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

Diff of /code/trunk/ChangeLog

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

revision 11 by nigel, Sat Feb 24 21:38:17 2007 UTC revision 216 by ph10, Wed Aug 15 14:35:57 2007 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4    Version 7.3 09-Aug-07
5    ---------------------
6    
7     1. In the rejigging of the build system that eventually resulted in 7.1, the
8        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
9        brackets there is not right, since it causes compilers to look for an
10        installed pcre.h, not the version that is in the source that is being
11        compiled (which of course may be different). I have changed it back to:
12    
13          #include "pcre.h"
14    
15        I have a vague recollection that the change was concerned with compiling in
16        different directories, but in the new build system, that is taken care of
17        by the VPATH setting the Makefile.
18    
19     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
20        when the subject happened to end in the byte 0x85 (e.g. if the last
21        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
22        characters but of course it shouldn't be taken as a newline when it is part
23        of another character. The bug was that, for an unlimited repeat of . in
24        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
25        characters when looking for a newline.
26    
27     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
28    
29     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
30        in debug output.
31    
32     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
33        long printing in the pcrecpp unittest when running under MinGW.
34    
35     6. ESC_K was left out of the EBCDIC table.
36    
37     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
38        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
39        limit also applies to "virtual nesting" when a pattern is recursive, and in
40        this case 1000 isn't so big. I have been able to remove this limit at the
41        expense of backing off one optimization in certain circumstances. Normally,
42        when pcre_exec() would call its internal match() function recursively and
43        immediately return the result unconditionally, it uses a "tail recursion"
44        feature to save stack. However, when a subpattern that can match an empty
45        string has an unlimited repetition quantifier, it no longer makes this
46        optimization. That gives it a stack frame in which to save the data for
47        checking that an empty string has been matched. Previously this was taken
48        from the 1000-entry workspace that had been reserved. So now there is no
49        explicit limit, but more stack is used.
50    
51     8. Applied Daniel's patches to solve problems with the import/export magic
52        syntax that is required for Windows, and which was going wrong for the
53        pcreposix and pcrecpp parts of the library. These were overlooked when this
54        problem was solved for the main library.
55    
56     9. There were some crude static tests to avoid integer overflow when computing
57        the size of patterns that contain repeated groups with explicit upper
58        limits. As the maximum quantifier is 65535, the maximum group length was
59        set at 30,000 so that the product of these two numbers did not overflow a
60        32-bit integer. However, it turns out that people want to use groups that
61        are longer than 30,000 bytes (though not repeat them that many times).
62        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
63        made it possible to implement the integer overflow checks in a much more
64        dynamic way, which I have now done. The artificial limitation on group
65        length has been removed - we now have only the limit on the total length of
66        the compiled pattern, which depends on the LINK_SIZE setting.
67    
68    10. Fixed a bug in the documentation for get/copy named substring when
69        duplicate names are permitted. If none of the named substrings are set, the
70        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
71        empty string.
72    
73    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
74        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
75        because the ] is interpreted as the first data character and the
76        terminating ] is not found. PCRE has been made compatible with Perl in this
77        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
78        cause memory overwriting.
79    
80    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
81        string has been matched (to stop an infinite loop). It was not recognizing
82        a conditional subpattern that could match an empty string if that
83        subpattern was within another subpattern. For example, it looped when
84        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
85        condition was not nested. This bug has been fixed.
86    
87    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
88        past the start of the subject in the presence of bytes with the top bit
89        set, for example "\x8aBCD".
90    
91    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
92        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
93    
94    14. Optimized (?!) to (*FAIL).
95    
96    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
97        This restricts code points to be within the range 0 to 0x10FFFF, excluding
98        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
99        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
100        does: it's just the validity check that is more restrictive.
101    
102    16. Inserted checks for integer overflows during escape sequence (backslash)
103        processing, and also fixed erroneous offset values for syntax errors during
104        backslash processing.
105    
106    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
107        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
108    
109    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
110        caused an overrun.
111    
112    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
113        something other than just ASCII characters) inside a group that had an
114        unlimited repeat caused a loop at compile time (while checking to see
115        whether the group could match an empty string).
116    
117    
118    Version 7.2 19-Jun-07
119    ---------------------
120    
121     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
122        which is apparently normally available under Windows.
123    
124     2. Re-jig the pcregrep tests with different newline settings in an attempt
125        to make them independent of the local environment's newline setting.
126    
127     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
128    
129     4. Some of the "internals" tests were previously cut out when the link size
130        was not 2, because the output contained actual offsets. The recent new
131        "Z" feature of pcretest means that these can be cut out, making the tests
132        usable with all link sizes.
133    
134     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
135        stack recursion. This gives a massive performance boost under BSD, but just
136        a small improvement under Linux. However, it saves one field in the frame
137        in all cases.
138    
139     6. Added more features from the forthcoming Perl 5.10:
140    
141        (a) (?-n) (where n is a string of digits) is a relative subroutine or
142            recursion call. It refers to the nth most recently opened parentheses.
143    
144        (b) (?+n) is also a relative subroutine call; it refers to the nth next
145            to be opened parentheses.
146    
147        (c) Conditions that refer to capturing parentheses can be specified
148            relatively, for example, (?(-2)... or (?(+3)...
149    
150        (d) \K resets the start of the current match so that everything before
151            is not part of it.
152    
153        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
154    
155        (f) \g{name} is another synonym - part of Perl 5.10's unification of
156            reference syntax.
157    
158        (g) (?| introduces a group in which the numbering of parentheses in each
159            alternative starts with the same number.
160    
161        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
162    
163     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
164        PCRE_INFO_JCHANGED.
165    
166     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
167        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
168        for detecting groups that can match an empty string.
169    
170     9. A pattern with a very large number of alternatives (more than several
171        hundred) was running out of internal workspace during the pre-compile
172        phase, where pcre_compile() figures out how much memory will be needed. A
173        bit of new cunning has reduced the workspace needed for groups with
174        alternatives. The 1000-alternative test pattern now uses 12 bytes of
175        workspace instead of running out of the 4096 that are available.
176    
177    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
178    
179    11. Applied patch from Google to remove an optimization that didn't quite work.
180        The report of the bug said:
181    
182          pcrecpp::RE("a*").FullMatch("aaa") matches, while
183          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
184          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
185    
186    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
187        it matched the wrong number of bytes.
188    
189    
190    Version 7.1 24-Apr-07
191    ---------------------
192    
193     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
194        that is more "standard", making use of automake and other Autotools. There
195        is some re-arrangement of the files and adjustment of comments consequent
196        on this.
197    
198     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
199        for recursive directory scanning broke on some systems because the files
200        are not scanned in any specific order and on different systems the order
201        was different. A call to "sort" has been inserted into RunGrepTest for the
202        approprate test as a short-term fix. In the longer term there may be an
203        alternative.
204    
205     3. I had an email from Eric Raymond about problems translating some of PCRE's
206        man pages to HTML (despite the fact that I distribute HTML pages, some
207        people do their own conversions for various reasons). The problems
208        concerned the use of low-level troff macros .br and .in. I have therefore
209        removed all such uses from the man pages (some were redundant, some could
210        be replaced by .nf/.fi pairs). The 132html script that I use to generate
211        HTML has been updated to handle .nf/.fi and to complain if it encounters
212        .br or .in.
213    
214     4. Updated comments in configure.ac that get placed in config.h.in and also
215        arranged for config.h to be included in the distribution, with the name
216        config.h.generic, for the benefit of those who have to compile without
217        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
218    
219     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
220        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
221        makevp.bat for the current PCRE, using the additional files
222        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
223    
224     6. A Windows user reported a minor discrepancy with test 2, which turned out
225        to be caused by a trailing space on an input line that had got lost in his
226        copy. The trailing space was an accident, so I've just removed it.
227    
228     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
229        that is needed.
230    
231     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
232        as "const" (a) because they are and (b) because it helps the PHP
233        maintainers who have recently made a script to detect big data structures
234        in the php code that should be moved to the .rodata section. I remembered
235        to update Builducptable as well, so it won't revert if ucptable.h is ever
236        re-created.
237    
238     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
239        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
240        order to be able to cut out the UTF-8 tables in the latter when UTF-8
241        support is not required. This saves 1.5-2K of code, which is important in
242        some applications.
243    
244        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
245        so as not to refer to the tables, even though these functions will never be
246        called when UTF-8 support is disabled. Otherwise there are problems with a
247        shared library.
248    
249    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
250    
251        (a) It was defining its arguments as char * instead of void *.
252    
253        (b) It was assuming that all moves were upwards in memory; this was true
254            a long time ago when I wrote it, but is no longer the case.
255    
256        The emulated memove() is provided for those environments that have neither
257        memmove() nor bcopy(). I didn't think anyone used it these days, but that
258        is clearly not the case, as these two bugs were recently reported.
259    
260    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
261        and Detrail to create the HTML documentation, the .txt form of the man
262        pages, and it removes trailing spaces from listed files. It also creates
263        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
264        case, it wraps all the #defines with #ifndefs. This script should be run
265        before "make dist".
266    
267    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
268        with Unicode property support.
269    
270        (a) For a maximizing quantifier, if the two different cases of the
271            character were of different lengths in their UTF-8 codings (there are
272            some cases like this - I found 11), and the matching function had to
273            back up over a mixture of the two cases, it incorrectly assumed they
274            were both the same length.
275    
276        (b) When PCRE was configured to use the heap rather than the stack for
277            recursion during matching, it was not correctly preserving the data for
278            the other case of a UTF-8 character when checking ahead for a match
279            while processing a minimizing repeat. If the check also involved
280            matching a wide character, but failed, corruption could cause an
281            erroneous result when trying to check for a repeat of the original
282            character.
283    
284    13. Some tidying changes to the testing mechanism:
285    
286        (a) The RunTest script now detects the internal link size and whether there
287            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
288            values substituted by "configure". (The RunGrepTest script already did
289            this for UTF-8.) The configure.ac script no longer substitutes the
290            relevant variables.
291    
292        (b) The debugging options /B and /D in pcretest show the compiled bytecode
293            with length and offset values. This means that the output is different
294            for different internal link sizes. Test 2 is skipped for link sizes
295            other than 2 because of this, bypassing the problem. Unfortunately,
296            there was also a test in test 3 (the locale tests) that used /B and
297            failed for link sizes other than 2. Rather than cut the whole test out,
298            I have added a new /Z option to pcretest that replaces the length and
299            offset values with spaces. This is now used to make test 3 independent
300            of link size. (Test 2 will be tidied up later.)
301    
302    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
303        segmentation fault instead of returning the appropriate error message.
304    
305    15. In multiline mode when the newline sequence was set to "any", the pattern
306        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
307        This doesn't seem right; it now treats the CRLF combination as the line
308        ending, and so does not match in that case. It's only a pattern such as ^$
309        that would hit this one: something like ^ABC$ would have failed after \r
310        and then tried again after \r\n.
311    
312    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
313        in an attempt to make files that differ only in their line terminators
314        compare equal. This works on Linux.
315    
316    17. Under certain error circumstances pcregrep might try to free random memory
317        as it exited. This is now fixed, thanks to valgrind.
318    
319    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
320        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
321        was because its rules for how to advance for /g after matching an empty
322        string at the end of a line did not allow for this case. They now check for
323        it specially.
324    
325    20. pcretest is supposed to handle patterns and data of any length, by
326        extending its buffers when necessary. It was getting this wrong when the
327        buffer for a data line had to be extended.
328    
329    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
330        CRLF as a newline sequence.
331    
332    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
333        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
334        I have nevertheless tidied it up.
335    
336    23. Added some casts to kill warnings from HP-UX ia64 compiler.
337    
338    24. Added a man page for pcre-config.
339    
340    
341    Version 7.0 19-Dec-06
342    ---------------------
343    
344     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
345        moving to gcc 4.1.1.
346    
347     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
348        sys/time.h, which is documented as needed for this function. It doesn't
349        seem to matter on Linux, but it showed up on some releases of OS X.
350    
351     3. It seems that there are systems where bytes whose values are greater than
352        127 match isprint() in the "C" locale. The "C" locale should be the
353        default when a C program starts up. In most systems, only ASCII printing
354        characters match isprint(). This difference caused the output from pcretest
355        to vary, making some of the tests fail. I have changed pcretest so that:
356    
357        (a) When it is outputting text in the compiled version of a pattern, bytes
358            other than 32-126 are always shown as hex escapes.
359    
360        (b) When it is outputting text that is a matched part of a subject string,
361            it does the same, unless a different locale has been set for the match
362            (using the /L modifier). In this case, it uses isprint() to decide.
363    
364     4. Fixed a major bug that caused incorrect computation of the amount of memory
365        required for a compiled pattern when options that changed within the
366        pattern affected the logic of the preliminary scan that determines the
367        length. The relevant options are -x, and -i in UTF-8 mode. The result was
368        that the computed length was too small. The symptoms of this bug were
369        either the PCRE error "internal error: code overflow" from pcre_compile(),
370        or a glibc crash with a message such as "pcretest: free(): invalid next
371        size (fast)". Examples of patterns that provoked this bug (shown in
372        pcretest format) are:
373    
374          /(?-x: )/x
375          /(?x)(?-x: \s*#\s*)/
376          /((?i)[\x{c0}])/8
377          /(?i:[\x{c0}])/8
378    
379        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
380        is now done differently.
381    
382     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
383        wrapper classes; (b) implement a new function in the C++ scanner that is
384        more efficient than the old way of doing things because it avoids levels of
385        recursion in the regex matching; (c) add a paragraph to the documentation
386        for the FullMatch() function.
387    
388     6. The escape sequence \n was being treated as whatever was defined as
389        "newline". Not only was this contrary to the documentation, which states
390        that \n is character 10 (hex 0A), but it also went horribly wrong when
391        "newline" was defined as CRLF. This has been fixed.
392    
393     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
394        was being set to -1 for the "end of line" case (supposedly a value that no
395        character can have). Though this value is never used (the check for end of
396        line is "zero bytes in current character"), it caused compiler complaints.
397        I've changed it to 0xffffffff.
398    
399     8. In pcre_version.c, the version string was being built by a sequence of
400        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
401        string (as it is for production releases) called a macro with an empty
402        argument. The C standard says the result of this is undefined. The gcc
403        compiler treats it as an empty string (which was what was wanted) but it is
404        reported that Visual C gives an error. The source has been hacked around to
405        avoid this problem.
406    
407     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
408        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
409        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
410        of them did).
411    
412    10. Originally, pcretest opened its input and output without "b"; then I was
413        told that "b" was needed in some environments, so it was added for release
414        5.0 to both the input and output. (It makes no difference on Unix-like
415        systems.) Later I was told that it is wrong for the input on Windows. I've
416        now abstracted the modes into two macros, to make it easier to fiddle with
417        them, and removed "b" from the input mode under Windows.
418    
419    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
420    
421    12. Added -help and --help to pcretest as an official way of being reminded
422        of the options.
423    
424    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
425        and pcrecpp.cc because they annoy compilers at high warning levels.
426    
427    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
428    
429    15. Fixed an occurrence of == in configure.ac that should have been = (shell
430        scripts are not C programs :-) and which was not noticed because it works
431        on Linux.
432    
433    16. pcretest is supposed to handle any length of pattern and data line (as one
434        line or as a continued sequence of lines) by extending its input buffer if
435        necessary. This feature was broken for very long pattern lines, leading to
436        a string of junk being passed to pcre_compile() if the pattern was longer
437        than about 50K.
438    
439    17. I have done a major re-factoring of the way pcre_compile() computes the
440        amount of memory needed for a compiled pattern. Previously, there was code
441        that made a preliminary scan of the pattern in order to do this. That was
442        OK when PCRE was new, but as the facilities have expanded, it has become
443        harder and harder to keep it in step with the real compile phase, and there
444        have been a number of bugs (see for example, 4 above). I have now found a
445        cunning way of running the real compile function in a "fake" mode that
446        enables it to compute how much memory it would need, while actually only
447        ever using a few hundred bytes of working memory and without too many
448        tests of the mode. This should make future maintenance and development
449        easier. A side effect of this work is that the limit of 200 on the nesting
450        depth of parentheses has been removed (though this was never a serious
451        limitation, I suspect). However, there is a downside: pcre_compile() now
452        runs more slowly than before (30% or more, depending on the pattern). I
453        hope this isn't a big issue. There is no effect on runtime performance.
454    
455    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
456        newline (only possible for the last line of a file) and it was a
457        pattern that set a locale (followed by /Lsomething), pcretest crashed.
458    
459    19. Added additional timing features to pcretest. (1) The -tm option now times
460        matching only, not compiling. (2) Both -t and -tm can be followed, as a
461        separate command line item, by a number that specifies the number of
462        repeats to use when timing. The default is 50000; this gives better
463        precision, but takes uncomfortably long for very large patterns.
464    
465    20. Extended pcre_study() to be more clever in cases where a branch of a
466        subpattern has no definite first character. For example, (a*|b*)[cd] would
467        previously give no result from pcre_study(). Now it recognizes that the
468        first character must be a, b, c, or d.
469    
470    21. There was an incorrect error "recursive call could loop indefinitely" if
471        a subpattern (or the entire pattern) that was being tested for matching an
472        empty string contained only one non-empty item after a nested subpattern.
473        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
474        incorrectly, because the \d was being skipped in the check.
475    
476    22. The pcretest program now has a new pattern option /B and a command line
477        option -b, which is equivalent to adding /B to every pattern. This causes
478        it to show the compiled bytecode, without the additional information that
479        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
480        is the same as /B/I).
481    
482    23. A new optimization is now able automatically to treat some sequences such
483        as a*b as a*+b. More specifically, if something simple (such as a character
484        or a simple class like \d) has an unlimited quantifier, and is followed by
485        something that cannot possibly match the quantified thing, the quantifier
486        is automatically "possessified".
487    
488    24. A recursive reference to a subpattern whose number was greater than 39
489        went wrong under certain circumstances in UTF-8 mode. This bug could also
490        have affected the operation of pcre_study().
491    
492    25. Realized that a little bit of performance could be had by replacing
493        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
494    
495    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
496    
497    27. Possessive quantifiers such as a++ were previously implemented by turning
498        them into atomic groups such as ($>a+). Now they have their own opcodes,
499        which improves performance. This includes the automatically created ones
500        from 23 above.
501    
502    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
503        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
504        the first matched character to be a colon. This applied both to named and
505        numbered groups.
506    
507    29. The ucpinternal.h header file was missing its idempotency #ifdef.
508    
509    30. I was sent a "project" file called libpcre.a.dev which I understand makes
510        building PCRE on Windows easier, so I have included it in the distribution.
511    
512    31. There is now a check in pcretest against a ridiculously large number being
513        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
514        loop, the loop is abandoned.
515    
516    32. Forward references to subpatterns in conditions such as (?(2)...) where
517        subpattern 2 is defined later cause pcre_compile() to search forwards in
518        the pattern for the relevant set of parentheses. This search went wrong
519        when there were unescaped parentheses in a character class, parentheses
520        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
521    
522    33. "Subroutine" calls and backreferences were previously restricted to
523        referencing subpatterns earlier in the regex. This restriction has now
524        been removed.
525    
526    34. Added a number of extra features that are going to be in Perl 5.10. On the
527        whole, these are just syntactic alternatives for features that PCRE had
528        previously implemented using the Python syntax or my own invention. The
529        other formats are all retained for compatibility.
530    
531        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
532            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
533            also .NET compatible.
534    
535        (b) A recursion or subroutine call to a named group can now be defined as
536            (?&name) as well as (?P>name).
537    
538        (c) A backreference to a named group can now be defined as \k<name> or
539            \k'name' as well as (?P=name). The new forms, as well as being in Perl
540            5.10, are also .NET compatible.
541    
542        (d) A conditional reference to a named group can now use the syntax
543            (?(<name>) or (?('name') as well as (?(name).
544    
545        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
546            groups (named and numbered) that are never evaluated inline, but can be
547            called as "subroutines" from elsewhere. In effect, the DEFINE condition
548            is always false. There may be only one alternative in such a group.
549    
550        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
551            as the simple (?(R). The condition is true only if the most recent
552            recursion is that of the given number or name. It does not search out
553            through the entire recursion stack.
554    
555        (g) The escape \gN or \g{N} has been added, where N is a positive or
556            negative number, specifying an absolute or relative reference.
557    
558    35. Tidied to get rid of some further signed/unsigned compiler warnings and
559        some "unreachable code" warnings.
560    
561    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
562        things, this adds five new scripts.
563    
564    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
565        There were also incompatibilities regarding the handling of \Q..\E inside
566        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
567        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
568    
569    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
570        matches an empty string, and forcibly breaks the loop. There were bugs in
571        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
572        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
573        separate and independent bugs (that affected different cases) have been
574        fixed.
575    
576    39. Refactored the code to abolish the use of different opcodes for small
577        capturing bracket numbers. This is a tidy that I avoided doing when I
578        removed the limit on the number of capturing brackets for 3.5 back in 2001.
579        The new approach is not only tidier, it makes it possible to reduce the
580        memory needed to fix the previous bug (38).
581    
582    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
583        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
584        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
585        mode.
586    
587    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
588        report.
589    
590    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
591        copy construction and assignment in the C++ wrapper.
592    
593    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
594        couple of bugs that could have given wrong results in the "--newline=crlf"
595        case.
596    
597    44. Added a number of casts and did some reorganization of signed/unsigned int
598        variables following suggestions from Dair Grant. Also renamed the variable
599        "this" as "item" because it is a C++ keyword.
600    
601    45. Arranged for dftables to add
602    
603          #include "pcre_internal.h"
604    
605        to pcre_chartables.c because without it, gcc 4.x may remove the array
606        definition from the final binary if PCRE is built into a static library and
607        dead code stripping is activated.
608    
609    46. For an unanchored pattern, if a match attempt fails at the start of a
610        newline sequence, and the newline setting is CRLF or ANY, and the next two
611        characters are CRLF, advance by two characters instead of one.
612    
613    
614    Version 6.7 04-Jul-06
615    ---------------------
616    
617     1. In order to handle tests when input lines are enormously long, pcretest has
618        been re-factored so that it automatically extends its buffers when
619        necessary. The code is crude, but this _is_ just a test program. The
620        default size has been increased from 32K to 50K.
621    
622     2. The code in pcre_study() was using the value of the re argument before
623        testing it for NULL. (Of course, in any sensible call of the function, it
624        won't be NULL.)
625    
626     3. The memmove() emulation function in pcre_internal.h, which is used on
627        systems that lack both memmove() and bcopy() - that is, hardly ever -
628        was missing a "static" storage class specifier.
629    
630     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
631        containing an extended class (one that cannot be represented by a bitmap
632        because it contains high-valued characters or Unicode property items, e.g.
633        [\pZ]). Almost always one would set UTF-8 mode when processing such a
634        pattern, but PCRE should not loop if you do not (it no longer does).
635        [Detail: two cases were found: (a) a repeated subpattern containing an
636        extended class; (b) a recursive reference to a subpattern that followed a
637        previous extended class. It wasn't skipping over the extended class
638        correctly when UTF-8 mode was not set.]
639    
640     5. A negated single-character class was not being recognized as fixed-length
641        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
642        compile error "lookbehind assertion is not fixed length".
643    
644     6. The RunPerlTest auxiliary script was showing an unexpected difference
645        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
646        write a Perl script that can interpret lines of an input file either as
647        byte characters or as UTF-8, which is what "perltest" was being required to
648        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
649        can't do is switch easily at run time between having the "use utf8;" pragma
650        or not. In the end, I fudged it by using the RunPerlTest script to insert
651        "use utf8;" explicitly for the UTF-8 tests.
652    
653     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
654        the end of the subject string, contrary to the documentation and to what
655        Perl does. This was true of both matching functions. Now it matches only at
656        the start of the subject and immediately after *internal* newlines.
657    
658     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
659        a pointer to an int instead of a pointer to an unsigned long int. This
660        caused problems on 64-bit systems.
661    
662     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
663        instance of the 'standard' template library not being so standard".
664    
665    10. There was no check on the number of named subpatterns nor the maximum
666        length of a subpattern name. The product of these values is used to compute
667        the size of the memory block for a compiled pattern. By supplying a very
668        long subpattern name and a large number of named subpatterns, the size
669        computation could be caused to overflow. This is now prevented by limiting
670        the length of names to 32 characters, and the number of named subpatterns
671        to 10,000.
672    
673    11. Subpatterns that are repeated with specific counts have to be replicated in
674        the compiled pattern. The size of memory for this was computed from the
675        length of the subpattern and the repeat count. The latter is limited to
676        65535, but there was no limit on the former, meaning that integer overflow
677        could in principle occur. The compiled length of a repeated subpattern is
678        now limited to 30,000 bytes in order to prevent this.
679    
680    12. Added the optional facility to have named substrings with the same name.
681    
682    13. Added the ability to use a named substring as a condition, using the
683        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
684        are numbers (not recommended). Forward references are permitted.
685    
686    14. Added forward references in named backreferences (if you see what I mean).
687    
688    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
689        pattern could run off the end of the subject. For example, the pattern
690        "(?s)(.{1,5})"8 did this with the subject "ab".
691    
692    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
693        PCRE_CASELESS was set when matching characters that were quantified with ?
694        or *.
695    
696    17. A character class other than a single negated character that had a minimum
697        but no maximum quantifier - for example [ab]{6,} - was not handled
698        correctly by pce_dfa_exec(). It would match only one character.
699    
700    18. A valid (though odd) pattern that looked like a POSIX character
701        class but used an invalid character after [ (for example [[,abc,]]) caused
702        pcre_compile() to give the error "Failed: internal error: code overflow" or
703        in some cases to crash with a glibc free() error. This could even happen if
704        the pattern terminated after [[ but there just happened to be a sequence of
705        letters, a binary zero, and a closing ] in the memory that followed.
706    
707    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
708        over the years. Originally (before any Unicode support), just the bottom 8
709        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
710        output from "man perlunicode" includes this:
711    
712          The regular expression compiler produces polymorphic opcodes.  That
713          is, the pattern adapts to the data and automatically switches to
714          the Unicode character scheme when presented with Unicode data--or
715          instead uses a traditional byte scheme when presented with byte
716          data.
717    
718        Sadly, a wide octal escape does not cause a switch, and in a string with
719        no other multibyte characters, these octal escapes are treated as before.
720        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
721        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
722        Unicode string.
723    
724        I have not perpetrated such confusion in PCRE. Up till now, it took just
725        the bottom 8 bits, as in old Perl. I have now made octal escapes with
726        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
727        translate to the appropriate multibyte character.
728    
729    29. Applied some refactoring to reduce the number of warnings from Microsoft
730        and Borland compilers. This has included removing the fudge introduced
731        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
732        a warning about an unused variable.
733    
734    21. PCRE has not included VT (character 0x0b) in the set of whitespace
735        characters since release 4.0, because Perl (from release 5.004) does not.
736        [Or at least, is documented not to: some releases seem to be in conflict
737        with the documentation.] However, when a pattern was studied with
738        pcre_study() and all its branches started with \s, PCRE still included VT
739        as a possible starting character. Of course, this did no harm; it just
740        caused an unnecessary match attempt.
741    
742    22. Removed a now-redundant internal flag bit that recorded the fact that case
743        dependency changed within the pattern. This was once needed for "required
744        byte" processing, but is no longer used. This recovers a now-scarce options
745        bit. Also moved the least significant internal flag bit to the most-
746        significant bit of the word, which was not previously used (hangover from
747        the days when it was an int rather than a uint) to free up another bit for
748        the future.
749    
750    23. Added support for CRLF line endings as well as CR and LF. As well as the
751        default being selectable at build time, it can now be changed at runtime
752        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
753        specify that it is scanning data with non-default line endings.
754    
755    24. Changed the definition of CXXLINK to make it agree with the definition of
756        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
757    
758    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
759        recursions. This makes a big different to stack usage for some patterns.
760    
761    26. If a subpattern containing a named recursion or subroutine reference such
762        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
763        the space required for the compiled pattern went wrong and gave too small a
764        value. Depending on the environment, this could lead to "Failed: internal
765        error: code overflow at offset 49" or "glibc detected double free or
766        corruption" errors.
767    
768    27. Applied patches from Google (a) to support the new newline modes and (b) to
769        advance over multibyte UTF-8 characters in GlobalReplace.
770    
771    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
772        difference for some implementation of PCRE in some Windows version.
773    
774    29. Added some extra testing facilities to pcretest:
775    
776        \q<number>   in a data line sets the "match limit" value
777        \Q<number>   in a data line sets the "match recursion limt" value
778        -S <number>  sets the stack size, where <number> is in megabytes
779    
780        The -S option isn't available for Windows.
781    
782    
783    Version 6.6 06-Feb-06
784    ---------------------
785    
786     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
787        in pcreposix.h. I have copied the definition from pcre.h.
788    
789     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
790        because pcre.h is no longer a built file.
791    
792     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
793        not normally included in the compiled code.
794    
795    
796    Version 6.5 01-Feb-06
797    ---------------------
798    
799     1. When using the partial match feature with pcre_dfa_exec(), it was not
800        anchoring the second and subsequent partial matches at the new starting
801        point. This could lead to incorrect results. For example, with the pattern
802        /1234/, partially matching against "123" and then "a4" gave a match.
803    
804     2. Changes to pcregrep:
805    
806        (a) All non-match returns from pcre_exec() were being treated as failures
807            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
808            error message is output. Some extra information is given for the
809            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
810            probably the only errors that are likely to be caused by users (by
811            specifying a regex that has nested indefinite repeats, for instance).
812            If there are more than 20 of these errors, pcregrep is abandoned.
813    
814        (b) A binary zero was treated as data while matching, but terminated the
815            output line if it was written out. This has been fixed: binary zeroes
816            are now no different to any other data bytes.
817    
818        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
819            used to set a locale for matching. The --locale=xxxx long option has
820            been added (no short equivalent) to specify a locale explicitly on the
821            pcregrep command, overriding the environment variables.
822    
823        (d) When -B was used with -n, some line numbers in the output were one less
824            than they should have been.
825    
826        (e) Added the -o (--only-matching) option.
827    
828        (f) If -A or -C was used with -c (count only), some lines of context were
829            accidentally printed for the final match.
830    
831        (g) Added the -H (--with-filename) option.
832    
833        (h) The combination of options -rh failed to suppress file names for files
834            that were found from directory arguments.
835    
836        (i) Added the -D (--devices) and -d (--directories) options.
837    
838        (j) Added the -F (--fixed-strings) option.
839    
840        (k) Allow "-" to be used as a file name for -f as well as for a data file.
841    
842        (l) Added the --colo(u)r option.
843    
844        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
845            is not present by default.
846    
847     3. A nasty bug was discovered in the handling of recursive patterns, that is,
848        items such as (?R) or (?1), when the recursion could match a number of
849        alternatives. If it matched one of the alternatives, but subsequently,
850        outside the recursion, there was a failure, the code tried to back up into
851        the recursion. However, because of the way PCRE is implemented, this is not
852        possible, and the result was an incorrect result from the match.
853    
854        In order to prevent this happening, the specification of recursion has
855        been changed so that all such subpatterns are automatically treated as
856        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
857    
858     4. I had overlooked the fact that, in some locales, there are characters for
859        which isalpha() is true but neither isupper() nor islower() are true. In
860        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
861        and ordfeminine) are like this. This affected the treatment of \w and \W
862        when they appeared in character classes, but not when they appeared outside
863        a character class. The bit map for "word" characters is now created
864        separately from the results of isalnum() instead of just taking it from the
865        upper, lower, and digit maps. (Plus the underscore character, of course.)
866    
867     5. The above bug also affected the handling of POSIX character classes such as
868        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
869        permanent tables. Instead, the bit maps for such a class were previously
870        created as the appropriate unions of the upper, lower, and digit bitmaps.
871        Now they are created by subtraction from the [[:word:]] class, which has
872        its own bitmap.
873    
874     6. The [[:blank:]] character class matches horizontal, but not vertical space.
875        It is created by subtracting the vertical space characters (\x09, \x0a,
876        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
877        subtraction was done in the overall bitmap for a character class, meaning
878        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
879        be recognized. This bug has been fixed.
880    
881     7. Patches from the folks at Google:
882    
883          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
884          real life, but is still worth protecting against".
885    
886          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
887          regular expressions".
888    
889          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
890          have it.
891    
892          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
893          "configure" and the latter not, in order to fix a problem somebody had
894          with compiling the Arg class on HP-UX.
895    
896          (e) Improve the error-handling of the C++ wrapper a little bit.
897    
898          (f) New tests for checking recursion limiting.
899    
900     8. The pcre_memmove() function, which is used only if the environment does not
901        have a standard memmove() function (and is therefore rarely compiled),
902        contained two bugs: (a) use of int instead of size_t, and (b) it was not
903        returning a result (though PCRE never actually uses the result).
904    
905     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
906        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
907        returned instead of calling malloc() with an overflowing number that would
908        most likely cause subsequent chaos.
909    
910    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
911    
912    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
913        with this option is matched, the nmatch and pmatch options of regexec() are
914        ignored.
915    
916    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
917        provided in case anyone wants to the the POSIX interface with UTF-8
918        strings.
919    
920    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
921        C++ linking (needed for some HP-UX environments).
922    
923    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
924        (unused parameter) and in the pcre_printint() function (omitted "default"
925        switch label when the default is to do nothing).
926    
927    15. Added some code to make it possible, when PCRE is compiled as a C++
928        library, to replace subject pointers for pcre_exec() with a smart pointer
929        class, thus making it possible to process discontinuous strings.
930    
931    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
932        much the same function. They were added by different people who were trying
933        to make PCRE easy to compile on non-Unix systems. It has been suggested
934        that PCRE_EXPORT be abolished now that there is more automatic apparatus
935        for compiling on Windows systems. I have therefore replaced it with
936        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
937        defaults to "extern" for C or "extern C" for C++, which works fine on
938        Unix-like systems. It is now possible to override the value of PCRE_DATA_
939        SCOPE with something explicit in config.h. In addition:
940    
941        (a) pcreposix.h still had just "extern" instead of either of these macros;
942            I have replaced it with PCRE_DATA_SCOPE.
943    
944        (b) Functions such as _pcre_xclass(), which are internal to the library,
945            but external in the C sense, all had PCRE_EXPORT in their definitions.
946            This is apparently wrong for the Windows case, so I have removed it.
947            (It makes no difference on Unix-like systems.)
948    
949    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
950        of recursive calls to match(). This is different to MATCH_LIMIT because
951        that limits the total number of calls to match(), not all of which increase
952        the depth of recursion. Limiting the recursion depth limits the amount of
953        stack (or heap if NO_RECURSE is set) that is used. The default can be set
954        when PCRE is compiled, and changed at run time. A patch from Google adds
955        this functionality to the C++ interface.
956    
957    18. Changes to the handling of Unicode character properties:
958    
959        (a) Updated the table to Unicode 4.1.0.
960    
961        (b) Recognize characters that are not in the table as "Cn" (undefined).
962    
963        (c) I revised the way the table is implemented to a much improved format
964            which includes recognition of ranges. It now supports the ranges that
965            are defined in UnicodeData.txt, and it also amalgamates other
966            characters into ranges. This has reduced the number of entries in the
967            table from around 16,000 to around 3,000, thus reducing its size
968            considerably. I realized I did not need to use a tree structure after
969            all - a binary chop search is just as efficient. Having reduced the
970            number of entries, I extended their size from 6 bytes to 8 bytes to
971            allow for more data.
972    
973        (d) Added support for Unicode script names via properties such as \p{Han}.
974    
975    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
976        matching that character.
977    
978    20. When matching a repeated Unicode property with a minimum greater than zero,
979        (for example \pL{2,}), PCRE could look past the end of the subject if it
980        reached it while seeking the minimum number of characters. This could
981        happen only if some of the characters were more than one byte long, because
982        there is a check for at least the minimum number of bytes.
983    
984    21. Refactored the implementation of \p and \P so as to be more general, to
985        allow for more different types of property in future. This has changed the
986        compiled form incompatibly. Anybody with saved compiled patterns that use
987        \p or \P will have to recompile them.
988    
989    22. Added "Any" and "L&" to the supported property types.
990    
991    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
992        but give a compile time error if the value is greater than 0xff.
993    
994    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
995        accidentally not being installed or uninstalled.
996    
997    25. The pcre.h file was built from pcre.h.in, but the only changes that were
998        made were to insert the current release number. This seemed silly, because
999        it made things harder for people building PCRE on systems that don't run
1000        "configure". I have turned pcre.h into a distributed file, no longer built
1001        by "configure", with the version identification directly included. There is
1002        no longer a pcre.h.in file.
1003    
1004        However, this change necessitated a change to the pcre-config script as
1005        well. It is built from pcre-config.in, and one of the substitutions was the
1006        release number. I have updated configure.ac so that ./configure now finds
1007        the release number by grepping pcre.h.
1008    
1009    26. Added the ability to run the tests under valgrind.
1010    
1011    
1012    Version 6.4 05-Sep-05
1013    ---------------------
1014    
1015     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1016        "--" to be printed when multiple files were scanned, even when none of the
1017        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1018        consider it to be a bug, and have restored the previous behaviour.
1019    
1020     2. A couple of code tidies to get rid of compiler warnings.
1021    
1022     3. The pcretest program used to cheat by referring to symbols in the library
1023        whose names begin with _pcre_. These are internal symbols that are not
1024        really supposed to be visible externally, and in some environments it is
1025        possible to suppress them. The cheating is now confined to including
1026        certain files from the library's source, which is a bit cleaner.
1027    
1028     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1029        file's purpose clearer.
1030    
1031     5. Reorganized pcre_ucp_findchar().
1032    
1033    
1034    Version 6.3 15-Aug-05
1035    ---------------------
1036    
1037     1. The file libpcre.pc.in did not have general read permission in the tarball.
1038    
1039     2. There were some problems when building without C++ support:
1040    
1041        (a) If C++ support was not built, "make install" and "make test" still
1042            tried to test it.
1043    
1044        (b) There were problems when the value of CXX was explicitly set. Some
1045            changes have been made to try to fix these, and ...
1046    
1047        (c) --disable-cpp can now be used to explicitly disable C++ support.
1048    
1049        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1050            backslash in a target when C++ was disabled. This confuses some
1051            versions of "make", apparently. Using an intermediate variable solves
1052            this. (Same for CPP_LOBJ.)
1053    
1054     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1055        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1056        necessary on certain architectures.
1057    
1058     4. Added a setting of -export-symbols-regex to the link command to remove
1059        those symbols that are exported in the C sense, but actually are local
1060        within the library, and not documented. Their names all begin with
1061        "_pcre_". This is not a perfect job, because (a) we have to except some
1062        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1063        available (and never for static libraries). I have made a note to try to
1064        find a way round (a) in the future.
1065    
1066    
1067    Version 6.2 01-Aug-05
1068    ---------------------
1069    
1070     1. There was no test for integer overflow of quantifier values. A construction
1071        such as {1111111111111111} would give undefined results. What is worse, if
1072        a minimum quantifier for a parenthesized subpattern overflowed and became
1073        negative, the calculation of the memory size went wrong. This could have
1074        led to memory overwriting.
1075    
1076     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1077    
1078     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1079        operating environments where this matters.
1080    
1081     4. Applied Giuseppe Maxia's patch to add additional features for controlling
1082        PCRE options from within the C++ wrapper.
1083    
1084     5. Named capturing subpatterns were not being correctly counted when a pattern
1085        was compiled. This caused two problems: (a) If there were more than 100
1086        such subpatterns, the calculation of the memory needed for the whole
1087        compiled pattern went wrong, leading to an overflow error. (b) Numerical
1088        back references of the form \12, where the number was greater than 9, were
1089        not recognized as back references, even though there were sufficient
1090        previous subpatterns.
1091    
1092     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1093        versions of gcc, e.g. 2.95.4.
1094    
1095    
1096    Version 6.1 21-Jun-05
1097    ---------------------
1098    
1099     1. There was one reference to the variable "posix" in pcretest.c that was not
1100        surrounded by "#if !defined NOPOSIX".
1101    
1102     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1103        the cross-check on the old pcre_info() function, for the benefit of the
1104        cut-down version of PCRE that is currently imported into Exim.
1105    
1106     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1107        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1108        patterns that start (?i) but I don't think that matters. The use of (?i) is
1109        just an example; this all applies to the other options as well.
1110    
1111     4. Since libtool seems to echo the compile commands it is issuing, the output
1112        from "make" can be reduced a bit by putting "@" in front of each libtool
1113        compile command.
1114    
1115     5. Patch from the folks at Google for configure.in to be a bit more thorough
1116        in checking for a suitable C++ installation before trying to compile the
1117        C++ stuff. This should fix a reported problem when a compiler was present,
1118        but no suitable headers.
1119    
1120     6. The man pages all had just "PCRE" as their title. I have changed them to
1121        be the relevant file name. I have also arranged that these names are
1122        retained in the file doc/pcre.txt, which is a concatenation in text format
1123        of all the man pages except the little individual ones for each function.
1124    
1125     7. The NON-UNIX-USE file had not been updated for the different set of source
1126        files that come with release 6. I also added a few comments about the C++
1127        wrapper.
1128    
1129    
1130    Version 6.0 07-Jun-05
1131    ---------------------
1132    
1133     1. Some minor internal re-organization to help with my DFA experiments.
1134    
1135     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1136        didn't matter for the library itself when fully configured, but did matter
1137        when compiling without UCP support, or within Exim, where the ucp files are
1138        not imported.
1139    
1140     3. Refactoring of the library code to split up the various functions into
1141        different source modules. The addition of the new DFA matching code (see
1142        below) to a single monolithic source would have made it really too
1143        unwieldy, quite apart from causing all the code to be include in a
1144        statically linked application, when only some functions are used. This is
1145        relevant even without the DFA addition now that patterns can be compiled in
1146        one application and matched in another.
1147    
1148        The downside of splitting up is that there have to be some external
1149        functions and data tables that are used internally in different modules of
1150        the library but which are not part of the API. These have all had their
1151        names changed to start with "_pcre_" so that they are unlikely to clash
1152        with other external names.
1153    
1154     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1155        a different (DFA) algorithm. Although it is slower than the original
1156        function, it does have some advantages for certain types of matching
1157        problem.
1158    
1159     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1160        including restarting after a partial match.
1161    
1162     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1163        defined when compiling for Windows was sent to me. I have put it into the
1164        code, though I have no means of testing or verifying it.
1165    
1166     7. Added the pcre_refcount() auxiliary function.
1167    
1168     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1169        match before or at the first newline in the subject string. In pcretest,
1170        the /f option on a pattern can be used to set this.
1171    
1172     9. A repeated \w when used in UTF-8 mode with characters greater than 256
1173        would behave wrongly. This has been present in PCRE since release 4.0.
1174    
1175    10. A number of changes to the pcregrep command:
1176    
1177        (a) Refactored how -x works; insert ^(...)$ instead of setting
1178            PCRE_ANCHORED and checking the length, in preparation for adding
1179            something similar for -w.
1180    
1181        (b) Added the -w (match as a word) option.
1182    
1183        (c) Refactored the way lines are read and buffered so as to have more
1184            than one at a time available.
1185    
1186        (d) Implemented a pcregrep test script.
1187    
1188        (e) Added the -M (multiline match) option. This allows patterns to match
1189            over several lines of the subject. The buffering ensures that at least
1190            8K, or the rest of the document (whichever is the shorter) is available
1191            for matching (and similarly the previous 8K for lookbehind assertions).
1192    
1193        (f) Changed the --help output so that it now says
1194    
1195              -w, --word-regex(p)
1196    
1197            instead of two lines, one with "regex" and the other with "regexp"
1198            because that confused at least one person since the short forms are the
1199            same. (This required a bit of code, as the output is generated
1200            automatically from a table. It wasn't just a text change.)
1201    
1202        (g) -- can be used to terminate pcregrep options if the next thing isn't an
1203            option but starts with a hyphen. Could be a pattern or a path name
1204            starting with a hyphen, for instance.
1205    
1206        (h) "-" can be given as a file name to represent stdin.
1207    
1208        (i) When file names are being printed, "(standard input)" is used for
1209            the standard input, for compatibility with GNU grep. Previously
1210            "<stdin>" was used.
1211    
1212        (j) The option --label=xxx can be used to supply a name to be used for
1213            stdin when file names are being printed. There is no short form.
1214    
1215        (k) Re-factored the options decoding logic because we are going to add
1216            two more options that take data. Such options can now be given in four
1217            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1218    
1219        (l) Added the -A, -B, and -C options for requesting that lines of context
1220            around matches be printed.
1221    
1222        (m) Added the -L option to print the names of files that do not contain
1223            any matching lines, that is, the complement of -l.
1224    
1225        (n) The return code is 2 if any file cannot be opened, but pcregrep does
1226            continue to scan other files.
1227    
1228        (o) The -s option was incorrectly implemented. For compatibility with other
1229            greps, it now suppresses the error message for a non-existent or non-
1230            accessible file (but not the return code). There is a new option called
1231            -q that suppresses the output of matching lines, which was what -s was
1232            previously doing.
1233    
1234        (p) Added --include and --exclude options to specify files for inclusion
1235            and exclusion when recursing.
1236    
1237    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1238        Hopefully, it now does.
1239    
1240    12. Missing cast in pcre_study().
1241    
1242    13. Added an "uninstall" target to the makefile.
1243    
1244    14. Replaced "extern" in the function prototypes in Makefile.in with
1245        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1246        world, but is set differently for Windows.
1247    
1248    15. Added a second compiling function called pcre_compile2(). The only
1249        difference is that it has an extra argument, which is a pointer to an
1250        integer error code. When there is a compile-time failure, this is set
1251        non-zero, in addition to the error test pointer being set to point to an
1252        error message. The new argument may be NULL if no error number is required
1253        (but then you may as well call pcre_compile(), which is now just a
1254        wrapper). This facility is provided because some applications need a
1255        numeric error indication, but it has also enabled me to tidy up the way
1256        compile-time errors are handled in the POSIX wrapper.
1257    
1258    16. Added VPATH=.libs to the makefile; this should help when building with one
1259        prefix path and installing with another. (Or so I'm told by someone who
1260        knows more about this stuff than I do.)
1261    
1262    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1263        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1264        match everything, including newlines. This is not POSIX-compatible, but
1265        somebody wanted the feature. From pcretest it can be activated by using
1266        both the P and the s flags.
1267    
1268    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1269    
1270    19. libpcre.pc was being incorrectly installed as executable.
1271    
1272    20. A couple of places in pcretest check for end-of-line by looking for '\n';
1273        it now also looks for '\r' so that it will work unmodified on Windows.
1274    
1275    21. Added Google's contributed C++ wrapper to the distribution.
1276    
1277    22. Added some untidy missing memory free() calls in pcretest, to keep
1278        Electric Fence happy when testing.
1279    
1280    
1281    
1282    Version 5.0 13-Sep-04
1283    ---------------------
1284    
1285     1. Internal change: literal characters are no longer packed up into items
1286        containing multiple characters in a single byte-string. Each character
1287        is now matched using a separate opcode. However, there may be more than one
1288        byte in the character in UTF-8 mode.
1289    
1290     2. The pcre_callout_block structure has two new fields: pattern_position and
1291        next_item_length. These contain the offset in the pattern to the next match
1292        item, and its length, respectively.
1293    
1294     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1295        insertion of callouts before each pattern item. Added the /C option to
1296        pcretest to make use of this.
1297    
1298     4. On the advice of a Windows user, the lines
1299    
1300          #if defined(_WIN32) || defined(WIN32)
1301          _setmode( _fileno( stdout ), 0x8000 );
1302          #endif  /* defined(_WIN32) || defined(WIN32) */
1303    
1304        have been added to the source of pcretest. This apparently does useful
1305        magic in relation to line terminators.
1306    
1307     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1308        for the benefit of those environments where the "b" makes a difference.
1309    
1310     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1311        to know about it. I have put a hack into configure.in that adds in code
1312        to set GCC=yes if CC=icc. This seems to end up at a point in the
1313        generated configure script that is early enough to affect the setting of
1314        compiler options, which is what is needed, but I have no means of testing
1315        whether it really works. (The user who reported this had patched the
1316        generated configure script, which of course I cannot do.)
1317    
1318        LATER: After change 22 below (new libtool files), the configure script
1319        seems to know about icc (and also ecc). Therefore, I have commented out
1320        this hack in configure.in.
1321    
1322     7. Added support for pkg-config (2 patches were sent in).
1323    
1324     8. Negated POSIX character classes that used a combination of internal tables
1325        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1326        [[:^ascii]]. Typically, they would match almost any characters. The other
1327        POSIX classes were not broken in this way.
1328    
1329     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1330        to find the match, as PCRE was deluded into thinking that the match had to
1331        start at the start point or following a newline. The same bug applied to
1332        patterns with negative forward assertions or any backward assertions
1333        preceding ".*" at the start, unless the pattern required a fixed first
1334        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1335    
1336    10. In UTF-8 mode, when moving forwards in the subject after a failed match
1337        starting at the last subject character, bytes beyond the end of the subject
1338        string were read.
1339    
1340    11. Renamed the variable "class" as "classbits" to make life easier for C++
1341        users. (Previously there was a macro definition, but it apparently wasn't
1342        enough.)
1343    
1344    12. Added the new field "tables" to the extra data so that tables can be passed
1345        in at exec time, or the internal tables can be re-selected. This allows
1346        a compiled regex to be saved and re-used at a later time by a different
1347        program that might have everything at different addresses.
1348    
1349    13. Modified the pcre-config script so that, when run on Solaris, it shows a
1350        -R library as well as a -L library.
1351    
1352    14. The debugging options of pcretest (-d on the command line or D on a
1353        pattern) showed incorrect output for anything following an extended class
1354        that contained multibyte characters and which was followed by a quantifier.
1355    
1356    15. Added optional support for general category Unicode character properties
1357        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1358        support. It adds about 90K to the size of the library. The meanings of the
1359        inbuilt class escapes such as \d and \s have NOT been changed.
1360    
1361    16. Updated pcredemo.c to include calls to free() to release the memory for the
1362        compiled pattern.
1363    
1364    17. The generated file chartables.c was being created in the source directory
1365        instead of in the building directory. This caused the build to fail if the
1366        source directory was different from the building directory, and was
1367        read-only.
1368    
1369    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1370        file. No doubt somebody will tell me if they don't make sense... Also added
1371        Dan Mooney's comments about building on OpenVMS.
1372    
1373    19. Added support for partial matching via the PCRE_PARTIAL option for
1374        pcre_exec() and the \P data escape in pcretest.
1375    
1376    20. Extended pcretest with 3 new pattern features:
1377    
1378        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
1379              write the compiled pattern to the file whose name is "rest-of-line".
1380              This is a straight binary dump of the data, with the saved pointer to
1381              the character tables forced to be NULL. The study data, if any, is
1382              written too. After writing, pcretest reads a new pattern.
1383    
1384        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1385              compiled pattern from the given file. There must not be any
1386              occurrences of "<" in the file name (pretty unlikely); if there are,
1387              pcretest will instead treat the initial "<" as a pattern delimiter.
1388              After reading in the pattern, pcretest goes on to read data lines as
1389              usual.
1390    
1391        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1392              and 16-bit fields in a compiled pattern, to simulate a pattern that
1393              was compiled on a host of opposite endianness.
1394    
1395    21. The pcre-exec() function can now cope with patterns that were compiled on
1396        hosts of opposite endianness, with this restriction:
1397    
1398          As for any compiled expression that is saved and used later, the tables
1399          pointer field cannot be preserved; the extra_data field in the arguments
1400          to pcre_exec() should be used to pass in a tables address if a value
1401          other than the default internal tables were used at compile time.
1402    
1403    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1404        now diagnosed as an error. Previously, most of the time, a negative number
1405        would have been treated as zero, but if in addition "ovector" was passed as
1406        NULL, a crash could occur.
1407    
1408    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1409        new versions from the libtool 1.5 distribution (the last one is a copy of
1410        a file called libtool.m4). This seems to have fixed the need to patch
1411        "configure" to support Darwin 1.3 (which I used to do). However, I still
1412        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1413        workstation).
1414    
1415    24. Changed the PCRE licence to be the more standard "BSD" licence.
1416    
1417    
1418    Version 4.5 01-Dec-03
1419    ---------------------
1420    
1421     1. There has been some re-arrangement of the code for the match() function so
1422        that it can be compiled in a version that does not call itself recursively.
1423        Instead, it keeps those local variables that need separate instances for
1424        each "recursion" in a frame on the heap, and gets/frees frames whenever it
1425        needs to "recurse". Keeping track of where control must go is done by means
1426        of setjmp/longjmp. The whole thing is implemented by a set of macros that
1427        hide most of the details from the main code, and operates only if
1428        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1429        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1430        operating.
1431    
1432        To make it easier for callers to provide specially tailored get/free
1433        functions for this usage, two new functions, pcre_stack_malloc, and
1434        pcre_stack_free, are used. They are always called in strict stacking order,
1435        and the size of block requested is always the same.
1436    
1437        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1438        PCRE has been compiled to use the stack or the heap for recursion. The
1439        -C option of pcretest uses this to show which version is compiled.
1440    
1441        A new data escape \S, is added to pcretest; it causes the amounts of store
1442        obtained and freed by both kinds of malloc/free at match time to be added
1443        to the output.
1444    
1445     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1446        what's available on my current Linux desktop machine.
1447    
1448     3. When matching a UTF-8 string, the test for a valid string at the start has
1449        been extended. If start_offset is not zero, PCRE now checks that it points
1450        to a byte that is the start of a UTF-8 character. If not, it returns
1451        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1452        this is necessary because there may be backward assertions in the pattern.
1453        When matching the same subject several times, it may save resources to use
1454        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1455    
1456     4. The code for checking the validity of UTF-8 strings has been tightened so
1457        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1458        containing "overlong sequences".
1459    
1460     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1461        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1462        should have been "&", but it just so happened that all the cases this let
1463        through by mistake were picked up later in the function.
1464    
1465     6. I had used a variable called "isblank" - this is a C99 function, causing
1466        some compilers to warn. To avoid this, I renamed it (as "blankclass").
1467    
1468     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1469        prompting; (b) run "./pcretest /dev/null" at the start of the test script
1470        so the version is shown; (c) stop "make test" echoing "./RunTest".
1471    
1472     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1473    
1474     9. The prototype for memmove() for systems that don't have it was using
1475        size_t, but the inclusion of the header that defines size_t was later. I've
1476        moved the #includes for the C headers earlier to avoid this.
1477    
1478    10. Added some adjustments to the code to make it easier to compiler on certain
1479        special systems:
1480    
1481          (a) Some "const" qualifiers were missing.
1482          (b) Added the macro EXPORT before all exported functions; by default this
1483              is defined to be empty.
1484          (c) Changed the dftables auxiliary program (that builds chartables.c) so
1485              that it reads its output file name as an argument instead of writing
1486              to the standard output and assuming this can be redirected.
1487    
1488    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
1489        class containing characters with values greater than 255, PCRE compilation
1490        went into a loop.
1491    
1492    12. A recursive reference to a subpattern that was within another subpattern
1493        that had a minimum quantifier of zero caused PCRE to crash. For example,
1494        (x(y(?2))z)? provoked this bug with a subject that got as far as the
1495        recursion. If the recursively-called subpattern itself had a zero repeat,
1496        that was OK.
1497    
1498    13. In pcretest, the buffer for reading a data line was set at 30K, but the
1499        buffer into which it was copied (for escape processing) was still set at
1500        1024, so long lines caused crashes.
1501    
1502    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
1503        "internal error: code overflow...". This applied to any character class
1504        that was followed by a possessive quantifier.
1505    
1506    15. Modified the Makefile to add libpcre.la as a prerequisite for
1507        libpcreposix.la because I was told this is needed for a parallel build to
1508        work.
1509    
1510    16. If a pattern that contained .* following optional items at the start was
1511        studied, the wrong optimizing data was generated, leading to matching
1512        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
1513        matching string must start with a or b or c. The correct conclusion for
1514        this pattern is that a match can start with any character.
1515    
1516    
1517    Version 4.4 13-Aug-03
1518    ---------------------
1519    
1520     1. In UTF-8 mode, a character class containing characters with values between
1521        127 and 255 was not handled correctly if the compiled pattern was studied.
1522        In fixing this, I have also improved the studying algorithm for such
1523        classes (slightly).
1524    
1525     2. Three internal functions had redundant arguments passed to them. Removal
1526        might give a very teeny performance improvement.
1527    
1528     3. Documentation bug: the value of the capture_top field in a callout is *one
1529        more than* the number of the hightest numbered captured substring.
1530    
1531     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
1532        in incorrectly linking with a previously installed version. They now link
1533        explicitly with libpcre.la.
1534    
1535     5. configure.in no longer needs to recognize Cygwin specially.
1536    
1537     6. A problem in pcre.in for Windows platforms is fixed.
1538    
1539     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
1540        pcretest, it used to include the size of the study block as part of its
1541        output. Unfortunately, the structure contains a field that has a different
1542        size on different hardware architectures. This meant that the tests that
1543        showed this size failed. As the block is currently always of a fixed size,
1544        this information isn't actually particularly useful in pcretest output, so
1545        I have just removed it.
1546    
1547     8. Three pre-processor statements accidentally did not start in column 1.
1548        Sadly, there are *still* compilers around that complain, even though
1549        standard C has not required this for well over a decade. Sigh.
1550    
1551     9. In pcretest, the code for checking callouts passed small integers in the
1552        callout_data field, which is a void * field. However, some picky compilers
1553        complained about the casts involved for this on 64-bit systems. Now
1554        pcretest passes the address of the small integer instead, which should get
1555        rid of the warnings.
1556    
1557    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
1558        both compile and run time, and gives an error if an invalid UTF-8 sequence
1559        is found. There is a option for disabling this check in cases where the
1560        string is known to be correct and/or the maximum performance is wanted.
1561    
1562    11. In response to a bug report, I changed one line in Makefile.in from
1563    
1564            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
1565        to
1566            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
1567    
1568        to look similar to other lines, but I have no way of telling whether this
1569        is the right thing to do, as I do not use Windows. No doubt I'll get told
1570        if it's wrong...
1571    
1572    
1573    Version 4.3 21-May-03
1574    ---------------------
1575    
1576    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
1577       Makefile.
1578    
1579    2. Some refactoring to improve the quality of the code:
1580    
1581       (i)   The utf8_table... variables are now declared "const".
1582    
1583       (ii)  The code for \cx, which used the "case flipping" table to upper case
1584             lower case letters, now just substracts 32. This is ASCII-specific,
1585             but the whole concept of \cx is ASCII-specific, so it seems
1586             reasonable.
1587    
1588       (iii) PCRE was using its character types table to recognize decimal and
1589             hexadecimal digits in the pattern. This is silly, because it handles
1590             only 0-9, a-f, and A-F, but the character types table is locale-
1591             specific, which means strange things might happen. A private
1592             table is now used for this - though it costs 256 bytes, a table is
1593             much faster than multiple explicit tests. Of course, the standard
1594             character types table is still used for matching digits in subject
1595             strings against \d.
1596    
1597       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
1598             ending in _t are). So I've renamed it as ESC_tee.
1599    
1600    3. The first argument for regexec() in the POSIX wrapper should have been
1601       defined as "const".
1602    
1603    4. Changed pcretest to use malloc() for its buffers so that they can be
1604       Electric Fenced for debugging.
1605    
1606    5. There were several places in the code where, in UTF-8 mode, PCRE would try
1607       to read one or more bytes before the start of the subject string. Often this
1608       had no effect on PCRE's behaviour, but in some circumstances it could
1609       provoke a segmentation fault.
1610    
1611    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
1612       to try to read one or more bytes before the start of the subject string.
1613    
1614    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
1615       UTF-8 support could misbehave in various ways if the subject string
1616       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
1617       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
1618       back over UTF-8 characters.)
1619    
1620    
1621    Version 4.2 14-Apr-03
1622    ---------------------
1623    
1624    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
1625    
1626    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
1627         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
1628         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
1629         [WIN_PREFIX]: new variable, "cyg" for Cygwin
1630         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
1631           and BUILD_EXEEXT
1632         Note: automatic setting of the BUILD variables is not yet working
1633         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
1634           compile-time but not at link-time
1635         [LINK]: use for linking executables only
1636         make different versions for Windows and non-Windows
1637         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
1638           libraries
1639         [LINK_FOR_BUILD]: new variable
1640         [OBJEXT]: use throughout
1641         [EXEEXT]: use throughout
1642         <winshared>: new target
1643         <wininstall>: new target
1644         <dftables.o>: use native compiler
1645         <dftables>: use native linker
1646         <install>: handle Windows platform correctly
1647         <clean>: ditto
1648         <check>: ditto
1649         copy DLL to top builddir before testing
1650    
1651       As part of these changes, -no-undefined was removed again. This was reported
1652       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
1653       in any case.
1654    
1655    3. Some tidies to get rid of compiler warnings:
1656    
1657       . In the match_data structure, match_limit was an unsigned long int, whereas
1658         match_call_count was an int. I've made them both unsigned long ints.
1659    
1660       . In pcretest the fact that a const uschar * doesn't automatically cast to
1661         a void * provoked a warning.
1662    
1663       . Turning on some more compiler warnings threw up some "shadow" variables
1664         and a few more missing casts.
1665    
1666    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
1667       option, a class that contained a single character with a value between 128
1668       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
1669    
1670    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
1671       option, a class that contained several characters, but with at least one
1672       whose value was between 128 and 255 caused PCRE to crash.
1673    
1674    
1675    Version 4.1 12-Mar-03
1676    ---------------------
1677    
1678    1. Compiling with gcc -pedantic found a couple of places where casts were
1679    needed, and a string in dftables.c that was longer than standard compilers are
1680    required to support.
1681    
1682    2. Compiling with Sun's compiler found a few more places where the code could
1683    be tidied up in order to avoid warnings.
1684    
1685    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
1686    first of these names is deprecated in the latest Autoconf in favour of the name
1687    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
1688    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
1689    analogy I have changed it to CFLAGS_FOR_BUILD.
1690    
1691    4. Added -no-undefined to the linking command in the Makefile, because this is
1692    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
1693    linking step for the pcreposix library.
1694    
1695    5. PCRE was failing to diagnose the case of two named groups with the same
1696    name.
1697    
1698    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
1699    literal character that is needed in the subject for a match, and scans along to
1700    ensure that it is present before embarking on the full matching process. This
1701    saves time in cases of nested unlimited repeats that are never going to match.
1702    Problem: the scan can take a lot of time if the subject is very long (e.g.
1703    megabytes), thus penalizing straightforward matches. It is now done only if the
1704    amount of subject to be scanned is less than 1000 bytes.
1705    
1706    7. A lesser problem with the same optimization is that it was recording the
1707    first character of an anchored pattern as "needed", thus provoking a search
1708    right along the subject, even when the first match of the pattern was going to
1709    fail. The "needed" character is now not set for anchored patterns, unless it
1710    follows something in the pattern that is of non-fixed length. Thus, it still
1711    fulfils its original purpose of finding quick non-matches in cases of nested
1712    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
1713    
1714    
1715    Version 4.0 17-Feb-03
1716    ---------------------
1717    
1718    1. If a comment in an extended regex that started immediately after a meta-item
1719    extended to the end of string, PCRE compiled incorrect data. This could lead to
1720    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
1721    
1722    2. Moved to autoconf 2.53 and libtool 1.4.2.
1723    
1724    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
1725    the special perltest8 script is no longer needed - all the tests can be run
1726    from a single perltest script.
1727    
1728    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
1729    by \s. It has now been removed in PCRE. This means it isn't recognized as
1730    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
1731    class [:space:] *does* include VT, thereby creating a mess.
1732    
1733    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
1734    space and tab.
1735    
1736    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
1737    its new features into the main test script, reducing the number of scripts.
1738    
1739    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
1740    were backward compatible, and made the (?i) apply to the whole pattern, as if
1741    /i were given. Now it behaves more logically, and applies the option setting
1742    only to what follows. PCRE has been changed to follow suit. However, if it
1743    finds options settings right at the start of the pattern, it extracts them into
1744    the global options, as before. Thus, they show up in the info data.
1745    
1746    8. Added support for the \Q...\E escape sequence. Characters in between are
1747    treated as literals. This is slightly different from Perl in that $ and @ are
1748    also handled as literals inside the quotes. In Perl, they will cause variable
1749    interpolation. Note the following examples:
1750    
1751        Pattern            PCRE matches      Perl matches
1752    
1753        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
1754        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
1755        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
1756    
1757    For compatibility with Perl, \Q...\E sequences are recognized inside character
1758    classes as well as outside them.
1759    
1760    9. Re-organized 3 code statements in pcretest to avoid "overflow in
1761    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
1762    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
1763    signed/unsigned warnings.
1764    
1765    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
1766    option for pcretest, so I've replaced it by a simple function that does just
1767    that job.
1768    
1769    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
1770    "pcregrep -".
1771    
1772    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
1773    Java package. This provides some syntactic sugar for simple cases of what my
1774    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
1775    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
1776    item, you can use this simplified notation. Note that only makes sense with
1777    greedy quantifiers. Consequently, the use of the possessive quantifier forces
1778    greediness, whatever the setting of the PCRE_UNGREEDY option.
1779    
1780    13. A change of greediness default within a pattern was not taking effect at
1781    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
1782    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
1783    was abstracted outside.
1784    
1785    14. PCRE now supports the \G assertion. It is true when the current matching
1786    position is at the start point of the match. This differs from \A when the
1787    starting offset is non-zero. Used with the /g option of pcretest (or similar
1788    code), it works in the same way as it does for Perl's /g option. If all
1789    alternatives of a regex begin with \G, the expression is anchored to the start
1790    match position, and the "anchored" flag is set in the compiled expression.
1791    
1792    15. Some bugs concerning the handling of certain option changes within patterns
1793    have been fixed. These applied to options other than (?ims). For example,
1794    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
1795    been the other way round. Some of this was related to change 7 above.
1796    
1797    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
1798    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
1799    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
1800    POSIX classes only within a class (e.g. /[[:alpha:]]/).
1801    
1802    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
1803    mode. Unlike ".", it always matches newline, whatever the setting of
1804    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
1805    assertions. Perl allows it, but it doesn't (in general) work because it can't
1806    calculate the length of the lookbehind. At least, that's the case for Perl
1807    5.8.0 - I've been told they are going to document that it doesn't work in
1808    future.
1809    
1810    18. Added an error diagnosis for escapes that PCRE does not support: these are
1811    \L, \l, \N, \P, \p, \U, \u, and \X.
1812    
1813    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
1814    reading past the end of the pattern in cases such as /[abcd/.
1815    
1816    20. PCRE was getting more memory than necessary for patterns with classes that
1817    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
1818    
1819    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
1820    compiling PCRE for use with Virtual Pascal.
1821    
1822    22. Small fix to the Makefile to make it work properly if the build is done
1823    outside the source tree.
1824    
1825    23. Added a new extension: a condition to go with recursion. If a conditional
1826    subpattern starts with (?(R) the "true" branch is used if recursion has
1827    happened, whereas the "false" branch is used only at the top level.
1828    
1829    24. When there was a very long string of literal characters (over 255 bytes
1830    without UTF support, over 250 bytes with UTF support), the computation of how
1831    much memory was required could be incorrect, leading to segfaults or other
1832    strange effects.
1833    
1834    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
1835    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
1836    there was a subsequent back reference to those brackets. This meant that, for
1837    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
1838    possible to check for precisely this case. All we can do is abandon the
1839    optimization if .* occurs inside capturing brackets when there are any back
1840    references whatsoever. (See below for a better fix that came later.)
1841    
1842    26. The handling of the optimization for finding the first character of a
1843    non-anchored pattern, and for finding a character that is required later in the
1844    match were failing in some cases. This didn't break the matching; it just
1845    failed to optimize when it could. The way this is done has been re-implemented.
1846    
1847    27. Fixed typo in error message for invalid (?R item (it said "(?p").
1848    
1849    28. Added a new feature that provides some of the functionality that Perl
1850    provides with (?{...}). The facility is termed a "callout". The way it is done
1851    in PCRE is for the caller to provide an optional function, by setting
1852    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
1853    global variable. By default it is unset, which disables all calling out. To get
1854    the function called, the regex must include (?C) at appropriate points. This
1855    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
1856    This provides a means of identifying different callout points. When PCRE
1857    reaches such a point in the regex, if pcre_callout has been set, the external
1858    function is called. It is provided with data in a structure called
1859    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
1860    matching continues; if it returns a non-zero value, the match at the current
1861    point fails. However, backtracking will occur if possible. [This was changed
1862    later and other features added - see item 49 below.]
1863    
1864    29. pcretest is upgraded to test the callout functionality. It provides a
1865    callout function that displays information. By default, it shows the start of
1866    the match and the current position in the text. There are some new data escapes
1867    to vary what happens:
1868    
1869        \C+         in addition, show current contents of captured substrings
1870        \C-         do not supply a callout function
1871        \C!n        return 1 when callout number n is reached
1872        \C!n!m      return 1 when callout number n is reached for the mth time
1873    
1874    30. If pcregrep was called with the -l option and just a single file name, it
1875    output "<stdin>" if a match was found, instead of the file name.
1876    
1877    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
1878    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
1879    pcre_exec(). This saves a malloc/free per call. The default value of
1880    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
1881    when configuring.
1882    
1883    32. The default maximum size of a compiled pattern is 64K. There have been a
1884    few cases of people hitting this limit. The code now uses macros to handle the
1885    storing of links as offsets within the compiled pattern. It defaults to 2-byte
1886    links, but this can be changed to 3 or 4 bytes by --with-link-size when
1887    configuring. Tests 2 and 5 work only with 2-byte links because they output
1888    debugging information about compiled patterns.
1889    
1890    33. Internal code re-arrangements:
1891    
1892    (a) Moved the debugging function for printing out a compiled regex into
1893        its own source file (printint.c) and used #include to pull it into
1894        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
1895        separate copies.
1896    
1897    (b) Defined the list of op-code names for debugging as a macro in
1898        internal.h so that it is next to the definition of the opcodes.
1899    
1900    (c) Defined a table of op-code lengths for simpler skipping along compiled
1901        code. This is again a macro in internal.h so that it is next to the
1902        definition of the opcodes.
1903    
1904    34. Added support for recursive calls to individual subpatterns, along the
1905    lines of Robin Houston's patch (but implemented somewhat differently).
1906    
1907    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
1908    allow it to read and process whole directories in Win32. This code was
1909    contributed by Lionel Fourquaux; it has not been tested by me.
1910    
1911    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
1912    used to name a group. Names consist of alphanumerics and underscores, and must
1913    be unique. Back references use the syntax (?P=name) and recursive calls use
1914    (?P>name) which is a PCRE extension to the Python extension. Groups still have
1915    numbers. The function pcre_fullinfo() can be used after compilation to extract
1916    a name/number map. There are three relevant calls:
1917    
1918      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
1919      PCRE_INFO_NAMECOUNT            yields the number of entries
1920      PCRE_INFO_NAMETABLE            yields a pointer to the map.
1921    
1922    The map is a vector of fixed-size entries. The size of each entry depends on
1923    the length of the longest name used. The first two bytes of each entry are the
1924    group number, most significant byte first. There follows the corresponding
1925    name, zero terminated. The names are in alphabetical order.
1926    
1927    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
1928    case instead of 255. Making it the same both with and without UTF-8 support
1929    means that the same test output works with both.
1930    
1931    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
1932    calling malloc() with a zero argument.
1933    
1934    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
1935    optimization. I've improved things by keeping a bitmap of backreferences with
1936    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
1937    fact referenced, the optimization can be applied. It is unlikely that a
1938    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
1939    the match to follow \n) will appear inside brackets with a number greater than
1940    31, but if it does, any back reference > 31 suppresses the optimization.
1941    
1942    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
1943    of disabling numbered capturing parentheses. Any opening parenthesis that is
1944    not followed by ? behaves as if it were followed by ?: but named parentheses
1945    can still be used for capturing (and they will acquire numbers in the usual
1946    way).
1947    
1948    41. Redesigned the return codes from the match() function into yes/no/error so
1949    that errors can be passed back from deep inside the nested calls. A malloc
1950    failure while inside a recursive subpattern call now causes the
1951    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
1952    
1953    42. It is now possible to set a limit on the number of times the match()
1954    function is called in a call to pcre_exec(). This facility makes it possible to
1955    limit the amount of recursion and backtracking, though not in a directly
1956    obvious way, because the match() function is used in a number of different
1957    circumstances. The count starts from zero for each position in the subject
1958    string (for non-anchored patterns). The default limit is, for compatibility, a
1959    large number, namely 10 000 000. You can change this in two ways:
1960    
1961    (a) When configuring PCRE before making, you can use --with-match-limit=n
1962        to set a default value for the compiled library.
1963    
1964    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
1965        a different value is set. See 45 below.
1966    
1967    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
1968    
1969    43. Added a new function pcre_config(int, void *) to enable run-time extraction
1970    of things that can be changed at compile time. The first argument specifies
1971    what is wanted and the second points to where the information is to be placed.
1972    The current list of available information is:
1973    
1974      PCRE_CONFIG_UTF8
1975    
1976    The output is an integer that is set to one if UTF-8 support is available;
1977    otherwise it is set to zero.
1978    
1979      PCRE_CONFIG_NEWLINE
1980    
1981    The output is an integer that it set to the value of the code that is used for
1982    newline. It is either LF (10) or CR (13).
1983    
1984      PCRE_CONFIG_LINK_SIZE
1985    
1986    The output is an integer that contains the number of bytes used for internal
1987    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
1988    
1989      PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
1990    
1991    The output is an integer that contains the threshold above which the POSIX
1992    interface uses malloc() for output vectors. See item 31 above.
1993    
1994      PCRE_CONFIG_MATCH_LIMIT
1995    
1996    The output is an unsigned integer that contains the default limit of the number
1997    of match() calls in a pcre_exec() execution. See 42 above.
1998    
1999    44. pcretest has been upgraded by the addition of the -C option. This causes it
2000    to extract all the available output from the new pcre_config() function, and to
2001    output it. The program then exits immediately.
2002    
2003    45. A need has arisen to pass over additional data with calls to pcre_exec() in
2004    order to support additional features. One way would have been to define
2005    pcre_exec2() (for example) with extra arguments, but this would not have been
2006    extensible, and would also have required all calls to the original function to
2007    be mapped to the new one. Instead, I have chosen to extend the mechanism that
2008    is used for passing in "extra" data from pcre_study().
2009    
2010    The pcre_extra structure is now exposed and defined in pcre.h. It currently
2011    contains the following fields:
2012    
2013      flags         a bitmap indicating which of the following fields are set
2014      study_data    opaque data from pcre_study()
2015      match_limit   a way of specifying a limit on match() calls for a specific
2016                      call to pcre_exec()
2017      callout_data  data for callouts (see 49 below)
2018    
2019    The flag bits are also defined in pcre.h, and are
2020    
2021      PCRE_EXTRA_STUDY_DATA
2022      PCRE_EXTRA_MATCH_LIMIT
2023      PCRE_EXTRA_CALLOUT_DATA
2024    
2025    The pcre_study() function now returns one of these new pcre_extra blocks, with
2026    the actual study data pointed to by the study_data field, and the
2027    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
2028    before. That is, this change is entirely upwards-compatible and requires no
2029    change to existing code.
2030    
2031    If you want to pass in additional data to pcre_exec(), you can either place it
2032    in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
2033    block.
2034    
2035    46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
2036    data string contains the escape sequence \M, pcretest calls pcre_exec() several
2037    times with different match limits, until it finds the minimum value needed for
2038    pcre_exec() to complete. The value is then output. This can be instructive; for
2039    most simple matches the number is quite small, but for pathological cases it
2040    gets very large very quickly.
2041    
2042    47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
2043    returns the size of the data block pointed to by the study_data field in a
2044    pcre_extra block, that is, the value that was passed as the argument to
2045    pcre_malloc() when PCRE was getting memory in which to place the information
2046    created by pcre_study(). The fourth argument should point to a size_t variable.
2047    pcretest has been extended so that this information is shown after a successful
2048    pcre_study() call when information about the compiled regex is being displayed.
2049    
2050    48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
2051    because what follows is always an absolute path. (Later: it turns out that this
2052    is more than cosmetic for MinGW, because it doesn't like empty path
2053    components.)
2054    
2055    49. Some changes have been made to the callout feature (see 28 above):
2056    
2057    (i)  A callout function now has three choices for what it returns:
2058    
2059           0  =>  success, carry on matching
2060         > 0  =>  failure at this point, but backtrack if possible
2061         < 0  =>  serious error, return this value from pcre_exec()
2062    
2063         Negative values should normally be chosen from the set of PCRE_ERROR_xxx
2064         values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
2065         "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
2066         use by callout functions. It will never be used by PCRE itself.
2067    
2068    (ii) The pcre_extra structure (see 45 above) has a void * field called
2069         callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
2070         pcre_callout_block structure has a field of the same name. The contents of
2071         the field passed in the pcre_extra structure are passed to the callout
2072         function in the corresponding field in the callout block. This makes it
2073         easier to use the same callout-containing regex from multiple threads. For
2074         testing, the pcretest program has a new data escape
2075    
2076           \C*n        pass the number n (may be negative) as callout_data
2077    
2078         If the callout function in pcretest receives a non-zero value as
2079         callout_data, it returns that value.
2080    
2081    50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
2082    there were some redundant $(CFLAGS) in commands that are now specified as
2083    $(LINK), which already includes $(CFLAGS).
2084    
2085    51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
2086    has been compiled with UTF-8 support *and* pcre_compile() has been compiled
2087    with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
2088    one-byte characters throughout. Note that case-insensitive matching applies
2089    only to characters whose values are less than 256. PCRE doesn't support the
2090    notion of cases for higher-valued characters.
2091    
2092    (i)   A character class whose characters are all within 0-255 is handled as
2093          a bit map, and the map is inverted for negative classes. Previously, a
2094          character > 255 always failed to match such a class; however it should
2095          match if the class was a negative one (e.g. [^ab]). This has been fixed.
2096    
2097    (ii)  A negated character class with a single character < 255 is coded as
2098          "not this character" (OP_NOT). This wasn't working properly when the test
2099          character was multibyte, either singly or repeated.
2100    
2101    (iii) Repeats of multibyte characters are now handled correctly in UTF-8
2102          mode, for example: \x{100}{2,3}.
2103    
2104    (iv)  The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
2105          singly or repeated) now correctly test multibyte characters. However,
2106          PCRE doesn't recognize any characters with values greater than 255 as
2107          digits, spaces, or word characters. Such characters always match \D, \S,
2108          and \W, and never match \d, \s, or \w.
2109    
2110    (v)   Classes may now contain characters and character ranges with values
2111          greater than 255. For example: [ab\x{100}-\x{400}].
2112    
2113    (vi)  pcregrep now has a --utf-8 option (synonym -u) which makes it call
2114          PCRE in UTF-8 mode.
2115    
2116    52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
2117    PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
2118    retained for backwards compatibility. (Note that LASTLITERAL is also a byte
2119    value.)
2120    
2121    53. The single man page has become too large. I have therefore split it up into
2122    a number of separate man pages. These also give rise to individual HTML pages;
2123    these are now put in a separate directory, and there is an index.html page that
2124    lists them all. Some hyperlinking between the pages has been installed.
2125    
2126    54. Added convenience functions for handling named capturing parentheses.
2127    
2128    55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2129    aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2130    true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2131    are faulted.
2132    
2133    56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2134    calling configure. These values are used when compiling the dftables.c program
2135    which is run to generate the source of the default character tables. They
2136    default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2137    you will need to set these values.
2138    
2139    57. Updated the building process for Windows DLL, as provided by Fred Cox.
2140    
2141    
2142    Version 3.9 02-Jan-02
2143    ---------------------
2144    
2145    1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2146    
2147    2. If --disable-static was given, the building process failed when trying to
2148    build pcretest and pcregrep. (For some reason it was using libtool to compile
2149    them, which is not right, as they aren't part of the library.)
2150    
2151    
2152    Version 3.8 18-Dec-01
2153    ---------------------
2154    
2155    1. The experimental UTF-8 code was completely screwed up. It was packing the
2156    bytes in the wrong order. How dumb can you get?
2157    
2158    
2159    Version 3.7 29-Oct-01
2160    ---------------------
2161    
2162    1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2163    This caused pcretest, when used on the test data, to segfault. Unfortunately,
2164    this didn't happen under Solaris 8, where I normally test things.
2165    
2166    2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2167    doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2168    isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2169    this fix an hour or so after the initial 3.7 release.)
2170    
2171    
2172    Version 3.6 23-Oct-01
2173    ---------------------
2174    
2175    1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2176    offsets passed as NULL with zero offset count.
2177    
2178    2. The config.guess and config.sub files had not been updated when I moved to
2179    the latest autoconf.
2180    
2181    
2182    Version 3.5 15-Aug-01
2183    ---------------------
2184    
2185    1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2186    had been forgotten.
2187    
2188    2. By using declared but undefined structures, we can avoid using "void"
2189    definitions in pcre.h while keeping the internal definitions of the structures
2190    private.
2191    
2192    3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2193    user point of view, this means that both static and shared libraries are built
2194    by default, but this can be individually controlled. More of the work of
2195    handling this static/shared cases is now inside libtool instead of PCRE's make
2196    file.
2197    
2198    4. The pcretest utility is now installed along with pcregrep because it is
2199    useful for users (to test regexs) and by doing this, it automatically gets
2200    relinked by libtool. The documentation has been turned into a man page, so
2201    there are now .1, .txt, and .html versions in /doc.
2202    
2203    5. Upgrades to pcregrep:
2204       (i)   Added long-form option names like gnu grep.
2205       (ii)  Added --help to list all options with an explanatory phrase.
2206       (iii) Added -r, --recursive to recurse into sub-directories.
2207       (iv)  Added -f, --file to read patterns from a file.
2208    
2209    6. pcre_exec() was referring to its "code" argument before testing that
2210    argument for NULL (and giving an error if it was NULL).
2211    
2212    7. Upgraded Makefile.in to allow for compiling in a different directory from
2213    the source directory.
2214    
2215    8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2216    options bits, the pointer it was passed was to an int instead of to an unsigned
2217    long int. This mattered only on 64-bit systems.
2218    
2219    9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2220    generated) instead of pcre.in, which it its source. Also made the same change
2221    in several of the .c files.
2222    
2223    10. A new release of gcc defines printf() as a macro, which broke pcretest
2224    because it had an ifdef in the middle of a string argument for printf(). Fixed
2225    by using separate calls to printf().
2226    
2227    11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2228    script, to force use of CR or LF instead of \n in the source. On non-Unix
2229    systems, the value can be set in config.h.
2230    
2231    12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2232    absolute limit. Changed the text of the error message to make this clear, and
2233    likewise updated the man page.
2234    
2235    13. The limit of 99 on the number of capturing subpatterns has been removed.
2236    The new limit is 65535, which I hope will not be a "real" limit.
2237    
2238    
2239    Version 3.4 22-Aug-00
2240    ---------------------
2241    
2242    1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2243    
2244    2. Diagnose condition (?(0) as an error instead of crashing on matching.
2245    
2246    
2247    Version 3.3 01-Aug-00
2248    ---------------------
2249    
2250    1. If an octal character was given, but the value was greater than \377, it
2251    was not getting masked to the least significant bits, as documented. This could
2252    lead to crashes in some systems.
2253    
2254    2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
2255    the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
2256    
2257    3. Added the functions pcre_free_substring() and pcre_free_substring_list().
2258    These just pass their arguments on to (pcre_free)(), but they are provided
2259    because some uses of PCRE bind it to non-C systems that can call its functions,
2260    but cannot call free() or pcre_free() directly.
2261    
2262    4. Add "make test" as a synonym for "make check". Corrected some comments in
2263    the Makefile.
2264    
2265    5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
2266    Makefile.
2267    
2268    6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
2269    command called pgrep for grepping around the active processes.
2270    
2271    7. Added the beginnings of support for UTF-8 character strings.
2272    
2273    8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
2274    RANLIB to ./ltconfig so that they are used by libtool. I think these are all
2275    the relevant ones. (AR is not passed because ./ltconfig does its own figuring
2276    out for the ar command.)
2277    
2278    
2279    Version 3.2 12-May-00
2280    ---------------------
2281    
2282    This is purely a bug fixing release.
2283    
2284    1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
2285    of ZA. This was just one example of several cases that could provoke this bug,
2286    which was introduced by change 9 of version 2.00. The code for breaking
2287    infinite loops after an iteration that matches an empty string was't working
2288    correctly.
2289    
2290    2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
2291    when matched against abbab (for example). After matching an empty string, it
2292    wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
2293    caused it to match further down the string than it should.
2294    
2295    3. The code contained an inclusion of sys/types.h. It isn't clear why this
2296    was there because it doesn't seem to be needed, and it causes trouble on some
2297    systems, as it is not a Standard C header. It has been removed.
2298    
2299    4. Made 4 silly changes to the source to avoid stupid compiler warnings that
2300    were reported on the Macintosh. The changes were from
2301    
2302      while ((c = *(++ptr)) != 0 && c != '\n');
2303    to
2304      while ((c = *(++ptr)) != 0 && c != '\n') ;
2305    
2306    Totally extraordinary, but if that's what it takes...
2307    
2308    5. PCRE is being used in one environment where neither memmove() nor bcopy() is
2309    available. Added HAVE_BCOPY and an autoconf test for it; if neither
2310    HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
2311    assumes the way PCRE uses memmove() (always moving upwards).
2312    
2313    6. PCRE is being used in one environment where strchr() is not available. There
2314    was only one use in pcre.c, and writing it out to avoid strchr() probably gives
2315    faster code anyway.
2316    
2317    
2318    Version 3.1 09-Feb-00
2319    ---------------------
2320    
2321    The only change in this release is the fixing of some bugs in Makefile.in for
2322    the "install" target:
2323    
2324    (1) It was failing to install pcreposix.h.
2325    
2326    (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
2327    
2328    
2329    Version 3.0 01-Feb-00
2330    ---------------------
2331    
2332    1. Add support for the /+ modifier to perltest (to output $` like it does in
2333    pcretest).
2334    
2335    2. Add support for the /g modifier to perltest.
2336    
2337    3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
2338    matches null strings.
2339    
2340    4. Fix perltest so that it doesn't do unwanted things when fed an empty
2341    pattern. Perl treats empty patterns specially - it reuses the most recent
2342    pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
2343    effect.
2344    
2345    5. The POSIX interface was broken in that it was just handing over the POSIX
2346    captured string vector to pcre_exec(), but (since release 2.00) PCRE has
2347    required a bigger vector, with some working space on the end. This means that
2348    the POSIX wrapper now has to get and free some memory, and copy the results.
2349    
2350    6. Added some simple autoconf support, placing the test data and the
2351    documentation in separate directories, re-organizing some of the
2352    information files, and making it build pcre-config (a GNU standard). Also added
2353    libtool support for building PCRE as a shared library, which is now the
2354    default.
2355    
2356    7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
2357    09 are not valid octal constants. Single digits will be used for minor values
2358    less than 10.
2359    
2360    8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
2361    existing programs that set these in the POSIX interface can use PCRE without
2362    modification.
2363    
2364    9. Added a new function, pcre_fullinfo() with an extensible interface. It can
2365    return all that pcre_info() returns, plus additional data. The pcre_info()
2366    function is retained for compatibility, but is considered to be obsolete.
2367    
2368    10. Added experimental recursion feature (?R) to handle one common case that
2369    Perl 5.6 will be able to do with (?p{...}).
2370    
2371    11. Added support for POSIX character classes like [:alpha:], which Perl is
2372    adopting.
2373    
2374    
2375    Version 2.08 31-Aug-99
2376    ----------------------
2377    
2378    1. When startoffset was not zero and the pattern began with ".*", PCRE was not
2379    trying to match at the startoffset position, but instead was moving forward to
2380    the next newline as if a previous match had failed.
2381    
2382    2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
2383    and could get into a loop if a null string was matched other than at the start
2384    of the subject.
2385    
2386    3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
2387    be distinguished at compile time, and for completeness also added PCRE_DATE.
2388    
2389    5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
2390    in GnuWin32 environments.
2391    
2392    
2393    Version 2.07 29-Jul-99
2394    ----------------------
2395    
2396    1. The documentation is now supplied in plain text form and HTML as well as in
2397    the form of man page sources.
2398    
2399    2. C++ compilers don't like assigning (void *) values to other pointer types.
2400    In particular this affects malloc(). Although there is no problem in Standard
2401    C, I've put in casts to keep C++ compilers happy.
2402    
2403    3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
2404    should be (const char *).
2405    
2406    4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
2407    be useful for non-Unix systems who don't want to bother with the POSIX stuff.
2408    However, I haven't made this a standard facility. The documentation doesn't
2409    mention it, and the Makefile doesn't support it.
2410    
2411    5. The Makefile now contains an "install" target, with editable destinations at
2412    the top of the file. The pcretest program is not installed.
2413    
2414    6. pgrep -V now gives the PCRE version number and date.
2415    
2416    7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
2417    causing the entire string to be ignored, instead of just the last character.
2418    
2419    8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
2420    non-matching string, it can take a very, very long time, even for strings of
2421    quite modest length, because of the nested recursion. PCRE now does better in
2422    some of these cases. It does this by remembering the last required literal
2423    character in the pattern, and pre-searching the subject to ensure it is present
2424    before running the real match. In other words, it applies a heuristic to detect
2425    some types of certain failure quickly, and in the above example, if presented
2426    with a string that has no trailing " it gives "no match" very quickly.
2427    
2428    9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
2429    other alternatives are tried instead.
2430    
2431    
2432    Version 2.06 09-Jun-99
2433    ----------------------
2434    
2435    1. Change pcretest's output for amount of store used to show just the code
2436    space, because the remainder (the data block) varies in size between 32-bit and
2437    64-bit systems.
2438    
2439    2. Added an extra argument to pcre_exec() to supply an offset in the subject to
2440    start matching at. This allows lookbehinds to work when searching for multiple
2441    occurrences in a string.
2442    
2443    3. Added additional options to pcretest for testing multiple occurrences:
2444    
2445       /+   outputs the rest of the string that follows a match
2446       /g   loops for multiple occurrences, using the new startoffset argument
2447       /G   loops for multiple occurrences by passing an incremented pointer
2448    
2449    4. PCRE wasn't doing the "first character" optimization for patterns starting
2450    with \b or \B, though it was doing it for other lookbehind assertions. That is,
2451    it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
2452    the letter 'x'. On long subject strings, this gives a significant speed-up.
2453    
2454    
2455    Version 2.05 21-Apr-99
2456    ----------------------
2457    
2458    1. Changed the type of magic_number from int to long int so that it works
2459    properly on 16-bit systems.
2460    
2461    2. Fixed a bug which caused patterns starting with .* not to work correctly
2462    when the subject string contained newline characters. PCRE was assuming
2463    anchoring for such patterns in all cases, which is not correct because .* will
2464    not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
2465    DOTALL is set at top level; otherwise it knows that patterns starting with .*
2466    must be retried after every newline in the subject.
2467    
2468    
2469    Version 2.04 18-Feb-99
2470    ----------------------
2471    
2472    1. For parenthesized subpatterns with repeats whose minimum was zero, the
2473    computation of the store needed to hold the pattern was incorrect (too large).
2474    If such patterns were nested a few deep, this could multiply and become a real
2475    problem.
2476    
2477    2. Added /M option to pcretest to show the memory requirement of a specific
2478    pattern. Made -m a synonym of -s (which does this globally) for compatibility.
2479    
2480    3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
2481    compiled in such a way that the backtracking after subsequent failure was
2482    pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
2483    ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
2484    
2485    
2486    Version 2.03 02-Feb-99
2487    ----------------------
2488    
2489    1. Fixed typo and small mistake in man page.
2490    
2491    2. Added 4th condition (GPL supersedes if conflict) and created separate
2492    LICENCE file containing the conditions.
2493    
2494    3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
2495    Perl, that is the internal \ allows the delimiter to be included in the
2496    pattern. Locked out the use of \ as a delimiter. If \ immediately follows
2497    the final delimiter, add \ to the end of the pattern (to test the error).
2498    
2499    4. Added the convenience functions for extracting substrings after a successful
2500    match. Updated pcretest to make it able to test these functions.
2501    
2502    
2503    Version 2.02 14-Jan-99
2504    ----------------------
2505    
2506    1. Initialized the working variables associated with each extraction so that
2507    their saving and restoring doesn't refer to uninitialized store.
2508    
2509    2. Put dummy code into study.c in order to trick the optimizer of the IBM C
2510    compiler for OS/2 into generating correct code. Apparently IBM isn't going to
2511    fix the problem.
2512    
2513    3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
2514    calls, and wasn't printing the correct value for compiling calls. Increased the
2515    default value of LOOPREPEAT, and the number of significant figures in the
2516    times.
2517    
2518    4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
2519    
2520    5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
2521    a building problem on Windows NT with a FAT file system.
2522    
2523    
2524    Version 2.01 21-Oct-98
2525    ----------------------
2526    
2527    1. Changed the API for pcre_compile() to allow for the provision of a pointer
2528    to character tables built by pcre_maketables() in the current locale. If NULL
2529    is passed, the default tables are used.
2530    
2531    
2532    Version 2.00 24-Sep-98
2533    ----------------------
2534    
2535    1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
2536    it any more.
2537    
2538    2. Allow quantification of (?>) groups, and make it work correctly.
2539    
2540    3. The first character computation wasn't working for (?>) groups.
2541    
2542    4. Correct the implementation of \Z (it is permitted to match on the \n at the
2543    end of the subject) and add 5.005's \z, which really does match only at the
2544    very end of the subject.
2545    
2546    5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
2547    
2548    6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
2549    DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
2550    localized options. All options to pcre_study() were also removed.
2551    
2552    7. Add other new features from 5.005:
2553    
2554       $(?<=           positive lookbehind
2555       $(?<!           negative lookbehind
2556       (?imsx-imsx)    added the unsetting capability
2557                       such a setting is global if at outer level; local otherwise
2558       (?imsx-imsx:)   non-capturing groups with option setting
2559       (?(cond)re|re)  conditional pattern matching
2560    
2561       A backreference to itself in a repeated group matches the previous
2562       captured string.
2563    
2564    8. General tidying up of studying (both automatic and via "study")
2565    consequential on the addition of new assertions.
2566    
2567    9. As in 5.005, unlimited repeated groups that could match an empty substring
2568    are no longer faulted at compile time. Instead, the loop is forcibly broken at
2569    runtime if any iteration does actually match an empty substring.
2570    
2571    10. Include the RunTest script in the distribution.
2572    
2573    11. Added tests from the Perl 5.005_02 distribution. This showed up a few
2574    discrepancies, some of which were old and were also with respect to 5.004. They
2575    have now been fixed.
2576    
2577    
2578    Version 1.09 28-Apr-98
2579    ----------------------
2580    
2581    1. A negated single character class followed by a quantifier with a minimum
2582    value of one (e.g.  [^x]{1,6}  ) was not compiled correctly. This could lead to
2583    program crashes, or just wrong answers. This did not apply to negated classes
2584    containing more than one character, or to minima other than one.
2585    
2586    
2587    Version 1.08 27-Mar-98
2588    ----------------------
2589    
2590    1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
2591    
2592    2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
2593    latter must appear before anything that relies on it in the pattern.
2594    
2595    
2596    Version 1.07 16-Feb-98
2597    ----------------------
2598    
2599    1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
2600    repeat of a potentially empty string).
2601    
2602    
2603    Version 1.06 23-Jan-98
2604    ----------------------
2605    
2606    1. Added Markus Oberhumer's little patches for C++.
2607    
2608    2. Literal strings longer than 255 characters were broken.
2609    
2610    
2611    Version 1.05 23-Dec-97
2612    ----------------------
2613    
2614    1. Negated character classes containing more than one character were failing if
2615    PCRE_CASELESS was set at run time.
2616    
2617    
2618  Version 1.04 19-Dec-97  Version 1.04 19-Dec-97
2619  ----------------------  ----------------------

Legend:
Removed from v.11  
changed lines
  Added in v.216

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12