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

Diff of /code/trunk/ChangeLog

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

revision 23 by nigel, Sat Feb 24 21:38:41 2007 UTC revision 226 by ph10, Tue Aug 21 11:46:08 2007 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4    Version 7.3 20-Aug-07
5    ---------------------
6    
7     1. In the rejigging of the build system that eventually resulted in 7.1, the
8        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
9        brackets there is not right, since it causes compilers to look for an
10        installed pcre.h, not the version that is in the source that is being
11        compiled (which of course may be different). I have changed it back to:
12    
13          #include "pcre.h"
14    
15        I have a vague recollection that the change was concerned with compiling in
16        different directories, but in the new build system, that is taken care of
17        by the VPATH setting the Makefile.
18    
19     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
20        when the subject happened to end in the byte 0x85 (e.g. if the last
21        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
22        characters but of course it shouldn't be taken as a newline when it is part
23        of another character. The bug was that, for an unlimited repeat of . in
24        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
25        characters when looking for a newline.
26    
27     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
28    
29     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
30        in debug output.
31    
32     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
33        long printing in the pcrecpp unittest when running under MinGW.
34    
35     6. ESC_K was left out of the EBCDIC table.
36    
37     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
38        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
39        limit also applies to "virtual nesting" when a pattern is recursive, and in
40        this case 1000 isn't so big. I have been able to remove this limit at the
41        expense of backing off one optimization in certain circumstances. Normally,
42        when pcre_exec() would call its internal match() function recursively and
43        immediately return the result unconditionally, it uses a "tail recursion"
44        feature to save stack. However, when a subpattern that can match an empty
45        string has an unlimited repetition quantifier, it no longer makes this
46        optimization. That gives it a stack frame in which to save the data for
47        checking that an empty string has been matched. Previously this was taken
48        from the 1000-entry workspace that had been reserved. So now there is no
49        explicit limit, but more stack is used.
50    
51     8. Applied Daniel's patches to solve problems with the import/export magic
52        syntax that is required for Windows, and which was going wrong for the
53        pcreposix and pcrecpp parts of the library. These were overlooked when this
54        problem was solved for the main library.
55    
56     9. There were some crude static tests to avoid integer overflow when computing
57        the size of patterns that contain repeated groups with explicit upper
58        limits. As the maximum quantifier is 65535, the maximum group length was
59        set at 30,000 so that the product of these two numbers did not overflow a
60        32-bit integer. However, it turns out that people want to use groups that
61        are longer than 30,000 bytes (though not repeat them that many times).
62        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
63        made it possible to implement the integer overflow checks in a much more
64        dynamic way, which I have now done. The artificial limitation on group
65        length has been removed - we now have only the limit on the total length of
66        the compiled pattern, which depends on the LINK_SIZE setting.
67    
68    10. Fixed a bug in the documentation for get/copy named substring when
69        duplicate names are permitted. If none of the named substrings are set, the
70        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
71        empty string.
72    
73    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
74        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
75        because the ] is interpreted as the first data character and the
76        terminating ] is not found. PCRE has been made compatible with Perl in this
77        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
78        cause memory overwriting.
79    
80    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
81        string has been matched (to stop an infinite loop). It was not recognizing
82        a conditional subpattern that could match an empty string if that
83        subpattern was within another subpattern. For example, it looped when
84        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
85        condition was not nested. This bug has been fixed.
86    
87    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
88        past the start of the subject in the presence of bytes with the top bit
89        set, for example "\x8aBCD".
90    
91    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
92        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
93    
94    14. Optimized (?!) to (*FAIL).
95    
96    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
97        This restricts code points to be within the range 0 to 0x10FFFF, excluding
98        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
99        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
100        does: it's just the validity check that is more restrictive.
101    
102    16. Inserted checks for integer overflows during escape sequence (backslash)
103        processing, and also fixed erroneous offset values for syntax errors during
104        backslash processing.
105    
106    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
107        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
108    
109    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
110        caused an overrun.
111    
112    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
113        something other than just ASCII characters) inside a group that had an
114        unlimited repeat caused a loop at compile time (while checking to see
115        whether the group could match an empty string).
116    
117    20. Debugging a pattern containing \p or \P could cause a crash. For example,
118        [\P{Any}] did so. (Error in the code for printing property names.)
119    
120    21. An orphan \E inside a character class could cause a crash.
121    
122    22. A repeated capturing bracket such as (A)? could cause a wild memory
123        reference during compilation.
124    
125    23. There are several functions in pcre_compile() that scan along a compiled
126        expression for various reasons (e.g. to see if it's fixed length for look
127        behind). There were bugs in these functions when a repeated \p or \P was
128        present in the pattern. These operators have additional parameters compared
129        with \d, etc, and these were not being taken into account when moving along
130        the compiled data. Specifically:
131    
132        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
133            length.
134    
135        (b) An item such as \pL+ within a repeated group could cause crashes or
136            loops.
137    
138        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
139            "reference to non-existent subpattern" error.
140    
141        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
142    
143    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
144        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
145    
146    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
147    
148    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
149        character were causing crashes (broken optimization).
150    
151    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
152        \p or \P) caused a compile-time loop.
153    
154    28. More problems have arisen in unanchored patterns when CRLF is a valid line
155        break. For example, the unstudied pattern [\r\n]A does not match the string
156        "\r\nA" because change 7.0/46 below moves the current point on by two
157        characters after failing to match at the start. However, the pattern \nA
158        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
159        the same is true. There doesn't seem any very clean way out of this, but
160        what I have chosen to do makes the common cases work: PCRE now takes note
161        of whether there can be an explicit match for \r or \n anywhere in the
162        pattern, and if so, 7.0/46 no longer applies. As part of this change,
163        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
164        pattern has explicit CR or LF references.
165    
166    
167    Version 7.2 19-Jun-07
168    ---------------------
169    
170     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
171        which is apparently normally available under Windows.
172    
173     2. Re-jig the pcregrep tests with different newline settings in an attempt
174        to make them independent of the local environment's newline setting.
175    
176     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
177    
178     4. Some of the "internals" tests were previously cut out when the link size
179        was not 2, because the output contained actual offsets. The recent new
180        "Z" feature of pcretest means that these can be cut out, making the tests
181        usable with all link sizes.
182    
183     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
184        stack recursion. This gives a massive performance boost under BSD, but just
185        a small improvement under Linux. However, it saves one field in the frame
186        in all cases.
187    
188     6. Added more features from the forthcoming Perl 5.10:
189    
190        (a) (?-n) (where n is a string of digits) is a relative subroutine or
191            recursion call. It refers to the nth most recently opened parentheses.
192    
193        (b) (?+n) is also a relative subroutine call; it refers to the nth next
194            to be opened parentheses.
195    
196        (c) Conditions that refer to capturing parentheses can be specified
197            relatively, for example, (?(-2)... or (?(+3)...
198    
199        (d) \K resets the start of the current match so that everything before
200            is not part of it.
201    
202        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
203    
204        (f) \g{name} is another synonym - part of Perl 5.10's unification of
205            reference syntax.
206    
207        (g) (?| introduces a group in which the numbering of parentheses in each
208            alternative starts with the same number.
209    
210        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
211    
212     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
213        PCRE_INFO_JCHANGED.
214    
215     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
216        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
217        for detecting groups that can match an empty string.
218    
219     9. A pattern with a very large number of alternatives (more than several
220        hundred) was running out of internal workspace during the pre-compile
221        phase, where pcre_compile() figures out how much memory will be needed. A
222        bit of new cunning has reduced the workspace needed for groups with
223        alternatives. The 1000-alternative test pattern now uses 12 bytes of
224        workspace instead of running out of the 4096 that are available.
225    
226    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
227    
228    11. Applied patch from Google to remove an optimization that didn't quite work.
229        The report of the bug said:
230    
231          pcrecpp::RE("a*").FullMatch("aaa") matches, while
232          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
233          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
234    
235    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
236        it matched the wrong number of bytes.
237    
238    
239    Version 7.1 24-Apr-07
240    ---------------------
241    
242     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
243        that is more "standard", making use of automake and other Autotools. There
244        is some re-arrangement of the files and adjustment of comments consequent
245        on this.
246    
247     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
248        for recursive directory scanning broke on some systems because the files
249        are not scanned in any specific order and on different systems the order
250        was different. A call to "sort" has been inserted into RunGrepTest for the
251        approprate test as a short-term fix. In the longer term there may be an
252        alternative.
253    
254     3. I had an email from Eric Raymond about problems translating some of PCRE's
255        man pages to HTML (despite the fact that I distribute HTML pages, some
256        people do their own conversions for various reasons). The problems
257        concerned the use of low-level troff macros .br and .in. I have therefore
258        removed all such uses from the man pages (some were redundant, some could
259        be replaced by .nf/.fi pairs). The 132html script that I use to generate
260        HTML has been updated to handle .nf/.fi and to complain if it encounters
261        .br or .in.
262    
263     4. Updated comments in configure.ac that get placed in config.h.in and also
264        arranged for config.h to be included in the distribution, with the name
265        config.h.generic, for the benefit of those who have to compile without
266        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
267    
268     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
269        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
270        makevp.bat for the current PCRE, using the additional files
271        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
272    
273     6. A Windows user reported a minor discrepancy with test 2, which turned out
274        to be caused by a trailing space on an input line that had got lost in his
275        copy. The trailing space was an accident, so I've just removed it.
276    
277     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
278        that is needed.
279    
280     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
281        as "const" (a) because they are and (b) because it helps the PHP
282        maintainers who have recently made a script to detect big data structures
283        in the php code that should be moved to the .rodata section. I remembered
284        to update Builducptable as well, so it won't revert if ucptable.h is ever
285        re-created.
286    
287     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
288        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
289        order to be able to cut out the UTF-8 tables in the latter when UTF-8
290        support is not required. This saves 1.5-2K of code, which is important in
291        some applications.
292    
293        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
294        so as not to refer to the tables, even though these functions will never be
295        called when UTF-8 support is disabled. Otherwise there are problems with a
296        shared library.
297    
298    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
299    
300        (a) It was defining its arguments as char * instead of void *.
301    
302        (b) It was assuming that all moves were upwards in memory; this was true
303            a long time ago when I wrote it, but is no longer the case.
304    
305        The emulated memove() is provided for those environments that have neither
306        memmove() nor bcopy(). I didn't think anyone used it these days, but that
307        is clearly not the case, as these two bugs were recently reported.
308    
309    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
310        and Detrail to create the HTML documentation, the .txt form of the man
311        pages, and it removes trailing spaces from listed files. It also creates
312        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
313        case, it wraps all the #defines with #ifndefs. This script should be run
314        before "make dist".
315    
316    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
317        with Unicode property support.
318    
319        (a) For a maximizing quantifier, if the two different cases of the
320            character were of different lengths in their UTF-8 codings (there are
321            some cases like this - I found 11), and the matching function had to
322            back up over a mixture of the two cases, it incorrectly assumed they
323            were both the same length.
324    
325        (b) When PCRE was configured to use the heap rather than the stack for
326            recursion during matching, it was not correctly preserving the data for
327            the other case of a UTF-8 character when checking ahead for a match
328            while processing a minimizing repeat. If the check also involved
329            matching a wide character, but failed, corruption could cause an
330            erroneous result when trying to check for a repeat of the original
331            character.
332    
333    13. Some tidying changes to the testing mechanism:
334    
335        (a) The RunTest script now detects the internal link size and whether there
336            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
337            values substituted by "configure". (The RunGrepTest script already did
338            this for UTF-8.) The configure.ac script no longer substitutes the
339            relevant variables.
340    
341        (b) The debugging options /B and /D in pcretest show the compiled bytecode
342            with length and offset values. This means that the output is different
343            for different internal link sizes. Test 2 is skipped for link sizes
344            other than 2 because of this, bypassing the problem. Unfortunately,
345            there was also a test in test 3 (the locale tests) that used /B and
346            failed for link sizes other than 2. Rather than cut the whole test out,
347            I have added a new /Z option to pcretest that replaces the length and
348            offset values with spaces. This is now used to make test 3 independent
349            of link size. (Test 2 will be tidied up later.)
350    
351    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
352        segmentation fault instead of returning the appropriate error message.
353    
354    15. In multiline mode when the newline sequence was set to "any", the pattern
355        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
356        This doesn't seem right; it now treats the CRLF combination as the line
357        ending, and so does not match in that case. It's only a pattern such as ^$
358        that would hit this one: something like ^ABC$ would have failed after \r
359        and then tried again after \r\n.
360    
361    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
362        in an attempt to make files that differ only in their line terminators
363        compare equal. This works on Linux.
364    
365    17. Under certain error circumstances pcregrep might try to free random memory
366        as it exited. This is now fixed, thanks to valgrind.
367    
368    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
369        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
370        was because its rules for how to advance for /g after matching an empty
371        string at the end of a line did not allow for this case. They now check for
372        it specially.
373    
374    20. pcretest is supposed to handle patterns and data of any length, by
375        extending its buffers when necessary. It was getting this wrong when the
376        buffer for a data line had to be extended.
377    
378    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
379        CRLF as a newline sequence.
380    
381    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
382        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
383        I have nevertheless tidied it up.
384    
385    23. Added some casts to kill warnings from HP-UX ia64 compiler.
386    
387    24. Added a man page for pcre-config.
388    
389    
390    Version 7.0 19-Dec-06
391    ---------------------
392    
393     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
394        moving to gcc 4.1.1.
395    
396     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
397        sys/time.h, which is documented as needed for this function. It doesn't
398        seem to matter on Linux, but it showed up on some releases of OS X.
399    
400     3. It seems that there are systems where bytes whose values are greater than
401        127 match isprint() in the "C" locale. The "C" locale should be the
402        default when a C program starts up. In most systems, only ASCII printing
403        characters match isprint(). This difference caused the output from pcretest
404        to vary, making some of the tests fail. I have changed pcretest so that:
405    
406        (a) When it is outputting text in the compiled version of a pattern, bytes
407            other than 32-126 are always shown as hex escapes.
408    
409        (b) When it is outputting text that is a matched part of a subject string,
410            it does the same, unless a different locale has been set for the match
411            (using the /L modifier). In this case, it uses isprint() to decide.
412    
413     4. Fixed a major bug that caused incorrect computation of the amount of memory
414        required for a compiled pattern when options that changed within the
415        pattern affected the logic of the preliminary scan that determines the
416        length. The relevant options are -x, and -i in UTF-8 mode. The result was
417        that the computed length was too small. The symptoms of this bug were
418        either the PCRE error "internal error: code overflow" from pcre_compile(),
419        or a glibc crash with a message such as "pcretest: free(): invalid next
420        size (fast)". Examples of patterns that provoked this bug (shown in
421        pcretest format) are:
422    
423          /(?-x: )/x
424          /(?x)(?-x: \s*#\s*)/
425          /((?i)[\x{c0}])/8
426          /(?i:[\x{c0}])/8
427    
428        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
429        is now done differently.
430    
431     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
432        wrapper classes; (b) implement a new function in the C++ scanner that is
433        more efficient than the old way of doing things because it avoids levels of
434        recursion in the regex matching; (c) add a paragraph to the documentation
435        for the FullMatch() function.
436    
437     6. The escape sequence \n was being treated as whatever was defined as
438        "newline". Not only was this contrary to the documentation, which states
439        that \n is character 10 (hex 0A), but it also went horribly wrong when
440        "newline" was defined as CRLF. This has been fixed.
441    
442     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
443        was being set to -1 for the "end of line" case (supposedly a value that no
444        character can have). Though this value is never used (the check for end of
445        line is "zero bytes in current character"), it caused compiler complaints.
446        I've changed it to 0xffffffff.
447    
448     8. In pcre_version.c, the version string was being built by a sequence of
449        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
450        string (as it is for production releases) called a macro with an empty
451        argument. The C standard says the result of this is undefined. The gcc
452        compiler treats it as an empty string (which was what was wanted) but it is
453        reported that Visual C gives an error. The source has been hacked around to
454        avoid this problem.
455    
456     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
457        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
458        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
459        of them did).
460    
461    10. Originally, pcretest opened its input and output without "b"; then I was
462        told that "b" was needed in some environments, so it was added for release
463        5.0 to both the input and output. (It makes no difference on Unix-like
464        systems.) Later I was told that it is wrong for the input on Windows. I've
465        now abstracted the modes into two macros, to make it easier to fiddle with
466        them, and removed "b" from the input mode under Windows.
467    
468    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
469    
470    12. Added -help and --help to pcretest as an official way of being reminded
471        of the options.
472    
473    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
474        and pcrecpp.cc because they annoy compilers at high warning levels.
475    
476    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
477    
478    15. Fixed an occurrence of == in configure.ac that should have been = (shell
479        scripts are not C programs :-) and which was not noticed because it works
480        on Linux.
481    
482    16. pcretest is supposed to handle any length of pattern and data line (as one
483        line or as a continued sequence of lines) by extending its input buffer if
484        necessary. This feature was broken for very long pattern lines, leading to
485        a string of junk being passed to pcre_compile() if the pattern was longer
486        than about 50K.
487    
488    17. I have done a major re-factoring of the way pcre_compile() computes the
489        amount of memory needed for a compiled pattern. Previously, there was code
490        that made a preliminary scan of the pattern in order to do this. That was
491        OK when PCRE was new, but as the facilities have expanded, it has become
492        harder and harder to keep it in step with the real compile phase, and there
493        have been a number of bugs (see for example, 4 above). I have now found a
494        cunning way of running the real compile function in a "fake" mode that
495        enables it to compute how much memory it would need, while actually only
496        ever using a few hundred bytes of working memory and without too many
497        tests of the mode. This should make future maintenance and development
498        easier. A side effect of this work is that the limit of 200 on the nesting
499        depth of parentheses has been removed (though this was never a serious
500        limitation, I suspect). However, there is a downside: pcre_compile() now
501        runs more slowly than before (30% or more, depending on the pattern). I
502        hope this isn't a big issue. There is no effect on runtime performance.
503    
504    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
505        newline (only possible for the last line of a file) and it was a
506        pattern that set a locale (followed by /Lsomething), pcretest crashed.
507    
508    19. Added additional timing features to pcretest. (1) The -tm option now times
509        matching only, not compiling. (2) Both -t and -tm can be followed, as a
510        separate command line item, by a number that specifies the number of
511        repeats to use when timing. The default is 50000; this gives better
512        precision, but takes uncomfortably long for very large patterns.
513    
514    20. Extended pcre_study() to be more clever in cases where a branch of a
515        subpattern has no definite first character. For example, (a*|b*)[cd] would
516        previously give no result from pcre_study(). Now it recognizes that the
517        first character must be a, b, c, or d.
518    
519    21. There was an incorrect error "recursive call could loop indefinitely" if
520        a subpattern (or the entire pattern) that was being tested for matching an
521        empty string contained only one non-empty item after a nested subpattern.
522        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
523        incorrectly, because the \d was being skipped in the check.
524    
525    22. The pcretest program now has a new pattern option /B and a command line
526        option -b, which is equivalent to adding /B to every pattern. This causes
527        it to show the compiled bytecode, without the additional information that
528        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
529        is the same as /B/I).
530    
531    23. A new optimization is now able automatically to treat some sequences such
532        as a*b as a*+b. More specifically, if something simple (such as a character
533        or a simple class like \d) has an unlimited quantifier, and is followed by
534        something that cannot possibly match the quantified thing, the quantifier
535        is automatically "possessified".
536    
537    24. A recursive reference to a subpattern whose number was greater than 39
538        went wrong under certain circumstances in UTF-8 mode. This bug could also
539        have affected the operation of pcre_study().
540    
541    25. Realized that a little bit of performance could be had by replacing
542        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
543    
544    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
545    
546    27. Possessive quantifiers such as a++ were previously implemented by turning
547        them into atomic groups such as ($>a+). Now they have their own opcodes,
548        which improves performance. This includes the automatically created ones
549        from 23 above.
550    
551    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
552        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
553        the first matched character to be a colon. This applied both to named and
554        numbered groups.
555    
556    29. The ucpinternal.h header file was missing its idempotency #ifdef.
557    
558    30. I was sent a "project" file called libpcre.a.dev which I understand makes
559        building PCRE on Windows easier, so I have included it in the distribution.
560    
561    31. There is now a check in pcretest against a ridiculously large number being
562        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
563        loop, the loop is abandoned.
564    
565    32. Forward references to subpatterns in conditions such as (?(2)...) where
566        subpattern 2 is defined later cause pcre_compile() to search forwards in
567        the pattern for the relevant set of parentheses. This search went wrong
568        when there were unescaped parentheses in a character class, parentheses
569        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
570    
571    33. "Subroutine" calls and backreferences were previously restricted to
572        referencing subpatterns earlier in the regex. This restriction has now
573        been removed.
574    
575    34. Added a number of extra features that are going to be in Perl 5.10. On the
576        whole, these are just syntactic alternatives for features that PCRE had
577        previously implemented using the Python syntax or my own invention. The
578        other formats are all retained for compatibility.
579    
580        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
581            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
582            also .NET compatible.
583    
584        (b) A recursion or subroutine call to a named group can now be defined as
585            (?&name) as well as (?P>name).
586    
587        (c) A backreference to a named group can now be defined as \k<name> or
588            \k'name' as well as (?P=name). The new forms, as well as being in Perl
589            5.10, are also .NET compatible.
590    
591        (d) A conditional reference to a named group can now use the syntax
592            (?(<name>) or (?('name') as well as (?(name).
593    
594        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
595            groups (named and numbered) that are never evaluated inline, but can be
596            called as "subroutines" from elsewhere. In effect, the DEFINE condition
597            is always false. There may be only one alternative in such a group.
598    
599        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
600            as the simple (?(R). The condition is true only if the most recent
601            recursion is that of the given number or name. It does not search out
602            through the entire recursion stack.
603    
604        (g) The escape \gN or \g{N} has been added, where N is a positive or
605            negative number, specifying an absolute or relative reference.
606    
607    35. Tidied to get rid of some further signed/unsigned compiler warnings and
608        some "unreachable code" warnings.
609    
610    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
611        things, this adds five new scripts.
612    
613    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
614        There were also incompatibilities regarding the handling of \Q..\E inside
615        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
616        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
617    
618    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
619        matches an empty string, and forcibly breaks the loop. There were bugs in
620        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
621        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
622        separate and independent bugs (that affected different cases) have been
623        fixed.
624    
625    39. Refactored the code to abolish the use of different opcodes for small
626        capturing bracket numbers. This is a tidy that I avoided doing when I
627        removed the limit on the number of capturing brackets for 3.5 back in 2001.
628        The new approach is not only tidier, it makes it possible to reduce the
629        memory needed to fix the previous bug (38).
630    
631    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
632        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
633        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
634        mode.
635    
636    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
637        report.
638    
639    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
640        copy construction and assignment in the C++ wrapper.
641    
642    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
643        couple of bugs that could have given wrong results in the "--newline=crlf"
644        case.
645    
646    44. Added a number of casts and did some reorganization of signed/unsigned int
647        variables following suggestions from Dair Grant. Also renamed the variable
648        "this" as "item" because it is a C++ keyword.
649    
650    45. Arranged for dftables to add
651    
652          #include "pcre_internal.h"
653    
654        to pcre_chartables.c because without it, gcc 4.x may remove the array
655        definition from the final binary if PCRE is built into a static library and
656        dead code stripping is activated.
657    
658    46. For an unanchored pattern, if a match attempt fails at the start of a
659        newline sequence, and the newline setting is CRLF or ANY, and the next two
660        characters are CRLF, advance by two characters instead of one.
661    
662    
663    Version 6.7 04-Jul-06
664    ---------------------
665    
666     1. In order to handle tests when input lines are enormously long, pcretest has
667        been re-factored so that it automatically extends its buffers when
668        necessary. The code is crude, but this _is_ just a test program. The
669        default size has been increased from 32K to 50K.
670    
671     2. The code in pcre_study() was using the value of the re argument before
672        testing it for NULL. (Of course, in any sensible call of the function, it
673        won't be NULL.)
674    
675     3. The memmove() emulation function in pcre_internal.h, which is used on
676        systems that lack both memmove() and bcopy() - that is, hardly ever -
677        was missing a "static" storage class specifier.
678    
679     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
680        containing an extended class (one that cannot be represented by a bitmap
681        because it contains high-valued characters or Unicode property items, e.g.
682        [\pZ]). Almost always one would set UTF-8 mode when processing such a
683        pattern, but PCRE should not loop if you do not (it no longer does).
684        [Detail: two cases were found: (a) a repeated subpattern containing an
685        extended class; (b) a recursive reference to a subpattern that followed a
686        previous extended class. It wasn't skipping over the extended class
687        correctly when UTF-8 mode was not set.]
688    
689     5. A negated single-character class was not being recognized as fixed-length
690        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
691        compile error "lookbehind assertion is not fixed length".
692    
693     6. The RunPerlTest auxiliary script was showing an unexpected difference
694        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
695        write a Perl script that can interpret lines of an input file either as
696        byte characters or as UTF-8, which is what "perltest" was being required to
697        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
698        can't do is switch easily at run time between having the "use utf8;" pragma
699        or not. In the end, I fudged it by using the RunPerlTest script to insert
700        "use utf8;" explicitly for the UTF-8 tests.
701    
702     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
703        the end of the subject string, contrary to the documentation and to what
704        Perl does. This was true of both matching functions. Now it matches only at
705        the start of the subject and immediately after *internal* newlines.
706    
707     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
708        a pointer to an int instead of a pointer to an unsigned long int. This
709        caused problems on 64-bit systems.
710    
711     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
712        instance of the 'standard' template library not being so standard".
713    
714    10. There was no check on the number of named subpatterns nor the maximum
715        length of a subpattern name. The product of these values is used to compute
716        the size of the memory block for a compiled pattern. By supplying a very
717        long subpattern name and a large number of named subpatterns, the size
718        computation could be caused to overflow. This is now prevented by limiting
719        the length of names to 32 characters, and the number of named subpatterns
720        to 10,000.
721    
722    11. Subpatterns that are repeated with specific counts have to be replicated in
723        the compiled pattern. The size of memory for this was computed from the
724        length of the subpattern and the repeat count. The latter is limited to
725        65535, but there was no limit on the former, meaning that integer overflow
726        could in principle occur. The compiled length of a repeated subpattern is
727        now limited to 30,000 bytes in order to prevent this.
728    
729    12. Added the optional facility to have named substrings with the same name.
730    
731    13. Added the ability to use a named substring as a condition, using the
732        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
733        are numbers (not recommended). Forward references are permitted.
734    
735    14. Added forward references in named backreferences (if you see what I mean).
736    
737    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
738        pattern could run off the end of the subject. For example, the pattern
739        "(?s)(.{1,5})"8 did this with the subject "ab".
740    
741    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
742        PCRE_CASELESS was set when matching characters that were quantified with ?
743        or *.
744    
745    17. A character class other than a single negated character that had a minimum
746        but no maximum quantifier - for example [ab]{6,} - was not handled
747        correctly by pce_dfa_exec(). It would match only one character.
748    
749    18. A valid (though odd) pattern that looked like a POSIX character
750        class but used an invalid character after [ (for example [[,abc,]]) caused
751        pcre_compile() to give the error "Failed: internal error: code overflow" or
752        in some cases to crash with a glibc free() error. This could even happen if
753        the pattern terminated after [[ but there just happened to be a sequence of
754        letters, a binary zero, and a closing ] in the memory that followed.
755    
756    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
757        over the years. Originally (before any Unicode support), just the bottom 8
758        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
759        output from "man perlunicode" includes this:
760    
761          The regular expression compiler produces polymorphic opcodes.  That
762          is, the pattern adapts to the data and automatically switches to
763          the Unicode character scheme when presented with Unicode data--or
764          instead uses a traditional byte scheme when presented with byte
765          data.
766    
767        Sadly, a wide octal escape does not cause a switch, and in a string with
768        no other multibyte characters, these octal escapes are treated as before.
769        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
770        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
771        Unicode string.
772    
773        I have not perpetrated such confusion in PCRE. Up till now, it took just
774        the bottom 8 bits, as in old Perl. I have now made octal escapes with
775        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
776        translate to the appropriate multibyte character.
777    
778    29. Applied some refactoring to reduce the number of warnings from Microsoft
779        and Borland compilers. This has included removing the fudge introduced
780        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
781        a warning about an unused variable.
782    
783    21. PCRE has not included VT (character 0x0b) in the set of whitespace
784        characters since release 4.0, because Perl (from release 5.004) does not.
785        [Or at least, is documented not to: some releases seem to be in conflict
786        with the documentation.] However, when a pattern was studied with
787        pcre_study() and all its branches started with \s, PCRE still included VT
788        as a possible starting character. Of course, this did no harm; it just
789        caused an unnecessary match attempt.
790    
791    22. Removed a now-redundant internal flag bit that recorded the fact that case
792        dependency changed within the pattern. This was once needed for "required
793        byte" processing, but is no longer used. This recovers a now-scarce options
794        bit. Also moved the least significant internal flag bit to the most-
795        significant bit of the word, which was not previously used (hangover from
796        the days when it was an int rather than a uint) to free up another bit for
797        the future.
798    
799    23. Added support for CRLF line endings as well as CR and LF. As well as the
800        default being selectable at build time, it can now be changed at runtime
801        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
802        specify that it is scanning data with non-default line endings.
803    
804    24. Changed the definition of CXXLINK to make it agree with the definition of
805        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
806    
807    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
808        recursions. This makes a big different to stack usage for some patterns.
809    
810    26. If a subpattern containing a named recursion or subroutine reference such
811        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
812        the space required for the compiled pattern went wrong and gave too small a
813        value. Depending on the environment, this could lead to "Failed: internal
814        error: code overflow at offset 49" or "glibc detected double free or
815        corruption" errors.
816    
817    27. Applied patches from Google (a) to support the new newline modes and (b) to
818        advance over multibyte UTF-8 characters in GlobalReplace.
819    
820    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
821        difference for some implementation of PCRE in some Windows version.
822    
823    29. Added some extra testing facilities to pcretest:
824    
825        \q<number>   in a data line sets the "match limit" value
826        \Q<number>   in a data line sets the "match recursion limt" value
827        -S <number>  sets the stack size, where <number> is in megabytes
828    
829        The -S option isn't available for Windows.
830    
831    
832    Version 6.6 06-Feb-06
833    ---------------------
834    
835     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
836        in pcreposix.h. I have copied the definition from pcre.h.
837    
838     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
839        because pcre.h is no longer a built file.
840    
841     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
842        not normally included in the compiled code.
843    
844    
845    Version 6.5 01-Feb-06
846    ---------------------
847    
848     1. When using the partial match feature with pcre_dfa_exec(), it was not
849        anchoring the second and subsequent partial matches at the new starting
850        point. This could lead to incorrect results. For example, with the pattern
851        /1234/, partially matching against "123" and then "a4" gave a match.
852    
853     2. Changes to pcregrep:
854    
855        (a) All non-match returns from pcre_exec() were being treated as failures
856            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
857            error message is output. Some extra information is given for the
858            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
859            probably the only errors that are likely to be caused by users (by
860            specifying a regex that has nested indefinite repeats, for instance).
861            If there are more than 20 of these errors, pcregrep is abandoned.
862    
863        (b) A binary zero was treated as data while matching, but terminated the
864            output line if it was written out. This has been fixed: binary zeroes
865            are now no different to any other data bytes.
866    
867        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
868            used to set a locale for matching. The --locale=xxxx long option has
869            been added (no short equivalent) to specify a locale explicitly on the
870            pcregrep command, overriding the environment variables.
871    
872        (d) When -B was used with -n, some line numbers in the output were one less
873            than they should have been.
874    
875        (e) Added the -o (--only-matching) option.
876    
877        (f) If -A or -C was used with -c (count only), some lines of context were
878            accidentally printed for the final match.
879    
880        (g) Added the -H (--with-filename) option.
881    
882        (h) The combination of options -rh failed to suppress file names for files
883            that were found from directory arguments.
884    
885        (i) Added the -D (--devices) and -d (--directories) options.
886    
887        (j) Added the -F (--fixed-strings) option.
888    
889        (k) Allow "-" to be used as a file name for -f as well as for a data file.
890    
891        (l) Added the --colo(u)r option.
892    
893        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
894            is not present by default.
895    
896     3. A nasty bug was discovered in the handling of recursive patterns, that is,
897        items such as (?R) or (?1), when the recursion could match a number of
898        alternatives. If it matched one of the alternatives, but subsequently,
899        outside the recursion, there was a failure, the code tried to back up into
900        the recursion. However, because of the way PCRE is implemented, this is not
901        possible, and the result was an incorrect result from the match.
902    
903        In order to prevent this happening, the specification of recursion has
904        been changed so that all such subpatterns are automatically treated as
905        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
906    
907     4. I had overlooked the fact that, in some locales, there are characters for
908        which isalpha() is true but neither isupper() nor islower() are true. In
909        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
910        and ordfeminine) are like this. This affected the treatment of \w and \W
911        when they appeared in character classes, but not when they appeared outside
912        a character class. The bit map for "word" characters is now created
913        separately from the results of isalnum() instead of just taking it from the
914        upper, lower, and digit maps. (Plus the underscore character, of course.)
915    
916     5. The above bug also affected the handling of POSIX character classes such as
917        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
918        permanent tables. Instead, the bit maps for such a class were previously
919        created as the appropriate unions of the upper, lower, and digit bitmaps.
920        Now they are created by subtraction from the [[:word:]] class, which has
921        its own bitmap.
922    
923     6. The [[:blank:]] character class matches horizontal, but not vertical space.
924        It is created by subtracting the vertical space characters (\x09, \x0a,
925        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
926        subtraction was done in the overall bitmap for a character class, meaning
927        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
928        be recognized. This bug has been fixed.
929    
930     7. Patches from the folks at Google:
931    
932          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
933          real life, but is still worth protecting against".
934    
935          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
936          regular expressions".
937    
938          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
939          have it.
940    
941          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
942          "configure" and the latter not, in order to fix a problem somebody had
943          with compiling the Arg class on HP-UX.
944    
945          (e) Improve the error-handling of the C++ wrapper a little bit.
946    
947          (f) New tests for checking recursion limiting.
948    
949     8. The pcre_memmove() function, which is used only if the environment does not
950        have a standard memmove() function (and is therefore rarely compiled),
951        contained two bugs: (a) use of int instead of size_t, and (b) it was not
952        returning a result (though PCRE never actually uses the result).
953    
954     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
955        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
956        returned instead of calling malloc() with an overflowing number that would
957        most likely cause subsequent chaos.
958    
959    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
960    
961    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
962        with this option is matched, the nmatch and pmatch options of regexec() are
963        ignored.
964    
965    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
966        provided in case anyone wants to the the POSIX interface with UTF-8
967        strings.
968    
969    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
970        C++ linking (needed for some HP-UX environments).
971    
972    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
973        (unused parameter) and in the pcre_printint() function (omitted "default"
974        switch label when the default is to do nothing).
975    
976    15. Added some code to make it possible, when PCRE is compiled as a C++
977        library, to replace subject pointers for pcre_exec() with a smart pointer
978        class, thus making it possible to process discontinuous strings.
979    
980    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
981        much the same function. They were added by different people who were trying
982        to make PCRE easy to compile on non-Unix systems. It has been suggested
983        that PCRE_EXPORT be abolished now that there is more automatic apparatus
984        for compiling on Windows systems. I have therefore replaced it with
985        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
986        defaults to "extern" for C or "extern C" for C++, which works fine on
987        Unix-like systems. It is now possible to override the value of PCRE_DATA_
988        SCOPE with something explicit in config.h. In addition:
989    
990        (a) pcreposix.h still had just "extern" instead of either of these macros;
991            I have replaced it with PCRE_DATA_SCOPE.
992    
993        (b) Functions such as _pcre_xclass(), which are internal to the library,
994            but external in the C sense, all had PCRE_EXPORT in their definitions.
995            This is apparently wrong for the Windows case, so I have removed it.
996            (It makes no difference on Unix-like systems.)
997    
998    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
999        of recursive calls to match(). This is different to MATCH_LIMIT because
1000        that limits the total number of calls to match(), not all of which increase
1001        the depth of recursion. Limiting the recursion depth limits the amount of
1002        stack (or heap if NO_RECURSE is set) that is used. The default can be set
1003        when PCRE is compiled, and changed at run time. A patch from Google adds
1004        this functionality to the C++ interface.
1005    
1006    18. Changes to the handling of Unicode character properties:
1007    
1008        (a) Updated the table to Unicode 4.1.0.
1009    
1010        (b) Recognize characters that are not in the table as "Cn" (undefined).
1011    
1012        (c) I revised the way the table is implemented to a much improved format
1013            which includes recognition of ranges. It now supports the ranges that
1014            are defined in UnicodeData.txt, and it also amalgamates other
1015            characters into ranges. This has reduced the number of entries in the
1016            table from around 16,000 to around 3,000, thus reducing its size
1017            considerably. I realized I did not need to use a tree structure after
1018            all - a binary chop search is just as efficient. Having reduced the
1019            number of entries, I extended their size from 6 bytes to 8 bytes to
1020            allow for more data.
1021    
1022        (d) Added support for Unicode script names via properties such as \p{Han}.
1023    
1024    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1025        matching that character.
1026    
1027    20. When matching a repeated Unicode property with a minimum greater than zero,
1028        (for example \pL{2,}), PCRE could look past the end of the subject if it
1029        reached it while seeking the minimum number of characters. This could
1030        happen only if some of the characters were more than one byte long, because
1031        there is a check for at least the minimum number of bytes.
1032    
1033    21. Refactored the implementation of \p and \P so as to be more general, to
1034        allow for more different types of property in future. This has changed the
1035        compiled form incompatibly. Anybody with saved compiled patterns that use
1036        \p or \P will have to recompile them.
1037    
1038    22. Added "Any" and "L&" to the supported property types.
1039    
1040    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1041        but give a compile time error if the value is greater than 0xff.
1042    
1043    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1044        accidentally not being installed or uninstalled.
1045    
1046    25. The pcre.h file was built from pcre.h.in, but the only changes that were
1047        made were to insert the current release number. This seemed silly, because
1048        it made things harder for people building PCRE on systems that don't run
1049        "configure". I have turned pcre.h into a distributed file, no longer built
1050        by "configure", with the version identification directly included. There is
1051        no longer a pcre.h.in file.
1052    
1053        However, this change necessitated a change to the pcre-config script as
1054        well. It is built from pcre-config.in, and one of the substitutions was the
1055        release number. I have updated configure.ac so that ./configure now finds
1056        the release number by grepping pcre.h.
1057    
1058    26. Added the ability to run the tests under valgrind.
1059    
1060    
1061    Version 6.4 05-Sep-05
1062    ---------------------
1063    
1064     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1065        "--" to be printed when multiple files were scanned, even when none of the
1066        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1067        consider it to be a bug, and have restored the previous behaviour.
1068    
1069     2. A couple of code tidies to get rid of compiler warnings.
1070    
1071     3. The pcretest program used to cheat by referring to symbols in the library
1072        whose names begin with _pcre_. These are internal symbols that are not
1073        really supposed to be visible externally, and in some environments it is
1074        possible to suppress them. The cheating is now confined to including
1075        certain files from the library's source, which is a bit cleaner.
1076    
1077     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1078        file's purpose clearer.
1079    
1080     5. Reorganized pcre_ucp_findchar().
1081    
1082    
1083    Version 6.3 15-Aug-05
1084    ---------------------
1085    
1086     1. The file libpcre.pc.in did not have general read permission in the tarball.
1087    
1088     2. There were some problems when building without C++ support:
1089    
1090        (a) If C++ support was not built, "make install" and "make test" still
1091            tried to test it.
1092    
1093        (b) There were problems when the value of CXX was explicitly set. Some
1094            changes have been made to try to fix these, and ...
1095    
1096        (c) --disable-cpp can now be used to explicitly disable C++ support.
1097    
1098        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1099            backslash in a target when C++ was disabled. This confuses some
1100            versions of "make", apparently. Using an intermediate variable solves
1101            this. (Same for CPP_LOBJ.)
1102    
1103     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1104        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1105        necessary on certain architectures.
1106    
1107     4. Added a setting of -export-symbols-regex to the link command to remove
1108        those symbols that are exported in the C sense, but actually are local
1109        within the library, and not documented. Their names all begin with
1110        "_pcre_". This is not a perfect job, because (a) we have to except some
1111        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1112        available (and never for static libraries). I have made a note to try to
1113        find a way round (a) in the future.
1114    
1115    
1116    Version 6.2 01-Aug-05
1117    ---------------------
1118    
1119     1. There was no test for integer overflow of quantifier values. A construction
1120        such as {1111111111111111} would give undefined results. What is worse, if
1121        a minimum quantifier for a parenthesized subpattern overflowed and became
1122        negative, the calculation of the memory size went wrong. This could have
1123        led to memory overwriting.
1124    
1125     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1126    
1127     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1128        operating environments where this matters.
1129    
1130     4. Applied Giuseppe Maxia's patch to add additional features for controlling
1131        PCRE options from within the C++ wrapper.
1132    
1133     5. Named capturing subpatterns were not being correctly counted when a pattern
1134        was compiled. This caused two problems: (a) If there were more than 100
1135        such subpatterns, the calculation of the memory needed for the whole
1136        compiled pattern went wrong, leading to an overflow error. (b) Numerical
1137        back references of the form \12, where the number was greater than 9, were
1138        not recognized as back references, even though there were sufficient
1139        previous subpatterns.
1140    
1141     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1142        versions of gcc, e.g. 2.95.4.
1143    
1144    
1145    Version 6.1 21-Jun-05
1146    ---------------------
1147    
1148     1. There was one reference to the variable "posix" in pcretest.c that was not
1149        surrounded by "#if !defined NOPOSIX".
1150    
1151     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1152        the cross-check on the old pcre_info() function, for the benefit of the
1153        cut-down version of PCRE that is currently imported into Exim.
1154    
1155     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1156        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1157        patterns that start (?i) but I don't think that matters. The use of (?i) is
1158        just an example; this all applies to the other options as well.
1159    
1160     4. Since libtool seems to echo the compile commands it is issuing, the output
1161        from "make" can be reduced a bit by putting "@" in front of each libtool
1162        compile command.
1163    
1164     5. Patch from the folks at Google for configure.in to be a bit more thorough
1165        in checking for a suitable C++ installation before trying to compile the
1166        C++ stuff. This should fix a reported problem when a compiler was present,
1167        but no suitable headers.
1168    
1169     6. The man pages all had just "PCRE" as their title. I have changed them to
1170        be the relevant file name. I have also arranged that these names are
1171        retained in the file doc/pcre.txt, which is a concatenation in text format
1172        of all the man pages except the little individual ones for each function.
1173    
1174     7. The NON-UNIX-USE file had not been updated for the different set of source
1175        files that come with release 6. I also added a few comments about the C++
1176        wrapper.
1177    
1178    
1179    Version 6.0 07-Jun-05
1180    ---------------------
1181    
1182     1. Some minor internal re-organization to help with my DFA experiments.
1183    
1184     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1185        didn't matter for the library itself when fully configured, but did matter
1186        when compiling without UCP support, or within Exim, where the ucp files are
1187        not imported.
1188    
1189     3. Refactoring of the library code to split up the various functions into
1190        different source modules. The addition of the new DFA matching code (see
1191        below) to a single monolithic source would have made it really too
1192        unwieldy, quite apart from causing all the code to be include in a
1193        statically linked application, when only some functions are used. This is
1194        relevant even without the DFA addition now that patterns can be compiled in
1195        one application and matched in another.
1196    
1197        The downside of splitting up is that there have to be some external
1198        functions and data tables that are used internally in different modules of
1199        the library but which are not part of the API. These have all had their
1200        names changed to start with "_pcre_" so that they are unlikely to clash
1201        with other external names.
1202    
1203     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1204        a different (DFA) algorithm. Although it is slower than the original
1205        function, it does have some advantages for certain types of matching
1206        problem.
1207    
1208     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1209        including restarting after a partial match.
1210    
1211     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1212        defined when compiling for Windows was sent to me. I have put it into the
1213        code, though I have no means of testing or verifying it.
1214    
1215     7. Added the pcre_refcount() auxiliary function.
1216    
1217     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1218        match before or at the first newline in the subject string. In pcretest,
1219        the /f option on a pattern can be used to set this.
1220    
1221     9. A repeated \w when used in UTF-8 mode with characters greater than 256
1222        would behave wrongly. This has been present in PCRE since release 4.0.
1223    
1224    10. A number of changes to the pcregrep command:
1225    
1226        (a) Refactored how -x works; insert ^(...)$ instead of setting
1227            PCRE_ANCHORED and checking the length, in preparation for adding
1228            something similar for -w.
1229    
1230        (b) Added the -w (match as a word) option.
1231    
1232        (c) Refactored the way lines are read and buffered so as to have more
1233            than one at a time available.
1234    
1235        (d) Implemented a pcregrep test script.
1236    
1237        (e) Added the -M (multiline match) option. This allows patterns to match
1238            over several lines of the subject. The buffering ensures that at least
1239            8K, or the rest of the document (whichever is the shorter) is available
1240            for matching (and similarly the previous 8K for lookbehind assertions).
1241    
1242        (f) Changed the --help output so that it now says
1243    
1244              -w, --word-regex(p)
1245    
1246            instead of two lines, one with "regex" and the other with "regexp"
1247            because that confused at least one person since the short forms are the
1248            same. (This required a bit of code, as the output is generated
1249            automatically from a table. It wasn't just a text change.)
1250    
1251        (g) -- can be used to terminate pcregrep options if the next thing isn't an
1252            option but starts with a hyphen. Could be a pattern or a path name
1253            starting with a hyphen, for instance.
1254    
1255        (h) "-" can be given as a file name to represent stdin.
1256    
1257        (i) When file names are being printed, "(standard input)" is used for
1258            the standard input, for compatibility with GNU grep. Previously
1259            "<stdin>" was used.
1260    
1261        (j) The option --label=xxx can be used to supply a name to be used for
1262            stdin when file names are being printed. There is no short form.
1263    
1264        (k) Re-factored the options decoding logic because we are going to add
1265            two more options that take data. Such options can now be given in four
1266            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1267    
1268        (l) Added the -A, -B, and -C options for requesting that lines of context
1269            around matches be printed.
1270    
1271        (m) Added the -L option to print the names of files that do not contain
1272            any matching lines, that is, the complement of -l.
1273    
1274        (n) The return code is 2 if any file cannot be opened, but pcregrep does
1275            continue to scan other files.
1276    
1277        (o) The -s option was incorrectly implemented. For compatibility with other
1278            greps, it now suppresses the error message for a non-existent or non-
1279            accessible file (but not the return code). There is a new option called
1280            -q that suppresses the output of matching lines, which was what -s was
1281            previously doing.
1282    
1283        (p) Added --include and --exclude options to specify files for inclusion
1284            and exclusion when recursing.
1285    
1286    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1287        Hopefully, it now does.
1288    
1289    12. Missing cast in pcre_study().
1290    
1291    13. Added an "uninstall" target to the makefile.
1292    
1293    14. Replaced "extern" in the function prototypes in Makefile.in with
1294        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1295        world, but is set differently for Windows.
1296    
1297    15. Added a second compiling function called pcre_compile2(). The only
1298        difference is that it has an extra argument, which is a pointer to an
1299        integer error code. When there is a compile-time failure, this is set
1300        non-zero, in addition to the error test pointer being set to point to an
1301        error message. The new argument may be NULL if no error number is required
1302        (but then you may as well call pcre_compile(), which is now just a
1303        wrapper). This facility is provided because some applications need a
1304        numeric error indication, but it has also enabled me to tidy up the way
1305        compile-time errors are handled in the POSIX wrapper.
1306    
1307    16. Added VPATH=.libs to the makefile; this should help when building with one
1308        prefix path and installing with another. (Or so I'm told by someone who
1309        knows more about this stuff than I do.)
1310    
1311    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1312        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1313        match everything, including newlines. This is not POSIX-compatible, but
1314        somebody wanted the feature. From pcretest it can be activated by using
1315        both the P and the s flags.
1316    
1317    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1318    
1319    19. libpcre.pc was being incorrectly installed as executable.
1320    
1321    20. A couple of places in pcretest check for end-of-line by looking for '\n';
1322        it now also looks for '\r' so that it will work unmodified on Windows.
1323    
1324    21. Added Google's contributed C++ wrapper to the distribution.
1325    
1326    22. Added some untidy missing memory free() calls in pcretest, to keep
1327        Electric Fence happy when testing.
1328    
1329    
1330    
1331    Version 5.0 13-Sep-04
1332    ---------------------
1333    
1334     1. Internal change: literal characters are no longer packed up into items
1335        containing multiple characters in a single byte-string. Each character
1336        is now matched using a separate opcode. However, there may be more than one
1337        byte in the character in UTF-8 mode.
1338    
1339     2. The pcre_callout_block structure has two new fields: pattern_position and
1340        next_item_length. These contain the offset in the pattern to the next match
1341        item, and its length, respectively.
1342    
1343     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1344        insertion of callouts before each pattern item. Added the /C option to
1345        pcretest to make use of this.
1346    
1347     4. On the advice of a Windows user, the lines
1348    
1349          #if defined(_WIN32) || defined(WIN32)
1350          _setmode( _fileno( stdout ), 0x8000 );
1351          #endif  /* defined(_WIN32) || defined(WIN32) */
1352    
1353        have been added to the source of pcretest. This apparently does useful
1354        magic in relation to line terminators.
1355    
1356     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1357        for the benefit of those environments where the "b" makes a difference.
1358    
1359     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1360        to know about it. I have put a hack into configure.in that adds in code
1361        to set GCC=yes if CC=icc. This seems to end up at a point in the
1362        generated configure script that is early enough to affect the setting of
1363        compiler options, which is what is needed, but I have no means of testing
1364        whether it really works. (The user who reported this had patched the
1365        generated configure script, which of course I cannot do.)
1366    
1367        LATER: After change 22 below (new libtool files), the configure script
1368        seems to know about icc (and also ecc). Therefore, I have commented out
1369        this hack in configure.in.
1370    
1371     7. Added support for pkg-config (2 patches were sent in).
1372    
1373     8. Negated POSIX character classes that used a combination of internal tables
1374        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1375        [[:^ascii]]. Typically, they would match almost any characters. The other
1376        POSIX classes were not broken in this way.
1377    
1378     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1379        to find the match, as PCRE was deluded into thinking that the match had to
1380        start at the start point or following a newline. The same bug applied to
1381        patterns with negative forward assertions or any backward assertions
1382        preceding ".*" at the start, unless the pattern required a fixed first
1383        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1384    
1385    10. In UTF-8 mode, when moving forwards in the subject after a failed match
1386        starting at the last subject character, bytes beyond the end of the subject
1387        string were read.
1388    
1389    11. Renamed the variable "class" as "classbits" to make life easier for C++
1390        users. (Previously there was a macro definition, but it apparently wasn't
1391        enough.)
1392    
1393    12. Added the new field "tables" to the extra data so that tables can be passed
1394        in at exec time, or the internal tables can be re-selected. This allows
1395        a compiled regex to be saved and re-used at a later time by a different
1396        program that might have everything at different addresses.
1397    
1398    13. Modified the pcre-config script so that, when run on Solaris, it shows a
1399        -R library as well as a -L library.
1400    
1401    14. The debugging options of pcretest (-d on the command line or D on a
1402        pattern) showed incorrect output for anything following an extended class
1403        that contained multibyte characters and which was followed by a quantifier.
1404    
1405    15. Added optional support for general category Unicode character properties
1406        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1407        support. It adds about 90K to the size of the library. The meanings of the
1408        inbuilt class escapes such as \d and \s have NOT been changed.
1409    
1410    16. Updated pcredemo.c to include calls to free() to release the memory for the
1411        compiled pattern.
1412    
1413    17. The generated file chartables.c was being created in the source directory
1414        instead of in the building directory. This caused the build to fail if the
1415        source directory was different from the building directory, and was
1416        read-only.
1417    
1418    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1419        file. No doubt somebody will tell me if they don't make sense... Also added
1420        Dan Mooney's comments about building on OpenVMS.
1421    
1422    19. Added support for partial matching via the PCRE_PARTIAL option for
1423        pcre_exec() and the \P data escape in pcretest.
1424    
1425    20. Extended pcretest with 3 new pattern features:
1426    
1427        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
1428              write the compiled pattern to the file whose name is "rest-of-line".
1429              This is a straight binary dump of the data, with the saved pointer to
1430              the character tables forced to be NULL. The study data, if any, is
1431              written too. After writing, pcretest reads a new pattern.
1432    
1433        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1434              compiled pattern from the given file. There must not be any
1435              occurrences of "<" in the file name (pretty unlikely); if there are,
1436              pcretest will instead treat the initial "<" as a pattern delimiter.
1437              After reading in the pattern, pcretest goes on to read data lines as
1438              usual.
1439    
1440        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1441              and 16-bit fields in a compiled pattern, to simulate a pattern that
1442              was compiled on a host of opposite endianness.
1443    
1444    21. The pcre-exec() function can now cope with patterns that were compiled on
1445        hosts of opposite endianness, with this restriction:
1446    
1447          As for any compiled expression that is saved and used later, the tables
1448          pointer field cannot be preserved; the extra_data field in the arguments
1449          to pcre_exec() should be used to pass in a tables address if a value
1450          other than the default internal tables were used at compile time.
1451    
1452    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1453        now diagnosed as an error. Previously, most of the time, a negative number
1454        would have been treated as zero, but if in addition "ovector" was passed as
1455        NULL, a crash could occur.
1456    
1457    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1458        new versions from the libtool 1.5 distribution (the last one is a copy of
1459        a file called libtool.m4). This seems to have fixed the need to patch
1460        "configure" to support Darwin 1.3 (which I used to do). However, I still
1461        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1462        workstation).
1463    
1464    24. Changed the PCRE licence to be the more standard "BSD" licence.
1465    
1466    
1467    Version 4.5 01-Dec-03
1468    ---------------------
1469    
1470     1. There has been some re-arrangement of the code for the match() function so
1471        that it can be compiled in a version that does not call itself recursively.
1472        Instead, it keeps those local variables that need separate instances for
1473        each "recursion" in a frame on the heap, and gets/frees frames whenever it
1474        needs to "recurse". Keeping track of where control must go is done by means
1475        of setjmp/longjmp. The whole thing is implemented by a set of macros that
1476        hide most of the details from the main code, and operates only if
1477        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1478        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1479        operating.
1480    
1481        To make it easier for callers to provide specially tailored get/free
1482        functions for this usage, two new functions, pcre_stack_malloc, and
1483        pcre_stack_free, are used. They are always called in strict stacking order,
1484        and the size of block requested is always the same.
1485    
1486        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1487        PCRE has been compiled to use the stack or the heap for recursion. The
1488        -C option of pcretest uses this to show which version is compiled.
1489    
1490        A new data escape \S, is added to pcretest; it causes the amounts of store
1491        obtained and freed by both kinds of malloc/free at match time to be added
1492        to the output.
1493    
1494     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1495        what's available on my current Linux desktop machine.
1496    
1497     3. When matching a UTF-8 string, the test for a valid string at the start has
1498        been extended. If start_offset is not zero, PCRE now checks that it points
1499        to a byte that is the start of a UTF-8 character. If not, it returns
1500        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1501        this is necessary because there may be backward assertions in the pattern.
1502        When matching the same subject several times, it may save resources to use
1503        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1504    
1505     4. The code for checking the validity of UTF-8 strings has been tightened so
1506        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1507        containing "overlong sequences".
1508    
1509     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1510        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1511        should have been "&", but it just so happened that all the cases this let
1512        through by mistake were picked up later in the function.
1513    
1514     6. I had used a variable called "isblank" - this is a C99 function, causing
1515        some compilers to warn. To avoid this, I renamed it (as "blankclass").
1516    
1517     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1518        prompting; (b) run "./pcretest /dev/null" at the start of the test script
1519        so the version is shown; (c) stop "make test" echoing "./RunTest".
1520    
1521     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1522    
1523     9. The prototype for memmove() for systems that don't have it was using
1524        size_t, but the inclusion of the header that defines size_t was later. I've
1525        moved the #includes for the C headers earlier to avoid this.
1526    
1527    10. Added some adjustments to the code to make it easier to compiler on certain
1528        special systems:
1529    
1530          (a) Some "const" qualifiers were missing.
1531          (b) Added the macro EXPORT before all exported functions; by default this
1532              is defined to be empty.
1533          (c) Changed the dftables auxiliary program (that builds chartables.c) so
1534              that it reads its output file name as an argument instead of writing
1535              to the standard output and assuming this can be redirected.
1536    
1537    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
1538        class containing characters with values greater than 255, PCRE compilation
1539        went into a loop.
1540    
1541    12. A recursive reference to a subpattern that was within another subpattern
1542        that had a minimum quantifier of zero caused PCRE to crash. For example,
1543        (x(y(?2))z)? provoked this bug with a subject that got as far as the
1544        recursion. If the recursively-called subpattern itself had a zero repeat,
1545        that was OK.
1546    
1547    13. In pcretest, the buffer for reading a data line was set at 30K, but the
1548        buffer into which it was copied (for escape processing) was still set at
1549        1024, so long lines caused crashes.
1550    
1551    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
1552        "internal error: code overflow...". This applied to any character class
1553        that was followed by a possessive quantifier.
1554    
1555    15. Modified the Makefile to add libpcre.la as a prerequisite for
1556        libpcreposix.la because I was told this is needed for a parallel build to
1557        work.
1558    
1559    16. If a pattern that contained .* following optional items at the start was
1560        studied, the wrong optimizing data was generated, leading to matching
1561        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
1562        matching string must start with a or b or c. The correct conclusion for
1563        this pattern is that a match can start with any character.
1564    
1565    
1566    Version 4.4 13-Aug-03
1567    ---------------------
1568    
1569     1. In UTF-8 mode, a character class containing characters with values between
1570        127 and 255 was not handled correctly if the compiled pattern was studied.
1571        In fixing this, I have also improved the studying algorithm for such
1572        classes (slightly).
1573    
1574     2. Three internal functions had redundant arguments passed to them. Removal
1575        might give a very teeny performance improvement.
1576    
1577     3. Documentation bug: the value of the capture_top field in a callout is *one
1578        more than* the number of the hightest numbered captured substring.
1579    
1580     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
1581        in incorrectly linking with a previously installed version. They now link
1582        explicitly with libpcre.la.
1583    
1584     5. configure.in no longer needs to recognize Cygwin specially.
1585    
1586     6. A problem in pcre.in for Windows platforms is fixed.
1587    
1588     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
1589        pcretest, it used to include the size of the study block as part of its
1590        output. Unfortunately, the structure contains a field that has a different
1591        size on different hardware architectures. This meant that the tests that
1592        showed this size failed. As the block is currently always of a fixed size,
1593        this information isn't actually particularly useful in pcretest output, so
1594        I have just removed it.
1595    
1596     8. Three pre-processor statements accidentally did not start in column 1.
1597        Sadly, there are *still* compilers around that complain, even though
1598        standard C has not required this for well over a decade. Sigh.
1599    
1600     9. In pcretest, the code for checking callouts passed small integers in the
1601        callout_data field, which is a void * field. However, some picky compilers
1602        complained about the casts involved for this on 64-bit systems. Now
1603        pcretest passes the address of the small integer instead, which should get
1604        rid of the warnings.
1605    
1606    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
1607        both compile and run time, and gives an error if an invalid UTF-8 sequence
1608        is found. There is a option for disabling this check in cases where the
1609        string is known to be correct and/or the maximum performance is wanted.
1610    
1611    11. In response to a bug report, I changed one line in Makefile.in from
1612    
1613            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
1614        to
1615            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
1616    
1617        to look similar to other lines, but I have no way of telling whether this
1618        is the right thing to do, as I do not use Windows. No doubt I'll get told
1619        if it's wrong...
1620    
1621    
1622    Version 4.3 21-May-03
1623    ---------------------
1624    
1625    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
1626       Makefile.
1627    
1628    2. Some refactoring to improve the quality of the code:
1629    
1630       (i)   The utf8_table... variables are now declared "const".
1631    
1632       (ii)  The code for \cx, which used the "case flipping" table to upper case
1633             lower case letters, now just substracts 32. This is ASCII-specific,
1634             but the whole concept of \cx is ASCII-specific, so it seems
1635             reasonable.
1636    
1637       (iii) PCRE was using its character types table to recognize decimal and
1638             hexadecimal digits in the pattern. This is silly, because it handles
1639             only 0-9, a-f, and A-F, but the character types table is locale-
1640             specific, which means strange things might happen. A private
1641             table is now used for this - though it costs 256 bytes, a table is
1642             much faster than multiple explicit tests. Of course, the standard
1643             character types table is still used for matching digits in subject
1644             strings against \d.
1645    
1646       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
1647             ending in _t are). So I've renamed it as ESC_tee.
1648    
1649    3. The first argument for regexec() in the POSIX wrapper should have been
1650       defined as "const".
1651    
1652    4. Changed pcretest to use malloc() for its buffers so that they can be
1653       Electric Fenced for debugging.
1654    
1655    5. There were several places in the code where, in UTF-8 mode, PCRE would try
1656       to read one or more bytes before the start of the subject string. Often this
1657       had no effect on PCRE's behaviour, but in some circumstances it could
1658       provoke a segmentation fault.
1659    
1660    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
1661       to try to read one or more bytes before the start of the subject string.
1662    
1663    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
1664       UTF-8 support could misbehave in various ways if the subject string
1665       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
1666       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
1667       back over UTF-8 characters.)
1668    
1669    
1670    Version 4.2 14-Apr-03
1671    ---------------------
1672    
1673    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
1674    
1675    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
1676         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
1677         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
1678         [WIN_PREFIX]: new variable, "cyg" for Cygwin
1679         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
1680           and BUILD_EXEEXT
1681         Note: automatic setting of the BUILD variables is not yet working
1682         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
1683           compile-time but not at link-time
1684         [LINK]: use for linking executables only
1685         make different versions for Windows and non-Windows
1686         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
1687           libraries
1688         [LINK_FOR_BUILD]: new variable
1689         [OBJEXT]: use throughout
1690         [EXEEXT]: use throughout
1691         <winshared>: new target
1692         <wininstall>: new target
1693         <dftables.o>: use native compiler
1694         <dftables>: use native linker
1695         <install>: handle Windows platform correctly
1696         <clean>: ditto
1697         <check>: ditto
1698         copy DLL to top builddir before testing
1699    
1700       As part of these changes, -no-undefined was removed again. This was reported
1701       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
1702       in any case.
1703    
1704    3. Some tidies to get rid of compiler warnings:
1705    
1706       . In the match_data structure, match_limit was an unsigned long int, whereas
1707         match_call_count was an int. I've made them both unsigned long ints.
1708    
1709       . In pcretest the fact that a const uschar * doesn't automatically cast to
1710         a void * provoked a warning.
1711    
1712       . Turning on some more compiler warnings threw up some "shadow" variables
1713         and a few more missing casts.
1714    
1715    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
1716       option, a class that contained a single character with a value between 128
1717       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
1718    
1719    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
1720       option, a class that contained several characters, but with at least one
1721       whose value was between 128 and 255 caused PCRE to crash.
1722    
1723    
1724    Version 4.1 12-Mar-03
1725    ---------------------
1726    
1727    1. Compiling with gcc -pedantic found a couple of places where casts were
1728    needed, and a string in dftables.c that was longer than standard compilers are
1729    required to support.
1730    
1731    2. Compiling with Sun's compiler found a few more places where the code could
1732    be tidied up in order to avoid warnings.
1733    
1734    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
1735    first of these names is deprecated in the latest Autoconf in favour of the name
1736    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
1737    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
1738    analogy I have changed it to CFLAGS_FOR_BUILD.
1739    
1740    4. Added -no-undefined to the linking command in the Makefile, because this is
1741    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
1742    linking step for the pcreposix library.
1743    
1744    5. PCRE was failing to diagnose the case of two named groups with the same
1745    name.
1746    
1747    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
1748    literal character that is needed in the subject for a match, and scans along to
1749    ensure that it is present before embarking on the full matching process. This
1750    saves time in cases of nested unlimited repeats that are never going to match.
1751    Problem: the scan can take a lot of time if the subject is very long (e.g.
1752    megabytes), thus penalizing straightforward matches. It is now done only if the
1753    amount of subject to be scanned is less than 1000 bytes.
1754    
1755    7. A lesser problem with the same optimization is that it was recording the
1756    first character of an anchored pattern as "needed", thus provoking a search
1757    right along the subject, even when the first match of the pattern was going to
1758    fail. The "needed" character is now not set for anchored patterns, unless it
1759    follows something in the pattern that is of non-fixed length. Thus, it still
1760    fulfils its original purpose of finding quick non-matches in cases of nested
1761    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
1762    
1763    
1764    Version 4.0 17-Feb-03
1765    ---------------------
1766    
1767    1. If a comment in an extended regex that started immediately after a meta-item
1768    extended to the end of string, PCRE compiled incorrect data. This could lead to
1769    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
1770    
1771    2. Moved to autoconf 2.53 and libtool 1.4.2.
1772    
1773    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
1774    the special perltest8 script is no longer needed - all the tests can be run
1775    from a single perltest script.
1776    
1777    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
1778    by \s. It has now been removed in PCRE. This means it isn't recognized as
1779    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
1780    class [:space:] *does* include VT, thereby creating a mess.
1781    
1782    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
1783    space and tab.
1784    
1785    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
1786    its new features into the main test script, reducing the number of scripts.
1787    
1788    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
1789    were backward compatible, and made the (?i) apply to the whole pattern, as if
1790    /i were given. Now it behaves more logically, and applies the option setting
1791    only to what follows. PCRE has been changed to follow suit. However, if it
1792    finds options settings right at the start of the pattern, it extracts them into
1793    the global options, as before. Thus, they show up in the info data.
1794    
1795    8. Added support for the \Q...\E escape sequence. Characters in between are
1796    treated as literals. This is slightly different from Perl in that $ and @ are
1797    also handled as literals inside the quotes. In Perl, they will cause variable
1798    interpolation. Note the following examples:
1799    
1800        Pattern            PCRE matches      Perl matches
1801    
1802        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
1803        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
1804        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
1805    
1806    For compatibility with Perl, \Q...\E sequences are recognized inside character
1807    classes as well as outside them.
1808    
1809    9. Re-organized 3 code statements in pcretest to avoid "overflow in
1810    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
1811    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
1812    signed/unsigned warnings.
1813    
1814    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
1815    option for pcretest, so I've replaced it by a simple function that does just
1816    that job.
1817    
1818    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
1819    "pcregrep -".
1820    
1821    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
1822    Java package. This provides some syntactic sugar for simple cases of what my
1823    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
1824    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
1825    item, you can use this simplified notation. Note that only makes sense with
1826    greedy quantifiers. Consequently, the use of the possessive quantifier forces
1827    greediness, whatever the setting of the PCRE_UNGREEDY option.
1828    
1829    13. A change of greediness default within a pattern was not taking effect at
1830    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
1831    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
1832    was abstracted outside.
1833    
1834    14. PCRE now supports the \G assertion. It is true when the current matching
1835    position is at the start point of the match. This differs from \A when the
1836    starting offset is non-zero. Used with the /g option of pcretest (or similar
1837    code), it works in the same way as it does for Perl's /g option. If all
1838    alternatives of a regex begin with \G, the expression is anchored to the start
1839    match position, and the "anchored" flag is set in the compiled expression.
1840    
1841    15. Some bugs concerning the handling of certain option changes within patterns
1842    have been fixed. These applied to options other than (?ims). For example,
1843    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
1844    been the other way round. Some of this was related to change 7 above.
1845    
1846    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
1847    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
1848    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
1849    POSIX classes only within a class (e.g. /[[:alpha:]]/).
1850    
1851    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
1852    mode. Unlike ".", it always matches newline, whatever the setting of
1853    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
1854    assertions. Perl allows it, but it doesn't (in general) work because it can't
1855    calculate the length of the lookbehind. At least, that's the case for Perl
1856    5.8.0 - I've been told they are going to document that it doesn't work in
1857    future.
1858    
1859    18. Added an error diagnosis for escapes that PCRE does not support: these are
1860    \L, \l, \N, \P, \p, \U, \u, and \X.
1861    
1862    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
1863    reading past the end of the pattern in cases such as /[abcd/.
1864    
1865    20. PCRE was getting more memory than necessary for patterns with classes that
1866    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
1867    
1868    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
1869    compiling PCRE for use with Virtual Pascal.
1870    
1871    22. Small fix to the Makefile to make it work properly if the build is done
1872    outside the source tree.
1873    
1874    23. Added a new extension: a condition to go with recursion. If a conditional
1875    subpattern starts with (?(R) the "true" branch is used if recursion has
1876    happened, whereas the "false" branch is used only at the top level.
1877    
1878    24. When there was a very long string of literal characters (over 255 bytes
1879    without UTF support, over 250 bytes with UTF support), the computation of how
1880    much memory was required could be incorrect, leading to segfaults or other
1881    strange effects.
1882    
1883    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
1884    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
1885    there was a subsequent back reference to those brackets. This meant that, for
1886    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
1887    possible to check for precisely this case. All we can do is abandon the
1888    optimization if .* occurs inside capturing brackets when there are any back
1889    references whatsoever. (See below for a better fix that came later.)
1890    
1891    26. The handling of the optimization for finding the first character of a
1892    non-anchored pattern, and for finding a character that is required later in the
1893    match were failing in some cases. This didn't break the matching; it just
1894    failed to optimize when it could. The way this is done has been re-implemented.
1895    
1896    27. Fixed typo in error message for invalid (?R item (it said "(?p").
1897    
1898    28. Added a new feature that provides some of the functionality that Perl
1899    provides with (?{...}). The facility is termed a "callout". The way it is done
1900    in PCRE is for the caller to provide an optional function, by setting
1901    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
1902    global variable. By default it is unset, which disables all calling out. To get
1903    the function called, the regex must include (?C) at appropriate points. This
1904    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
1905    This provides a means of identifying different callout points. When PCRE
1906    reaches such a point in the regex, if pcre_callout has been set, the external
1907    function is called. It is provided with data in a structure called
1908    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
1909    matching continues; if it returns a non-zero value, the match at the current
1910    point fails. However, backtracking will occur if possible. [This was changed
1911    later and other features added - see item 49 below.]
1912    
1913    29. pcretest is upgraded to test the callout functionality. It provides a
1914    callout function that displays information. By default, it shows the start of
1915    the match and the current position in the text. There are some new data escapes
1916    to vary what happens:
1917    
1918        \C+         in addition, show current contents of captured substrings
1919        \C-         do not supply a callout function
1920        \C!n        return 1 when callout number n is reached
1921        \C!n!m      return 1 when callout number n is reached for the mth time
1922    
1923    30. If pcregrep was called with the -l option and just a single file name, it
1924    output "<stdin>" if a match was found, instead of the file name.
1925    
1926    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
1927    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
1928    pcre_exec(). This saves a malloc/free per call. The default value of
1929    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
1930    when configuring.
1931    
1932    32. The default maximum size of a compiled pattern is 64K. There have been a
1933    few cases of people hitting this limit. The code now uses macros to handle the
1934    storing of links as offsets within the compiled pattern. It defaults to 2-byte
1935    links, but this can be changed to 3 or 4 bytes by --with-link-size when
1936    configuring. Tests 2 and 5 work only with 2-byte links because they output
1937    debugging information about compiled patterns.
1938    
1939    33. Internal code re-arrangements:
1940    
1941    (a) Moved the debugging function for printing out a compiled regex into
1942        its own source file (printint.c) and used #include to pull it into
1943        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
1944        separate copies.
1945    
1946    (b) Defined the list of op-code names for debugging as a macro in
1947        internal.h so that it is next to the definition of the opcodes.
1948    
1949    (c) Defined a table of op-code lengths for simpler skipping along compiled
1950        code. This is again a macro in internal.h so that it is next to the
1951        definition of the opcodes.
1952    
1953    34. Added support for recursive calls to individual subpatterns, along the
1954    lines of Robin Houston's patch (but implemented somewhat differently).
1955    
1956    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
1957    allow it to read and process whole directories in Win32. This code was
1958    contributed by Lionel Fourquaux; it has not been tested by me.
1959    
1960    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
1961    used to name a group. Names consist of alphanumerics and underscores, and must
1962    be unique. Back references use the syntax (?P=name) and recursive calls use
1963    (?P>name) which is a PCRE extension to the Python extension. Groups still have
1964    numbers. The function pcre_fullinfo() can be used after compilation to extract
1965    a name/number map. There are three relevant calls:
1966    
1967      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
1968      PCRE_INFO_NAMECOUNT            yields the number of entries
1969      PCRE_INFO_NAMETABLE            yields a pointer to the map.
1970    
1971    The map is a vector of fixed-size entries. The size of each entry depends on
1972    the length of the longest name used. The first two bytes of each entry are the
1973    group number, most significant byte first. There follows the corresponding
1974    name, zero terminated. The names are in alphabetical order.
1975    
1976    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
1977    case instead of 255. Making it the same both with and without UTF-8 support
1978    means that the same test output works with both.
1979    
1980    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
1981    calling malloc() with a zero argument.
1982    
1983    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
1984    optimization. I've improved things by keeping a bitmap of backreferences with
1985    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
1986    fact referenced, the optimization can be applied. It is unlikely that a
1987    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
1988    the match to follow \n) will appear inside brackets with a number greater than
1989    31, but if it does, any back reference > 31 suppresses the optimization.
1990    
1991    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
1992    of disabling numbered capturing parentheses. Any opening parenthesis that is
1993    not followed by ? behaves as if it were followed by ?: but named parentheses
1994    can still be used for capturing (and they will acquire numbers in the usual
1995    way).
1996    
1997    41. Redesigned the return codes from the match() function into yes/no/error so
1998    that errors can be passed back from deep inside the nested calls. A malloc
1999    failure while inside a recursive subpattern call now causes the
2000    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
2001    
2002    42. It is now possible to set a limit on the number of times the match()
2003    function is called in a call to pcre_exec(). This facility makes it possible to
2004    limit the amount of recursion and backtracking, though not in a directly
2005    obvious way, because the match() function is used in a number of different
2006    circumstances. The count starts from zero for each position in the subject
2007    string (for non-anchored patterns). The default limit is, for compatibility, a
2008    large number, namely 10 000 000. You can change this in two ways:
2009    
2010    (a) When configuring PCRE before making, you can use --with-match-limit=n
2011        to set a default value for the compiled library.
2012    
2013    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
2014        a different value is set. See 45 below.
2015    
2016    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
2017    
2018    43. Added a new function pcre_config(int, void *) to enable run-time extraction
2019    of things that can be changed at compile time. The first argument specifies
2020    what is wanted and the second points to where the information is to be placed.
2021    The current list of available information is:
2022    
2023      PCRE_CONFIG_UTF8
2024    
2025    The output is an integer that is set to one if UTF-8 support is available;
2026    otherwise it is set to zero.
2027    
2028      PCRE_CONFIG_NEWLINE
2029    
2030    The output is an integer that it set to the value of the code that is used for
2031    newline. It is either LF (10) or CR (13).
2032    
2033      PCRE_CONFIG_LINK_SIZE
2034    
2035    The output is an integer that contains the number of bytes used for internal
2036    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
2037    
2038      PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
2039    
2040    The output is an integer that contains the threshold above which the POSIX
2041    interface uses malloc() for output vectors. See item 31 above.
2042    
2043      PCRE_CONFIG_MATCH_LIMIT
2044    
2045    The output is an unsigned integer that contains the default limit of the number
2046    of match() calls in a pcre_exec() execution. See 42 above.
2047    
2048    44. pcretest has been upgraded by the addition of the -C option. This causes it
2049    to extract all the available output from the new pcre_config() function, and to
2050    output it. The program then exits immediately.
2051    
2052    45. A need has arisen to pass over additional data with calls to pcre_exec() in
2053    order to support additional features. One way would have been to define
2054    pcre_exec2() (for example) with extra arguments, but this would not have been
2055    extensible, and would also have required all calls to the original function to
2056    be mapped to the new one. Instead, I have chosen to extend the mechanism that
2057    is used for passing in "extra" data from pcre_study().
2058    
2059    The pcre_extra structure is now exposed and defined in pcre.h. It currently
2060    contains the following fields:
2061    
2062      flags         a bitmap indicating which of the following fields are set
2063      study_data    opaque data from pcre_study()
2064      match_limit   a way of specifying a limit on match() calls for a specific
2065                      call to pcre_exec()
2066      callout_data  data for callouts (see 49 below)
2067    
2068    The flag bits are also defined in pcre.h, and are
2069    
2070      PCRE_EXTRA_STUDY_DATA
2071      PCRE_EXTRA_MATCH_LIMIT
2072      PCRE_EXTRA_CALLOUT_DATA
2073    
2074    The pcre_study() function now returns one of these new pcre_extra blocks, with
2075    the actual study data pointed to by the study_data field, and the
2076    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
2077    before. That is, this change is entirely upwards-compatible and requires no
2078    change to existing code.
2079    
2080    If you want to pass in additional data to pcre_exec(), you can either place it
2081    in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
2082    block.
2083    
2084    46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
2085    data string contains the escape sequence \M, pcretest calls pcre_exec() several
2086    times with different match limits, until it finds the minimum value needed for
2087    pcre_exec() to complete. The value is then output. This can be instructive; for
2088    most simple matches the number is quite small, but for pathological cases it
2089    gets very large very quickly.
2090    
2091    47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
2092    returns the size of the data block pointed to by the study_data field in a
2093    pcre_extra block, that is, the value that was passed as the argument to
2094    pcre_malloc() when PCRE was getting memory in which to place the information
2095    created by pcre_study(). The fourth argument should point to a size_t variable.
2096    pcretest has been extended so that this information is shown after a successful
2097    pcre_study() call when information about the compiled regex is being displayed.
2098    
2099    48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
2100    because what follows is always an absolute path. (Later: it turns out that this
2101    is more than cosmetic for MinGW, because it doesn't like empty path
2102    components.)
2103    
2104    49. Some changes have been made to the callout feature (see 28 above):
2105    
2106    (i)  A callout function now has three choices for what it returns:
2107    
2108           0  =>  success, carry on matching
2109         > 0  =>  failure at this point, but backtrack if possible
2110         < 0  =>  serious error, return this value from pcre_exec()
2111    
2112         Negative values should normally be chosen from the set of PCRE_ERROR_xxx
2113         values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
2114         "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
2115         use by callout functions. It will never be used by PCRE itself.
2116    
2117    (ii) The pcre_extra structure (see 45 above) has a void * field called
2118         callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
2119         pcre_callout_block structure has a field of the same name. The contents of
2120         the field passed in the pcre_extra structure are passed to the callout
2121         function in the corresponding field in the callout block. This makes it
2122         easier to use the same callout-containing regex from multiple threads. For
2123         testing, the pcretest program has a new data escape
2124    
2125           \C*n        pass the number n (may be negative) as callout_data
2126    
2127         If the callout function in pcretest receives a non-zero value as
2128         callout_data, it returns that value.
2129    
2130    50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
2131    there were some redundant $(CFLAGS) in commands that are now specified as
2132    $(LINK), which already includes $(CFLAGS).
2133    
2134    51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
2135    has been compiled with UTF-8 support *and* pcre_compile() has been compiled
2136    with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
2137    one-byte characters throughout. Note that case-insensitive matching applies
2138    only to characters whose values are less than 256. PCRE doesn't support the
2139    notion of cases for higher-valued characters.
2140    
2141    (i)   A character class whose characters are all within 0-255 is handled as
2142          a bit map, and the map is inverted for negative classes. Previously, a
2143          character > 255 always failed to match such a class; however it should
2144          match if the class was a negative one (e.g. [^ab]). This has been fixed.
2145    
2146    (ii)  A negated character class with a single character < 255 is coded as
2147          "not this character" (OP_NOT). This wasn't working properly when the test
2148          character was multibyte, either singly or repeated.
2149    
2150    (iii) Repeats of multibyte characters are now handled correctly in UTF-8
2151          mode, for example: \x{100}{2,3}.
2152    
2153    (iv)  The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
2154          singly or repeated) now correctly test multibyte characters. However,
2155          PCRE doesn't recognize any characters with values greater than 255 as
2156          digits, spaces, or word characters. Such characters always match \D, \S,
2157          and \W, and never match \d, \s, or \w.
2158    
2159    (v)   Classes may now contain characters and character ranges with values
2160          greater than 255. For example: [ab\x{100}-\x{400}].
2161    
2162    (vi)  pcregrep now has a --utf-8 option (synonym -u) which makes it call
2163          PCRE in UTF-8 mode.
2164    
2165    52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
2166    PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
2167    retained for backwards compatibility. (Note that LASTLITERAL is also a byte
2168    value.)
2169    
2170    53. The single man page has become too large. I have therefore split it up into
2171    a number of separate man pages. These also give rise to individual HTML pages;
2172    these are now put in a separate directory, and there is an index.html page that
2173    lists them all. Some hyperlinking between the pages has been installed.
2174    
2175    54. Added convenience functions for handling named capturing parentheses.
2176    
2177    55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2178    aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2179    true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2180    are faulted.
2181    
2182    56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2183    calling configure. These values are used when compiling the dftables.c program
2184    which is run to generate the source of the default character tables. They
2185    default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2186    you will need to set these values.
2187    
2188    57. Updated the building process for Windows DLL, as provided by Fred Cox.
2189    
2190    
2191    Version 3.9 02-Jan-02
2192    ---------------------
2193    
2194    1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2195    
2196    2. If --disable-static was given, the building process failed when trying to
2197    build pcretest and pcregrep. (For some reason it was using libtool to compile
2198    them, which is not right, as they aren't part of the library.)
2199    
2200    
2201    Version 3.8 18-Dec-01
2202    ---------------------
2203    
2204    1. The experimental UTF-8 code was completely screwed up. It was packing the
2205    bytes in the wrong order. How dumb can you get?
2206    
2207    
2208    Version 3.7 29-Oct-01
2209    ---------------------
2210    
2211    1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2212    This caused pcretest, when used on the test data, to segfault. Unfortunately,
2213    this didn't happen under Solaris 8, where I normally test things.
2214    
2215    2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2216    doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2217    isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2218    this fix an hour or so after the initial 3.7 release.)
2219    
2220    
2221    Version 3.6 23-Oct-01
2222    ---------------------
2223    
2224    1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2225    offsets passed as NULL with zero offset count.
2226    
2227    2. The config.guess and config.sub files had not been updated when I moved to
2228    the latest autoconf.
2229    
2230    
2231    Version 3.5 15-Aug-01
2232    ---------------------
2233    
2234    1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2235    had been forgotten.
2236    
2237    2. By using declared but undefined structures, we can avoid using "void"
2238    definitions in pcre.h while keeping the internal definitions of the structures
2239    private.
2240    
2241    3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2242    user point of view, this means that both static and shared libraries are built
2243    by default, but this can be individually controlled. More of the work of
2244    handling this static/shared cases is now inside libtool instead of PCRE's make
2245    file.
2246    
2247    4. The pcretest utility is now installed along with pcregrep because it is
2248    useful for users (to test regexs) and by doing this, it automatically gets
2249    relinked by libtool. The documentation has been turned into a man page, so
2250    there are now .1, .txt, and .html versions in /doc.
2251    
2252    5. Upgrades to pcregrep:
2253       (i)   Added long-form option names like gnu grep.
2254       (ii)  Added --help to list all options with an explanatory phrase.
2255       (iii) Added -r, --recursive to recurse into sub-directories.
2256       (iv)  Added -f, --file to read patterns from a file.
2257    
2258    6. pcre_exec() was referring to its "code" argument before testing that
2259    argument for NULL (and giving an error if it was NULL).
2260    
2261    7. Upgraded Makefile.in to allow for compiling in a different directory from
2262    the source directory.
2263    
2264    8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2265    options bits, the pointer it was passed was to an int instead of to an unsigned
2266    long int. This mattered only on 64-bit systems.
2267    
2268    9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2269    generated) instead of pcre.in, which it its source. Also made the same change
2270    in several of the .c files.
2271    
2272    10. A new release of gcc defines printf() as a macro, which broke pcretest
2273    because it had an ifdef in the middle of a string argument for printf(). Fixed
2274    by using separate calls to printf().
2275    
2276    11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2277    script, to force use of CR or LF instead of \n in the source. On non-Unix
2278    systems, the value can be set in config.h.
2279    
2280    12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2281    absolute limit. Changed the text of the error message to make this clear, and
2282    likewise updated the man page.
2283    
2284    13. The limit of 99 on the number of capturing subpatterns has been removed.
2285    The new limit is 65535, which I hope will not be a "real" limit.
2286    
2287    
2288    Version 3.4 22-Aug-00
2289    ---------------------
2290    
2291    1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2292    
2293    2. Diagnose condition (?(0) as an error instead of crashing on matching.
2294    
2295    
2296    Version 3.3 01-Aug-00
2297    ---------------------
2298    
2299    1. If an octal character was given, but the value was greater than \377, it
2300    was not getting masked to the least significant bits, as documented. This could
2301    lead to crashes in some systems.
2302    
2303    2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
2304    the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
2305    
2306    3. Added the functions pcre_free_substring() and pcre_free_substring_list().
2307    These just pass their arguments on to (pcre_free)(), but they are provided
2308    because some uses of PCRE bind it to non-C systems that can call its functions,
2309    but cannot call free() or pcre_free() directly.
2310    
2311    4. Add "make test" as a synonym for "make check". Corrected some comments in
2312    the Makefile.
2313    
2314    5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
2315    Makefile.
2316    
2317    6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
2318    command called pgrep for grepping around the active processes.
2319    
2320    7. Added the beginnings of support for UTF-8 character strings.
2321    
2322    8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
2323    RANLIB to ./ltconfig so that they are used by libtool. I think these are all
2324    the relevant ones. (AR is not passed because ./ltconfig does its own figuring
2325    out for the ar command.)
2326    
2327    
2328    Version 3.2 12-May-00
2329    ---------------------
2330    
2331    This is purely a bug fixing release.
2332    
2333    1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
2334    of ZA. This was just one example of several cases that could provoke this bug,
2335    which was introduced by change 9 of version 2.00. The code for breaking
2336    infinite loops after an iteration that matches an empty string was't working
2337    correctly.
2338    
2339    2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
2340    when matched against abbab (for example). After matching an empty string, it
2341    wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
2342    caused it to match further down the string than it should.
2343    
2344    3. The code contained an inclusion of sys/types.h. It isn't clear why this
2345    was there because it doesn't seem to be needed, and it causes trouble on some
2346    systems, as it is not a Standard C header. It has been removed.
2347    
2348    4. Made 4 silly changes to the source to avoid stupid compiler warnings that
2349    were reported on the Macintosh. The changes were from
2350    
2351      while ((c = *(++ptr)) != 0 && c != '\n');
2352    to
2353      while ((c = *(++ptr)) != 0 && c != '\n') ;
2354    
2355    Totally extraordinary, but if that's what it takes...
2356    
2357    5. PCRE is being used in one environment where neither memmove() nor bcopy() is
2358    available. Added HAVE_BCOPY and an autoconf test for it; if neither
2359    HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
2360    assumes the way PCRE uses memmove() (always moving upwards).
2361    
2362    6. PCRE is being used in one environment where strchr() is not available. There
2363    was only one use in pcre.c, and writing it out to avoid strchr() probably gives
2364    faster code anyway.
2365    
2366    
2367    Version 3.1 09-Feb-00
2368    ---------------------
2369    
2370    The only change in this release is the fixing of some bugs in Makefile.in for
2371    the "install" target:
2372    
2373    (1) It was failing to install pcreposix.h.
2374    
2375    (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
2376    
2377    
2378    Version 3.0 01-Feb-00
2379    ---------------------
2380    
2381    1. Add support for the /+ modifier to perltest (to output $` like it does in
2382    pcretest).
2383    
2384    2. Add support for the /g modifier to perltest.
2385    
2386    3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
2387    matches null strings.
2388    
2389    4. Fix perltest so that it doesn't do unwanted things when fed an empty
2390    pattern. Perl treats empty patterns specially - it reuses the most recent
2391    pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
2392    effect.
2393    
2394    5. The POSIX interface was broken in that it was just handing over the POSIX
2395    captured string vector to pcre_exec(), but (since release 2.00) PCRE has
2396    required a bigger vector, with some working space on the end. This means that
2397    the POSIX wrapper now has to get and free some memory, and copy the results.
2398    
2399    6. Added some simple autoconf support, placing the test data and the
2400    documentation in separate directories, re-organizing some of the
2401    information files, and making it build pcre-config (a GNU standard). Also added
2402    libtool support for building PCRE as a shared library, which is now the
2403    default.
2404    
2405    7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
2406    09 are not valid octal constants. Single digits will be used for minor values
2407    less than 10.
2408    
2409    8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
2410    existing programs that set these in the POSIX interface can use PCRE without
2411    modification.
2412    
2413    9. Added a new function, pcre_fullinfo() with an extensible interface. It can
2414    return all that pcre_info() returns, plus additional data. The pcre_info()
2415    function is retained for compatibility, but is considered to be obsolete.
2416    
2417    10. Added experimental recursion feature (?R) to handle one common case that
2418    Perl 5.6 will be able to do with (?p{...}).
2419    
2420    11. Added support for POSIX character classes like [:alpha:], which Perl is
2421    adopting.
2422    
2423    
2424    Version 2.08 31-Aug-99
2425    ----------------------
2426    
2427    1. When startoffset was not zero and the pattern began with ".*", PCRE was not
2428    trying to match at the startoffset position, but instead was moving forward to
2429    the next newline as if a previous match had failed.
2430    
2431    2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
2432    and could get into a loop if a null string was matched other than at the start
2433    of the subject.
2434    
2435    3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
2436    be distinguished at compile time, and for completeness also added PCRE_DATE.
2437    
2438    5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
2439    in GnuWin32 environments.
2440    
2441    
2442    Version 2.07 29-Jul-99
2443    ----------------------
2444    
2445    1. The documentation is now supplied in plain text form and HTML as well as in
2446    the form of man page sources.
2447    
2448    2. C++ compilers don't like assigning (void *) values to other pointer types.
2449    In particular this affects malloc(). Although there is no problem in Standard
2450    C, I've put in casts to keep C++ compilers happy.
2451    
2452    3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
2453    should be (const char *).
2454    
2455    4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
2456    be useful for non-Unix systems who don't want to bother with the POSIX stuff.
2457    However, I haven't made this a standard facility. The documentation doesn't
2458    mention it, and the Makefile doesn't support it.
2459    
2460    5. The Makefile now contains an "install" target, with editable destinations at
2461    the top of the file. The pcretest program is not installed.
2462    
2463    6. pgrep -V now gives the PCRE version number and date.
2464    
2465    7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
2466    causing the entire string to be ignored, instead of just the last character.
2467    
2468    8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
2469    non-matching string, it can take a very, very long time, even for strings of
2470    quite modest length, because of the nested recursion. PCRE now does better in
2471    some of these cases. It does this by remembering the last required literal
2472    character in the pattern, and pre-searching the subject to ensure it is present
2473    before running the real match. In other words, it applies a heuristic to detect
2474    some types of certain failure quickly, and in the above example, if presented
2475    with a string that has no trailing " it gives "no match" very quickly.
2476    
2477    9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
2478    other alternatives are tried instead.
2479    
2480    
2481    Version 2.06 09-Jun-99
2482    ----------------------
2483    
2484    1. Change pcretest's output for amount of store used to show just the code
2485    space, because the remainder (the data block) varies in size between 32-bit and
2486    64-bit systems.
2487    
2488    2. Added an extra argument to pcre_exec() to supply an offset in the subject to
2489    start matching at. This allows lookbehinds to work when searching for multiple
2490    occurrences in a string.
2491    
2492    3. Added additional options to pcretest for testing multiple occurrences:
2493    
2494       /+   outputs the rest of the string that follows a match
2495       /g   loops for multiple occurrences, using the new startoffset argument
2496       /G   loops for multiple occurrences by passing an incremented pointer
2497    
2498    4. PCRE wasn't doing the "first character" optimization for patterns starting
2499    with \b or \B, though it was doing it for other lookbehind assertions. That is,
2500    it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
2501    the letter 'x'. On long subject strings, this gives a significant speed-up.
2502    
2503    
2504    Version 2.05 21-Apr-99
2505    ----------------------
2506    
2507    1. Changed the type of magic_number from int to long int so that it works
2508    properly on 16-bit systems.
2509    
2510    2. Fixed a bug which caused patterns starting with .* not to work correctly
2511    when the subject string contained newline characters. PCRE was assuming
2512    anchoring for such patterns in all cases, which is not correct because .* will
2513    not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
2514    DOTALL is set at top level; otherwise it knows that patterns starting with .*
2515    must be retried after every newline in the subject.
2516    
2517    
2518    Version 2.04 18-Feb-99
2519    ----------------------
2520    
2521    1. For parenthesized subpatterns with repeats whose minimum was zero, the
2522    computation of the store needed to hold the pattern was incorrect (too large).
2523    If such patterns were nested a few deep, this could multiply and become a real
2524    problem.
2525    
2526    2. Added /M option to pcretest to show the memory requirement of a specific
2527    pattern. Made -m a synonym of -s (which does this globally) for compatibility.
2528    
2529    3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
2530    compiled in such a way that the backtracking after subsequent failure was
2531    pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
2532    ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
2533    
2534    
2535    Version 2.03 02-Feb-99
2536    ----------------------
2537    
2538    1. Fixed typo and small mistake in man page.
2539    
2540    2. Added 4th condition (GPL supersedes if conflict) and created separate
2541    LICENCE file containing the conditions.
2542    
2543    3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
2544    Perl, that is the internal \ allows the delimiter to be included in the
2545    pattern. Locked out the use of \ as a delimiter. If \ immediately follows
2546    the final delimiter, add \ to the end of the pattern (to test the error).
2547    
2548    4. Added the convenience functions for extracting substrings after a successful
2549    match. Updated pcretest to make it able to test these functions.
2550    
2551    
2552    Version 2.02 14-Jan-99
2553    ----------------------
2554    
2555    1. Initialized the working variables associated with each extraction so that
2556    their saving and restoring doesn't refer to uninitialized store.
2557    
2558    2. Put dummy code into study.c in order to trick the optimizer of the IBM C
2559    compiler for OS/2 into generating correct code. Apparently IBM isn't going to
2560    fix the problem.
2561    
2562    3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
2563    calls, and wasn't printing the correct value for compiling calls. Increased the
2564    default value of LOOPREPEAT, and the number of significant figures in the
2565    times.
2566    
2567    4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
2568    
2569    5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
2570    a building problem on Windows NT with a FAT file system.
2571    
2572    
2573    Version 2.01 21-Oct-98
2574    ----------------------
2575    
2576    1. Changed the API for pcre_compile() to allow for the provision of a pointer
2577    to character tables built by pcre_maketables() in the current locale. If NULL
2578    is passed, the default tables are used.
2579    
2580    
2581  Version 2.00 24-Sep-98  Version 2.00 24-Sep-98
2582  ----------------------  ----------------------

Legend:
Removed from v.23  
changed lines
  Added in v.226

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12