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

Diff of /code/trunk/ChangeLog

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

revision 61 by nigel, Sat Feb 24 21:39:58 2007 UTC revision 189 by ph10, Thu Jul 5 11:50:49 2007 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4  Version 3.0 02-Jan-02  Version 7.3 05-Jul-07
5    ---------------------
6    
7     1. In the rejigging of the build system that eventually resulted in 7.1, the
8        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
9        brackets there is not right, since it causes compilers to look for an
10        installed pcre.h, not the version that is in the source that is being
11        compiled (which of course may be different). I have changed it back to:
12    
13          #include "pcre.h"
14    
15        I have a vague recollection that the change was concerned with compiling in
16        different directories, but in the new build system, that is taken care of
17        by the VPATH setting the Makefile.
18    
19    
20    
21    Version 7.2 19-Jun-07
22    ---------------------
23    
24     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
25        which is apparently normally available under Windows.
26    
27     2. Re-jig the pcregrep tests with different newline settings in an attempt
28        to make them independent of the local environment's newline setting.
29    
30     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
31    
32     4. Some of the "internals" tests were previously cut out when the link size
33        was not 2, because the output contained actual offsets. The recent new
34        "Z" feature of pcretest means that these can be cut out, making the tests
35        usable with all link sizes.
36    
37     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
38        stack recursion. This gives a massive performance boost under BSD, but just
39        a small improvement under Linux. However, it saves one field in the frame
40        in all cases.
41    
42     6. Added more features from the forthcoming Perl 5.10:
43    
44        (a) (?-n) (where n is a string of digits) is a relative subroutine or
45            recursion call. It refers to the nth most recently opened parentheses.
46    
47        (b) (?+n) is also a relative subroutine call; it refers to the nth next
48            to be opened parentheses.
49    
50        (c) Conditions that refer to capturing parentheses can be specified
51            relatively, for example, (?(-2)... or (?(+3)...
52    
53        (d) \K resets the start of the current match so that everything before
54            is not part of it.
55    
56        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
57    
58        (f) \g{name} is another synonym - part of Perl 5.10's unification of
59            reference syntax.
60    
61        (g) (?| introduces a group in which the numbering of parentheses in each
62            alternative starts with the same number.
63    
64        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
65    
66     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
67        PCRE_INFO_JCHANGED.
68    
69     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
70        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
71        for detecting groups that can match an empty string.
72    
73     9. A pattern with a very large number of alternatives (more than several
74        hundred) was running out of internal workspace during the pre-compile
75        phase, where pcre_compile() figures out how much memory will be needed. A
76        bit of new cunning has reduced the workspace needed for groups with
77        alternatives. The 1000-alternative test pattern now uses 12 bytes of
78        workspace instead of running out of the 4096 that are available.
79    
80    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
81    
82    11. Applied patch from Google to remove an optimization that didn't quite work.
83        The report of the bug said:
84    
85          pcrecpp::RE("a*").FullMatch("aaa") matches, while
86          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
87          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
88    
89    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
90        it matched the wrong number of bytes.
91    
92    
93    Version 7.1 24-Apr-07
94    ---------------------
95    
96     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
97        that is more "standard", making use of automake and other Autotools. There
98        is some re-arrangement of the files and adjustment of comments consequent
99        on this.
100    
101     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
102        for recursive directory scanning broke on some systems because the files
103        are not scanned in any specific order and on different systems the order
104        was different. A call to "sort" has been inserted into RunGrepTest for the
105        approprate test as a short-term fix. In the longer term there may be an
106        alternative.
107    
108     3. I had an email from Eric Raymond about problems translating some of PCRE's
109        man pages to HTML (despite the fact that I distribute HTML pages, some
110        people do their own conversions for various reasons). The problems
111        concerned the use of low-level troff macros .br and .in. I have therefore
112        removed all such uses from the man pages (some were redundant, some could
113        be replaced by .nf/.fi pairs). The 132html script that I use to generate
114        HTML has been updated to handle .nf/.fi and to complain if it encounters
115        .br or .in.
116    
117     4. Updated comments in configure.ac that get placed in config.h.in and also
118        arranged for config.h to be included in the distribution, with the name
119        config.h.generic, for the benefit of those who have to compile without
120        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
121    
122     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
123        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
124        makevp.bat for the current PCRE, using the additional files
125        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
126    
127     6. A Windows user reported a minor discrepancy with test 2, which turned out
128        to be caused by a trailing space on an input line that had got lost in his
129        copy. The trailing space was an accident, so I've just removed it.
130    
131     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
132        that is needed.
133    
134     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
135        as "const" (a) because they are and (b) because it helps the PHP
136        maintainers who have recently made a script to detect big data structures
137        in the php code that should be moved to the .rodata section. I remembered
138        to update Builducptable as well, so it won't revert if ucptable.h is ever
139        re-created.
140    
141     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
142        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
143        order to be able to cut out the UTF-8 tables in the latter when UTF-8
144        support is not required. This saves 1.5-2K of code, which is important in
145        some applications.
146    
147        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
148        so as not to refer to the tables, even though these functions will never be
149        called when UTF-8 support is disabled. Otherwise there are problems with a
150        shared library.
151    
152    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
153    
154        (a) It was defining its arguments as char * instead of void *.
155    
156        (b) It was assuming that all moves were upwards in memory; this was true
157            a long time ago when I wrote it, but is no longer the case.
158    
159        The emulated memove() is provided for those environments that have neither
160        memmove() nor bcopy(). I didn't think anyone used it these days, but that
161        is clearly not the case, as these two bugs were recently reported.
162    
163    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
164        and Detrail to create the HTML documentation, the .txt form of the man
165        pages, and it removes trailing spaces from listed files. It also creates
166        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
167        case, it wraps all the #defines with #ifndefs. This script should be run
168        before "make dist".
169    
170    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
171        with Unicode property support.
172    
173        (a) For a maximizing quantifier, if the two different cases of the
174            character were of different lengths in their UTF-8 codings (there are
175            some cases like this - I found 11), and the matching function had to
176            back up over a mixture of the two cases, it incorrectly assumed they
177            were both the same length.
178    
179        (b) When PCRE was configured to use the heap rather than the stack for
180            recursion during matching, it was not correctly preserving the data for
181            the other case of a UTF-8 character when checking ahead for a match
182            while processing a minimizing repeat. If the check also involved
183            matching a wide character, but failed, corruption could cause an
184            erroneous result when trying to check for a repeat of the original
185            character.
186    
187    13. Some tidying changes to the testing mechanism:
188    
189        (a) The RunTest script now detects the internal link size and whether there
190            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
191            values substituted by "configure". (The RunGrepTest script already did
192            this for UTF-8.) The configure.ac script no longer substitutes the
193            relevant variables.
194    
195        (b) The debugging options /B and /D in pcretest show the compiled bytecode
196            with length and offset values. This means that the output is different
197            for different internal link sizes. Test 2 is skipped for link sizes
198            other than 2 because of this, bypassing the problem. Unfortunately,
199            there was also a test in test 3 (the locale tests) that used /B and
200            failed for link sizes other than 2. Rather than cut the whole test out,
201            I have added a new /Z option to pcretest that replaces the length and
202            offset values with spaces. This is now used to make test 3 independent
203            of link size. (Test 2 will be tidied up later.)
204    
205    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
206        segmentation fault instead of returning the appropriate error message.
207    
208    15. In multiline mode when the newline sequence was set to "any", the pattern
209        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
210        This doesn't seem right; it now treats the CRLF combination as the line
211        ending, and so does not match in that case. It's only a pattern such as ^$
212        that would hit this one: something like ^ABC$ would have failed after \r
213        and then tried again after \r\n.
214    
215    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
216        in an attempt to make files that differ only in their line terminators
217        compare equal. This works on Linux.
218    
219    17. Under certain error circumstances pcregrep might try to free random memory
220        as it exited. This is now fixed, thanks to valgrind.
221    
222    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
223        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
224        was because its rules for how to advance for /g after matching an empty
225        string at the end of a line did not allow for this case. They now check for
226        it specially.
227    
228    20. pcretest is supposed to handle patterns and data of any length, by
229        extending its buffers when necessary. It was getting this wrong when the
230        buffer for a data line had to be extended.
231    
232    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
233        CRLF as a newline sequence.
234    
235    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
236        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
237        I have nevertheless tidied it up.
238    
239    23. Added some casts to kill warnings from HP-UX ia64 compiler.
240    
241    24. Added a man page for pcre-config.
242    
243    
244    Version 7.0 19-Dec-06
245    ---------------------
246    
247     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
248        moving to gcc 4.1.1.
249    
250     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
251        sys/time.h, which is documented as needed for this function. It doesn't
252        seem to matter on Linux, but it showed up on some releases of OS X.
253    
254     3. It seems that there are systems where bytes whose values are greater than
255        127 match isprint() in the "C" locale. The "C" locale should be the
256        default when a C program starts up. In most systems, only ASCII printing
257        characters match isprint(). This difference caused the output from pcretest
258        to vary, making some of the tests fail. I have changed pcretest so that:
259    
260        (a) When it is outputting text in the compiled version of a pattern, bytes
261            other than 32-126 are always shown as hex escapes.
262    
263        (b) When it is outputting text that is a matched part of a subject string,
264            it does the same, unless a different locale has been set for the match
265            (using the /L modifier). In this case, it uses isprint() to decide.
266    
267     4. Fixed a major bug that caused incorrect computation of the amount of memory
268        required for a compiled pattern when options that changed within the
269        pattern affected the logic of the preliminary scan that determines the
270        length. The relevant options are -x, and -i in UTF-8 mode. The result was
271        that the computed length was too small. The symptoms of this bug were
272        either the PCRE error "internal error: code overflow" from pcre_compile(),
273        or a glibc crash with a message such as "pcretest: free(): invalid next
274        size (fast)". Examples of patterns that provoked this bug (shown in
275        pcretest format) are:
276    
277          /(?-x: )/x
278          /(?x)(?-x: \s*#\s*)/
279          /((?i)[\x{c0}])/8
280          /(?i:[\x{c0}])/8
281    
282        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
283        is now done differently.
284    
285     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
286        wrapper classes; (b) implement a new function in the C++ scanner that is
287        more efficient than the old way of doing things because it avoids levels of
288        recursion in the regex matching; (c) add a paragraph to the documentation
289        for the FullMatch() function.
290    
291     6. The escape sequence \n was being treated as whatever was defined as
292        "newline". Not only was this contrary to the documentation, which states
293        that \n is character 10 (hex 0A), but it also went horribly wrong when
294        "newline" was defined as CRLF. This has been fixed.
295    
296     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
297        was being set to -1 for the "end of line" case (supposedly a value that no
298        character can have). Though this value is never used (the check for end of
299        line is "zero bytes in current character"), it caused compiler complaints.
300        I've changed it to 0xffffffff.
301    
302     8. In pcre_version.c, the version string was being built by a sequence of
303        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
304        string (as it is for production releases) called a macro with an empty
305        argument. The C standard says the result of this is undefined. The gcc
306        compiler treats it as an empty string (which was what was wanted) but it is
307        reported that Visual C gives an error. The source has been hacked around to
308        avoid this problem.
309    
310     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
311        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
312        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
313        of them did).
314    
315    10. Originally, pcretest opened its input and output without "b"; then I was
316        told that "b" was needed in some environments, so it was added for release
317        5.0 to both the input and output. (It makes no difference on Unix-like
318        systems.) Later I was told that it is wrong for the input on Windows. I've
319        now abstracted the modes into two macros, to make it easier to fiddle with
320        them, and removed "b" from the input mode under Windows.
321    
322    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
323    
324    12. Added -help and --help to pcretest as an official way of being reminded
325        of the options.
326    
327    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
328        and pcrecpp.cc because they annoy compilers at high warning levels.
329    
330    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
331    
332    15. Fixed an occurrence of == in configure.ac that should have been = (shell
333        scripts are not C programs :-) and which was not noticed because it works
334        on Linux.
335    
336    16. pcretest is supposed to handle any length of pattern and data line (as one
337        line or as a continued sequence of lines) by extending its input buffer if
338        necessary. This feature was broken for very long pattern lines, leading to
339        a string of junk being passed to pcre_compile() if the pattern was longer
340        than about 50K.
341    
342    17. I have done a major re-factoring of the way pcre_compile() computes the
343        amount of memory needed for a compiled pattern. Previously, there was code
344        that made a preliminary scan of the pattern in order to do this. That was
345        OK when PCRE was new, but as the facilities have expanded, it has become
346        harder and harder to keep it in step with the real compile phase, and there
347        have been a number of bugs (see for example, 4 above). I have now found a
348        cunning way of running the real compile function in a "fake" mode that
349        enables it to compute how much memory it would need, while actually only
350        ever using a few hundred bytes of working memory and without too many
351        tests of the mode. This should make future maintenance and development
352        easier. A side effect of this work is that the limit of 200 on the nesting
353        depth of parentheses has been removed (though this was never a serious
354        limitation, I suspect). However, there is a downside: pcre_compile() now
355        runs more slowly than before (30% or more, depending on the pattern). I
356        hope this isn't a big issue. There is no effect on runtime performance.
357    
358    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
359        newline (only possible for the last line of a file) and it was a
360        pattern that set a locale (followed by /Lsomething), pcretest crashed.
361    
362    19. Added additional timing features to pcretest. (1) The -tm option now times
363        matching only, not compiling. (2) Both -t and -tm can be followed, as a
364        separate command line item, by a number that specifies the number of
365        repeats to use when timing. The default is 50000; this gives better
366        precision, but takes uncomfortably long for very large patterns.
367    
368    20. Extended pcre_study() to be more clever in cases where a branch of a
369        subpattern has no definite first character. For example, (a*|b*)[cd] would
370        previously give no result from pcre_study(). Now it recognizes that the
371        first character must be a, b, c, or d.
372    
373    21. There was an incorrect error "recursive call could loop indefinitely" if
374        a subpattern (or the entire pattern) that was being tested for matching an
375        empty string contained only one non-empty item after a nested subpattern.
376        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
377        incorrectly, because the \d was being skipped in the check.
378    
379    22. The pcretest program now has a new pattern option /B and a command line
380        option -b, which is equivalent to adding /B to every pattern. This causes
381        it to show the compiled bytecode, without the additional information that
382        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
383        is the same as /B/I).
384    
385    23. A new optimization is now able automatically to treat some sequences such
386        as a*b as a*+b. More specifically, if something simple (such as a character
387        or a simple class like \d) has an unlimited quantifier, and is followed by
388        something that cannot possibly match the quantified thing, the quantifier
389        is automatically "possessified".
390    
391    24. A recursive reference to a subpattern whose number was greater than 39
392        went wrong under certain circumstances in UTF-8 mode. This bug could also
393        have affected the operation of pcre_study().
394    
395    25. Realized that a little bit of performance could be had by replacing
396        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
397    
398    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
399    
400    27. Possessive quantifiers such as a++ were previously implemented by turning
401        them into atomic groups such as ($>a+). Now they have their own opcodes,
402        which improves performance. This includes the automatically created ones
403        from 23 above.
404    
405    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
406        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
407        the first matched character to be a colon. This applied both to named and
408        numbered groups.
409    
410    29. The ucpinternal.h header file was missing its idempotency #ifdef.
411    
412    30. I was sent a "project" file called libpcre.a.dev which I understand makes
413        building PCRE on Windows easier, so I have included it in the distribution.
414    
415    31. There is now a check in pcretest against a ridiculously large number being
416        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
417        loop, the loop is abandoned.
418    
419    32. Forward references to subpatterns in conditions such as (?(2)...) where
420        subpattern 2 is defined later cause pcre_compile() to search forwards in
421        the pattern for the relevant set of parentheses. This search went wrong
422        when there were unescaped parentheses in a character class, parentheses
423        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
424    
425    33. "Subroutine" calls and backreferences were previously restricted to
426        referencing subpatterns earlier in the regex. This restriction has now
427        been removed.
428    
429    34. Added a number of extra features that are going to be in Perl 5.10. On the
430        whole, these are just syntactic alternatives for features that PCRE had
431        previously implemented using the Python syntax or my own invention. The
432        other formats are all retained for compatibility.
433    
434        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
435            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
436            also .NET compatible.
437    
438        (b) A recursion or subroutine call to a named group can now be defined as
439            (?&name) as well as (?P>name).
440    
441        (c) A backreference to a named group can now be defined as \k<name> or
442            \k'name' as well as (?P=name). The new forms, as well as being in Perl
443            5.10, are also .NET compatible.
444    
445        (d) A conditional reference to a named group can now use the syntax
446            (?(<name>) or (?('name') as well as (?(name).
447    
448        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
449            groups (named and numbered) that are never evaluated inline, but can be
450            called as "subroutines" from elsewhere. In effect, the DEFINE condition
451            is always false. There may be only one alternative in such a group.
452    
453        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
454            as the simple (?(R). The condition is true only if the most recent
455            recursion is that of the given number or name. It does not search out
456            through the entire recursion stack.
457    
458        (g) The escape \gN or \g{N} has been added, where N is a positive or
459            negative number, specifying an absolute or relative reference.
460    
461    35. Tidied to get rid of some further signed/unsigned compiler warnings and
462        some "unreachable code" warnings.
463    
464    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
465        things, this adds five new scripts.
466    
467    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
468        There were also incompatibilities regarding the handling of \Q..\E inside
469        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
470        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
471    
472    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
473        matches an empty string, and forcibly breaks the loop. There were bugs in
474        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
475        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
476        separate and independent bugs (that affected different cases) have been
477        fixed.
478    
479    39. Refactored the code to abolish the use of different opcodes for small
480        capturing bracket numbers. This is a tidy that I avoided doing when I
481        removed the limit on the number of capturing brackets for 3.5 back in 2001.
482        The new approach is not only tidier, it makes it possible to reduce the
483        memory needed to fix the previous bug (38).
484    
485    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
486        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
487        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
488        mode.
489    
490    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
491        report.
492    
493    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
494        copy construction and assignment in the C++ wrapper.
495    
496    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
497        couple of bugs that could have given wrong results in the "--newline=crlf"
498        case.
499    
500    44. Added a number of casts and did some reorganization of signed/unsigned int
501        variables following suggestions from Dair Grant. Also renamed the variable
502        "this" as "item" because it is a C++ keyword.
503    
504    45. Arranged for dftables to add
505    
506          #include "pcre_internal.h"
507    
508        to pcre_chartables.c because without it, gcc 4.x may remove the array
509        definition from the final binary if PCRE is built into a static library and
510        dead code stripping is activated.
511    
512    46. For an unanchored pattern, if a match attempt fails at the start of a
513        newline sequence, and the newline setting is CRLF or ANY, and the next two
514        characters are CRLF, advance by two characters instead of one.
515    
516    
517    Version 6.7 04-Jul-06
518    ---------------------
519    
520     1. In order to handle tests when input lines are enormously long, pcretest has
521        been re-factored so that it automatically extends its buffers when
522        necessary. The code is crude, but this _is_ just a test program. The
523        default size has been increased from 32K to 50K.
524    
525     2. The code in pcre_study() was using the value of the re argument before
526        testing it for NULL. (Of course, in any sensible call of the function, it
527        won't be NULL.)
528    
529     3. The memmove() emulation function in pcre_internal.h, which is used on
530        systems that lack both memmove() and bcopy() - that is, hardly ever -
531        was missing a "static" storage class specifier.
532    
533     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
534        containing an extended class (one that cannot be represented by a bitmap
535        because it contains high-valued characters or Unicode property items, e.g.
536        [\pZ]). Almost always one would set UTF-8 mode when processing such a
537        pattern, but PCRE should not loop if you do not (it no longer does).
538        [Detail: two cases were found: (a) a repeated subpattern containing an
539        extended class; (b) a recursive reference to a subpattern that followed a
540        previous extended class. It wasn't skipping over the extended class
541        correctly when UTF-8 mode was not set.]
542    
543     5. A negated single-character class was not being recognized as fixed-length
544        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
545        compile error "lookbehind assertion is not fixed length".
546    
547     6. The RunPerlTest auxiliary script was showing an unexpected difference
548        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
549        write a Perl script that can interpret lines of an input file either as
550        byte characters or as UTF-8, which is what "perltest" was being required to
551        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
552        can't do is switch easily at run time between having the "use utf8;" pragma
553        or not. In the end, I fudged it by using the RunPerlTest script to insert
554        "use utf8;" explicitly for the UTF-8 tests.
555    
556     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
557        the end of the subject string, contrary to the documentation and to what
558        Perl does. This was true of both matching functions. Now it matches only at
559        the start of the subject and immediately after *internal* newlines.
560    
561     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
562        a pointer to an int instead of a pointer to an unsigned long int. This
563        caused problems on 64-bit systems.
564    
565     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
566        instance of the 'standard' template library not being so standard".
567    
568    10. There was no check on the number of named subpatterns nor the maximum
569        length of a subpattern name. The product of these values is used to compute
570        the size of the memory block for a compiled pattern. By supplying a very
571        long subpattern name and a large number of named subpatterns, the size
572        computation could be caused to overflow. This is now prevented by limiting
573        the length of names to 32 characters, and the number of named subpatterns
574        to 10,000.
575    
576    11. Subpatterns that are repeated with specific counts have to be replicated in
577        the compiled pattern. The size of memory for this was computed from the
578        length of the subpattern and the repeat count. The latter is limited to
579        65535, but there was no limit on the former, meaning that integer overflow
580        could in principle occur. The compiled length of a repeated subpattern is
581        now limited to 30,000 bytes in order to prevent this.
582    
583    12. Added the optional facility to have named substrings with the same name.
584    
585    13. Added the ability to use a named substring as a condition, using the
586        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
587        are numbers (not recommended). Forward references are permitted.
588    
589    14. Added forward references in named backreferences (if you see what I mean).
590    
591    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
592        pattern could run off the end of the subject. For example, the pattern
593        "(?s)(.{1,5})"8 did this with the subject "ab".
594    
595    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
596        PCRE_CASELESS was set when matching characters that were quantified with ?
597        or *.
598    
599    17. A character class other than a single negated character that had a minimum
600        but no maximum quantifier - for example [ab]{6,} - was not handled
601        correctly by pce_dfa_exec(). It would match only one character.
602    
603    18. A valid (though odd) pattern that looked like a POSIX character
604        class but used an invalid character after [ (for example [[,abc,]]) caused
605        pcre_compile() to give the error "Failed: internal error: code overflow" or
606        in some cases to crash with a glibc free() error. This could even happen if
607        the pattern terminated after [[ but there just happened to be a sequence of
608        letters, a binary zero, and a closing ] in the memory that followed.
609    
610    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
611        over the years. Originally (before any Unicode support), just the bottom 8
612        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
613        output from "man perlunicode" includes this:
614    
615          The regular expression compiler produces polymorphic opcodes.  That
616          is, the pattern adapts to the data and automatically switches to
617          the Unicode character scheme when presented with Unicode data--or
618          instead uses a traditional byte scheme when presented with byte
619          data.
620    
621        Sadly, a wide octal escape does not cause a switch, and in a string with
622        no other multibyte characters, these octal escapes are treated as before.
623        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
624        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
625        Unicode string.
626    
627        I have not perpetrated such confusion in PCRE. Up till now, it took just
628        the bottom 8 bits, as in old Perl. I have now made octal escapes with
629        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
630        translate to the appropriate multibyte character.
631    
632    29. Applied some refactoring to reduce the number of warnings from Microsoft
633        and Borland compilers. This has included removing the fudge introduced
634        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
635        a warning about an unused variable.
636    
637    21. PCRE has not included VT (character 0x0b) in the set of whitespace
638        characters since release 4.0, because Perl (from release 5.004) does not.
639        [Or at least, is documented not to: some releases seem to be in conflict
640        with the documentation.] However, when a pattern was studied with
641        pcre_study() and all its branches started with \s, PCRE still included VT
642        as a possible starting character. Of course, this did no harm; it just
643        caused an unnecessary match attempt.
644    
645    22. Removed a now-redundant internal flag bit that recorded the fact that case
646        dependency changed within the pattern. This was once needed for "required
647        byte" processing, but is no longer used. This recovers a now-scarce options
648        bit. Also moved the least significant internal flag bit to the most-
649        significant bit of the word, which was not previously used (hangover from
650        the days when it was an int rather than a uint) to free up another bit for
651        the future.
652    
653    23. Added support for CRLF line endings as well as CR and LF. As well as the
654        default being selectable at build time, it can now be changed at runtime
655        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
656        specify that it is scanning data with non-default line endings.
657    
658    24. Changed the definition of CXXLINK to make it agree with the definition of
659        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
660    
661    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
662        recursions. This makes a big different to stack usage for some patterns.
663    
664    26. If a subpattern containing a named recursion or subroutine reference such
665        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
666        the space required for the compiled pattern went wrong and gave too small a
667        value. Depending on the environment, this could lead to "Failed: internal
668        error: code overflow at offset 49" or "glibc detected double free or
669        corruption" errors.
670    
671    27. Applied patches from Google (a) to support the new newline modes and (b) to
672        advance over multibyte UTF-8 characters in GlobalReplace.
673    
674    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
675        difference for some implementation of PCRE in some Windows version.
676    
677    29. Added some extra testing facilities to pcretest:
678    
679        \q<number>   in a data line sets the "match limit" value
680        \Q<number>   in a data line sets the "match recursion limt" value
681        -S <number>  sets the stack size, where <number> is in megabytes
682    
683        The -S option isn't available for Windows.
684    
685    
686    Version 6.6 06-Feb-06
687    ---------------------
688    
689     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
690        in pcreposix.h. I have copied the definition from pcre.h.
691    
692     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
693        because pcre.h is no longer a built file.
694    
695     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
696        not normally included in the compiled code.
697    
698    
699    Version 6.5 01-Feb-06
700    ---------------------
701    
702     1. When using the partial match feature with pcre_dfa_exec(), it was not
703        anchoring the second and subsequent partial matches at the new starting
704        point. This could lead to incorrect results. For example, with the pattern
705        /1234/, partially matching against "123" and then "a4" gave a match.
706    
707     2. Changes to pcregrep:
708    
709        (a) All non-match returns from pcre_exec() were being treated as failures
710            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
711            error message is output. Some extra information is given for the
712            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
713            probably the only errors that are likely to be caused by users (by
714            specifying a regex that has nested indefinite repeats, for instance).
715            If there are more than 20 of these errors, pcregrep is abandoned.
716    
717        (b) A binary zero was treated as data while matching, but terminated the
718            output line if it was written out. This has been fixed: binary zeroes
719            are now no different to any other data bytes.
720    
721        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
722            used to set a locale for matching. The --locale=xxxx long option has
723            been added (no short equivalent) to specify a locale explicitly on the
724            pcregrep command, overriding the environment variables.
725    
726        (d) When -B was used with -n, some line numbers in the output were one less
727            than they should have been.
728    
729        (e) Added the -o (--only-matching) option.
730    
731        (f) If -A or -C was used with -c (count only), some lines of context were
732            accidentally printed for the final match.
733    
734        (g) Added the -H (--with-filename) option.
735    
736        (h) The combination of options -rh failed to suppress file names for files
737            that were found from directory arguments.
738    
739        (i) Added the -D (--devices) and -d (--directories) options.
740    
741        (j) Added the -F (--fixed-strings) option.
742    
743        (k) Allow "-" to be used as a file name for -f as well as for a data file.
744    
745        (l) Added the --colo(u)r option.
746    
747        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
748            is not present by default.
749    
750     3. A nasty bug was discovered in the handling of recursive patterns, that is,
751        items such as (?R) or (?1), when the recursion could match a number of
752        alternatives. If it matched one of the alternatives, but subsequently,
753        outside the recursion, there was a failure, the code tried to back up into
754        the recursion. However, because of the way PCRE is implemented, this is not
755        possible, and the result was an incorrect result from the match.
756    
757        In order to prevent this happening, the specification of recursion has
758        been changed so that all such subpatterns are automatically treated as
759        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
760    
761     4. I had overlooked the fact that, in some locales, there are characters for
762        which isalpha() is true but neither isupper() nor islower() are true. In
763        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
764        and ordfeminine) are like this. This affected the treatment of \w and \W
765        when they appeared in character classes, but not when they appeared outside
766        a character class. The bit map for "word" characters is now created
767        separately from the results of isalnum() instead of just taking it from the
768        upper, lower, and digit maps. (Plus the underscore character, of course.)
769    
770     5. The above bug also affected the handling of POSIX character classes such as
771        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
772        permanent tables. Instead, the bit maps for such a class were previously
773        created as the appropriate unions of the upper, lower, and digit bitmaps.
774        Now they are created by subtraction from the [[:word:]] class, which has
775        its own bitmap.
776    
777     6. The [[:blank:]] character class matches horizontal, but not vertical space.
778        It is created by subtracting the vertical space characters (\x09, \x0a,
779        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
780        subtraction was done in the overall bitmap for a character class, meaning
781        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
782        be recognized. This bug has been fixed.
783    
784     7. Patches from the folks at Google:
785    
786          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
787          real life, but is still worth protecting against".
788    
789          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
790          regular expressions".
791    
792          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
793          have it.
794    
795          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
796          "configure" and the latter not, in order to fix a problem somebody had
797          with compiling the Arg class on HP-UX.
798    
799          (e) Improve the error-handling of the C++ wrapper a little bit.
800    
801          (f) New tests for checking recursion limiting.
802    
803     8. The pcre_memmove() function, which is used only if the environment does not
804        have a standard memmove() function (and is therefore rarely compiled),
805        contained two bugs: (a) use of int instead of size_t, and (b) it was not
806        returning a result (though PCRE never actually uses the result).
807    
808     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
809        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
810        returned instead of calling malloc() with an overflowing number that would
811        most likely cause subsequent chaos.
812    
813    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
814    
815    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
816        with this option is matched, the nmatch and pmatch options of regexec() are
817        ignored.
818    
819    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
820        provided in case anyone wants to the the POSIX interface with UTF-8
821        strings.
822    
823    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
824        C++ linking (needed for some HP-UX environments).
825    
826    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
827        (unused parameter) and in the pcre_printint() function (omitted "default"
828        switch label when the default is to do nothing).
829    
830    15. Added some code to make it possible, when PCRE is compiled as a C++
831        library, to replace subject pointers for pcre_exec() with a smart pointer
832        class, thus making it possible to process discontinuous strings.
833    
834    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
835        much the same function. They were added by different people who were trying
836        to make PCRE easy to compile on non-Unix systems. It has been suggested
837        that PCRE_EXPORT be abolished now that there is more automatic apparatus
838        for compiling on Windows systems. I have therefore replaced it with
839        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
840        defaults to "extern" for C or "extern C" for C++, which works fine on
841        Unix-like systems. It is now possible to override the value of PCRE_DATA_
842        SCOPE with something explicit in config.h. In addition:
843    
844        (a) pcreposix.h still had just "extern" instead of either of these macros;
845            I have replaced it with PCRE_DATA_SCOPE.
846    
847        (b) Functions such as _pcre_xclass(), which are internal to the library,
848            but external in the C sense, all had PCRE_EXPORT in their definitions.
849            This is apparently wrong for the Windows case, so I have removed it.
850            (It makes no difference on Unix-like systems.)
851    
852    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
853        of recursive calls to match(). This is different to MATCH_LIMIT because
854        that limits the total number of calls to match(), not all of which increase
855        the depth of recursion. Limiting the recursion depth limits the amount of
856        stack (or heap if NO_RECURSE is set) that is used. The default can be set
857        when PCRE is compiled, and changed at run time. A patch from Google adds
858        this functionality to the C++ interface.
859    
860    18. Changes to the handling of Unicode character properties:
861    
862        (a) Updated the table to Unicode 4.1.0.
863    
864        (b) Recognize characters that are not in the table as "Cn" (undefined).
865    
866        (c) I revised the way the table is implemented to a much improved format
867            which includes recognition of ranges. It now supports the ranges that
868            are defined in UnicodeData.txt, and it also amalgamates other
869            characters into ranges. This has reduced the number of entries in the
870            table from around 16,000 to around 3,000, thus reducing its size
871            considerably. I realized I did not need to use a tree structure after
872            all - a binary chop search is just as efficient. Having reduced the
873            number of entries, I extended their size from 6 bytes to 8 bytes to
874            allow for more data.
875    
876        (d) Added support for Unicode script names via properties such as \p{Han}.
877    
878    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
879        matching that character.
880    
881    20. When matching a repeated Unicode property with a minimum greater than zero,
882        (for example \pL{2,}), PCRE could look past the end of the subject if it
883        reached it while seeking the minimum number of characters. This could
884        happen only if some of the characters were more than one byte long, because
885        there is a check for at least the minimum number of bytes.
886    
887    21. Refactored the implementation of \p and \P so as to be more general, to
888        allow for more different types of property in future. This has changed the
889        compiled form incompatibly. Anybody with saved compiled patterns that use
890        \p or \P will have to recompile them.
891    
892    22. Added "Any" and "L&" to the supported property types.
893    
894    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
895        but give a compile time error if the value is greater than 0xff.
896    
897    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
898        accidentally not being installed or uninstalled.
899    
900    25. The pcre.h file was built from pcre.h.in, but the only changes that were
901        made were to insert the current release number. This seemed silly, because
902        it made things harder for people building PCRE on systems that don't run
903        "configure". I have turned pcre.h into a distributed file, no longer built
904        by "configure", with the version identification directly included. There is
905        no longer a pcre.h.in file.
906    
907        However, this change necessitated a change to the pcre-config script as
908        well. It is built from pcre-config.in, and one of the substitutions was the
909        release number. I have updated configure.ac so that ./configure now finds
910        the release number by grepping pcre.h.
911    
912    26. Added the ability to run the tests under valgrind.
913    
914    
915    Version 6.4 05-Sep-05
916    ---------------------
917    
918     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
919        "--" to be printed when multiple files were scanned, even when none of the
920        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
921        consider it to be a bug, and have restored the previous behaviour.
922    
923     2. A couple of code tidies to get rid of compiler warnings.
924    
925     3. The pcretest program used to cheat by referring to symbols in the library
926        whose names begin with _pcre_. These are internal symbols that are not
927        really supposed to be visible externally, and in some environments it is
928        possible to suppress them. The cheating is now confined to including
929        certain files from the library's source, which is a bit cleaner.
930    
931     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
932        file's purpose clearer.
933    
934     5. Reorganized pcre_ucp_findchar().
935    
936    
937    Version 6.3 15-Aug-05
938    ---------------------
939    
940     1. The file libpcre.pc.in did not have general read permission in the tarball.
941    
942     2. There were some problems when building without C++ support:
943    
944        (a) If C++ support was not built, "make install" and "make test" still
945            tried to test it.
946    
947        (b) There were problems when the value of CXX was explicitly set. Some
948            changes have been made to try to fix these, and ...
949    
950        (c) --disable-cpp can now be used to explicitly disable C++ support.
951    
952        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
953            backslash in a target when C++ was disabled. This confuses some
954            versions of "make", apparently. Using an intermediate variable solves
955            this. (Same for CPP_LOBJ.)
956    
957     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
958        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
959        necessary on certain architectures.
960    
961     4. Added a setting of -export-symbols-regex to the link command to remove
962        those symbols that are exported in the C sense, but actually are local
963        within the library, and not documented. Their names all begin with
964        "_pcre_". This is not a perfect job, because (a) we have to except some
965        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
966        available (and never for static libraries). I have made a note to try to
967        find a way round (a) in the future.
968    
969    
970    Version 6.2 01-Aug-05
971    ---------------------
972    
973     1. There was no test for integer overflow of quantifier values. A construction
974        such as {1111111111111111} would give undefined results. What is worse, if
975        a minimum quantifier for a parenthesized subpattern overflowed and became
976        negative, the calculation of the memory size went wrong. This could have
977        led to memory overwriting.
978    
979     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
980    
981     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
982        operating environments where this matters.
983    
984     4. Applied Giuseppe Maxia's patch to add additional features for controlling
985        PCRE options from within the C++ wrapper.
986    
987     5. Named capturing subpatterns were not being correctly counted when a pattern
988        was compiled. This caused two problems: (a) If there were more than 100
989        such subpatterns, the calculation of the memory needed for the whole
990        compiled pattern went wrong, leading to an overflow error. (b) Numerical
991        back references of the form \12, where the number was greater than 9, were
992        not recognized as back references, even though there were sufficient
993        previous subpatterns.
994    
995     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
996        versions of gcc, e.g. 2.95.4.
997    
998    
999    Version 6.1 21-Jun-05
1000    ---------------------
1001    
1002     1. There was one reference to the variable "posix" in pcretest.c that was not
1003        surrounded by "#if !defined NOPOSIX".
1004    
1005     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1006        the cross-check on the old pcre_info() function, for the benefit of the
1007        cut-down version of PCRE that is currently imported into Exim.
1008    
1009     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1010        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1011        patterns that start (?i) but I don't think that matters. The use of (?i) is
1012        just an example; this all applies to the other options as well.
1013    
1014     4. Since libtool seems to echo the compile commands it is issuing, the output
1015        from "make" can be reduced a bit by putting "@" in front of each libtool
1016        compile command.
1017    
1018     5. Patch from the folks at Google for configure.in to be a bit more thorough
1019        in checking for a suitable C++ installation before trying to compile the
1020        C++ stuff. This should fix a reported problem when a compiler was present,
1021        but no suitable headers.
1022    
1023     6. The man pages all had just "PCRE" as their title. I have changed them to
1024        be the relevant file name. I have also arranged that these names are
1025        retained in the file doc/pcre.txt, which is a concatenation in text format
1026        of all the man pages except the little individual ones for each function.
1027    
1028     7. The NON-UNIX-USE file had not been updated for the different set of source
1029        files that come with release 6. I also added a few comments about the C++
1030        wrapper.
1031    
1032    
1033    Version 6.0 07-Jun-05
1034    ---------------------
1035    
1036     1. Some minor internal re-organization to help with my DFA experiments.
1037    
1038     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1039        didn't matter for the library itself when fully configured, but did matter
1040        when compiling without UCP support, or within Exim, where the ucp files are
1041        not imported.
1042    
1043     3. Refactoring of the library code to split up the various functions into
1044        different source modules. The addition of the new DFA matching code (see
1045        below) to a single monolithic source would have made it really too
1046        unwieldy, quite apart from causing all the code to be include in a
1047        statically linked application, when only some functions are used. This is
1048        relevant even without the DFA addition now that patterns can be compiled in
1049        one application and matched in another.
1050    
1051        The downside of splitting up is that there have to be some external
1052        functions and data tables that are used internally in different modules of
1053        the library but which are not part of the API. These have all had their
1054        names changed to start with "_pcre_" so that they are unlikely to clash
1055        with other external names.
1056    
1057     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1058        a different (DFA) algorithm. Although it is slower than the original
1059        function, it does have some advantages for certain types of matching
1060        problem.
1061    
1062     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1063        including restarting after a partial match.
1064    
1065     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1066        defined when compiling for Windows was sent to me. I have put it into the
1067        code, though I have no means of testing or verifying it.
1068    
1069     7. Added the pcre_refcount() auxiliary function.
1070    
1071     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1072        match before or at the first newline in the subject string. In pcretest,
1073        the /f option on a pattern can be used to set this.
1074    
1075     9. A repeated \w when used in UTF-8 mode with characters greater than 256
1076        would behave wrongly. This has been present in PCRE since release 4.0.
1077    
1078    10. A number of changes to the pcregrep command:
1079    
1080        (a) Refactored how -x works; insert ^(...)$ instead of setting
1081            PCRE_ANCHORED and checking the length, in preparation for adding
1082            something similar for -w.
1083    
1084        (b) Added the -w (match as a word) option.
1085    
1086        (c) Refactored the way lines are read and buffered so as to have more
1087            than one at a time available.
1088    
1089        (d) Implemented a pcregrep test script.
1090    
1091        (e) Added the -M (multiline match) option. This allows patterns to match
1092            over several lines of the subject. The buffering ensures that at least
1093            8K, or the rest of the document (whichever is the shorter) is available
1094            for matching (and similarly the previous 8K for lookbehind assertions).
1095    
1096        (f) Changed the --help output so that it now says
1097    
1098              -w, --word-regex(p)
1099    
1100            instead of two lines, one with "regex" and the other with "regexp"
1101            because that confused at least one person since the short forms are the
1102            same. (This required a bit of code, as the output is generated
1103            automatically from a table. It wasn't just a text change.)
1104    
1105        (g) -- can be used to terminate pcregrep options if the next thing isn't an
1106            option but starts with a hyphen. Could be a pattern or a path name
1107            starting with a hyphen, for instance.
1108    
1109        (h) "-" can be given as a file name to represent stdin.
1110    
1111        (i) When file names are being printed, "(standard input)" is used for
1112            the standard input, for compatibility with GNU grep. Previously
1113            "<stdin>" was used.
1114    
1115        (j) The option --label=xxx can be used to supply a name to be used for
1116            stdin when file names are being printed. There is no short form.
1117    
1118        (k) Re-factored the options decoding logic because we are going to add
1119            two more options that take data. Such options can now be given in four
1120            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1121    
1122        (l) Added the -A, -B, and -C options for requesting that lines of context
1123            around matches be printed.
1124    
1125        (m) Added the -L option to print the names of files that do not contain
1126            any matching lines, that is, the complement of -l.
1127    
1128        (n) The return code is 2 if any file cannot be opened, but pcregrep does
1129            continue to scan other files.
1130    
1131        (o) The -s option was incorrectly implemented. For compatibility with other
1132            greps, it now suppresses the error message for a non-existent or non-
1133            accessible file (but not the return code). There is a new option called
1134            -q that suppresses the output of matching lines, which was what -s was
1135            previously doing.
1136    
1137        (p) Added --include and --exclude options to specify files for inclusion
1138            and exclusion when recursing.
1139    
1140    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1141        Hopefully, it now does.
1142    
1143    12. Missing cast in pcre_study().
1144    
1145    13. Added an "uninstall" target to the makefile.
1146    
1147    14. Replaced "extern" in the function prototypes in Makefile.in with
1148        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1149        world, but is set differently for Windows.
1150    
1151    15. Added a second compiling function called pcre_compile2(). The only
1152        difference is that it has an extra argument, which is a pointer to an
1153        integer error code. When there is a compile-time failure, this is set
1154        non-zero, in addition to the error test pointer being set to point to an
1155        error message. The new argument may be NULL if no error number is required
1156        (but then you may as well call pcre_compile(), which is now just a
1157        wrapper). This facility is provided because some applications need a
1158        numeric error indication, but it has also enabled me to tidy up the way
1159        compile-time errors are handled in the POSIX wrapper.
1160    
1161    16. Added VPATH=.libs to the makefile; this should help when building with one
1162        prefix path and installing with another. (Or so I'm told by someone who
1163        knows more about this stuff than I do.)
1164    
1165    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1166        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1167        match everything, including newlines. This is not POSIX-compatible, but
1168        somebody wanted the feature. From pcretest it can be activated by using
1169        both the P and the s flags.
1170    
1171    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1172    
1173    19. libpcre.pc was being incorrectly installed as executable.
1174    
1175    20. A couple of places in pcretest check for end-of-line by looking for '\n';
1176        it now also looks for '\r' so that it will work unmodified on Windows.
1177    
1178    21. Added Google's contributed C++ wrapper to the distribution.
1179    
1180    22. Added some untidy missing memory free() calls in pcretest, to keep
1181        Electric Fence happy when testing.
1182    
1183    
1184    
1185    Version 5.0 13-Sep-04
1186    ---------------------
1187    
1188     1. Internal change: literal characters are no longer packed up into items
1189        containing multiple characters in a single byte-string. Each character
1190        is now matched using a separate opcode. However, there may be more than one
1191        byte in the character in UTF-8 mode.
1192    
1193     2. The pcre_callout_block structure has two new fields: pattern_position and
1194        next_item_length. These contain the offset in the pattern to the next match
1195        item, and its length, respectively.
1196    
1197     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1198        insertion of callouts before each pattern item. Added the /C option to
1199        pcretest to make use of this.
1200    
1201     4. On the advice of a Windows user, the lines
1202    
1203          #if defined(_WIN32) || defined(WIN32)
1204          _setmode( _fileno( stdout ), 0x8000 );
1205          #endif  /* defined(_WIN32) || defined(WIN32) */
1206    
1207        have been added to the source of pcretest. This apparently does useful
1208        magic in relation to line terminators.
1209    
1210     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1211        for the benefit of those environments where the "b" makes a difference.
1212    
1213     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1214        to know about it. I have put a hack into configure.in that adds in code
1215        to set GCC=yes if CC=icc. This seems to end up at a point in the
1216        generated configure script that is early enough to affect the setting of
1217        compiler options, which is what is needed, but I have no means of testing
1218        whether it really works. (The user who reported this had patched the
1219        generated configure script, which of course I cannot do.)
1220    
1221        LATER: After change 22 below (new libtool files), the configure script
1222        seems to know about icc (and also ecc). Therefore, I have commented out
1223        this hack in configure.in.
1224    
1225     7. Added support for pkg-config (2 patches were sent in).
1226    
1227     8. Negated POSIX character classes that used a combination of internal tables
1228        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1229        [[:^ascii]]. Typically, they would match almost any characters. The other
1230        POSIX classes were not broken in this way.
1231    
1232     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1233        to find the match, as PCRE was deluded into thinking that the match had to
1234        start at the start point or following a newline. The same bug applied to
1235        patterns with negative forward assertions or any backward assertions
1236        preceding ".*" at the start, unless the pattern required a fixed first
1237        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1238    
1239    10. In UTF-8 mode, when moving forwards in the subject after a failed match
1240        starting at the last subject character, bytes beyond the end of the subject
1241        string were read.
1242    
1243    11. Renamed the variable "class" as "classbits" to make life easier for C++
1244        users. (Previously there was a macro definition, but it apparently wasn't
1245        enough.)
1246    
1247    12. Added the new field "tables" to the extra data so that tables can be passed
1248        in at exec time, or the internal tables can be re-selected. This allows
1249        a compiled regex to be saved and re-used at a later time by a different
1250        program that might have everything at different addresses.
1251    
1252    13. Modified the pcre-config script so that, when run on Solaris, it shows a
1253        -R library as well as a -L library.
1254    
1255    14. The debugging options of pcretest (-d on the command line or D on a
1256        pattern) showed incorrect output for anything following an extended class
1257        that contained multibyte characters and which was followed by a quantifier.
1258    
1259    15. Added optional support for general category Unicode character properties
1260        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1261        support. It adds about 90K to the size of the library. The meanings of the
1262        inbuilt class escapes such as \d and \s have NOT been changed.
1263    
1264    16. Updated pcredemo.c to include calls to free() to release the memory for the
1265        compiled pattern.
1266    
1267    17. The generated file chartables.c was being created in the source directory
1268        instead of in the building directory. This caused the build to fail if the
1269        source directory was different from the building directory, and was
1270        read-only.
1271    
1272    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1273        file. No doubt somebody will tell me if they don't make sense... Also added
1274        Dan Mooney's comments about building on OpenVMS.
1275    
1276    19. Added support for partial matching via the PCRE_PARTIAL option for
1277        pcre_exec() and the \P data escape in pcretest.
1278    
1279    20. Extended pcretest with 3 new pattern features:
1280    
1281        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
1282              write the compiled pattern to the file whose name is "rest-of-line".
1283              This is a straight binary dump of the data, with the saved pointer to
1284              the character tables forced to be NULL. The study data, if any, is
1285              written too. After writing, pcretest reads a new pattern.
1286    
1287        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1288              compiled pattern from the given file. There must not be any
1289              occurrences of "<" in the file name (pretty unlikely); if there are,
1290              pcretest will instead treat the initial "<" as a pattern delimiter.
1291              After reading in the pattern, pcretest goes on to read data lines as
1292              usual.
1293    
1294        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1295              and 16-bit fields in a compiled pattern, to simulate a pattern that
1296              was compiled on a host of opposite endianness.
1297    
1298    21. The pcre-exec() function can now cope with patterns that were compiled on
1299        hosts of opposite endianness, with this restriction:
1300    
1301          As for any compiled expression that is saved and used later, the tables
1302          pointer field cannot be preserved; the extra_data field in the arguments
1303          to pcre_exec() should be used to pass in a tables address if a value
1304          other than the default internal tables were used at compile time.
1305    
1306    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1307        now diagnosed as an error. Previously, most of the time, a negative number
1308        would have been treated as zero, but if in addition "ovector" was passed as
1309        NULL, a crash could occur.
1310    
1311    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1312        new versions from the libtool 1.5 distribution (the last one is a copy of
1313        a file called libtool.m4). This seems to have fixed the need to patch
1314        "configure" to support Darwin 1.3 (which I used to do). However, I still
1315        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1316        workstation).
1317    
1318    24. Changed the PCRE licence to be the more standard "BSD" licence.
1319    
1320    
1321    Version 4.5 01-Dec-03
1322    ---------------------
1323    
1324     1. There has been some re-arrangement of the code for the match() function so
1325        that it can be compiled in a version that does not call itself recursively.
1326        Instead, it keeps those local variables that need separate instances for
1327        each "recursion" in a frame on the heap, and gets/frees frames whenever it
1328        needs to "recurse". Keeping track of where control must go is done by means
1329        of setjmp/longjmp. The whole thing is implemented by a set of macros that
1330        hide most of the details from the main code, and operates only if
1331        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1332        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1333        operating.
1334    
1335        To make it easier for callers to provide specially tailored get/free
1336        functions for this usage, two new functions, pcre_stack_malloc, and
1337        pcre_stack_free, are used. They are always called in strict stacking order,
1338        and the size of block requested is always the same.
1339    
1340        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1341        PCRE has been compiled to use the stack or the heap for recursion. The
1342        -C option of pcretest uses this to show which version is compiled.
1343    
1344        A new data escape \S, is added to pcretest; it causes the amounts of store
1345        obtained and freed by both kinds of malloc/free at match time to be added
1346        to the output.
1347    
1348     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1349        what's available on my current Linux desktop machine.
1350    
1351     3. When matching a UTF-8 string, the test for a valid string at the start has
1352        been extended. If start_offset is not zero, PCRE now checks that it points
1353        to a byte that is the start of a UTF-8 character. If not, it returns
1354        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1355        this is necessary because there may be backward assertions in the pattern.
1356        When matching the same subject several times, it may save resources to use
1357        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1358    
1359     4. The code for checking the validity of UTF-8 strings has been tightened so
1360        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1361        containing "overlong sequences".
1362    
1363     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1364        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1365        should have been "&", but it just so happened that all the cases this let
1366        through by mistake were picked up later in the function.
1367    
1368     6. I had used a variable called "isblank" - this is a C99 function, causing
1369        some compilers to warn. To avoid this, I renamed it (as "blankclass").
1370    
1371     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1372        prompting; (b) run "./pcretest /dev/null" at the start of the test script
1373        so the version is shown; (c) stop "make test" echoing "./RunTest".
1374    
1375     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1376    
1377     9. The prototype for memmove() for systems that don't have it was using
1378        size_t, but the inclusion of the header that defines size_t was later. I've
1379        moved the #includes for the C headers earlier to avoid this.
1380    
1381    10. Added some adjustments to the code to make it easier to compiler on certain
1382        special systems:
1383    
1384          (a) Some "const" qualifiers were missing.
1385          (b) Added the macro EXPORT before all exported functions; by default this
1386              is defined to be empty.
1387          (c) Changed the dftables auxiliary program (that builds chartables.c) so
1388              that it reads its output file name as an argument instead of writing
1389              to the standard output and assuming this can be redirected.
1390    
1391    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
1392        class containing characters with values greater than 255, PCRE compilation
1393        went into a loop.
1394    
1395    12. A recursive reference to a subpattern that was within another subpattern
1396        that had a minimum quantifier of zero caused PCRE to crash. For example,
1397        (x(y(?2))z)? provoked this bug with a subject that got as far as the
1398        recursion. If the recursively-called subpattern itself had a zero repeat,
1399        that was OK.
1400    
1401    13. In pcretest, the buffer for reading a data line was set at 30K, but the
1402        buffer into which it was copied (for escape processing) was still set at
1403        1024, so long lines caused crashes.
1404    
1405    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
1406        "internal error: code overflow...". This applied to any character class
1407        that was followed by a possessive quantifier.
1408    
1409    15. Modified the Makefile to add libpcre.la as a prerequisite for
1410        libpcreposix.la because I was told this is needed for a parallel build to
1411        work.
1412    
1413    16. If a pattern that contained .* following optional items at the start was
1414        studied, the wrong optimizing data was generated, leading to matching
1415        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
1416        matching string must start with a or b or c. The correct conclusion for
1417        this pattern is that a match can start with any character.
1418    
1419    
1420    Version 4.4 13-Aug-03
1421    ---------------------
1422    
1423     1. In UTF-8 mode, a character class containing characters with values between
1424        127 and 255 was not handled correctly if the compiled pattern was studied.
1425        In fixing this, I have also improved the studying algorithm for such
1426        classes (slightly).
1427    
1428     2. Three internal functions had redundant arguments passed to them. Removal
1429        might give a very teeny performance improvement.
1430    
1431     3. Documentation bug: the value of the capture_top field in a callout is *one
1432        more than* the number of the hightest numbered captured substring.
1433    
1434     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
1435        in incorrectly linking with a previously installed version. They now link
1436        explicitly with libpcre.la.
1437    
1438     5. configure.in no longer needs to recognize Cygwin specially.
1439    
1440     6. A problem in pcre.in for Windows platforms is fixed.
1441    
1442     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
1443        pcretest, it used to include the size of the study block as part of its
1444        output. Unfortunately, the structure contains a field that has a different
1445        size on different hardware architectures. This meant that the tests that
1446        showed this size failed. As the block is currently always of a fixed size,
1447        this information isn't actually particularly useful in pcretest output, so
1448        I have just removed it.
1449    
1450     8. Three pre-processor statements accidentally did not start in column 1.
1451        Sadly, there are *still* compilers around that complain, even though
1452        standard C has not required this for well over a decade. Sigh.
1453    
1454     9. In pcretest, the code for checking callouts passed small integers in the
1455        callout_data field, which is a void * field. However, some picky compilers
1456        complained about the casts involved for this on 64-bit systems. Now
1457        pcretest passes the address of the small integer instead, which should get
1458        rid of the warnings.
1459    
1460    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
1461        both compile and run time, and gives an error if an invalid UTF-8 sequence
1462        is found. There is a option for disabling this check in cases where the
1463        string is known to be correct and/or the maximum performance is wanted.
1464    
1465    11. In response to a bug report, I changed one line in Makefile.in from
1466    
1467            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
1468        to
1469            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
1470    
1471        to look similar to other lines, but I have no way of telling whether this
1472        is the right thing to do, as I do not use Windows. No doubt I'll get told
1473        if it's wrong...
1474    
1475    
1476    Version 4.3 21-May-03
1477    ---------------------
1478    
1479    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
1480       Makefile.
1481    
1482    2. Some refactoring to improve the quality of the code:
1483    
1484       (i)   The utf8_table... variables are now declared "const".
1485    
1486       (ii)  The code for \cx, which used the "case flipping" table to upper case
1487             lower case letters, now just substracts 32. This is ASCII-specific,
1488             but the whole concept of \cx is ASCII-specific, so it seems
1489             reasonable.
1490    
1491       (iii) PCRE was using its character types table to recognize decimal and
1492             hexadecimal digits in the pattern. This is silly, because it handles
1493             only 0-9, a-f, and A-F, but the character types table is locale-
1494             specific, which means strange things might happen. A private
1495             table is now used for this - though it costs 256 bytes, a table is
1496             much faster than multiple explicit tests. Of course, the standard
1497             character types table is still used for matching digits in subject
1498             strings against \d.
1499    
1500       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
1501             ending in _t are). So I've renamed it as ESC_tee.
1502    
1503    3. The first argument for regexec() in the POSIX wrapper should have been
1504       defined as "const".
1505    
1506    4. Changed pcretest to use malloc() for its buffers so that they can be
1507       Electric Fenced for debugging.
1508    
1509    5. There were several places in the code where, in UTF-8 mode, PCRE would try
1510       to read one or more bytes before the start of the subject string. Often this
1511       had no effect on PCRE's behaviour, but in some circumstances it could
1512       provoke a segmentation fault.
1513    
1514    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
1515       to try to read one or more bytes before the start of the subject string.
1516    
1517    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
1518       UTF-8 support could misbehave in various ways if the subject string
1519       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
1520       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
1521       back over UTF-8 characters.)
1522    
1523    
1524    Version 4.2 14-Apr-03
1525    ---------------------
1526    
1527    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
1528    
1529    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
1530         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
1531         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
1532         [WIN_PREFIX]: new variable, "cyg" for Cygwin
1533         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
1534           and BUILD_EXEEXT
1535         Note: automatic setting of the BUILD variables is not yet working
1536         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
1537           compile-time but not at link-time
1538         [LINK]: use for linking executables only
1539         make different versions for Windows and non-Windows
1540         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
1541           libraries
1542         [LINK_FOR_BUILD]: new variable
1543         [OBJEXT]: use throughout
1544         [EXEEXT]: use throughout
1545         <winshared>: new target
1546         <wininstall>: new target
1547         <dftables.o>: use native compiler
1548         <dftables>: use native linker
1549         <install>: handle Windows platform correctly
1550         <clean>: ditto
1551         <check>: ditto
1552         copy DLL to top builddir before testing
1553    
1554       As part of these changes, -no-undefined was removed again. This was reported
1555       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
1556       in any case.
1557    
1558    3. Some tidies to get rid of compiler warnings:
1559    
1560       . In the match_data structure, match_limit was an unsigned long int, whereas
1561         match_call_count was an int. I've made them both unsigned long ints.
1562    
1563       . In pcretest the fact that a const uschar * doesn't automatically cast to
1564         a void * provoked a warning.
1565    
1566       . Turning on some more compiler warnings threw up some "shadow" variables
1567         and a few more missing casts.
1568    
1569    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
1570       option, a class that contained a single character with a value between 128
1571       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
1572    
1573    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
1574       option, a class that contained several characters, but with at least one
1575       whose value was between 128 and 255 caused PCRE to crash.
1576    
1577    
1578    Version 4.1 12-Mar-03
1579    ---------------------
1580    
1581    1. Compiling with gcc -pedantic found a couple of places where casts were
1582    needed, and a string in dftables.c that was longer than standard compilers are
1583    required to support.
1584    
1585    2. Compiling with Sun's compiler found a few more places where the code could
1586    be tidied up in order to avoid warnings.
1587    
1588    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
1589    first of these names is deprecated in the latest Autoconf in favour of the name
1590    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
1591    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
1592    analogy I have changed it to CFLAGS_FOR_BUILD.
1593    
1594    4. Added -no-undefined to the linking command in the Makefile, because this is
1595    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
1596    linking step for the pcreposix library.
1597    
1598    5. PCRE was failing to diagnose the case of two named groups with the same
1599    name.
1600    
1601    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
1602    literal character that is needed in the subject for a match, and scans along to
1603    ensure that it is present before embarking on the full matching process. This
1604    saves time in cases of nested unlimited repeats that are never going to match.
1605    Problem: the scan can take a lot of time if the subject is very long (e.g.
1606    megabytes), thus penalizing straightforward matches. It is now done only if the
1607    amount of subject to be scanned is less than 1000 bytes.
1608    
1609    7. A lesser problem with the same optimization is that it was recording the
1610    first character of an anchored pattern as "needed", thus provoking a search
1611    right along the subject, even when the first match of the pattern was going to
1612    fail. The "needed" character is now not set for anchored patterns, unless it
1613    follows something in the pattern that is of non-fixed length. Thus, it still
1614    fulfils its original purpose of finding quick non-matches in cases of nested
1615    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
1616    
1617    
1618    Version 4.0 17-Feb-03
1619    ---------------------
1620    
1621    1. If a comment in an extended regex that started immediately after a meta-item
1622    extended to the end of string, PCRE compiled incorrect data. This could lead to
1623    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
1624    
1625    2. Moved to autoconf 2.53 and libtool 1.4.2.
1626    
1627    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
1628    the special perltest8 script is no longer needed - all the tests can be run
1629    from a single perltest script.
1630    
1631    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
1632    by \s. It has now been removed in PCRE. This means it isn't recognized as
1633    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
1634    class [:space:] *does* include VT, thereby creating a mess.
1635    
1636    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
1637    space and tab.
1638    
1639    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
1640    its new features into the main test script, reducing the number of scripts.
1641    
1642    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
1643    were backward compatible, and made the (?i) apply to the whole pattern, as if
1644    /i were given. Now it behaves more logically, and applies the option setting
1645    only to what follows. PCRE has been changed to follow suit. However, if it
1646    finds options settings right at the start of the pattern, it extracts them into
1647    the global options, as before. Thus, they show up in the info data.
1648    
1649    8. Added support for the \Q...\E escape sequence. Characters in between are
1650    treated as literals. This is slightly different from Perl in that $ and @ are
1651    also handled as literals inside the quotes. In Perl, they will cause variable
1652    interpolation. Note the following examples:
1653    
1654        Pattern            PCRE matches      Perl matches
1655    
1656        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
1657        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
1658        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
1659    
1660    For compatibility with Perl, \Q...\E sequences are recognized inside character
1661    classes as well as outside them.
1662    
1663    9. Re-organized 3 code statements in pcretest to avoid "overflow in
1664    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
1665    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
1666    signed/unsigned warnings.
1667    
1668    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
1669    option for pcretest, so I've replaced it by a simple function that does just
1670    that job.
1671    
1672    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
1673    "pcregrep -".
1674    
1675    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
1676    Java package. This provides some syntactic sugar for simple cases of what my
1677    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
1678    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
1679    item, you can use this simplified notation. Note that only makes sense with
1680    greedy quantifiers. Consequently, the use of the possessive quantifier forces
1681    greediness, whatever the setting of the PCRE_UNGREEDY option.
1682    
1683    13. A change of greediness default within a pattern was not taking effect at
1684    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
1685    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
1686    was abstracted outside.
1687    
1688    14. PCRE now supports the \G assertion. It is true when the current matching
1689    position is at the start point of the match. This differs from \A when the
1690    starting offset is non-zero. Used with the /g option of pcretest (or similar
1691    code), it works in the same way as it does for Perl's /g option. If all
1692    alternatives of a regex begin with \G, the expression is anchored to the start
1693    match position, and the "anchored" flag is set in the compiled expression.
1694    
1695    15. Some bugs concerning the handling of certain option changes within patterns
1696    have been fixed. These applied to options other than (?ims). For example,
1697    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
1698    been the other way round. Some of this was related to change 7 above.
1699    
1700    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
1701    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
1702    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
1703    POSIX classes only within a class (e.g. /[[:alpha:]]/).
1704    
1705    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
1706    mode. Unlike ".", it always matches newline, whatever the setting of
1707    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
1708    assertions. Perl allows it, but it doesn't (in general) work because it can't
1709    calculate the length of the lookbehind. At least, that's the case for Perl
1710    5.8.0 - I've been told they are going to document that it doesn't work in
1711    future.
1712    
1713    18. Added an error diagnosis for escapes that PCRE does not support: these are
1714    \L, \l, \N, \P, \p, \U, \u, and \X.
1715    
1716    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
1717    reading past the end of the pattern in cases such as /[abcd/.
1718    
1719    20. PCRE was getting more memory than necessary for patterns with classes that
1720    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
1721    
1722    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
1723    compiling PCRE for use with Virtual Pascal.
1724    
1725    22. Small fix to the Makefile to make it work properly if the build is done
1726    outside the source tree.
1727    
1728    23. Added a new extension: a condition to go with recursion. If a conditional
1729    subpattern starts with (?(R) the "true" branch is used if recursion has
1730    happened, whereas the "false" branch is used only at the top level.
1731    
1732    24. When there was a very long string of literal characters (over 255 bytes
1733    without UTF support, over 250 bytes with UTF support), the computation of how
1734    much memory was required could be incorrect, leading to segfaults or other
1735    strange effects.
1736    
1737    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
1738    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
1739    there was a subsequent back reference to those brackets. This meant that, for
1740    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
1741    possible to check for precisely this case. All we can do is abandon the
1742    optimization if .* occurs inside capturing brackets when there are any back
1743    references whatsoever. (See below for a better fix that came later.)
1744    
1745    26. The handling of the optimization for finding the first character of a
1746    non-anchored pattern, and for finding a character that is required later in the
1747    match were failing in some cases. This didn't break the matching; it just
1748    failed to optimize when it could. The way this is done has been re-implemented.
1749    
1750    27. Fixed typo in error message for invalid (?R item (it said "(?p").
1751    
1752    28. Added a new feature that provides some of the functionality that Perl
1753    provides with (?{...}). The facility is termed a "callout". The way it is done
1754    in PCRE is for the caller to provide an optional function, by setting
1755    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
1756    global variable. By default it is unset, which disables all calling out. To get
1757    the function called, the regex must include (?C) at appropriate points. This
1758    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
1759    This provides a means of identifying different callout points. When PCRE
1760    reaches such a point in the regex, if pcre_callout has been set, the external
1761    function is called. It is provided with data in a structure called
1762    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
1763    matching continues; if it returns a non-zero value, the match at the current
1764    point fails. However, backtracking will occur if possible. [This was changed
1765    later and other features added - see item 49 below.]
1766    
1767    29. pcretest is upgraded to test the callout functionality. It provides a
1768    callout function that displays information. By default, it shows the start of
1769    the match and the current position in the text. There are some new data escapes
1770    to vary what happens:
1771    
1772        \C+         in addition, show current contents of captured substrings
1773        \C-         do not supply a callout function
1774        \C!n        return 1 when callout number n is reached
1775        \C!n!m      return 1 when callout number n is reached for the mth time
1776    
1777    30. If pcregrep was called with the -l option and just a single file name, it
1778    output "<stdin>" if a match was found, instead of the file name.
1779    
1780    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
1781    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
1782    pcre_exec(). This saves a malloc/free per call. The default value of
1783    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
1784    when configuring.
1785    
1786    32. The default maximum size of a compiled pattern is 64K. There have been a
1787    few cases of people hitting this limit. The code now uses macros to handle the
1788    storing of links as offsets within the compiled pattern. It defaults to 2-byte
1789    links, but this can be changed to 3 or 4 bytes by --with-link-size when
1790    configuring. Tests 2 and 5 work only with 2-byte links because they output
1791    debugging information about compiled patterns.
1792    
1793    33. Internal code re-arrangements:
1794    
1795    (a) Moved the debugging function for printing out a compiled regex into
1796        its own source file (printint.c) and used #include to pull it into
1797        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
1798        separate copies.
1799    
1800    (b) Defined the list of op-code names for debugging as a macro in
1801        internal.h so that it is next to the definition of the opcodes.
1802    
1803    (c) Defined a table of op-code lengths for simpler skipping along compiled
1804        code. This is again a macro in internal.h so that it is next to the
1805        definition of the opcodes.
1806    
1807    34. Added support for recursive calls to individual subpatterns, along the
1808    lines of Robin Houston's patch (but implemented somewhat differently).
1809    
1810    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
1811    allow it to read and process whole directories in Win32. This code was
1812    contributed by Lionel Fourquaux; it has not been tested by me.
1813    
1814    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
1815    used to name a group. Names consist of alphanumerics and underscores, and must
1816    be unique. Back references use the syntax (?P=name) and recursive calls use
1817    (?P>name) which is a PCRE extension to the Python extension. Groups still have
1818    numbers. The function pcre_fullinfo() can be used after compilation to extract
1819    a name/number map. There are three relevant calls:
1820    
1821      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
1822      PCRE_INFO_NAMECOUNT            yields the number of entries
1823      PCRE_INFO_NAMETABLE            yields a pointer to the map.
1824    
1825    The map is a vector of fixed-size entries. The size of each entry depends on
1826    the length of the longest name used. The first two bytes of each entry are the
1827    group number, most significant byte first. There follows the corresponding
1828    name, zero terminated. The names are in alphabetical order.
1829    
1830    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
1831    case instead of 255. Making it the same both with and without UTF-8 support
1832    means that the same test output works with both.
1833    
1834    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
1835    calling malloc() with a zero argument.
1836    
1837    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
1838    optimization. I've improved things by keeping a bitmap of backreferences with
1839    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
1840    fact referenced, the optimization can be applied. It is unlikely that a
1841    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
1842    the match to follow \n) will appear inside brackets with a number greater than
1843    31, but if it does, any back reference > 31 suppresses the optimization.
1844    
1845    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
1846    of disabling numbered capturing parentheses. Any opening parenthesis that is
1847    not followed by ? behaves as if it were followed by ?: but named parentheses
1848    can still be used for capturing (and they will acquire numbers in the usual
1849    way).
1850    
1851    41. Redesigned the return codes from the match() function into yes/no/error so
1852    that errors can be passed back from deep inside the nested calls. A malloc
1853    failure while inside a recursive subpattern call now causes the
1854    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
1855    
1856    42. It is now possible to set a limit on the number of times the match()
1857    function is called in a call to pcre_exec(). This facility makes it possible to
1858    limit the amount of recursion and backtracking, though not in a directly
1859    obvious way, because the match() function is used in a number of different
1860    circumstances. The count starts from zero for each position in the subject
1861    string (for non-anchored patterns). The default limit is, for compatibility, a
1862    large number, namely 10 000 000. You can change this in two ways:
1863    
1864    (a) When configuring PCRE before making, you can use --with-match-limit=n
1865        to set a default value for the compiled library.
1866    
1867    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
1868        a different value is set. See 45 below.
1869    
1870    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
1871    
1872    43. Added a new function pcre_config(int, void *) to enable run-time extraction
1873    of things that can be changed at compile time. The first argument specifies
1874    what is wanted and the second points to where the information is to be placed.
1875    The current list of available information is:
1876    
1877      PCRE_CONFIG_UTF8
1878    
1879    The output is an integer that is set to one if UTF-8 support is available;
1880    otherwise it is set to zero.
1881    
1882      PCRE_CONFIG_NEWLINE
1883    
1884    The output is an integer that it set to the value of the code that is used for
1885    newline. It is either LF (10) or CR (13).
1886    
1887      PCRE_CONFIG_LINK_SIZE
1888    
1889    The output is an integer that contains the number of bytes used for internal
1890    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
1891    
1892      PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
1893    
1894    The output is an integer that contains the threshold above which the POSIX
1895    interface uses malloc() for output vectors. See item 31 above.
1896    
1897      PCRE_CONFIG_MATCH_LIMIT
1898    
1899    The output is an unsigned integer that contains the default limit of the number
1900    of match() calls in a pcre_exec() execution. See 42 above.
1901    
1902    44. pcretest has been upgraded by the addition of the -C option. This causes it
1903    to extract all the available output from the new pcre_config() function, and to
1904    output it. The program then exits immediately.
1905    
1906    45. A need has arisen to pass over additional data with calls to pcre_exec() in
1907    order to support additional features. One way would have been to define
1908    pcre_exec2() (for example) with extra arguments, but this would not have been
1909    extensible, and would also have required all calls to the original function to
1910    be mapped to the new one. Instead, I have chosen to extend the mechanism that
1911    is used for passing in "extra" data from pcre_study().
1912    
1913    The pcre_extra structure is now exposed and defined in pcre.h. It currently
1914    contains the following fields:
1915    
1916      flags         a bitmap indicating which of the following fields are set
1917      study_data    opaque data from pcre_study()
1918      match_limit   a way of specifying a limit on match() calls for a specific
1919                      call to pcre_exec()
1920      callout_data  data for callouts (see 49 below)
1921    
1922    The flag bits are also defined in pcre.h, and are
1923    
1924      PCRE_EXTRA_STUDY_DATA
1925      PCRE_EXTRA_MATCH_LIMIT
1926      PCRE_EXTRA_CALLOUT_DATA
1927    
1928    The pcre_study() function now returns one of these new pcre_extra blocks, with
1929    the actual study data pointed to by the study_data field, and the
1930    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
1931    before. That is, this change is entirely upwards-compatible and requires no
1932    change to existing code.
1933    
1934    If you want to pass in additional data to pcre_exec(), you can either place it
1935    in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
1936    block.
1937    
1938    46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
1939    data string contains the escape sequence \M, pcretest calls pcre_exec() several
1940    times with different match limits, until it finds the minimum value needed for
1941    pcre_exec() to complete. The value is then output. This can be instructive; for
1942    most simple matches the number is quite small, but for pathological cases it
1943    gets very large very quickly.
1944    
1945    47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
1946    returns the size of the data block pointed to by the study_data field in a
1947    pcre_extra block, that is, the value that was passed as the argument to
1948    pcre_malloc() when PCRE was getting memory in which to place the information
1949    created by pcre_study(). The fourth argument should point to a size_t variable.
1950    pcretest has been extended so that this information is shown after a successful
1951    pcre_study() call when information about the compiled regex is being displayed.
1952    
1953    48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
1954    because what follows is always an absolute path. (Later: it turns out that this
1955    is more than cosmetic for MinGW, because it doesn't like empty path
1956    components.)
1957    
1958    49. Some changes have been made to the callout feature (see 28 above):
1959    
1960    (i)  A callout function now has three choices for what it returns:
1961    
1962           0  =>  success, carry on matching
1963         > 0  =>  failure at this point, but backtrack if possible
1964         < 0  =>  serious error, return this value from pcre_exec()
1965    
1966         Negative values should normally be chosen from the set of PCRE_ERROR_xxx
1967         values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
1968         "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
1969         use by callout functions. It will never be used by PCRE itself.
1970    
1971    (ii) The pcre_extra structure (see 45 above) has a void * field called
1972         callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
1973         pcre_callout_block structure has a field of the same name. The contents of
1974         the field passed in the pcre_extra structure are passed to the callout
1975         function in the corresponding field in the callout block. This makes it
1976         easier to use the same callout-containing regex from multiple threads. For
1977         testing, the pcretest program has a new data escape
1978    
1979           \C*n        pass the number n (may be negative) as callout_data
1980    
1981         If the callout function in pcretest receives a non-zero value as
1982         callout_data, it returns that value.
1983    
1984    50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
1985    there were some redundant $(CFLAGS) in commands that are now specified as
1986    $(LINK), which already includes $(CFLAGS).
1987    
1988    51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
1989    has been compiled with UTF-8 support *and* pcre_compile() has been compiled
1990    with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
1991    one-byte characters throughout. Note that case-insensitive matching applies
1992    only to characters whose values are less than 256. PCRE doesn't support the
1993    notion of cases for higher-valued characters.
1994    
1995    (i)   A character class whose characters are all within 0-255 is handled as
1996          a bit map, and the map is inverted for negative classes. Previously, a
1997          character > 255 always failed to match such a class; however it should
1998          match if the class was a negative one (e.g. [^ab]). This has been fixed.
1999    
2000    (ii)  A negated character class with a single character < 255 is coded as
2001          "not this character" (OP_NOT). This wasn't working properly when the test
2002          character was multibyte, either singly or repeated.
2003    
2004    (iii) Repeats of multibyte characters are now handled correctly in UTF-8
2005          mode, for example: \x{100}{2,3}.
2006    
2007    (iv)  The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
2008          singly or repeated) now correctly test multibyte characters. However,
2009          PCRE doesn't recognize any characters with values greater than 255 as
2010          digits, spaces, or word characters. Such characters always match \D, \S,
2011          and \W, and never match \d, \s, or \w.
2012    
2013    (v)   Classes may now contain characters and character ranges with values
2014          greater than 255. For example: [ab\x{100}-\x{400}].
2015    
2016    (vi)  pcregrep now has a --utf-8 option (synonym -u) which makes it call
2017          PCRE in UTF-8 mode.
2018    
2019    52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
2020    PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
2021    retained for backwards compatibility. (Note that LASTLITERAL is also a byte
2022    value.)
2023    
2024    53. The single man page has become too large. I have therefore split it up into
2025    a number of separate man pages. These also give rise to individual HTML pages;
2026    these are now put in a separate directory, and there is an index.html page that
2027    lists them all. Some hyperlinking between the pages has been installed.
2028    
2029    54. Added convenience functions for handling named capturing parentheses.
2030    
2031    55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2032    aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2033    true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2034    are faulted.
2035    
2036    56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2037    calling configure. These values are used when compiling the dftables.c program
2038    which is run to generate the source of the default character tables. They
2039    default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2040    you will need to set these values.
2041    
2042    57. Updated the building process for Windows DLL, as provided by Fred Cox.
2043    
2044    
2045    Version 3.9 02-Jan-02
2046  ---------------------  ---------------------
2047    
2048  1. A bit of extraneous text had somehow crept into the pcregrep documentation.  1. A bit of extraneous text had somehow crept into the pcregrep documentation.

Legend:
Removed from v.61  
changed lines
  Added in v.189

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12