/[pcre]/code/trunk/ChangeLog
ViewVC logotype

Diff of /code/trunk/ChangeLog

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

revision 21 by nigel, Sat Feb 24 21:38:37 2007 UTC revision 143 by ph10, Mon Apr 2 10:08:14 2007 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4    Version 7.1 12-Mar-07
5    ---------------------
6    
7  Version 1.09 28 Apr-98   1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
8        that is more "standard", making use of automake and other Autotools. There
9        is some re-arrangement of the files and adjustment of comments consequent
10        on this.
11    
12     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
13        for recursive directory scanning broke on some systems because the files
14        are not scanned in any specific order and on different systems the order
15        was different. A call to "sort" has been inserted into RunGrepTest for the
16        approprate test as a short-term fix. In the longer term there may be an
17        alternative.
18    
19     3. I had an email from Eric Raymond about problems translating some of PCRE's
20        man pages to HTML (despite the fact that I distribute HTML pages, some
21        people do their own conversions for various reasons). The problems
22        concerned the use of low-level troff macros .br and .in. I have therefore
23        removed all such uses from the man pages (some were redundant, some could
24        be replaced by .nf/.fi pairs). The 132html script that I use to generate
25        HTML has been updated to handle .nf/.fi and to complain if it encounters
26        .br or .in.
27    
28     4. Updated comments in configure.ac that get placed in config.h.in and also
29        arranged for config.h to be included in the distribution, with the name
30        config.h.generic, for the benefit of those who have to compile without
31        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
32    
33     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
34        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
35        makevp.bat for the current PCRE, using the additional files
36        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
37    
38     6. A Windows user reported a minor discrepancy with test 2, which turned out
39        to be caused by a trailing space on an input line that had got lost in his
40        copy. The trailing space was an accident, so I've just removed it.
41    
42     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
43        that is needed.
44    
45     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
46        as "const" (a) because they are and (b) because it helps the PHP
47        maintainers who have recently made a script to detect big data structures
48        in the php code that should be moved to the .rodata section. I remembered
49        to update Builducptable as well, so it won't revert if ucptable.h is ever
50        re-created.
51    
52     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
53        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
54        order to be able to cut out the UTF-8 tables in the latter when UTF-8
55        support is not required. This saves 1.5-2K of code, which is important in
56        some applications.
57    
58        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
59        so as not to refer to the tables, even though these functions will never be
60        called when UTF-8 support is disabled. Otherwise there are problems with a
61        shared library.
62    
63    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
64    
65        (a) It was defining its arguments as char * instead of void *.
66    
67        (b) It was assuming that all moves were upwards in memory; this was true
68            a long time ago when I wrote it, but is no longer the case.
69    
70        The emulated memove() is provided for those environments that have neither
71        memmove() nor bcopy(). I didn't think anyone used it these days, but that
72        is clearly not the case, as these two bugs were recently reported.
73    
74    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
75        and Detrail to create the HTML documentation, the .txt form of the man
76        pages, and it removes trailing spaces from listed files. It also creates
77        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
78        case, it wraps all the #defines with #ifndefs. This script should be run
79        before "make dist".
80    
81    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
82        with Unicode property support.
83    
84        (a) For a maximizing quantifier, if the two different cases of the
85            character were of different lengths in their UTF-8 codings (there are
86            some cases like this - I found 11), and the matching function had to
87            back up over a mixture of the two cases, it incorrectly assumed they
88            were both the same length.
89    
90        (b) When PCRE was configured to use the heap rather than the stack for
91            recursion during matching, it was not correctly preserving the data for
92            the other case of a UTF-8 character when checking ahead for a match
93            while processing a minimizing repeat. If the check also involved
94            matching a wide character, but failed, corruption could cause an
95            erroneous result when trying to check for a repeat of the original
96            character.
97    
98    13. Some tidying changes to the testing mechanism:
99    
100        (a) The RunTest script now detects the internal link size and whether there
101            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
102            values substituted by "configure". (The RunGrepTest script already did
103            this for UTF-8.) The configure.ac script no longer substitutes the
104            relevant variables.
105    
106        (b) The debugging options /B and /D in pcretest show the compiled bytecode
107            with length and offset values. This means that the output is different
108            for different internal link sizes. Test 2 is skipped for link sizes
109            other than 2 because of this, bypassing the problem. Unfortunately,
110            there was also a test in test 3 (the locale tests) that used /B and
111            failed for link sizes other than 2. Rather than cut the whole test out,
112            I have added a new /Z option to pcretest that replaces the length and
113            offset values with spaces. This is now used to make test 3 independent
114            of link size. (Test 2 will be tidied up later.)
115    
116    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
117        segmentation fault instead of returning the appropriate error message.
118    
119    15. In multiline mode when the newline sequence was set to "any", the pattern
120        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
121        This doesn't seem right; it now treats the CRLF combination as the line
122        ending, and so does not match in that case. It's only a pattern such as ^$
123        that would hit this one: something like ^ABC$ would have failed after \r
124        and then tried again after \r\n.
125    
126    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
127        in an attempt to make files that differ only in their line terminators
128        compare equal. This works on Linux.
129    
130    17. Under certain error circumstances pcregrep might try to free random memory
131        as it exited. This is now fixed, thanks to valgrind.
132    
133    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
134        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
135        was because its rules for how to advance for /g after matching an empty
136        string at the end of a line did not allow for this case. They now check for
137        it specially.
138    
139    
140    Version 7.0 19-Dec-06
141    ---------------------
142    
143     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
144        moving to gcc 4.1.1.
145    
146     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
147        sys/time.h, which is documented as needed for this function. It doesn't
148        seem to matter on Linux, but it showed up on some releases of OS X.
149    
150     3. It seems that there are systems where bytes whose values are greater than
151        127 match isprint() in the "C" locale. The "C" locale should be the
152        default when a C program starts up. In most systems, only ASCII printing
153        characters match isprint(). This difference caused the output from pcretest
154        to vary, making some of the tests fail. I have changed pcretest so that:
155    
156        (a) When it is outputting text in the compiled version of a pattern, bytes
157            other than 32-126 are always shown as hex escapes.
158    
159        (b) When it is outputting text that is a matched part of a subject string,
160            it does the same, unless a different locale has been set for the match
161            (using the /L modifier). In this case, it uses isprint() to decide.
162    
163     4. Fixed a major bug that caused incorrect computation of the amount of memory
164        required for a compiled pattern when options that changed within the
165        pattern affected the logic of the preliminary scan that determines the
166        length. The relevant options are -x, and -i in UTF-8 mode. The result was
167        that the computed length was too small. The symptoms of this bug were
168        either the PCRE error "internal error: code overflow" from pcre_compile(),
169        or a glibc crash with a message such as "pcretest: free(): invalid next
170        size (fast)". Examples of patterns that provoked this bug (shown in
171        pcretest format) are:
172    
173          /(?-x: )/x
174          /(?x)(?-x: \s*#\s*)/
175          /((?i)[\x{c0}])/8
176          /(?i:[\x{c0}])/8
177    
178        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
179        is now done differently.
180    
181     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
182        wrapper classes; (b) implement a new function in the C++ scanner that is
183        more efficient than the old way of doing things because it avoids levels of
184        recursion in the regex matching; (c) add a paragraph to the documentation
185        for the FullMatch() function.
186    
187     6. The escape sequence \n was being treated as whatever was defined as
188        "newline". Not only was this contrary to the documentation, which states
189        that \n is character 10 (hex 0A), but it also went horribly wrong when
190        "newline" was defined as CRLF. This has been fixed.
191    
192     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
193        was being set to -1 for the "end of line" case (supposedly a value that no
194        character can have). Though this value is never used (the check for end of
195        line is "zero bytes in current character"), it caused compiler complaints.
196        I've changed it to 0xffffffff.
197    
198     8. In pcre_version.c, the version string was being built by a sequence of
199        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
200        string (as it is for production releases) called a macro with an empty
201        argument. The C standard says the result of this is undefined. The gcc
202        compiler treats it as an empty string (which was what was wanted) but it is
203        reported that Visual C gives an error. The source has been hacked around to
204        avoid this problem.
205    
206     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
207        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
208        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
209        of them did).
210    
211    10. Originally, pcretest opened its input and output without "b"; then I was
212        told that "b" was needed in some environments, so it was added for release
213        5.0 to both the input and output. (It makes no difference on Unix-like
214        systems.) Later I was told that it is wrong for the input on Windows. I've
215        now abstracted the modes into two macros, to make it easier to fiddle with
216        them, and removed "b" from the input mode under Windows.
217    
218    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
219    
220    12. Added -help and --help to pcretest as an official way of being reminded
221        of the options.
222    
223    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
224        and pcrecpp.cc because they annoy compilers at high warning levels.
225    
226    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
227    
228    15. Fixed an occurrence of == in configure.ac that should have been = (shell
229        scripts are not C programs :-) and which was not noticed because it works
230        on Linux.
231    
232    16. pcretest is supposed to handle any length of pattern and data line (as one
233        line or as a continued sequence of lines) by extending its input buffer if
234        necessary. This feature was broken for very long pattern lines, leading to
235        a string of junk being passed to pcre_compile() if the pattern was longer
236        than about 50K.
237    
238    17. I have done a major re-factoring of the way pcre_compile() computes the
239        amount of memory needed for a compiled pattern. Previously, there was code
240        that made a preliminary scan of the pattern in order to do this. That was
241        OK when PCRE was new, but as the facilities have expanded, it has become
242        harder and harder to keep it in step with the real compile phase, and there
243        have been a number of bugs (see for example, 4 above). I have now found a
244        cunning way of running the real compile function in a "fake" mode that
245        enables it to compute how much memory it would need, while actually only
246        ever using a few hundred bytes of working memory and without too many
247        tests of the mode. This should make future maintenance and development
248        easier. A side effect of this work is that the limit of 200 on the nesting
249        depth of parentheses has been removed (though this was never a serious
250        limitation, I suspect). However, there is a downside: pcre_compile() now
251        runs more slowly than before (30% or more, depending on the pattern). I
252        hope this isn't a big issue. There is no effect on runtime performance.
253    
254    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
255        newline (only possible for the last line of a file) and it was a
256        pattern that set a locale (followed by /Lsomething), pcretest crashed.
257    
258    19. Added additional timing features to pcretest. (1) The -tm option now times
259        matching only, not compiling. (2) Both -t and -tm can be followed, as a
260        separate command line item, by a number that specifies the number of
261        repeats to use when timing. The default is 50000; this gives better
262        precision, but takes uncomfortably long for very large patterns.
263    
264    20. Extended pcre_study() to be more clever in cases where a branch of a
265        subpattern has no definite first character. For example, (a*|b*)[cd] would
266        previously give no result from pcre_study(). Now it recognizes that the
267        first character must be a, b, c, or d.
268    
269    21. There was an incorrect error "recursive call could loop indefinitely" if
270        a subpattern (or the entire pattern) that was being tested for matching an
271        empty string contained only one non-empty item after a nested subpattern.
272        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
273        incorrectly, because the \d was being skipped in the check.
274    
275    22. The pcretest program now has a new pattern option /B and a command line
276        option -b, which is equivalent to adding /B to every pattern. This causes
277        it to show the compiled bytecode, without the additional information that
278        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
279        is the same as /B/I).
280    
281    23. A new optimization is now able automatically to treat some sequences such
282        as a*b as a*+b. More specifically, if something simple (such as a character
283        or a simple class like \d) has an unlimited quantifier, and is followed by
284        something that cannot possibly match the quantified thing, the quantifier
285        is automatically "possessified".
286    
287    24. A recursive reference to a subpattern whose number was greater than 39
288        went wrong under certain circumstances in UTF-8 mode. This bug could also
289        have affected the operation of pcre_study().
290    
291    25. Realized that a little bit of performance could be had by replacing
292        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
293    
294    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
295    
296    27. Possessive quantifiers such as a++ were previously implemented by turning
297        them into atomic groups such as ($>a+). Now they have their own opcodes,
298        which improves performance. This includes the automatically created ones
299        from 23 above.
300    
301    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
302        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
303        the first matched character to be a colon. This applied both to named and
304        numbered groups.
305    
306    29. The ucpinternal.h header file was missing its idempotency #ifdef.
307    
308    30. I was sent a "project" file called libpcre.a.dev which I understand makes
309        building PCRE on Windows easier, so I have included it in the distribution.
310    
311    31. There is now a check in pcretest against a ridiculously large number being
312        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
313        loop, the loop is abandoned.
314    
315    32. Forward references to subpatterns in conditions such as (?(2)...) where
316        subpattern 2 is defined later cause pcre_compile() to search forwards in
317        the pattern for the relevant set of parentheses. This search went wrong
318        when there were unescaped parentheses in a character class, parentheses
319        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
320    
321    33. "Subroutine" calls and backreferences were previously restricted to
322        referencing subpatterns earlier in the regex. This restriction has now
323        been removed.
324    
325    34. Added a number of extra features that are going to be in Perl 5.10. On the
326        whole, these are just syntactic alternatives for features that PCRE had
327        previously implemented using the Python syntax or my own invention. The
328        other formats are all retained for compatibility.
329    
330        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
331            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
332            also .NET compatible.
333    
334        (b) A recursion or subroutine call to a named group can now be defined as
335            (?&name) as well as (?P>name).
336    
337        (c) A backreference to a named group can now be defined as \k<name> or
338            \k'name' as well as (?P=name). The new forms, as well as being in Perl
339            5.10, are also .NET compatible.
340    
341        (d) A conditional reference to a named group can now use the syntax
342            (?(<name>) or (?('name') as well as (?(name).
343    
344        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
345            groups (named and numbered) that are never evaluated inline, but can be
346            called as "subroutines" from elsewhere. In effect, the DEFINE condition
347            is always false. There may be only one alternative in such a group.
348    
349        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
350            as the simple (?(R). The condition is true only if the most recent
351            recursion is that of the given number or name. It does not search out
352            through the entire recursion stack.
353    
354        (g) The escape \gN or \g{N} has been added, where N is a positive or
355            negative number, specifying an absolute or relative reference.
356    
357    35. Tidied to get rid of some further signed/unsigned compiler warnings and
358        some "unreachable code" warnings.
359    
360    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
361        things, this adds five new scripts.
362    
363    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
364        There were also incompatibilities regarding the handling of \Q..\E inside
365        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
366        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
367    
368    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
369        matches an empty string, and forcibly breaks the loop. There were bugs in
370        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
371        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
372        separate and independent bugs (that affected different cases) have been
373        fixed.
374    
375    39. Refactored the code to abolish the use of different opcodes for small
376        capturing bracket numbers. This is a tidy that I avoided doing when I
377        removed the limit on the number of capturing brackets for 3.5 back in 2001.
378        The new approach is not only tidier, it makes it possible to reduce the
379        memory needed to fix the previous bug (38).
380    
381    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
382        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
383        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
384        mode.
385    
386    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
387        report.
388    
389    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
390        copy construction and assignment in the C++ wrapper.
391    
392    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
393        couple of bugs that could have given wrong results in the "--newline=crlf"
394        case.
395    
396    44. Added a number of casts and did some reorganization of signed/unsigned int
397        variables following suggestions from Dair Grant. Also renamed the variable
398        "this" as "item" because it is a C++ keyword.
399    
400    45. Arranged for dftables to add
401    
402          #include "pcre_internal.h"
403    
404        to pcre_chartables.c because without it, gcc 4.x may remove the array
405        definition from the final binary if PCRE is built into a static library and
406        dead code stripping is activated.
407    
408    46. For an unanchored pattern, if a match attempt fails at the start of a
409        newline sequence, and the newline setting is CRLF or ANY, and the next two
410        characters are CRLF, advance by two characters instead of one.
411    
412    
413    Version 6.7 04-Jul-06
414    ---------------------
415    
416     1. In order to handle tests when input lines are enormously long, pcretest has
417        been re-factored so that it automatically extends its buffers when
418        necessary. The code is crude, but this _is_ just a test program. The
419        default size has been increased from 32K to 50K.
420    
421     2. The code in pcre_study() was using the value of the re argument before
422        testing it for NULL. (Of course, in any sensible call of the function, it
423        won't be NULL.)
424    
425     3. The memmove() emulation function in pcre_internal.h, which is used on
426        systems that lack both memmove() and bcopy() - that is, hardly ever -
427        was missing a "static" storage class specifier.
428    
429     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
430        containing an extended class (one that cannot be represented by a bitmap
431        because it contains high-valued characters or Unicode property items, e.g.
432        [\pZ]). Almost always one would set UTF-8 mode when processing such a
433        pattern, but PCRE should not loop if you do not (it no longer does).
434        [Detail: two cases were found: (a) a repeated subpattern containing an
435        extended class; (b) a recursive reference to a subpattern that followed a
436        previous extended class. It wasn't skipping over the extended class
437        correctly when UTF-8 mode was not set.]
438    
439     5. A negated single-character class was not being recognized as fixed-length
440        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
441        compile error "lookbehind assertion is not fixed length".
442    
443     6. The RunPerlTest auxiliary script was showing an unexpected difference
444        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
445        write a Perl script that can interpret lines of an input file either as
446        byte characters or as UTF-8, which is what "perltest" was being required to
447        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
448        can't do is switch easily at run time between having the "use utf8;" pragma
449        or not. In the end, I fudged it by using the RunPerlTest script to insert
450        "use utf8;" explicitly for the UTF-8 tests.
451    
452     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
453        the end of the subject string, contrary to the documentation and to what
454        Perl does. This was true of both matching functions. Now it matches only at
455        the start of the subject and immediately after *internal* newlines.
456    
457     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
458        a pointer to an int instead of a pointer to an unsigned long int. This
459        caused problems on 64-bit systems.
460    
461     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
462        instance of the 'standard' template library not being so standard".
463    
464    10. There was no check on the number of named subpatterns nor the maximum
465        length of a subpattern name. The product of these values is used to compute
466        the size of the memory block for a compiled pattern. By supplying a very
467        long subpattern name and a large number of named subpatterns, the size
468        computation could be caused to overflow. This is now prevented by limiting
469        the length of names to 32 characters, and the number of named subpatterns
470        to 10,000.
471    
472    11. Subpatterns that are repeated with specific counts have to be replicated in
473        the compiled pattern. The size of memory for this was computed from the
474        length of the subpattern and the repeat count. The latter is limited to
475        65535, but there was no limit on the former, meaning that integer overflow
476        could in principle occur. The compiled length of a repeated subpattern is
477        now limited to 30,000 bytes in order to prevent this.
478    
479    12. Added the optional facility to have named substrings with the same name.
480    
481    13. Added the ability to use a named substring as a condition, using the
482        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
483        are numbers (not recommended). Forward references are permitted.
484    
485    14. Added forward references in named backreferences (if you see what I mean).
486    
487    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
488        pattern could run off the end of the subject. For example, the pattern
489        "(?s)(.{1,5})"8 did this with the subject "ab".
490    
491    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
492        PCRE_CASELESS was set when matching characters that were quantified with ?
493        or *.
494    
495    17. A character class other than a single negated character that had a minimum
496        but no maximum quantifier - for example [ab]{6,} - was not handled
497        correctly by pce_dfa_exec(). It would match only one character.
498    
499    18. A valid (though odd) pattern that looked like a POSIX character
500        class but used an invalid character after [ (for example [[,abc,]]) caused
501        pcre_compile() to give the error "Failed: internal error: code overflow" or
502        in some cases to crash with a glibc free() error. This could even happen if
503        the pattern terminated after [[ but there just happened to be a sequence of
504        letters, a binary zero, and a closing ] in the memory that followed.
505    
506    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
507        over the years. Originally (before any Unicode support), just the bottom 8
508        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
509        output from "man perlunicode" includes this:
510    
511          The regular expression compiler produces polymorphic opcodes.  That
512          is, the pattern adapts to the data and automatically switches to
513          the Unicode character scheme when presented with Unicode data--or
514          instead uses a traditional byte scheme when presented with byte
515          data.
516    
517        Sadly, a wide octal escape does not cause a switch, and in a string with
518        no other multibyte characters, these octal escapes are treated as before.
519        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
520        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
521        Unicode string.
522    
523        I have not perpetrated such confusion in PCRE. Up till now, it took just
524        the bottom 8 bits, as in old Perl. I have now made octal escapes with
525        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
526        translate to the appropriate multibyte character.
527    
528    29. Applied some refactoring to reduce the number of warnings from Microsoft
529        and Borland compilers. This has included removing the fudge introduced
530        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
531        a warning about an unused variable.
532    
533    21. PCRE has not included VT (character 0x0b) in the set of whitespace
534        characters since release 4.0, because Perl (from release 5.004) does not.
535        [Or at least, is documented not to: some releases seem to be in conflict
536        with the documentation.] However, when a pattern was studied with
537        pcre_study() and all its branches started with \s, PCRE still included VT
538        as a possible starting character. Of course, this did no harm; it just
539        caused an unnecessary match attempt.
540    
541    22. Removed a now-redundant internal flag bit that recorded the fact that case
542        dependency changed within the pattern. This was once needed for "required
543        byte" processing, but is no longer used. This recovers a now-scarce options
544        bit. Also moved the least significant internal flag bit to the most-
545        significant bit of the word, which was not previously used (hangover from
546        the days when it was an int rather than a uint) to free up another bit for
547        the future.
548    
549    23. Added support for CRLF line endings as well as CR and LF. As well as the
550        default being selectable at build time, it can now be changed at runtime
551        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
552        specify that it is scanning data with non-default line endings.
553    
554    24. Changed the definition of CXXLINK to make it agree with the definition of
555        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
556    
557    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
558        recursions. This makes a big different to stack usage for some patterns.
559    
560    26. If a subpattern containing a named recursion or subroutine reference such
561        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
562        the space required for the compiled pattern went wrong and gave too small a
563        value. Depending on the environment, this could lead to "Failed: internal
564        error: code overflow at offset 49" or "glibc detected double free or
565        corruption" errors.
566    
567    27. Applied patches from Google (a) to support the new newline modes and (b) to
568        advance over multibyte UTF-8 characters in GlobalReplace.
569    
570    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
571        difference for some implementation of PCRE in some Windows version.
572    
573    29. Added some extra testing facilities to pcretest:
574    
575        \q<number>   in a data line sets the "match limit" value
576        \Q<number>   in a data line sets the "match recursion limt" value
577        -S <number>  sets the stack size, where <number> is in megabytes
578    
579        The -S option isn't available for Windows.
580    
581    
582    Version 6.6 06-Feb-06
583    ---------------------
584    
585     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
586        in pcreposix.h. I have copied the definition from pcre.h.
587    
588     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
589        because pcre.h is no longer a built file.
590    
591     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
592        not normally included in the compiled code.
593    
594    
595    Version 6.5 01-Feb-06
596    ---------------------
597    
598     1. When using the partial match feature with pcre_dfa_exec(), it was not
599        anchoring the second and subsequent partial matches at the new starting
600        point. This could lead to incorrect results. For example, with the pattern
601        /1234/, partially matching against "123" and then "a4" gave a match.
602    
603     2. Changes to pcregrep:
604    
605        (a) All non-match returns from pcre_exec() were being treated as failures
606            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
607            error message is output. Some extra information is given for the
608            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
609            probably the only errors that are likely to be caused by users (by
610            specifying a regex that has nested indefinite repeats, for instance).
611            If there are more than 20 of these errors, pcregrep is abandoned.
612    
613        (b) A binary zero was treated as data while matching, but terminated the
614            output line if it was written out. This has been fixed: binary zeroes
615            are now no different to any other data bytes.
616    
617        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
618            used to set a locale for matching. The --locale=xxxx long option has
619            been added (no short equivalent) to specify a locale explicitly on the
620            pcregrep command, overriding the environment variables.
621    
622        (d) When -B was used with -n, some line numbers in the output were one less
623            than they should have been.
624    
625        (e) Added the -o (--only-matching) option.
626    
627        (f) If -A or -C was used with -c (count only), some lines of context were
628            accidentally printed for the final match.
629    
630        (g) Added the -H (--with-filename) option.
631    
632        (h) The combination of options -rh failed to suppress file names for files
633            that were found from directory arguments.
634    
635        (i) Added the -D (--devices) and -d (--directories) options.
636    
637        (j) Added the -F (--fixed-strings) option.
638    
639        (k) Allow "-" to be used as a file name for -f as well as for a data file.
640    
641        (l) Added the --colo(u)r option.
642    
643        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
644            is not present by default.
645    
646     3. A nasty bug was discovered in the handling of recursive patterns, that is,
647        items such as (?R) or (?1), when the recursion could match a number of
648        alternatives. If it matched one of the alternatives, but subsequently,
649        outside the recursion, there was a failure, the code tried to back up into
650        the recursion. However, because of the way PCRE is implemented, this is not
651        possible, and the result was an incorrect result from the match.
652    
653        In order to prevent this happening, the specification of recursion has
654        been changed so that all such subpatterns are automatically treated as
655        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
656    
657     4. I had overlooked the fact that, in some locales, there are characters for
658        which isalpha() is true but neither isupper() nor islower() are true. In
659        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
660        and ordfeminine) are like this. This affected the treatment of \w and \W
661        when they appeared in character classes, but not when they appeared outside
662        a character class. The bit map for "word" characters is now created
663        separately from the results of isalnum() instead of just taking it from the
664        upper, lower, and digit maps. (Plus the underscore character, of course.)
665    
666     5. The above bug also affected the handling of POSIX character classes such as
667        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
668        permanent tables. Instead, the bit maps for such a class were previously
669        created as the appropriate unions of the upper, lower, and digit bitmaps.
670        Now they are created by subtraction from the [[:word:]] class, which has
671        its own bitmap.
672    
673     6. The [[:blank:]] character class matches horizontal, but not vertical space.
674        It is created by subtracting the vertical space characters (\x09, \x0a,
675        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
676        subtraction was done in the overall bitmap for a character class, meaning
677        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
678        be recognized. This bug has been fixed.
679    
680     7. Patches from the folks at Google:
681    
682          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
683          real life, but is still worth protecting against".
684    
685          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
686          regular expressions".
687    
688          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
689          have it.
690    
691          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
692          "configure" and the latter not, in order to fix a problem somebody had
693          with compiling the Arg class on HP-UX.
694    
695          (e) Improve the error-handling of the C++ wrapper a little bit.
696    
697          (f) New tests for checking recursion limiting.
698    
699     8. The pcre_memmove() function, which is used only if the environment does not
700        have a standard memmove() function (and is therefore rarely compiled),
701        contained two bugs: (a) use of int instead of size_t, and (b) it was not
702        returning a result (though PCRE never actually uses the result).
703    
704     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
705        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
706        returned instead of calling malloc() with an overflowing number that would
707        most likely cause subsequent chaos.
708    
709    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
710    
711    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
712        with this option is matched, the nmatch and pmatch options of regexec() are
713        ignored.
714    
715    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
716        provided in case anyone wants to the the POSIX interface with UTF-8
717        strings.
718    
719    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
720        C++ linking (needed for some HP-UX environments).
721    
722    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
723        (unused parameter) and in the pcre_printint() function (omitted "default"
724        switch label when the default is to do nothing).
725    
726    15. Added some code to make it possible, when PCRE is compiled as a C++
727        library, to replace subject pointers for pcre_exec() with a smart pointer
728        class, thus making it possible to process discontinuous strings.
729    
730    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
731        much the same function. They were added by different people who were trying
732        to make PCRE easy to compile on non-Unix systems. It has been suggested
733        that PCRE_EXPORT be abolished now that there is more automatic apparatus
734        for compiling on Windows systems. I have therefore replaced it with
735        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
736        defaults to "extern" for C or "extern C" for C++, which works fine on
737        Unix-like systems. It is now possible to override the value of PCRE_DATA_
738        SCOPE with something explicit in config.h. In addition:
739    
740        (a) pcreposix.h still had just "extern" instead of either of these macros;
741            I have replaced it with PCRE_DATA_SCOPE.
742    
743        (b) Functions such as _pcre_xclass(), which are internal to the library,
744            but external in the C sense, all had PCRE_EXPORT in their definitions.
745            This is apparently wrong for the Windows case, so I have removed it.
746            (It makes no difference on Unix-like systems.)
747    
748    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
749        of recursive calls to match(). This is different to MATCH_LIMIT because
750        that limits the total number of calls to match(), not all of which increase
751        the depth of recursion. Limiting the recursion depth limits the amount of
752        stack (or heap if NO_RECURSE is set) that is used. The default can be set
753        when PCRE is compiled, and changed at run time. A patch from Google adds
754        this functionality to the C++ interface.
755    
756    18. Changes to the handling of Unicode character properties:
757    
758        (a) Updated the table to Unicode 4.1.0.
759    
760        (b) Recognize characters that are not in the table as "Cn" (undefined).
761    
762        (c) I revised the way the table is implemented to a much improved format
763            which includes recognition of ranges. It now supports the ranges that
764            are defined in UnicodeData.txt, and it also amalgamates other
765            characters into ranges. This has reduced the number of entries in the
766            table from around 16,000 to around 3,000, thus reducing its size
767            considerably. I realized I did not need to use a tree structure after
768            all - a binary chop search is just as efficient. Having reduced the
769            number of entries, I extended their size from 6 bytes to 8 bytes to
770            allow for more data.
771    
772        (d) Added support for Unicode script names via properties such as \p{Han}.
773    
774    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
775        matching that character.
776    
777    20. When matching a repeated Unicode property with a minimum greater than zero,
778        (for example \pL{2,}), PCRE could look past the end of the subject if it
779        reached it while seeking the minimum number of characters. This could
780        happen only if some of the characters were more than one byte long, because
781        there is a check for at least the minimum number of bytes.
782    
783    21. Refactored the implementation of \p and \P so as to be more general, to
784        allow for more different types of property in future. This has changed the
785        compiled form incompatibly. Anybody with saved compiled patterns that use
786        \p or \P will have to recompile them.
787    
788    22. Added "Any" and "L&" to the supported property types.
789    
790    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
791        but give a compile time error if the value is greater than 0xff.
792    
793    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
794        accidentally not being installed or uninstalled.
795    
796    25. The pcre.h file was built from pcre.h.in, but the only changes that were
797        made were to insert the current release number. This seemed silly, because
798        it made things harder for people building PCRE on systems that don't run
799        "configure". I have turned pcre.h into a distributed file, no longer built
800        by "configure", with the version identification directly included. There is
801        no longer a pcre.h.in file.
802    
803        However, this change necessitated a change to the pcre-config script as
804        well. It is built from pcre-config.in, and one of the substitutions was the
805        release number. I have updated configure.ac so that ./configure now finds
806        the release number by grepping pcre.h.
807    
808    26. Added the ability to run the tests under valgrind.
809    
810    
811    Version 6.4 05-Sep-05
812    ---------------------
813    
814     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
815        "--" to be printed when multiple files were scanned, even when none of the
816        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
817        consider it to be a bug, and have restored the previous behaviour.
818    
819     2. A couple of code tidies to get rid of compiler warnings.
820    
821     3. The pcretest program used to cheat by referring to symbols in the library
822        whose names begin with _pcre_. These are internal symbols that are not
823        really supposed to be visible externally, and in some environments it is
824        possible to suppress them. The cheating is now confined to including
825        certain files from the library's source, which is a bit cleaner.
826    
827     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
828        file's purpose clearer.
829    
830     5. Reorganized pcre_ucp_findchar().
831    
832    
833    Version 6.3 15-Aug-05
834    ---------------------
835    
836     1. The file libpcre.pc.in did not have general read permission in the tarball.
837    
838     2. There were some problems when building without C++ support:
839    
840        (a) If C++ support was not built, "make install" and "make test" still
841            tried to test it.
842    
843        (b) There were problems when the value of CXX was explicitly set. Some
844            changes have been made to try to fix these, and ...
845    
846        (c) --disable-cpp can now be used to explicitly disable C++ support.
847    
848        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
849            backslash in a target when C++ was disabled. This confuses some
850            versions of "make", apparently. Using an intermediate variable solves
851            this. (Same for CPP_LOBJ.)
852    
853     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
854        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
855        necessary on certain architectures.
856    
857     4. Added a setting of -export-symbols-regex to the link command to remove
858        those symbols that are exported in the C sense, but actually are local
859        within the library, and not documented. Their names all begin with
860        "_pcre_". This is not a perfect job, because (a) we have to except some
861        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
862        available (and never for static libraries). I have made a note to try to
863        find a way round (a) in the future.
864    
865    
866    Version 6.2 01-Aug-05
867    ---------------------
868    
869     1. There was no test for integer overflow of quantifier values. A construction
870        such as {1111111111111111} would give undefined results. What is worse, if
871        a minimum quantifier for a parenthesized subpattern overflowed and became
872        negative, the calculation of the memory size went wrong. This could have
873        led to memory overwriting.
874    
875     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
876    
877     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
878        operating environments where this matters.
879    
880     4. Applied Giuseppe Maxia's patch to add additional features for controlling
881        PCRE options from within the C++ wrapper.
882    
883     5. Named capturing subpatterns were not being correctly counted when a pattern
884        was compiled. This caused two problems: (a) If there were more than 100
885        such subpatterns, the calculation of the memory needed for the whole
886        compiled pattern went wrong, leading to an overflow error. (b) Numerical
887        back references of the form \12, where the number was greater than 9, were
888        not recognized as back references, even though there were sufficient
889        previous subpatterns.
890    
891     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
892        versions of gcc, e.g. 2.95.4.
893    
894    
895    Version 6.1 21-Jun-05
896    ---------------------
897    
898     1. There was one reference to the variable "posix" in pcretest.c that was not
899        surrounded by "#if !defined NOPOSIX".
900    
901     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
902        the cross-check on the old pcre_info() function, for the benefit of the
903        cut-down version of PCRE that is currently imported into Exim.
904    
905     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
906        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
907        patterns that start (?i) but I don't think that matters. The use of (?i) is
908        just an example; this all applies to the other options as well.
909    
910     4. Since libtool seems to echo the compile commands it is issuing, the output
911        from "make" can be reduced a bit by putting "@" in front of each libtool
912        compile command.
913    
914     5. Patch from the folks at Google for configure.in to be a bit more thorough
915        in checking for a suitable C++ installation before trying to compile the
916        C++ stuff. This should fix a reported problem when a compiler was present,
917        but no suitable headers.
918    
919     6. The man pages all had just "PCRE" as their title. I have changed them to
920        be the relevant file name. I have also arranged that these names are
921        retained in the file doc/pcre.txt, which is a concatenation in text format
922        of all the man pages except the little individual ones for each function.
923    
924     7. The NON-UNIX-USE file had not been updated for the different set of source
925        files that come with release 6. I also added a few comments about the C++
926        wrapper.
927    
928    
929    Version 6.0 07-Jun-05
930    ---------------------
931    
932     1. Some minor internal re-organization to help with my DFA experiments.
933    
934     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
935        didn't matter for the library itself when fully configured, but did matter
936        when compiling without UCP support, or within Exim, where the ucp files are
937        not imported.
938    
939     3. Refactoring of the library code to split up the various functions into
940        different source modules. The addition of the new DFA matching code (see
941        below) to a single monolithic source would have made it really too
942        unwieldy, quite apart from causing all the code to be include in a
943        statically linked application, when only some functions are used. This is
944        relevant even without the DFA addition now that patterns can be compiled in
945        one application and matched in another.
946    
947        The downside of splitting up is that there have to be some external
948        functions and data tables that are used internally in different modules of
949        the library but which are not part of the API. These have all had their
950        names changed to start with "_pcre_" so that they are unlikely to clash
951        with other external names.
952    
953     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
954        a different (DFA) algorithm. Although it is slower than the original
955        function, it does have some advantages for certain types of matching
956        problem.
957    
958     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
959        including restarting after a partial match.
960    
961     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
962        defined when compiling for Windows was sent to me. I have put it into the
963        code, though I have no means of testing or verifying it.
964    
965     7. Added the pcre_refcount() auxiliary function.
966    
967     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
968        match before or at the first newline in the subject string. In pcretest,
969        the /f option on a pattern can be used to set this.
970    
971     9. A repeated \w when used in UTF-8 mode with characters greater than 256
972        would behave wrongly. This has been present in PCRE since release 4.0.
973    
974    10. A number of changes to the pcregrep command:
975    
976        (a) Refactored how -x works; insert ^(...)$ instead of setting
977            PCRE_ANCHORED and checking the length, in preparation for adding
978            something similar for -w.
979    
980        (b) Added the -w (match as a word) option.
981    
982        (c) Refactored the way lines are read and buffered so as to have more
983            than one at a time available.
984    
985        (d) Implemented a pcregrep test script.
986    
987        (e) Added the -M (multiline match) option. This allows patterns to match
988            over several lines of the subject. The buffering ensures that at least
989            8K, or the rest of the document (whichever is the shorter) is available
990            for matching (and similarly the previous 8K for lookbehind assertions).
991    
992        (f) Changed the --help output so that it now says
993    
994              -w, --word-regex(p)
995    
996            instead of two lines, one with "regex" and the other with "regexp"
997            because that confused at least one person since the short forms are the
998            same. (This required a bit of code, as the output is generated
999            automatically from a table. It wasn't just a text change.)
1000    
1001        (g) -- can be used to terminate pcregrep options if the next thing isn't an
1002            option but starts with a hyphen. Could be a pattern or a path name
1003            starting with a hyphen, for instance.
1004    
1005        (h) "-" can be given as a file name to represent stdin.
1006    
1007        (i) When file names are being printed, "(standard input)" is used for
1008            the standard input, for compatibility with GNU grep. Previously
1009            "<stdin>" was used.
1010    
1011        (j) The option --label=xxx can be used to supply a name to be used for
1012            stdin when file names are being printed. There is no short form.
1013    
1014        (k) Re-factored the options decoding logic because we are going to add
1015            two more options that take data. Such options can now be given in four
1016            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1017    
1018        (l) Added the -A, -B, and -C options for requesting that lines of context
1019            around matches be printed.
1020    
1021        (m) Added the -L option to print the names of files that do not contain
1022            any matching lines, that is, the complement of -l.
1023    
1024        (n) The return code is 2 if any file cannot be opened, but pcregrep does
1025            continue to scan other files.
1026    
1027        (o) The -s option was incorrectly implemented. For compatibility with other
1028            greps, it now suppresses the error message for a non-existent or non-
1029            accessible file (but not the return code). There is a new option called
1030            -q that suppresses the output of matching lines, which was what -s was
1031            previously doing.
1032    
1033        (p) Added --include and --exclude options to specify files for inclusion
1034            and exclusion when recursing.
1035    
1036    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1037        Hopefully, it now does.
1038    
1039    12. Missing cast in pcre_study().
1040    
1041    13. Added an "uninstall" target to the makefile.
1042    
1043    14. Replaced "extern" in the function prototypes in Makefile.in with
1044        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1045        world, but is set differently for Windows.
1046    
1047    15. Added a second compiling function called pcre_compile2(). The only
1048        difference is that it has an extra argument, which is a pointer to an
1049        integer error code. When there is a compile-time failure, this is set
1050        non-zero, in addition to the error test pointer being set to point to an
1051        error message. The new argument may be NULL if no error number is required
1052        (but then you may as well call pcre_compile(), which is now just a
1053        wrapper). This facility is provided because some applications need a
1054        numeric error indication, but it has also enabled me to tidy up the way
1055        compile-time errors are handled in the POSIX wrapper.
1056    
1057    16. Added VPATH=.libs to the makefile; this should help when building with one
1058        prefix path and installing with another. (Or so I'm told by someone who
1059        knows more about this stuff than I do.)
1060    
1061    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1062        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1063        match everything, including newlines. This is not POSIX-compatible, but
1064        somebody wanted the feature. From pcretest it can be activated by using
1065        both the P and the s flags.
1066    
1067    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1068    
1069    19. libpcre.pc was being incorrectly installed as executable.
1070    
1071    20. A couple of places in pcretest check for end-of-line by looking for '\n';
1072        it now also looks for '\r' so that it will work unmodified on Windows.
1073    
1074    21. Added Google's contributed C++ wrapper to the distribution.
1075    
1076    22. Added some untidy missing memory free() calls in pcretest, to keep
1077        Electric Fence happy when testing.
1078    
1079    
1080    
1081    Version 5.0 13-Sep-04
1082    ---------------------
1083    
1084     1. Internal change: literal characters are no longer packed up into items
1085        containing multiple characters in a single byte-string. Each character
1086        is now matched using a separate opcode. However, there may be more than one
1087        byte in the character in UTF-8 mode.
1088    
1089     2. The pcre_callout_block structure has two new fields: pattern_position and
1090        next_item_length. These contain the offset in the pattern to the next match
1091        item, and its length, respectively.
1092    
1093     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1094        insertion of callouts before each pattern item. Added the /C option to
1095        pcretest to make use of this.
1096    
1097     4. On the advice of a Windows user, the lines
1098    
1099          #if defined(_WIN32) || defined(WIN32)
1100          _setmode( _fileno( stdout ), 0x8000 );
1101          #endif  /* defined(_WIN32) || defined(WIN32) */
1102    
1103        have been added to the source of pcretest. This apparently does useful
1104        magic in relation to line terminators.
1105    
1106     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1107        for the benefit of those environments where the "b" makes a difference.
1108    
1109     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1110        to know about it. I have put a hack into configure.in that adds in code
1111        to set GCC=yes if CC=icc. This seems to end up at a point in the
1112        generated configure script that is early enough to affect the setting of
1113        compiler options, which is what is needed, but I have no means of testing
1114        whether it really works. (The user who reported this had patched the
1115        generated configure script, which of course I cannot do.)
1116    
1117        LATER: After change 22 below (new libtool files), the configure script
1118        seems to know about icc (and also ecc). Therefore, I have commented out
1119        this hack in configure.in.
1120    
1121     7. Added support for pkg-config (2 patches were sent in).
1122    
1123     8. Negated POSIX character classes that used a combination of internal tables
1124        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1125        [[:^ascii]]. Typically, they would match almost any characters. The other
1126        POSIX classes were not broken in this way.
1127    
1128     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1129        to find the match, as PCRE was deluded into thinking that the match had to
1130        start at the start point or following a newline. The same bug applied to
1131        patterns with negative forward assertions or any backward assertions
1132        preceding ".*" at the start, unless the pattern required a fixed first
1133        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1134    
1135    10. In UTF-8 mode, when moving forwards in the subject after a failed match
1136        starting at the last subject character, bytes beyond the end of the subject
1137        string were read.
1138    
1139    11. Renamed the variable "class" as "classbits" to make life easier for C++
1140        users. (Previously there was a macro definition, but it apparently wasn't
1141        enough.)
1142    
1143    12. Added the new field "tables" to the extra data so that tables can be passed
1144        in at exec time, or the internal tables can be re-selected. This allows
1145        a compiled regex to be saved and re-used at a later time by a different
1146        program that might have everything at different addresses.
1147    
1148    13. Modified the pcre-config script so that, when run on Solaris, it shows a
1149        -R library as well as a -L library.
1150    
1151    14. The debugging options of pcretest (-d on the command line or D on a
1152        pattern) showed incorrect output for anything following an extended class
1153        that contained multibyte characters and which was followed by a quantifier.
1154    
1155    15. Added optional support for general category Unicode character properties
1156        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1157        support. It adds about 90K to the size of the library. The meanings of the
1158        inbuilt class escapes such as \d and \s have NOT been changed.
1159    
1160    16. Updated pcredemo.c to include calls to free() to release the memory for the
1161        compiled pattern.
1162    
1163    17. The generated file chartables.c was being created in the source directory
1164        instead of in the building directory. This caused the build to fail if the
1165        source directory was different from the building directory, and was
1166        read-only.
1167    
1168    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1169        file. No doubt somebody will tell me if they don't make sense... Also added
1170        Dan Mooney's comments about building on OpenVMS.
1171    
1172    19. Added support for partial matching via the PCRE_PARTIAL option for
1173        pcre_exec() and the \P data escape in pcretest.
1174    
1175    20. Extended pcretest with 3 new pattern features:
1176    
1177        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
1178              write the compiled pattern to the file whose name is "rest-of-line".
1179              This is a straight binary dump of the data, with the saved pointer to
1180              the character tables forced to be NULL. The study data, if any, is
1181              written too. After writing, pcretest reads a new pattern.
1182    
1183        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1184              compiled pattern from the given file. There must not be any
1185              occurrences of "<" in the file name (pretty unlikely); if there are,
1186              pcretest will instead treat the initial "<" as a pattern delimiter.
1187              After reading in the pattern, pcretest goes on to read data lines as
1188              usual.
1189    
1190        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1191              and 16-bit fields in a compiled pattern, to simulate a pattern that
1192              was compiled on a host of opposite endianness.
1193    
1194    21. The pcre-exec() function can now cope with patterns that were compiled on
1195        hosts of opposite endianness, with this restriction:
1196    
1197          As for any compiled expression that is saved and used later, the tables
1198          pointer field cannot be preserved; the extra_data field in the arguments
1199          to pcre_exec() should be used to pass in a tables address if a value
1200          other than the default internal tables were used at compile time.
1201    
1202    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1203        now diagnosed as an error. Previously, most of the time, a negative number
1204        would have been treated as zero, but if in addition "ovector" was passed as
1205        NULL, a crash could occur.
1206    
1207    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1208        new versions from the libtool 1.5 distribution (the last one is a copy of
1209        a file called libtool.m4). This seems to have fixed the need to patch
1210        "configure" to support Darwin 1.3 (which I used to do). However, I still
1211        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1212        workstation).
1213    
1214    24. Changed the PCRE licence to be the more standard "BSD" licence.
1215    
1216    
1217    Version 4.5 01-Dec-03
1218    ---------------------
1219    
1220     1. There has been some re-arrangement of the code for the match() function so
1221        that it can be compiled in a version that does not call itself recursively.
1222        Instead, it keeps those local variables that need separate instances for
1223        each "recursion" in a frame on the heap, and gets/frees frames whenever it
1224        needs to "recurse". Keeping track of where control must go is done by means
1225        of setjmp/longjmp. The whole thing is implemented by a set of macros that
1226        hide most of the details from the main code, and operates only if
1227        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1228        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1229        operating.
1230    
1231        To make it easier for callers to provide specially tailored get/free
1232        functions for this usage, two new functions, pcre_stack_malloc, and
1233        pcre_stack_free, are used. They are always called in strict stacking order,
1234        and the size of block requested is always the same.
1235    
1236        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1237        PCRE has been compiled to use the stack or the heap for recursion. The
1238        -C option of pcretest uses this to show which version is compiled.
1239    
1240        A new data escape \S, is added to pcretest; it causes the amounts of store
1241        obtained and freed by both kinds of malloc/free at match time to be added
1242        to the output.
1243    
1244     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1245        what's available on my current Linux desktop machine.
1246    
1247     3. When matching a UTF-8 string, the test for a valid string at the start has
1248        been extended. If start_offset is not zero, PCRE now checks that it points
1249        to a byte that is the start of a UTF-8 character. If not, it returns
1250        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1251        this is necessary because there may be backward assertions in the pattern.
1252        When matching the same subject several times, it may save resources to use
1253        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1254    
1255     4. The code for checking the validity of UTF-8 strings has been tightened so
1256        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1257        containing "overlong sequences".
1258    
1259     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1260        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1261        should have been "&", but it just so happened that all the cases this let
1262        through by mistake were picked up later in the function.
1263    
1264     6. I had used a variable called "isblank" - this is a C99 function, causing
1265        some compilers to warn. To avoid this, I renamed it (as "blankclass").
1266    
1267     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1268        prompting; (b) run "./pcretest /dev/null" at the start of the test script
1269        so the version is shown; (c) stop "make test" echoing "./RunTest".
1270    
1271     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1272    
1273     9. The prototype for memmove() for systems that don't have it was using
1274        size_t, but the inclusion of the header that defines size_t was later. I've
1275        moved the #includes for the C headers earlier to avoid this.
1276    
1277    10. Added some adjustments to the code to make it easier to compiler on certain
1278        special systems:
1279    
1280          (a) Some "const" qualifiers were missing.
1281          (b) Added the macro EXPORT before all exported functions; by default this
1282              is defined to be empty.
1283          (c) Changed the dftables auxiliary program (that builds chartables.c) so
1284              that it reads its output file name as an argument instead of writing
1285              to the standard output and assuming this can be redirected.
1286    
1287    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
1288        class containing characters with values greater than 255, PCRE compilation
1289        went into a loop.
1290    
1291    12. A recursive reference to a subpattern that was within another subpattern
1292        that had a minimum quantifier of zero caused PCRE to crash. For example,
1293        (x(y(?2))z)? provoked this bug with a subject that got as far as the
1294        recursion. If the recursively-called subpattern itself had a zero repeat,
1295        that was OK.
1296    
1297    13. In pcretest, the buffer for reading a data line was set at 30K, but the
1298        buffer into which it was copied (for escape processing) was still set at
1299        1024, so long lines caused crashes.
1300    
1301    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
1302        "internal error: code overflow...". This applied to any character class
1303        that was followed by a possessive quantifier.
1304    
1305    15. Modified the Makefile to add libpcre.la as a prerequisite for
1306        libpcreposix.la because I was told this is needed for a parallel build to
1307        work.
1308    
1309    16. If a pattern that contained .* following optional items at the start was
1310        studied, the wrong optimizing data was generated, leading to matching
1311        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
1312        matching string must start with a or b or c. The correct conclusion for
1313        this pattern is that a match can start with any character.
1314    
1315    
1316    Version 4.4 13-Aug-03
1317    ---------------------
1318    
1319     1. In UTF-8 mode, a character class containing characters with values between
1320        127 and 255 was not handled correctly if the compiled pattern was studied.
1321        In fixing this, I have also improved the studying algorithm for such
1322        classes (slightly).
1323    
1324     2. Three internal functions had redundant arguments passed to them. Removal
1325        might give a very teeny performance improvement.
1326    
1327     3. Documentation bug: the value of the capture_top field in a callout is *one
1328        more than* the number of the hightest numbered captured substring.
1329    
1330     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
1331        in incorrectly linking with a previously installed version. They now link
1332        explicitly with libpcre.la.
1333    
1334     5. configure.in no longer needs to recognize Cygwin specially.
1335    
1336     6. A problem in pcre.in for Windows platforms is fixed.
1337    
1338     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
1339        pcretest, it used to include the size of the study block as part of its
1340        output. Unfortunately, the structure contains a field that has a different
1341        size on different hardware architectures. This meant that the tests that
1342        showed this size failed. As the block is currently always of a fixed size,
1343        this information isn't actually particularly useful in pcretest output, so
1344        I have just removed it.
1345    
1346     8. Three pre-processor statements accidentally did not start in column 1.
1347        Sadly, there are *still* compilers around that complain, even though
1348        standard C has not required this for well over a decade. Sigh.
1349    
1350     9. In pcretest, the code for checking callouts passed small integers in the
1351        callout_data field, which is a void * field. However, some picky compilers
1352        complained about the casts involved for this on 64-bit systems. Now
1353        pcretest passes the address of the small integer instead, which should get
1354        rid of the warnings.
1355    
1356    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
1357        both compile and run time, and gives an error if an invalid UTF-8 sequence
1358        is found. There is a option for disabling this check in cases where the
1359        string is known to be correct and/or the maximum performance is wanted.
1360    
1361    11. In response to a bug report, I changed one line in Makefile.in from
1362    
1363            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
1364        to
1365            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
1366    
1367        to look similar to other lines, but I have no way of telling whether this
1368        is the right thing to do, as I do not use Windows. No doubt I'll get told
1369        if it's wrong...
1370    
1371    
1372    Version 4.3 21-May-03
1373    ---------------------
1374    
1375    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
1376       Makefile.
1377    
1378    2. Some refactoring to improve the quality of the code:
1379    
1380       (i)   The utf8_table... variables are now declared "const".
1381    
1382       (ii)  The code for \cx, which used the "case flipping" table to upper case
1383             lower case letters, now just substracts 32. This is ASCII-specific,
1384             but the whole concept of \cx is ASCII-specific, so it seems
1385             reasonable.
1386    
1387       (iii) PCRE was using its character types table to recognize decimal and
1388             hexadecimal digits in the pattern. This is silly, because it handles
1389             only 0-9, a-f, and A-F, but the character types table is locale-
1390             specific, which means strange things might happen. A private
1391             table is now used for this - though it costs 256 bytes, a table is
1392             much faster than multiple explicit tests. Of course, the standard
1393             character types table is still used for matching digits in subject
1394             strings against \d.
1395    
1396       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
1397             ending in _t are). So I've renamed it as ESC_tee.
1398    
1399    3. The first argument for regexec() in the POSIX wrapper should have been
1400       defined as "const".
1401    
1402    4. Changed pcretest to use malloc() for its buffers so that they can be
1403       Electric Fenced for debugging.
1404    
1405    5. There were several places in the code where, in UTF-8 mode, PCRE would try
1406       to read one or more bytes before the start of the subject string. Often this
1407       had no effect on PCRE's behaviour, but in some circumstances it could
1408       provoke a segmentation fault.
1409    
1410    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
1411       to try to read one or more bytes before the start of the subject string.
1412    
1413    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
1414       UTF-8 support could misbehave in various ways if the subject string
1415       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
1416       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
1417       back over UTF-8 characters.)
1418    
1419    
1420    Version 4.2 14-Apr-03
1421    ---------------------
1422    
1423    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
1424    
1425    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
1426         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
1427         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
1428         [WIN_PREFIX]: new variable, "cyg" for Cygwin
1429         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
1430           and BUILD_EXEEXT
1431         Note: automatic setting of the BUILD variables is not yet working
1432         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
1433           compile-time but not at link-time
1434         [LINK]: use for linking executables only
1435         make different versions for Windows and non-Windows
1436         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
1437           libraries
1438         [LINK_FOR_BUILD]: new variable
1439         [OBJEXT]: use throughout
1440         [EXEEXT]: use throughout
1441         <winshared>: new target
1442         <wininstall>: new target
1443         <dftables.o>: use native compiler
1444         <dftables>: use native linker
1445         <install>: handle Windows platform correctly
1446         <clean>: ditto
1447         <check>: ditto
1448         copy DLL to top builddir before testing
1449    
1450       As part of these changes, -no-undefined was removed again. This was reported
1451       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
1452       in any case.
1453    
1454    3. Some tidies to get rid of compiler warnings:
1455    
1456       . In the match_data structure, match_limit was an unsigned long int, whereas
1457         match_call_count was an int. I've made them both unsigned long ints.
1458    
1459       . In pcretest the fact that a const uschar * doesn't automatically cast to
1460         a void * provoked a warning.
1461    
1462       . Turning on some more compiler warnings threw up some "shadow" variables
1463         and a few more missing casts.
1464    
1465    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
1466       option, a class that contained a single character with a value between 128
1467       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
1468    
1469    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
1470       option, a class that contained several characters, but with at least one
1471       whose value was between 128 and 255 caused PCRE to crash.
1472    
1473    
1474    Version 4.1 12-Mar-03
1475    ---------------------
1476    
1477    1. Compiling with gcc -pedantic found a couple of places where casts were
1478    needed, and a string in dftables.c that was longer than standard compilers are
1479    required to support.
1480    
1481    2. Compiling with Sun's compiler found a few more places where the code could
1482    be tidied up in order to avoid warnings.
1483    
1484    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
1485    first of these names is deprecated in the latest Autoconf in favour of the name
1486    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
1487    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
1488    analogy I have changed it to CFLAGS_FOR_BUILD.
1489    
1490    4. Added -no-undefined to the linking command in the Makefile, because this is
1491    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
1492    linking step for the pcreposix library.
1493    
1494    5. PCRE was failing to diagnose the case of two named groups with the same
1495    name.
1496    
1497    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
1498    literal character that is needed in the subject for a match, and scans along to
1499    ensure that it is present before embarking on the full matching process. This
1500    saves time in cases of nested unlimited repeats that are never going to match.
1501    Problem: the scan can take a lot of time if the subject is very long (e.g.
1502    megabytes), thus penalizing straightforward matches. It is now done only if the
1503    amount of subject to be scanned is less than 1000 bytes.
1504    
1505    7. A lesser problem with the same optimization is that it was recording the
1506    first character of an anchored pattern as "needed", thus provoking a search
1507    right along the subject, even when the first match of the pattern was going to
1508    fail. The "needed" character is now not set for anchored patterns, unless it
1509    follows something in the pattern that is of non-fixed length. Thus, it still
1510    fulfils its original purpose of finding quick non-matches in cases of nested
1511    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
1512    
1513    
1514    Version 4.0 17-Feb-03
1515    ---------------------
1516    
1517    1. If a comment in an extended regex that started immediately after a meta-item
1518    extended to the end of string, PCRE compiled incorrect data. This could lead to
1519    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
1520    
1521    2. Moved to autoconf 2.53 and libtool 1.4.2.
1522    
1523    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
1524    the special perltest8 script is no longer needed - all the tests can be run
1525    from a single perltest script.
1526    
1527    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
1528    by \s. It has now been removed in PCRE. This means it isn't recognized as
1529    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
1530    class [:space:] *does* include VT, thereby creating a mess.
1531    
1532    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
1533    space and tab.
1534    
1535    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
1536    its new features into the main test script, reducing the number of scripts.
1537    
1538    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
1539    were backward compatible, and made the (?i) apply to the whole pattern, as if
1540    /i were given. Now it behaves more logically, and applies the option setting
1541    only to what follows. PCRE has been changed to follow suit. However, if it
1542    finds options settings right at the start of the pattern, it extracts them into
1543    the global options, as before. Thus, they show up in the info data.
1544    
1545    8. Added support for the \Q...\E escape sequence. Characters in between are
1546    treated as literals. This is slightly different from Perl in that $ and @ are
1547    also handled as literals inside the quotes. In Perl, they will cause variable
1548    interpolation. Note the following examples:
1549    
1550        Pattern            PCRE matches      Perl matches
1551    
1552        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
1553        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
1554        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
1555    
1556    For compatibility with Perl, \Q...\E sequences are recognized inside character
1557    classes as well as outside them.
1558    
1559    9. Re-organized 3 code statements in pcretest to avoid "overflow in
1560    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
1561    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
1562    signed/unsigned warnings.
1563    
1564    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
1565    option for pcretest, so I've replaced it by a simple function that does just
1566    that job.
1567    
1568    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
1569    "pcregrep -".
1570    
1571    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
1572    Java package. This provides some syntactic sugar for simple cases of what my
1573    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
1574    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
1575    item, you can use this simplified notation. Note that only makes sense with
1576    greedy quantifiers. Consequently, the use of the possessive quantifier forces
1577    greediness, whatever the setting of the PCRE_UNGREEDY option.
1578    
1579    13. A change of greediness default within a pattern was not taking effect at
1580    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
1581    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
1582    was abstracted outside.
1583    
1584    14. PCRE now supports the \G assertion. It is true when the current matching
1585    position is at the start point of the match. This differs from \A when the
1586    starting offset is non-zero. Used with the /g option of pcretest (or similar
1587    code), it works in the same way as it does for Perl's /g option. If all
1588    alternatives of a regex begin with \G, the expression is anchored to the start
1589    match position, and the "anchored" flag is set in the compiled expression.
1590    
1591    15. Some bugs concerning the handling of certain option changes within patterns
1592    have been fixed. These applied to options other than (?ims). For example,
1593    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
1594    been the other way round. Some of this was related to change 7 above.
1595    
1596    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
1597    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
1598    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
1599    POSIX classes only within a class (e.g. /[[:alpha:]]/).
1600    
1601    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
1602    mode. Unlike ".", it always matches newline, whatever the setting of
1603    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
1604    assertions. Perl allows it, but it doesn't (in general) work because it can't
1605    calculate the length of the lookbehind. At least, that's the case for Perl
1606    5.8.0 - I've been told they are going to document that it doesn't work in
1607    future.
1608    
1609    18. Added an error diagnosis for escapes that PCRE does not support: these are
1610    \L, \l, \N, \P, \p, \U, \u, and \X.
1611    
1612    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
1613    reading past the end of the pattern in cases such as /[abcd/.
1614    
1615    20. PCRE was getting more memory than necessary for patterns with classes that
1616    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
1617    
1618    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
1619    compiling PCRE for use with Virtual Pascal.
1620    
1621    22. Small fix to the Makefile to make it work properly if the build is done
1622    outside the source tree.
1623    
1624    23. Added a new extension: a condition to go with recursion. If a conditional
1625    subpattern starts with (?(R) the "true" branch is used if recursion has
1626    happened, whereas the "false" branch is used only at the top level.
1627    
1628    24. When there was a very long string of literal characters (over 255 bytes
1629    without UTF support, over 250 bytes with UTF support), the computation of how
1630    much memory was required could be incorrect, leading to segfaults or other
1631    strange effects.
1632    
1633    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
1634    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
1635    there was a subsequent back reference to those brackets. This meant that, for
1636    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
1637    possible to check for precisely this case. All we can do is abandon the
1638    optimization if .* occurs inside capturing brackets when there are any back
1639    references whatsoever. (See below for a better fix that came later.)
1640    
1641    26. The handling of the optimization for finding the first character of a
1642    non-anchored pattern, and for finding a character that is required later in the
1643    match were failing in some cases. This didn't break the matching; it just
1644    failed to optimize when it could. The way this is done has been re-implemented.
1645    
1646    27. Fixed typo in error message for invalid (?R item (it said "(?p").
1647    
1648    28. Added a new feature that provides some of the functionality that Perl
1649    provides with (?{...}). The facility is termed a "callout". The way it is done
1650    in PCRE is for the caller to provide an optional function, by setting
1651    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
1652    global variable. By default it is unset, which disables all calling out. To get
1653    the function called, the regex must include (?C) at appropriate points. This
1654    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
1655    This provides a means of identifying different callout points. When PCRE
1656    reaches such a point in the regex, if pcre_callout has been set, the external
1657    function is called. It is provided with data in a structure called
1658    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
1659    matching continues; if it returns a non-zero value, the match at the current
1660    point fails. However, backtracking will occur if possible. [This was changed
1661    later and other features added - see item 49 below.]
1662    
1663    29. pcretest is upgraded to test the callout functionality. It provides a
1664    callout function that displays information. By default, it shows the start of
1665    the match and the current position in the text. There are some new data escapes
1666    to vary what happens:
1667    
1668        \C+         in addition, show current contents of captured substrings
1669        \C-         do not supply a callout function
1670        \C!n        return 1 when callout number n is reached
1671        \C!n!m      return 1 when callout number n is reached for the mth time
1672    
1673    30. If pcregrep was called with the -l option and just a single file name, it
1674    output "<stdin>" if a match was found, instead of the file name.
1675    
1676    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
1677    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
1678    pcre_exec(). This saves a malloc/free per call. The default value of
1679    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
1680    when configuring.
1681    
1682    32. The default maximum size of a compiled pattern is 64K. There have been a
1683    few cases of people hitting this limit. The code now uses macros to handle the
1684    storing of links as offsets within the compiled pattern. It defaults to 2-byte
1685    links, but this can be changed to 3 or 4 bytes by --with-link-size when
1686    configuring. Tests 2 and 5 work only with 2-byte links because they output
1687    debugging information about compiled patterns.
1688    
1689    33. Internal code re-arrangements:
1690    
1691    (a) Moved the debugging function for printing out a compiled regex into
1692        its own source file (printint.c) and used #include to pull it into
1693        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
1694        separate copies.
1695    
1696    (b) Defined the list of op-code names for debugging as a macro in
1697        internal.h so that it is next to the definition of the opcodes.
1698    
1699    (c) Defined a table of op-code lengths for simpler skipping along compiled
1700        code. This is again a macro in internal.h so that it is next to the
1701        definition of the opcodes.
1702    
1703    34. Added support for recursive calls to individual subpatterns, along the
1704    lines of Robin Houston's patch (but implemented somewhat differently).
1705    
1706    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
1707    allow it to read and process whole directories in Win32. This code was
1708    contributed by Lionel Fourquaux; it has not been tested by me.
1709    
1710    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
1711    used to name a group. Names consist of alphanumerics and underscores, and must
1712    be unique. Back references use the syntax (?P=name) and recursive calls use
1713    (?P>name) which is a PCRE extension to the Python extension. Groups still have
1714    numbers. The function pcre_fullinfo() can be used after compilation to extract
1715    a name/number map. There are three relevant calls:
1716    
1717      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
1718      PCRE_INFO_NAMECOUNT            yields the number of entries
1719      PCRE_INFO_NAMETABLE            yields a pointer to the map.
1720    
1721    The map is a vector of fixed-size entries. The size of each entry depends on
1722    the length of the longest name used. The first two bytes of each entry are the
1723    group number, most significant byte first. There follows the corresponding
1724    name, zero terminated. The names are in alphabetical order.
1725    
1726    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
1727    case instead of 255. Making it the same both with and without UTF-8 support
1728    means that the same test output works with both.
1729    
1730    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
1731    calling malloc() with a zero argument.
1732    
1733    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
1734    optimization. I've improved things by keeping a bitmap of backreferences with
1735    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
1736    fact referenced, the optimization can be applied. It is unlikely that a
1737    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
1738    the match to follow \n) will appear inside brackets with a number greater than
1739    31, but if it does, any back reference > 31 suppresses the optimization.
1740    
1741    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
1742    of disabling numbered capturing parentheses. Any opening parenthesis that is
1743    not followed by ? behaves as if it were followed by ?: but named parentheses
1744    can still be used for capturing (and they will acquire numbers in the usual
1745    way).
1746    
1747    41. Redesigned the return codes from the match() function into yes/no/error so
1748    that errors can be passed back from deep inside the nested calls. A malloc
1749    failure while inside a recursive subpattern call now causes the
1750    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
1751    
1752    42. It is now possible to set a limit on the number of times the match()
1753    function is called in a call to pcre_exec(). This facility makes it possible to
1754    limit the amount of recursion and backtracking, though not in a directly
1755    obvious way, because the match() function is used in a number of different
1756    circumstances. The count starts from zero for each position in the subject
1757    string (for non-anchored patterns). The default limit is, for compatibility, a
1758    large number, namely 10 000 000. You can change this in two ways:
1759    
1760    (a) When configuring PCRE before making, you can use --with-match-limit=n
1761        to set a default value for the compiled library.
1762    
1763    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
1764        a different value is set. See 45 below.
1765    
1766    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
1767    
1768    43. Added a new function pcre_config(int, void *) to enable run-time extraction
1769    of things that can be changed at compile time. The first argument specifies
1770    what is wanted and the second points to where the information is to be placed.
1771    The current list of available information is:
1772    
1773      PCRE_CONFIG_UTF8
1774    
1775    The output is an integer that is set to one if UTF-8 support is available;
1776    otherwise it is set to zero.
1777    
1778      PCRE_CONFIG_NEWLINE
1779    
1780    The output is an integer that it set to the value of the code that is used for
1781    newline. It is either LF (10) or CR (13).
1782    
1783      PCRE_CONFIG_LINK_SIZE
1784    
1785    The output is an integer that contains the number of bytes used for internal
1786    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
1787    
1788      PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
1789    
1790    The output is an integer that contains the threshold above which the POSIX
1791    interface uses malloc() for output vectors. See item 31 above.
1792    
1793      PCRE_CONFIG_MATCH_LIMIT
1794    
1795    The output is an unsigned integer that contains the default limit of the number
1796    of match() calls in a pcre_exec() execution. See 42 above.
1797    
1798    44. pcretest has been upgraded by the addition of the -C option. This causes it
1799    to extract all the available output from the new pcre_config() function, and to
1800    output it. The program then exits immediately.
1801    
1802    45. A need has arisen to pass over additional data with calls to pcre_exec() in
1803    order to support additional features. One way would have been to define
1804    pcre_exec2() (for example) with extra arguments, but this would not have been
1805    extensible, and would also have required all calls to the original function to
1806    be mapped to the new one. Instead, I have chosen to extend the mechanism that
1807    is used for passing in "extra" data from pcre_study().
1808    
1809    The pcre_extra structure is now exposed and defined in pcre.h. It currently
1810    contains the following fields:
1811    
1812      flags         a bitmap indicating which of the following fields are set
1813      study_data    opaque data from pcre_study()
1814      match_limit   a way of specifying a limit on match() calls for a specific
1815                      call to pcre_exec()
1816      callout_data  data for callouts (see 49 below)
1817    
1818    The flag bits are also defined in pcre.h, and are
1819    
1820      PCRE_EXTRA_STUDY_DATA
1821      PCRE_EXTRA_MATCH_LIMIT
1822      PCRE_EXTRA_CALLOUT_DATA
1823    
1824    The pcre_study() function now returns one of these new pcre_extra blocks, with
1825    the actual study data pointed to by the study_data field, and the
1826    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
1827    before. That is, this change is entirely upwards-compatible and requires no
1828    change to existing code.
1829    
1830    If you want to pass in additional data to pcre_exec(), you can either place it
1831    in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
1832    block.
1833    
1834    46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
1835    data string contains the escape sequence \M, pcretest calls pcre_exec() several
1836    times with different match limits, until it finds the minimum value needed for
1837    pcre_exec() to complete. The value is then output. This can be instructive; for
1838    most simple matches the number is quite small, but for pathological cases it
1839    gets very large very quickly.
1840    
1841    47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
1842    returns the size of the data block pointed to by the study_data field in a
1843    pcre_extra block, that is, the value that was passed as the argument to
1844    pcre_malloc() when PCRE was getting memory in which to place the information
1845    created by pcre_study(). The fourth argument should point to a size_t variable.
1846    pcretest has been extended so that this information is shown after a successful
1847    pcre_study() call when information about the compiled regex is being displayed.
1848    
1849    48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
1850    because what follows is always an absolute path. (Later: it turns out that this
1851    is more than cosmetic for MinGW, because it doesn't like empty path
1852    components.)
1853    
1854    49. Some changes have been made to the callout feature (see 28 above):
1855    
1856    (i)  A callout function now has three choices for what it returns:
1857    
1858           0  =>  success, carry on matching
1859         > 0  =>  failure at this point, but backtrack if possible
1860         < 0  =>  serious error, return this value from pcre_exec()
1861    
1862         Negative values should normally be chosen from the set of PCRE_ERROR_xxx
1863         values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
1864         "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
1865         use by callout functions. It will never be used by PCRE itself.
1866    
1867    (ii) The pcre_extra structure (see 45 above) has a void * field called
1868         callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
1869         pcre_callout_block structure has a field of the same name. The contents of
1870         the field passed in the pcre_extra structure are passed to the callout
1871         function in the corresponding field in the callout block. This makes it
1872         easier to use the same callout-containing regex from multiple threads. For
1873         testing, the pcretest program has a new data escape
1874    
1875           \C*n        pass the number n (may be negative) as callout_data
1876    
1877         If the callout function in pcretest receives a non-zero value as
1878         callout_data, it returns that value.
1879    
1880    50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
1881    there were some redundant $(CFLAGS) in commands that are now specified as
1882    $(LINK), which already includes $(CFLAGS).
1883    
1884    51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
1885    has been compiled with UTF-8 support *and* pcre_compile() has been compiled
1886    with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
1887    one-byte characters throughout. Note that case-insensitive matching applies
1888    only to characters whose values are less than 256. PCRE doesn't support the
1889    notion of cases for higher-valued characters.
1890    
1891    (i)   A character class whose characters are all within 0-255 is handled as
1892          a bit map, and the map is inverted for negative classes. Previously, a
1893          character > 255 always failed to match such a class; however it should
1894          match if the class was a negative one (e.g. [^ab]). This has been fixed.
1895    
1896    (ii)  A negated character class with a single character < 255 is coded as
1897          "not this character" (OP_NOT). This wasn't working properly when the test
1898          character was multibyte, either singly or repeated.
1899    
1900    (iii) Repeats of multibyte characters are now handled correctly in UTF-8
1901          mode, for example: \x{100}{2,3}.
1902    
1903    (iv)  The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
1904          singly or repeated) now correctly test multibyte characters. However,
1905          PCRE doesn't recognize any characters with values greater than 255 as
1906          digits, spaces, or word characters. Such characters always match \D, \S,
1907          and \W, and never match \d, \s, or \w.
1908    
1909    (v)   Classes may now contain characters and character ranges with values
1910          greater than 255. For example: [ab\x{100}-\x{400}].
1911    
1912    (vi)  pcregrep now has a --utf-8 option (synonym -u) which makes it call
1913          PCRE in UTF-8 mode.
1914    
1915    52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
1916    PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
1917    retained for backwards compatibility. (Note that LASTLITERAL is also a byte
1918    value.)
1919    
1920    53. The single man page has become too large. I have therefore split it up into
1921    a number of separate man pages. These also give rise to individual HTML pages;
1922    these are now put in a separate directory, and there is an index.html page that
1923    lists them all. Some hyperlinking between the pages has been installed.
1924    
1925    54. Added convenience functions for handling named capturing parentheses.
1926    
1927    55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
1928    aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
1929    true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
1930    are faulted.
1931    
1932    56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
1933    calling configure. These values are used when compiling the dftables.c program
1934    which is run to generate the source of the default character tables. They
1935    default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
1936    you will need to set these values.
1937    
1938    57. Updated the building process for Windows DLL, as provided by Fred Cox.
1939    
1940    
1941    Version 3.9 02-Jan-02
1942    ---------------------
1943    
1944    1. A bit of extraneous text had somehow crept into the pcregrep documentation.
1945    
1946    2. If --disable-static was given, the building process failed when trying to
1947    build pcretest and pcregrep. (For some reason it was using libtool to compile
1948    them, which is not right, as they aren't part of the library.)
1949    
1950    
1951    Version 3.8 18-Dec-01
1952    ---------------------
1953    
1954    1. The experimental UTF-8 code was completely screwed up. It was packing the
1955    bytes in the wrong order. How dumb can you get?
1956    
1957    
1958    Version 3.7 29-Oct-01
1959    ---------------------
1960    
1961    1. In updating pcretest to check change 1 of version 3.6, I screwed up.
1962    This caused pcretest, when used on the test data, to segfault. Unfortunately,
1963    this didn't happen under Solaris 8, where I normally test things.
1964    
1965    2. The Makefile had to be changed to make it work on BSD systems, where 'make'
1966    doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
1967    isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
1968    this fix an hour or so after the initial 3.7 release.)
1969    
1970    
1971    Version 3.6 23-Oct-01
1972    ---------------------
1973    
1974    1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
1975    offsets passed as NULL with zero offset count.
1976    
1977    2. The config.guess and config.sub files had not been updated when I moved to
1978    the latest autoconf.
1979    
1980    
1981    Version 3.5 15-Aug-01
1982    ---------------------
1983    
1984    1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
1985    had been forgotten.
1986    
1987    2. By using declared but undefined structures, we can avoid using "void"
1988    definitions in pcre.h while keeping the internal definitions of the structures
1989    private.
1990    
1991    3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
1992    user point of view, this means that both static and shared libraries are built
1993    by default, but this can be individually controlled. More of the work of
1994    handling this static/shared cases is now inside libtool instead of PCRE's make
1995    file.
1996    
1997    4. The pcretest utility is now installed along with pcregrep because it is
1998    useful for users (to test regexs) and by doing this, it automatically gets
1999    relinked by libtool. The documentation has been turned into a man page, so
2000    there are now .1, .txt, and .html versions in /doc.
2001    
2002    5. Upgrades to pcregrep:
2003       (i)   Added long-form option names like gnu grep.
2004       (ii)  Added --help to list all options with an explanatory phrase.
2005       (iii) Added -r, --recursive to recurse into sub-directories.
2006       (iv)  Added -f, --file to read patterns from a file.
2007    
2008    6. pcre_exec() was referring to its "code" argument before testing that
2009    argument for NULL (and giving an error if it was NULL).
2010    
2011    7. Upgraded Makefile.in to allow for compiling in a different directory from
2012    the source directory.
2013    
2014    8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2015    options bits, the pointer it was passed was to an int instead of to an unsigned
2016    long int. This mattered only on 64-bit systems.
2017    
2018    9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2019    generated) instead of pcre.in, which it its source. Also made the same change
2020    in several of the .c files.
2021    
2022    10. A new release of gcc defines printf() as a macro, which broke pcretest
2023    because it had an ifdef in the middle of a string argument for printf(). Fixed
2024    by using separate calls to printf().
2025    
2026    11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2027    script, to force use of CR or LF instead of \n in the source. On non-Unix
2028    systems, the value can be set in config.h.
2029    
2030    12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2031    absolute limit. Changed the text of the error message to make this clear, and
2032    likewise updated the man page.
2033    
2034    13. The limit of 99 on the number of capturing subpatterns has been removed.
2035    The new limit is 65535, which I hope will not be a "real" limit.
2036    
2037    
2038    Version 3.4 22-Aug-00
2039    ---------------------
2040    
2041    1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2042    
2043    2. Diagnose condition (?(0) as an error instead of crashing on matching.
2044    
2045    
2046    Version 3.3 01-Aug-00
2047    ---------------------
2048    
2049    1. If an octal character was given, but the value was greater than \377, it
2050    was not getting masked to the least significant bits, as documented. This could
2051    lead to crashes in some systems.
2052    
2053    2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
2054    the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
2055    
2056    3. Added the functions pcre_free_substring() and pcre_free_substring_list().
2057    These just pass their arguments on to (pcre_free)(), but they are provided
2058    because some uses of PCRE bind it to non-C systems that can call its functions,
2059    but cannot call free() or pcre_free() directly.
2060    
2061    4. Add "make test" as a synonym for "make check". Corrected some comments in
2062    the Makefile.
2063    
2064    5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
2065    Makefile.
2066    
2067    6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
2068    command called pgrep for grepping around the active processes.
2069    
2070    7. Added the beginnings of support for UTF-8 character strings.
2071    
2072    8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
2073    RANLIB to ./ltconfig so that they are used by libtool. I think these are all
2074    the relevant ones. (AR is not passed because ./ltconfig does its own figuring
2075    out for the ar command.)
2076    
2077    
2078    Version 3.2 12-May-00
2079    ---------------------
2080    
2081    This is purely a bug fixing release.
2082    
2083    1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
2084    of ZA. This was just one example of several cases that could provoke this bug,
2085    which was introduced by change 9 of version 2.00. The code for breaking
2086    infinite loops after an iteration that matches an empty string was't working
2087    correctly.
2088    
2089    2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
2090    when matched against abbab (for example). After matching an empty string, it
2091    wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
2092    caused it to match further down the string than it should.
2093    
2094    3. The code contained an inclusion of sys/types.h. It isn't clear why this
2095    was there because it doesn't seem to be needed, and it causes trouble on some
2096    systems, as it is not a Standard C header. It has been removed.
2097    
2098    4. Made 4 silly changes to the source to avoid stupid compiler warnings that
2099    were reported on the Macintosh. The changes were from
2100    
2101      while ((c = *(++ptr)) != 0 && c != '\n');
2102    to
2103      while ((c = *(++ptr)) != 0 && c != '\n') ;
2104    
2105    Totally extraordinary, but if that's what it takes...
2106    
2107    5. PCRE is being used in one environment where neither memmove() nor bcopy() is
2108    available. Added HAVE_BCOPY and an autoconf test for it; if neither
2109    HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
2110    assumes the way PCRE uses memmove() (always moving upwards).
2111    
2112    6. PCRE is being used in one environment where strchr() is not available. There
2113    was only one use in pcre.c, and writing it out to avoid strchr() probably gives
2114    faster code anyway.
2115    
2116    
2117    Version 3.1 09-Feb-00
2118    ---------------------
2119    
2120    The only change in this release is the fixing of some bugs in Makefile.in for
2121    the "install" target:
2122    
2123    (1) It was failing to install pcreposix.h.
2124    
2125    (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
2126    
2127    
2128    Version 3.0 01-Feb-00
2129    ---------------------
2130    
2131    1. Add support for the /+ modifier to perltest (to output $` like it does in
2132    pcretest).
2133    
2134    2. Add support for the /g modifier to perltest.
2135    
2136    3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
2137    matches null strings.
2138    
2139    4. Fix perltest so that it doesn't do unwanted things when fed an empty
2140    pattern. Perl treats empty patterns specially - it reuses the most recent
2141    pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
2142    effect.
2143    
2144    5. The POSIX interface was broken in that it was just handing over the POSIX
2145    captured string vector to pcre_exec(), but (since release 2.00) PCRE has
2146    required a bigger vector, with some working space on the end. This means that
2147    the POSIX wrapper now has to get and free some memory, and copy the results.
2148    
2149    6. Added some simple autoconf support, placing the test data and the
2150    documentation in separate directories, re-organizing some of the
2151    information files, and making it build pcre-config (a GNU standard). Also added
2152    libtool support for building PCRE as a shared library, which is now the
2153    default.
2154    
2155    7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
2156    09 are not valid octal constants. Single digits will be used for minor values
2157    less than 10.
2158    
2159    8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
2160    existing programs that set these in the POSIX interface can use PCRE without
2161    modification.
2162    
2163    9. Added a new function, pcre_fullinfo() with an extensible interface. It can
2164    return all that pcre_info() returns, plus additional data. The pcre_info()
2165    function is retained for compatibility, but is considered to be obsolete.
2166    
2167    10. Added experimental recursion feature (?R) to handle one common case that
2168    Perl 5.6 will be able to do with (?p{...}).
2169    
2170    11. Added support for POSIX character classes like [:alpha:], which Perl is
2171    adopting.
2172    
2173    
2174    Version 2.08 31-Aug-99
2175    ----------------------
2176    
2177    1. When startoffset was not zero and the pattern began with ".*", PCRE was not
2178    trying to match at the startoffset position, but instead was moving forward to
2179    the next newline as if a previous match had failed.
2180    
2181    2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
2182    and could get into a loop if a null string was matched other than at the start
2183    of the subject.
2184    
2185    3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
2186    be distinguished at compile time, and for completeness also added PCRE_DATE.
2187    
2188    5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
2189    in GnuWin32 environments.
2190    
2191    
2192    Version 2.07 29-Jul-99
2193    ----------------------
2194    
2195    1. The documentation is now supplied in plain text form and HTML as well as in
2196    the form of man page sources.
2197    
2198    2. C++ compilers don't like assigning (void *) values to other pointer types.
2199    In particular this affects malloc(). Although there is no problem in Standard
2200    C, I've put in casts to keep C++ compilers happy.
2201    
2202    3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
2203    should be (const char *).
2204    
2205    4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
2206    be useful for non-Unix systems who don't want to bother with the POSIX stuff.
2207    However, I haven't made this a standard facility. The documentation doesn't
2208    mention it, and the Makefile doesn't support it.
2209    
2210    5. The Makefile now contains an "install" target, with editable destinations at
2211    the top of the file. The pcretest program is not installed.
2212    
2213    6. pgrep -V now gives the PCRE version number and date.
2214    
2215    7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
2216    causing the entire string to be ignored, instead of just the last character.
2217    
2218    8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
2219    non-matching string, it can take a very, very long time, even for strings of
2220    quite modest length, because of the nested recursion. PCRE now does better in
2221    some of these cases. It does this by remembering the last required literal
2222    character in the pattern, and pre-searching the subject to ensure it is present
2223    before running the real match. In other words, it applies a heuristic to detect
2224    some types of certain failure quickly, and in the above example, if presented
2225    with a string that has no trailing " it gives "no match" very quickly.
2226    
2227    9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
2228    other alternatives are tried instead.
2229    
2230    
2231    Version 2.06 09-Jun-99
2232    ----------------------
2233    
2234    1. Change pcretest's output for amount of store used to show just the code
2235    space, because the remainder (the data block) varies in size between 32-bit and
2236    64-bit systems.
2237    
2238    2. Added an extra argument to pcre_exec() to supply an offset in the subject to
2239    start matching at. This allows lookbehinds to work when searching for multiple
2240    occurrences in a string.
2241    
2242    3. Added additional options to pcretest for testing multiple occurrences:
2243    
2244       /+   outputs the rest of the string that follows a match
2245       /g   loops for multiple occurrences, using the new startoffset argument
2246       /G   loops for multiple occurrences by passing an incremented pointer
2247    
2248    4. PCRE wasn't doing the "first character" optimization for patterns starting
2249    with \b or \B, though it was doing it for other lookbehind assertions. That is,
2250    it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
2251    the letter 'x'. On long subject strings, this gives a significant speed-up.
2252    
2253    
2254    Version 2.05 21-Apr-99
2255    ----------------------
2256    
2257    1. Changed the type of magic_number from int to long int so that it works
2258    properly on 16-bit systems.
2259    
2260    2. Fixed a bug which caused patterns starting with .* not to work correctly
2261    when the subject string contained newline characters. PCRE was assuming
2262    anchoring for such patterns in all cases, which is not correct because .* will
2263    not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
2264    DOTALL is set at top level; otherwise it knows that patterns starting with .*
2265    must be retried after every newline in the subject.
2266    
2267    
2268    Version 2.04 18-Feb-99
2269    ----------------------
2270    
2271    1. For parenthesized subpatterns with repeats whose minimum was zero, the
2272    computation of the store needed to hold the pattern was incorrect (too large).
2273    If such patterns were nested a few deep, this could multiply and become a real
2274    problem.
2275    
2276    2. Added /M option to pcretest to show the memory requirement of a specific
2277    pattern. Made -m a synonym of -s (which does this globally) for compatibility.
2278    
2279    3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
2280    compiled in such a way that the backtracking after subsequent failure was
2281    pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
2282    ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
2283    
2284    
2285    Version 2.03 02-Feb-99
2286    ----------------------
2287    
2288    1. Fixed typo and small mistake in man page.
2289    
2290    2. Added 4th condition (GPL supersedes if conflict) and created separate
2291    LICENCE file containing the conditions.
2292    
2293    3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
2294    Perl, that is the internal \ allows the delimiter to be included in the
2295    pattern. Locked out the use of \ as a delimiter. If \ immediately follows
2296    the final delimiter, add \ to the end of the pattern (to test the error).
2297    
2298    4. Added the convenience functions for extracting substrings after a successful
2299    match. Updated pcretest to make it able to test these functions.
2300    
2301    
2302    Version 2.02 14-Jan-99
2303    ----------------------
2304    
2305    1. Initialized the working variables associated with each extraction so that
2306    their saving and restoring doesn't refer to uninitialized store.
2307    
2308    2. Put dummy code into study.c in order to trick the optimizer of the IBM C
2309    compiler for OS/2 into generating correct code. Apparently IBM isn't going to
2310    fix the problem.
2311    
2312    3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
2313    calls, and wasn't printing the correct value for compiling calls. Increased the
2314    default value of LOOPREPEAT, and the number of significant figures in the
2315    times.
2316    
2317    4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
2318    
2319    5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
2320    a building problem on Windows NT with a FAT file system.
2321    
2322    
2323    Version 2.01 21-Oct-98
2324    ----------------------
2325    
2326    1. Changed the API for pcre_compile() to allow for the provision of a pointer
2327    to character tables built by pcre_maketables() in the current locale. If NULL
2328    is passed, the default tables are used.
2329    
2330    
2331    Version 2.00 24-Sep-98
2332    ----------------------
2333    
2334    1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
2335    it any more.
2336    
2337    2. Allow quantification of (?>) groups, and make it work correctly.
2338    
2339    3. The first character computation wasn't working for (?>) groups.
2340    
2341    4. Correct the implementation of \Z (it is permitted to match on the \n at the
2342    end of the subject) and add 5.005's \z, which really does match only at the
2343    very end of the subject.
2344    
2345    5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
2346    
2347    6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
2348    DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
2349    localized options. All options to pcre_study() were also removed.
2350    
2351    7. Add other new features from 5.005:
2352    
2353       $(?<=           positive lookbehind
2354       $(?<!           negative lookbehind
2355       (?imsx-imsx)    added the unsetting capability
2356                       such a setting is global if at outer level; local otherwise
2357       (?imsx-imsx:)   non-capturing groups with option setting
2358       (?(cond)re|re)  conditional pattern matching
2359    
2360       A backreference to itself in a repeated group matches the previous
2361       captured string.
2362    
2363    8. General tidying up of studying (both automatic and via "study")
2364    consequential on the addition of new assertions.
2365    
2366    9. As in 5.005, unlimited repeated groups that could match an empty substring
2367    are no longer faulted at compile time. Instead, the loop is forcibly broken at
2368    runtime if any iteration does actually match an empty substring.
2369    
2370    10. Include the RunTest script in the distribution.
2371    
2372    11. Added tests from the Perl 5.005_02 distribution. This showed up a few
2373    discrepancies, some of which were old and were also with respect to 5.004. They
2374    have now been fixed.
2375    
2376    
2377    Version 1.09 28-Apr-98
2378  ----------------------  ----------------------
2379    
2380  1. A negated single character class followed by a quantifier with a minimum  1. A negated single character class followed by a quantifier with a minimum

Legend:
Removed from v.21  
changed lines
  Added in v.143

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12