled. This caused two problems: (a) If there were more than 100
960        such subpatterns, the calculation of the memory needed for the whole
961        compiled pattern went wrong, leading to an overflow error. (b) Numerical
962        back references of the form \12, where the number was greater than 9, were
963        not recognized as back references, even though there were sufficient
964        previous subpatterns.
966     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
967        versions of gcc, e.g. 2.95.4.
970    Version 6.1 21-Jun-05
971    ---------------------
973     1. There was one reference to the variable "posix" in pcretest.c that was not
974        surrounded by "#if !defined NOPOSIX".
976     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
977        the cross-check on the old pcre_info() function, for the benefit of the
978        cut-down version of PCRE that is currently imported into Exim.
980     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
981        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
982        patterns that start (?i) but I don't think that matters. The use of (?i) is
983        just an example; this all applies to the other options as well.
985     4. Since libtool seems to echo the compile commands it is issuing, the output
986        from "make" can be reduced a bit by putting "@" in front of each libtool
987        compile command.
989     5. Patch from the folks at Google for configure.in to be a bit more thorough
990        in checking for a suitable C++ installation before trying to compile the
991        C++ stuff. This should fix a reported problem when a compiler was present,
992        but no suitable headers.
994     6. The man pages all had just "PCRE" as their title. I have changed them to
995        be the relevant file name. I have also arranged that these names are
996        retained in the file doc/pcre.txt, which is a concatenation in text format
997        of all the man pages except the little individual ones for each function.
999     7. The NON-UNIX-USE file had not been updated for the different set of source
1000        files that come with release 6. I also added a few comments about the C++
1001        wrapper.
1004    Version 6.0 07-Jun-05
1005    ---------------------
1007     1. Some minor internal re-organization to help with my DFA experiments.
1009     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1010        didn't matter for the library itself when fully configured, but did matter
1011        when compiling without UCP support, or within Exim, where the ucp files are
1012        not imported.
1014     3. Refactoring of the library code to split up the various functions into
1015        different source modules. The addition of the new DFA matching code (see
1016        below) to a single monolithic source would have made it really too
1017        unwieldy, quite apart from causing all the code to be include in a
1018        statically linked application, when only some functions are used. This is
1019        relevant even without the DFA addition now that patterns can be compiled in
1020        one application and matched in another.
1022        The downside of splitting up is that there have to be some external
1023        functions and data tables that are used internally in different modules of
1024        the library but which are not part of the API. These have all had their
1025        names changed to start with "_pcre_" so that they are unlikely to clash
1026        with other external names.
1028     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1029        a different (DFA) algorithm. Although it is slower than the original
1030        function, it does have some advantages for certain types of matching
1031        problem.
1033     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1034        including restarting after a partial match.
1036     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1037        defined when compiling for Windows was sent to me. I have put it into the
1038        code, though I have no means of testing or verifying it.
1040     7. Added the pcre_refcount() auxiliary function.
1042     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1043        match before or at the first newline in the subject string. In pcretest,
1044        the /f option on a pattern can be used to set this.
1046     9. A repeated \w when used in UTF-8 mode with characters greater than 256
1047        would behave wrongly. This has been present in PCRE since release 4.0.
1049    10. A number of changes to the pcregrep command:
1051        (a) Refactored how -x works; insert ^(...)$ instead of setting
1052            PCRE_ANCHORED and checking the length, in preparation for adding
1053            something similar for -w.
1055        (b) Added the -w (match as a word) option.
1057        (c) Refactored the way lines are read and buffered so as to have more
1058            than one at a time available.
1060        (d) Implemented a pcregrep test script.
1062        (e) Added the -M (multiline match) option. This allows patterns to match
1063            over several lines of the subject. The buffering ensures that at least
1064            8K, or the rest of the document (whichever is the shorter) is available
1065            for matching (and similarly the previous 8K for lookbehind assertions).
1067        (f) Changed the --help output so that it now says
1069              -w, --word-regex(p)
1071            instead of two lines, one with "regex" and the other with "regexp"
1072            because that confused at least one person since the short forms are the
1073            same. (This required a bit of code, as the output is generated
1074            automatically from a table. It wasn't just a text change.)
1076        (g) -- can be used to terminate pcregrep options if the next thing isn't an
1077            option but starts with a hyphen. Could be a pattern or a path name
1078            starting with a hyphen, for instance.
1080        (h) "-" can be given as a file name to represent stdin.
1082        (i) When file names are being printed, "(standard input)" is used for
1083            the standard input, for compatibility with GNU grep. Previously
1084            "<stdin>" was used.
1086        (j) The option --label=xxx can be used to supply a name to be used for
1087            stdin when file names are being printed. There is no short form.
1089        (k) Re-factored the options decoding logic because we are going to add
1090            two more options that take data. Such options can now be given in four
1091            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1093        (l) Added the -A, -B, and -C options for requesting that lines of context
1094            around matches be printed.
1096        (m) Added the -L option to print the names of files that do not contain
1097            any matching lines, that is, the complement of -l.
1099        (n) The return code is 2 if any file cannot be opened, but pcregrep does
1100            continue to scan other files.
1102        (o) The -s option was incorrectly implemented. For compatibility with other
1103            greps, it now suppresses the error message for a non-existent or non-
1104            accessible file (but not the return code). There is a new option called
1105            -q that suppresses the output of matching lines, which was what -s was
1106            previously doing.
1108        (p) Added --include and --exclude options to specify files for inclusion
1109            and exclusion when recursing.
1111    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1112        Hopefully, it now does.
1114    12. Missing cast in pcre_study().
1116    13. Added an "uninstall" target to the makefile.
1118    14. Replaced "extern" in the function prototypes in Makefile.in with
1119        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1120        world, but is set differently for Windows.
1122    15. Added a second compiling function called pcre_compile2(). The only
1123        difference is that it has an extra argument, which is a pointer to an
1124        integer error code. When there is a compile-time failure, this is set
1125        non-zero, in addition to the error test pointer being set to point to an
1126        error message. The new argument may be NULL if no error number is required
1127        (but then you may as well call pcre_compile(), which is now just a
1128        wrapper). This facility is provided because some applications need a
1129        numeric error indication, but it has also enabled me to tidy up the way
1130        compile-time errors are handled in the POSIX wrapper.
1132    16. Added VPATH=.libs to the makefile; this should help when building with one
1133        prefix path and installing with another. (Or so I'm told by someone who
1134        knows more about this stuff than I do.)
1136    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1137        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1138        match everything, including newlines. This is not POSIX-compatible, but
1139        somebody wanted the feature. From pcretest it can be activated by using
1140        both the P and the s flags.
1142    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1144    19. libpcre.pc was being incorrectly installed as executable.
1146    20. A couple of places in pcretest check for end-of-line by looking for '\n';
1147        it now also looks for '\r' so that it will work unmodified on Windows.
1149    21. Added Google's contributed C++ wrapper to the distribution.
1151    22. Added some untidy missing memory free() calls in pcretest, to keep
1152        Electric Fence happy when testing.
1156    Version 5.0 13-Sep-04
1157    ---------------------
1159     1. Internal change: literal characters are no longer packed up into items
1160        containing multiple characters in a single byte-string. Each character
1161        is now matched using a separate opcode. However, there may be more than one
1162        byte in the character in UTF-8 mode.
1164     2. The pcre_callout_block structure has two new fields: pattern_position and
1165        next_item_length. These contain the offset in the pattern to the next match
1166        item, and its length, respectively.
1168     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1169        insertion of callouts before each pattern item. Added the /C option to
1170        pcretest to make use of this.
1172     4. On the advice of a Windows user, the lines
1174          #if defined(_WIN32) || defined(WIN32)
1175          _setmode( _fileno( stdout ), 0x8000 );
1176          #endif  /* defined(_WIN32) || defined(WIN32) */
1178        have been added to the source of pcretest. This apparently does useful
1179        magic in relation to line terminators.
1181     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1182        for the benefit of those environments where the "b" makes a difference.
1184     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1185        to know about it. I have put a hack into configure.in that adds in code
1186        to set GCC=yes if CC=icc. This seems to end up at a point in the
1187        generated configure script that is early enough to affect the setting of
1188        compiler options, which is what is needed, but I have no means of testing
1189        whether it really works. (The user who reported this had patched the
1190        generated configure script, which of course I cannot do.)
1192        LATER: After change 22 below (new libtool files), the configure script
1193        seems to know about icc (and also ecc). Therefore, I have commented out
1194        this hack in configure.in.
1196     7. Added support for pkg-config (2 patches were sent in).
1198     8. Negated POSIX character classes that used a combination of internal tables
1199        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1200        [[:^ascii]]. Typically, they would match almost any characters. The other
1201        POSIX classes were not broken in this way.
1203     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1204        to find the match, as PCRE was deluded into thinking that the match had to
1205        start at the start point or following a newline. The same bug applied to
1206        patterns with negative forward assertions or any backward assertions
1207        preceding ".*" at the start, unless the pattern required a fixed first
1208        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1210    10. In UTF-8 mode, when moving forwards in the subject after a failed match
1211        starting at the last subject character, bytes beyond the end of the subject
1212        string were read.
1214    11. Renamed the variable "class" as "classbits" to make life easier for C++
1215        users. (Previously there was a macro definition, but it apparently wasn't
1216        enough.)
1218    12. Added the new field "tables" to the extra data so that tables can be passed
1219        in at exec time, or the internal tables can be re-selected. This allows
1220        a compiled regex to be saved and re-used at a later time by a different
1221        program that might have everything at different addresses.
1223    13. Modified the pcre-config script so that, when run on Solaris, it shows a
1224        -R library as well as a -L library.
1226    14. The debugging options of pcretest (-d on the command line or D on a
1227        pattern) showed incorrect output for anything following an extended class
1228        that contained multibyte characters and which was followed by a quantifier.
1230    15. Added optional support for general category Unicode character properties
1231        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1232        support. It adds about 90K to the size of the library. The meanings of the
1233        inbuilt class escapes such as \d and \s have NOT been changed.
1235    16. Updated pcredemo.c to include calls to free() to release the memory for the
1236        compiled pattern.
1238    17. The generated file chartables.c was being created in the source directory
1239        instead of in the building directory. This caused the build to fail if the
1240        source directory was different from the building directory, and was
1241        read-only.
1243    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1244        file. No doubt somebody will tell me if they don't make sense... Also added
1245        Dan Mooney's comments about building on OpenVMS.
1247    19. Added support for partial matching via the PCRE_PARTIAL option for
1248        pcre_exec() and the \P data escape in pcretest.
1250    20. Extended pcretest with 3 new pattern features:
1252        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
1253              write the compiled pattern to the file whose name is "rest-of-line".
1254              This is a straight binary dump of the data, with the saved pointer to
1255              the character tables forced to be NULL. The study data, if any, is
1256              written too. After writing, pcretest reads a new pattern.
1258        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1259              compiled pattern from the given file. There must not be any
1260              occurrences of "<" in the file name (pretty unlikely); if there are,
1261              pcretest will instead treat the initial "<" as a pattern delimiter.
1262              After reading in the pattern, pcretest goes on to read data lines as
1263              usual.
1265        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1266              and 16-bit fields in a compiled pattern, to simulate a pattern that
1267              was compiled on a host of opposite endianness.
1269    21. The pcre-exec() function can now cope with patterns that were compiled on
1270        hosts of opposite endianness, with this restriction:
1272          As for any compiled expression that is saved and used later, the tables
1273          pointer field cannot be preserved; the extra_data field in the arguments
1274          to pcre_exec() should be used to pass in a tables address if a value
1275          other than the default internal tables were used at compile time.
1277    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1278        now diagnosed as an error. Previously, most of the time, a negative number
1279        would have been treated as zero, but if in addition "ovector" was passed as
1280        NULL, a crash could occur.
1282    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1283        new versions from the libtool 1.5 distribution (the last one is a copy of
1284        a file called libtool.m4). This seems to have fixed the need to patch
1285        "configure" to support Darwin 1.3 (which I used to do). However, I still
1286        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1287        workstation).
1289    24. Changed the PCRE licence to be the more standard "BSD" licence.
1292    Version 4.5 01-Dec-03
1293    ---------------------
1295     1. There has been some re-arrangement of the code for the match() function so
1296        that it can be compiled in a version that does not call itself recursively.
1297        Instead, it keeps those local variables that need separate instances for
1298        each "recursion" in a frame on the heap, and gets/frees frames whenever it
1299        needs to "recurse". Keeping track of where control must go is done by means
1300        of setjmp/longjmp. The whole thing is implemented by a set of macros that
1301        hide most of the details from the main code, and operates only if
1302        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1303        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1304        operating.
1306        To make it easier for callers to provide specially tailored get/free
1307        functions for this usage, two new functions, pcre_stack_malloc, and
1308        pcre_stack_free, are used. They are always called in strict stacking order,
1309        and the size of block requested is always the same.
1311        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1312        PCRE has been compiled to use the stack or the heap for recursion. The
1313        -C option of pcretest uses this to show which version is compiled.
1315        A new data escape \S, is added to pcretest; it causes the amounts of store
1316        obtained and freed by both kinds of malloc/free at match time to be added
1317        to the output.
1319     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1320        what's available on my current Linux desktop machine.
1322     3. When matching a UTF-8 string, the test for a valid string at the start has
1323        been extended. If start_offset is not zero, PCRE now checks that it points
1324        to a byte that is the start of a UTF-8 character. If not, it returns
1325        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1326        this is necessary because there may be backward assertions in the pattern.
1327        When matching the same subject several times, it may save resources to use
1328        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1330     4. The code for checking the validity of UTF-8 strings has been tightened so
1331        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1332        containing "overlong sequences".
1334     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1335        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1336        should have been "&", but it just so happened that all the cases this let
1337        through by mistake were picked up later in the function.
1339     6. I had used a variable called "isblank" - this is a C99 function, causing
1340        some compilers to warn. To avoid this, I renamed it (as "blankclass").
1342     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1343        prompting; (b) run "./pcretest /dev/null" at the start of the test script
1344        so the version is shown; (c) stop "make test" echoing "./RunTest".
1346     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1348     9. The prototype for memmove() for systems that don't have it was using
1349        size_t, but the inclusion of the header that defines size_t was later. I've
1350        moved the #includes for the C headers earlier to avoid this.
1352    10. Added some adjustments to the code to make it easier to compiler on certain
1353        special systems:
1355          (a) Some "const" qualifiers were missing.
1356          (b) Added the macro EXPORT before all exported functions; by default this
1357              is defined to be empty.
1358          (c) Changed the dftables auxiliary program (that builds chartables.c) so
1359              that it reads its output file name as an argument instead of writing
1360              to the standard output and assuming this can be redirected.
1362    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
1363        class containing characters with values greater than 255, PCRE compilation
1364        went into a loop.
1366    12. A recursive reference to a subpattern that was within another subpattern
1367        that had a minimum quantifier of zero caused PCRE to crash. For example,
1368        (x(y(?2))z)? provoked this bug with a subject that got as far as the
1369        recursion. If the recursively-called subpattern itself had a zero repeat,
1370        that was OK.
1372    13. In pcretest, the buffer for reading a data line was set at 30K, but the
1373        buffer into which it was copied (for escape processing) was still set at
1374        1024, so long lines caused crashes.
1376    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
1377        "internal error: code overflow...". This applied to any character class
1378        that was followed by a possessive quantifier.
1380    15. Modified the Makefile to add libpcre.la as a prerequisite for
1381        libpcreposix.la because I was told this is needed for a parallel build to
1382        work.
1384    16. If a pattern that contained .* following optional items at the start was
1385        studied, the wrong optimizing data was generated, leading to matching
1386        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
1387        matching string must start with a or b or c. The correct conclusion for
1388        this pattern is that a match can start with any character.
1391    Version 4.4 13-Aug-03
1392    ---------------------
1394     1. In UTF-8 mode, a character class containing characters with values between
1395        127 and 255 was not handled correctly if the compiled pattern was studied.
1396        In fixing this, I have also improved the studying algorithm for such
1397        classes (slightly).
1399     2. Three internal functions had redundant arguments passed to them. Removal
1400        might give a very teeny performance improvement.
1402     3. Documentation bug: the value of the capture_top field in a callout is *one
1403        more than* the number of the hightest numbered captured substring.
1405     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
1406        in incorrectly linking with a previously installed version. They now link
1407        explicitly with libpcre.la.
1409     5. configure.in no longer needs to recognize Cygwin specially.
1411     6. A problem in pcre.in for Windows platforms is fixed.
1413     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
1414        pcretest, it used to include the size of the study block as part of its
1415        output. Unfortunately, the structure contains a field that has a different
1416        size on different hardware architectures. This meant that the tests that
1417        showed this size failed. As the block is currently always of a fixed size,
1418        this information isn't actually particularly useful in pcretest output, so
1419        I have just removed it.
1421     8. Three pre-processor statements accidentally did not start in column 1.
1422        Sadly, there are *still* compilers around that complain, even though
1423        standard C has not required this for well over a decade. Sigh.
1425     9. In pcretest, the code for checking callouts passed small integers in the
1426        callout_data field, which is a void * field. However, some picky compilers
1427        complained about the casts involved for this on 64-bit systems. Now
1428        pcretest passes the address of the small integer instead, which should get
1429        rid of the warnings.
1431    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
1432        both compile and run time, and gives an error if an invalid UTF-8 sequence
1433        is found. There is a option for disabling this check in cases where the
1434        string is known to be correct and/or the maximum performance is wanted.
1436    11. In response to a bug report, I changed one line in Makefile.in from
1438            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
1439        to
1440            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
1442        to look similar to other lines, but I have no way of telling whether this
1443        is the right thing to do, as I do not use Windows. No doubt I'll get told
1444        if it's wrong...
1447    Version 4.3 21-May-03
1448    ---------------------
1450    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
1451       Makefile.
1453    2. Some refactoring to improve the quality of the code:
1455       (i)   The utf8_table... variables are now declared "const".
1457       (ii)  The code for \cx, which used the "case flipping" table to upper case
1458             lower case letters, now just substracts 32. This is ASCII-specific,
1459             but the whole concept of \cx is ASCII-specific, so it seems
1460             reasonable.
1462       (iii) PCRE was using its character types table to recognize decimal and
1463             hexadecimal digits in the pattern. This is silly, because it handles
1464             only 0-9, a-f, and A-F, but the character types table is locale-
1465             specific, which means strange things might happen. A private
1466             table is now used for this - though it costs 256 bytes, a table is
1467             much faster than multiple explicit tests. Of course, the standard
1468             character types table is still used for matching digits in subject
1469             strings against \d.
1471       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
1472             ending in _t are). So I've renamed it as ESC_tee.
1474    3. The first argument for regexec() in the POSIX wrapper should have been
1475       defined as "const".
1477    4. Changed pcretest to use malloc() for its buffers so that they can be
1478       Electric Fenced for debugging.
1480    5. There were several places in the code where, in UTF-8 mode, PCRE would try
1481       to read one or more bytes before the start of the subject string. Often this
1482       had no effect on PCRE's behaviour, but in some circumstances it could
1483       provoke a segmentation fault.
1485    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
1486       to try to read one or more bytes before the start of the subject string.
1488    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
1489       UTF-8 support could misbehave in various ways if the subject string
1490       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
1491       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
1492       back over UTF-8 characters.)
1495    Version 4.2 14-Apr-03
1496    ---------------------
1498    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
1500    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
1501         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
1502         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
1503         [WIN_PREFIX]: new variable, "cyg" for Cygwin
1504         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
1505           and BUILD_EXEEXT
1506         Note: automatic setting of the BUILD variables is not yet working
1507         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
1508           compile-time but not at link-time
1509         [LINK]: use for linking executables only
1510         make different versions for Windows and non-Windows
1511         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
1512           libraries
1513         [LINK_FOR_BUILD]: new variable
1514         [OBJEXT]: use throughout
1515         [EXEEXT]: use throughout
1516         <winshared>: new target
1517         <wininstall>: new target
1518         <dftables.o>: use native compiler
1519         <dftables>: use native linker
1520         <install>: handle Windows platform correctly
1521         <clean>: ditto
1522         <check>: ditto
1523         copy DLL to top builddir before testing
1525       As part of these changes, -no-undefined was removed again. This was reported
1526       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
1527       in any case.
1529    3. Some tidies to get rid of compiler warnings:
1531       . In the match_data structure, match_limit was an unsigned long int, whereas
1532         match_call_count was an int. I've made them both unsigned long ints.
1534       . In pcretest the fact that a const uschar * doesn't automatically cast to
1535         a void * provoked a warning.
1537       . Turning on some more compiler warnings threw up some "shadow" variables
1538         and a few more missing casts.
1540    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
1541       option, a class that contained a single character with a value between 128
1542       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
1544    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
1545       option, a class that contained several characters, but with at least one
1546       whose value was between 128 and 255 caused PCRE to crash.
1549    Version 4.1 12-Mar-03
1550    ---------------------
1552    1. Compiling with gcc -pedantic found a couple of places where casts were
1553    needed, and a string in dftables.c that was longer than standard compilers are
1554    required to support.
1556    2. Compiling with Sun's compiler found a few more places where the code could
1557    be tidied up in order to avoid warnings.
1559    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
1560    first of these names is deprecated in the latest Autoconf in favour of the name
1561    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
1562    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
1563    analogy I have changed it to CFLAGS_FOR_BUILD.
1565    4. Added -no-undefined to the linking command in the Makefile, because this is
1566    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
1567    linking step for the pcreposix library.
1569    5. PCRE was failing to diagnose the case of two named groups with the same
1570    name.
1572    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
1573    literal character that is needed in the subject for a match, and scans along to
1574    ensure that it is present before embarking on the full matching process. This
1575    saves time in cases of nested unlimited repeats that are never going to match.
1576    Problem: the scan can take a lot of time if the subject is very long (e.g.
1577    megabytes), thus penalizing straightforward matches. It is now done only if the
1578    amount of subject to be scanned is less than 1000 bytes.
1580    7. A lesser problem with the same optimization is that it was recording the
1581    first character of an anchored pattern as "needed", thus provoking a search
1582    right along the subject, even when the first match of the pattern was going to
1583    fail. The "needed" character is now not set for anchored patterns, unless it
1584    follows something in the pattern that is of non-fixed length. Thus, it still
1585    fulfils its original purpose of finding quick non-matches in cases of nested
1586    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
1589    Version 4.0 17-Feb-03
1590    ---------------------
1592    1. If a comment in an extended regex that started immediately after a meta-item
1593    extended to the end of string, PCRE compiled incorrect data. This could lead to
1594    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
1596    2. Moved to autoconf 2.53 and libtool 1.4.2.
1598    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
1599    the special perltest8 script is no longer needed - all the tests can be run
1600    from a single perltest script.
1602    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
1603    by \s. It has now been removed in PCRE. This means it isn't recognized as
1604    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
1605    class [:space:] *does* include VT, thereby creating a mess.
1607    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
1608    space and tab.
1610    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
1611    its new features into the main test script, reducing the number of scripts.
1613    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
1614    were backward compatible, and made the (?i) apply to the whole pattern, as if
1615    /i were given. Now it behaves more logically, and applies the option setting
1616    only to what follows. PCRE has been changed to follow suit. However, if it
1617    finds options settings right at the start of the pattern, it extracts them into
1618    the global options, as before. Thus, they show up in the info data.
1620    8. Added support for the \Q...\E escape sequence. Characters in between are
1621    treated as literals. This is slightly different from Perl in that $ and @ are
1622    also handled as literals inside the quotes. In Perl, they will cause variable
1623    interpolation. Note the following examples:
1625        Pattern            PCRE matches      Perl matches
1627        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
1628        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
1629        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
1631    For compatibility with Perl, \Q...\E sequences are recognized inside character
1632    classes as well as outside them.
1634    9. Re-organized 3 code statements in pcretest to avoid "overflow in
1635    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
1636    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
1637    signed/unsigned warnings.
1639    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
1640    option for pcretest, so I've replaced it by a simple function that does just
1641    that job.
1643    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
1644    "pcregrep -".
1646    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
1647    Java package. This provides some syntactic sugar for simple cases of what my
1648    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
1649    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
1650    item, you can use this simplified notation. Note that only makes sense with
1651    greedy quantifiers. Consequently, the use of the possessive quantifier forces
1652    greediness, whatever the setting of the PCRE_UNGREEDY option.
1654    13. A change of greediness default within a pattern was not taking effect at
1655    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
1656    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
1657    was abstracted outside.
1659    14. PCRE now supports the \G assertion. It is true when the current matching
1660    position is at the start point of the match. This differs from \A when the
1661    starting offset is non-zero. Used with the /g option of pcretest (or similar
1662    code), it works in the same way as it does for Perl's /g option. If all
1663    alternatives of a regex begin with \G, the expression is anchored to the start
1664    match position, and the "anchored" flag is set in the compiled expression.
1666    15. Some bugs concerning the handling of certain option changes within patterns
1667    have been fixed. These applied to options other than (?ims). For example,
1668    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
1669    been the other way round. Some of this was related to change 7 above.
1671    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
1672    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
1673    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
1674    POSIX classes only within a class (e.g. /[[:alpha:]]/).
1676    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
1677    mode. Unlike ".", it always matches newline, whatever the setting of
1678    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
1679    assertions. Perl allows it, but it doesn't (in general) work because it can't
1680    calculate the length of the lookbehind. At least, that's the case for Perl
1681    5.8.0 - I've been told they are going to document that it doesn't work in
1682    future.
1684    18. Added an error diagnosis for escapes that PCRE does not support: these are
1685    \L, \l, \N, \P, \p, \U, \u, and \X.
1687    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
1688    reading past the end of the pattern in cases such as /[abcd/.
1690    20. PCRE was getting more memory than necessary for patterns with classes that
1691    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
1693    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
1694    compiling PCRE for use with Virtual Pascal.
1696    22. Small fix to the Makefile to make it work properly if the build is done
1697    outside the source tree.
1699    23. Added a new extension: a condition to go with recursion. If a conditional
1700    subpattern starts with (?(R) the "true" branch is used if recursion has
1701    happened, whereas the "false" branch is used only at the top level.
1703    24. When there was a very long string of literal characters (over 255 bytes
1704    without UTF support, over 250 bytes with UTF support), the computation of how
1705    much memory was required could be incorrect, leading to segfaults or other
1706    strange effects.
1708    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
1709    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
1710    there was a subsequent back reference to those brackets. This meant that, for
1711    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
1712    possible to check for precisely this case. All we can do is abandon the
1713    optimization if .* occurs inside capturing brackets when there are any back
1714    references whatsoever. (See below for a better fix that came later.)
1716    26. The handling of the optimization for finding the first character of a
1717    non-anchored pattern, and for finding a character that is required later in the
1718    match were failing in some cases. This didn't break the matching; it just
1719    failed to optimize when it could. The way this is done has been re-implemented.
1721    27. Fixed typo in error message for invalid (?R item (it said "(?p").
1723    28. Added a new feature that provides some of the functionality that Perl
1724    provides with (?{...}). The facility is termed a "callout". The way it is done
1725    in PCRE is for the caller to provide an optional function, by setting
1726    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
1727    global variable. By default it is unset, which disables all calling out. To get
1728    the function called, the regex must include (?C) at appropriate points. This
1729    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
1730    This provides a means of identifying different callout points. When PCRE
1731    reaches such a point in the regex, if pcre_callout has been set, the external
1732    function is called. It is provided with data in a structure called
1733    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
1734    matching continues; if it returns a non-zero value, the match at the current
1735    point fails. However, backtracking will occur if possible. [This was changed
1736    later and other features added - see item 49 below.]
1738    29. pcretest is upgraded to test the callout functionality. It provides a
1739    callout function that displays information. By default, it shows the start of
1740    the match and the current position in the text. There are some new data escapes
1741    to vary what happens:
1743        \C+         in addition, show current contents of captured substrings
1744        \C-         do not supply a callout function
1745        \C!n        return 1 when callout number n is reached
1746        \C!n!m      return 1 when callout number n is reached for the mth time
1748    30. If pcregrep was called with the -l option and just a single file name, it
1749    output "<stdin>" if a match was found, instead of the file name.
1751    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
1752    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
1753    pcre_exec(). This saves a malloc/free per call. The default value of
1754    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
1755    when configuring.
1757    32. The default maximum size of a compiled pattern is 64K. There have been a
1758    few cases of people hitting this limit. The code now uses macros to handle the
1759    storing of links as offsets within the compiled pattern. It defaults to 2-byte
1760    links, but this can be changed to 3 or 4 bytes by --with-link-size when
1761    configuring. Tests 2 and 5 work only with 2-byte links because they output
1762    debugging information about compiled patterns.
1764    33. Internal code re-arrangements:
1766    (a) Moved the debugging function for printing out a compiled regex into
1767        its own source file (printint.c) and used #include to pull it into
1768        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
1769        separate copies.
1771    (b) Defined the list of op-code names for debugging as a macro in
1772        internal.h so that it is next to the definition of the opcodes.
1774    (c) Defined a table of op-code lengths for simpler skipping along compiled
1775        code. This is again a macro in internal.h so that it is next to the
1776        definition of the opcodes.
1778    34. Added support for recursive calls to individual subpatterns, along the
1779    lines of Robin Houston's patch (but implemented somewhat differently).
1781    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
1782    allow it to read and process whole directories in Win32. This code was
1783    contributed by Lionel Fourquaux; it has not been tested by me.
1785    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
1786    used to name a group. Names consist of alphanumerics and underscores, and must
1787    be unique. Back references use the syntax (?P=name) and recursive calls use
1788    (?P>name) which is a PCRE extension to the Python extension. Groups still have
1789    numbers. The function pcre_fullinfo() can be used after compilation to extract
1790    a name/number map. There are three relevant calls:
1792      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
1793      PCRE_INFO_NAMECOUNT            yields the number of entries
1794      PCRE_INFO_NAMETABLE            yields a pointer to the map.
1796    The map is a vector of fixed-size entries. The size of each entry depends on
1797    the length of the longest name used. The first two bytes of each entry are the
1798    group number, most significant byte first. There follows the corresponding
1799    name, zero terminated. The names are in alphabetical order.
1801    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
1802    case instead of 255. Making it the same both with and without UTF-8 support
1803    means that the same test output works with both.
1805    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
1806    calling malloc() with a zero argument.
1808    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
1809    optimization. I've improved things by keeping a bitmap of backreferences with
1810    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
1811    fact referenced, the optimization can be applied. It is unlikely that a
1812    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
1813    the match to follow \n) will appear inside brackets with a number greater than
1814    31, but if it does, any back reference > 31 suppresses the optimization.
1816    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
1817    of disabling numbered capturing parentheses. Any opening parenthesis that is
1818    not followed by ? behaves as if it were followed by ?: but named parentheses
1819    can still be used for capturing (and they will acquire numbers in the usual
1820    way).
1822    41. Redesigned the return codes from the match() function into yes/no/error so
1823    that errors can be passed back from deep inside the nested calls. A malloc
1824    failure while inside a recursive subpattern call now causes the
1825    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
1827    42. It is now possible to set a limit on the number of times the match()
1828    function is called in a call to pcre_exec(). This facility makes it possible to
1829    limit the amount of recursion and backtracking, though not in a directly
1830    obvious way, because the match() function is used in a number of different
1831    circumstances. The count starts from zero for each position in the subject
1832    string (for non-anchored patterns). The default limit is, for compatibility, a
1833    large number, namely 10 000 000. You can change this in two ways:
1835    (a) When configuring PCRE before making, you can use --with-match-limit=n
1836        to set a default value for the compiled library.
1838    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
1839        a different value is set. See 45 below.
1841    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
1843    43. Added a new function pcre_config(int, void *) to enable run-time extraction
1844    of things that can be changed at compile time. The first argument specifies
1845    what is wanted and the second points to where the information is to be placed.
1846    The current list of available information is:
1848      PCRE_CONFIG_UTF8
1850    The output is an integer that is set to one if UTF-8 support is available;
1851    otherwise it is set to zero.
1855    The output is an integer that it set to the value of the code that is used for
1856    newline. It is either LF (10) or CR (13).
1860    The output is an integer that contains the number of bytes used for internal
1861    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
1865    The output is an integer that contains the threshold above which the POSIX
1866    interface uses malloc() for output vectors. See item 31 above.
1870    The output is an unsigned integer that contains the default limit of the number
1871    of match() calls in a pcre_exec() execution. See 42 above.
1873    44. pcretest has been upgraded by the addition of the -C option. This causes it
1874    to extract all the available output from the new pcre_config() function, and to
1875    output it. The program then exits immediately.
1877    45. A need has arisen to pass over additional data with calls to pcre_exec() in
1878    order to support additional features. One way would have been to define
1879    pcre_exec2() (for example) with extra arguments, but this would not have been
1880    extensible, and would also have required all calls to the original function to
1881    be mapped to the new one. Instead, I have chosen to extend the mechanism that
1882    is used for passing in "extra" data from pcre_study().
1884    The pcre_extra structure is now exposed and defined in pcre.h. It currently
1885    contains the following fields:
1887      flags         a bitmap indicating which of the following fields are set
1888      study_data    opaque data from pcre_study()
1889      match_limit   a way of specifying a limit on match() calls for a specific
1890                      call to pcre_exec()
1891      callout_data  data for callouts (see 49 below)
1893    The flag bits are also defined in pcre.h, and are
1899    The pcre_study() function now returns one of these new pcre_extra blocks, with
1900    the actual study data pointed to by the study_data field, and the
1901    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
1902    before. That is, this change is entirely upwards-compatible and requires no
1903    change to existing code.
1905    If you want to pass in additional data to pcre_exec(), you can either place it
1906    in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
1907    block.
1909    46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
1910    data string contains the escape sequence \M, pcretest calls pcre_exec() several
1911    times with different match limits, until it finds the minimum value needed for
1912    pcre_exec() to complete. The value is then output. This can be instructive; for
1913    most simple matches the number is quite small, but for pathological cases it
1914    gets very large very quickly.
1916    47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
1917    returns the size of the data block pointed to by the study_data field in a
1918    pcre_extra block, that is, the value that was passed as the argument to
1919    pcre_malloc() when PCRE was getting memory in which to place the information
1920    created by pcre_study(). The fourth argument should point to a size_t variable.
1921    pcretest has been extended so that this information is shown after a successful
1922    pcre_study() call when information about the compiled regex is being displayed.
1924    48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
1925    because what follows is always an absolute path. (Later: it turns out that this
1926    is more than cosmetic for MinGW, because it doesn't like empty path
1927    components.)
1929    49. Some changes have been made to the callout feature (see 28 above):
1931    (i)  A callout function now has three choices for what it returns:
1933           0  =>  success, carry on matching
1934         > 0  =>  failure at this point, but backtrack if possible
1935         < 0  =>  serious error, return this value from pcre_exec()
1937         Negative values should normally be chosen from the set of PCRE_ERROR_xxx
1938         values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
1939         "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
1940         use by callout functions. It will never be used by PCRE itself.
1942    (ii) The pcre_extra structure (see 45 above) has a void * field called
1943         callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
1944         pcre_callout_block structure has a field of the same name. The contents of
1945         the field passed in the pcre_extra structure are passed to the callout
1946         function in the corresponding field in the callout block. This makes it
1947         easier to use the same callout-containing regex from multiple threads. For
1948         testing, the pcretest program has a new data escape
1950           \C*n        pass the number n (may be negative) as callout_data
1952         If the callout function in pcretest receives a non-zero value as
1953         callout_data, it returns that value.
1955    50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
1956    there were some redundant $(CFLAGS) in commands that are now specified as
1957    $(LINK), which already includes $(CFLAGS).
1959    51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
1960    has been compiled with UTF-8 support *and* pcre_compile() has been compiled
1961    with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
1962    one-byte characters throughout. Note that case-insensitive matching applies
1963    only to characters whose values are less than 256. PCRE doesn't support the
1964    notion of cases for higher-valued characters.
1966    (i)   A character class whose characters are all within 0-255 is handled as
1967          a bit map, and the map is inverted for negative classes. Previously, a
1968          character > 255 always failed to match such a class; however it should
1969          match if the class was a negative one (e.g. [^ab]). This has been fixed.
1971    (ii)  A negated character class with a single character < 255 is coded as
1972          "not this character" (OP_NOT). This wasn't working properly when the test
1973          character was multibyte, either singly or repeated.
1975    (iii) Repeats of multibyte characters are now handled correctly in UTF-8
1976          mode, for example: \x{100}{2,3}.
1978    (iv)  The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
1979          singly or repeated) now correctly test multibyte characters. However,
1980          PCRE doesn't recognize any characters with values greater than 255 as
1981          digits, spaces, or word characters. Such characters always match \D, \S,
1982          and \W, and never match \d, \s, or \w.
1984    (v)   Classes may now contain characters and character ranges with values
1985          greater than 255. For example: [ab\x{100}-\x{400}].
1987    (vi)  pcregrep now has a --utf-8 option (synonym -u) which makes it call
1988          PCRE in UTF-8 mode.
1990    52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
1991    PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
1992    retained for backwards compatibility. (Note that LASTLITERAL is also a byte
1993    value.)
1995    53. The single man page has become too large. I have therefore split it up into
1996    a number of separate man pages. These also give rise to individual HTML pages;
1997    these are now put in a separate directory, and there is an index.html page that
1998    lists them all. Some hyperlinking between the pages has been installed.
2000    54. Added convenience functions for handling named capturing parentheses.
2002    55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2003    aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2004    true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2005    are faulted.
2007    56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2008    calling configure. These values are used when compiling the dftables.c program
2009    which is run to generate the source of the default character tables. They
2010    default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2011    you will need to set these values.
2013    57. Updated the building process for Windows DLL, as provided by Fred Cox.
2016    Version 3.9 02-Jan-02
2017    ---------------------
2019    1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2021    2. If --disable-static was given, the building process failed when trying to
2022    build pcretest and pcregrep. (For some reason it was using libtool to compile
2023    them, which is not right, as they aren't part of the library.)
2026    Version 3.8 18-Dec-01
2027    ---------------------
2029    1. The experimental UTF-8 code was completely screwed up. It was packing the
2030    bytes in the wrong order. How dumb can you get?
2033    Version 3.7 29-Oct-01
2034    ---------------------
2036    1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2037    This caused pcretest, when used on the test data, to segfault. Unfortunately,
2038    this didn't happen under Solaris 8, where I normally test things.
2040    2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2041    doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2042    isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2043    this fix an hour or so after the initial 3.7 release.)
2046    Version 3.6 23-Oct-01
2047    ---------------------
2049    1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2050    offsets passed as NULL with zero offset count.
2052    2. The config.guess and config.sub files had not been updated when I moved to
2053    the latest autoconf.
2056    Version 3.5 15-Aug-01
2057    ---------------------
2059    1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2060    had been forgotten.
2062    2. By using declared but undefined structures, we can avoid using "void"
2063    definitions in pcre.h while keeping the internal definitions of the structures
2064    private.
2066    3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2067    user point of view, this means that both static and shared libraries are built
2068    by default, but this can be individually controlled. More of the work of
2069    handling this static/shared cases is now inside libtool instead of PCRE's make
2070    file.
2072    4. The pcretest utility is now installed along with pcregrep because it is
2073    useful for users (to test regexs) and by doing this, it automatically gets
2074    relinked by libtool. The documentation has been turned into a man page, so
2075    there are now .1, .txt, and .html versions in /doc.
2077    5. Upgrades to pcregrep:
2078       (i)   Added long-form option names like gnu grep.
2079       (ii)  Added --help to list all options with an explanatory phrase.
2080       (iii) Added -r, --recursive to recurse into sub-directories.
2081       (iv)  Added -f, --file to read patterns from a file.
2083    6. pcre_exec() was referring to its "code" argument before testing that
2084    argument for NULL (and giving an error if it was NULL).
2086    7. Upgraded Makefile.in to allow for compiling in a different directory from
2087    the source directory.
2089    8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2090    options bits, the pointer it was passed was to an int instead of to an unsigned
2091    long int. This mattered only on 64-bit systems.
2093    9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2094    generated) instead of pcre.in, which it its source. Also made the same change
2095    in several of the .c files.
2097    10. A new release of gcc defines printf() as a macro, which broke pcretest
2098    because it had an ifdef in the middle of a string argument for printf(). Fixed
2099    by using separate calls to printf().
2101    11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2102    script, to force use of CR or LF instead of \n in the source. On non-Unix
2103    systems, the value can be set in config.h.
2105    12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2106    absolute limit. Changed the text of the error message to make this clear, and
2107    likewise updated the man page.
2109    13. The limit of 99 on the number of capturing subpatterns has been removed.
2110    The new limit is 65535, which I hope will not be a "real" limit.
2113    Version 3.4 22-Aug-00
2114    ---------------------
2116    1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2118    2. Diagnose condition (?(0) as an error instead of crashing on matching.
2121    Version 3.3 01-Aug-00
2122    ---------------------
2124    1. If an octal character was given, but the value was greater than \377, it
2125    was not getting masked to the least significant bits, as documented. This could
2126    lead to crashes in some systems.
2128    2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
2129    the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
2131    3. Added the functions pcre_free_substring() and pcre_free_substring_list().
2132    These just pass their arguments on to (pcre_free)(), but they are provided
2133    because some uses of PCRE bind it to non-C systems that can call its functions,
2134    but cannot call free() or pcre_free() directly.
2136    4. Add "make test" as a synonym for "make check". Corrected some comments in
2137    the Makefile.
2139    5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
2140    Makefile.
2142    6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
2143    command called pgrep for grepping around the active processes.
2145    7. Added the beginnings of support for UTF-8 character strings.
2147    8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
2148    RANLIB to ./ltconfig so that they are used by libtool. I think these are all
2149    the relevant ones. (AR is not passed because ./ltconfig does its own figuring
2150    out for the ar command.)
2153    Version 3.2 12-May-00
2154    ---------------------
2156    This is purely a bug fixing release.
2158    1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
2159    of ZA. This was just one example of several cases that could provoke this bug,
2160    which was introduced by change 9 of version 2.00. The code for breaking
2161    infinite loops after an iteration that matches an empty string was't working
2162    correctly.
2164    2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
2165    when matched against abbab (for example). After matching an empty string, it
2166    wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
2167    caused it to match further down the string than it should.
2169    3. The code contained an inclusion of sys/types.h. It isn't clear why this
2170    was there because it doesn't seem to be needed, and it causes trouble on some
2171    systems, as it is not a Standard C header. It has been removed.
2173    4. Made 4 silly changes to the source to avoid stupid compiler warnings that
2174    were reported on the Macintosh. The changes were from
2176      while ((c = *(++ptr)) != 0 && c != '\n');
2177    to
2178      while ((c = *(++ptr)) != 0 && c != '\n') ;
2180    Totally extraordinary, but if that's what it takes...
2182    5. PCRE is being used in one environment where neither memmove() nor bcopy() is
2183    available. Added HAVE_BCOPY and an autoconf test for it; if neither
2184    HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
2185    assumes the way PCRE uses memmove() (always moving upwards).
2187    6. PCRE is being used in one environment where strchr() is not available. There
2188    was only one use in pcre.c, and writing it out to avoid strchr() probably gives
2189    faster code anyway.
2192    Version 3.1 09-Feb-00
2193    ---------------------
2195    The only change in this release is the fixing of some bugs in Makefile.in for
2196    the "install" target:
2198    (1) It was failing to install pcreposix.h.
2200    (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
2203    Version 3.0 01-Feb-00
2204    ---------------------
2206    1. Add support for the /+ modifier to perltest (to output $` like it does in
2207    pcretest).
2209    2. Add support for the /g modifier to perltest.
2211    3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
2212    matches null strings.
2214    4. Fix perltest so that it doesn't do unwanted things when fed an empty
2215    pattern. Perl treats empty patterns specially - it reuses the most recent
2216    pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
2217    effect.
2219    5. The POSIX interface was broken in that it was just handing over the POSIX
2220    captured string vector to pcre_exec(), but (since release 2.00) PCRE has
2221    required a bigger vector, with some working space on the end. This means that
2222    the POSIX wrapper now has to get and free some memory, and copy the results.
2224    6. Added some simple autoconf support, placing the test data and the
2225    documentation in separate directories, re-organizing some of the
2226    information files, and making it build pcre-config (a GNU standard). Also added
2227    libtool support for building PCRE as a shared library, which is now the
2228    default.
2230    7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
2231    09 are not valid octal constants. Single digits will be used for minor values
2232    less than 10.
2234    8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
2235    existing programs that set these in the POSIX interface can use PCRE without
2236    modification.
2238    9. Added a new function, pcre_fullinfo() with an extensible interface. It can
2239    return all that pcre_info() returns, plus additional data. The pcre_info()
2240    function is retained for compatibility, but is considered to be obsolete.
2242    10. Added experimental recursion feature (?R) to handle one common case that
2243    Perl 5.6 will be able to do with (?p{...}).
2245    11. Added support for POSIX character classes like [:alpha:], which Perl is
2246    adopting.
2249    Version 2.08 31-Aug-99
2250    ----------------------
2252    1. When startoffset was not zero and the pattern began with ".*", PCRE was not
2253    trying to match at the startoffset position, but instead was moving forward to
2254    the next newline as if a previous match had failed.
2256    2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
2257    and could get into a loop if a null string was matched other than at the start
2258    of the subject.
2260    3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
2261    be distinguished at compile time, and for completeness also added PCRE_DATE.
2263    5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
2264    in GnuWin32 environments.
2267    Version 2.07 29-Jul-99
2268    ----------------------
2270    1. The documentation is now supplied in plain text form and HTML as well as in
2271    the form of man page sources.
2273    2. C++ compilers don't like assigning (void *) values to other pointer types.
2274    In particular this affects malloc(). Although there is no problem in Standard
2275    C, I've put in casts to keep C++ compilers happy.
2277    3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
2278    should be (const char *).
2280    4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
2281    be useful for non-Unix systems who don't want to bother with the POSIX stuff.
2282    However, I haven't made this a standard facility. The documentation doesn't
2283    mention it, and the Makefile doesn't support it.
2285    5. The Makefile now contains an "install" target, with editable destinations at
2286    the top of the file. The pcretest program is not installed.
2288    6. pgrep -V now gives the PCRE version number and date.
2290    7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
2291    causing the entire string to be ignored, instead of just the last character.
2293    8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
2294    non-matching string, it can take a very, very long time, even for strings of
2295    quite modest length, because of the nested recursion. PCRE now does better in
2296    some of these cases. It does this by remembering the last required literal
2297    character in the pattern, and pre-searching the subject to ensure it is present
2298    before running the real match. In other words, it applies a heuristic to detect
2299    some types of certain failure quickly, and in the above example, if presented
2300    with a string that has no trailing " it gives "no match" very quickly.
2302    9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
2303    other alternatives are tried instead.
2306    Version 2.06 09-Jun-99
2307    ----------------------
2309    1. Change pcretest's output for amount of store used to show just the code
2310    space, because the remainder (the data block) varies in size between 32-bit and
2311    64-bit systems.
2313    2. Added an extra argument to pcre_exec() to supply an offset in the subject to
2314    start matching at. This allows lookbehinds to work when searching for multiple
2315    occurrences in a string.
2317    3. Added additional options to pcretest for testing multiple occurrences:
2319       /+   outputs the rest of the string that follows a match
2320       /g   loops for multiple occurrences, using the new startoffset argument
2321       /G   loops for multiple occurrences by passing an incremented pointer
2323    4. PCRE wasn't doing the "first character" optimization for patterns starting
2324    with \b or \B, though it was doing it for other lookbehind assertions. That is,
2325    it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
2326    the letter 'x'. On long subject strings, this gives a significant speed-up.
2329    Version 2.05 21-Apr-99
2330    ----------------------
2332    1. Changed the type of magic_number from int to long int so that it works
2333    properly on 16-bit systems.
2335    2. Fixed a bug which caused patterns starting with .* not to work correctly
2336    when the subject string contained newline characters. PCRE was assuming
2337    anchoring for such patterns in all cases, which is not correct because .* will
2338    not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
2339    DOTALL is set at top level; otherwise it knows that patterns starting with .*
2340    must be retried after every newline in the subject.
2343    Version 2.04 18-Feb-99
2344    ----------------------
2346    1. For parenthesized subpatterns with repeats whose minimum was zero, the
2347    computation of the store needed to hold the pattern was incorrect (too large).
2348    If such patterns were nested a few deep, this could multiply and become a real
2349    problem.
2351    2. Added /M option to pcretest to show the memory requirement of a specific
2352    pattern. Made -m a synonym of -s (which does this globally) for compatibility.
2354    3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
2355    compiled in such a way that the backtracking after subsequent failure was
2356    pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
2357    ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
2360    Version 2.03 02-Feb-99
2361    ----------------------
2363    1. Fixed typo and small mistake in man page.
2365    2. Added 4th condition (GPL supersedes if conflict) and created separate
2366    LICENCE file containing the conditions.
2368    3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
2369    Perl, that is the internal \ allows the delimiter to be included in the
2370    pattern. Locked out the use of \ as a delimiter. If \ immediately follows
2371    the final delimiter, add \ to the end of the pattern (to test the error).
2373    4. Added the convenience functions for extracting substrings after a successful
2374    match. Updated pcretest to make it able to test these functions.
2377  Version 2.02 14-Jan-99  Version 2.02 14-Jan-99
2378  ----------------------  ----------------------

Removed from v.27  
changed lines
  Added in v.176

  ViewVC Help
Powered by ViewVC 1.1.5