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

Diff of /code/trunk/ChangeLog

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

revision 59 by nigel, Sat Feb 24 21:39:54 2007 UTC revision 184 by ph10, Tue Jun 19 13:26:46 2007 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4    Version 7.2 19-June-07
5    ---------------------
6    
7     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
8        which is apparently normally available under Windows.
9    
10     2. Re-jig the pcregrep tests with different newline settings in an attempt
11        to make them independent of the local environment's newline setting.
12    
13     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
14    
15     4. Some of the "internals" tests were previously cut out when the link size
16        was not 2, because the output contained actual offsets. The recent new
17        "Z" feature of pcretest means that these can be cut out, making the tests
18        usable with all link sizes.
19    
20     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
21        stack recursion. This gives a massive performance boost under BSD, but just
22        a small improvement under Linux. However, it saves one field in the frame
23        in all cases.
24    
25     6. Added more features from the forthcoming Perl 5.10:
26    
27        (a) (?-n) (where n is a string of digits) is a relative subroutine or
28            recursion call. It refers to the nth most recently opened parentheses.
29    
30        (b) (?+n) is also a relative subroutine call; it refers to the nth next
31            to be opened parentheses.
32    
33        (c) Conditions that refer to capturing parentheses can be specified
34            relatively, for example, (?(-2)... or (?(+3)...
35    
36        (d) \K resets the start of the current match so that everything before
37            is not part of it.
38    
39        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
40    
41        (f) \g{name} is another synonym - part of Perl 5.10's unification of
42            reference syntax.
43    
44        (g) (?| introduces a group in which the numbering of parentheses in each
45            alternative starts with the same number.
46    
47        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
48    
49     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
50        PCRE_INFO_JCHANGED.
51    
52     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
53        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
54        for detecting groups that can match an empty string.
55    
56     9. A pattern with a very large number of alternatives (more than several
57        hundred) was running out of internal workspace during the pre-compile
58        phase, where pcre_compile() figures out how much memory will be needed. A
59        bit of new cunning has reduced the workspace needed for groups with
60        alternatives. The 1000-alternative test pattern now uses 12 bytes of
61        workspace instead of running out of the 4096 that are available.
62    
63    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
64    
65    11. Applied patch from Google to remove an optimization that didn't quite work.
66        The report of the bug said:
67    
68          pcrecpp::RE("a*").FullMatch("aaa") matches, while
69          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
70          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
71    
72    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
73        it matched the wrong number of bytes.
74    
75    
76    Version 7.1 24-Apr-07
77    ---------------------
78    
79     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
80        that is more "standard", making use of automake and other Autotools. There
81        is some re-arrangement of the files and adjustment of comments consequent
82        on this.
83    
84     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
85        for recursive directory scanning broke on some systems because the files
86        are not scanned in any specific order and on different systems the order
87        was different. A call to "sort" has been inserted into RunGrepTest for the
88        approprate test as a short-term fix. In the longer term there may be an
89        alternative.
90    
91     3. I had an email from Eric Raymond about problems translating some of PCRE's
92        man pages to HTML (despite the fact that I distribute HTML pages, some
93        people do their own conversions for various reasons). The problems
94        concerned the use of low-level troff macros .br and .in. I have therefore
95        removed all such uses from the man pages (some were redundant, some could
96        be replaced by .nf/.fi pairs). The 132html script that I use to generate
97        HTML has been updated to handle .nf/.fi and to complain if it encounters
98        .br or .in.
99    
100     4. Updated comments in configure.ac that get placed in config.h.in and also
101        arranged for config.h to be included in the distribution, with the name
102        config.h.generic, for the benefit of those who have to compile without
103        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
104    
105     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
106        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
107        makevp.bat for the current PCRE, using the additional files
108        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
109    
110     6. A Windows user reported a minor discrepancy with test 2, which turned out
111        to be caused by a trailing space on an input line that had got lost in his
112        copy. The trailing space was an accident, so I've just removed it.
113    
114     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
115        that is needed.
116    
117     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
118        as "const" (a) because they are and (b) because it helps the PHP
119        maintainers who have recently made a script to detect big data structures
120        in the php code that should be moved to the .rodata section. I remembered
121        to update Builducptable as well, so it won't revert if ucptable.h is ever
122        re-created.
123    
124     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
125        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
126        order to be able to cut out the UTF-8 tables in the latter when UTF-8
127        support is not required. This saves 1.5-2K of code, which is important in
128        some applications.
129    
130        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
131        so as not to refer to the tables, even though these functions will never be
132        called when UTF-8 support is disabled. Otherwise there are problems with a
133        shared library.
134    
135    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
136    
137        (a) It was defining its arguments as char * instead of void *.
138    
139        (b) It was assuming that all moves were upwards in memory; this was true
140            a long time ago when I wrote it, but is no longer the case.
141    
142        The emulated memove() is provided for those environments that have neither
143        memmove() nor bcopy(). I didn't think anyone used it these days, but that
144        is clearly not the case, as these two bugs were recently reported.
145    
146    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
147        and Detrail to create the HTML documentation, the .txt form of the man
148        pages, and it removes trailing spaces from listed files. It also creates
149        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
150        case, it wraps all the #defines with #ifndefs. This script should be run
151        before "make dist".
152    
153    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
154        with Unicode property support.
155    
156        (a) For a maximizing quantifier, if the two different cases of the
157            character were of different lengths in their UTF-8 codings (there are
158            some cases like this - I found 11), and the matching function had to
159            back up over a mixture of the two cases, it incorrectly assumed they
160            were both the same length.
161    
162        (b) When PCRE was configured to use the heap rather than the stack for
163            recursion during matching, it was not correctly preserving the data for
164            the other case of a UTF-8 character when checking ahead for a match
165            while processing a minimizing repeat. If the check also involved
166            matching a wide character, but failed, corruption could cause an
167            erroneous result when trying to check for a repeat of the original
168            character.
169    
170    13. Some tidying changes to the testing mechanism:
171    
172        (a) The RunTest script now detects the internal link size and whether there
173            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
174            values substituted by "configure". (The RunGrepTest script already did
175            this for UTF-8.) The configure.ac script no longer substitutes the
176            relevant variables.
177    
178        (b) The debugging options /B and /D in pcretest show the compiled bytecode
179            with length and offset values. This means that the output is different
180            for different internal link sizes. Test 2 is skipped for link sizes
181            other than 2 because of this, bypassing the problem. Unfortunately,
182            there was also a test in test 3 (the locale tests) that used /B and
183            failed for link sizes other than 2. Rather than cut the whole test out,
184            I have added a new /Z option to pcretest that replaces the length and
185            offset values with spaces. This is now used to make test 3 independent
186            of link size. (Test 2 will be tidied up later.)
187    
188    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
189        segmentation fault instead of returning the appropriate error message.
190    
191    15. In multiline mode when the newline sequence was set to "any", the pattern
192        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
193        This doesn't seem right; it now treats the CRLF combination as the line
194        ending, and so does not match in that case. It's only a pattern such as ^$
195        that would hit this one: something like ^ABC$ would have failed after \r
196        and then tried again after \r\n.
197    
198    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
199        in an attempt to make files that differ only in their line terminators
200        compare equal. This works on Linux.
201    
202    17. Under certain error circumstances pcregrep might try to free random memory
203        as it exited. This is now fixed, thanks to valgrind.
204    
205    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
206        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
207        was because its rules for how to advance for /g after matching an empty
208        string at the end of a line did not allow for this case. They now check for
209        it specially.
210    
211    20. pcretest is supposed to handle patterns and data of any length, by
212        extending its buffers when necessary. It was getting this wrong when the
213        buffer for a data line had to be extended.
214    
215    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
216        CRLF as a newline sequence.
217    
218    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
219        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
220        I have nevertheless tidied it up.
221    
222    23. Added some casts to kill warnings from HP-UX ia64 compiler.
223    
224    24. Added a man page for pcre-config.
225    
226    
227    Version 7.0 19-Dec-06
228    ---------------------
229    
230     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
231        moving to gcc 4.1.1.
232    
233     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
234        sys/time.h, which is documented as needed for this function. It doesn't
235        seem to matter on Linux, but it showed up on some releases of OS X.
236    
237     3. It seems that there are systems where bytes whose values are greater than
238        127 match isprint() in the "C" locale. The "C" locale should be the
239        default when a C program starts up. In most systems, only ASCII printing
240        characters match isprint(). This difference caused the output from pcretest
241        to vary, making some of the tests fail. I have changed pcretest so that:
242    
243        (a) When it is outputting text in the compiled version of a pattern, bytes
244            other than 32-126 are always shown as hex escapes.
245    
246        (b) When it is outputting text that is a matched part of a subject string,
247            it does the same, unless a different locale has been set for the match
248            (using the /L modifier). In this case, it uses isprint() to decide.
249    
250     4. Fixed a major bug that caused incorrect computation of the amount of memory
251        required for a compiled pattern when options that changed within the
252        pattern affected the logic of the preliminary scan that determines the
253        length. The relevant options are -x, and -i in UTF-8 mode. The result was
254        that the computed length was too small. The symptoms of this bug were
255        either the PCRE error "internal error: code overflow" from pcre_compile(),
256        or a glibc crash with a message such as "pcretest: free(): invalid next
257        size (fast)". Examples of patterns that provoked this bug (shown in
258        pcretest format) are:
259    
260          /(?-x: )/x
261          /(?x)(?-x: \s*#\s*)/
262          /((?i)[\x{c0}])/8
263          /(?i:[\x{c0}])/8
264    
265        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
266        is now done differently.
267    
268     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
269        wrapper classes; (b) implement a new function in the C++ scanner that is
270        more efficient than the old way of doing things because it avoids levels of
271        recursion in the regex matching; (c) add a paragraph to the documentation
272        for the FullMatch() function.
273    
274     6. The escape sequence \n was being treated as whatever was defined as
275        "newline". Not only was this contrary to the documentation, which states
276        that \n is character 10 (hex 0A), but it also went horribly wrong when
277        "newline" was defined as CRLF. This has been fixed.
278    
279     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
280        was being set to -1 for the "end of line" case (supposedly a value that no
281        character can have). Though this value is never used (the check for end of
282        line is "zero bytes in current character"), it caused compiler complaints.
283        I've changed it to 0xffffffff.
284    
285     8. In pcre_version.c, the version string was being built by a sequence of
286        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
287        string (as it is for production releases) called a macro with an empty
288        argument. The C standard says the result of this is undefined. The gcc
289        compiler treats it as an empty string (which was what was wanted) but it is
290        reported that Visual C gives an error. The source has been hacked around to
291        avoid this problem.
292    
293     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
294        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
295        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
296        of them did).
297    
298    10. Originally, pcretest opened its input and output without "b"; then I was
299        told that "b" was needed in some environments, so it was added for release
300        5.0 to both the input and output. (It makes no difference on Unix-like
301        systems.) Later I was told that it is wrong for the input on Windows. I've
302        now abstracted the modes into two macros, to make it easier to fiddle with
303        them, and removed "b" from the input mode under Windows.
304    
305    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
306    
307    12. Added -help and --help to pcretest as an official way of being reminded
308        of the options.
309    
310    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
311        and pcrecpp.cc because they annoy compilers at high warning levels.
312    
313    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
314    
315    15. Fixed an occurrence of == in configure.ac that should have been = (shell
316        scripts are not C programs :-) and which was not noticed because it works
317        on Linux.
318    
319    16. pcretest is supposed to handle any length of pattern and data line (as one
320        line or as a continued sequence of lines) by extending its input buffer if
321        necessary. This feature was broken for very long pattern lines, leading to
322        a string of junk being passed to pcre_compile() if the pattern was longer
323        than about 50K.
324    
325    17. I have done a major re-factoring of the way pcre_compile() computes the
326        amount of memory needed for a compiled pattern. Previously, there was code
327        that made a preliminary scan of the pattern in order to do this. That was
328        OK when PCRE was new, but as the facilities have expanded, it has become
329        harder and harder to keep it in step with the real compile phase, and there
330        have been a number of bugs (see for example, 4 above). I have now found a
331        cunning way of running the real compile function in a "fake" mode that
332        enables it to compute how much memory it would need, while actually only
333        ever using a few hundred bytes of working memory and without too many
334        tests of the mode. This should make future maintenance and development
335        easier. A side effect of this work is that the limit of 200 on the nesting
336        depth of parentheses has been removed (though this was never a serious
337        limitation, I suspect). However, there is a downside: pcre_compile() now
338        runs more slowly than before (30% or more, depending on the pattern). I
339        hope this isn't a big issue. There is no effect on runtime performance.
340    
341    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
342        newline (only possible for the last line of a file) and it was a
343        pattern that set a locale (followed by /Lsomething), pcretest crashed.
344    
345    19. Added additional timing features to pcretest. (1) The -tm option now times
346        matching only, not compiling. (2) Both -t and -tm can be followed, as a
347        separate command line item, by a number that specifies the number of
348        repeats to use when timing. The default is 50000; this gives better
349        precision, but takes uncomfortably long for very large patterns.
350    
351    20. Extended pcre_study() to be more clever in cases where a branch of a
352        subpattern has no definite first character. For example, (a*|b*)[cd] would
353        previously give no result from pcre_study(). Now it recognizes that the
354        first character must be a, b, c, or d.
355    
356    21. There was an incorrect error "recursive call could loop indefinitely" if
357        a subpattern (or the entire pattern) that was being tested for matching an
358        empty string contained only one non-empty item after a nested subpattern.
359        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
360        incorrectly, because the \d was being skipped in the check.
361    
362    22. The pcretest program now has a new pattern option /B and a command line
363        option -b, which is equivalent to adding /B to every pattern. This causes
364        it to show the compiled bytecode, without the additional information that
365        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
366        is the same as /B/I).
367    
368    23. A new optimization is now able automatically to treat some sequences such
369        as a*b as a*+b. More specifically, if something simple (such as a character
370        or a simple class like \d) has an unlimited quantifier, and is followed by
371        something that cannot possibly match the quantified thing, the quantifier
372        is automatically "possessified".
373    
374    24. A recursive reference to a subpattern whose number was greater than 39
375        went wrong under certain circumstances in UTF-8 mode. This bug could also
376        have affected the operation of pcre_study().
377    
378    25. Realized that a little bit of performance could be had by replacing
379        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
380    
381    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
382    
383    27. Possessive quantifiers such as a++ were previously implemented by turning
384        them into atomic groups such as ($>a+). Now they have their own opcodes,
385        which improves performance. This includes the automatically created ones
386        from 23 above.
387    
388    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
389        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
390        the first matched character to be a colon. This applied both to named and
391        numbered groups.
392    
393    29. The ucpinternal.h header file was missing its idempotency #ifdef.
394    
395    30. I was sent a "project" file called libpcre.a.dev which I understand makes
396        building PCRE on Windows easier, so I have included it in the distribution.
397    
398    31. There is now a check in pcretest against a ridiculously large number being
399        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
400        loop, the loop is abandoned.
401    
402    32. Forward references to subpatterns in conditions such as (?(2)...) where
403        subpattern 2 is defined later cause pcre_compile() to search forwards in
404        the pattern for the relevant set of parentheses. This search went wrong
405        when there were unescaped parentheses in a character class, parentheses
406        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
407    
408    33. "Subroutine" calls and backreferences were previously restricted to
409        referencing subpatterns earlier in the regex. This restriction has now
410        been removed.
411    
412    34. Added a number of extra features that are going to be in Perl 5.10. On the
413        whole, these are just syntactic alternatives for features that PCRE had
414        previously implemented using the Python syntax or my own invention. The
415        other formats are all retained for compatibility.
416    
417        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
418            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
419            also .NET compatible.
420    
421        (b) A recursion or subroutine call to a named group can now be defined as
422            (?&name) as well as (?P>name).
423    
424        (c) A backreference to a named group can now be defined as \k<name> or
425            \k'name' as well as (?P=name). The new forms, as well as being in Perl
426            5.10, are also .NET compatible.
427    
428        (d) A conditional reference to a named group can now use the syntax
429            (?(<name>) or (?('name') as well as (?(name).
430    
431        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
432            groups (named and numbered) that are never evaluated inline, but can be
433            called as "subroutines" from elsewhere. In effect, the DEFINE condition
434            is always false. There may be only one alternative in such a group.
435    
436        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
437            as the simple (?(R). The condition is true only if the most recent
438            recursion is that of the given number or name. It does not search out
439            through the entire recursion stack.
440    
441        (g) The escape \gN or \g{N} has been added, where N is a positive or
442            negative number, specifying an absolute or relative reference.
443    
444    35. Tidied to get rid of some further signed/unsigned compiler warnings and
445        some "unreachable code" warnings.
446    
447    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
448        things, this adds five new scripts.
449    
450    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
451        There were also incompatibilities regarding the handling of \Q..\E inside
452        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
453        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
454    
455    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
456        matches an empty string, and forcibly breaks the loop. There were bugs in
457        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
458        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
459        separate and independent bugs (that affected different cases) have been
460        fixed.
461    
462    39. Refactored the code to abolish the use of different opcodes for small
463        capturing bracket numbers. This is a tidy that I avoided doing when I
464        removed the limit on the number of capturing brackets for 3.5 back in 2001.
465        The new approach is not only tidier, it makes it possible to reduce the
466        memory needed to fix the previous bug (38).
467    
468    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
469        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
470        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
471        mode.
472    
473    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
474        report.
475    
476    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
477        copy construction and assignment in the C++ wrapper.
478    
479    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
480        couple of bugs that could have given wrong results in the "--newline=crlf"
481        case.
482    
483    44. Added a number of casts and did some reorganization of signed/unsigned int
484        variables following suggestions from Dair Grant. Also renamed the variable
485        "this" as "item" because it is a C++ keyword.
486    
487    45. Arranged for dftables to add
488    
489          #include "pcre_internal.h"
490    
491        to pcre_chartables.c because without it, gcc 4.x may remove the array
492        definition from the final binary if PCRE is built into a static library and
493        dead code stripping is activated.
494    
495    46. For an unanchored pattern, if a match attempt fails at the start of a
496        newline sequence, and the newline setting is CRLF or ANY, and the next two
497        characters are CRLF, advance by two characters instead of one.
498    
499    
500    Version 6.7 04-Jul-06
501    ---------------------
502    
503     1. In order to handle tests when input lines are enormously long, pcretest has
504        been re-factored so that it automatically extends its buffers when
505        necessary. The code is crude, but this _is_ just a test program. The
506        default size has been increased from 32K to 50K.
507    
508     2. The code in pcre_study() was using the value of the re argument before
509        testing it for NULL. (Of course, in any sensible call of the function, it
510        won't be NULL.)
511    
512     3. The memmove() emulation function in pcre_internal.h, which is used on
513        systems that lack both memmove() and bcopy() - that is, hardly ever -
514        was missing a "static" storage class specifier.
515    
516     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
517        containing an extended class (one that cannot be represented by a bitmap
518        because it contains high-valued characters or Unicode property items, e.g.
519        [\pZ]). Almost always one would set UTF-8 mode when processing such a
520        pattern, but PCRE should not loop if you do not (it no longer does).
521        [Detail: two cases were found: (a) a repeated subpattern containing an
522        extended class; (b) a recursive reference to a subpattern that followed a
523        previous extended class. It wasn't skipping over the extended class
524        correctly when UTF-8 mode was not set.]
525    
526     5. A negated single-character class was not being recognized as fixed-length
527        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
528        compile error "lookbehind assertion is not fixed length".
529    
530     6. The RunPerlTest auxiliary script was showing an unexpected difference
531        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
532        write a Perl script that can interpret lines of an input file either as
533        byte characters or as UTF-8, which is what "perltest" was being required to
534        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
535        can't do is switch easily at run time between having the "use utf8;" pragma
536        or not. In the end, I fudged it by using the RunPerlTest script to insert
537        "use utf8;" explicitly for the UTF-8 tests.
538    
539     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
540        the end of the subject string, contrary to the documentation and to what
541        Perl does. This was true of both matching functions. Now it matches only at
542        the start of the subject and immediately after *internal* newlines.
543    
544     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
545        a pointer to an int instead of a pointer to an unsigned long int. This
546        caused problems on 64-bit systems.
547    
548     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
549        instance of the 'standard' template library not being so standard".
550    
551    10. There was no check on the number of named subpatterns nor the maximum
552        length of a subpattern name. The product of these values is used to compute
553        the size of the memory block for a compiled pattern. By supplying a very
554        long subpattern name and a large number of named subpatterns, the size
555        computation could be caused to overflow. This is now prevented by limiting
556        the length of names to 32 characters, and the number of named subpatterns
557        to 10,000.
558    
559    11. Subpatterns that are repeated with specific counts have to be replicated in
560        the compiled pattern. The size of memory for this was computed from the
561        length of the subpattern and the repeat count. The latter is limited to
562        65535, but there was no limit on the former, meaning that integer overflow
563        could in principle occur. The compiled length of a repeated subpattern is
564        now limited to 30,000 bytes in order to prevent this.
565    
566    12. Added the optional facility to have named substrings with the same name.
567    
568    13. Added the ability to use a named substring as a condition, using the
569        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
570        are numbers (not recommended). Forward references are permitted.
571    
572    14. Added forward references in named backreferences (if you see what I mean).
573    
574    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
575        pattern could run off the end of the subject. For example, the pattern
576        "(?s)(.{1,5})"8 did this with the subject "ab".
577    
578    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
579        PCRE_CASELESS was set when matching characters that were quantified with ?
580        or *.
581    
582    17. A character class other than a single negated character that had a minimum
583        but no maximum quantifier - for example [ab]{6,} - was not handled
584        correctly by pce_dfa_exec(). It would match only one character.
585    
586    18. A valid (though odd) pattern that looked like a POSIX character
587        class but used an invalid character after [ (for example [[,abc,]]) caused
588        pcre_compile() to give the error "Failed: internal error: code overflow" or
589        in some cases to crash with a glibc free() error. This could even happen if
590        the pattern terminated after [[ but there just happened to be a sequence of
591        letters, a binary zero, and a closing ] in the memory that followed.
592    
593    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
594        over the years. Originally (before any Unicode support), just the bottom 8
595        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
596        output from "man perlunicode" includes this:
597    
598          The regular expression compiler produces polymorphic opcodes.  That
599          is, the pattern adapts to the data and automatically switches to
600          the Unicode character scheme when presented with Unicode data--or
601          instead uses a traditional byte scheme when presented with byte
602          data.
603    
604        Sadly, a wide octal escape does not cause a switch, and in a string with
605        no other multibyte characters, these octal escapes are treated as before.
606        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
607        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
608        Unicode string.
609    
610        I have not perpetrated such confusion in PCRE. Up till now, it took just
611        the bottom 8 bits, as in old Perl. I have now made octal escapes with
612        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
613        translate to the appropriate multibyte character.
614    
615    29. Applied some refactoring to reduce the number of warnings from Microsoft
616        and Borland compilers. This has included removing the fudge introduced
617        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
618        a warning about an unused variable.
619    
620    21. PCRE has not included VT (character 0x0b) in the set of whitespace
621        characters since release 4.0, because Perl (from release 5.004) does not.
622        [Or at least, is documented not to: some releases seem to be in conflict
623        with the documentation.] However, when a pattern was studied with
624        pcre_study() and all its branches started with \s, PCRE still included VT
625        as a possible starting character. Of course, this did no harm; it just
626        caused an unnecessary match attempt.
627    
628    22. Removed a now-redundant internal flag bit that recorded the fact that case
629        dependency changed within the pattern. This was once needed for "required
630        byte" processing, but is no longer used. This recovers a now-scarce options
631        bit. Also moved the least significant internal flag bit to the most-
632        significant bit of the word, which was not previously used (hangover from
633        the days when it was an int rather than a uint) to free up another bit for
634        the future.
635    
636    23. Added support for CRLF line endings as well as CR and LF. As well as the
637        default being selectable at build time, it can now be changed at runtime
638        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
639        specify that it is scanning data with non-default line endings.
640    
641    24. Changed the definition of CXXLINK to make it agree with the definition of
642        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
643    
644    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
645        recursions. This makes a big different to stack usage for some patterns.
646    
647    26. If a subpattern containing a named recursion or subroutine reference such
648        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
649        the space required for the compiled pattern went wrong and gave too small a
650        value. Depending on the environment, this could lead to "Failed: internal
651        error: code overflow at offset 49" or "glibc detected double free or
652        corruption" errors.
653    
654    27. Applied patches from Google (a) to support the new newline modes and (b) to
655        advance over multibyte UTF-8 characters in GlobalReplace.
656    
657    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
658        difference for some implementation of PCRE in some Windows version.
659    
660    29. Added some extra testing facilities to pcretest:
661    
662        \q<number>   in a data line sets the "match limit" value
663        \Q<number>   in a data line sets the "match recursion limt" value
664        -S <number>  sets the stack size, where <number> is in megabytes
665    
666        The -S option isn't available for Windows.
667    
668    
669    Version 6.6 06-Feb-06
670    ---------------------
671    
672     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
673        in pcreposix.h. I have copied the definition from pcre.h.
674    
675     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
676        because pcre.h is no longer a built file.
677    
678     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
679        not normally included in the compiled code.
680    
681    
682    Version 6.5 01-Feb-06
683    ---------------------
684    
685     1. When using the partial match feature with pcre_dfa_exec(), it was not
686        anchoring the second and subsequent partial matches at the new starting
687        point. This could lead to incorrect results. For example, with the pattern
688        /1234/, partially matching against "123" and then "a4" gave a match.
689    
690     2. Changes to pcregrep:
691    
692        (a) All non-match returns from pcre_exec() were being treated as failures
693            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
694            error message is output. Some extra information is given for the
695            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
696            probably the only errors that are likely to be caused by users (by
697            specifying a regex that has nested indefinite repeats, for instance).
698            If there are more than 20 of these errors, pcregrep is abandoned.
699    
700        (b) A binary zero was treated as data while matching, but terminated the
701            output line if it was written out. This has been fixed: binary zeroes
702            are now no different to any other data bytes.
703    
704        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
705            used to set a locale for matching. The --locale=xxxx long option has
706            been added (no short equivalent) to specify a locale explicitly on the
707            pcregrep command, overriding the environment variables.
708    
709        (d) When -B was used with -n, some line numbers in the output were one less
710            than they should have been.
711    
712        (e) Added the -o (--only-matching) option.
713    
714        (f) If -A or -C was used with -c (count only), some lines of context were
715            accidentally printed for the final match.
716    
717        (g) Added the -H (--with-filename) option.
718    
719        (h) The combination of options -rh failed to suppress file names for files
720            that were found from directory arguments.
721    
722        (i) Added the -D (--devices) and -d (--directories) options.
723    
724        (j) Added the -F (--fixed-strings) option.
725    
726        (k) Allow "-" to be used as a file name for -f as well as for a data file.
727    
728        (l) Added the --colo(u)r option.
729    
730        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
731            is not present by default.
732    
733     3. A nasty bug was discovered in the handling of recursive patterns, that is,
734        items such as (?R) or (?1), when the recursion could match a number of
735        alternatives. If it matched one of the alternatives, but subsequently,
736        outside the recursion, there was a failure, the code tried to back up into
737        the recursion. However, because of the way PCRE is implemented, this is not
738        possible, and the result was an incorrect result from the match.
739    
740        In order to prevent this happening, the specification of recursion has
741        been changed so that all such subpatterns are automatically treated as
742        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
743    
744     4. I had overlooked the fact that, in some locales, there are characters for
745        which isalpha() is true but neither isupper() nor islower() are true. In
746        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
747        and ordfeminine) are like this. This affected the treatment of \w and \W
748        when they appeared in character classes, but not when they appeared outside
749        a character class. The bit map for "word" characters is now created
750        separately from the results of isalnum() instead of just taking it from the
751        upper, lower, and digit maps. (Plus the underscore character, of course.)
752    
753     5. The above bug also affected the handling of POSIX character classes such as
754        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
755        permanent tables. Instead, the bit maps for such a class were previously
756        created as the appropriate unions of the upper, lower, and digit bitmaps.
757        Now they are created by subtraction from the [[:word:]] class, which has
758        its own bitmap.
759    
760     6. The [[:blank:]] character class matches horizontal, but not vertical space.
761        It is created by subtracting the vertical space characters (\x09, \x0a,
762        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
763        subtraction was done in the overall bitmap for a character class, meaning
764        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
765        be recognized. This bug has been fixed.
766    
767     7. Patches from the folks at Google:
768    
769          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
770          real life, but is still worth protecting against".
771    
772          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
773          regular expressions".
774    
775          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
776          have it.
777    
778          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
779          "configure" and the latter not, in order to fix a problem somebody had
780          with compiling the Arg class on HP-UX.
781    
782          (e) Improve the error-handling of the C++ wrapper a little bit.
783    
784          (f) New tests for checking recursion limiting.
785    
786     8. The pcre_memmove() function, which is used only if the environment does not
787        have a standard memmove() function (and is therefore rarely compiled),
788        contained two bugs: (a) use of int instead of size_t, and (b) it was not
789        returning a result (though PCRE never actually uses the result).
790    
791     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
792        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
793        returned instead of calling malloc() with an overflowing number that would
794        most likely cause subsequent chaos.
795    
796    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
797    
798    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
799        with this option is matched, the nmatch and pmatch options of regexec() are
800        ignored.
801    
802    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
803        provided in case anyone wants to the the POSIX interface with UTF-8
804        strings.
805    
806    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
807        C++ linking (needed for some HP-UX environments).
808    
809    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
810        (unused parameter) and in the pcre_printint() function (omitted "default"
811        switch label when the default is to do nothing).
812    
813    15. Added some code to make it possible, when PCRE is compiled as a C++
814        library, to replace subject pointers for pcre_exec() with a smart pointer
815        class, thus making it possible to process discontinuous strings.
816    
817    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
818        much the same function. They were added by different people who were trying
819        to make PCRE easy to compile on non-Unix systems. It has been suggested
820        that PCRE_EXPORT be abolished now that there is more automatic apparatus
821        for compiling on Windows systems. I have therefore replaced it with
822        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
823        defaults to "extern" for C or "extern C" for C++, which works fine on
824        Unix-like systems. It is now possible to override the value of PCRE_DATA_
825        SCOPE with something explicit in config.h. In addition:
826    
827        (a) pcreposix.h still had just "extern" instead of either of these macros;
828            I have replaced it with PCRE_DATA_SCOPE.
829    
830        (b) Functions such as _pcre_xclass(), which are internal to the library,
831            but external in the C sense, all had PCRE_EXPORT in their definitions.
832            This is apparently wrong for the Windows case, so I have removed it.
833            (It makes no difference on Unix-like systems.)
834    
835    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
836        of recursive calls to match(). This is different to MATCH_LIMIT because
837        that limits the total number of calls to match(), not all of which increase
838        the depth of recursion. Limiting the recursion depth limits the amount of
839        stack (or heap if NO_RECURSE is set) that is used. The default can be set
840        when PCRE is compiled, and changed at run time. A patch from Google adds
841        this functionality to the C++ interface.
842    
843    18. Changes to the handling of Unicode character properties:
844    
845        (a) Updated the table to Unicode 4.1.0.
846    
847        (b) Recognize characters that are not in the table as "Cn" (undefined).
848    
849        (c) I revised the way the table is implemented to a much improved format
850            which includes recognition of ranges. It now supports the ranges that
851            are defined in UnicodeData.txt, and it also amalgamates other
852            characters into ranges. This has reduced the number of entries in the
853            table from around 16,000 to around 3,000, thus reducing its size
854            considerably. I realized I did not need to use a tree structure after
855            all - a binary chop search is just as efficient. Having reduced the
856            number of entries, I extended their size from 6 bytes to 8 bytes to
857            allow for more data.
858    
859        (d) Added support for Unicode script names via properties such as \p{Han}.
860    
861    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
862        matching that character.
863    
864    20. When matching a repeated Unicode property with a minimum greater than zero,
865        (for example \pL{2,}), PCRE could look past the end of the subject if it
866        reached it while seeking the minimum number of characters. This could
867        happen only if some of the characters were more than one byte long, because
868        there is a check for at least the minimum number of bytes.
869    
870    21. Refactored the implementation of \p and \P so as to be more general, to
871        allow for more different types of property in future. This has changed the
872        compiled form incompatibly. Anybody with saved compiled patterns that use
873        \p or \P will have to recompile them.
874    
875    22. Added "Any" and "L&" to the supported property types.
876    
877    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
878        but give a compile time error if the value is greater than 0xff.
879    
880    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
881        accidentally not being installed or uninstalled.
882    
883    25. The pcre.h file was built from pcre.h.in, but the only changes that were
884        made were to insert the current release number. This seemed silly, because
885        it made things harder for people building PCRE on systems that don't run
886        "configure". I have turned pcre.h into a distributed file, no longer built
887        by "configure", with the version identification directly included. There is
888        no longer a pcre.h.in file.
889    
890        However, this change necessitated a change to the pcre-config script as
891        well. It is built from pcre-config.in, and one of the substitutions was the
892        release number. I have updated configure.ac so that ./configure now finds
893        the release number by grepping pcre.h.
894    
895    26. Added the ability to run the tests under valgrind.
896    
897    
898    Version 6.4 05-Sep-05
899    ---------------------
900    
901     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
902        "--" to be printed when multiple files were scanned, even when none of the
903        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
904        consider it to be a bug, and have restored the previous behaviour.
905    
906     2. A couple of code tidies to get rid of compiler warnings.
907    
908     3. The pcretest program used to cheat by referring to symbols in the library
909        whose names begin with _pcre_. These are internal symbols that are not
910        really supposed to be visible externally, and in some environments it is
911        possible to suppress them. The cheating is now confined to including
912        certain files from the library's source, which is a bit cleaner.
913    
914     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
915        file's purpose clearer.
916    
917     5. Reorganized pcre_ucp_findchar().
918    
919    
920    Version 6.3 15-Aug-05
921    ---------------------
922    
923     1. The file libpcre.pc.in did not have general read permission in the tarball.
924    
925     2. There were some problems when building without C++ support:
926    
927        (a) If C++ support was not built, "make install" and "make test" still
928            tried to test it.
929    
930        (b) There were problems when the value of CXX was explicitly set. Some
931            changes have been made to try to fix these, and ...
932    
933        (c) --disable-cpp can now be used to explicitly disable C++ support.
934    
935        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
936            backslash in a target when C++ was disabled. This confuses some
937            versions of "make", apparently. Using an intermediate variable solves
938            this. (Same for CPP_LOBJ.)
939    
940     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
941        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
942        necessary on certain architectures.
943    
944     4. Added a setting of -export-symbols-regex to the link command to remove
945        those symbols that are exported in the C sense, but actually are local
946        within the library, and not documented. Their names all begin with
947        "_pcre_". This is not a perfect job, because (a) we have to except some
948        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
949        available (and never for static libraries). I have made a note to try to
950        find a way round (a) in the future.
951    
952    
953    Version 6.2 01-Aug-05
954    ---------------------
955    
956     1. There was no test for integer overflow of quantifier values. A construction
957        such as {1111111111111111} would give undefined results. What is worse, if
958        a minimum quantifier for a parenthesized subpattern overflowed and became
959        negative, the calculation of the memory size went wrong. This could have
960        led to memory overwriting.
961    
962     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
963    
964     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
965        operating environments where this matters.
966    
967     4. Applied Giuseppe Maxia's patch to add additional features for controlling
968        PCRE options from within the C++ wrapper.
969    
970     5. Named capturing subpatterns were not being correctly counted when a pattern
971        was compiled. This caused two problems: (a) If there were more than 100
972        such subpatterns, the calculation of the memory needed for the whole
973        compiled pattern went wrong, leading to an overflow error. (b) Numerical
974        back references of the form \12, where the number was greater than 9, were
975        not recognized as back references, even though there were sufficient
976        previous subpatterns.
977    
978     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
979        versions of gcc, e.g. 2.95.4.
980    
981    
982    Version 6.1 21-Jun-05
983    ---------------------
984    
985     1. There was one reference to the variable "posix" in pcretest.c that was not
986        surrounded by "#if !defined NOPOSIX".
987    
988     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
989        the cross-check on the old pcre_info() function, for the benefit of the
990        cut-down version of PCRE that is currently imported into Exim.
991    
992     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
993        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
994        patterns that start (?i) but I don't think that matters. The use of (?i) is
995        just an example; this all applies to the other options as well.
996    
997     4. Since libtool seems to echo the compile commands it is issuing, the output
998        from "make" can be reduced a bit by putting "@" in front of each libtool
999        compile command.
1000    
1001     5. Patch from the folks at Google for configure.in to be a bit more thorough
1002        in checking for a suitable C++ installation before trying to compile the
1003        C++ stuff. This should fix a reported problem when a compiler was present,
1004        but no suitable headers.
1005    
1006     6. The man pages all had just "PCRE" as their title. I have changed them to
1007        be the relevant file name. I have also arranged that these names are
1008        retained in the file doc/pcre.txt, which is a concatenation in text format
1009        of all the man pages except the little individual ones for each function.
1010    
1011     7. The NON-UNIX-USE file had not been updated for the different set of source
1012        files that come with release 6. I also added a few comments about the C++
1013        wrapper.
1014    
1015    
1016    Version 6.0 07-Jun-05
1017    ---------------------
1018    
1019     1. Some minor internal re-organization to help with my DFA experiments.
1020    
1021     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1022        didn't matter for the library itself when fully configured, but did matter
1023        when compiling without UCP support, or within Exim, where the ucp files are
1024        not imported.
1025    
1026     3. Refactoring of the library code to split up the various functions into
1027        different source modules. The addition of the new DFA matching code (see
1028        below) to a single monolithic source would have made it really too
1029        unwieldy, quite apart from causing all the code to be include in a
1030        statically linked application, when only some functions are used. This is
1031        relevant even without the DFA addition now that patterns can be compiled in
1032        one application and matched in another.
1033    
1034        The downside of splitting up is that there have to be some external
1035        functions and data tables that are used internally in different modules of
1036        the library but which are not part of the API. These have all had their
1037        names changed to start with "_pcre_" so that they are unlikely to clash
1038        with other external names.
1039    
1040     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1041        a different (DFA) algorithm. Although it is slower than the original
1042        function, it does have some advantages for certain types of matching
1043        problem.
1044    
1045     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1046        including restarting after a partial match.
1047    
1048     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1049        defined when compiling for Windows was sent to me. I have put it into the
1050        code, though I have no means of testing or verifying it.
1051    
1052     7. Added the pcre_refcount() auxiliary function.
1053    
1054     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1055        match before or at the first newline in the subject string. In pcretest,
1056        the /f option on a pattern can be used to set this.
1057    
1058     9. A repeated \w when used in UTF-8 mode with characters greater than 256
1059        would behave wrongly. This has been present in PCRE since release 4.0.
1060    
1061    10. A number of changes to the pcregrep command:
1062    
1063        (a) Refactored how -x works; insert ^(...)$ instead of setting
1064            PCRE_ANCHORED and checking the length, in preparation for adding
1065            something similar for -w.
1066    
1067        (b) Added the -w (match as a word) option.
1068    
1069        (c) Refactored the way lines are read and buffered so as to have more
1070            than one at a time available.
1071    
1072        (d) Implemented a pcregrep test script.
1073    
1074        (e) Added the -M (multiline match) option. This allows patterns to match
1075            over several lines of the subject. The buffering ensures that at least
1076            8K, or the rest of the document (whichever is the shorter) is available
1077            for matching (and similarly the previous 8K for lookbehind assertions).
1078    
1079        (f) Changed the --help output so that it now says
1080    
1081              -w, --word-regex(p)
1082    
1083            instead of two lines, one with "regex" and the other with "regexp"
1084            because that confused at least one person since the short forms are the
1085            same. (This required a bit of code, as the output is generated
1086            automatically from a table. It wasn't just a text change.)
1087    
1088        (g) -- can be used to terminate pcregrep options if the next thing isn't an
1089            option but starts with a hyphen. Could be a pattern or a path name
1090            starting with a hyphen, for instance.
1091    
1092        (h) "-" can be given as a file name to represent stdin.
1093    
1094        (i) When file names are being printed, "(standard input)" is used for
1095            the standard input, for compatibility with GNU grep. Previously
1096            "<stdin>" was used.
1097    
1098        (j) The option --label=xxx can be used to supply a name to be used for
1099            stdin when file names are being printed. There is no short form.
1100    
1101        (k) Re-factored the options decoding logic because we are going to add
1102            two more options that take data. Such options can now be given in four
1103            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1104    
1105        (l) Added the -A, -B, and -C options for requesting that lines of context
1106            around matches be printed.
1107    
1108        (m) Added the -L option to print the names of files that do not contain
1109            any matching lines, that is, the complement of -l.
1110    
1111        (n) The return code is 2 if any file cannot be opened, but pcregrep does
1112            continue to scan other files.
1113    
1114        (o) The -s option was incorrectly implemented. For compatibility with other
1115            greps, it now suppresses the error message for a non-existent or non-
1116            accessible file (but not the return code). There is a new option called
1117            -q that suppresses the output of matching lines, which was what -s was
1118            previously doing.
1119    
1120        (p) Added --include and --exclude options to specify files for inclusion
1121            and exclusion when recursing.
1122    
1123    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1124        Hopefully, it now does.
1125    
1126    12. Missing cast in pcre_study().
1127    
1128    13. Added an "uninstall" target to the makefile.
1129    
1130    14. Replaced "extern" in the function prototypes in Makefile.in with
1131        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1132        world, but is set differently for Windows.
1133    
1134    15. Added a second compiling function called pcre_compile2(). The only
1135        difference is that it has an extra argument, which is a pointer to an
1136        integer error code. When there is a compile-time failure, this is set
1137        non-zero, in addition to the error test pointer being set to point to an
1138        error message. The new argument may be NULL if no error number is required
1139        (but then you may as well call pcre_compile(), which is now just a
1140        wrapper). This facility is provided because some applications need a
1141        numeric error indication, but it has also enabled me to tidy up the way
1142        compile-time errors are handled in the POSIX wrapper.
1143    
1144    16. Added VPATH=.libs to the makefile; this should help when building with one
1145        prefix path and installing with another. (Or so I'm told by someone who
1146        knows more about this stuff than I do.)
1147    
1148    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1149        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1150        match everything, including newlines. This is not POSIX-compatible, but
1151        somebody wanted the feature. From pcretest it can be activated by using
1152        both the P and the s flags.
1153    
1154    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1155    
1156    19. libpcre.pc was being incorrectly installed as executable.
1157    
1158    20. A couple of places in pcretest check for end-of-line by looking for '\n';
1159        it now also looks for '\r' so that it will work unmodified on Windows.
1160    
1161    21. Added Google's contributed C++ wrapper to the distribution.
1162    
1163    22. Added some untidy missing memory free() calls in pcretest, to keep
1164        Electric Fence happy when testing.
1165    
1166    
1167    
1168    Version 5.0 13-Sep-04
1169    ---------------------
1170    
1171     1. Internal change: literal characters are no longer packed up into items
1172        containing multiple characters in a single byte-string. Each character
1173        is now matched using a separate opcode. However, there may be more than one
1174        byte in the character in UTF-8 mode.
1175    
1176     2. The pcre_callout_block structure has two new fields: pattern_position and
1177        next_item_length. These contain the offset in the pattern to the next match
1178        item, and its length, respectively.
1179    
1180     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1181        insertion of callouts before each pattern item. Added the /C option to
1182        pcretest to make use of this.
1183    
1184     4. On the advice of a Windows user, the lines
1185    
1186          #if defined(_WIN32) || defined(WIN32)
1187          _setmode( _fileno( stdout ), 0x8000 );
1188          #endif  /* defined(_WIN32) || defined(WIN32) */
1189    
1190        have been added to the source of pcretest. This apparently does useful
1191        magic in relation to line terminators.
1192    
1193     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1194        for the benefit of those environments where the "b" makes a difference.
1195    
1196     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1197        to know about it. I have put a hack into configure.in that adds in code
1198        to set GCC=yes if CC=icc. This seems to end up at a point in the
1199        generated configure script that is early enough to affect the setting of
1200        compiler options, which is what is needed, but I have no means of testing
1201        whether it really works. (The user who reported this had patched the
1202        generated configure script, which of course I cannot do.)
1203    
1204        LATER: After change 22 below (new libtool files), the configure script
1205        seems to know about icc (and also ecc). Therefore, I have commented out
1206        this hack in configure.in.
1207    
1208     7. Added support for pkg-config (2 patches were sent in).
1209    
1210     8. Negated POSIX character classes that used a combination of internal tables
1211        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1212        [[:^ascii]]. Typically, they would match almost any characters. The other
1213        POSIX classes were not broken in this way.
1214    
1215     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1216        to find the match, as PCRE was deluded into thinking that the match had to
1217        start at the start point or following a newline. The same bug applied to
1218        patterns with negative forward assertions or any backward assertions
1219        preceding ".*" at the start, unless the pattern required a fixed first
1220        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1221    
1222    10. In UTF-8 mode, when moving forwards in the subject after a failed match
1223        starting at the last subject character, bytes beyond the end of the subject
1224        string were read.
1225    
1226    11. Renamed the variable "class" as "classbits" to make life easier for C++
1227        users. (Previously there was a macro definition, but it apparently wasn't
1228        enough.)
1229    
1230    12. Added the new field "tables" to the extra data so that tables can be passed
1231        in at exec time, or the internal tables can be re-selected. This allows
1232        a compiled regex to be saved and re-used at a later time by a different
1233        program that might have everything at different addresses.
1234    
1235    13. Modified the pcre-config script so that, when run on Solaris, it shows a
1236        -R library as well as a -L library.
1237    
1238    14. The debugging options of pcretest (-d on the command line or D on a
1239        pattern) showed incorrect output for anything following an extended class
1240        that contained multibyte characters and which was followed by a quantifier.
1241    
1242    15. Added optional support for general category Unicode character properties
1243        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1244        support. It adds about 90K to the size of the library. The meanings of the
1245        inbuilt class escapes such as \d and \s have NOT been changed.
1246    
1247    16. Updated pcredemo.c to include calls to free() to release the memory for the
1248        compiled pattern.
1249    
1250    17. The generated file chartables.c was being created in the source directory
1251        instead of in the building directory. This caused the build to fail if the
1252        source directory was different from the building directory, and was
1253        read-only.
1254    
1255    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1256        file. No doubt somebody will tell me if they don't make sense... Also added
1257        Dan Mooney's comments about building on OpenVMS.
1258    
1259    19. Added support for partial matching via the PCRE_PARTIAL option for
1260        pcre_exec() and the \P data escape in pcretest.
1261    
1262    20. Extended pcretest with 3 new pattern features:
1263    
1264        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
1265              write the compiled pattern to the file whose name is "rest-of-line".
1266              This is a straight binary dump of the data, with the saved pointer to
1267              the character tables forced to be NULL. The study data, if any, is
1268              written too. After writing, pcretest reads a new pattern.
1269    
1270        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1271              compiled pattern from the given file. There must not be any
1272              occurrences of "<" in the file name (pretty unlikely); if there are,
1273              pcretest will instead treat the initial "<" as a pattern delimiter.
1274              After reading in the pattern, pcretest goes on to read data lines as
1275              usual.
1276    
1277        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1278              and 16-bit fields in a compiled pattern, to simulate a pattern that
1279              was compiled on a host of opposite endianness.
1280    
1281    21. The pcre-exec() function can now cope with patterns that were compiled on
1282        hosts of opposite endianness, with this restriction:
1283    
1284          As for any compiled expression that is saved and used later, the tables
1285          pointer field cannot be preserved; the extra_data field in the arguments
1286          to pcre_exec() should be used to pass in a tables address if a value
1287          other than the default internal tables were used at compile time.
1288    
1289    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1290        now diagnosed as an error. Previously, most of the time, a negative number
1291        would have been treated as zero, but if in addition "ovector" was passed as
1292        NULL, a crash could occur.
1293    
1294    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1295        new versions from the libtool 1.5 distribution (the last one is a copy of
1296        a file called libtool.m4). This seems to have fixed the need to patch
1297        "configure" to support Darwin 1.3 (which I used to do). However, I still
1298        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1299        workstation).
1300    
1301    24. Changed the PCRE licence to be the more standard "BSD" licence.
1302    
1303    
1304    Version 4.5 01-Dec-03
1305    ---------------------
1306    
1307     1. There has been some re-arrangement of the code for the match() function so
1308        that it can be compiled in a version that does not call itself recursively.
1309        Instead, it keeps those local variables that need separate instances for
1310        each "recursion" in a frame on the heap, and gets/frees frames whenever it
1311        needs to "recurse". Keeping track of where control must go is done by means
1312        of setjmp/longjmp. The whole thing is implemented by a set of macros that
1313        hide most of the details from the main code, and operates only if
1314        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1315        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1316        operating.
1317    
1318        To make it easier for callers to provide specially tailored get/free
1319        functions for this usage, two new functions, pcre_stack_malloc, and
1320        pcre_stack_free, are used. They are always called in strict stacking order,
1321        and the size of block requested is always the same.
1322    
1323        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1324        PCRE has been compiled to use the stack or the heap for recursion. The
1325        -C option of pcretest uses this to show which version is compiled.
1326    
1327        A new data escape \S, is added to pcretest; it causes the amounts of store
1328        obtained and freed by both kinds of malloc/free at match time to be added
1329        to the output.
1330    
1331     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1332        what's available on my current Linux desktop machine.
1333    
1334     3. When matching a UTF-8 string, the test for a valid string at the start has
1335        been extended. If start_offset is not zero, PCRE now checks that it points
1336        to a byte that is the start of a UTF-8 character. If not, it returns
1337        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1338        this is necessary because there may be backward assertions in the pattern.
1339        When matching the same subject several times, it may save resources to use
1340        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1341    
1342     4. The code for checking the validity of UTF-8 strings has been tightened so
1343        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1344        containing "overlong sequences".
1345    
1346     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1347        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1348        should have been "&", but it just so happened that all the cases this let
1349        through by mistake were picked up later in the function.
1350    
1351     6. I had used a variable called "isblank" - this is a C99 function, causing
1352        some compilers to warn. To avoid this, I renamed it (as "blankclass").
1353    
1354     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1355        prompting; (b) run "./pcretest /dev/null" at the start of the test script
1356        so the version is shown; (c) stop "make test" echoing "./RunTest".
1357    
1358     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1359    
1360     9. The prototype for memmove() for systems that don't have it was using
1361        size_t, but the inclusion of the header that defines size_t was later. I've
1362        moved the #includes for the C headers earlier to avoid this.
1363    
1364    10. Added some adjustments to the code to make it easier to compiler on certain
1365        special systems:
1366    
1367          (a) Some "const" qualifiers were missing.
1368          (b) Added the macro EXPORT before all exported functions; by default this
1369              is defined to be empty.
1370          (c) Changed the dftables auxiliary program (that builds chartables.c) so
1371              that it reads its output file name as an argument instead of writing
1372              to the standard output and assuming this can be redirected.
1373    
1374    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
1375        class containing characters with values greater than 255, PCRE compilation
1376        went into a loop.
1377    
1378    12. A recursive reference to a subpattern that was within another subpattern
1379        that had a minimum quantifier of zero caused PCRE to crash. For example,
1380        (x(y(?2))z)? provoked this bug with a subject that got as far as the
1381        recursion. If the recursively-called subpattern itself had a zero repeat,
1382        that was OK.
1383    
1384    13. In pcretest, the buffer for reading a data line was set at 30K, but the
1385        buffer into which it was copied (for escape processing) was still set at
1386        1024, so long lines caused crashes.
1387    
1388    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
1389        "internal error: code overflow...". This applied to any character class
1390        that was followed by a possessive quantifier.
1391    
1392    15. Modified the Makefile to add libpcre.la as a prerequisite for
1393        libpcreposix.la because I was told this is needed for a parallel build to
1394        work.
1395    
1396    16. If a pattern that contained .* following optional items at the start was
1397        studied, the wrong optimizing data was generated, leading to matching
1398        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
1399        matching string must start with a or b or c. The correct conclusion for
1400        this pattern is that a match can start with any character.
1401    
1402    
1403    Version 4.4 13-Aug-03
1404    ---------------------
1405    
1406     1. In UTF-8 mode, a character class containing characters with values between
1407        127 and 255 was not handled correctly if the compiled pattern was studied.
1408        In fixing this, I have also improved the studying algorithm for such
1409        classes (slightly).
1410    
1411     2. Three internal functions had redundant arguments passed to them. Removal
1412        might give a very teeny performance improvement.
1413    
1414     3. Documentation bug: the value of the capture_top field in a callout is *one
1415        more than* the number of the hightest numbered captured substring.
1416    
1417     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
1418        in incorrectly linking with a previously installed version. They now link
1419        explicitly with libpcre.la.
1420    
1421     5. configure.in no longer needs to recognize Cygwin specially.
1422    
1423     6. A problem in pcre.in for Windows platforms is fixed.
1424    
1425     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
1426        pcretest, it used to include the size of the study block as part of its
1427        output. Unfortunately, the structure contains a field that has a different
1428        size on different hardware architectures. This meant that the tests that
1429        showed this size failed. As the block is currently always of a fixed size,
1430        this information isn't actually particularly useful in pcretest output, so
1431        I have just removed it.
1432    
1433     8. Three pre-processor statements accidentally did not start in column 1.
1434        Sadly, there are *still* compilers around that complain, even though
1435        standard C has not required this for well over a decade. Sigh.
1436    
1437     9. In pcretest, the code for checking callouts passed small integers in the
1438        callout_data field, which is a void * field. However, some picky compilers
1439        complained about the casts involved for this on 64-bit systems. Now
1440        pcretest passes the address of the small integer instead, which should get
1441        rid of the warnings.
1442    
1443    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
1444        both compile and run time, and gives an error if an invalid UTF-8 sequence
1445        is found. There is a option for disabling this check in cases where the
1446        string is known to be correct and/or the maximum performance is wanted.
1447    
1448    11. In response to a bug report, I changed one line in Makefile.in from
1449    
1450            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
1451        to
1452            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
1453    
1454        to look similar to other lines, but I have no way of telling whether this
1455        is the right thing to do, as I do not use Windows. No doubt I'll get told
1456        if it's wrong...
1457    
1458    
1459    Version 4.3 21-May-03
1460    ---------------------
1461    
1462    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
1463       Makefile.
1464    
1465    2. Some refactoring to improve the quality of the code:
1466    
1467       (i)   The utf8_table... variables are now declared "const".
1468    
1469       (ii)  The code for \cx, which used the "case flipping" table to upper case
1470             lower case letters, now just substracts 32. This is ASCII-specific,
1471             but the whole concept of \cx is ASCII-specific, so it seems
1472             reasonable.
1473    
1474       (iii) PCRE was using its character types table to recognize decimal and
1475             hexadecimal digits in the pattern. This is silly, because it handles
1476             only 0-9, a-f, and A-F, but the character types table is locale-
1477             specific, which means strange things might happen. A private
1478             table is now used for this - though it costs 256 bytes, a table is
1479             much faster than multiple explicit tests. Of course, the standard
1480             character types table is still used for matching digits in subject
1481             strings against \d.
1482    
1483       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
1484             ending in _t are). So I've renamed it as ESC_tee.
1485    
1486    3. The first argument for regexec() in the POSIX wrapper should have been
1487       defined as "const".
1488    
1489    4. Changed pcretest to use malloc() for its buffers so that they can be
1490       Electric Fenced for debugging.
1491    
1492    5. There were several places in the code where, in UTF-8 mode, PCRE would try
1493       to read one or more bytes before the start of the subject string. Often this
1494       had no effect on PCRE's behaviour, but in some circumstances it could
1495       provoke a segmentation fault.
1496    
1497    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
1498       to try to read one or more bytes before the start of the subject string.
1499    
1500    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
1501       UTF-8 support could misbehave in various ways if the subject string
1502       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
1503       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
1504       back over UTF-8 characters.)
1505    
1506    
1507    Version 4.2 14-Apr-03
1508    ---------------------
1509    
1510    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
1511    
1512    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
1513         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
1514         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
1515         [WIN_PREFIX]: new variable, "cyg" for Cygwin
1516         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
1517           and BUILD_EXEEXT
1518         Note: automatic setting of the BUILD variables is not yet working
1519         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
1520           compile-time but not at link-time
1521         [LINK]: use for linking executables only
1522         make different versions for Windows and non-Windows
1523         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
1524           libraries
1525         [LINK_FOR_BUILD]: new variable
1526         [OBJEXT]: use throughout
1527         [EXEEXT]: use throughout
1528         <winshared>: new target
1529         <wininstall>: new target
1530         <dftables.o>: use native compiler
1531         <dftables>: use native linker
1532         <install>: handle Windows platform correctly
1533         <clean>: ditto
1534         <check>: ditto
1535         copy DLL to top builddir before testing
1536    
1537       As part of these changes, -no-undefined was removed again. This was reported
1538       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
1539       in any case.
1540    
1541    3. Some tidies to get rid of compiler warnings:
1542    
1543       . In the match_data structure, match_limit was an unsigned long int, whereas
1544         match_call_count was an int. I've made them both unsigned long ints.
1545    
1546       . In pcretest the fact that a const uschar * doesn't automatically cast to
1547         a void * provoked a warning.
1548    
1549       . Turning on some more compiler warnings threw up some "shadow" variables
1550         and a few more missing casts.
1551    
1552    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
1553       option, a class that contained a single character with a value between 128
1554       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
1555    
1556    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
1557       option, a class that contained several characters, but with at least one
1558       whose value was between 128 and 255 caused PCRE to crash.
1559    
1560    
1561    Version 4.1 12-Mar-03
1562    ---------------------
1563    
1564    1. Compiling with gcc -pedantic found a couple of places where casts were
1565    needed, and a string in dftables.c that was longer than standard compilers are
1566    required to support.
1567    
1568    2. Compiling with Sun's compiler found a few more places where the code could
1569    be tidied up in order to avoid warnings.
1570    
1571    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
1572    first of these names is deprecated in the latest Autoconf in favour of the name
1573    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
1574    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
1575    analogy I have changed it to CFLAGS_FOR_BUILD.
1576    
1577    4. Added -no-undefined to the linking command in the Makefile, because this is
1578    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
1579    linking step for the pcreposix library.
1580    
1581    5. PCRE was failing to diagnose the case of two named groups with the same
1582    name.
1583    
1584    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
1585    literal character that is needed in the subject for a match, and scans along to
1586    ensure that it is present before embarking on the full matching process. This
1587    saves time in cases of nested unlimited repeats that are never going to match.
1588    Problem: the scan can take a lot of time if the subject is very long (e.g.
1589    megabytes), thus penalizing straightforward matches. It is now done only if the
1590    amount of subject to be scanned is less than 1000 bytes.
1591    
1592    7. A lesser problem with the same optimization is that it was recording the
1593    first character of an anchored pattern as "needed", thus provoking a search
1594    right along the subject, even when the first match of the pattern was going to
1595    fail. The "needed" character is now not set for anchored patterns, unless it
1596    follows something in the pattern that is of non-fixed length. Thus, it still
1597    fulfils its original purpose of finding quick non-matches in cases of nested
1598    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
1599    
1600    
1601    Version 4.0 17-Feb-03
1602    ---------------------
1603    
1604    1. If a comment in an extended regex that started immediately after a meta-item
1605    extended to the end of string, PCRE compiled incorrect data. This could lead to
1606    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
1607    
1608    2. Moved to autoconf 2.53 and libtool 1.4.2.
1609    
1610    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
1611    the special perltest8 script is no longer needed - all the tests can be run
1612    from a single perltest script.
1613    
1614    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
1615    by \s. It has now been removed in PCRE. This means it isn't recognized as
1616    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
1617    class [:space:] *does* include VT, thereby creating a mess.
1618    
1619    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
1620    space and tab.
1621    
1622    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
1623    its new features into the main test script, reducing the number of scripts.
1624    
1625    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
1626    were backward compatible, and made the (?i) apply to the whole pattern, as if
1627    /i were given. Now it behaves more logically, and applies the option setting
1628    only to what follows. PCRE has been changed to follow suit. However, if it
1629    finds options settings right at the start of the pattern, it extracts them into
1630    the global options, as before. Thus, they show up in the info data.
1631    
1632    8. Added support for the \Q...\E escape sequence. Characters in between are
1633    treated as literals. This is slightly different from Perl in that $ and @ are
1634    also handled as literals inside the quotes. In Perl, they will cause variable
1635    interpolation. Note the following examples:
1636    
1637        Pattern            PCRE matches      Perl matches
1638    
1639        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
1640        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
1641        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
1642    
1643    For compatibility with Perl, \Q...\E sequences are recognized inside character
1644    classes as well as outside them.
1645    
1646    9. Re-organized 3 code statements in pcretest to avoid "overflow in
1647    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
1648    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
1649    signed/unsigned warnings.
1650    
1651    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
1652    option for pcretest, so I've replaced it by a simple function that does just
1653    that job.
1654    
1655    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
1656    "pcregrep -".
1657    
1658    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
1659    Java package. This provides some syntactic sugar for simple cases of what my
1660    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
1661    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
1662    item, you can use this simplified notation. Note that only makes sense with
1663    greedy quantifiers. Consequently, the use of the possessive quantifier forces
1664    greediness, whatever the setting of the PCRE_UNGREEDY option.
1665    
1666    13. A change of greediness default within a pattern was not taking effect at
1667    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
1668    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
1669    was abstracted outside.
1670    
1671    14. PCRE now supports the \G assertion. It is true when the current matching
1672    position is at the start point of the match. This differs from \A when the
1673    starting offset is non-zero. Used with the /g option of pcretest (or similar
1674    code), it works in the same way as it does for Perl's /g option. If all
1675    alternatives of a regex begin with \G, the expression is anchored to the start
1676    match position, and the "anchored" flag is set in the compiled expression.
1677    
1678    15. Some bugs concerning the handling of certain option changes within patterns
1679    have been fixed. These applied to options other than (?ims). For example,
1680    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
1681    been the other way round. Some of this was related to change 7 above.
1682    
1683    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
1684    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
1685    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
1686    POSIX classes only within a class (e.g. /[[:alpha:]]/).
1687    
1688    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
1689    mode. Unlike ".", it always matches newline, whatever the setting of
1690    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
1691    assertions. Perl allows it, but it doesn't (in general) work because it can't
1692    calculate the length of the lookbehind. At least, that's the case for Perl
1693    5.8.0 - I've been told they are going to document that it doesn't work in
1694    future.
1695    
1696    18. Added an error diagnosis for escapes that PCRE does not support: these are
1697    \L, \l, \N, \P, \p, \U, \u, and \X.
1698    
1699    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
1700    reading past the end of the pattern in cases such as /[abcd/.
1701    
1702    20. PCRE was getting more memory than necessary for patterns with classes that
1703    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
1704    
1705    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
1706    compiling PCRE for use with Virtual Pascal.
1707    
1708    22. Small fix to the Makefile to make it work properly if the build is done
1709    outside the source tree.
1710    
1711    23. Added a new extension: a condition to go with recursion. If a conditional
1712    subpattern starts with (?(R) the "true" branch is used if recursion has
1713    happened, whereas the "false" branch is used only at the top level.
1714    
1715    24. When there was a very long string of literal characters (over 255 bytes
1716    without UTF support, over 250 bytes with UTF support), the computation of how
1717    much memory was required could be incorrect, leading to segfaults or other
1718    strange effects.
1719    
1720    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
1721    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
1722    there was a subsequent back reference to those brackets. This meant that, for
1723    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
1724    possible to check for precisely this case. All we can do is abandon the
1725    optimization if .* occurs inside capturing brackets when there are any back
1726    references whatsoever. (See below for a better fix that came later.)
1727    
1728    26. The handling of the optimization for finding the first character of a
1729    non-anchored pattern, and for finding a character that is required later in the
1730    match were failing in some cases. This didn't break the matching; it just
1731    failed to optimize when it could. The way this is done has been re-implemented.
1732    
1733    27. Fixed typo in error message for invalid (?R item (it said "(?p").
1734    
1735    28. Added a new feature that provides some of the functionality that Perl
1736    provides with (?{...}). The facility is termed a "callout". The way it is done
1737    in PCRE is for the caller to provide an optional function, by setting
1738    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
1739    global variable. By default it is unset, which disables all calling out. To get
1740    the function called, the regex must include (?C) at appropriate points. This
1741    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
1742    This provides a means of identifying different callout points. When PCRE
1743    reaches such a point in the regex, if pcre_callout has been set, the external
1744    function is called. It is provided with data in a structure called
1745    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
1746    matching continues; if it returns a non-zero value, the match at the current
1747    point fails. However, backtracking will occur if possible. [This was changed
1748    later and other features added - see item 49 below.]
1749    
1750    29. pcretest is upgraded to test the callout functionality. It provides a
1751    callout function that displays information. By default, it shows the start of
1752    the match and the current position in the text. There are some new data escapes
1753    to vary what happens:
1754    
1755        \C+         in addition, show current contents of captured substrings
1756        \C-         do not supply a callout function
1757        \C!n        return 1 when callout number n is reached
1758        \C!n!m      return 1 when callout number n is reached for the mth time
1759    
1760    30. If pcregrep was called with the -l option and just a single file name, it
1761    output "<stdin>" if a match was found, instead of the file name.
1762    
1763    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
1764    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
1765    pcre_exec(). This saves a malloc/free per call. The default value of
1766    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
1767    when configuring.
1768    
1769    32. The default maximum size of a compiled pattern is 64K. There have been a
1770    few cases of people hitting this limit. The code now uses macros to handle the
1771    storing of links as offsets within the compiled pattern. It defaults to 2-byte
1772    links, but this can be changed to 3 or 4 bytes by --with-link-size when
1773    configuring. Tests 2 and 5 work only with 2-byte links because they output
1774    debugging information about compiled patterns.
1775    
1776    33. Internal code re-arrangements:
1777    
1778    (a) Moved the debugging function for printing out a compiled regex into
1779        its own source file (printint.c) and used #include to pull it into
1780        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
1781        separate copies.
1782    
1783    (b) Defined the list of op-code names for debugging as a macro in
1784        internal.h so that it is next to the definition of the opcodes.
1785    
1786    (c) Defined a table of op-code lengths for simpler skipping along compiled
1787        code. This is again a macro in internal.h so that it is next to the
1788        definition of the opcodes.
1789    
1790    34. Added support for recursive calls to individual subpatterns, along the
1791    lines of Robin Houston's patch (but implemented somewhat differently).
1792    
1793    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
1794    allow it to read and process whole directories in Win32. This code was
1795    contributed by Lionel Fourquaux; it has not been tested by me.
1796    
1797    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
1798    used to name a group. Names consist of alphanumerics and underscores, and must
1799    be unique. Back references use the syntax (?P=name) and recursive calls use
1800    (?P>name) which is a PCRE extension to the Python extension. Groups still have
1801    numbers. The function pcre_fullinfo() can be used after compilation to extract
1802    a name/number map. There are three relevant calls:
1803    
1804      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
1805      PCRE_INFO_NAMECOUNT            yields the number of entries
1806      PCRE_INFO_NAMETABLE            yields a pointer to the map.
1807    
1808    The map is a vector of fixed-size entries. The size of each entry depends on
1809    the length of the longest name used. The first two bytes of each entry are the
1810    group number, most significant byte first. There follows the corresponding
1811    name, zero terminated. The names are in alphabetical order.
1812    
1813    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
1814    case instead of 255. Making it the same both with and without UTF-8 support
1815    means that the same test output works with both.
1816    
1817    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
1818    calling malloc() with a zero argument.
1819    
1820    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
1821    optimization. I've improved things by keeping a bitmap of backreferences with
1822    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
1823    fact referenced, the optimization can be applied. It is unlikely that a
1824    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
1825    the match to follow \n) will appear inside brackets with a number greater than
1826    31, but if it does, any back reference > 31 suppresses the optimization.
1827    
1828    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
1829    of disabling numbered capturing parentheses. Any opening parenthesis that is
1830    not followed by ? behaves as if it were followed by ?: but named parentheses
1831    can still be used for capturing (and they will acquire numbers in the usual
1832    way).
1833    
1834    41. Redesigned the return codes from the match() function into yes/no/error so
1835    that errors can be passed back from deep inside the nested calls. A malloc
1836    failure while inside a recursive subpattern call now causes the
1837    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
1838    
1839    42. It is now possible to set a limit on the number of times the match()
1840    function is called in a call to pcre_exec(). This facility makes it possible to
1841    limit the amount of recursion and backtracking, though not in a directly
1842    obvious way, because the match() function is used in a number of different
1843    circumstances. The count starts from zero for each position in the subject
1844    string (for non-anchored patterns). The default limit is, for compatibility, a
1845    large number, namely 10 000 000. You can change this in two ways:
1846    
1847    (a) When configuring PCRE before making, you can use --with-match-limit=n
1848        to set a default value for the compiled library.
1849    
1850    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
1851        a different value is set. See 45 below.
1852    
1853    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
1854    
1855    43. Added a new function pcre_config(int, void *) to enable run-time extraction
1856    of things that can be changed at compile time. The first argument specifies
1857    what is wanted and the second points to where the information is to be placed.
1858    The current list of available information is:
1859    
1860      PCRE_CONFIG_UTF8
1861    
1862    The output is an integer that is set to one if UTF-8 support is available;
1863    otherwise it is set to zero.
1864    
1865      PCRE_CONFIG_NEWLINE
1866    
1867    The output is an integer that it set to the value of the code that is used for
1868    newline. It is either LF (10) or CR (13).
1869    
1870      PCRE_CONFIG_LINK_SIZE
1871    
1872    The output is an integer that contains the number of bytes used for internal
1873    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
1874    
1875      PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
1876    
1877    The output is an integer that contains the threshold above which the POSIX
1878    interface uses malloc() for output vectors. See item 31 above.
1879    
1880      PCRE_CONFIG_MATCH_LIMIT
1881    
1882    The output is an unsigned integer that contains the default limit of the number
1883    of match() calls in a pcre_exec() execution. See 42 above.
1884    
1885    44. pcretest has been upgraded by the addition of the -C option. This causes it
1886    to extract all the available output from the new pcre_config() function, and to
1887    output it. The program then exits immediately.
1888    
1889    45. A need has arisen to pass over additional data with calls to pcre_exec() in
1890    order to support additional features. One way would have been to define
1891    pcre_exec2() (for example) with extra arguments, but this would not have been
1892    extensible, and would also have required all calls to the original function to
1893    be mapped to the new one. Instead, I have chosen to extend the mechanism that
1894    is used for passing in "extra" data from pcre_study().
1895    
1896    The pcre_extra structure is now exposed and defined in pcre.h. It currently
1897    contains the following fields:
1898    
1899      flags         a bitmap indicating which of the following fields are set
1900      study_data    opaque data from pcre_study()
1901      match_limit   a way of specifying a limit on match() calls for a specific
1902                      call to pcre_exec()
1903      callout_data  data for callouts (see 49 below)
1904    
1905    The flag bits are also defined in pcre.h, and are
1906    
1907      PCRE_EXTRA_STUDY_DATA
1908      PCRE_EXTRA_MATCH_LIMIT
1909      PCRE_EXTRA_CALLOUT_DATA
1910    
1911    The pcre_study() function now returns one of these new pcre_extra blocks, with
1912    the actual study data pointed to by the study_data field, and the
1913    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
1914    before. That is, this change is entirely upwards-compatible and requires no
1915    change to existing code.
1916    
1917    If you want to pass in additional data to pcre_exec(), you can either place it
1918    in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
1919    block.
1920    
1921    46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
1922    data string contains the escape sequence \M, pcretest calls pcre_exec() several
1923    times with different match limits, until it finds the minimum value needed for
1924    pcre_exec() to complete. The value is then output. This can be instructive; for
1925    most simple matches the number is quite small, but for pathological cases it
1926    gets very large very quickly.
1927    
1928    47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
1929    returns the size of the data block pointed to by the study_data field in a
1930    pcre_extra block, that is, the value that was passed as the argument to
1931    pcre_malloc() when PCRE was getting memory in which to place the information
1932    created by pcre_study(). The fourth argument should point to a size_t variable.
1933    pcretest has been extended so that this information is shown after a successful
1934    pcre_study() call when information about the compiled regex is being displayed.
1935    
1936    48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
1937    because what follows is always an absolute path. (Later: it turns out that this
1938    is more than cosmetic for MinGW, because it doesn't like empty path
1939    components.)
1940    
1941    49. Some changes have been made to the callout feature (see 28 above):
1942    
1943    (i)  A callout function now has three choices for what it returns:
1944    
1945           0  =>  success, carry on matching
1946         > 0  =>  failure at this point, but backtrack if possible
1947         < 0  =>  serious error, return this value from pcre_exec()
1948    
1949         Negative values should normally be chosen from the set of PCRE_ERROR_xxx
1950         values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
1951         "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
1952         use by callout functions. It will never be used by PCRE itself.
1953    
1954    (ii) The pcre_extra structure (see 45 above) has a void * field called
1955         callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
1956         pcre_callout_block structure has a field of the same name. The contents of
1957         the field passed in the pcre_extra structure are passed to the callout
1958         function in the corresponding field in the callout block. This makes it
1959         easier to use the same callout-containing regex from multiple threads. For
1960         testing, the pcretest program has a new data escape
1961    
1962           \C*n        pass the number n (may be negative) as callout_data
1963    
1964         If the callout function in pcretest receives a non-zero value as
1965         callout_data, it returns that value.
1966    
1967    50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
1968    there were some redundant $(CFLAGS) in commands that are now specified as
1969    $(LINK), which already includes $(CFLAGS).
1970    
1971    51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
1972    has been compiled with UTF-8 support *and* pcre_compile() has been compiled
1973    with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
1974    one-byte characters throughout. Note that case-insensitive matching applies
1975    only to characters whose values are less than 256. PCRE doesn't support the
1976    notion of cases for higher-valued characters.
1977    
1978    (i)   A character class whose characters are all within 0-255 is handled as
1979          a bit map, and the map is inverted for negative classes. Previously, a
1980          character > 255 always failed to match such a class; however it should
1981          match if the class was a negative one (e.g. [^ab]). This has been fixed.
1982    
1983    (ii)  A negated character class with a single character < 255 is coded as
1984          "not this character" (OP_NOT). This wasn't working properly when the test
1985          character was multibyte, either singly or repeated.
1986    
1987    (iii) Repeats of multibyte characters are now handled correctly in UTF-8
1988          mode, for example: \x{100}{2,3}.
1989    
1990    (iv)  The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
1991          singly or repeated) now correctly test multibyte characters. However,
1992          PCRE doesn't recognize any characters with values greater than 255 as
1993          digits, spaces, or word characters. Such characters always match \D, \S,
1994          and \W, and never match \d, \s, or \w.
1995    
1996    (v)   Classes may now contain characters and character ranges with values
1997          greater than 255. For example: [ab\x{100}-\x{400}].
1998    
1999    (vi)  pcregrep now has a --utf-8 option (synonym -u) which makes it call
2000          PCRE in UTF-8 mode.
2001    
2002    52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
2003    PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
2004    retained for backwards compatibility. (Note that LASTLITERAL is also a byte
2005    value.)
2006    
2007    53. The single man page has become too large. I have therefore split it up into
2008    a number of separate man pages. These also give rise to individual HTML pages;
2009    these are now put in a separate directory, and there is an index.html page that
2010    lists them all. Some hyperlinking between the pages has been installed.
2011    
2012    54. Added convenience functions for handling named capturing parentheses.
2013    
2014    55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2015    aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2016    true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2017    are faulted.
2018    
2019    56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2020    calling configure. These values are used when compiling the dftables.c program
2021    which is run to generate the source of the default character tables. They
2022    default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2023    you will need to set these values.
2024    
2025    57. Updated the building process for Windows DLL, as provided by Fred Cox.
2026    
2027    
2028    Version 3.9 02-Jan-02
2029    ---------------------
2030    
2031    1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2032    
2033    2. If --disable-static was given, the building process failed when trying to
2034    build pcretest and pcregrep. (For some reason it was using libtool to compile
2035    them, which is not right, as they aren't part of the library.)
2036    
2037    
2038  Version 3.8 18-Dec-01  Version 3.8 18-Dec-01
2039  ---------------------  ---------------------
2040    

Legend:
Removed from v.59  
changed lines
  Added in v.184

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12