ViewVC logotype

Diff of /code/trunk/ChangeLog

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

revision 49 by nigel, Sat Feb 24 21:39:33 2007 UTC revision 161 by ph10, Thu May 3 10:28:07 2007 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
4    Version 7.2 01-May-07
5    ---------------------
7     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
8        which is apparently normally available under Windows.
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.
21    Version 7.1 24-Apr-07
22    ---------------------
24     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
25        that is more "standard", making use of automake and other Autotools. There
26        is some re-arrangement of the files and adjustment of comments consequent
27        on this.
29     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
30        for recursive directory scanning broke on some systems because the files
31        are not scanned in any specific order and on different systems the order
32        was different. A call to "sort" has been inserted into RunGrepTest for the
33        approprate test as a short-term fix. In the longer term there may be an
34        alternative.
36     3. I had an email from Eric Raymond about problems translating some of PCRE's
37        man pages to HTML (despite the fact that I distribute HTML pages, some
38        people do their own conversions for various reasons). The problems
39        concerned the use of low-level troff macros .br and .in. I have therefore
40        removed all such uses from the man pages (some were redundant, some could
41        be replaced by .nf/.fi pairs). The 132html script that I use to generate
42        HTML has been updated to handle .nf/.fi and to complain if it encounters
43        .br or .in.
45     4. Updated comments in configure.ac that get placed in config.h.in and also
46        arranged for config.h to be included in the distribution, with the name
47        config.h.generic, for the benefit of those who have to compile without
48        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
50     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
51        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
52        makevp.bat for the current PCRE, using the additional files
53        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
55     6. A Windows user reported a minor discrepancy with test 2, which turned out
56        to be caused by a trailing space on an input line that had got lost in his
57        copy. The trailing space was an accident, so I've just removed it.
59     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
60        that is needed.
62     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
63        as "const" (a) because they are and (b) because it helps the PHP
64        maintainers who have recently made a script to detect big data structures
65        in the php code that should be moved to the .rodata section. I remembered
66        to update Builducptable as well, so it won't revert if ucptable.h is ever
67        re-created.
69     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
70        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
71        order to be able to cut out the UTF-8 tables in the latter when UTF-8
72        support is not required. This saves 1.5-2K of code, which is important in
73        some applications.
75        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
76        so as not to refer to the tables, even though these functions will never be
77        called when UTF-8 support is disabled. Otherwise there are problems with a
78        shared library.
80    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
82        (a) It was defining its arguments as char * instead of void *.
84        (b) It was assuming that all moves were upwards in memory; this was true
85            a long time ago when I wrote it, but is no longer the case.
87        The emulated memove() is provided for those environments that have neither
88        memmove() nor bcopy(). I didn't think anyone used it these days, but that
89        is clearly not the case, as these two bugs were recently reported.
91    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
92        and Detrail to create the HTML documentation, the .txt form of the man
93        pages, and it removes trailing spaces from listed files. It also creates
94        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
95        case, it wraps all the #defines with #ifndefs. This script should be run
96        before "make dist".
98    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
99        with Unicode property support.
101        (a) For a maximizing quantifier, if the two different cases of the
102            character were of different lengths in their UTF-8 codings (there are
103            some cases like this - I found 11), and the matching function had to
104            back up over a mixture of the two cases, it incorrectly assumed they
105            were both the same length.
107        (b) When PCRE was configured to use the heap rather than the stack for
108            recursion during matching, it was not correctly preserving the data for
109            the other case of a UTF-8 character when checking ahead for a match
110            while processing a minimizing repeat. If the check also involved
111            matching a wide character, but failed, corruption could cause an
112            erroneous result when trying to check for a repeat of the original
113            character.
115    13. Some tidying changes to the testing mechanism:
117        (a) The RunTest script now detects the internal link size and whether there
118            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
119            values substituted by "configure". (The RunGrepTest script already did
120            this for UTF-8.) The configure.ac script no longer substitutes the
121            relevant variables.
123        (b) The debugging options /B and /D in pcretest show the compiled bytecode
124            with length and offset values. This means that the output is different
125            for different internal link sizes. Test 2 is skipped for link sizes
126            other than 2 because of this, bypassing the problem. Unfortunately,
127            there was also a test in test 3 (the locale tests) that used /B and
128            failed for link sizes other than 2. Rather than cut the whole test out,
129            I have added a new /Z option to pcretest that replaces the length and
130            offset values with spaces. This is now used to make test 3 independent
131            of link size. (Test 2 will be tidied up later.)
133    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
134        segmentation fault instead of returning the appropriate error message.
136    15. In multiline mode when the newline sequence was set to "any", the pattern
137        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
138        This doesn't seem right; it now treats the CRLF combination as the line
139        ending, and so does not match in that case. It's only a pattern such as ^$
140        that would hit this one: something like ^ABC$ would have failed after \r
141        and then tried again after \r\n.
143    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
144        in an attempt to make files that differ only in their line terminators
145        compare equal. This works on Linux.
147    17. Under certain error circumstances pcregrep might try to free random memory
148        as it exited. This is now fixed, thanks to valgrind.
150    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
151        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
152        was because its rules for how to advance for /g after matching an empty
153        string at the end of a line did not allow for this case. They now check for
154        it specially.
156    20. pcretest is supposed to handle patterns and data of any length, by
157        extending its buffers when necessary. It was getting this wrong when the
158        buffer for a data line had to be extended.
160    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
161        CRLF as a newline sequence.
163    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
164        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
165        I have nevertheless tidied it up.
167    23. Added some casts to kill warnings from HP-UX ia64 compiler.
169    24. Added a man page for pcre-config.
172    Version 7.0 19-Dec-06
173    ---------------------
175     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
176        moving to gcc 4.1.1.
178     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
179        sys/time.h, which is documented as needed for this function. It doesn't
180        seem to matter on Linux, but it showed up on some releases of OS X.
182     3. It seems that there are systems where bytes whose values are greater than
183        127 match isprint() in the "C" locale. The "C" locale should be the
184        default when a C program starts up. In most systems, only ASCII printing
185        characters match isprint(). This difference caused the output from pcretest
186        to vary, making some of the tests fail. I have changed pcretest so that:
188        (a) When it is outputting text in the compiled version of a pattern, bytes
189            other than 32-126 are always shown as hex escapes.
191        (b) When it is outputting text that is a matched part of a subject string,
192            it does the same, unless a different locale has been set for the match
193            (using the /L modifier). In this case, it uses isprint() to decide.
195     4. Fixed a major bug that caused incorrect computation of the amount of memory
196        required for a compiled pattern when options that changed within the
197        pattern affected the logic of the preliminary scan that determines the
198        length. The relevant options are -x, and -i in UTF-8 mode. The result was
199        that the computed length was too small. The symptoms of this bug were
200        either the PCRE error "internal error: code overflow" from pcre_compile(),
201        or a glibc crash with a message such as "pcretest: free(): invalid next
202        size (fast)". Examples of patterns that provoked this bug (shown in
203        pcretest format) are:
205          /(?-x: )/x
206          /(?x)(?-x: \s*#\s*)/
207          /((?i)[\x{c0}])/8
208          /(?i:[\x{c0}])/8
210        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
211        is now done differently.
213     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
214        wrapper classes; (b) implement a new function in the C++ scanner that is
215        more efficient than the old way of doing things because it avoids levels of
216        recursion in the regex matching; (c) add a paragraph to the documentation
217        for the FullMatch() function.
219     6. The escape sequence \n was being treated as whatever was defined as
220        "newline". Not only was this contrary to the documentation, which states
221        that \n is character 10 (hex 0A), but it also went horribly wrong when
222        "newline" was defined as CRLF. This has been fixed.
224     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
225        was being set to -1 for the "end of line" case (supposedly a value that no
226        character can have). Though this value is never used (the check for end of
227        line is "zero bytes in current character"), it caused compiler complaints.
228        I've changed it to 0xffffffff.
230     8. In pcre_version.c, the version string was being built by a sequence of
231        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
232        string (as it is for production releases) called a macro with an empty
233        argument. The C standard says the result of this is undefined. The gcc
234        compiler treats it as an empty string (which was what was wanted) but it is
235        reported that Visual C gives an error. The source has been hacked around to
236        avoid this problem.
238     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
239        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
240        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
241        of them did).
243    10. Originally, pcretest opened its input and output without "b"; then I was
244        told that "b" was needed in some environments, so it was added for release
245        5.0 to both the input and output. (It makes no difference on Unix-like
246        systems.) Later I was told that it is wrong for the input on Windows. I've
247        now abstracted the modes into two macros, to make it easier to fiddle with
248        them, and removed "b" from the input mode under Windows.
250    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
252    12. Added -help and --help to pcretest as an official way of being reminded
253        of the options.
255    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
256        and pcrecpp.cc because they annoy compilers at high warning levels.
258    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
260    15. Fixed an occurrence of == in configure.ac that should have been = (shell
261        scripts are not C programs :-) and which was not noticed because it works
262        on Linux.
264    16. pcretest is supposed to handle any length of pattern and data line (as one
265        line or as a continued sequence of lines) by extending its input buffer if
266        necessary. This feature was broken for very long pattern lines, leading to
267        a string of junk being passed to pcre_compile() if the pattern was longer
268        than about 50K.
270    17. I have done a major re-factoring of the way pcre_compile() computes the
271        amount of memory needed for a compiled pattern. Previously, there was code
272        that made a preliminary scan of the pattern in order to do this. That was
273        OK when PCRE was new, but as the facilities have expanded, it has become
274        harder and harder to keep it in step with the real compile phase, and there
275        have been a number of bugs (see for example, 4 above). I have now found a
276        cunning way of running the real compile function in a "fake" mode that
277        enables it to compute how much memory it would need, while actually only
278        ever using a few hundred bytes of working memory and without too many
279        tests of the mode. This should make future maintenance and development
280        easier. A side effect of this work is that the limit of 200 on the nesting
281        depth of parentheses has been removed (though this was never a serious
282        limitation, I suspect). However, there is a downside: pcre_compile() now
283        runs more slowly than before (30% or more, depending on the pattern). I
284        hope this isn't a big issue. There is no effect on runtime performance.
286    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
287        newline (only possible for the last line of a file) and it was a
288        pattern that set a locale (followed by /Lsomething), pcretest crashed.
290    19. Added additional timing features to pcretest. (1) The -tm option now times
291        matching only, not compiling. (2) Both -t and -tm can be followed, as a
292        separate command line item, by a number that specifies the number of
293        repeats to use when timing. The default is 50000; this gives better
294        precision, but takes uncomfortably long for very large patterns.
296    20. Extended pcre_study() to be more clever in cases where a branch of a
297        subpattern has no definite first character. For example, (a*|b*)[cd] would
298        previously give no result from pcre_study(). Now it recognizes that the
299        first character must be a, b, c, or d.
301    21. There was an incorrect error "recursive call could loop indefinitely" if
302        a subpattern (or the entire pattern) that was being tested for matching an
303        empty string contained only one non-empty item after a nested subpattern.
304        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
305        incorrectly, because the \d was being skipped in the check.
307    22. The pcretest program now has a new pattern option /B and a command line
308        option -b, which is equivalent to adding /B to every pattern. This causes
309        it to show the compiled bytecode, without the additional information that
310        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
311        is the same as /B/I).
313    23. A new optimization is now able automatically to treat some sequences such
314        as a*b as a*+b. More specifically, if something simple (such as a character
315        or a simple class like \d) has an unlimited quantifier, and is followed by
316        something that cannot possibly match the quantified thing, the quantifier
317        is automatically "possessified".
319    24. A recursive reference to a subpattern whose number was greater than 39
320        went wrong under certain circumstances in UTF-8 mode. This bug could also
321        have affected the operation of pcre_study().
323    25. Realized that a little bit of performance could be had by replacing
324        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
326    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
328    27. Possessive quantifiers such as a++ were previously implemented by turning
329        them into atomic groups such as ($>a+). Now they have their own opcodes,
330        which improves performance. This includes the automatically created ones
331        from 23 above.
333    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
334        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
335        the first matched character to be a colon. This applied both to named and
336        numbered groups.
338    29. The ucpinternal.h header file was missing its idempotency #ifdef.
340    30. I was sent a "project" file called libpcre.a.dev which I understand makes
341        building PCRE on Windows easier, so I have included it in the distribution.
343    31. There is now a check in pcretest against a ridiculously large number being
344        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
345        loop, the loop is abandoned.
347    32. Forward references to subpatterns in conditions such as (?(2)...) where
348        subpattern 2 is defined later cause pcre_compile() to search forwards in
349        the pattern for the relevant set of parentheses. This search went wrong
350        when there were unescaped parentheses in a character class, parentheses
351        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
353    33. "Subroutine" calls and backreferences were previously restricted to
354        referencing subpatterns earlier in the regex. This restriction has now
355        been removed.
357    34. Added a number of extra features that are going to be in Perl 5.10. On the
358        whole, these are just syntactic alternatives for features that PCRE had
359        previously implemented using the Python syntax or my own invention. The
360        other formats are all retained for compatibility.
362        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
363            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
364            also .NET compatible.
366        (b) A recursion or subroutine call to a named group can now be defined as
367            (?&name) as well as (?P>name).
369        (c) A backreference to a named group can now be defined as \k<name> or
370            \k'name' as well as (?P=name). The new forms, as well as being in Perl
371            5.10, are also .NET compatible.
373        (d) A conditional reference to a named group can now use the syntax
374            (?(<name>) or (?('name') as well as (?(name).
376        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
377            groups (named and numbered) that are never evaluated inline, but can be
378            called as "subroutines" from elsewhere. In effect, the DEFINE condition
379            is always false. There may be only one alternative in such a group.
381        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
382            as the simple (?(R). The condition is true only if the most recent
383            recursion is that of the given number or name. It does not search out
384            through the entire recursion stack.
386        (g) The escape \gN or \g{N} has been added, where N is a positive or
387            negative number, specifying an absolute or relative reference.
389    35. Tidied to get rid of some further signed/unsigned compiler warnings and
390        some "unreachable code" warnings.
392    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
393        things, this adds five new scripts.
395    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
396        There were also incompatibilities regarding the handling of \Q..\E inside
397        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
398        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
400    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
401        matches an empty string, and forcibly breaks the loop. There were bugs in
402        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
403        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
404        separate and independent bugs (that affected different cases) have been
405        fixed.
407    39. Refactored the code to abolish the use of different opcodes for small
408        capturing bracket numbers. This is a tidy that I avoided doing when I
409        removed the limit on the number of capturing brackets for 3.5 back in 2001.
410        The new approach is not only tidier, it makes it possible to reduce the
411        memory needed to fix the previous bug (38).
413    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
414        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
415        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
416        mode.
418    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
419        report.
421    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
422        copy construction and assignment in the C++ wrapper.
424    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
425        couple of bugs that could have given wrong results in the "--newline=crlf"
426        case.
428    44. Added a number of casts and did some reorganization of signed/unsigned int
429        variables following suggestions from Dair Grant. Also renamed the variable
430        "this" as "item" because it is a C++ keyword.
432    45. Arranged for dftables to add
434          #include "pcre_internal.h"
436        to pcre_chartables.c because without it, gcc 4.x may remove the array
437        definition from the final binary if PCRE is built into a static library and
438        dead code stripping is activated.
440    46. For an unanchored pattern, if a match attempt fails at the start of a
441        newline sequence, and the newline setting is CRLF or ANY, and the next two
442        characters are CRLF, advance by two characters instead of one.
445    Version 6.7 04-Jul-06
446    ---------------------
448     1. In order to handle tests when input lines are enormously long, pcretest has
449        been re-factored so that it automatically extends its buffers when
450        necessary. The code is crude, but this _is_ just a test program. The
451        default size has been increased from 32K to 50K.
453     2. The code in pcre_study() was using the value of the re argument before
454        testing it for NULL. (Of course, in any sensible call of the function, it
455        won't be NULL.)
457     3. The memmove() emulation function in pcre_internal.h, which is used on
458        systems that lack both memmove() and bcopy() - that is, hardly ever -
459        was missing a "static" storage class specifier.
461     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
462        containing an extended class (one that cannot be represented by a bitmap
463        because it contains high-valued characters or Unicode property items, e.g.
464        [\pZ]). Almost always one would set UTF-8 mode when processing such a
465        pattern, but PCRE should not loop if you do not (it no longer does).
466        [Detail: two cases were found: (a) a repeated subpattern containing an
467        extended class; (b) a recursive reference to a subpattern that followed a
468        previous extended class. It wasn't skipping over the extended class
469        correctly when UTF-8 mode was not set.]
471     5. A negated single-character class was not being recognized as fixed-length
472        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
473        compile error "lookbehind assertion is not fixed length".
475     6. The RunPerlTest auxiliary script was showing an unexpected difference
476        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
477        write a Perl script that can interpret lines of an input file either as
478        byte characters or as UTF-8, which is what "perltest" was being required to
479        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
480        can't do is switch easily at run time between having the "use utf8;" pragma
481        or not. In the end, I fudged it by using the RunPerlTest script to insert
482        "use utf8;" explicitly for the UTF-8 tests.
484     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
485        the end of the subject string, contrary to the documentation and to what
486        Perl does. This was true of both matching functions. Now it matches only at
487        the start of the subject and immediately after *internal* newlines.
489     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
490        a pointer to an int instead of a pointer to an unsigned long int. This
491        caused problems on 64-bit systems.
493     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
494        instance of the 'standard' template library not being so standard".
496    10. There was no check on the number of named subpatterns nor the maximum
497        length of a subpattern name. The product of these values is used to compute
498        the size of the memory block for a compiled pattern. By supplying a very
499        long subpattern name and a large number of named subpatterns, the size
500        computation could be caused to overflow. This is now prevented by limiting
501        the length of names to 32 characters, and the number of named subpatterns
502        to 10,000.
504    11. Subpatterns that are repeated with specific counts have to be replicated in
505        the compiled pattern. The size of memory for this was computed from the
506        length of the subpattern and the repeat count. The latter is limited to
507        65535, but there was no limit on the former, meaning that integer overflow
508        could in principle occur. The compiled length of a repeated subpattern is
509        now limited to 30,000 bytes in order to prevent this.
511    12. Added the optional facility to have named substrings with the same name.
513    13. Added the ability to use a named substring as a condition, using the
514        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
515        are numbers (not recommended). Forward references are permitted.
517    14. Added forward references in named backreferences (if you see what I mean).
519    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
520        pattern could run off the end of the subject. For example, the pattern
521        "(?s)(.{1,5})"8 did this with the subject "ab".
523    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
524        PCRE_CASELESS was set when matching characters that were quantified with ?
525        or *.
527    17. A character class other than a single negated character that had a minimum
528        but no maximum quantifier - for example [ab]{6,} - was not handled
529        correctly by pce_dfa_exec(). It would match only one character.
531    18. A valid (though odd) pattern that looked like a POSIX character
532        class but used an invalid character after [ (for example [[,abc,]]) caused
533        pcre_compile() to give the error "Failed: internal error: code overflow" or
534        in some cases to crash with a glibc free() error. This could even happen if
535        the pattern terminated after [[ but there just happened to be a sequence of
536        letters, a binary zero, and a closing ] in the memory that followed.
538    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
539        over the years. Originally (before any Unicode support), just the bottom 8
540        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
541        output from "man perlunicode" includes this:
543          The regular expression compiler produces polymorphic opcodes.  That
544          is, the pattern adapts to the data and automatically switches to
545          the Unicode character scheme when presented with Unicode data--or
546          instead uses a traditional byte scheme when presented with byte
547          data.
549        Sadly, a wide octal escape does not cause a switch, and in a string with
550        no other multibyte characters, these octal escapes are treated as before.
551        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
552        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
553        Unicode string.
555        I have not perpetrated such confusion in PCRE. Up till now, it took just
556        the bottom 8 bits, as in old Perl. I have now made octal escapes with
557        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
558        translate to the appropriate multibyte character.
560    29. Applied some refactoring to reduce the number of warnings from Microsoft
561        and Borland compilers. This has included removing the fudge introduced
562        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
563        a warning about an unused variable.
565    21. PCRE has not included VT (character 0x0b) in the set of whitespace
566        characters since release 4.0, because Perl (from release 5.004) does not.
567        [Or at least, is documented not to: some releases seem to be in conflict
568        with the documentation.] However, when a pattern was studied with
569        pcre_study() and all its branches started with \s, PCRE still included VT
570        as a possible starting character. Of course, this did no harm; it just
571        caused an unnecessary match attempt.
573    22. Removed a now-redundant internal flag bit that recorded the fact that case
574        dependency changed within the pattern. This was once needed for "required
575        byte" processing, but is no longer used. This recovers a now-scarce options
576        bit. Also moved the least significant internal flag bit to the most-
577        significant bit of the word, which was not previously used (hangover from
578        the days when it was an int rather than a uint) to free up another bit for
579        the future.
581    23. Added support for CRLF line endings as well as CR and LF. As well as the
582        default being selectable at build time, it can now be changed at runtime
583        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
584        specify that it is scanning data with non-default line endings.
586    24. Changed the definition of CXXLINK to make it agree with the definition of
587        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
589    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
590        recursions. This makes a big different to stack usage for some patterns.
592    26. If a subpattern containing a named recursion or subroutine reference such
593        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
594        the space required for the compiled pattern went wrong and gave too small a
595        value. Depending on the environment, this could lead to "Failed: internal
596        error: code overflow at offset 49" or "glibc detected double free or
597        corruption" errors.
599    27. Applied patches from Google (a) to support the new newline modes and (b) to
600        advance over multibyte UTF-8 characters in GlobalReplace.
602    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
603        difference for some implementation of PCRE in some Windows version.
605    29. Added some extra testing facilities to pcretest:
607        \q<number>   in a data line sets the "match limit" value
608        \Q<number>   in a data line sets the "match recursion limt" value
609        -S <number>  sets the stack size, where <number> is in megabytes
611        The -S option isn't available for Windows.
614    Version 6.6 06-Feb-06
615    ---------------------
617     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
618        in pcreposix.h. I have copied the definition from pcre.h.
620     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
621        because pcre.h is no longer a built file.
623     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
624        not normally included in the compiled code.
627    Version 6.5 01-Feb-06
628    ---------------------
630     1. When using the partial match feature with pcre_dfa_exec(), it was not
631        anchoring the second and subsequent partial matches at the new starting
632        point. This could lead to incorrect results. For example, with the pattern
633        /1234/, partially matching against "123" and then "a4" gave a match.
635     2. Changes to pcregrep:
637        (a) All non-match returns from pcre_exec() were being treated as failures
638            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
639            error message is output. Some extra information is given for the
640            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
641            probably the only errors that are likely to be caused by users (by
642            specifying a regex that has nested indefinite repeats, for instance).
643            If there are more than 20 of these errors, pcregrep is abandoned.
645        (b) A binary zero was treated as data while matching, but terminated the
646            output line if it was written out. This has been fixed: binary zeroes
647            are now no different to any other data bytes.
649        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
650            used to set a locale for matching. The --locale=xxxx long option has
651            been added (no short equivalent) to specify a locale explicitly on the
652            pcregrep command, overriding the environment variables.
654        (d) When -B was used with -n, some line numbers in the output were one less
655            than they should have been.
657        (e) Added the -o (--only-matching) option.
659        (f) If -A or -C was used with -c (count only), some lines of context were
660            accidentally printed for the final match.
662        (g) Added the -H (--with-filename) option.
664        (h) The combination of options -rh failed to suppress file names for files
665            that were found from directory arguments.
667        (i) Added the -D (--devices) and -d (--directories) options.
669        (j) Added the -F (--fixed-strings) option.
671        (k) Allow "-" to be used as a file name for -f as well as for a data file.
673        (l) Added the --colo(u)r option.
675        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
676            is not present by default.
678     3. A nasty bug was discovered in the handling of recursive patterns, that is,
679        items such as (?R) or (?1), when the recursion could match a number of
680        alternatives. If it matched one of the alternatives, but subsequently,
681        outside the recursion, there was a failure, the code tried to back up into
682        the recursion. However, because of the way PCRE is implemented, this is not
683        possible, and the result was an incorrect result from the match.
685        In order to prevent this happening, the specification of recursion has
686        been changed so that all such subpatterns are automatically treated as
687        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
689     4. I had overlooked the fact that, in some locales, there are characters for
690        which isalpha() is true but neither isupper() nor islower() are true. In
691        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
692        and ordfeminine) are like this. This affected the treatment of \w and \W
693        when they appeared in character classes, but not when they appeared outside
694        a character class. The bit map for "word" characters is now created
695        separately from the results of isalnum() instead of just taking it from the
696        upper, lower, and digit maps. (Plus the underscore character, of course.)
698     5. The above bug also affected the handling of POSIX character classes such as
699        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
700        permanent tables. Instead, the bit maps for such a class were previously
701        created as the appropriate unions of the upper, lower, and digit bitmaps.
702        Now they are created by subtraction from the [[:word:]] class, which has
703        its own bitmap.
705     6. The [[:blank:]] character class matches horizontal, but not vertical space.
706        It is created by subtracting the vertical space characters (\x09, \x0a,
707        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
708        subtraction was done in the overall bitmap for a character class, meaning
709        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
710        be recognized. This bug has been fixed.
712     7. Patches from the folks at Google:
714          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
715          real life, but is still worth protecting against".
717          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
718          regular expressions".
720          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
721          have it.
723          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
724          "configure" and the latter not, in order to fix a problem somebody had
725          with compiling the Arg class on HP-UX.
727          (e) Improve the error-handling of the C++ wrapper a little bit.
729          (f) New tests for checking recursion limiting.
731     8. The pcre_memmove() function, which is used only if the environment does not
732        have a standard memmove() function (and is therefore rarely compiled),
733        contained two bugs: (a) use of int instead of size_t, and (b) it was not
734        returning a result (though PCRE never actually uses the result).
736     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
737        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
738        returned instead of calling malloc() with an overflowing number that would
739        most likely cause subsequent chaos.
741    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
743    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
744        with this option is matched, the nmatch and pmatch options of regexec() are
745        ignored.
747    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
748        provided in case anyone wants to the the POSIX interface with UTF-8
749        strings.
751    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
752        C++ linking (needed for some HP-UX environments).
754    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
755        (unused parameter) and in the pcre_printint() function (omitted "default"
756        switch label when the default is to do nothing).
758    15. Added some code to make it possible, when PCRE is compiled as a C++
759        library, to replace subject pointers for pcre_exec() with a smart pointer
760        class, thus making it possible to process discontinuous strings.
762    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
763        much the same function. They were added by different people who were trying
764        to make PCRE easy to compile on non-Unix systems. It has been suggested
765        that PCRE_EXPORT be abolished now that there is more automatic apparatus
766        for compiling on Windows systems. I have therefore replaced it with
767        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
768        defaults to "extern" for C or "extern C" for C++, which works fine on
769        Unix-like systems. It is now possible to override the value of PCRE_DATA_
770        SCOPE with something explicit in config.h. In addition:
772        (a) pcreposix.h still had just "extern" instead of either of these macros;
773            I have replaced it with PCRE_DATA_SCOPE.
775        (b) Functions such as _pcre_xclass(), which are internal to the library,
776            but external in the C sense, all had PCRE_EXPORT in their definitions.
777            This is apparently wrong for the Windows case, so I have removed it.
778            (It makes no difference on Unix-like systems.)
780    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
781        of recursive calls to match(). This is different to MATCH_LIMIT because
782        that limits the total number of calls to match(), not all of which increase
783        the depth of recursion. Limiting the recursion depth limits the amount of
784        stack (or heap if NO_RECURSE is set) that is used. The default can be set
785        when PCRE is compiled, and changed at run time. A patch from Google adds
786        this functionality to the C++ interface.
788    18. Changes to the handling of Unicode character properties:
790        (a) Updated the table to Unicode 4.1.0.
792        (b) Recognize characters that are not in the table as "Cn" (undefined).
794        (c) I revised the way the table is implemented to a much improved format
795            which includes recognition of ranges. It now supports the ranges that
796            are defined in UnicodeData.txt, and it also amalgamates other
797            characters into ranges. This has reduced the number of entries in the
798            table from around 16,000 to around 3,000, thus reducing its size
799            considerably. I realized I did not need to use a tree structure after
800            all - a binary chop search is just as efficient. Having reduced the
801            number of entries, I extended their size from 6 bytes to 8 bytes to
802            allow for more data.
804        (d) Added support for Unicode script names via properties such as \p{Han}.
806    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
807        matching that character.
809    20. When matching a repeated Unicode property with a minimum greater than zero,
810        (for example \pL{2,}), PCRE could look past the end of the subject if it
811        reached it while seeking the minimum number of characters. This could
812        happen only if some of the characters were more than one byte long, because
813        there is a check for at least the minimum number of bytes.
815    21. Refactored the implementation of \p and \P so as to be more general, to
816        allow for more different types of property in future. This has changed the
817        compiled form incompatibly. Anybody with saved compiled patterns that use
818        \p or \P will have to recompile them.
820    22. Added "Any" and "L&" to the supported property types.
822    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
823        but give a compile time error if the value is greater than 0xff.
825    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
826        accidentally not being installed or uninstalled.
828    25. The pcre.h file was built from pcre.h.in, but the only changes that were
829        made were to insert the current release number. This seemed silly, because
830        it made things harder for people building PCRE on systems that don't run
831        "configure". I have turned pcre.h into a distributed file, no longer built
832        by "configure", with the version identification directly included. There is
833        no longer a pcre.h.in file.
835        However, this change necessitated a change to the pcre-config script as
836        well. It is built from pcre-config.in, and one of the substitutions was the
837        release number. I have updated configure.ac so that ./configure now finds
838        the release number by grepping pcre.h.
840    26. Added the ability to run the tests under valgrind.
843    Version 6.4 05-Sep-05
844    ---------------------
846     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
847        "--" to be printed when multiple files were scanned, even when none of the
848        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
849        consider it to be a bug, and have restored the previous behaviour.
851     2. A couple of code tidies to get rid of compiler warnings.
853     3. The pcretest program used to cheat by referring to symbols in the library
854        whose names begin with _pcre_. These are internal symbols that are not
855        really supposed to be visible externally, and in some environments it is
856        possible to suppress them. The cheating is now confined to including
857        certain files from the library's source, which is a bit cleaner.
859     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
860        file's purpose clearer.
862     5. Reorganized pcre_ucp_findchar().
865    Version 6.3 15-Aug-05
866    ---------------------
868     1. The file libpcre.pc.in did not have general read permission in the tarball.
870     2. There were some problems when building without C++ support:
872        (a) If C++ support was not built, "make install" and "make test" still
873            tried to test it.
875        (b) There were problems when the value of CXX was explicitly set. Some
876            changes have been made to try to fix these, and ...
878        (c) --disable-cpp can now be used to explicitly disable C++ support.
880        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
881            backslash in a target when C++ was disabled. This confuses some
882            versions of "make", apparently. Using an intermediate variable solves
883            this. (Same for CPP_LOBJ.)
885     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
886        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
887        necessary on certain architectures.
889     4. Added a setting of -export-symbols-regex to the link command to remove
890        those symbols that are exported in the C sense, but actually are local
891        within the library, and not documented. Their names all begin with
892        "_pcre_". This is not a perfect job, because (a) we have to except some
893        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
894        available (and never for static libraries). I have made a note to try to
895        find a way round (a) in the future.
898    Version 6.2 01-Aug-05
899    ---------------------
901     1. There was no test for integer overflow of quantifier values. A construction
902        such as {1111111111111111} would give undefined results. What is worse, if
903        a minimum quantifier for a parenthesized subpattern overflowed and became
904        negative, the calculation of the memory size went wrong. This could have
905        led to memory overwriting.
907     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
909     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
910        operating environments where this matters.
912     4. Applied Giuseppe Maxia's patch to add additional features for controlling
913        PCRE options from within the C++ wrapper.
915     5. Named capturing subpatterns were not being correctly counted when a pattern
916        was compiled. This caused two problems: (a) If there were more than 100
917        such subpatterns, the calculation of the memory needed for the whole
918        compiled pattern went wrong, leading to an overflow error. (b) Numerical
919        back references of the form \12, where the number was greater than 9, were
920        not recognized as back references, even though there were sufficient
921        previous subpatterns.
923     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
924        versions of gcc, e.g. 2.95.4.
927    Version 6.1 21-Jun-05
928    ---------------------
930     1. There was one reference to the variable "posix" in pcretest.c that was not
931        surrounded by "#if !defined NOPOSIX".
933     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
934        the cross-check on the old pcre_info() function, for the benefit of the
935        cut-down version of PCRE that is currently imported into Exim.
937     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
938        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
939        patterns that start (?i) but I don't think that matters. The use of (?i) is
940        just an example; this all applies to the other options as well.
942     4. Since libtool seems to echo the compile commands it is issuing, the output
943        from "make" can be reduced a bit by putting "@" in front of each libtool
944        compile command.
946     5. Patch from the folks at Google for configure.in to be a bit more thorough
947        in checking for a suitable C++ installation before trying to compile the
948        C++ stuff. This should fix a reported problem when a compiler was present,
949        but no suitable headers.
951     6. The man pages all had just "PCRE" as their title. I have changed them to
952        be the relevant file name. I have also arranged that these names are
953        retained in the file doc/pcre.txt, which is a concatenation in text format
954        of all the man pages except the little individual ones for each function.
956     7. The NON-UNIX-USE file had not been updated for the different set of source
957        files that come with release 6. I also added a few comments about the C++
958        wrapper.
961    Version 6.0 07-Jun-05
962    ---------------------
964     1. Some minor internal re-organization to help with my DFA experiments.
966     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
967        didn't matter for the library itself when fully configured, but did matter
968        when compiling without UCP support, or within Exim, where the ucp files are
969        not imported.
971     3. Refactoring of the library code to split up the various functions into
972        different source modules. The addition of the new DFA matching code (see
973        below) to a single monolithic source would have made it really too
974        unwieldy, quite apart from causing all the code to be include in a
975        statically linked application, when only some functions are used. This is
976        relevant even without the DFA addition now that patterns can be compiled in
977        one application and matched in another.
979        The downside of splitting up is that there have to be some external
980        functions and data tables that are used internally in different modules of
981        the library but which are not part of the API. These have all had their
982        names changed to start with "_pcre_" so that they are unlikely to clash
983        with other external names.
985     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
986        a different (DFA) algorithm. Although it is slower than the original
987        function, it does have some advantages for certain types of matching
988        problem.
990     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
991        including restarting after a partial match.
993     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
994        defined when compiling for Windows was sent to me. I have put it into the
995        code, though I have no means of testing or verifying it.
997     7. Added the pcre_refcount() auxiliary function.
999     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1000        match before or at the first newline in the subject string. In pcretest,
1001        the /f option on a pattern can be used to set this.
1003     9. A repeated \w when used in UTF-8 mode with characters greater than 256
1004        would behave wrongly. This has been present in PCRE since release 4.0.
1006    10. A number of changes to the pcregrep command:
1008        (a) Refactored how -x works; insert ^(...)$ instead of setting
1009            PCRE_ANCHORED and checking the length, in preparation for adding
1010            something similar for -w.
1012        (b) Added the -w (match as a word) option.
1014        (c) Refactored the way lines are read and buffered so as to have more
1015            than one at a time available.
1017        (d) Implemented a pcregrep test script.
1019        (e) Added the -M (multiline match) option. This allows patterns to match
1020            over several lines of the subject. The buffering ensures that at least
1021            8K, or the rest of the document (whichever is the shorter) is available
1022            for matching (and similarly the previous 8K for lookbehind assertions).
1024        (f) Changed the --help output so that it now says
1026              -w, --word-regex(p)
1028            instead of two lines, one with "regex" and the other with "regexp"
1029            because that confused at least one person since the short forms are the
1030            same. (This required a bit of code, as the output is generated
1031            automatically from a table. It wasn't just a text change.)
1033        (g) -- can be used to terminate pcregrep options if the next thing isn't an
1034            option but starts with a hyphen. Could be a pattern or a path name
1035            starting with a hyphen, for instance.
1037        (h) "-" can be given as a file name to represent stdin.
1039        (i) When file names are being printed, "(standard input)" is used for
1040            the standard input, for compatibility with GNU grep. Previously
1041            "<stdin>" was used.
1043        (j) The option --label=xxx can be used to supply a name to be used for
1044            stdin when file names are being printed. There is no short form.
1046        (k) Re-factored the options decoding logic because we are going to add
1047            two more options that take data. Such options can now be given in four
1048            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1050        (l) Added the -A, -B, and -C options for requesting that lines of context
1051            around matches be printed.
1053        (m) Added the -L option to print the names of files that do not contain
1054            any matching lines, that is, the complement of -l.
1056        (n) The return code is 2 if any file cannot be opened, but pcregrep does
1057            continue to scan other files.
1059        (o) The -s option was incorrectly implemented. For compatibility with other
1060            greps, it now suppresses the error message for a non-existent or non-
1061            accessible file (but not the return code). There is a new option called
1062            -q that suppresses the output of matching lines, which was what -s was
1063            previously doing.
1065        (p) Added --include and --exclude options to specify files for inclusion
1066            and exclusion when recursing.
1068    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1069        Hopefully, it now does.
1071    12. Missing cast in pcre_study().
1073    13. Added an "uninstall" target to the makefile.
1075    14. Replaced "extern" in the function prototypes in Makefile.in with
1076        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1077        world, but is set differently for Windows.
1079    15. Added a second compiling function called pcre_compile2(). The only
1080        difference is that it has an extra argument, which is a pointer to an
1081        integer error code. When there is a compile-time failure, this is set
1082        non-zero, in addition to the error test pointer being set to point to an
1083        error message. The new argument may be NULL if no error number is required
1084        (but then you may as well call pcre_compile(), which is now just a
1085        wrapper). This facility is provided because some applications need a
1086        numeric error indication, but it has also enabled me to tidy up the way
1087        compile-time errors are handled in the POSIX wrapper.
1089    16. Added VPATH=.libs to the makefile; this should help when building with one
1090        prefix path and installing with another. (Or so I'm told by someone who
1091        knows more about this stuff than I do.)
1093    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1094        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1095        match everything, including newlines. This is not POSIX-compatible, but
1096        somebody wanted the feature. From pcretest it can be activated by using
1097        both the P and the s flags.
1099    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1101    19. libpcre.pc was being incorrectly installed as executable.
1103    20. A couple of places in pcretest check for end-of-line by looking for '\n';
1104        it now also looks for '\r' so that it will work unmodified on Windows.
1106    21. Added Google's contributed C++ wrapper to the distribution.
1108    22. Added some untidy missing memory free() calls in pcretest, to keep
1109        Electric Fence happy when testing.
1113    Version 5.0 13-Sep-04
1114    ---------------------
1116     1. Internal change: literal characters are no longer packed up into items
1117        containing multiple characters in a single byte-string. Each character
1118        is now matched using a separate opcode. However, there may be more than one
1119        byte in the character in UTF-8 mode.
1121     2. The pcre_callout_block structure has two new fields: pattern_position and
1122        next_item_length. These contain the offset in the pattern to the next match
1123        item, and its length, respectively.
1125     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1126        insertion of callouts before each pattern item. Added the /C option to
1127        pcretest to make use of this.
1129     4. On the advice of a Windows user, the lines
1131          #if defined(_WIN32) || defined(WIN32)
1132          _setmode( _fileno( stdout ), 0x8000 );
1133          #endif  /* defined(_WIN32) || defined(WIN32) */
1135        have been added to the source of pcretest. This apparently does useful
1136        magic in relation to line terminators.
1138     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1139        for the benefit of those environments where the "b" makes a difference.
1141     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1142        to know about it. I have put a hack into configure.in that adds in code
1143        to set GCC=yes if CC=icc. This seems to end up at a point in the
1144        generated configure script that is early enough to affect the setting of
1145        compiler options, which is what is needed, but I have no means of testing
1146        whether it really works. (The user who reported this had patched the
1147        generated configure script, which of course I cannot do.)
1149        LATER: After change 22 below (new libtool files), the configure script
1150        seems to know about icc (and also ecc). Therefore, I have commented out
1151        this hack in configure.in.
1153     7. Added support for pkg-config (2 patches were sent in).
1155     8. Negated POSIX character classes that used a combination of internal tables
1156        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1157        [[:^ascii]]. Typically, they would match almost any characters. The other
1158        POSIX classes were not broken in this way.
1160     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1161        to find the match, as PCRE was deluded into thinking that the match had to
1162        start at the start point or following a newline. The same bug applied to
1163        patterns with negative forward assertions or any backward assertions
1164        preceding ".*" at the start, unless the pattern required a fixed first
1165        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1167    10. In UTF-8 mode, when moving forwards in the subject after a failed match
1168        starting at the last subject character, bytes beyond the end of the subject
1169        string were read.
1171    11. Renamed the variable "class" as "classbits" to make life easier for C++
1172        users. (Previously there was a macro definition, but it apparently wasn't
1173        enough.)
1175    12. Added the new field "tables" to the extra data so that tables can be passed
1176        in at exec time, or the internal tables can be re-selected. This allows
1177        a compiled regex to be saved and re-used at a later time by a different
1178        program that might have everything at different addresses.
1180    13. Modified the pcre-config script so that, when run on Solaris, it shows a
1181        -R library as well as a -L library.
1183    14. The debugging options of pcretest (-d on the command line or D on a
1184        pattern) showed incorrect output for anything following an extended class
1185        that contained multibyte characters and which was followed by a quantifier.
1187    15. Added optional support for general category Unicode character properties
1188        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1189        support. It adds about 90K to the size of the library. The meanings of the
1190        inbuilt class escapes such as \d and \s have NOT been changed.
1192    16. Updated pcredemo.c to include calls to free() to release the memory for the
1193        compiled pattern.
1195    17. The generated file chartables.c was being created in the source directory
1196        instead of in the building directory. This caused the build to fail if the
1197        source directory was different from the building directory, and was
1198        read-only.
1200    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1201        file. No doubt somebody will tell me if they don't make sense... Also added
1202        Dan Mooney's comments about building on OpenVMS.
1204    19. Added support for partial matching via the PCRE_PARTIAL option for
1205        pcre_exec() and the \P data escape in pcretest.
1207    20. Extended pcretest with 3 new pattern features:
1209        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
1210              write the compiled pattern to the file whose name is "rest-of-line".
1211              This is a straight binary dump of the data, with the saved pointer to
1212              the character tables forced to be NULL. The study data, if any, is
1213              written too. After writing, pcretest reads a new pattern.
1215        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1216              compiled pattern from the given file. There must not be any
1217              occurrences of "<" in the file name (pretty unlikely); if there are,
1218              pcretest will instead treat the initial "<" as a pattern delimiter.
1219              After reading in the pattern, pcretest goes on to read data lines as
1220              usual.
1222        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1223              and 16-bit fields in a compiled pattern, to simulate a pattern that
1224              was compiled on a host of opposite endianness.
1226    21. The pcre-exec() function can now cope with patterns that were compiled on
1227        hosts of opposite endianness, with this restriction:
1229          As for any compiled expression that is saved and used later, the tables
1230          pointer field cannot be preserved; the extra_data field in the arguments
1231          to pcre_exec() should be used to pass in a tables address if a value
1232          other than the default internal tables were used at compile time.
1234    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1235        now diagnosed as an error. Previously, most of the time, a negative number
1236        would have been treated as zero, but if in addition "ovector" was passed as
1237        NULL, a crash could occur.
1239    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1240        new versions from the libtool 1.5 distribution (the last one is a copy of
1241        a file called libtool.m4). This seems to have fixed the need to patch
1242        "configure" to support Darwin 1.3 (which I used to do). However, I still
1243        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1244        workstation).
1246    24. Changed the PCRE licence to be the more standard "BSD" licence.
1249    Version 4.5 01-Dec-03
1250    ---------------------
1252     1. There has been some re-arrangement of the code for the match() function so
1253        that it can be compiled in a version that does not call itself recursively.
1254        Instead, it keeps those local variables that need separate instances for
1255        each "recursion" in a frame on the heap, and gets/frees frames whenever it
1256        needs to "recurse". Keeping track of where control must go is done by means
1257        of setjmp/longjmp. The whole thing is implemented by a set of macros that
1258        hide most of the details from the main code, and operates only if
1259        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1260        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1261        operating.
1263        To make it easier for callers to provide specially tailored get/free
1264        functions for this usage, two new functions, pcre_stack_malloc, and
1265        pcre_stack_free, are used. They are always called in strict stacking order,
1266        and the size of block requested is always the same.
1268        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1269        PCRE has been compiled to use the stack or the heap for recursion. The
1270        -C option of pcretest uses this to show which version is compiled.
1272        A new data escape \S, is added to pcretest; it causes the amounts of store
1273        obtained and freed by both kinds of malloc/free at match time to be added
1274        to the output.
1276     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1277        what's available on my current Linux desktop machine.
1279     3. When matching a UTF-8 string, the test for a valid string at the start has
1280        been extended. If start_offset is not zero, PCRE now checks that it points
1281        to a byte that is the start of a UTF-8 character. If not, it returns
1282        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1283        this is necessary because there may be backward assertions in the pattern.
1284        When matching the same subject several times, it may save resources to use
1285        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1287     4. The code for checking the validity of UTF-8 strings has been tightened so
1288        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1289        containing "overlong sequences".
1291     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1292        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1293        should have been "&", but it just so happened that all the cases this let
1294        through by mistake were picked up later in the function.
1296     6. I had used a variable called "isblank" - this is a C99 function, causing
1297        some compilers to warn. To avoid this, I renamed it (as "blankclass").
1299     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1300        prompting; (b) run "./pcretest /dev/null" at the start of the test script
1301        so the version is shown; (c) stop "make test" echoing "./RunTest".
1303     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1305     9. The prototype for memmove() for systems that don't have it was using
1306        size_t, but the inclusion of the header that defines size_t was later. I've
1307        moved the #includes for the C headers earlier to avoid this.
1309    10. Added some adjustments to the code to make it easier to compiler on certain
1310        special systems:
1312          (a) Some "const" qualifiers were missing.
1313          (b) Added the macro EXPORT before all exported functions; by default this
1314              is defined to be empty.
1315          (c) Changed the dftables auxiliary program (that builds chartables.c) so
1316              that it reads its output file name as an argument instead of writing
1317              to the standard output and assuming this can be redirected.
1319    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
1320        class containing characters with values greater than 255, PCRE compilation
1321        went into a loop.
1323    12. A recursive reference to a subpattern that was within another subpattern
1324        that had a minimum quantifier of zero caused PCRE to crash. For example,
1325        (x(y(?2))z)? provoked this bug with a subject that got as far as the
1326        recursion. If the recursively-called subpattern itself had a zero repeat,
1327        that was OK.
1329    13. In pcretest, the buffer for reading a data line was set at 30K, but the
1330        buffer into which it was copied (for escape processing) was still set at
1331        1024, so long lines caused crashes.
1333    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
1334        "internal error: code overflow...". This applied to any character class
1335        that was followed by a possessive quantifier.
1337    15. Modified the Makefile to add libpcre.la as a prerequisite for
1338        libpcreposix.la because I was told this is needed for a parallel build to
1339        work.
1341    16. If a pattern that contained .* following optional items at the start was
1342        studied, the wrong optimizing data was generated, leading to matching
1343        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
1344        matching string must start with a or b or c. The correct conclusion for
1345        this pattern is that a match can start with any character.
1348    Version 4.4 13-Aug-03
1349    ---------------------
1351     1. In UTF-8 mode, a character class containing characters with values between
1352        127 and 255 was not handled correctly if the compiled pattern was studied.
1353        In fixing this, I have also improved the studying algorithm for such
1354        classes (slightly).
1356     2. Three internal functions had redundant arguments passed to them. Removal
1357        might give a very teeny performance improvement.
1359     3. Documentation bug: the value of the capture_top field in a callout is *one
1360        more than* the number of the hightest numbered captured substring.
1362     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
1363        in incorrectly linking with a previously installed version. They now link
1364        explicitly with libpcre.la.
1366     5. configure.in no longer needs to recognize Cygwin specially.
1368     6. A problem in pcre.in for Windows platforms is fixed.
1370     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
1371        pcretest, it used to include the size of the study block as part of its
1372        output. Unfortunately, the structure contains a field that has a different
1373        size on different hardware architectures. This meant that the tests that
1374        showed this size failed. As the block is currently always of a fixed size,
1375        this information isn't actually particularly useful in pcretest output, so
1376        I have just removed it.
1378     8. Three pre-processor statements accidentally did not start in column 1.
1379        Sadly, there are *still* compilers around that complain, even though
1380        standard C has not required this for well over a decade. Sigh.
1382     9. In pcretest, the code for checking callouts passed small integers in the
1383        callout_data field, which is a void * field. However, some picky compilers
1384        complained about the casts involved for this on 64-bit systems. Now
1385        pcretest passes the address of the small integer instead, which should get
1386        rid of the warnings.
1388    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
1389        both compile and run time, and gives an error if an invalid UTF-8 sequence
1390        is found. There is a option for disabling this check in cases where the
1391        string is known to be correct and/or the maximum performance is wanted.
1393    11. In response to a bug report, I changed one line in Makefile.in from
1395            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
1396        to
1397            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
1399        to look similar to other lines, but I have no way of telling whether this
1400        is the right thing to do, as I do not use Windows. No doubt I'll get told
1401        if it's wrong...
1404    Version 4.3 21-May-03
1405    ---------------------
1407    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
1408       Makefile.
1410    2. Some refactoring to improve the quality of the code:
1412       (i)   The utf8_table... variables are now declared "const".
1414       (ii)  The code for \cx, which used the "case flipping" table to upper case
1415             lower case letters, now just substracts 32. This is ASCII-specific,
1416             but the whole concept of \cx is ASCII-specific, so it seems
1417             reasonable.
1419       (iii) PCRE was using its character types table to recognize decimal and
1420             hexadecimal digits in the pattern. This is silly, because it handles
1421             only 0-9, a-f, and A-F, but the character types table is locale-
1422             specific, which means strange things might happen. A private
1423             table is now used for this - though it costs 256 bytes, a table is
1424             much faster than multiple explicit tests. Of course, the standard
1425             character types table is still used for matching digits in subject
1426             strings against \d.
1428       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
1429             ending in _t are). So I've renamed it as ESC_tee.
1431    3. The first argument for regexec() in the POSIX wrapper should have been
1432       defined as "const".
1434    4. Changed pcretest to use malloc() for its buffers so that they can be
1435       Electric Fenced for debugging.
1437    5. There were several places in the code where, in UTF-8 mode, PCRE would try
1438       to read one or more bytes before the start of the subject string. Often this
1439       had no effect on PCRE's behaviour, but in some circumstances it could
1440       provoke a segmentation fault.
1442    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
1443       to try to read one or more bytes before the start of the subject string.
1445    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
1446       UTF-8 support could misbehave in various ways if the subject string
1447       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
1448       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
1449       back over UTF-8 characters.)
1452    Version 4.2 14-Apr-03
1453    ---------------------
1455    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
1457    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
1458         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
1459         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
1460         [WIN_PREFIX]: new variable, "cyg" for Cygwin
1461         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
1462           and BUILD_EXEEXT
1463         Note: automatic setting of the BUILD variables is not yet working
1464         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
1465           compile-time but not at link-time
1466         [LINK]: use for linking executables only
1467         make different versions for Windows and non-Windows
1468         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
1469           libraries
1470         [LINK_FOR_BUILD]: new variable
1471         [OBJEXT]: use throughout
1472         [EXEEXT]: use throughout
1473         <winshared>: new target
1474         <wininstall>: new target
1475         <dftables.o>: use native compiler
1476         <dftables>: use native linker
1477         <install>: handle Windows platform correctly
1478         <clean>: ditto
1479         <check>: ditto
1480         copy DLL to top builddir before testing
1482       As part of these changes, -no-undefined was removed again. This was reported
1483       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
1484       in any case.
1486    3. Some tidies to get rid of compiler warnings:
1488       . In the match_data structure, match_limit was an unsigned long int, whereas
1489         match_call_count was an int. I've made them both unsigned long ints.
1491       . In pcretest the fact that a const uschar * doesn't automatically cast to
1492         a void * provoked a warning.
1494       . Turning on some more compiler warnings threw up some "shadow" variables
1495         and a few more missing casts.
1497    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
1498       option, a class that contained a single character with a value between 128
1499       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
1501    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
1502       option, a class that contained several characters, but with at least one
1503       whose value was between 128 and 255 caused PCRE to crash.
1506    Version 4.1 12-Mar-03
1507    ---------------------
1509    1. Compiling with gcc -pedantic found a couple of places where casts were
1510    needed, and a string in dftables.c that was longer than standard compilers are
1511    required to support.
1513    2. Compiling with Sun's compiler found a few more places where the code could
1514    be tidied up in order to avoid warnings.
1516    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
1517    first of these names is deprecated in the latest Autoconf in favour of the name
1518    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
1519    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
1520    analogy I have changed it to CFLAGS_FOR_BUILD.
1522    4. Added -no-undefined to the linking command in the Makefile, because this is
1523    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
1524    linking step for the pcreposix library.
1526    5. PCRE was failing to diagnose the case of two named groups with the same
1527    name.
1529    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
1530    literal character that is needed in the subject for a match, and scans along to
1531    ensure that it is present before embarking on the full matching process. This
1532    saves time in cases of nested unlimited repeats that are never going to match.
1533    Problem: the scan can take a lot of time if the subject is very long (e.g.
1534    megabytes), thus penalizing straightforward matches. It is now done only if the
1535    amount of subject to be scanned is less than 1000 bytes.
1537    7. A lesser problem with the same optimization is that it was recording the
1538    first character of an anchored pattern as "needed", thus provoking a search
1539    right along the subject, even when the first match of the pattern was going to
1540    fail. The "needed" character is now not set for anchored patterns, unless it
1541    follows something in the pattern that is of non-fixed length. Thus, it still
1542    fulfils its original purpose of finding quick non-matches in cases of nested
1543    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
1546    Version 4.0 17-Feb-03
1547    ---------------------
1549    1. If a comment in an extended regex that started immediately after a meta-item
1550    extended to the end of string, PCRE compiled incorrect data. This could lead to
1551    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
1553    2. Moved to autoconf 2.53 and libtool 1.4.2.
1555    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
1556    the special perltest8 script is no longer needed - all the tests can be run
1557    from a single perltest script.
1559    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
1560    by \s. It has now been removed in PCRE. This means it isn't recognized as
1561    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
1562    class [:space:] *does* include VT, thereby creating a mess.
1564    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
1565    space and tab.
1567    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
1568    its new features into the main test script, reducing the number of scripts.
1570    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
1571    were backward compatible, and made the (?i) apply to the whole pattern, as if
1572    /i were given. Now it behaves more logically, and applies the option setting
1573    only to what follows. PCRE has been changed to follow suit. However, if it
1574    finds options settings right at the start of the pattern, it extracts them into
1575    the global options, as before. Thus, they show up in the info data.
1577    8. Added support for the \Q...\E escape sequence. Characters in between are
1578    treated as literals. This is slightly different from Perl in that $ and @ are
1579    also handled as literals inside the quotes. In Perl, they will cause variable
1580    interpolation. Note the following examples:
1582        Pattern            PCRE matches      Perl matches
1584        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
1585        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
1586        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
1588    For compatibility with Perl, \Q...\E sequences are recognized inside character
1589    classes as well as outside them.
1591    9. Re-organized 3 code statements in pcretest to avoid "overflow in
1592    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
1593    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
1594    signed/unsigned warnings.
1596    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
1597    option for pcretest, so I've replaced it by a simple function that does just
1598    that job.
1600    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
1601    "pcregrep -".
1603    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
1604    Java package. This provides some syntactic sugar for simple cases of what my
1605    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
1606    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
1607    item, you can use this simplified notation. Note that only makes sense with
1608    greedy quantifiers. Consequently, the use of the possessive quantifier forces
1609    greediness, whatever the setting of the PCRE_UNGREEDY option.
1611    13. A change of greediness default within a pattern was not taking effect at
1612    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
1613    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
1614    was abstracted outside.
1616    14. PCRE now supports the \G assertion. It is true when the current matching
1617    position is at the start point of the match. This differs from \A when the
1618    starting offset is non-zero. Used with the /g option of pcretest (or similar
1619    code), it works in the same way as it does for Perl's /g option. If all
1620    alternatives of a regex begin with \G, the expression is anchored to the start
1621    match position, and the "anchored" flag is set in the compiled expression.
1623    15. Some bugs concerning the handling of certain option changes within patterns
1624    have been fixed. These applied to options other than (?ims). For example,
1625    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
1626    been the other way round. Some of this was related to change 7 above.
1628    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
1629    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
1630    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
1631    POSIX classes only within a class (e.g. /[[:alpha:]]/).
1633    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
1634    mode. Unlike ".", it always matches newline, whatever the setting of
1635    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
1636    assertions. Perl allows it, but it doesn't (in general) work because it can't
1637    calculate the length of the lookbehind. At least, that's the case for Perl
1638    5.8.0 - I've been told they are going to document that it doesn't work in
1639    future.
1641    18. Added an error diagnosis for escapes that PCRE does not support: these are
1642    \L, \l, \N, \P, \p, \U, \u, and \X.
1644    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
1645    reading past the end of the pattern in cases such as /[abcd/.
1647    20. PCRE was getting more memory than necessary for patterns with classes that
1648    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
1650    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
1651    compiling PCRE for use with Virtual Pascal.
1653    22. Small fix to the Makefile to make it work properly if the build is done
1654    outside the source tree.
1656    23. Added a new extension: a condition to go with recursion. If a conditional
1657    subpattern starts with (?(R) the "true" branch is used if recursion has
1658    happened, whereas the "false" branch is used only at the top level.
1660    24. When there was a very long string of literal characters (over 255 bytes
1661    without UTF support, over 250 bytes with UTF support), the computation of how
1662    much memory was required could be incorrect, leading to segfaults or other
1663    strange effects.
1665    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
1666    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
1667    there was a subsequent back reference to those brackets. This meant that, for
1668    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
1669    possible to check for precisely this case. All we can do is abandon the
1670    optimization if .* occurs inside capturing brackets when there are any back
1671    references whatsoever. (See below for a better fix that came later.)
1673    26. The handling of the optimization for finding the first character of a
1674    non-anchored pattern, and for finding a character that is required later in the
1675    match were failing in some cases. This didn't break the matching; it just
1676    failed to optimize when it could. The way this is done has been re-implemented.
1678    27. Fixed typo in error message for invalid (?R item (it said "(?p").
1680    28. Added a new feature that provides some of the functionality that Perl
1681    provides with (?{...}). The facility is termed a "callout". The way it is done
1682    in PCRE is for the caller to provide an optional function, by setting
1683    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
1684    global variable. By default it is unset, which disables all calling out. To get
1685    the function called, the regex must include (?C) at appropriate points. This
1686    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
1687    This provides a means of identifying different callout points. When PCRE
1688    reaches such a point in the regex, if pcre_callout has been set, the external
1689    function is called. It is provided with data in a structure called
1690    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
1691    matching continues; if it returns a non-zero value, the match at the current
1692    point fails. However, backtracking will occur if possible. [This was changed
1693    later and other features added - see item 49 below.]
1695    29. pcretest is upgraded to test the callout functionality. It provides a
1696    callout function that displays information. By default, it shows the start of
1697    the match and the current position in the text. There are some new data escapes
1698    to vary what happens:
1700        \C+         in addition, show current contents of captured substrings
1701        \C-         do not supply a callout function
1702        \C!n        return 1 when callout number n is reached
1703        \C!n!m      return 1 when callout number n is reached for the mth time
1705    30. If pcregrep was called with the -l option and just a single file name, it
1706    output "<stdin>" if a match was found, instead of the file name.
1708    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
1709    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
1710    pcre_exec(). This saves a malloc/free per call. The default value of
1711    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
1712    when configuring.
1714    32. The default maximum size of a compiled pattern is 64K. There have been a
1715    few cases of people hitting this limit. The code now uses macros to handle the
1716    storing of links as offsets within the compiled pattern. It defaults to 2-byte
1717    links, but this can be changed to 3 or 4 bytes by --with-link-size when
1718    configuring. Tests 2 and 5 work only with 2-byte links because they output
1719    debugging information about compiled patterns.
1721    33. Internal code re-arrangements:
1723    (a) Moved the debugging function for printing out a compiled regex into
1724        its own source file (printint.c) and used #include to pull it into
1725        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
1726        separate copies.
1728    (b) Defined the list of op-code names for debugging as a macro in
1729        internal.h so that it is next to the definition of the opcodes.
1731    (c) Defined a table of op-code lengths for simpler skipping along compiled
1732        code. This is again a macro in internal.h so that it is next to the
1733        definition of the opcodes.
1735    34. Added support for recursive calls to individual subpatterns, along the
1736    lines of Robin Houston's patch (but implemented somewhat differently).
1738    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
1739    allow it to read and process whole directories in Win32. This code was
1740    contributed by Lionel Fourquaux; it has not been tested by me.
1742    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
1743    used to name a group. Names consist of alphanumerics and underscores, and must
1744    be unique. Back references use the syntax (?P=name) and recursive calls use
1745    (?P>name) which is a PCRE extension to the Python extension. Groups still have
1746    numbers. The function pcre_fullinfo() can be used after compilation to extract
1747    a name/number map. There are three relevant calls:
1749      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
1750      PCRE_INFO_NAMECOUNT            yields the number of entries
1751      PCRE_INFO_NAMETABLE            yields a pointer to the map.
1753    The map is a vector of fixed-size entries. The size of each entry depends on
1754    the length of the longest name used. The first two bytes of each entry are the
1755    group number, most significant byte first. There follows the corresponding
1756    name, zero terminated. The names are in alphabetical order.
1758    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
1759    case instead of 255. Making it the same both with and without UTF-8 support
1760    means that the same test output works with both.
1762    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
1763    calling malloc() with a zero argument.
1765    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
1766    optimization. I've improved things by keeping a bitmap of backreferences with
1767    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
1768    fact referenced, the optimization can be applied. It is unlikely that a
1769    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
1770    the match to follow \n) will appear inside brackets with a number greater than
1771    31, but if it does, any back reference > 31 suppresses the optimization.
1773    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
1774    of disabling numbered capturing parentheses. Any opening parenthesis that is
1775    not followed by ? behaves as if it were followed by ?: but named parentheses
1776    can still be used for capturing (and they will acquire numbers in the usual
1777    way).
1779    41. Redesigned the return codes from the match() function into yes/no/error so
1780    that errors can be passed back from deep inside the nested calls. A malloc
1781    failure while inside a recursive subpattern call now causes the
1782    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
1784    42. It is now possible to set a limit on the number of times the match()
1785    function is called in a call to pcre_exec(). This facility makes it possible to
1786    limit the amount of recursion and backtracking, though not in a directly
1787    obvious way, because the match() function is used in a number of different
1788    circumstances. The count starts from zero for each position in the subject
1789    string (for non-anchored patterns). The default limit is, for compatibility, a
1790    large number, namely 10 000 000. You can change this in two ways:
1792    (a) When configuring PCRE before making, you can use --with-match-limit=n
1793        to set a default value for the compiled library.
1795    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
1796        a different value is set. See 45 below.
1798    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
1800    43. Added a new function pcre_config(int, void *) to enable run-time extraction
1801    of things that can be changed at compile time. The first argument specifies
1802    what is wanted and the second points to where the information is to be placed.
1803    The current list of available information is:
1805      PCRE_CONFIG_UTF8
1807    The output is an integer that is set to one if UTF-8 support is available;
1808    otherwise it is set to zero.
1812    The output is an integer that it set to the value of the code that is used for
1813    newline. It is either LF (10) or CR (13).
1817    The output is an integer that contains the number of bytes used for internal
1818    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
1822    The output is an integer that contains the threshold above which the POSIX
1823    interface uses malloc() for output vectors. See item 31 above.
1827    The output is an unsigned integer that contains the default limit of the number
1828    of match() calls in a pcre_exec() execution. See 42 above.
1830    44. pcretest has been upgraded by the addition of the -C option. This causes it
1831    to extract all the available output from the new pcre_config() function, and to
1832    output it. The program then exits immediately.
1834    45. A need has arisen to pass over additional data with calls to pcre_exec() in
1835    order to support additional features. One way would have been to define
1836    pcre_exec2() (for example) with extra arguments, but this would not have been
1837    extensible, and would also have required all calls to the original function to
1838    be mapped to the new one. Instead, I have chosen to extend the mechanism that
1839    is used for passing in "extra" data from pcre_study().
1841    The pcre_extra structure is now exposed and defined in pcre.h. It currently
1842    contains the following fields:
1844      flags         a bitmap indicating which of the following fields are set
1845      study_data    opaque data from pcre_study()
1846      match_limit   a way of specifying a limit on match() calls for a specific
1847                      call to pcre_exec()
1848      callout_data  data for callouts (see 49 below)
1850    The flag bits are also defined in pcre.h, and are
1856    The pcre_study() function now returns one of these new pcre_extra blocks, with
1857    the actual study data pointed to by the study_data field, and the
1858    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
1859    before. That is, this change is entirely upwards-compatible and requires no
1860    change to existing code.
1862    If you want to pass in additional data to pcre_exec(), you can either place it
1863    in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
1864    block.
1866    46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
1867    data string contains the escape sequence \M, pcretest calls pcre_exec() several
1868    times with different match limits, until it finds the minimum value needed for
1869    pcre_exec() to complete. The value is then output. This can be instructive; for
1870    most simple matches the number is quite small, but for pathological cases it
1871    gets very large very quickly.
1873    47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
1874    returns the size of the data block pointed to by the study_data field in a
1875    pcre_extra block, that is, the value that was passed as the argument to
1876    pcre_malloc() when PCRE was getting memory in which to place the information
1877    created by pcre_study(). The fourth argument should point to a size_t variable.
1878    pcretest has been extended so that this information is shown after a successful
1879    pcre_study() call when information about the compiled regex is being displayed.
1881    48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
1882    because what follows is always an absolute path. (Later: it turns out that this
1883    is more than cosmetic for MinGW, because it doesn't like empty path
1884    components.)
1886    49. Some changes have been made to the callout feature (see 28 above):
1888    (i)  A callout function now has three choices for what it returns:
1890           0  =>  success, carry on matching
1891         > 0  =>  failure at this point, but backtrack if possible
1892         < 0  =>  serious error, return this value from pcre_exec()
1894         Negative values should normally be chosen from the set of PCRE_ERROR_xxx
1895         values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
1896         "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
1897         use by callout functions. It will never be used by PCRE itself.
1899    (ii) The pcre_extra structure (see 45 above) has a void * field called
1900         callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
1901         pcre_callout_block structure has a field of the same name. The contents of
1902         the field passed in the pcre_extra structure are passed to the callout
1903         function in the corresponding field in the callout block. This makes it
1904         easier to use the same callout-containing regex from multiple threads. For
1905         testing, the pcretest program has a new data escape
1907           \C*n        pass the number n (may be negative) as callout_data
1909         If the callout function in pcretest receives a non-zero value as
1910         callout_data, it returns that value.
1912    50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
1913    there were some redundant $(CFLAGS) in commands that are now specified as
1914    $(LINK), which already includes $(CFLAGS).
1916    51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
1917    has been compiled with UTF-8 support *and* pcre_compile() has been compiled
1918    with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
1919    one-byte characters throughout. Note that case-insensitive matching applies
1920    only to characters whose values are less than 256. PCRE doesn't support the
1921    notion of cases for higher-valued characters.
1923    (i)   A character class whose characters are all within 0-255 is handled as
1924          a bit map, and the map is inverted for negative classes. Previously, a
1925          character > 255 always failed to match such a class; however it should
1926          match if the class was a negative one (e.g. [^ab]). This has been fixed.
1928    (ii)  A negated character class with a single character < 255 is coded as
1929          "not this character" (OP_NOT). This wasn't working properly when the test
1930          character was multibyte, either singly or repeated.
1932    (iii) Repeats of multibyte characters are now handled correctly in UTF-8
1933          mode, for example: \x{100}{2,3}.
1935    (iv)  The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
1936          singly or repeated) now correctly test multibyte characters. However,
1937          PCRE doesn't recognize any characters with values greater than 255 as
1938          digits, spaces, or word characters. Such characters always match \D, \S,
1939          and \W, and never match \d, \s, or \w.
1941    (v)   Classes may now contain characters and character ranges with values
1942          greater than 255. For example: [ab\x{100}-\x{400}].
1944    (vi)  pcregrep now has a --utf-8 option (synonym -u) which makes it call
1945          PCRE in UTF-8 mode.
1947    52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
1948    PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
1949    retained for backwards compatibility. (Note that LASTLITERAL is also a byte
1950    value.)
1952    53. The single man page has become too large. I have therefore split it up into
1953    a number of separate man pages. These also give rise to individual HTML pages;
1954    these are now put in a separate directory, and there is an index.html page that
1955    lists them all. Some hyperlinking between the pages has been installed.
1957    54. Added convenience functions for handling named capturing parentheses.
1959    55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
1960    aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
1961    true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
1962    are faulted.
1964    56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
1965    calling configure. These values are used when compiling the dftables.c program
1966    which is run to generate the source of the default character tables. They
1967    default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
1968    you will need to set these values.
1970    57. Updated the building process for Windows DLL, as provided by Fred Cox.
1973    Version 3.9 02-Jan-02
1974    ---------------------
1976    1. A bit of extraneous text had somehow crept into the pcregrep documentation.
1978    2. If --disable-static was given, the building process failed when trying to
1979    build pcretest and pcregrep. (For some reason it was using libtool to compile
1980    them, which is not right, as they aren't part of the library.)
1983    Version 3.8 18-Dec-01
1984    ---------------------
1986    1. The experimental UTF-8 code was completely screwed up. It was packing the
1987    bytes in the wrong order. How dumb can you get?
1990    Version 3.7 29-Oct-01
1991    ---------------------
1993    1. In updating pcretest to check change 1 of version 3.6, I screwed up.
1994    This caused pcretest, when used on the test data, to segfault. Unfortunately,
1995    this didn't happen under Solaris 8, where I normally test things.
1997    2. The Makefile had to be changed to make it work on BSD systems, where 'make'
1998    doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
1999    isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2000    this fix an hour or so after the initial 3.7 release.)
2003    Version 3.6 23-Oct-01
2004    ---------------------
2006    1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2007    offsets passed as NULL with zero offset count.
2009    2. The config.guess and config.sub files had not been updated when I moved to
2010    the latest autoconf.
2013    Version 3.5 15-Aug-01
2014    ---------------------
2016    1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2017    had been forgotten.
2019    2. By using declared but undefined structures, we can avoid using "void"
2020    definitions in pcre.h while keeping the internal definitions of the structures
2021    private.
2023    3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2024    user point of view, this means that both static and shared libraries are built
2025    by default, but this can be individually controlled. More of the work of
2026    handling this static/shared cases is now inside libtool instead of PCRE's make
2027    file.
2029    4. The pcretest utility is now installed along with pcregrep because it is
2030    useful for users (to test regexs) and by doing this, it automatically gets
2031    relinked by libtool. The documentation has been turned into a man page, so
2032    there are now .1, .txt, and .html versions in /doc.
2034    5. Upgrades to pcregrep:
2035       (i)   Added long-form option names like gnu grep.
2036       (ii)  Added --help to list all options with an explanatory phrase.
2037       (iii) Added -r, --recursive to recurse into sub-directories.
2038       (iv)  Added -f, --file to read patterns from a file.
2040    6. pcre_exec() was referring to its "code" argument before testing that
2041    argument for NULL (and giving an error if it was NULL).
2043    7. Upgraded Makefile.in to allow for compiling in a different directory from
2044    the source directory.
2046    8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2047    options bits, the pointer it was passed was to an int instead of to an unsigned
2048    long int. This mattered only on 64-bit systems.
2050    9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2051    generated) instead of pcre.in, which it its source. Also made the same change
2052    in several of the .c files.
2054    10. A new release of gcc defines printf() as a macro, which broke pcretest
2055    because it had an ifdef in the middle of a string argument for printf(). Fixed
2056    by using separate calls to printf().
2058    11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2059    script, to force use of CR or LF instead of \n in the source. On non-Unix
2060    systems, the value can be set in config.h.
2062    12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2063    absolute limit. Changed the text of the error message to make this clear, and
2064    likewise updated the man page.
2066    13. The limit of 99 on the number of capturing subpatterns has been removed.
2067    The new limit is 65535, which I hope will not be a "real" limit.
2070    Version 3.4 22-Aug-00
2071    ---------------------
2073    1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2075    2. Diagnose condition (?(0) as an error instead of crashing on matching.
2078  Version 3.3 01-Aug-00  Version 3.3 01-Aug-00
2079  ---------------------  ---------------------

Removed from v.49  
changed lines
  Added in v.161

  ViewVC Help
Powered by ViewVC 1.1.5