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

Diff of /code/trunk/ChangeLog

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

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

Legend:
Removed from v.33  
changed lines
  Added in v.211

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12