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

Diff of /code/trunk/ChangeLog

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

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

Legend:
Removed from v.25  
changed lines
  Added in v.116

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12