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

Diff of /code/trunk/ChangeLog

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

revision 17 by nigel, Sat Feb 24 21:38:29 2007 UTC revision 167 by ph10, Wed May 9 15:53:54 2007 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4    Version 7.2 01-May-07
5    ---------------------
6    
7     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
8        which is apparently normally available under Windows.
9    
10     2. Re-jig the pcregrep tests with different newline settings in an attempt
11        to make them independent of the local environment's newline setting.
12    
13     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
14    
15     4. Some of the "internals" tests were previously cut out when the link size
16        was not 2, because the output contained actual offsets. The recent new
17        "Z" feature of pcretest means that these can be cut out, making the tests
18        usable with all link sizes.
19    
20     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
21        stack recursion. This gives a massive performance boost under BSD, but just
22        a small improvement under Linux. However, it saves one field in the frame
23        in all cases.
24    
25     6. Added more features from the forthcoming Perl 5.10:
26    
27        (a) (?-n) (where n is a string of digits) is a relative subroutine or
28            recursion call. It refers to the nth most recently opened parentheses.
29    
30        (b) (?+n) is also a relative subroutine call; it refers to the nth next
31            to be opened parentheses.
32    
33        (c) Conditions that refer to capturing parentheses can be specified
34            relatively, for example, (?(-2)... or (?(+3)...
35    
36    
37    Version 7.1 24-Apr-07
38    ---------------------
39    
40     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
41        that is more "standard", making use of automake and other Autotools. There
42        is some re-arrangement of the files and adjustment of comments consequent
43        on this.
44    
45     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
46        for recursive directory scanning broke on some systems because the files
47        are not scanned in any specific order and on different systems the order
48        was different. A call to "sort" has been inserted into RunGrepTest for the
49        approprate test as a short-term fix. In the longer term there may be an
50        alternative.
51    
52     3. I had an email from Eric Raymond about problems translating some of PCRE's
53        man pages to HTML (despite the fact that I distribute HTML pages, some
54        people do their own conversions for various reasons). The problems
55        concerned the use of low-level troff macros .br and .in. I have therefore
56        removed all such uses from the man pages (some were redundant, some could
57        be replaced by .nf/.fi pairs). The 132html script that I use to generate
58        HTML has been updated to handle .nf/.fi and to complain if it encounters
59        .br or .in.
60    
61     4. Updated comments in configure.ac that get placed in config.h.in and also
62        arranged for config.h to be included in the distribution, with the name
63        config.h.generic, for the benefit of those who have to compile without
64        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
65    
66     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
67        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
68        makevp.bat for the current PCRE, using the additional files
69        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
70    
71     6. A Windows user reported a minor discrepancy with test 2, which turned out
72        to be caused by a trailing space on an input line that had got lost in his
73        copy. The trailing space was an accident, so I've just removed it.
74    
75     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
76        that is needed.
77    
78     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
79        as "const" (a) because they are and (b) because it helps the PHP
80        maintainers who have recently made a script to detect big data structures
81        in the php code that should be moved to the .rodata section. I remembered
82        to update Builducptable as well, so it won't revert if ucptable.h is ever
83        re-created.
84    
85     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
86        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
87        order to be able to cut out the UTF-8 tables in the latter when UTF-8
88        support is not required. This saves 1.5-2K of code, which is important in
89        some applications.
90    
91        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
92        so as not to refer to the tables, even though these functions will never be
93        called when UTF-8 support is disabled. Otherwise there are problems with a
94        shared library.
95    
96    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
97    
98        (a) It was defining its arguments as char * instead of void *.
99    
100        (b) It was assuming that all moves were upwards in memory; this was true
101            a long time ago when I wrote it, but is no longer the case.
102    
103        The emulated memove() is provided for those environments that have neither
104        memmove() nor bcopy(). I didn't think anyone used it these days, but that
105        is clearly not the case, as these two bugs were recently reported.
106    
107    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
108        and Detrail to create the HTML documentation, the .txt form of the man
109        pages, and it removes trailing spaces from listed files. It also creates
110        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
111        case, it wraps all the #defines with #ifndefs. This script should be run
112        before "make dist".
113    
114    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
115        with Unicode property support.
116    
117        (a) For a maximizing quantifier, if the two different cases of the
118            character were of different lengths in their UTF-8 codings (there are
119            some cases like this - I found 11), and the matching function had to
120            back up over a mixture of the two cases, it incorrectly assumed they
121            were both the same length.
122    
123        (b) When PCRE was configured to use the heap rather than the stack for
124            recursion during matching, it was not correctly preserving the data for
125            the other case of a UTF-8 character when checking ahead for a match
126            while processing a minimizing repeat. If the check also involved
127            matching a wide character, but failed, corruption could cause an
128            erroneous result when trying to check for a repeat of the original
129            character.
130    
131    13. Some tidying changes to the testing mechanism:
132    
133        (a) The RunTest script now detects the internal link size and whether there
134            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
135            values substituted by "configure". (The RunGrepTest script already did
136            this for UTF-8.) The configure.ac script no longer substitutes the
137            relevant variables.
138    
139        (b) The debugging options /B and /D in pcretest show the compiled bytecode
140            with length and offset values. This means that the output is different
141            for different internal link sizes. Test 2 is skipped for link sizes
142            other than 2 because of this, bypassing the problem. Unfortunately,
143            there was also a test in test 3 (the locale tests) that used /B and
144            failed for link sizes other than 2. Rather than cut the whole test out,
145            I have added a new /Z option to pcretest that replaces the length and
146            offset values with spaces. This is now used to make test 3 independent
147            of link size. (Test 2 will be tidied up later.)
148    
149    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
150        segmentation fault instead of returning the appropriate error message.
151    
152    15. In multiline mode when the newline sequence was set to "any", the pattern
153        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
154        This doesn't seem right; it now treats the CRLF combination as the line
155        ending, and so does not match in that case. It's only a pattern such as ^$
156        that would hit this one: something like ^ABC$ would have failed after \r
157        and then tried again after \r\n.
158    
159    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
160        in an attempt to make files that differ only in their line terminators
161        compare equal. This works on Linux.
162    
163    17. Under certain error circumstances pcregrep might try to free random memory
164        as it exited. This is now fixed, thanks to valgrind.
165    
166    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
167        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
168        was because its rules for how to advance for /g after matching an empty
169        string at the end of a line did not allow for this case. They now check for
170        it specially.
171    
172    20. pcretest is supposed to handle patterns and data of any length, by
173        extending its buffers when necessary. It was getting this wrong when the
174        buffer for a data line had to be extended.
175    
176    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
177        CRLF as a newline sequence.
178    
179    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
180        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
181        I have nevertheless tidied it up.
182    
183    23. Added some casts to kill warnings from HP-UX ia64 compiler.
184    
185    24. Added a man page for pcre-config.
186    
187    
188    Version 7.0 19-Dec-06
189    ---------------------
190    
191     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
192        moving to gcc 4.1.1.
193    
194     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
195        sys/time.h, which is documented as needed for this function. It doesn't
196        seem to matter on Linux, but it showed up on some releases of OS X.
197    
198     3. It seems that there are systems where bytes whose values are greater than
199        127 match isprint() in the "C" locale. The "C" locale should be the
200        default when a C program starts up. In most systems, only ASCII printing
201        characters match isprint(). This difference caused the output from pcretest
202        to vary, making some of the tests fail. I have changed pcretest so that:
203    
204        (a) When it is outputting text in the compiled version of a pattern, bytes
205            other than 32-126 are always shown as hex escapes.
206    
207        (b) When it is outputting text that is a matched part of a subject string,
208            it does the same, unless a different locale has been set for the match
209            (using the /L modifier). In this case, it uses isprint() to decide.
210    
211     4. Fixed a major bug that caused incorrect computation of the amount of memory
212        required for a compiled pattern when options that changed within the
213        pattern affected the logic of the preliminary scan that determines the
214        length. The relevant options are -x, and -i in UTF-8 mode. The result was
215        that the computed length was too small. The symptoms of this bug were
216        either the PCRE error "internal error: code overflow" from pcre_compile(),
217        or a glibc crash with a message such as "pcretest: free(): invalid next
218        size (fast)". Examples of patterns that provoked this bug (shown in
219        pcretest format) are:
220    
221          /(?-x: )/x
222          /(?x)(?-x: \s*#\s*)/
223          /((?i)[\x{c0}])/8
224          /(?i:[\x{c0}])/8
225    
226        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
227        is now done differently.
228    
229     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
230        wrapper classes; (b) implement a new function in the C++ scanner that is
231        more efficient than the old way of doing things because it avoids levels of
232        recursion in the regex matching; (c) add a paragraph to the documentation
233        for the FullMatch() function.
234    
235     6. The escape sequence \n was being treated as whatever was defined as
236        "newline". Not only was this contrary to the documentation, which states
237        that \n is character 10 (hex 0A), but it also went horribly wrong when
238        "newline" was defined as CRLF. This has been fixed.
239    
240     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
241        was being set to -1 for the "end of line" case (supposedly a value that no
242        character can have). Though this value is never used (the check for end of
243        line is "zero bytes in current character"), it caused compiler complaints.
244        I've changed it to 0xffffffff.
245    
246     8. In pcre_version.c, the version string was being built by a sequence of
247        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
248        string (as it is for production releases) called a macro with an empty
249        argument. The C standard says the result of this is undefined. The gcc
250        compiler treats it as an empty string (which was what was wanted) but it is
251        reported that Visual C gives an error. The source has been hacked around to
252        avoid this problem.
253    
254     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
255        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
256        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
257        of them did).
258    
259    10. Originally, pcretest opened its input and output without "b"; then I was
260        told that "b" was needed in some environments, so it was added for release
261        5.0 to both the input and output. (It makes no difference on Unix-like
262        systems.) Later I was told that it is wrong for the input on Windows. I've
263        now abstracted the modes into two macros, to make it easier to fiddle with
264        them, and removed "b" from the input mode under Windows.
265    
266    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
267    
268    12. Added -help and --help to pcretest as an official way of being reminded
269        of the options.
270    
271    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
272        and pcrecpp.cc because they annoy compilers at high warning levels.
273    
274    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
275    
276    15. Fixed an occurrence of == in configure.ac that should have been = (shell
277        scripts are not C programs :-) and which was not noticed because it works
278        on Linux.
279    
280    16. pcretest is supposed to handle any length of pattern and data line (as one
281        line or as a continued sequence of lines) by extending its input buffer if
282        necessary. This feature was broken for very long pattern lines, leading to
283        a string of junk being passed to pcre_compile() if the pattern was longer
284        than about 50K.
285    
286    17. I have done a major re-factoring of the way pcre_compile() computes the
287        amount of memory needed for a compiled pattern. Previously, there was code
288        that made a preliminary scan of the pattern in order to do this. That was
289        OK when PCRE was new, but as the facilities have expanded, it has become
290        harder and harder to keep it in step with the real compile phase, and there
291        have been a number of bugs (see for example, 4 above). I have now found a
292        cunning way of running the real compile function in a "fake" mode that
293        enables it to compute how much memory it would need, while actually only
294        ever using a few hundred bytes of working memory and without too many
295        tests of the mode. This should make future maintenance and development
296        easier. A side effect of this work is that the limit of 200 on the nesting
297        depth of parentheses has been removed (though this was never a serious
298        limitation, I suspect). However, there is a downside: pcre_compile() now
299        runs more slowly than before (30% or more, depending on the pattern). I
300        hope this isn't a big issue. There is no effect on runtime performance.
301    
302    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
303        newline (only possible for the last line of a file) and it was a
304        pattern that set a locale (followed by /Lsomething), pcretest crashed.
305    
306    19. Added additional timing features to pcretest. (1) The -tm option now times
307        matching only, not compiling. (2) Both -t and -tm can be followed, as a
308        separate command line item, by a number that specifies the number of
309        repeats to use when timing. The default is 50000; this gives better
310        precision, but takes uncomfortably long for very large patterns.
311    
312    20. Extended pcre_study() to be more clever in cases where a branch of a
313        subpattern has no definite first character. For example, (a*|b*)[cd] would
314        previously give no result from pcre_study(). Now it recognizes that the
315        first character must be a, b, c, or d.
316    
317    21. There was an incorrect error "recursive call could loop indefinitely" if
318        a subpattern (or the entire pattern) that was being tested for matching an
319        empty string contained only one non-empty item after a nested subpattern.
320        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
321        incorrectly, because the \d was being skipped in the check.
322    
323    22. The pcretest program now has a new pattern option /B and a command line
324        option -b, which is equivalent to adding /B to every pattern. This causes
325        it to show the compiled bytecode, without the additional information that
326        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
327        is the same as /B/I).
328    
329    23. A new optimization is now able automatically to treat some sequences such
330        as a*b as a*+b. More specifically, if something simple (such as a character
331        or a simple class like \d) has an unlimited quantifier, and is followed by
332        something that cannot possibly match the quantified thing, the quantifier
333        is automatically "possessified".
334    
335    24. A recursive reference to a subpattern whose number was greater than 39
336        went wrong under certain circumstances in UTF-8 mode. This bug could also
337        have affected the operation of pcre_study().
338    
339    25. Realized that a little bit of performance could be had by replacing
340        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
341    
342    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
343    
344    27. Possessive quantifiers such as a++ were previously implemented by turning
345        them into atomic groups such as ($>a+). Now they have their own opcodes,
346        which improves performance. This includes the automatically created ones
347        from 23 above.
348    
349    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
350        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
351        the first matched character to be a colon. This applied both to named and
352        numbered groups.
353    
354    29. The ucpinternal.h header file was missing its idempotency #ifdef.
355    
356    30. I was sent a "project" file called libpcre.a.dev which I understand makes
357        building PCRE on Windows easier, so I have included it in the distribution.
358    
359    31. There is now a check in pcretest against a ridiculously large number being
360        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
361        loop, the loop is abandoned.
362    
363    32. Forward references to subpatterns in conditions such as (?(2)...) where
364        subpattern 2 is defined later cause pcre_compile() to search forwards in
365        the pattern for the relevant set of parentheses. This search went wrong
366        when there were unescaped parentheses in a character class, parentheses
367        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
368    
369    33. "Subroutine" calls and backreferences were previously restricted to
370        referencing subpatterns earlier in the regex. This restriction has now
371        been removed.
372    
373    34. Added a number of extra features that are going to be in Perl 5.10. On the
374        whole, these are just syntactic alternatives for features that PCRE had
375        previously implemented using the Python syntax or my own invention. The
376        other formats are all retained for compatibility.
377    
378        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
379            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
380            also .NET compatible.
381    
382        (b) A recursion or subroutine call to a named group can now be defined as
383            (?&name) as well as (?P>name).
384    
385        (c) A backreference to a named group can now be defined as \k<name> or
386            \k'name' as well as (?P=name). The new forms, as well as being in Perl
387            5.10, are also .NET compatible.
388    
389        (d) A conditional reference to a named group can now use the syntax
390            (?(<name>) or (?('name') as well as (?(name).
391    
392        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
393            groups (named and numbered) that are never evaluated inline, but can be
394            called as "subroutines" from elsewhere. In effect, the DEFINE condition
395            is always false. There may be only one alternative in such a group.
396    
397        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
398            as the simple (?(R). The condition is true only if the most recent
399            recursion is that of the given number or name. It does not search out
400            through the entire recursion stack.
401    
402        (g) The escape \gN or \g{N} has been added, where N is a positive or
403            negative number, specifying an absolute or relative reference.
404    
405    35. Tidied to get rid of some further signed/unsigned compiler warnings and
406        some "unreachable code" warnings.
407    
408    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
409        things, this adds five new scripts.
410    
411    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
412        There were also incompatibilities regarding the handling of \Q..\E inside
413        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
414        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
415    
416    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
417        matches an empty string, and forcibly breaks the loop. There were bugs in
418        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
419        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
420        separate and independent bugs (that affected different cases) have been
421        fixed.
422    
423    39. Refactored the code to abolish the use of different opcodes for small
424        capturing bracket numbers. This is a tidy that I avoided doing when I
425        removed the limit on the number of capturing brackets for 3.5 back in 2001.
426        The new approach is not only tidier, it makes it possible to reduce the
427        memory needed to fix the previous bug (38).
428    
429    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
430        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
431        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
432        mode.
433    
434    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
435        report.
436    
437    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
438        copy construction and assignment in the C++ wrapper.
439    
440    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
441        couple of bugs that could have given wrong results in the "--newline=crlf"
442        case.
443    
444    44. Added a number of casts and did some reorganization of signed/unsigned int
445        variables following suggestions from Dair Grant. Also renamed the variable
446        "this" as "item" because it is a C++ keyword.
447    
448    45. Arranged for dftables to add
449    
450          #include "pcre_internal.h"
451    
452        to pcre_chartables.c because without it, gcc 4.x may remove the array
453        definition from the final binary if PCRE is built into a static library and
454        dead code stripping is activated.
455    
456    46. For an unanchored pattern, if a match attempt fails at the start of a
457        newline sequence, and the newline setting is CRLF or ANY, and the next two
458        characters are CRLF, advance by two characters instead of one.
459    
460    
461    Version 6.7 04-Jul-06
462    ---------------------
463    
464     1. In order to handle tests when input lines are enormously long, pcretest has
465        been re-factored so that it automatically extends its buffers when
466        necessary. The code is crude, but this _is_ just a test program. The
467        default size has been increased from 32K to 50K.
468    
469     2. The code in pcre_study() was using the value of the re argument before
470        testing it for NULL. (Of course, in any sensible call of the function, it
471        won't be NULL.)
472    
473     3. The memmove() emulation function in pcre_internal.h, which is used on
474        systems that lack both memmove() and bcopy() - that is, hardly ever -
475        was missing a "static" storage class specifier.
476    
477     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
478        containing an extended class (one that cannot be represented by a bitmap
479        because it contains high-valued characters or Unicode property items, e.g.
480        [\pZ]). Almost always one would set UTF-8 mode when processing such a
481        pattern, but PCRE should not loop if you do not (it no longer does).
482        [Detail: two cases were found: (a) a repeated subpattern containing an
483        extended class; (b) a recursive reference to a subpattern that followed a
484        previous extended class. It wasn't skipping over the extended class
485        correctly when UTF-8 mode was not set.]
486    
487     5. A negated single-character class was not being recognized as fixed-length
488        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
489        compile error "lookbehind assertion is not fixed length".
490    
491     6. The RunPerlTest auxiliary script was showing an unexpected difference
492        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
493        write a Perl script that can interpret lines of an input file either as
494        byte characters or as UTF-8, which is what "perltest" was being required to
495        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
496        can't do is switch easily at run time between having the "use utf8;" pragma
497        or not. In the end, I fudged it by using the RunPerlTest script to insert
498        "use utf8;" explicitly for the UTF-8 tests.
499    
500     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
501        the end of the subject string, contrary to the documentation and to what
502        Perl does. This was true of both matching functions. Now it matches only at
503        the start of the subject and immediately after *internal* newlines.
504    
505     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
506        a pointer to an int instead of a pointer to an unsigned long int. This
507        caused problems on 64-bit systems.
508    
509     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
510        instance of the 'standard' template library not being so standard".
511    
512    10. There was no check on the number of named subpatterns nor the maximum
513        length of a subpattern name. The product of these values is used to compute
514        the size of the memory block for a compiled pattern. By supplying a very
515        long subpattern name and a large number of named subpatterns, the size
516        computation could be caused to overflow. This is now prevented by limiting
517        the length of names to 32 characters, and the number of named subpatterns
518        to 10,000.
519    
520    11. Subpatterns that are repeated with specific counts have to be replicated in
521        the compiled pattern. The size of memory for this was computed from the
522        length of the subpattern and the repeat count. The latter is limited to
523        65535, but there was no limit on the former, meaning that integer overflow
524        could in principle occur. The compiled length of a repeated subpattern is
525        now limited to 30,000 bytes in order to prevent this.
526    
527    12. Added the optional facility to have named substrings with the same name.
528    
529    13. Added the ability to use a named substring as a condition, using the
530        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
531        are numbers (not recommended). Forward references are permitted.
532    
533    14. Added forward references in named backreferences (if you see what I mean).
534    
535    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
536        pattern could run off the end of the subject. For example, the pattern
537        "(?s)(.{1,5})"8 did this with the subject "ab".
538    
539    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
540        PCRE_CASELESS was set when matching characters that were quantified with ?
541        or *.
542    
543    17. A character class other than a single negated character that had a minimum
544        but no maximum quantifier - for example [ab]{6,} - was not handled
545        correctly by pce_dfa_exec(). It would match only one character.
546    
547    18. A valid (though odd) pattern that looked like a POSIX character
548        class but used an invalid character after [ (for example [[,abc,]]) caused
549        pcre_compile() to give the error "Failed: internal error: code overflow" or
550        in some cases to crash with a glibc free() error. This could even happen if
551        the pattern terminated after [[ but there just happened to be a sequence of
552        letters, a binary zero, and a closing ] in the memory that followed.
553    
554    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
555        over the years. Originally (before any Unicode support), just the bottom 8
556        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
557        output from "man perlunicode" includes this:
558    
559          The regular expression compiler produces polymorphic opcodes.  That
560          is, the pattern adapts to the data and automatically switches to
561          the Unicode character scheme when presented with Unicode data--or
562          instead uses a traditional byte scheme when presented with byte
563          data.
564    
565        Sadly, a wide octal escape does not cause a switch, and in a string with
566        no other multibyte characters, these octal escapes are treated as before.
567        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
568        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
569        Unicode string.
570    
571        I have not perpetrated such confusion in PCRE. Up till now, it took just
572        the bottom 8 bits, as in old Perl. I have now made octal escapes with
573        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
574        translate to the appropriate multibyte character.
575    
576    29. Applied some refactoring to reduce the number of warnings from Microsoft
577        and Borland compilers. This has included removing the fudge introduced
578        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
579        a warning about an unused variable.
580    
581    21. PCRE has not included VT (character 0x0b) in the set of whitespace
582        characters since release 4.0, because Perl (from release 5.004) does not.
583        [Or at least, is documented not to: some releases seem to be in conflict
584        with the documentation.] However, when a pattern was studied with
585        pcre_study() and all its branches started with \s, PCRE still included VT
586        as a possible starting character. Of course, this did no harm; it just
587        caused an unnecessary match attempt.
588    
589    22. Removed a now-redundant internal flag bit that recorded the fact that case
590        dependency changed within the pattern. This was once needed for "required
591        byte" processing, but is no longer used. This recovers a now-scarce options
592        bit. Also moved the least significant internal flag bit to the most-
593        significant bit of the word, which was not previously used (hangover from
594        the days when it was an int rather than a uint) to free up another bit for
595        the future.
596    
597    23. Added support for CRLF line endings as well as CR and LF. As well as the
598        default being selectable at build time, it can now be changed at runtime
599        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
600        specify that it is scanning data with non-default line endings.
601    
602    24. Changed the definition of CXXLINK to make it agree with the definition of
603        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
604    
605    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
606        recursions. This makes a big different to stack usage for some patterns.
607    
608    26. If a subpattern containing a named recursion or subroutine reference such
609        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
610        the space required for the compiled pattern went wrong and gave too small a
611        value. Depending on the environment, this could lead to "Failed: internal
612        error: code overflow at offset 49" or "glibc detected double free or
613        corruption" errors.
614    
615    27. Applied patches from Google (a) to support the new newline modes and (b) to
616        advance over multibyte UTF-8 characters in GlobalReplace.
617    
618    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
619        difference for some implementation of PCRE in some Windows version.
620    
621    29. Added some extra testing facilities to pcretest:
622    
623        \q<number>   in a data line sets the "match limit" value
624        \Q<number>   in a data line sets the "match recursion limt" value
625        -S <number>  sets the stack size, where <number> is in megabytes
626    
627        The -S option isn't available for Windows.
628    
629    
630    Version 6.6 06-Feb-06
631    ---------------------
632    
633     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
634        in pcreposix.h. I have copied the definition from pcre.h.
635    
636     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
637        because pcre.h is no longer a built file.
638    
639     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
640        not normally included in the compiled code.
641    
642    
643    Version 6.5 01-Feb-06
644    ---------------------
645    
646     1. When using the partial match feature with pcre_dfa_exec(), it was not
647        anchoring the second and subsequent partial matches at the new starting
648        point. This could lead to incorrect results. For example, with the pattern
649        /1234/, partially matching against "123" and then "a4" gave a match.
650    
651     2. Changes to pcregrep:
652    
653        (a) All non-match returns from pcre_exec() were being treated as failures
654            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
655            error message is output. Some extra information is given for the
656            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
657            probably the only errors that are likely to be caused by users (by
658            specifying a regex that has nested indefinite repeats, for instance).
659            If there are more than 20 of these errors, pcregrep is abandoned.
660    
661        (b) A binary zero was treated as data while matching, but terminated the
662            output line if it was written out. This has been fixed: binary zeroes
663            are now no different to any other data bytes.
664    
665        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
666            used to set a locale for matching. The --locale=xxxx long option has
667            been added (no short equivalent) to specify a locale explicitly on the
668            pcregrep command, overriding the environment variables.
669    
670        (d) When -B was used with -n, some line numbers in the output were one less
671            than they should have been.
672    
673        (e) Added the -o (--only-matching) option.
674    
675        (f) If -A or -C was used with -c (count only), some lines of context were
676            accidentally printed for the final match.
677    
678        (g) Added the -H (--with-filename) option.
679    
680        (h) The combination of options -rh failed to suppress file names for files
681            that were found from directory arguments.
682    
683        (i) Added the -D (--devices) and -d (--directories) options.
684    
685        (j) Added the -F (--fixed-strings) option.
686    
687        (k) Allow "-" to be used as a file name for -f as well as for a data file.
688    
689        (l) Added the --colo(u)r option.
690    
691        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
692            is not present by default.
693    
694     3. A nasty bug was discovered in the handling of recursive patterns, that is,
695        items such as (?R) or (?1), when the recursion could match a number of
696        alternatives. If it matched one of the alternatives, but subsequently,
697        outside the recursion, there was a failure, the code tried to back up into
698        the recursion. However, because of the way PCRE is implemented, this is not
699        possible, and the result was an incorrect result from the match.
700    
701        In order to prevent this happening, the specification of recursion has
702        been changed so that all such subpatterns are automatically treated as
703        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
704    
705     4. I had overlooked the fact that, in some locales, there are characters for
706        which isalpha() is true but neither isupper() nor islower() are true. In
707        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
708        and ordfeminine) are like this. This affected the treatment of \w and \W
709        when they appeared in character classes, but not when they appeared outside
710        a character class. The bit map for "word" characters is now created
711        separately from the results of isalnum() instead of just taking it from the
712        upper, lower, and digit maps. (Plus the underscore character, of course.)
713    
714     5. The above bug also affected the handling of POSIX character classes such as
715        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
716        permanent tables. Instead, the bit maps for such a class were previously
717        created as the appropriate unions of the upper, lower, and digit bitmaps.
718        Now they are created by subtraction from the [[:word:]] class, which has
719        its own bitmap.
720    
721     6. The [[:blank:]] character class matches horizontal, but not vertical space.
722        It is created by subtracting the vertical space characters (\x09, \x0a,
723        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
724        subtraction was done in the overall bitmap for a character class, meaning
725        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
726        be recognized. This bug has been fixed.
727    
728     7. Patches from the folks at Google:
729    
730          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
731          real life, but is still worth protecting against".
732    
733          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
734          regular expressions".
735    
736          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
737          have it.
738    
739          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
740          "configure" and the latter not, in order to fix a problem somebody had
741          with compiling the Arg class on HP-UX.
742    
743          (e) Improve the error-handling of the C++ wrapper a little bit.
744    
745          (f) New tests for checking recursion limiting.
746    
747     8. The pcre_memmove() function, which is used only if the environment does not
748        have a standard memmove() function (and is therefore rarely compiled),
749        contained two bugs: (a) use of int instead of size_t, and (b) it was not
750        returning a result (though PCRE never actually uses the result).
751    
752     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
753        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
754        returned instead of calling malloc() with an overflowing number that would
755        most likely cause subsequent chaos.
756    
757    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
758    
759    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
760        with this option is matched, the nmatch and pmatch options of regexec() are
761        ignored.
762    
763    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
764        provided in case anyone wants to the the POSIX interface with UTF-8
765        strings.
766    
767    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
768        C++ linking (needed for some HP-UX environments).
769    
770    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
771        (unused parameter) and in the pcre_printint() function (omitted "default"
772        switch label when the default is to do nothing).
773    
774    15. Added some code to make it possible, when PCRE is compiled as a C++
775        library, to replace subject pointers for pcre_exec() with a smart pointer
776        class, thus making it possible to process discontinuous strings.
777    
778    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
779        much the same function. They were added by different people who were trying
780        to make PCRE easy to compile on non-Unix systems. It has been suggested
781        that PCRE_EXPORT be abolished now that there is more automatic apparatus
782        for compiling on Windows systems. I have therefore replaced it with
783        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
784        defaults to "extern" for C or "extern C" for C++, which works fine on
785        Unix-like systems. It is now possible to override the value of PCRE_DATA_
786        SCOPE with something explicit in config.h. In addition:
787    
788        (a) pcreposix.h still had just "extern" instead of either of these macros;
789            I have replaced it with PCRE_DATA_SCOPE.
790    
791        (b) Functions such as _pcre_xclass(), which are internal to the library,
792            but external in the C sense, all had PCRE_EXPORT in their definitions.
793            This is apparently wrong for the Windows case, so I have removed it.
794            (It makes no difference on Unix-like systems.)
795    
796    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
797        of recursive calls to match(). This is different to MATCH_LIMIT because
798        that limits the total number of calls to match(), not all of which increase
799        the depth of recursion. Limiting the recursion depth limits the amount of
800        stack (or heap if NO_RECURSE is set) that is used. The default can be set
801        when PCRE is compiled, and changed at run time. A patch from Google adds
802        this functionality to the C++ interface.
803    
804    18. Changes to the handling of Unicode character properties:
805    
806        (a) Updated the table to Unicode 4.1.0.
807    
808        (b) Recognize characters that are not in the table as "Cn" (undefined).
809    
810        (c) I revised the way the table is implemented to a much improved format
811            which includes recognition of ranges. It now supports the ranges that
812            are defined in UnicodeData.txt, and it also amalgamates other
813            characters into ranges. This has reduced the number of entries in the
814            table from around 16,000 to around 3,000, thus reducing its size
815            considerably. I realized I did not need to use a tree structure after
816            all - a binary chop search is just as efficient. Having reduced the
817            number of entries, I extended their size from 6 bytes to 8 bytes to
818            allow for more data.
819    
820        (d) Added support for Unicode script names via properties such as \p{Han}.
821    
822    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
823        matching that character.
824    
825    20. When matching a repeated Unicode property with a minimum greater than zero,
826        (for example \pL{2,}), PCRE could look past the end of the subject if it
827        reached it while seeking the minimum number of characters. This could
828        happen only if some of the characters were more than one byte long, because
829        there is a check for at least the minimum number of bytes.
830    
831    21. Refactored the implementation of \p and \P so as to be more general, to
832        allow for more different types of property in future. This has changed the
833        compiled form incompatibly. Anybody with saved compiled patterns that use
834        \p or \P will have to recompile them.
835    
836    22. Added "Any" and "L&" to the supported property types.
837    
838    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
839        but give a compile time error if the value is greater than 0xff.
840    
841    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
842        accidentally not being installed or uninstalled.
843    
844    25. The pcre.h file was built from pcre.h.in, but the only changes that were
845        made were to insert the current release number. This seemed silly, because
846        it made things harder for people building PCRE on systems that don't run
847        "configure". I have turned pcre.h into a distributed file, no longer built
848        by "configure", with the version identification directly included. There is
849        no longer a pcre.h.in file.
850    
851        However, this change necessitated a change to the pcre-config script as
852        well. It is built from pcre-config.in, and one of the substitutions was the
853        release number. I have updated configure.ac so that ./configure now finds
854        the release number by grepping pcre.h.
855    
856    26. Added the ability to run the tests under valgrind.
857    
858    
859    Version 6.4 05-Sep-05
860    ---------------------
861    
862     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
863        "--" to be printed when multiple files were scanned, even when none of the
864        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
865        consider it to be a bug, and have restored the previous behaviour.
866    
867     2. A couple of code tidies to get rid of compiler warnings.
868    
869     3. The pcretest program used to cheat by referring to symbols in the library
870        whose names begin with _pcre_. These are internal symbols that are not
871        really supposed to be visible externally, and in some environments it is
872        possible to suppress them. The cheating is now confined to including
873        certain files from the library's source, which is a bit cleaner.
874    
875     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
876        file's purpose clearer.
877    
878     5. Reorganized pcre_ucp_findchar().
879    
880    
881    Version 6.3 15-Aug-05
882    ---------------------
883    
884     1. The file libpcre.pc.in did not have general read permission in the tarball.
885    
886     2. There were some problems when building without C++ support:
887    
888        (a) If C++ support was not built, "make install" and "make test" still
889            tried to test it.
890    
891        (b) There were problems when the value of CXX was explicitly set. Some
892            changes have been made to try to fix these, and ...
893    
894        (c) --disable-cpp can now be used to explicitly disable C++ support.
895    
896        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
897            backslash in a target when C++ was disabled. This confuses some
898            versions of "make", apparently. Using an intermediate variable solves
899            this. (Same for CPP_LOBJ.)
900    
901     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
902        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
903        necessary on certain architectures.
904    
905     4. Added a setting of -export-symbols-regex to the link command to remove
906        those symbols that are exported in the C sense, but actually are local
907        within the library, and not documented. Their names all begin with
908        "_pcre_". This is not a perfect job, because (a) we have to except some
909        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
910        available (and never for static libraries). I have made a note to try to
911        find a way round (a) in the future.
912    
913    
914    Version 6.2 01-Aug-05
915    ---------------------
916    
917     1. There was no test for integer overflow of quantifier values. A construction
918        such as {1111111111111111} would give undefined results. What is worse, if
919        a minimum quantifier for a parenthesized subpattern overflowed and became
920        negative, the calculation of the memory size went wrong. This could have
921        led to memory overwriting.
922    
923     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
924    
925     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
926        operating environments where this matters.
927    
928     4. Applied Giuseppe Maxia's patch to add additional features for controlling
929        PCRE options from within the C++ wrapper.
930    
931     5. Named capturing subpatterns were not being correctly counted when a pattern
932        was compiled. This caused two problems: (a) If there were more than 100
933        such subpatterns, the calculation of the memory needed for the whole
934        compiled pattern went wrong, leading to an overflow error. (b) Numerical
935        back references of the form \12, where the number was greater than 9, were
936        not recognized as back references, even though there were sufficient
937        previous subpatterns.
938    
939     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
940        versions of gcc, e.g. 2.95.4.
941    
942    
943    Version 6.1 21-Jun-05
944    ---------------------
945    
946     1. There was one reference to the variable "posix" in pcretest.c that was not
947        surrounded by "#if !defined NOPOSIX".
948    
949     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
950        the cross-check on the old pcre_info() function, for the benefit of the
951        cut-down version of PCRE that is currently imported into Exim.
952    
953     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
954        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
955        patterns that start (?i) but I don't think that matters. The use of (?i) is
956        just an example; this all applies to the other options as well.
957    
958     4. Since libtool seems to echo the compile commands it is issuing, the output
959        from "make" can be reduced a bit by putting "@" in front of each libtool
960        compile command.
961    
962     5. Patch from the folks at Google for configure.in to be a bit more thorough
963        in checking for a suitable C++ installation before trying to compile the
964        C++ stuff. This should fix a reported problem when a compiler was present,
965        but no suitable headers.
966    
967     6. The man pages all had just "PCRE" as their title. I have changed them to
968        be the relevant file name. I have also arranged that these names are
969        retained in the file doc/pcre.txt, which is a concatenation in text format
970        of all the man pages except the little individual ones for each function.
971    
972     7. The NON-UNIX-USE file had not been updated for the different set of source
973        files that come with release 6. I also added a few comments about the C++
974        wrapper.
975    
976    
977    Version 6.0 07-Jun-05
978    ---------------------
979    
980     1. Some minor internal re-organization to help with my DFA experiments.
981    
982     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
983        didn't matter for the library itself when fully configured, but did matter
984        when compiling without UCP support, or within Exim, where the ucp files are
985        not imported.
986    
987     3. Refactoring of the library code to split up the various functions into
988        different source modules. The addition of the new DFA matching code (see
989        below) to a single monolithic source would have made it really too
990        unwieldy, quite apart from causing all the code to be include in a
991        statically linked application, when only some functions are used. This is
992        relevant even without the DFA addition now that patterns can be compiled in
993        one application and matched in another.
994    
995        The downside of splitting up is that there have to be some external
996        functions and data tables that are used internally in different modules of
997        the library but which are not part of the API. These have all had their
998        names changed to start with "_pcre_" so that they are unlikely to clash
999        with other external names.
1000    
1001     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1002        a different (DFA) algorithm. Although it is slower than the original
1003        function, it does have some advantages for certain types of matching
1004        problem.
1005    
1006     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1007        including restarting after a partial match.
1008    
1009     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1010        defined when compiling for Windows was sent to me. I have put it into the
1011        code, though I have no means of testing or verifying it.
1012    
1013     7. Added the pcre_refcount() auxiliary function.
1014    
1015     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1016        match before or at the first newline in the subject string. In pcretest,
1017        the /f option on a pattern can be used to set this.
1018    
1019     9. A repeated \w when used in UTF-8 mode with characters greater than 256
1020        would behave wrongly. This has been present in PCRE since release 4.0.
1021    
1022    10. A number of changes to the pcregrep command:
1023    
1024        (a) Refactored how -x works; insert ^(...)$ instead of setting
1025            PCRE_ANCHORED and checking the length, in preparation for adding
1026            something similar for -w.
1027    
1028        (b) Added the -w (match as a word) option.
1029    
1030        (c) Refactored the way lines are read and buffered so as to have more
1031            than one at a time available.
1032    
1033        (d) Implemented a pcregrep test script.
1034    
1035        (e) Added the -M (multiline match) option. This allows patterns to match
1036            over several lines of the subject. The buffering ensures that at least
1037            8K, or the rest of the document (whichever is the shorter) is available
1038            for matching (and similarly the previous 8K for lookbehind assertions).
1039    
1040        (f) Changed the --help output so that it now says
1041    
1042              -w, --word-regex(p)
1043    
1044            instead of two lines, one with "regex" and the other with "regexp"
1045            because that confused at least one person since the short forms are the
1046            same. (This required a bit of code, as the output is generated
1047            automatically from a table. It wasn't just a text change.)
1048    
1049        (g) -- can be used to terminate pcregrep options if the next thing isn't an
1050            option but starts with a hyphen. Could be a pattern or a path name
1051            starting with a hyphen, for instance.
1052    
1053        (h) "-" can be given as a file name to represent stdin.
1054    
1055        (i) When file names are being printed, "(standard input)" is used for
1056            the standard input, for compatibility with GNU grep. Previously
1057            "<stdin>" was used.
1058    
1059        (j) The option --label=xxx can be used to supply a name to be used for
1060            stdin when file names are being printed. There is no short form.
1061    
1062        (k) Re-factored the options decoding logic because we are going to add
1063            two more options that take data. Such options can now be given in four
1064            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1065    
1066        (l) Added the -A, -B, and -C options for requesting that lines of context
1067            around matches be printed.
1068    
1069        (m) Added the -L option to print the names of files that do not contain
1070            any matching lines, that is, the complement of -l.
1071    
1072        (n) The return code is 2 if any file cannot be opened, but pcregrep does
1073            continue to scan other files.
1074    
1075        (o) The -s option was incorrectly implemented. For compatibility with other
1076            greps, it now suppresses the error message for a non-existent or non-
1077            accessible file (but not the return code). There is a new option called
1078            -q that suppresses the output of matching lines, which was what -s was
1079            previously doing.
1080    
1081        (p) Added --include and --exclude options to specify files for inclusion
1082            and exclusion when recursing.
1083    
1084    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1085        Hopefully, it now does.
1086    
1087    12. Missing cast in pcre_study().
1088    
1089    13. Added an "uninstall" target to the makefile.
1090    
1091    14. Replaced "extern" in the function prototypes in Makefile.in with
1092        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1093        world, but is set differently for Windows.
1094    
1095    15. Added a second compiling function called pcre_compile2(). The only
1096        difference is that it has an extra argument, which is a pointer to an
1097        integer error code. When there is a compile-time failure, this is set
1098        non-zero, in addition to the error test pointer being set to point to an
1099        error message. The new argument may be NULL if no error number is required
1100        (but then you may as well call pcre_compile(), which is now just a
1101        wrapper). This facility is provided because some applications need a
1102        numeric error indication, but it has also enabled me to tidy up the way
1103        compile-time errors are handled in the POSIX wrapper.
1104    
1105    16. Added VPATH=.libs to the makefile; this should help when building with one
1106        prefix path and installing with another. (Or so I'm told by someone who
1107        knows more about this stuff than I do.)
1108    
1109    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1110        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1111        match everything, including newlines. This is not POSIX-compatible, but
1112        somebody wanted the feature. From pcretest it can be activated by using
1113        both the P and the s flags.
1114    
1115    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1116    
1117    19. libpcre.pc was being incorrectly installed as executable.
1118    
1119    20. A couple of places in pcretest check for end-of-line by looking for '\n';
1120        it now also looks for '\r' so that it will work unmodified on Windows.
1121    
1122    21. Added Google's contributed C++ wrapper to the distribution.
1123    
1124    22. Added some untidy missing memory free() calls in pcretest, to keep
1125        Electric Fence happy when testing.
1126    
1127    
1128    
1129    Version 5.0 13-Sep-04
1130    ---------------------
1131    
1132     1. Internal change: literal characters are no longer packed up into items
1133        containing multiple characters in a single byte-string. Each character
1134        is now matched using a separate opcode. However, there may be more than one
1135        byte in the character in UTF-8 mode.
1136    
1137     2. The pcre_callout_block structure has two new fields: pattern_position and
1138        next_item_length. These contain the offset in the pattern to the next match
1139        item, and its length, respectively.
1140    
1141     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1142        insertion of callouts before each pattern item. Added the /C option to
1143        pcretest to make use of this.
1144    
1145     4. On the advice of a Windows user, the lines
1146    
1147          #if defined(_WIN32) || defined(WIN32)
1148          _setmode( _fileno( stdout ), 0x8000 );
1149          #endif  /* defined(_WIN32) || defined(WIN32) */
1150    
1151        have been added to the source of pcretest. This apparently does useful
1152        magic in relation to line terminators.
1153    
1154     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1155        for the benefit of those environments where the "b" makes a difference.
1156    
1157     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1158        to know about it. I have put a hack into configure.in that adds in code
1159        to set GCC=yes if CC=icc. This seems to end up at a point in the
1160        generated configure script that is early enough to affect the setting of
1161        compiler options, which is what is needed, but I have no means of testing
1162        whether it really works. (The user who reported this had patched the
1163        generated configure script, which of course I cannot do.)
1164    
1165        LATER: After change 22 below (new libtool files), the configure script
1166        seems to know about icc (and also ecc). Therefore, I have commented out
1167        this hack in configure.in.
1168    
1169     7. Added support for pkg-config (2 patches were sent in).
1170    
1171     8. Negated POSIX character classes that used a combination of internal tables
1172        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1173        [[:^ascii]]. Typically, they would match almost any characters. The other
1174        POSIX classes were not broken in this way.
1175    
1176     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1177        to find the match, as PCRE was deluded into thinking that the match had to
1178        start at the start point or following a newline. The same bug applied to
1179        patterns with negative forward assertions or any backward assertions
1180        preceding ".*" at the start, unless the pattern required a fixed first
1181        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1182    
1183    10. In UTF-8 mode, when moving forwards in the subject after a failed match
1184        starting at the last subject character, bytes beyond the end of the subject
1185        string were read.
1186    
1187    11. Renamed the variable "class" as "classbits" to make life easier for C++
1188        users. (Previously there was a macro definition, but it apparently wasn't
1189        enough.)
1190    
1191    12. Added the new field "tables" to the extra data so that tables can be passed
1192        in at exec time, or the internal tables can be re-selected. This allows
1193        a compiled regex to be saved and re-used at a later time by a different
1194        program that might have everything at different addresses.
1195    
1196    13. Modified the pcre-config script so that, when run on Solaris, it shows a
1197        -R library as well as a -L library.
1198    
1199    14. The debugging options of pcretest (-d on the command line or D on a
1200        pattern) showed incorrect output for anything following an extended class
1201        that contained multibyte characters and which was followed by a quantifier.
1202    
1203    15. Added optional support for general category Unicode character properties
1204        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1205        support. It adds about 90K to the size of the library. The meanings of the
1206        inbuilt class escapes such as \d and \s have NOT been changed.
1207    
1208    16. Updated pcredemo.c to include calls to free() to release the memory for the
1209        compiled pattern.
1210    
1211    17. The generated file chartables.c was being created in the source directory
1212        instead of in the building directory. This caused the build to fail if the
1213        source directory was different from the building directory, and was
1214        read-only.
1215    
1216    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1217        file. No doubt somebody will tell me if they don't make sense... Also added
1218        Dan Mooney's comments about building on OpenVMS.
1219    
1220    19. Added support for partial matching via the PCRE_PARTIAL option for
1221        pcre_exec() and the \P data escape in pcretest.
1222    
1223    20. Extended pcretest with 3 new pattern features:
1224    
1225        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
1226              write the compiled pattern to the file whose name is "rest-of-line".
1227              This is a straight binary dump of the data, with the saved pointer to
1228              the character tables forced to be NULL. The study data, if any, is
1229              written too. After writing, pcretest reads a new pattern.
1230    
1231        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1232              compiled pattern from the given file. There must not be any
1233              occurrences of "<" in the file name (pretty unlikely); if there are,
1234              pcretest will instead treat the initial "<" as a pattern delimiter.
1235              After reading in the pattern, pcretest goes on to read data lines as
1236              usual.
1237    
1238        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1239              and 16-bit fields in a compiled pattern, to simulate a pattern that
1240              was compiled on a host of opposite endianness.
1241    
1242    21. The pcre-exec() function can now cope with patterns that were compiled on
1243        hosts of opposite endianness, with this restriction:
1244    
1245          As for any compiled expression that is saved and used later, the tables
1246          pointer field cannot be preserved; the extra_data field in the arguments
1247          to pcre_exec() should be used to pass in a tables address if a value
1248          other than the default internal tables were used at compile time.
1249    
1250    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1251        now diagnosed as an error. Previously, most of the time, a negative number
1252        would have been treated as zero, but if in addition "ovector" was passed as
1253        NULL, a crash could occur.
1254    
1255    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1256        new versions from the libtool 1.5 distribution (the last one is a copy of
1257        a file called libtool.m4). This seems to have fixed the need to patch
1258        "configure" to support Darwin 1.3 (which I used to do). However, I still
1259        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1260        workstation).
1261    
1262    24. Changed the PCRE licence to be the more standard "BSD" licence.
1263    
1264    
1265    Version 4.5 01-Dec-03
1266    ---------------------
1267    
1268     1. There has been some re-arrangement of the code for the match() function so
1269        that it can be compiled in a version that does not call itself recursively.
1270        Instead, it keeps those local variables that need separate instances for
1271        each "recursion" in a frame on the heap, and gets/frees frames whenever it
1272        needs to "recurse". Keeping track of where control must go is done by means
1273        of setjmp/longjmp. The whole thing is implemented by a set of macros that
1274        hide most of the details from the main code, and operates only if
1275        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1276        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1277        operating.
1278    
1279        To make it easier for callers to provide specially tailored get/free
1280        functions for this usage, two new functions, pcre_stack_malloc, and
1281        pcre_stack_free, are used. They are always called in strict stacking order,
1282        and the size of block requested is always the same.
1283    
1284        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1285        PCRE has been compiled to use the stack or the heap for recursion. The
1286        -C option of pcretest uses this to show which version is compiled.
1287    
1288        A new data escape \S, is added to pcretest; it causes the amounts of store
1289        obtained and freed by both kinds of malloc/free at match time to be added
1290        to the output.
1291    
1292     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1293        what's available on my current Linux desktop machine.
1294    
1295     3. When matching a UTF-8 string, the test for a valid string at the start has
1296        been extended. If start_offset is not zero, PCRE now checks that it points
1297        to a byte that is the start of a UTF-8 character. If not, it returns
1298        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1299        this is necessary because there may be backward assertions in the pattern.
1300        When matching the same subject several times, it may save resources to use
1301        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1302    
1303     4. The code for checking the validity of UTF-8 strings has been tightened so
1304        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1305        containing "overlong sequences".
1306    
1307     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1308        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1309        should have been "&", but it just so happened that all the cases this let
1310        through by mistake were picked up later in the function.
1311    
1312     6. I had used a variable called "isblank" - this is a C99 function, causing
1313        some compilers to warn. To avoid this, I renamed it (as "blankclass").
1314    
1315     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1316        prompting; (b) run "./pcretest /dev/null" at the start of the test script
1317        so the version is shown; (c) stop "make test" echoing "./RunTest".
1318    
1319     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1320    
1321     9. The prototype for memmove() for systems that don't have it was using
1322        size_t, but the inclusion of the header that defines size_t was later. I've
1323        moved the #includes for the C headers earlier to avoid this.
1324    
1325    10. Added some adjustments to the code to make it easier to compiler on certain
1326        special systems:
1327    
1328          (a) Some "const" qualifiers were missing.
1329          (b) Added the macro EXPORT before all exported functions; by default this
1330              is defined to be empty.
1331          (c) Changed the dftables auxiliary program (that builds chartables.c) so
1332              that it reads its output file name as an argument instead of writing
1333              to the standard output and assuming this can be redirected.
1334    
1335    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
1336        class containing characters with values greater than 255, PCRE compilation
1337        went into a loop.
1338    
1339    12. A recursive reference to a subpattern that was within another subpattern
1340        that had a minimum quantifier of zero caused PCRE to crash. For example,
1341        (x(y(?2))z)? provoked this bug with a subject that got as far as the
1342        recursion. If the recursively-called subpattern itself had a zero repeat,
1343        that was OK.
1344    
1345    13. In pcretest, the buffer for reading a data line was set at 30K, but the
1346        buffer into which it was copied (for escape processing) was still set at
1347        1024, so long lines caused crashes.
1348    
1349    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
1350        "internal error: code overflow...". This applied to any character class
1351        that was followed by a possessive quantifier.
1352    
1353    15. Modified the Makefile to add libpcre.la as a prerequisite for
1354        libpcreposix.la because I was told this is needed for a parallel build to
1355        work.
1356    
1357    16. If a pattern that contained .* following optional items at the start was
1358        studied, the wrong optimizing data was generated, leading to matching
1359        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
1360        matching string must start with a or b or c. The correct conclusion for
1361        this pattern is that a match can start with any character.
1362    
1363    
1364    Version 4.4 13-Aug-03
1365    ---------------------
1366    
1367     1. In UTF-8 mode, a character class containing characters with values between
1368        127 and 255 was not handled correctly if the compiled pattern was studied.
1369        In fixing this, I have also improved the studying algorithm for such
1370        classes (slightly).
1371    
1372     2. Three internal functions had redundant arguments passed to them. Removal
1373        might give a very teeny performance improvement.
1374    
1375     3. Documentation bug: the value of the capture_top field in a callout is *one
1376        more than* the number of the hightest numbered captured substring.
1377    
1378     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
1379        in incorrectly linking with a previously installed version. They now link
1380        explicitly with libpcre.la.
1381    
1382     5. configure.in no longer needs to recognize Cygwin specially.
1383    
1384     6. A problem in pcre.in for Windows platforms is fixed.
1385    
1386     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
1387        pcretest, it used to include the size of the study block as part of its
1388        output. Unfortunately, the structure contains a field that has a different
1389        size on different hardware architectures. This meant that the tests that
1390        showed this size failed. As the block is currently always of a fixed size,
1391        this information isn't actually particularly useful in pcretest output, so
1392        I have just removed it.
1393    
1394     8. Three pre-processor statements accidentally did not start in column 1.
1395        Sadly, there are *still* compilers around that complain, even though
1396        standard C has not required this for well over a decade. Sigh.
1397    
1398     9. In pcretest, the code for checking callouts passed small integers in the
1399        callout_data field, which is a void * field. However, some picky compilers
1400        complained about the casts involved for this on 64-bit systems. Now
1401        pcretest passes the address of the small integer instead, which should get
1402        rid of the warnings.
1403    
1404    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
1405        both compile and run time, and gives an error if an invalid UTF-8 sequence
1406        is found. There is a option for disabling this check in cases where the
1407        string is known to be correct and/or the maximum performance is wanted.
1408    
1409    11. In response to a bug report, I changed one line in Makefile.in from
1410    
1411            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
1412        to
1413            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
1414    
1415        to look similar to other lines, but I have no way of telling whether this
1416        is the right thing to do, as I do not use Windows. No doubt I'll get told
1417        if it's wrong...
1418    
1419    
1420    Version 4.3 21-May-03
1421    ---------------------
1422    
1423    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
1424       Makefile.
1425    
1426    2. Some refactoring to improve the quality of the code:
1427    
1428       (i)   The utf8_table... variables are now declared "const".
1429    
1430       (ii)  The code for \cx, which used the "case flipping" table to upper case
1431             lower case letters, now just substracts 32. This is ASCII-specific,
1432             but the whole concept of \cx is ASCII-specific, so it seems
1433             reasonable.
1434    
1435       (iii) PCRE was using its character types table to recognize decimal and
1436             hexadecimal digits in the pattern. This is silly, because it handles
1437             only 0-9, a-f, and A-F, but the character types table is locale-
1438             specific, which means strange things might happen. A private
1439             table is now used for this - though it costs 256 bytes, a table is
1440             much faster than multiple explicit tests. Of course, the standard
1441             character types table is still used for matching digits in subject
1442             strings against \d.
1443    
1444       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
1445             ending in _t are). So I've renamed it as ESC_tee.
1446    
1447    3. The first argument for regexec() in the POSIX wrapper should have been
1448       defined as "const".
1449    
1450    4. Changed pcretest to use malloc() for its buffers so that they can be
1451       Electric Fenced for debugging.
1452    
1453    5. There were several places in the code where, in UTF-8 mode, PCRE would try
1454       to read one or more bytes before the start of the subject string. Often this
1455       had no effect on PCRE's behaviour, but in some circumstances it could
1456       provoke a segmentation fault.
1457    
1458    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
1459       to try to read one or more bytes before the start of the subject string.
1460    
1461    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
1462       UTF-8 support could misbehave in various ways if the subject string
1463       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
1464       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
1465       back over UTF-8 characters.)
1466    
1467    
1468    Version 4.2 14-Apr-03
1469    ---------------------
1470    
1471    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
1472    
1473    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
1474         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
1475         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
1476         [WIN_PREFIX]: new variable, "cyg" for Cygwin
1477         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
1478           and BUILD_EXEEXT
1479         Note: automatic setting of the BUILD variables is not yet working
1480         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
1481           compile-time but not at link-time
1482         [LINK]: use for linking executables only
1483         make different versions for Windows and non-Windows
1484         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
1485           libraries
1486         [LINK_FOR_BUILD]: new variable
1487         [OBJEXT]: use throughout
1488         [EXEEXT]: use throughout
1489         <winshared>: new target
1490         <wininstall>: new target
1491         <dftables.o>: use native compiler
1492         <dftables>: use native linker
1493         <install>: handle Windows platform correctly
1494         <clean>: ditto
1495         <check>: ditto
1496         copy DLL to top builddir before testing
1497    
1498       As part of these changes, -no-undefined was removed again. This was reported
1499       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
1500       in any case.
1501    
1502    3. Some tidies to get rid of compiler warnings:
1503    
1504       . In the match_data structure, match_limit was an unsigned long int, whereas
1505         match_call_count was an int. I've made them both unsigned long ints.
1506    
1507       . In pcretest the fact that a const uschar * doesn't automatically cast to
1508         a void * provoked a warning.
1509    
1510       . Turning on some more compiler warnings threw up some "shadow" variables
1511         and a few more missing casts.
1512    
1513    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
1514       option, a class that contained a single character with a value between 128
1515       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
1516    
1517    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
1518       option, a class that contained several characters, but with at least one
1519       whose value was between 128 and 255 caused PCRE to crash.
1520    
1521    
1522    Version 4.1 12-Mar-03
1523    ---------------------
1524    
1525    1. Compiling with gcc -pedantic found a couple of places where casts were
1526    needed, and a string in dftables.c that was longer than standard compilers are
1527    required to support.
1528    
1529    2. Compiling with Sun's compiler found a few more places where the code could
1530    be tidied up in order to avoid warnings.
1531    
1532    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
1533    first of these names is deprecated in the latest Autoconf in favour of the name
1534    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
1535    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
1536    analogy I have changed it to CFLAGS_FOR_BUILD.
1537    
1538    4. Added -no-undefined to the linking command in the Makefile, because this is
1539    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
1540    linking step for the pcreposix library.
1541    
1542    5. PCRE was failing to diagnose the case of two named groups with the same
1543    name.
1544    
1545    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
1546    literal character that is needed in the subject for a match, and scans along to
1547    ensure that it is present before embarking on the full matching process. This
1548    saves time in cases of nested unlimited repeats that are never going to match.
1549    Problem: the scan can take a lot of time if the subject is very long (e.g.
1550    megabytes), thus penalizing straightforward matches. It is now done only if the
1551    amount of subject to be scanned is less than 1000 bytes.
1552    
1553    7. A lesser problem with the same optimization is that it was recording the
1554    first character of an anchored pattern as "needed", thus provoking a search
1555    right along the subject, even when the first match of the pattern was going to
1556    fail. The "needed" character is now not set for anchored patterns, unless it
1557    follows something in the pattern that is of non-fixed length. Thus, it still
1558    fulfils its original purpose of finding quick non-matches in cases of nested
1559    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
1560    
1561    
1562    Version 4.0 17-Feb-03
1563    ---------------------
1564    
1565    1. If a comment in an extended regex that started immediately after a meta-item
1566    extended to the end of string, PCRE compiled incorrect data. This could lead to
1567    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
1568    
1569    2. Moved to autoconf 2.53 and libtool 1.4.2.
1570    
1571    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
1572    the special perltest8 script is no longer needed - all the tests can be run
1573    from a single perltest script.
1574    
1575    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
1576    by \s. It has now been removed in PCRE. This means it isn't recognized as
1577    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
1578    class [:space:] *does* include VT, thereby creating a mess.
1579    
1580    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
1581    space and tab.
1582    
1583    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
1584    its new features into the main test script, reducing the number of scripts.
1585    
1586    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
1587    were backward compatible, and made the (?i) apply to the whole pattern, as if
1588    /i were given. Now it behaves more logically, and applies the option setting
1589    only to what follows. PCRE has been changed to follow suit. However, if it
1590    finds options settings right at the start of the pattern, it extracts them into
1591    the global options, as before. Thus, they show up in the info data.
1592    
1593    8. Added support for the \Q...\E escape sequence. Characters in between are
1594    treated as literals. This is slightly different from Perl in that $ and @ are
1595    also handled as literals inside the quotes. In Perl, they will cause variable
1596    interpolation. Note the following examples:
1597    
1598        Pattern            PCRE matches      Perl matches
1599    
1600        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
1601        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
1602        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
1603    
1604    For compatibility with Perl, \Q...\E sequences are recognized inside character
1605    classes as well as outside them.
1606    
1607    9. Re-organized 3 code statements in pcretest to avoid "overflow in
1608    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
1609    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
1610    signed/unsigned warnings.
1611    
1612    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
1613    option for pcretest, so I've replaced it by a simple function that does just
1614    that job.
1615    
1616    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
1617    "pcregrep -".
1618    
1619    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
1620    Java package. This provides some syntactic sugar for simple cases of what my
1621    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
1622    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
1623    item, you can use this simplified notation. Note that only makes sense with
1624    greedy quantifiers. Consequently, the use of the possessive quantifier forces
1625    greediness, whatever the setting of the PCRE_UNGREEDY option.
1626    
1627    13. A change of greediness default within a pattern was not taking effect at
1628    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
1629    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
1630    was abstracted outside.
1631    
1632    14. PCRE now supports the \G assertion. It is true when the current matching
1633    position is at the start point of the match. This differs from \A when the
1634    starting offset is non-zero. Used with the /g option of pcretest (or similar
1635    code), it works in the same way as it does for Perl's /g option. If all
1636    alternatives of a regex begin with \G, the expression is anchored to the start
1637    match position, and the "anchored" flag is set in the compiled expression.
1638    
1639    15. Some bugs concerning the handling of certain option changes within patterns
1640    have been fixed. These applied to options other than (?ims). For example,
1641    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
1642    been the other way round. Some of this was related to change 7 above.
1643    
1644    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
1645    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
1646    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
1647    POSIX classes only within a class (e.g. /[[:alpha:]]/).
1648    
1649    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
1650    mode. Unlike ".", it always matches newline, whatever the setting of
1651    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
1652    assertions. Perl allows it, but it doesn't (in general) work because it can't
1653    calculate the length of the lookbehind. At least, that's the case for Perl
1654    5.8.0 - I've been told they are going to document that it doesn't work in
1655    future.
1656    
1657    18. Added an error diagnosis for escapes that PCRE does not support: these are
1658    \L, \l, \N, \P, \p, \U, \u, and \X.
1659    
1660    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
1661    reading past the end of the pattern in cases such as /[abcd/.
1662    
1663    20. PCRE was getting more memory than necessary for patterns with classes that
1664    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
1665    
1666    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
1667    compiling PCRE for use with Virtual Pascal.
1668    
1669    22. Small fix to the Makefile to make it work properly if the build is done
1670    outside the source tree.
1671    
1672    23. Added a new extension: a condition to go with recursion. If a conditional
1673    subpattern starts with (?(R) the "true" branch is used if recursion has
1674    happened, whereas the "false" branch is used only at the top level.
1675    
1676    24. When there was a very long string of literal characters (over 255 bytes
1677    without UTF support, over 250 bytes with UTF support), the computation of how
1678    much memory was required could be incorrect, leading to segfaults or other
1679    strange effects.
1680    
1681    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
1682    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
1683    there was a subsequent back reference to those brackets. This meant that, for
1684    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
1685    possible to check for precisely this case. All we can do is abandon the
1686    optimization if .* occurs inside capturing brackets when there are any back
1687    references whatsoever. (See below for a better fix that came later.)
1688    
1689    26. The handling of the optimization for finding the first character of a
1690    non-anchored pattern, and for finding a character that is required later in the
1691    match were failing in some cases. This didn't break the matching; it just
1692    failed to optimize when it could. The way this is done has been re-implemented.
1693    
1694    27. Fixed typo in error message for invalid (?R item (it said "(?p").
1695    
1696    28. Added a new feature that provides some of the functionality that Perl
1697    provides with (?{...}). The facility is termed a "callout". The way it is done
1698    in PCRE is for the caller to provide an optional function, by setting
1699    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
1700    global variable. By default it is unset, which disables all calling out. To get
1701    the function called, the regex must include (?C) at appropriate points. This
1702    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
1703    This provides a means of identifying different callout points. When PCRE
1704    reaches such a point in the regex, if pcre_callout has been set, the external
1705    function is called. It is provided with data in a structure called
1706    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
1707    matching continues; if it returns a non-zero value, the match at the current
1708    point fails. However, backtracking will occur if possible. [This was changed
1709    later and other features added - see item 49 below.]
1710    
1711    29. pcretest is upgraded to test the callout functionality. It provides a
1712    callout function that displays information. By default, it shows the start of
1713    the match and the current position in the text. There are some new data escapes
1714    to vary what happens:
1715    
1716        \C+         in addition, show current contents of captured substrings
1717        \C-         do not supply a callout function
1718        \C!n        return 1 when callout number n is reached
1719        \C!n!m      return 1 when callout number n is reached for the mth time
1720    
1721    30. If pcregrep was called with the -l option and just a single file name, it
1722    output "<stdin>" if a match was found, instead of the file name.
1723    
1724    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
1725    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
1726    pcre_exec(). This saves a malloc/free per call. The default value of
1727    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
1728    when configuring.
1729    
1730    32. The default maximum size of a compiled pattern is 64K. There have been a
1731    few cases of people hitting this limit. The code now uses macros to handle the
1732    storing of links as offsets within the compiled pattern. It defaults to 2-byte
1733    links, but this can be changed to 3 or 4 bytes by --with-link-size when
1734    configuring. Tests 2 and 5 work only with 2-byte links because they output
1735    debugging information about compiled patterns.
1736    
1737    33. Internal code re-arrangements:
1738    
1739    (a) Moved the debugging function for printing out a compiled regex into
1740        its own source file (printint.c) and used #include to pull it into
1741        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
1742        separate copies.
1743    
1744    (b) Defined the list of op-code names for debugging as a macro in
1745        internal.h so that it is next to the definition of the opcodes.
1746    
1747    (c) Defined a table of op-code lengths for simpler skipping along compiled
1748        code. This is again a macro in internal.h so that it is next to the
1749        definition of the opcodes.
1750    
1751    34. Added support for recursive calls to individual subpatterns, along the
1752    lines of Robin Houston's patch (but implemented somewhat differently).
1753    
1754    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
1755    allow it to read and process whole directories in Win32. This code was
1756    contributed by Lionel Fourquaux; it has not been tested by me.
1757    
1758    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
1759    used to name a group. Names consist of alphanumerics and underscores, and must
1760    be unique. Back references use the syntax (?P=name) and recursive calls use
1761    (?P>name) which is a PCRE extension to the Python extension. Groups still have
1762    numbers. The function pcre_fullinfo() can be used after compilation to extract
1763    a name/number map. There are three relevant calls:
1764    
1765      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
1766      PCRE_INFO_NAMECOUNT            yields the number of entries
1767      PCRE_INFO_NAMETABLE            yields a pointer to the map.
1768    
1769    The map is a vector of fixed-size entries. The size of each entry depends on
1770    the length of the longest name used. The first two bytes of each entry are the
1771    group number, most significant byte first. There follows the corresponding
1772    name, zero terminated. The names are in alphabetical order.
1773    
1774    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
1775    case instead of 255. Making it the same both with and without UTF-8 support
1776    means that the same test output works with both.
1777    
1778    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
1779    calling malloc() with a zero argument.
1780    
1781    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
1782    optimization. I've improved things by keeping a bitmap of backreferences with
1783    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
1784    fact referenced, the optimization can be applied. It is unlikely that a
1785    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
1786    the match to follow \n) will appear inside brackets with a number greater than
1787    31, but if it does, any back reference > 31 suppresses the optimization.
1788    
1789    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
1790    of disabling numbered capturing parentheses. Any opening parenthesis that is
1791    not followed by ? behaves as if it were followed by ?: but named parentheses
1792    can still be used for capturing (and they will acquire numbers in the usual
1793    way).
1794    
1795    41. Redesigned the return codes from the match() function into yes/no/error so
1796    that errors can be passed back from deep inside the nested calls. A malloc
1797    failure while inside a recursive subpattern call now causes the
1798    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
1799    
1800    42. It is now possible to set a limit on the number of times the match()
1801    function is called in a call to pcre_exec(). This facility makes it possible to
1802    limit the amount of recursion and backtracking, though not in a directly
1803    obvious way, because the match() function is used in a number of different
1804    circumstances. The count starts from zero for each position in the subject
1805    string (for non-anchored patterns). The default limit is, for compatibility, a
1806    large number, namely 10 000 000. You can change this in two ways:
1807    
1808    (a) When configuring PCRE before making, you can use --with-match-limit=n
1809        to set a default value for the compiled library.
1810    
1811    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
1812        a different value is set. See 45 below.
1813    
1814    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
1815    
1816    43. Added a new function pcre_config(int, void *) to enable run-time extraction
1817    of things that can be changed at compile time. The first argument specifies
1818    what is wanted and the second points to where the information is to be placed.
1819    The current list of available information is:
1820    
1821      PCRE_CONFIG_UTF8
1822    
1823    The output is an integer that is set to one if UTF-8 support is available;
1824    otherwise it is set to zero.
1825    
1826      PCRE_CONFIG_NEWLINE
1827    
1828    The output is an integer that it set to the value of the code that is used for
1829    newline. It is either LF (10) or CR (13).
1830    
1831      PCRE_CONFIG_LINK_SIZE
1832    
1833    The output is an integer that contains the number of bytes used for internal
1834    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
1835    
1836      PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
1837    
1838    The output is an integer that contains the threshold above which the POSIX
1839    interface uses malloc() for output vectors. See item 31 above.
1840    
1841      PCRE_CONFIG_MATCH_LIMIT
1842    
1843    The output is an unsigned integer that contains the default limit of the number
1844    of match() calls in a pcre_exec() execution. See 42 above.
1845    
1846    44. pcretest has been upgraded by the addition of the -C option. This causes it
1847    to extract all the available output from the new pcre_config() function, and to
1848    output it. The program then exits immediately.
1849    
1850    45. A need has arisen to pass over additional data with calls to pcre_exec() in
1851    order to support additional features. One way would have been to define
1852    pcre_exec2() (for example) with extra arguments, but this would not have been
1853    extensible, and would also have required all calls to the original function to
1854    be mapped to the new one. Instead, I have chosen to extend the mechanism that
1855    is used for passing in "extra" data from pcre_study().
1856    
1857    The pcre_extra structure is now exposed and defined in pcre.h. It currently
1858    contains the following fields:
1859    
1860      flags         a bitmap indicating which of the following fields are set
1861      study_data    opaque data from pcre_study()
1862      match_limit   a way of specifying a limit on match() calls for a specific
1863                      call to pcre_exec()
1864      callout_data  data for callouts (see 49 below)
1865    
1866    The flag bits are also defined in pcre.h, and are
1867    
1868      PCRE_EXTRA_STUDY_DATA
1869      PCRE_EXTRA_MATCH_LIMIT
1870      PCRE_EXTRA_CALLOUT_DATA
1871    
1872    The pcre_study() function now returns one of these new pcre_extra blocks, with
1873    the actual study data pointed to by the study_data field, and the
1874    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
1875    before. That is, this change is entirely upwards-compatible and requires no
1876    change to existing code.
1877    
1878    If you want to pass in additional data to pcre_exec(), you can either place it
1879    in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
1880    block.
1881    
1882    46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
1883    data string contains the escape sequence \M, pcretest calls pcre_exec() several
1884    times with different match limits, until it finds the minimum value needed for
1885    pcre_exec() to complete. The value is then output. This can be instructive; for
1886    most simple matches the number is quite small, but for pathological cases it
1887    gets very large very quickly.
1888    
1889    47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
1890    returns the size of the data block pointed to by the study_data field in a
1891    pcre_extra block, that is, the value that was passed as the argument to
1892    pcre_malloc() when PCRE was getting memory in which to place the information
1893    created by pcre_study(). The fourth argument should point to a size_t variable.
1894    pcretest has been extended so that this information is shown after a successful
1895    pcre_study() call when information about the compiled regex is being displayed.
1896    
1897    48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
1898    because what follows is always an absolute path. (Later: it turns out that this
1899    is more than cosmetic for MinGW, because it doesn't like empty path
1900    components.)
1901    
1902    49. Some changes have been made to the callout feature (see 28 above):
1903    
1904    (i)  A callout function now has three choices for what it returns:
1905    
1906           0  =>  success, carry on matching
1907         > 0  =>  failure at this point, but backtrack if possible
1908         < 0  =>  serious error, return this value from pcre_exec()
1909    
1910         Negative values should normally be chosen from the set of PCRE_ERROR_xxx
1911         values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
1912         "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
1913         use by callout functions. It will never be used by PCRE itself.
1914    
1915    (ii) The pcre_extra structure (see 45 above) has a void * field called
1916         callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
1917         pcre_callout_block structure has a field of the same name. The contents of
1918         the field passed in the pcre_extra structure are passed to the callout
1919         function in the corresponding field in the callout block. This makes it
1920         easier to use the same callout-containing regex from multiple threads. For
1921         testing, the pcretest program has a new data escape
1922    
1923           \C*n        pass the number n (may be negative) as callout_data
1924    
1925         If the callout function in pcretest receives a non-zero value as
1926         callout_data, it returns that value.
1927    
1928    50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
1929    there were some redundant $(CFLAGS) in commands that are now specified as
1930    $(LINK), which already includes $(CFLAGS).
1931    
1932    51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
1933    has been compiled with UTF-8 support *and* pcre_compile() has been compiled
1934    with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
1935    one-byte characters throughout. Note that case-insensitive matching applies
1936    only to characters whose values are less than 256. PCRE doesn't support the
1937    notion of cases for higher-valued characters.
1938    
1939    (i)   A character class whose characters are all within 0-255 is handled as
1940          a bit map, and the map is inverted for negative classes. Previously, a
1941          character > 255 always failed to match such a class; however it should
1942          match if the class was a negative one (e.g. [^ab]). This has been fixed.
1943    
1944    (ii)  A negated character class with a single character < 255 is coded as
1945          "not this character" (OP_NOT). This wasn't working properly when the test
1946          character was multibyte, either singly or repeated.
1947    
1948    (iii) Repeats of multibyte characters are now handled correctly in UTF-8
1949          mode, for example: \x{100}{2,3}.
1950    
1951    (iv)  The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
1952          singly or repeated) now correctly test multibyte characters. However,
1953          PCRE doesn't recognize any characters with values greater than 255 as
1954          digits, spaces, or word characters. Such characters always match \D, \S,
1955          and \W, and never match \d, \s, or \w.
1956    
1957    (v)   Classes may now contain characters and character ranges with values
1958          greater than 255. For example: [ab\x{100}-\x{400}].
1959    
1960    (vi)  pcregrep now has a --utf-8 option (synonym -u) which makes it call
1961          PCRE in UTF-8 mode.
1962    
1963    52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
1964    PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
1965    retained for backwards compatibility. (Note that LASTLITERAL is also a byte
1966    value.)
1967    
1968    53. The single man page has become too large. I have therefore split it up into
1969    a number of separate man pages. These also give rise to individual HTML pages;
1970    these are now put in a separate directory, and there is an index.html page that
1971    lists them all. Some hyperlinking between the pages has been installed.
1972    
1973    54. Added convenience functions for handling named capturing parentheses.
1974    
1975    55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
1976    aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
1977    true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
1978    are faulted.
1979    
1980    56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
1981    calling configure. These values are used when compiling the dftables.c program
1982    which is run to generate the source of the default character tables. They
1983    default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
1984    you will need to set these values.
1985    
1986    57. Updated the building process for Windows DLL, as provided by Fred Cox.
1987    
1988    
1989    Version 3.9 02-Jan-02
1990    ---------------------
1991    
1992    1. A bit of extraneous text had somehow crept into the pcregrep documentation.
1993    
1994    2. If --disable-static was given, the building process failed when trying to
1995    build pcretest and pcregrep. (For some reason it was using libtool to compile
1996    them, which is not right, as they aren't part of the library.)
1997    
1998    
1999    Version 3.8 18-Dec-01
2000    ---------------------
2001    
2002    1. The experimental UTF-8 code was completely screwed up. It was packing the
2003    bytes in the wrong order. How dumb can you get?
2004    
2005    
2006    Version 3.7 29-Oct-01
2007    ---------------------
2008    
2009    1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2010    This caused pcretest, when used on the test data, to segfault. Unfortunately,
2011    this didn't happen under Solaris 8, where I normally test things.
2012    
2013    2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2014    doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2015    isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2016    this fix an hour or so after the initial 3.7 release.)
2017    
2018    
2019    Version 3.6 23-Oct-01
2020    ---------------------
2021    
2022    1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2023    offsets passed as NULL with zero offset count.
2024    
2025    2. The config.guess and config.sub files had not been updated when I moved to
2026    the latest autoconf.
2027    
2028    
2029    Version 3.5 15-Aug-01
2030    ---------------------
2031    
2032    1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2033    had been forgotten.
2034    
2035    2. By using declared but undefined structures, we can avoid using "void"
2036    definitions in pcre.h while keeping the internal definitions of the structures
2037    private.
2038    
2039    3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2040    user point of view, this means that both static and shared libraries are built
2041    by default, but this can be individually controlled. More of the work of
2042    handling this static/shared cases is now inside libtool instead of PCRE's make
2043    file.
2044    
2045    4. The pcretest utility is now installed along with pcregrep because it is
2046    useful for users (to test regexs) and by doing this, it automatically gets
2047    relinked by libtool. The documentation has been turned into a man page, so
2048    there are now .1, .txt, and .html versions in /doc.
2049    
2050    5. Upgrades to pcregrep:
2051       (i)   Added long-form option names like gnu grep.
2052       (ii)  Added --help to list all options with an explanatory phrase.
2053       (iii) Added -r, --recursive to recurse into sub-directories.
2054       (iv)  Added -f, --file to read patterns from a file.
2055    
2056    6. pcre_exec() was referring to its "code" argument before testing that
2057    argument for NULL (and giving an error if it was NULL).
2058    
2059    7. Upgraded Makefile.in to allow for compiling in a different directory from
2060    the source directory.
2061    
2062    8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2063    options bits, the pointer it was passed was to an int instead of to an unsigned
2064    long int. This mattered only on 64-bit systems.
2065    
2066    9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2067    generated) instead of pcre.in, which it its source. Also made the same change
2068    in several of the .c files.
2069    
2070    10. A new release of gcc defines printf() as a macro, which broke pcretest
2071    because it had an ifdef in the middle of a string argument for printf(). Fixed
2072    by using separate calls to printf().
2073    
2074    11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2075    script, to force use of CR or LF instead of \n in the source. On non-Unix
2076    systems, the value can be set in config.h.
2077    
2078    12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2079    absolute limit. Changed the text of the error message to make this clear, and
2080    likewise updated the man page.
2081    
2082    13. The limit of 99 on the number of capturing subpatterns has been removed.
2083    The new limit is 65535, which I hope will not be a "real" limit.
2084    
2085    
2086    Version 3.4 22-Aug-00
2087    ---------------------
2088    
2089    1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2090    
2091    2. Diagnose condition (?(0) as an error instead of crashing on matching.
2092    
2093    
2094    Version 3.3 01-Aug-00
2095    ---------------------
2096    
2097    1. If an octal character was given, but the value was greater than \377, it
2098    was not getting masked to the least significant bits, as documented. This could
2099    lead to crashes in some systems.
2100    
2101    2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
2102    the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
2103    
2104    3. Added the functions pcre_free_substring() and pcre_free_substring_list().
2105    These just pass their arguments on to (pcre_free)(), but they are provided
2106    because some uses of PCRE bind it to non-C systems that can call its functions,
2107    but cannot call free() or pcre_free() directly.
2108    
2109    4. Add "make test" as a synonym for "make check". Corrected some comments in
2110    the Makefile.
2111    
2112    5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
2113    Makefile.
2114    
2115    6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
2116    command called pgrep for grepping around the active processes.
2117    
2118    7. Added the beginnings of support for UTF-8 character strings.
2119    
2120    8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
2121    RANLIB to ./ltconfig so that they are used by libtool. I think these are all
2122    the relevant ones. (AR is not passed because ./ltconfig does its own figuring
2123    out for the ar command.)
2124    
2125    
2126    Version 3.2 12-May-00
2127    ---------------------
2128    
2129    This is purely a bug fixing release.
2130    
2131    1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
2132    of ZA. This was just one example of several cases that could provoke this bug,
2133    which was introduced by change 9 of version 2.00. The code for breaking
2134    infinite loops after an iteration that matches an empty string was't working
2135    correctly.
2136    
2137    2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
2138    when matched against abbab (for example). After matching an empty string, it
2139    wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
2140    caused it to match further down the string than it should.
2141    
2142    3. The code contained an inclusion of sys/types.h. It isn't clear why this
2143    was there because it doesn't seem to be needed, and it causes trouble on some
2144    systems, as it is not a Standard C header. It has been removed.
2145    
2146    4. Made 4 silly changes to the source to avoid stupid compiler warnings that
2147    were reported on the Macintosh. The changes were from
2148    
2149      while ((c = *(++ptr)) != 0 && c != '\n');
2150    to
2151      while ((c = *(++ptr)) != 0 && c != '\n') ;
2152    
2153    Totally extraordinary, but if that's what it takes...
2154    
2155    5. PCRE is being used in one environment where neither memmove() nor bcopy() is
2156    available. Added HAVE_BCOPY and an autoconf test for it; if neither
2157    HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
2158    assumes the way PCRE uses memmove() (always moving upwards).
2159    
2160    6. PCRE is being used in one environment where strchr() is not available. There
2161    was only one use in pcre.c, and writing it out to avoid strchr() probably gives
2162    faster code anyway.
2163    
2164    
2165    Version 3.1 09-Feb-00
2166    ---------------------
2167    
2168    The only change in this release is the fixing of some bugs in Makefile.in for
2169    the "install" target:
2170    
2171    (1) It was failing to install pcreposix.h.
2172    
2173    (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
2174    
2175    
2176    Version 3.0 01-Feb-00
2177    ---------------------
2178    
2179    1. Add support for the /+ modifier to perltest (to output $` like it does in
2180    pcretest).
2181    
2182    2. Add support for the /g modifier to perltest.
2183    
2184    3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
2185    matches null strings.
2186    
2187    4. Fix perltest so that it doesn't do unwanted things when fed an empty
2188    pattern. Perl treats empty patterns specially - it reuses the most recent
2189    pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
2190    effect.
2191    
2192    5. The POSIX interface was broken in that it was just handing over the POSIX
2193    captured string vector to pcre_exec(), but (since release 2.00) PCRE has
2194    required a bigger vector, with some working space on the end. This means that
2195    the POSIX wrapper now has to get and free some memory, and copy the results.
2196    
2197    6. Added some simple autoconf support, placing the test data and the
2198    documentation in separate directories, re-organizing some of the
2199    information files, and making it build pcre-config (a GNU standard). Also added
2200    libtool support for building PCRE as a shared library, which is now the
2201    default.
2202    
2203    7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
2204    09 are not valid octal constants. Single digits will be used for minor values
2205    less than 10.
2206    
2207    8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
2208    existing programs that set these in the POSIX interface can use PCRE without
2209    modification.
2210    
2211    9. Added a new function, pcre_fullinfo() with an extensible interface. It can
2212    return all that pcre_info() returns, plus additional data. The pcre_info()
2213    function is retained for compatibility, but is considered to be obsolete.
2214    
2215    10. Added experimental recursion feature (?R) to handle one common case that
2216    Perl 5.6 will be able to do with (?p{...}).
2217    
2218    11. Added support for POSIX character classes like [:alpha:], which Perl is
2219    adopting.
2220    
2221    
2222    Version 2.08 31-Aug-99
2223    ----------------------
2224    
2225    1. When startoffset was not zero and the pattern began with ".*", PCRE was not
2226    trying to match at the startoffset position, but instead was moving forward to
2227    the next newline as if a previous match had failed.
2228    
2229    2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
2230    and could get into a loop if a null string was matched other than at the start
2231    of the subject.
2232    
2233    3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
2234    be distinguished at compile time, and for completeness also added PCRE_DATE.
2235    
2236    5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
2237    in GnuWin32 environments.
2238    
2239    
2240    Version 2.07 29-Jul-99
2241    ----------------------
2242    
2243    1. The documentation is now supplied in plain text form and HTML as well as in
2244    the form of man page sources.
2245    
2246    2. C++ compilers don't like assigning (void *) values to other pointer types.
2247    In particular this affects malloc(). Although there is no problem in Standard
2248    C, I've put in casts to keep C++ compilers happy.
2249    
2250    3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
2251    should be (const char *).
2252    
2253    4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
2254    be useful for non-Unix systems who don't want to bother with the POSIX stuff.
2255    However, I haven't made this a standard facility. The documentation doesn't
2256    mention it, and the Makefile doesn't support it.
2257    
2258    5. The Makefile now contains an "install" target, with editable destinations at
2259    the top of the file. The pcretest program is not installed.
2260    
2261    6. pgrep -V now gives the PCRE version number and date.
2262    
2263    7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
2264    causing the entire string to be ignored, instead of just the last character.
2265    
2266    8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
2267    non-matching string, it can take a very, very long time, even for strings of
2268    quite modest length, because of the nested recursion. PCRE now does better in
2269    some of these cases. It does this by remembering the last required literal
2270    character in the pattern, and pre-searching the subject to ensure it is present
2271    before running the real match. In other words, it applies a heuristic to detect
2272    some types of certain failure quickly, and in the above example, if presented
2273    with a string that has no trailing " it gives "no match" very quickly.
2274    
2275    9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
2276    other alternatives are tried instead.
2277    
2278    
2279    Version 2.06 09-Jun-99
2280    ----------------------
2281    
2282    1. Change pcretest's output for amount of store used to show just the code
2283    space, because the remainder (the data block) varies in size between 32-bit and
2284    64-bit systems.
2285    
2286    2. Added an extra argument to pcre_exec() to supply an offset in the subject to
2287    start matching at. This allows lookbehinds to work when searching for multiple
2288    occurrences in a string.
2289    
2290    3. Added additional options to pcretest for testing multiple occurrences:
2291    
2292       /+   outputs the rest of the string that follows a match
2293       /g   loops for multiple occurrences, using the new startoffset argument
2294       /G   loops for multiple occurrences by passing an incremented pointer
2295    
2296    4. PCRE wasn't doing the "first character" optimization for patterns starting
2297    with \b or \B, though it was doing it for other lookbehind assertions. That is,
2298    it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
2299    the letter 'x'. On long subject strings, this gives a significant speed-up.
2300    
2301    
2302    Version 2.05 21-Apr-99
2303    ----------------------
2304    
2305    1. Changed the type of magic_number from int to long int so that it works
2306    properly on 16-bit systems.
2307    
2308    2. Fixed a bug which caused patterns starting with .* not to work correctly
2309    when the subject string contained newline characters. PCRE was assuming
2310    anchoring for such patterns in all cases, which is not correct because .* will
2311    not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
2312    DOTALL is set at top level; otherwise it knows that patterns starting with .*
2313    must be retried after every newline in the subject.
2314    
2315    
2316    Version 2.04 18-Feb-99
2317    ----------------------
2318    
2319    1. For parenthesized subpatterns with repeats whose minimum was zero, the
2320    computation of the store needed to hold the pattern was incorrect (too large).
2321    If such patterns were nested a few deep, this could multiply and become a real
2322    problem.
2323    
2324    2. Added /M option to pcretest to show the memory requirement of a specific
2325    pattern. Made -m a synonym of -s (which does this globally) for compatibility.
2326    
2327    3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
2328    compiled in such a way that the backtracking after subsequent failure was
2329    pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
2330    ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
2331    
2332    
2333    Version 2.03 02-Feb-99
2334    ----------------------
2335    
2336    1. Fixed typo and small mistake in man page.
2337    
2338    2. Added 4th condition (GPL supersedes if conflict) and created separate
2339    LICENCE file containing the conditions.
2340    
2341    3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
2342    Perl, that is the internal \ allows the delimiter to be included in the
2343    pattern. Locked out the use of \ as a delimiter. If \ immediately follows
2344    the final delimiter, add \ to the end of the pattern (to test the error).
2345    
2346    4. Added the convenience functions for extracting substrings after a successful
2347    match. Updated pcretest to make it able to test these functions.
2348    
2349    
2350    Version 2.02 14-Jan-99
2351    ----------------------
2352    
2353    1. Initialized the working variables associated with each extraction so that
2354    their saving and restoring doesn't refer to uninitialized store.
2355    
2356    2. Put dummy code into study.c in order to trick the optimizer of the IBM C
2357    compiler for OS/2 into generating correct code. Apparently IBM isn't going to
2358    fix the problem.
2359    
2360    3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
2361    calls, and wasn't printing the correct value for compiling calls. Increased the
2362    default value of LOOPREPEAT, and the number of significant figures in the
2363    times.
2364    
2365    4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
2366    
2367    5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
2368    a building problem on Windows NT with a FAT file system.
2369    
2370    
2371    Version 2.01 21-Oct-98
2372    ----------------------
2373    
2374    1. Changed the API for pcre_compile() to allow for the provision of a pointer
2375    to character tables built by pcre_maketables() in the current locale. If NULL
2376    is passed, the default tables are used.
2377    
2378    
2379    Version 2.00 24-Sep-98
2380    ----------------------
2381    
2382    1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
2383    it any more.
2384    
2385    2. Allow quantification of (?>) groups, and make it work correctly.
2386    
2387    3. The first character computation wasn't working for (?>) groups.
2388    
2389    4. Correct the implementation of \Z (it is permitted to match on the \n at the
2390    end of the subject) and add 5.005's \z, which really does match only at the
2391    very end of the subject.
2392    
2393    5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
2394    
2395    6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
2396    DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
2397    localized options. All options to pcre_study() were also removed.
2398    
2399    7. Add other new features from 5.005:
2400    
2401       $(?<=           positive lookbehind
2402       $(?<!           negative lookbehind
2403       (?imsx-imsx)    added the unsetting capability
2404                       such a setting is global if at outer level; local otherwise
2405       (?imsx-imsx:)   non-capturing groups with option setting
2406       (?(cond)re|re)  conditional pattern matching
2407    
2408       A backreference to itself in a repeated group matches the previous
2409       captured string.
2410    
2411    8. General tidying up of studying (both automatic and via "study")
2412    consequential on the addition of new assertions.
2413    
2414    9. As in 5.005, unlimited repeated groups that could match an empty substring
2415    are no longer faulted at compile time. Instead, the loop is forcibly broken at
2416    runtime if any iteration does actually match an empty substring.
2417    
2418    10. Include the RunTest script in the distribution.
2419    
2420    11. Added tests from the Perl 5.005_02 distribution. This showed up a few
2421    discrepancies, some of which were old and were also with respect to 5.004. They
2422    have now been fixed.
2423    
2424    
2425    Version 1.09 28-Apr-98
2426    ----------------------
2427    
2428    1. A negated single character class followed by a quantifier with a minimum
2429    value of one (e.g.  [^x]{1,6}  ) was not compiled correctly. This could lead to
2430    program crashes, or just wrong answers. This did not apply to negated classes
2431    containing more than one character, or to minima other than one.
2432    
2433    
2434    Version 1.08 27-Mar-98
2435    ----------------------
2436    
2437    1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
2438    
2439    2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
2440    latter must appear before anything that relies on it in the pattern.
2441    
2442    
2443  Version 1.07 16-Feb-98  Version 1.07 16-Feb-98
2444  ----------------------  ----------------------

Legend:
Removed from v.17  
changed lines
  Added in v.167

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12