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

Diff of /code/trunk/ChangeLog

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

revision 7 by nigel, Sat Feb 24 21:38:09 2007 UTC revision 275 by ph10, Wed Nov 21 15:35:09 2007 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4    Version 7.5 12-Nov-07
5    ---------------------
6    
7    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
8        values in parens when parsing an RE using the C++ wrapper."
9    
10    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
11        Characters greater than 255 were excluded from the class instead of being
12        included.
13    
14    3.  The same bug as (2) above applied to negated POSIX classes such as
15        [:^space:].
16    
17    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
18        defined or documented. It seems to have been a typo for PCRE_STATIC, so
19        I have changed it.
20    
21    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
22        first named subpattern) and a construct such as (?&a) would reference the
23        first named subpattern whose name started with "a" (in other words, the
24        length check was missing). Both these problems are fixed. "Subpattern name
25        expected" is now given for (?&) (a zero-length name), and this patch also
26        makes it give the same error for \k'' (previously it complained that that
27        was a reference to a non-existent subpattern).
28    
29    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
30        this is right because (?- can be followed by option settings as well as by
31        digits. I have, however, made the messages clearer.
32    
33    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
34        than the number used in the conditional) now cause a compile-time error.
35        This is actually not compatible with Perl, which accepts such patterns, but
36        treats the conditional as always being FALSE (as PCRE used to), but it
37        seems to me that giving a diagnostic is better.
38    
39    8.  Change "alphameric" to the more common word "alphanumeric" in comments
40        and messages.
41    
42    9.  Fix two occurrences of "backslash" in comments that should have been
43        "backspace".
44    
45    
46    Version 7.4 21-Sep-07
47    ---------------------
48    
49    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
50        means that a class such as [\s] counted as "explicit reference to CR or
51        LF". That isn't really right - the whole point of the change was to try to
52        help when there was an actual mention of one of the two characters. So now
53        the change happens only if \r or \n (or a literal CR or LF) character is
54        encountered.
55    
56    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
57        of both had grown to the point where there were only 3 bits left.
58        Fortunately, there was spare space in the data structure, and so I have
59        moved the internal flags into a new 16-bit field to free up more option
60        bits.
61    
62    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
63        but did not set the internal JCHANGED flag - either of these is enough to
64        control the way the "get" function works - but the PCRE_INFO_JCHANGED
65        facility is supposed to tell if (?J) was ever used, so now (?J) at the
66        start sets both bits.
67    
68    4.  Added options (at build time, compile time, exec time) to change \R from
69        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
70    
71    5.  doc/pcresyntax.html was missing from the distribution.
72    
73    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
74        compatibility, even though it is no longer used.
75    
76    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
77        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
78        windows.h file is present (where different names are used). [This was
79        reversed later after testing - see 16 below.]
80    
81    8.  Changed all #include <config.h> to #include "config.h". There were also
82        some further <pcre.h> cases that I changed to "pcre.h".
83    
84    9.  When pcregrep was used with the --colour option, it missed the line ending
85        sequence off the lines that it output.
86    
87    10. It was pointed out to me that arrays of string pointers cause lots of
88        relocations when a shared library is dynamically loaded. A technique of
89        using a single long string with a table of offsets can drastically reduce
90        these. I have refactored PCRE in four places to do this. The result is
91        dramatic:
92    
93          Originally:                          290
94          After changing UCP table:            187
95          After changing error message table:   43
96          After changing table of "verbs"       36
97          After changing table of Posix names   22
98    
99        Thanks to the folks working on Gregex for glib for this insight.
100    
101    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
102        unicode-properties was also set.
103    
104    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
105    
106    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
107        checked only for CRLF.
108    
109    14. Added casts to pcretest.c to avoid compiler warnings.
110    
111    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
112    
113    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
114        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
115        entirely. This removes changes made in 7 above.
116    
117    17. The CMake files have been updated, and there is now more information about
118        building with CMake in the NON-UNIX-USE document.
119    
120    
121    Version 7.3 28-Aug-07
122    ---------------------
123    
124     1. In the rejigging of the build system that eventually resulted in 7.1, the
125        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
126        brackets there is not right, since it causes compilers to look for an
127        installed pcre.h, not the version that is in the source that is being
128        compiled (which of course may be different). I have changed it back to:
129    
130          #include "pcre.h"
131    
132        I have a vague recollection that the change was concerned with compiling in
133        different directories, but in the new build system, that is taken care of
134        by the VPATH setting the Makefile.
135    
136     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
137        when the subject happened to end in the byte 0x85 (e.g. if the last
138        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
139        characters but of course it shouldn't be taken as a newline when it is part
140        of another character. The bug was that, for an unlimited repeat of . in
141        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
142        characters when looking for a newline.
143    
144     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
145    
146     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
147        in debug output.
148    
149     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
150        long printing in the pcrecpp unittest when running under MinGW.
151    
152     6. ESC_K was left out of the EBCDIC table.
153    
154     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
155        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
156        limit also applies to "virtual nesting" when a pattern is recursive, and in
157        this case 1000 isn't so big. I have been able to remove this limit at the
158        expense of backing off one optimization in certain circumstances. Normally,
159        when pcre_exec() would call its internal match() function recursively and
160        immediately return the result unconditionally, it uses a "tail recursion"
161        feature to save stack. However, when a subpattern that can match an empty
162        string has an unlimited repetition quantifier, it no longer makes this
163        optimization. That gives it a stack frame in which to save the data for
164        checking that an empty string has been matched. Previously this was taken
165        from the 1000-entry workspace that had been reserved. So now there is no
166        explicit limit, but more stack is used.
167    
168     8. Applied Daniel's patches to solve problems with the import/export magic
169        syntax that is required for Windows, and which was going wrong for the
170        pcreposix and pcrecpp parts of the library. These were overlooked when this
171        problem was solved for the main library.
172    
173     9. There were some crude static tests to avoid integer overflow when computing
174        the size of patterns that contain repeated groups with explicit upper
175        limits. As the maximum quantifier is 65535, the maximum group length was
176        set at 30,000 so that the product of these two numbers did not overflow a
177        32-bit integer. However, it turns out that people want to use groups that
178        are longer than 30,000 bytes (though not repeat them that many times).
179        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
180        made it possible to implement the integer overflow checks in a much more
181        dynamic way, which I have now done. The artificial limitation on group
182        length has been removed - we now have only the limit on the total length of
183        the compiled pattern, which depends on the LINK_SIZE setting.
184    
185    10. Fixed a bug in the documentation for get/copy named substring when
186        duplicate names are permitted. If none of the named substrings are set, the
187        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
188        empty string.
189    
190    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
191        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
192        because the ] is interpreted as the first data character and the
193        terminating ] is not found. PCRE has been made compatible with Perl in this
194        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
195        cause memory overwriting.
196    
197    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
198        string has been matched (to stop an infinite loop). It was not recognizing
199        a conditional subpattern that could match an empty string if that
200        subpattern was within another subpattern. For example, it looped when
201        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
202        condition was not nested. This bug has been fixed.
203    
204    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
205        past the start of the subject in the presence of bytes with the top bit
206        set, for example "\x8aBCD".
207    
208    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
209        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
210    
211    14. Optimized (?!) to (*FAIL).
212    
213    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
214        This restricts code points to be within the range 0 to 0x10FFFF, excluding
215        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
216        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
217        does: it's just the validity check that is more restrictive.
218    
219    16. Inserted checks for integer overflows during escape sequence (backslash)
220        processing, and also fixed erroneous offset values for syntax errors during
221        backslash processing.
222    
223    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
224        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
225    
226    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
227        caused an overrun.
228    
229    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
230        something other than just ASCII characters) inside a group that had an
231        unlimited repeat caused a loop at compile time (while checking to see
232        whether the group could match an empty string).
233    
234    20. Debugging a pattern containing \p or \P could cause a crash. For example,
235        [\P{Any}] did so. (Error in the code for printing property names.)
236    
237    21. An orphan \E inside a character class could cause a crash.
238    
239    22. A repeated capturing bracket such as (A)? could cause a wild memory
240        reference during compilation.
241    
242    23. There are several functions in pcre_compile() that scan along a compiled
243        expression for various reasons (e.g. to see if it's fixed length for look
244        behind). There were bugs in these functions when a repeated \p or \P was
245        present in the pattern. These operators have additional parameters compared
246        with \d, etc, and these were not being taken into account when moving along
247        the compiled data. Specifically:
248    
249        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
250            length.
251    
252        (b) An item such as \pL+ within a repeated group could cause crashes or
253            loops.
254    
255        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
256            "reference to non-existent subpattern" error.
257    
258        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
259    
260    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
261        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
262    
263    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
264    
265    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
266        character were causing crashes (broken optimization).
267    
268    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
269        \p or \P) caused a compile-time loop.
270    
271    28. More problems have arisen in unanchored patterns when CRLF is a valid line
272        break. For example, the unstudied pattern [\r\n]A does not match the string
273        "\r\nA" because change 7.0/46 below moves the current point on by two
274        characters after failing to match at the start. However, the pattern \nA
275        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
276        the same is true. There doesn't seem any very clean way out of this, but
277        what I have chosen to do makes the common cases work: PCRE now takes note
278        of whether there can be an explicit match for \r or \n anywhere in the
279        pattern, and if so, 7.0/46 no longer applies. As part of this change,
280        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
281        pattern has explicit CR or LF references.
282    
283    29. Added (*CR) etc for changing newline setting at start of pattern.
284    
285    
286    Version 7.2 19-Jun-07
287    ---------------------
288    
289     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
290        which is apparently normally available under Windows.
291    
292     2. Re-jig the pcregrep tests with different newline settings in an attempt
293        to make them independent of the local environment's newline setting.
294    
295     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
296    
297     4. Some of the "internals" tests were previously cut out when the link size
298        was not 2, because the output contained actual offsets. The recent new
299        "Z" feature of pcretest means that these can be cut out, making the tests
300        usable with all link sizes.
301    
302     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
303        stack recursion. This gives a massive performance boost under BSD, but just
304        a small improvement under Linux. However, it saves one field in the frame
305        in all cases.
306    
307     6. Added more features from the forthcoming Perl 5.10:
308    
309        (a) (?-n) (where n is a string of digits) is a relative subroutine or
310            recursion call. It refers to the nth most recently opened parentheses.
311    
312        (b) (?+n) is also a relative subroutine call; it refers to the nth next
313            to be opened parentheses.
314    
315        (c) Conditions that refer to capturing parentheses can be specified
316            relatively, for example, (?(-2)... or (?(+3)...
317    
318        (d) \K resets the start of the current match so that everything before
319            is not part of it.
320    
321        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
322    
323        (f) \g{name} is another synonym - part of Perl 5.10's unification of
324            reference syntax.
325    
326        (g) (?| introduces a group in which the numbering of parentheses in each
327            alternative starts with the same number.
328    
329        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
330    
331     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
332        PCRE_INFO_JCHANGED.
333    
334     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
335        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
336        for detecting groups that can match an empty string.
337    
338     9. A pattern with a very large number of alternatives (more than several
339        hundred) was running out of internal workspace during the pre-compile
340        phase, where pcre_compile() figures out how much memory will be needed. A
341        bit of new cunning has reduced the workspace needed for groups with
342        alternatives. The 1000-alternative test pattern now uses 12 bytes of
343        workspace instead of running out of the 4096 that are available.
344    
345    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
346    
347    11. Applied patch from Google to remove an optimization that didn't quite work.
348        The report of the bug said:
349    
350          pcrecpp::RE("a*").FullMatch("aaa") matches, while
351          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
352          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
353    
354    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
355        it matched the wrong number of bytes.
356    
357    
358    Version 7.1 24-Apr-07
359    ---------------------
360    
361     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
362        that is more "standard", making use of automake and other Autotools. There
363        is some re-arrangement of the files and adjustment of comments consequent
364        on this.
365    
366     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
367        for recursive directory scanning broke on some systems because the files
368        are not scanned in any specific order and on different systems the order
369        was different. A call to "sort" has been inserted into RunGrepTest for the
370        approprate test as a short-term fix. In the longer term there may be an
371        alternative.
372    
373     3. I had an email from Eric Raymond about problems translating some of PCRE's
374        man pages to HTML (despite the fact that I distribute HTML pages, some
375        people do their own conversions for various reasons). The problems
376        concerned the use of low-level troff macros .br and .in. I have therefore
377        removed all such uses from the man pages (some were redundant, some could
378        be replaced by .nf/.fi pairs). The 132html script that I use to generate
379        HTML has been updated to handle .nf/.fi and to complain if it encounters
380        .br or .in.
381    
382     4. Updated comments in configure.ac that get placed in config.h.in and also
383        arranged for config.h to be included in the distribution, with the name
384        config.h.generic, for the benefit of those who have to compile without
385        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
386    
387     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
388        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
389        makevp.bat for the current PCRE, using the additional files
390        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
391    
392     6. A Windows user reported a minor discrepancy with test 2, which turned out
393        to be caused by a trailing space on an input line that had got lost in his
394        copy. The trailing space was an accident, so I've just removed it.
395    
396     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
397        that is needed.
398    
399     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
400        as "const" (a) because they are and (b) because it helps the PHP
401        maintainers who have recently made a script to detect big data structures
402        in the php code that should be moved to the .rodata section. I remembered
403        to update Builducptable as well, so it won't revert if ucptable.h is ever
404        re-created.
405    
406     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
407        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
408        order to be able to cut out the UTF-8 tables in the latter when UTF-8
409        support is not required. This saves 1.5-2K of code, which is important in
410        some applications.
411    
412        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
413        so as not to refer to the tables, even though these functions will never be
414        called when UTF-8 support is disabled. Otherwise there are problems with a
415        shared library.
416    
417    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
418    
419        (a) It was defining its arguments as char * instead of void *.
420    
421        (b) It was assuming that all moves were upwards in memory; this was true
422            a long time ago when I wrote it, but is no longer the case.
423    
424        The emulated memove() is provided for those environments that have neither
425        memmove() nor bcopy(). I didn't think anyone used it these days, but that
426        is clearly not the case, as these two bugs were recently reported.
427    
428    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
429        and Detrail to create the HTML documentation, the .txt form of the man
430        pages, and it removes trailing spaces from listed files. It also creates
431        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
432        case, it wraps all the #defines with #ifndefs. This script should be run
433        before "make dist".
434    
435    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
436        with Unicode property support.
437    
438        (a) For a maximizing quantifier, if the two different cases of the
439            character were of different lengths in their UTF-8 codings (there are
440            some cases like this - I found 11), and the matching function had to
441            back up over a mixture of the two cases, it incorrectly assumed they
442            were both the same length.
443    
444        (b) When PCRE was configured to use the heap rather than the stack for
445            recursion during matching, it was not correctly preserving the data for
446            the other case of a UTF-8 character when checking ahead for a match
447            while processing a minimizing repeat. If the check also involved
448            matching a wide character, but failed, corruption could cause an
449            erroneous result when trying to check for a repeat of the original
450            character.
451    
452    13. Some tidying changes to the testing mechanism:
453    
454        (a) The RunTest script now detects the internal link size and whether there
455            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
456            values substituted by "configure". (The RunGrepTest script already did
457            this for UTF-8.) The configure.ac script no longer substitutes the
458            relevant variables.
459    
460        (b) The debugging options /B and /D in pcretest show the compiled bytecode
461            with length and offset values. This means that the output is different
462            for different internal link sizes. Test 2 is skipped for link sizes
463            other than 2 because of this, bypassing the problem. Unfortunately,
464            there was also a test in test 3 (the locale tests) that used /B and
465            failed for link sizes other than 2. Rather than cut the whole test out,
466            I have added a new /Z option to pcretest that replaces the length and
467            offset values with spaces. This is now used to make test 3 independent
468            of link size. (Test 2 will be tidied up later.)
469    
470    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
471        segmentation fault instead of returning the appropriate error message.
472    
473    15. In multiline mode when the newline sequence was set to "any", the pattern
474        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
475        This doesn't seem right; it now treats the CRLF combination as the line
476        ending, and so does not match in that case. It's only a pattern such as ^$
477        that would hit this one: something like ^ABC$ would have failed after \r
478        and then tried again after \r\n.
479    
480    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
481        in an attempt to make files that differ only in their line terminators
482        compare equal. This works on Linux.
483    
484    17. Under certain error circumstances pcregrep might try to free random memory
485        as it exited. This is now fixed, thanks to valgrind.
486    
487    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
488        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
489        was because its rules for how to advance for /g after matching an empty
490        string at the end of a line did not allow for this case. They now check for
491        it specially.
492    
493    20. pcretest is supposed to handle patterns and data of any length, by
494        extending its buffers when necessary. It was getting this wrong when the
495        buffer for a data line had to be extended.
496    
497    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
498        CRLF as a newline sequence.
499    
500    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
501        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
502        I have nevertheless tidied it up.
503    
504    23. Added some casts to kill warnings from HP-UX ia64 compiler.
505    
506    24. Added a man page for pcre-config.
507    
508    
509    Version 7.0 19-Dec-06
510    ---------------------
511    
512     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
513        moving to gcc 4.1.1.
514    
515     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
516        sys/time.h, which is documented as needed for this function. It doesn't
517        seem to matter on Linux, but it showed up on some releases of OS X.
518    
519     3. It seems that there are systems where bytes whose values are greater than
520        127 match isprint() in the "C" locale. The "C" locale should be the
521        default when a C program starts up. In most systems, only ASCII printing
522        characters match isprint(). This difference caused the output from pcretest
523        to vary, making some of the tests fail. I have changed pcretest so that:
524    
525        (a) When it is outputting text in the compiled version of a pattern, bytes
526            other than 32-126 are always shown as hex escapes.
527    
528        (b) When it is outputting text that is a matched part of a subject string,
529            it does the same, unless a different locale has been set for the match
530            (using the /L modifier). In this case, it uses isprint() to decide.
531    
532     4. Fixed a major bug that caused incorrect computation of the amount of memory
533        required for a compiled pattern when options that changed within the
534        pattern affected the logic of the preliminary scan that determines the
535        length. The relevant options are -x, and -i in UTF-8 mode. The result was
536        that the computed length was too small. The symptoms of this bug were
537        either the PCRE error "internal error: code overflow" from pcre_compile(),
538        or a glibc crash with a message such as "pcretest: free(): invalid next
539        size (fast)". Examples of patterns that provoked this bug (shown in
540        pcretest format) are:
541    
542          /(?-x: )/x
543          /(?x)(?-x: \s*#\s*)/
544          /((?i)[\x{c0}])/8
545          /(?i:[\x{c0}])/8
546    
547        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
548        is now done differently.
549    
550     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
551        wrapper classes; (b) implement a new function in the C++ scanner that is
552        more efficient than the old way of doing things because it avoids levels of
553        recursion in the regex matching; (c) add a paragraph to the documentation
554        for the FullMatch() function.
555    
556     6. The escape sequence \n was being treated as whatever was defined as
557        "newline". Not only was this contrary to the documentation, which states
558        that \n is character 10 (hex 0A), but it also went horribly wrong when
559        "newline" was defined as CRLF. This has been fixed.
560    
561     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
562        was being set to -1 for the "end of line" case (supposedly a value that no
563        character can have). Though this value is never used (the check for end of
564        line is "zero bytes in current character"), it caused compiler complaints.
565        I've changed it to 0xffffffff.
566    
567     8. In pcre_version.c, the version string was being built by a sequence of
568        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
569        string (as it is for production releases) called a macro with an empty
570        argument. The C standard says the result of this is undefined. The gcc
571        compiler treats it as an empty string (which was what was wanted) but it is
572        reported that Visual C gives an error. The source has been hacked around to
573        avoid this problem.
574    
575     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
576        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
577        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
578        of them did).
579    
580    10. Originally, pcretest opened its input and output without "b"; then I was
581        told that "b" was needed in some environments, so it was added for release
582        5.0 to both the input and output. (It makes no difference on Unix-like
583        systems.) Later I was told that it is wrong for the input on Windows. I've
584        now abstracted the modes into two macros, to make it easier to fiddle with
585        them, and removed "b" from the input mode under Windows.
586    
587    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
588    
589    12. Added -help and --help to pcretest as an official way of being reminded
590        of the options.
591    
592    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
593        and pcrecpp.cc because they annoy compilers at high warning levels.
594    
595    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
596    
597    15. Fixed an occurrence of == in configure.ac that should have been = (shell
598        scripts are not C programs :-) and which was not noticed because it works
599        on Linux.
600    
601    16. pcretest is supposed to handle any length of pattern and data line (as one
602        line or as a continued sequence of lines) by extending its input buffer if
603        necessary. This feature was broken for very long pattern lines, leading to
604        a string of junk being passed to pcre_compile() if the pattern was longer
605        than about 50K.
606    
607    17. I have done a major re-factoring of the way pcre_compile() computes the
608        amount of memory needed for a compiled pattern. Previously, there was code
609        that made a preliminary scan of the pattern in order to do this. That was
610        OK when PCRE was new, but as the facilities have expanded, it has become
611        harder and harder to keep it in step with the real compile phase, and there
612        have been a number of bugs (see for example, 4 above). I have now found a
613        cunning way of running the real compile function in a "fake" mode that
614        enables it to compute how much memory it would need, while actually only
615        ever using a few hundred bytes of working memory and without too many
616        tests of the mode. This should make future maintenance and development
617        easier. A side effect of this work is that the limit of 200 on the nesting
618        depth of parentheses has been removed (though this was never a serious
619        limitation, I suspect). However, there is a downside: pcre_compile() now
620        runs more slowly than before (30% or more, depending on the pattern). I
621        hope this isn't a big issue. There is no effect on runtime performance.
622    
623    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
624        newline (only possible for the last line of a file) and it was a
625        pattern that set a locale (followed by /Lsomething), pcretest crashed.
626    
627    19. Added additional timing features to pcretest. (1) The -tm option now times
628        matching only, not compiling. (2) Both -t and -tm can be followed, as a
629        separate command line item, by a number that specifies the number of
630        repeats to use when timing. The default is 50000; this gives better
631        precision, but takes uncomfortably long for very large patterns.
632    
633    20. Extended pcre_study() to be more clever in cases where a branch of a
634        subpattern has no definite first character. For example, (a*|b*)[cd] would
635        previously give no result from pcre_study(). Now it recognizes that the
636        first character must be a, b, c, or d.
637    
638    21. There was an incorrect error "recursive call could loop indefinitely" if
639        a subpattern (or the entire pattern) that was being tested for matching an
640        empty string contained only one non-empty item after a nested subpattern.
641        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
642        incorrectly, because the \d was being skipped in the check.
643    
644    22. The pcretest program now has a new pattern option /B and a command line
645        option -b, which is equivalent to adding /B to every pattern. This causes
646        it to show the compiled bytecode, without the additional information that
647        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
648        is the same as /B/I).
649    
650    23. A new optimization is now able automatically to treat some sequences such
651        as a*b as a*+b. More specifically, if something simple (such as a character
652        or a simple class like \d) has an unlimited quantifier, and is followed by
653        something that cannot possibly match the quantified thing, the quantifier
654        is automatically "possessified".
655    
656    24. A recursive reference to a subpattern whose number was greater than 39
657        went wrong under certain circumstances in UTF-8 mode. This bug could also
658        have affected the operation of pcre_study().
659    
660    25. Realized that a little bit of performance could be had by replacing
661        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
662    
663    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
664    
665    27. Possessive quantifiers such as a++ were previously implemented by turning
666        them into atomic groups such as ($>a+). Now they have their own opcodes,
667        which improves performance. This includes the automatically created ones
668        from 23 above.
669    
670    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
671        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
672        the first matched character to be a colon. This applied both to named and
673        numbered groups.
674    
675    29. The ucpinternal.h header file was missing its idempotency #ifdef.
676    
677    30. I was sent a "project" file called libpcre.a.dev which I understand makes
678        building PCRE on Windows easier, so I have included it in the distribution.
679    
680    31. There is now a check in pcretest against a ridiculously large number being
681        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
682        loop, the loop is abandoned.
683    
684    32. Forward references to subpatterns in conditions such as (?(2)...) where
685        subpattern 2 is defined later cause pcre_compile() to search forwards in
686        the pattern for the relevant set of parentheses. This search went wrong
687        when there were unescaped parentheses in a character class, parentheses
688        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
689    
690    33. "Subroutine" calls and backreferences were previously restricted to
691        referencing subpatterns earlier in the regex. This restriction has now
692        been removed.
693    
694    34. Added a number of extra features that are going to be in Perl 5.10. On the
695        whole, these are just syntactic alternatives for features that PCRE had
696        previously implemented using the Python syntax or my own invention. The
697        other formats are all retained for compatibility.
698    
699        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
700            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
701            also .NET compatible.
702    
703        (b) A recursion or subroutine call to a named group can now be defined as
704            (?&name) as well as (?P>name).
705    
706        (c) A backreference to a named group can now be defined as \k<name> or
707            \k'name' as well as (?P=name). The new forms, as well as being in Perl
708            5.10, are also .NET compatible.
709    
710        (d) A conditional reference to a named group can now use the syntax
711            (?(<name>) or (?('name') as well as (?(name).
712    
713        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
714            groups (named and numbered) that are never evaluated inline, but can be
715            called as "subroutines" from elsewhere. In effect, the DEFINE condition
716            is always false. There may be only one alternative in such a group.
717    
718        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
719            as the simple (?(R). The condition is true only if the most recent
720            recursion is that of the given number or name. It does not search out
721            through the entire recursion stack.
722    
723        (g) The escape \gN or \g{N} has been added, where N is a positive or
724            negative number, specifying an absolute or relative reference.
725    
726    35. Tidied to get rid of some further signed/unsigned compiler warnings and
727        some "unreachable code" warnings.
728    
729    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
730        things, this adds five new scripts.
731    
732    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
733        There were also incompatibilities regarding the handling of \Q..\E inside
734        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
735        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
736    
737    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
738        matches an empty string, and forcibly breaks the loop. There were bugs in
739        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
740        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
741        separate and independent bugs (that affected different cases) have been
742        fixed.
743    
744    39. Refactored the code to abolish the use of different opcodes for small
745        capturing bracket numbers. This is a tidy that I avoided doing when I
746        removed the limit on the number of capturing brackets for 3.5 back in 2001.
747        The new approach is not only tidier, it makes it possible to reduce the
748        memory needed to fix the previous bug (38).
749    
750    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
751        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
752        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
753        mode.
754    
755    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
756        report.
757    
758    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
759        copy construction and assignment in the C++ wrapper.
760    
761    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
762        couple of bugs that could have given wrong results in the "--newline=crlf"
763        case.
764    
765    44. Added a number of casts and did some reorganization of signed/unsigned int
766        variables following suggestions from Dair Grant. Also renamed the variable
767        "this" as "item" because it is a C++ keyword.
768    
769    45. Arranged for dftables to add
770    
771          #include "pcre_internal.h"
772    
773        to pcre_chartables.c because without it, gcc 4.x may remove the array
774        definition from the final binary if PCRE is built into a static library and
775        dead code stripping is activated.
776    
777    46. For an unanchored pattern, if a match attempt fails at the start of a
778        newline sequence, and the newline setting is CRLF or ANY, and the next two
779        characters are CRLF, advance by two characters instead of one.
780    
781    
782    Version 6.7 04-Jul-06
783    ---------------------
784    
785     1. In order to handle tests when input lines are enormously long, pcretest has
786        been re-factored so that it automatically extends its buffers when
787        necessary. The code is crude, but this _is_ just a test program. The
788        default size has been increased from 32K to 50K.
789    
790     2. The code in pcre_study() was using the value of the re argument before
791        testing it for NULL. (Of course, in any sensible call of the function, it
792        won't be NULL.)
793    
794     3. The memmove() emulation function in pcre_internal.h, which is used on
795        systems that lack both memmove() and bcopy() - that is, hardly ever -
796        was missing a "static" storage class specifier.
797    
798     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
799        containing an extended class (one that cannot be represented by a bitmap
800        because it contains high-valued characters or Unicode property items, e.g.
801        [\pZ]). Almost always one would set UTF-8 mode when processing such a
802        pattern, but PCRE should not loop if you do not (it no longer does).
803        [Detail: two cases were found: (a) a repeated subpattern containing an
804        extended class; (b) a recursive reference to a subpattern that followed a
805        previous extended class. It wasn't skipping over the extended class
806        correctly when UTF-8 mode was not set.]
807    
808     5. A negated single-character class was not being recognized as fixed-length
809        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
810        compile error "lookbehind assertion is not fixed length".
811    
812     6. The RunPerlTest auxiliary script was showing an unexpected difference
813        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
814        write a Perl script that can interpret lines of an input file either as
815        byte characters or as UTF-8, which is what "perltest" was being required to
816        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
817        can't do is switch easily at run time between having the "use utf8;" pragma
818        or not. In the end, I fudged it by using the RunPerlTest script to insert
819        "use utf8;" explicitly for the UTF-8 tests.
820    
821     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
822        the end of the subject string, contrary to the documentation and to what
823        Perl does. This was true of both matching functions. Now it matches only at
824        the start of the subject and immediately after *internal* newlines.
825    
826     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
827        a pointer to an int instead of a pointer to an unsigned long int. This
828        caused problems on 64-bit systems.
829    
830     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
831        instance of the 'standard' template library not being so standard".
832    
833    10. There was no check on the number of named subpatterns nor the maximum
834        length of a subpattern name. The product of these values is used to compute
835        the size of the memory block for a compiled pattern. By supplying a very
836        long subpattern name and a large number of named subpatterns, the size
837        computation could be caused to overflow. This is now prevented by limiting
838        the length of names to 32 characters, and the number of named subpatterns
839        to 10,000.
840    
841    11. Subpatterns that are repeated with specific counts have to be replicated in
842        the compiled pattern. The size of memory for this was computed from the
843        length of the subpattern and the repeat count. The latter is limited to
844        65535, but there was no limit on the former, meaning that integer overflow
845        could in principle occur. The compiled length of a repeated subpattern is
846        now limited to 30,000 bytes in order to prevent this.
847    
848    12. Added the optional facility to have named substrings with the same name.
849    
850    13. Added the ability to use a named substring as a condition, using the
851        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
852        are numbers (not recommended). Forward references are permitted.
853    
854    14. Added forward references in named backreferences (if you see what I mean).
855    
856    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
857        pattern could run off the end of the subject. For example, the pattern
858        "(?s)(.{1,5})"8 did this with the subject "ab".
859    
860    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
861        PCRE_CASELESS was set when matching characters that were quantified with ?
862        or *.
863    
864    17. A character class other than a single negated character that had a minimum
865        but no maximum quantifier - for example [ab]{6,} - was not handled
866        correctly by pce_dfa_exec(). It would match only one character.
867    
868    18. A valid (though odd) pattern that looked like a POSIX character
869        class but used an invalid character after [ (for example [[,abc,]]) caused
870        pcre_compile() to give the error "Failed: internal error: code overflow" or
871        in some cases to crash with a glibc free() error. This could even happen if
872        the pattern terminated after [[ but there just happened to be a sequence of
873        letters, a binary zero, and a closing ] in the memory that followed.
874    
875    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
876        over the years. Originally (before any Unicode support), just the bottom 8
877        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
878        output from "man perlunicode" includes this:
879    
880          The regular expression compiler produces polymorphic opcodes.  That
881          is, the pattern adapts to the data and automatically switches to
882          the Unicode character scheme when presented with Unicode data--or
883          instead uses a traditional byte scheme when presented with byte
884          data.
885    
886        Sadly, a wide octal escape does not cause a switch, and in a string with
887        no other multibyte characters, these octal escapes are treated as before.
888        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
889        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
890        Unicode string.
891    
892        I have not perpetrated such confusion in PCRE. Up till now, it took just
893        the bottom 8 bits, as in old Perl. I have now made octal escapes with
894        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
895        translate to the appropriate multibyte character.
896    
897    29. Applied some refactoring to reduce the number of warnings from Microsoft
898        and Borland compilers. This has included removing the fudge introduced
899        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
900        a warning about an unused variable.
901    
902    21. PCRE has not included VT (character 0x0b) in the set of whitespace
903        characters since release 4.0, because Perl (from release 5.004) does not.
904        [Or at least, is documented not to: some releases seem to be in conflict
905        with the documentation.] However, when a pattern was studied with
906        pcre_study() and all its branches started with \s, PCRE still included VT
907        as a possible starting character. Of course, this did no harm; it just
908        caused an unnecessary match attempt.
909    
910    22. Removed a now-redundant internal flag bit that recorded the fact that case
911        dependency changed within the pattern. This was once needed for "required
912        byte" processing, but is no longer used. This recovers a now-scarce options
913        bit. Also moved the least significant internal flag bit to the most-
914        significant bit of the word, which was not previously used (hangover from
915        the days when it was an int rather than a uint) to free up another bit for
916        the future.
917    
918    23. Added support for CRLF line endings as well as CR and LF. As well as the
919        default being selectable at build time, it can now be changed at runtime
920        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
921        specify that it is scanning data with non-default line endings.
922    
923    24. Changed the definition of CXXLINK to make it agree with the definition of
924        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
925    
926    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
927        recursions. This makes a big different to stack usage for some patterns.
928    
929    26. If a subpattern containing a named recursion or subroutine reference such
930        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
931        the space required for the compiled pattern went wrong and gave too small a
932        value. Depending on the environment, this could lead to "Failed: internal
933        error: code overflow at offset 49" or "glibc detected double free or
934        corruption" errors.
935    
936    27. Applied patches from Google (a) to support the new newline modes and (b) to
937        advance over multibyte UTF-8 characters in GlobalReplace.
938    
939    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
940        difference for some implementation of PCRE in some Windows version.
941    
942    29. Added some extra testing facilities to pcretest:
943    
944        \q<number>   in a data line sets the "match limit" value
945        \Q<number>   in a data line sets the "match recursion limt" value
946        -S <number>  sets the stack size, where <number> is in megabytes
947    
948        The -S option isn't available for Windows.
949    
950    
951    Version 6.6 06-Feb-06
952    ---------------------
953    
954     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
955        in pcreposix.h. I have copied the definition from pcre.h.
956    
957     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
958        because pcre.h is no longer a built file.
959    
960     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
961        not normally included in the compiled code.
962    
963    
964    Version 6.5 01-Feb-06
965    ---------------------
966    
967     1. When using the partial match feature with pcre_dfa_exec(), it was not
968        anchoring the second and subsequent partial matches at the new starting
969        point. This could lead to incorrect results. For example, with the pattern
970        /1234/, partially matching against "123" and then "a4" gave a match.
971    
972     2. Changes to pcregrep:
973    
974        (a) All non-match returns from pcre_exec() were being treated as failures
975            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
976            error message is output. Some extra information is given for the
977            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
978            probably the only errors that are likely to be caused by users (by
979            specifying a regex that has nested indefinite repeats, for instance).
980            If there are more than 20 of these errors, pcregrep is abandoned.
981    
982        (b) A binary zero was treated as data while matching, but terminated the
983            output line if it was written out. This has been fixed: binary zeroes
984            are now no different to any other data bytes.
985    
986        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
987            used to set a locale for matching. The --locale=xxxx long option has
988            been added (no short equivalent) to specify a locale explicitly on the
989            pcregrep command, overriding the environment variables.
990    
991        (d) When -B was used with -n, some line numbers in the output were one less
992            than they should have been.
993    
994        (e) Added the -o (--only-matching) option.
995    
996        (f) If -A or -C was used with -c (count only), some lines of context were
997            accidentally printed for the final match.
998    
999        (g) Added the -H (--with-filename) option.
1000    
1001        (h) The combination of options -rh failed to suppress file names for files
1002            that were found from directory arguments.
1003    
1004        (i) Added the -D (--devices) and -d (--directories) options.
1005    
1006        (j) Added the -F (--fixed-strings) option.
1007    
1008        (k) Allow "-" to be used as a file name for -f as well as for a data file.
1009    
1010        (l) Added the --colo(u)r option.
1011    
1012        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1013            is not present by default.
1014    
1015     3. A nasty bug was discovered in the handling of recursive patterns, that is,
1016        items such as (?R) or (?1), when the recursion could match a number of
1017        alternatives. If it matched one of the alternatives, but subsequently,
1018        outside the recursion, there was a failure, the code tried to back up into
1019        the recursion. However, because of the way PCRE is implemented, this is not
1020        possible, and the result was an incorrect result from the match.
1021    
1022        In order to prevent this happening, the specification of recursion has
1023        been changed so that all such subpatterns are automatically treated as
1024        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1025    
1026     4. I had overlooked the fact that, in some locales, there are characters for
1027        which isalpha() is true but neither isupper() nor islower() are true. In
1028        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1029        and ordfeminine) are like this. This affected the treatment of \w and \W
1030        when they appeared in character classes, but not when they appeared outside
1031        a character class. The bit map for "word" characters is now created
1032        separately from the results of isalnum() instead of just taking it from the
1033        upper, lower, and digit maps. (Plus the underscore character, of course.)
1034    
1035     5. The above bug also affected the handling of POSIX character classes such as
1036        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1037        permanent tables. Instead, the bit maps for such a class were previously
1038        created as the appropriate unions of the upper, lower, and digit bitmaps.
1039        Now they are created by subtraction from the [[:word:]] class, which has
1040        its own bitmap.
1041    
1042     6. The [[:blank:]] character class matches horizontal, but not vertical space.
1043        It is created by subtracting the vertical space characters (\x09, \x0a,
1044        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1045        subtraction was done in the overall bitmap for a character class, meaning
1046        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1047        be recognized. This bug has been fixed.
1048    
1049     7. Patches from the folks at Google:
1050    
1051          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1052          real life, but is still worth protecting against".
1053    
1054          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1055          regular expressions".
1056    
1057          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1058          have it.
1059    
1060          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1061          "configure" and the latter not, in order to fix a problem somebody had
1062          with compiling the Arg class on HP-UX.
1063    
1064          (e) Improve the error-handling of the C++ wrapper a little bit.
1065    
1066          (f) New tests for checking recursion limiting.
1067    
1068     8. The pcre_memmove() function, which is used only if the environment does not
1069        have a standard memmove() function (and is therefore rarely compiled),
1070        contained two bugs: (a) use of int instead of size_t, and (b) it was not
1071        returning a result (though PCRE never actually uses the result).
1072    
1073     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1074        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1075        returned instead of calling malloc() with an overflowing number that would
1076        most likely cause subsequent chaos.
1077    
1078    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1079    
1080    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1081        with this option is matched, the nmatch and pmatch options of regexec() are
1082        ignored.
1083    
1084    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1085        provided in case anyone wants to the the POSIX interface with UTF-8
1086        strings.
1087    
1088    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1089        C++ linking (needed for some HP-UX environments).
1090    
1091    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1092        (unused parameter) and in the pcre_printint() function (omitted "default"
1093        switch label when the default is to do nothing).
1094    
1095    15. Added some code to make it possible, when PCRE is compiled as a C++
1096        library, to replace subject pointers for pcre_exec() with a smart pointer
1097        class, thus making it possible to process discontinuous strings.
1098    
1099    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1100        much the same function. They were added by different people who were trying
1101        to make PCRE easy to compile on non-Unix systems. It has been suggested
1102        that PCRE_EXPORT be abolished now that there is more automatic apparatus
1103        for compiling on Windows systems. I have therefore replaced it with
1104        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1105        defaults to "extern" for C or "extern C" for C++, which works fine on
1106        Unix-like systems. It is now possible to override the value of PCRE_DATA_
1107        SCOPE with something explicit in config.h. In addition:
1108    
1109        (a) pcreposix.h still had just "extern" instead of either of these macros;
1110            I have replaced it with PCRE_DATA_SCOPE.
1111    
1112        (b) Functions such as _pcre_xclass(), which are internal to the library,
1113            but external in the C sense, all had PCRE_EXPORT in their definitions.
1114            This is apparently wrong for the Windows case, so I have removed it.
1115            (It makes no difference on Unix-like systems.)
1116    
1117    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1118        of recursive calls to match(). This is different to MATCH_LIMIT because
1119        that limits the total number of calls to match(), not all of which increase
1120        the depth of recursion. Limiting the recursion depth limits the amount of
1121        stack (or heap if NO_RECURSE is set) that is used. The default can be set
1122        when PCRE is compiled, and changed at run time. A patch from Google adds
1123        this functionality to the C++ interface.
1124    
1125    18. Changes to the handling of Unicode character properties:
1126    
1127        (a) Updated the table to Unicode 4.1.0.
1128    
1129        (b) Recognize characters that are not in the table as "Cn" (undefined).
1130    
1131        (c) I revised the way the table is implemented to a much improved format
1132            which includes recognition of ranges. It now supports the ranges that
1133            are defined in UnicodeData.txt, and it also amalgamates other
1134            characters into ranges. This has reduced the number of entries in the
1135            table from around 16,000 to around 3,000, thus reducing its size
1136            considerably. I realized I did not need to use a tree structure after
1137            all - a binary chop search is just as efficient. Having reduced the
1138            number of entries, I extended their size from 6 bytes to 8 bytes to
1139            allow for more data.
1140    
1141        (d) Added support for Unicode script names via properties such as \p{Han}.
1142    
1143    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1144        matching that character.
1145    
1146    20. When matching a repeated Unicode property with a minimum greater than zero,
1147        (for example \pL{2,}), PCRE could look past the end of the subject if it
1148        reached it while seeking the minimum number of characters. This could
1149        happen only if some of the characters were more than one byte long, because
1150        there is a check for at least the minimum number of bytes.
1151    
1152    21. Refactored the implementation of \p and \P so as to be more general, to
1153        allow for more different types of property in future. This has changed the
1154        compiled form incompatibly. Anybody with saved compiled patterns that use
1155        \p or \P will have to recompile them.
1156    
1157    22. Added "Any" and "L&" to the supported property types.
1158    
1159    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1160        but give a compile time error if the value is greater than 0xff.
1161    
1162    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1163        accidentally not being installed or uninstalled.
1164    
1165    25. The pcre.h file was built from pcre.h.in, but the only changes that were
1166        made were to insert the current release number. This seemed silly, because
1167        it made things harder for people building PCRE on systems that don't run
1168        "configure". I have turned pcre.h into a distributed file, no longer built
1169        by "configure", with the version identification directly included. There is
1170        no longer a pcre.h.in file.
1171    
1172        However, this change necessitated a change to the pcre-config script as
1173        well. It is built from pcre-config.in, and one of the substitutions was the
1174        release number. I have updated configure.ac so that ./configure now finds
1175        the release number by grepping pcre.h.
1176    
1177    26. Added the ability to run the tests under valgrind.
1178    
1179    
1180    Version 6.4 05-Sep-05
1181    ---------------------
1182    
1183     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1184        "--" to be printed when multiple files were scanned, even when none of the
1185        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1186        consider it to be a bug, and have restored the previous behaviour.
1187    
1188     2. A couple of code tidies to get rid of compiler warnings.
1189    
1190     3. The pcretest program used to cheat by referring to symbols in the library
1191        whose names begin with _pcre_. These are internal symbols that are not
1192        really supposed to be visible externally, and in some environments it is
1193        possible to suppress them. The cheating is now confined to including
1194        certain files from the library's source, which is a bit cleaner.
1195    
1196     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1197        file's purpose clearer.
1198    
1199     5. Reorganized pcre_ucp_findchar().
1200    
1201    
1202    Version 6.3 15-Aug-05
1203    ---------------------
1204    
1205     1. The file libpcre.pc.in did not have general read permission in the tarball.
1206    
1207     2. There were some problems when building without C++ support:
1208    
1209        (a) If C++ support was not built, "make install" and "make test" still
1210            tried to test it.
1211    
1212        (b) There were problems when the value of CXX was explicitly set. Some
1213            changes have been made to try to fix these, and ...
1214    
1215        (c) --disable-cpp can now be used to explicitly disable C++ support.
1216    
1217        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1218            backslash in a target when C++ was disabled. This confuses some
1219            versions of "make", apparently. Using an intermediate variable solves
1220            this. (Same for CPP_LOBJ.)
1221    
1222     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1223        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1224        necessary on certain architectures.
1225    
1226     4. Added a setting of -export-symbols-regex to the link command to remove
1227        those symbols that are exported in the C sense, but actually are local
1228        within the library, and not documented. Their names all begin with
1229        "_pcre_". This is not a perfect job, because (a) we have to except some
1230        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1231        available (and never for static libraries). I have made a note to try to
1232        find a way round (a) in the future.
1233    
1234    
1235    Version 6.2 01-Aug-05
1236    ---------------------
1237    
1238     1. There was no test for integer overflow of quantifier values. A construction
1239        such as {1111111111111111} would give undefined results. What is worse, if
1240        a minimum quantifier for a parenthesized subpattern overflowed and became
1241        negative, the calculation of the memory size went wrong. This could have
1242        led to memory overwriting.
1243    
1244     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1245    
1246     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1247        operating environments where this matters.
1248    
1249     4. Applied Giuseppe Maxia's patch to add additional features for controlling
1250        PCRE options from within the C++ wrapper.
1251    
1252     5. Named capturing subpatterns were not being correctly counted when a pattern
1253        was compiled. This caused two problems: (a) If there were more than 100
1254        such subpatterns, the calculation of the memory needed for the whole
1255        compiled pattern went wrong, leading to an overflow error. (b) Numerical
1256        back references of the form \12, where the number was greater than 9, were
1257        not recognized as back references, even though there were sufficient
1258        previous subpatterns.
1259    
1260     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1261        versions of gcc, e.g. 2.95.4.
1262    
1263    
1264    Version 6.1 21-Jun-05
1265    ---------------------
1266    
1267     1. There was one reference to the variable "posix" in pcretest.c that was not
1268        surrounded by "#if !defined NOPOSIX".
1269    
1270     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1271        the cross-check on the old pcre_info() function, for the benefit of the
1272        cut-down version of PCRE that is currently imported into Exim.
1273    
1274     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1275        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1276        patterns that start (?i) but I don't think that matters. The use of (?i) is
1277        just an example; this all applies to the other options as well.
1278    
1279     4. Since libtool seems to echo the compile commands it is issuing, the output
1280        from "make" can be reduced a bit by putting "@" in front of each libtool
1281        compile command.
1282    
1283     5. Patch from the folks at Google for configure.in to be a bit more thorough
1284        in checking for a suitable C++ installation before trying to compile the
1285        C++ stuff. This should fix a reported problem when a compiler was present,
1286        but no suitable headers.
1287    
1288     6. The man pages all had just "PCRE" as their title. I have changed them to
1289        be the relevant file name. I have also arranged that these names are
1290        retained in the file doc/pcre.txt, which is a concatenation in text format
1291        of all the man pages except the little individual ones for each function.
1292    
1293     7. The NON-UNIX-USE file had not been updated for the different set of source
1294        files that come with release 6. I also added a few comments about the C++
1295        wrapper.
1296    
1297    
1298    Version 6.0 07-Jun-05
1299    ---------------------
1300    
1301     1. Some minor internal re-organization to help with my DFA experiments.
1302    
1303     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1304        didn't matter for the library itself when fully configured, but did matter
1305        when compiling without UCP support, or within Exim, where the ucp files are
1306        not imported.
1307    
1308     3. Refactoring of the library code to split up the various functions into
1309        different source modules. The addition of the new DFA matching code (see
1310        below) to a single monolithic source would have made it really too
1311        unwieldy, quite apart from causing all the code to be include in a
1312        statically linked application, when only some functions are used. This is
1313        relevant even without the DFA addition now that patterns can be compiled in
1314        one application and matched in another.
1315    
1316        The downside of splitting up is that there have to be some external
1317        functions and data tables that are used internally in different modules of
1318        the library but which are not part of the API. These have all had their
1319        names changed to start with "_pcre_" so that they are unlikely to clash
1320        with other external names.
1321    
1322     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1323        a different (DFA) algorithm. Although it is slower than the original
1324        function, it does have some advantages for certain types of matching
1325        problem.
1326    
1327     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1328        including restarting after a partial match.
1329    
1330     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1331        defined when compiling for Windows was sent to me. I have put it into the
1332        code, though I have no means of testing or verifying it.
1333    
1334     7. Added the pcre_refcount() auxiliary function.
1335    
1336     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1337        match before or at the first newline in the subject string. In pcretest,
1338        the /f option on a pattern can be used to set this.
1339    
1340     9. A repeated \w when used in UTF-8 mode with characters greater than 256
1341        would behave wrongly. This has been present in PCRE since release 4.0.
1342    
1343    10. A number of changes to the pcregrep command:
1344    
1345        (a) Refactored how -x works; insert ^(...)$ instead of setting
1346            PCRE_ANCHORED and checking the length, in preparation for adding
1347            something similar for -w.
1348    
1349        (b) Added the -w (match as a word) option.
1350    
1351        (c) Refactored the way lines are read and buffered so as to have more
1352            than one at a time available.
1353    
1354        (d) Implemented a pcregrep test script.
1355    
1356        (e) Added the -M (multiline match) option. This allows patterns to match
1357            over several lines of the subject. The buffering ensures that at least
1358            8K, or the rest of the document (whichever is the shorter) is available
1359            for matching (and similarly the previous 8K for lookbehind assertions).
1360    
1361        (f) Changed the --help output so that it now says
1362    
1363              -w, --word-regex(p)
1364    
1365            instead of two lines, one with "regex" and the other with "regexp"
1366            because that confused at least one person since the short forms are the
1367            same. (This required a bit of code, as the output is generated
1368            automatically from a table. It wasn't just a text change.)
1369    
1370        (g) -- can be used to terminate pcregrep options if the next thing isn't an
1371            option but starts with a hyphen. Could be a pattern or a path name
1372            starting with a hyphen, for instance.
1373    
1374        (h) "-" can be given as a file name to represent stdin.
1375    
1376        (i) When file names are being printed, "(standard input)" is used for
1377            the standard input, for compatibility with GNU grep. Previously
1378            "<stdin>" was used.
1379    
1380        (j) The option --label=xxx can be used to supply a name to be used for
1381            stdin when file names are being printed. There is no short form.
1382    
1383        (k) Re-factored the options decoding logic because we are going to add
1384            two more options that take data. Such options can now be given in four
1385            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1386    
1387        (l) Added the -A, -B, and -C options for requesting that lines of context
1388            around matches be printed.
1389    
1390        (m) Added the -L option to print the names of files that do not contain
1391            any matching lines, that is, the complement of -l.
1392    
1393        (n) The return code is 2 if any file cannot be opened, but pcregrep does
1394            continue to scan other files.
1395    
1396        (o) The -s option was incorrectly implemented. For compatibility with other
1397            greps, it now suppresses the error message for a non-existent or non-
1398            accessible file (but not the return code). There is a new option called
1399            -q that suppresses the output of matching lines, which was what -s was
1400            previously doing.
1401    
1402        (p) Added --include and --exclude options to specify files for inclusion
1403            and exclusion when recursing.
1404    
1405    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1406        Hopefully, it now does.
1407    
1408    12. Missing cast in pcre_study().
1409    
1410    13. Added an "uninstall" target to the makefile.
1411    
1412    14. Replaced "extern" in the function prototypes in Makefile.in with
1413        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1414        world, but is set differently for Windows.
1415    
1416    15. Added a second compiling function called pcre_compile2(). The only
1417        difference is that it has an extra argument, which is a pointer to an
1418        integer error code. When there is a compile-time failure, this is set
1419        non-zero, in addition to the error test pointer being set to point to an
1420        error message. The new argument may be NULL if no error number is required
1421        (but then you may as well call pcre_compile(), which is now just a
1422        wrapper). This facility is provided because some applications need a
1423        numeric error indication, but it has also enabled me to tidy up the way
1424        compile-time errors are handled in the POSIX wrapper.
1425    
1426    16. Added VPATH=.libs to the makefile; this should help when building with one
1427        prefix path and installing with another. (Or so I'm told by someone who
1428        knows more about this stuff than I do.)
1429    
1430    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1431        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1432        match everything, including newlines. This is not POSIX-compatible, but
1433        somebody wanted the feature. From pcretest it can be activated by using
1434        both the P and the s flags.
1435    
1436    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1437    
1438    19. libpcre.pc was being incorrectly installed as executable.
1439    
1440    20. A couple of places in pcretest check for end-of-line by looking for '\n';
1441        it now also looks for '\r' so that it will work unmodified on Windows.
1442    
1443    21. Added Google's contributed C++ wrapper to the distribution.
1444    
1445    22. Added some untidy missing memory free() calls in pcretest, to keep
1446        Electric Fence happy when testing.
1447    
1448    
1449    
1450    Version 5.0 13-Sep-04
1451    ---------------------
1452    
1453     1. Internal change: literal characters are no longer packed up into items
1454        containing multiple characters in a single byte-string. Each character
1455        is now matched using a separate opcode. However, there may be more than one
1456        byte in the character in UTF-8 mode.
1457    
1458     2. The pcre_callout_block structure has two new fields: pattern_position and
1459        next_item_length. These contain the offset in the pattern to the next match
1460        item, and its length, respectively.
1461    
1462     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1463        insertion of callouts before each pattern item. Added the /C option to
1464        pcretest to make use of this.
1465    
1466     4. On the advice of a Windows user, the lines
1467    
1468          #if defined(_WIN32) || defined(WIN32)
1469          _setmode( _fileno( stdout ), 0x8000 );
1470          #endif  /* defined(_WIN32) || defined(WIN32) */
1471    
1472        have been added to the source of pcretest. This apparently does useful
1473        magic in relation to line terminators.
1474    
1475     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1476        for the benefit of those environments where the "b" makes a difference.
1477    
1478     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1479        to know about it. I have put a hack into configure.in that adds in code
1480        to set GCC=yes if CC=icc. This seems to end up at a point in the
1481        generated configure script that is early enough to affect the setting of
1482        compiler options, which is what is needed, but I have no means of testing
1483        whether it really works. (The user who reported this had patched the
1484        generated configure script, which of course I cannot do.)
1485    
1486        LATER: After change 22 below (new libtool files), the configure script
1487        seems to know about icc (and also ecc). Therefore, I have commented out
1488        this hack in configure.in.
1489    
1490     7. Added support for pkg-config (2 patches were sent in).
1491    
1492     8. Negated POSIX character classes that used a combination of internal tables
1493        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1494        [[:^ascii]]. Typically, they would match almost any characters. The other
1495        POSIX classes were not broken in this way.
1496    
1497     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1498        to find the match, as PCRE was deluded into thinking that the match had to
1499        start at the start point or following a newline. The same bug applied to
1500        patterns with negative forward assertions or any backward assertions
1501        preceding ".*" at the start, unless the pattern required a fixed first
1502        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1503    
1504    10. In UTF-8 mode, when moving forwards in the subject after a failed match
1505        starting at the last subject character, bytes beyond the end of the subject
1506        string were read.
1507    
1508    11. Renamed the variable "class" as "classbits" to make life easier for C++
1509        users. (Previously there was a macro definition, but it apparently wasn't
1510        enough.)
1511    
1512    12. Added the new field "tables" to the extra data so that tables can be passed
1513        in at exec time, or the internal tables can be re-selected. This allows
1514        a compiled regex to be saved and re-used at a later time by a different
1515        program that might have everything at different addresses.
1516    
1517    13. Modified the pcre-config script so that, when run on Solaris, it shows a
1518        -R library as well as a -L library.
1519    
1520    14. The debugging options of pcretest (-d on the command line or D on a
1521        pattern) showed incorrect output for anything following an extended class
1522        that contained multibyte characters and which was followed by a quantifier.
1523    
1524    15. Added optional support for general category Unicode character properties
1525        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1526        support. It adds about 90K to the size of the library. The meanings of the
1527        inbuilt class escapes such as \d and \s have NOT been changed.
1528    
1529    16. Updated pcredemo.c to include calls to free() to release the memory for the
1530        compiled pattern.
1531    
1532    17. The generated file chartables.c was being created in the source directory
1533        instead of in the building directory. This caused the build to fail if the
1534        source directory was different from the building directory, and was
1535        read-only.
1536    
1537    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1538        file. No doubt somebody will tell me if they don't make sense... Also added
1539        Dan Mooney's comments about building on OpenVMS.
1540    
1541    19. Added support for partial matching via the PCRE_PARTIAL option for
1542        pcre_exec() and the \P data escape in pcretest.
1543    
1544    20. Extended pcretest with 3 new pattern features:
1545    
1546        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
1547              write the compiled pattern to the file whose name is "rest-of-line".
1548              This is a straight binary dump of the data, with the saved pointer to
1549              the character tables forced to be NULL. The study data, if any, is
1550              written too. After writing, pcretest reads a new pattern.
1551    
1552        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1553              compiled pattern from the given file. There must not be any
1554              occurrences of "<" in the file name (pretty unlikely); if there are,
1555              pcretest will instead treat the initial "<" as a pattern delimiter.
1556              After reading in the pattern, pcretest goes on to read data lines as
1557              usual.
1558    
1559        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1560              and 16-bit fields in a compiled pattern, to simulate a pattern that
1561              was compiled on a host of opposite endianness.
1562    
1563    21. The pcre-exec() function can now cope with patterns that were compiled on
1564        hosts of opposite endianness, with this restriction:
1565    
1566          As for any compiled expression that is saved and used later, the tables
1567          pointer field cannot be preserved; the extra_data field in the arguments
1568          to pcre_exec() should be used to pass in a tables address if a value
1569          other than the default internal tables were used at compile time.
1570    
1571    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1572        now diagnosed as an error. Previously, most of the time, a negative number
1573        would have been treated as zero, but if in addition "ovector" was passed as
1574        NULL, a crash could occur.
1575    
1576    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1577        new versions from the libtool 1.5 distribution (the last one is a copy of
1578        a file called libtool.m4). This seems to have fixed the need to patch
1579        "configure" to support Darwin 1.3 (which I used to do). However, I still
1580        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1581        workstation).
1582    
1583    24. Changed the PCRE licence to be the more standard "BSD" licence.
1584    
1585    
1586    Version 4.5 01-Dec-03
1587    ---------------------
1588    
1589     1. There has been some re-arrangement of the code for the match() function so
1590        that it can be compiled in a version that does not call itself recursively.
1591        Instead, it keeps those local variables that need separate instances for
1592        each "recursion" in a frame on the heap, and gets/frees frames whenever it
1593        needs to "recurse". Keeping track of where control must go is done by means
1594        of setjmp/longjmp. The whole thing is implemented by a set of macros that
1595        hide most of the details from the main code, and operates only if
1596        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1597        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1598        operating.
1599    
1600        To make it easier for callers to provide specially tailored get/free
1601        functions for this usage, two new functions, pcre_stack_malloc, and
1602        pcre_stack_free, are used. They are always called in strict stacking order,
1603        and the size of block requested is always the same.
1604    
1605        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1606        PCRE has been compiled to use the stack or the heap for recursion. The
1607        -C option of pcretest uses this to show which version is compiled.
1608    
1609        A new data escape \S, is added to pcretest; it causes the amounts of store
1610        obtained and freed by both kinds of malloc/free at match time to be added
1611        to the output.
1612    
1613     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1614        what's available on my current Linux desktop machine.
1615    
1616     3. When matching a UTF-8 string, the test for a valid string at the start has
1617        been extended. If start_offset is not zero, PCRE now checks that it points
1618        to a byte that is the start of a UTF-8 character. If not, it returns
1619        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1620        this is necessary because there may be backward assertions in the pattern.
1621        When matching the same subject several times, it may save resources to use
1622        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1623    
1624     4. The code for checking the validity of UTF-8 strings has been tightened so
1625        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1626        containing "overlong sequences".
1627    
1628     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1629        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1630        should have been "&", but it just so happened that all the cases this let
1631        through by mistake were picked up later in the function.
1632    
1633     6. I had used a variable called "isblank" - this is a C99 function, causing
1634        some compilers to warn. To avoid this, I renamed it (as "blankclass").
1635    
1636     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1637        prompting; (b) run "./pcretest /dev/null" at the start of the test script
1638        so the version is shown; (c) stop "make test" echoing "./RunTest".
1639    
1640     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1641    
1642     9. The prototype for memmove() for systems that don't have it was using
1643        size_t, but the inclusion of the header that defines size_t was later. I've
1644        moved the #includes for the C headers earlier to avoid this.
1645    
1646    10. Added some adjustments to the code to make it easier to compiler on certain
1647        special systems:
1648    
1649          (a) Some "const" qualifiers were missing.
1650          (b) Added the macro EXPORT before all exported functions; by default this
1651              is defined to be empty.
1652          (c) Changed the dftables auxiliary program (that builds chartables.c) so
1653              that it reads its output file name as an argument instead of writing
1654              to the standard output and assuming this can be redirected.
1655    
1656    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
1657        class containing characters with values greater than 255, PCRE compilation
1658        went into a loop.
1659    
1660    12. A recursive reference to a subpattern that was within another subpattern
1661        that had a minimum quantifier of zero caused PCRE to crash. For example,
1662        (x(y(?2))z)? provoked this bug with a subject that got as far as the
1663        recursion. If the recursively-called subpattern itself had a zero repeat,
1664        that was OK.
1665    
1666    13. In pcretest, the buffer for reading a data line was set at 30K, but the
1667        buffer into which it was copied (for escape processing) was still set at
1668        1024, so long lines caused crashes.
1669    
1670    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
1671        "internal error: code overflow...". This applied to any character class
1672        that was followed by a possessive quantifier.
1673    
1674    15. Modified the Makefile to add libpcre.la as a prerequisite for
1675        libpcreposix.la because I was told this is needed for a parallel build to
1676        work.
1677    
1678    16. If a pattern that contained .* following optional items at the start was
1679        studied, the wrong optimizing data was generated, leading to matching
1680        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
1681        matching string must start with a or b or c. The correct conclusion for
1682        this pattern is that a match can start with any character.
1683    
1684    
1685    Version 4.4 13-Aug-03
1686    ---------------------
1687    
1688     1. In UTF-8 mode, a character class containing characters with values between
1689        127 and 255 was not handled correctly if the compiled pattern was studied.
1690        In fixing this, I have also improved the studying algorithm for such
1691        classes (slightly).
1692    
1693     2. Three internal functions had redundant arguments passed to them. Removal
1694        might give a very teeny performance improvement.
1695    
1696     3. Documentation bug: the value of the capture_top field in a callout is *one
1697        more than* the number of the hightest numbered captured substring.
1698    
1699     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
1700        in incorrectly linking with a previously installed version. They now link
1701        explicitly with libpcre.la.
1702    
1703     5. configure.in no longer needs to recognize Cygwin specially.
1704    
1705     6. A problem in pcre.in for Windows platforms is fixed.
1706    
1707     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
1708        pcretest, it used to include the size of the study block as part of its
1709        output. Unfortunately, the structure contains a field that has a different
1710        size on different hardware architectures. This meant that the tests that
1711        showed this size failed. As the block is currently always of a fixed size,
1712        this information isn't actually particularly useful in pcretest output, so
1713        I have just removed it.
1714    
1715     8. Three pre-processor statements accidentally did not start in column 1.
1716        Sadly, there are *still* compilers around that complain, even though
1717        standard C has not required this for well over a decade. Sigh.
1718    
1719     9. In pcretest, the code for checking callouts passed small integers in the
1720        callout_data field, which is a void * field. However, some picky compilers
1721        complained about the casts involved for this on 64-bit systems. Now
1722        pcretest passes the address of the small integer instead, which should get
1723        rid of the warnings.
1724    
1725    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
1726        both compile and run time, and gives an error if an invalid UTF-8 sequence
1727        is found. There is a option for disabling this check in cases where the
1728        string is known to be correct and/or the maximum performance is wanted.
1729    
1730    11. In response to a bug report, I changed one line in Makefile.in from
1731    
1732            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
1733        to
1734            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
1735    
1736        to look similar to other lines, but I have no way of telling whether this
1737        is the right thing to do, as I do not use Windows. No doubt I'll get told
1738        if it's wrong...
1739    
1740    
1741    Version 4.3 21-May-03
1742    ---------------------
1743    
1744    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
1745       Makefile.
1746    
1747    2. Some refactoring to improve the quality of the code:
1748    
1749       (i)   The utf8_table... variables are now declared "const".
1750    
1751       (ii)  The code for \cx, which used the "case flipping" table to upper case
1752             lower case letters, now just substracts 32. This is ASCII-specific,
1753             but the whole concept of \cx is ASCII-specific, so it seems
1754             reasonable.
1755    
1756       (iii) PCRE was using its character types table to recognize decimal and
1757             hexadecimal digits in the pattern. This is silly, because it handles
1758             only 0-9, a-f, and A-F, but the character types table is locale-
1759             specific, which means strange things might happen. A private
1760             table is now used for this - though it costs 256 bytes, a table is
1761             much faster than multiple explicit tests. Of course, the standard
1762             character types table is still used for matching digits in subject
1763             strings against \d.
1764    
1765       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
1766             ending in _t are). So I've renamed it as ESC_tee.
1767    
1768    3. The first argument for regexec() in the POSIX wrapper should have been
1769       defined as "const".
1770    
1771    4. Changed pcretest to use malloc() for its buffers so that they can be
1772       Electric Fenced for debugging.
1773    
1774    5. There were several places in the code where, in UTF-8 mode, PCRE would try
1775       to read one or more bytes before the start of the subject string. Often this
1776       had no effect on PCRE's behaviour, but in some circumstances it could
1777       provoke a segmentation fault.
1778    
1779    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
1780       to try to read one or more bytes before the start of the subject string.
1781    
1782    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
1783       UTF-8 support could misbehave in various ways if the subject string
1784       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
1785       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
1786       back over UTF-8 characters.)
1787    
1788    
1789    Version 4.2 14-Apr-03
1790    ---------------------
1791    
1792    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
1793    
1794    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
1795         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
1796         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
1797         [WIN_PREFIX]: new variable, "cyg" for Cygwin
1798         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
1799           and BUILD_EXEEXT
1800         Note: automatic setting of the BUILD variables is not yet working
1801         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
1802           compile-time but not at link-time
1803         [LINK]: use for linking executables only
1804         make different versions for Windows and non-Windows
1805         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
1806           libraries
1807         [LINK_FOR_BUILD]: new variable
1808         [OBJEXT]: use throughout
1809         [EXEEXT]: use throughout
1810         <winshared>: new target
1811         <wininstall>: new target
1812         <dftables.o>: use native compiler
1813         <dftables>: use native linker
1814         <install>: handle Windows platform correctly
1815         <clean>: ditto
1816         <check>: ditto
1817         copy DLL to top builddir before testing
1818    
1819       As part of these changes, -no-undefined was removed again. This was reported
1820       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
1821       in any case.
1822    
1823    3. Some tidies to get rid of compiler warnings:
1824    
1825       . In the match_data structure, match_limit was an unsigned long int, whereas
1826         match_call_count was an int. I've made them both unsigned long ints.
1827    
1828       . In pcretest the fact that a const uschar * doesn't automatically cast to
1829         a void * provoked a warning.
1830    
1831       . Turning on some more compiler warnings threw up some "shadow" variables
1832         and a few more missing casts.
1833    
1834    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
1835       option, a class that contained a single character with a value between 128
1836       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
1837    
1838    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
1839       option, a class that contained several characters, but with at least one
1840       whose value was between 128 and 255 caused PCRE to crash.
1841    
1842    
1843    Version 4.1 12-Mar-03
1844    ---------------------
1845    
1846    1. Compiling with gcc -pedantic found a couple of places where casts were
1847    needed, and a string in dftables.c that was longer than standard compilers are
1848    required to support.
1849    
1850    2. Compiling with Sun's compiler found a few more places where the code could
1851    be tidied up in order to avoid warnings.
1852    
1853    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
1854    first of these names is deprecated in the latest Autoconf in favour of the name
1855    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
1856    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
1857    analogy I have changed it to CFLAGS_FOR_BUILD.
1858    
1859    4. Added -no-undefined to the linking command in the Makefile, because this is
1860    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
1861    linking step for the pcreposix library.
1862    
1863    5. PCRE was failing to diagnose the case of two named groups with the same
1864    name.
1865    
1866    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
1867    literal character that is needed in the subject for a match, and scans along to
1868    ensure that it is present before embarking on the full matching process. This
1869    saves time in cases of nested unlimited repeats that are never going to match.
1870    Problem: the scan can take a lot of time if the subject is very long (e.g.
1871    megabytes), thus penalizing straightforward matches. It is now done only if the
1872    amount of subject to be scanned is less than 1000 bytes.
1873    
1874    7. A lesser problem with the same optimization is that it was recording the
1875    first character of an anchored pattern as "needed", thus provoking a search
1876    right along the subject, even when the first match of the pattern was going to
1877    fail. The "needed" character is now not set for anchored patterns, unless it
1878    follows something in the pattern that is of non-fixed length. Thus, it still
1879    fulfils its original purpose of finding quick non-matches in cases of nested
1880    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
1881    
1882    
1883    Version 4.0 17-Feb-03
1884    ---------------------
1885    
1886    1. If a comment in an extended regex that started immediately after a meta-item
1887    extended to the end of string, PCRE compiled incorrect data. This could lead to
1888    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
1889    
1890    2. Moved to autoconf 2.53 and libtool 1.4.2.
1891    
1892    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
1893    the special perltest8 script is no longer needed - all the tests can be run
1894    from a single perltest script.
1895    
1896    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
1897    by \s. It has now been removed in PCRE. This means it isn't recognized as
1898    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
1899    class [:space:] *does* include VT, thereby creating a mess.
1900    
1901    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
1902    space and tab.
1903    
1904    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
1905    its new features into the main test script, reducing the number of scripts.
1906    
1907    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
1908    were backward compatible, and made the (?i) apply to the whole pattern, as if
1909    /i were given. Now it behaves more logically, and applies the option setting
1910    only to what follows. PCRE has been changed to follow suit. However, if it
1911    finds options settings right at the start of the pattern, it extracts them into
1912    the global options, as before. Thus, they show up in the info data.
1913    
1914    8. Added support for the \Q...\E escape sequence. Characters in between are
1915    treated as literals. This is slightly different from Perl in that $ and @ are
1916    also handled as literals inside the quotes. In Perl, they will cause variable
1917    interpolation. Note the following examples:
1918    
1919        Pattern            PCRE matches      Perl matches
1920    
1921        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
1922        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
1923        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
1924    
1925    For compatibility with Perl, \Q...\E sequences are recognized inside character
1926    classes as well as outside them.
1927    
1928    9. Re-organized 3 code statements in pcretest to avoid "overflow in
1929    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
1930    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
1931    signed/unsigned warnings.
1932    
1933    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
1934    option for pcretest, so I've replaced it by a simple function that does just
1935    that job.
1936    
1937    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
1938    "pcregrep -".
1939    
1940    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
1941    Java package. This provides some syntactic sugar for simple cases of what my
1942    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
1943    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
1944    item, you can use this simplified notation. Note that only makes sense with
1945    greedy quantifiers. Consequently, the use of the possessive quantifier forces
1946    greediness, whatever the setting of the PCRE_UNGREEDY option.
1947    
1948    13. A change of greediness default within a pattern was not taking effect at
1949    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
1950    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
1951    was abstracted outside.
1952    
1953    14. PCRE now supports the \G assertion. It is true when the current matching
1954    position is at the start point of the match. This differs from \A when the
1955    starting offset is non-zero. Used with the /g option of pcretest (or similar
1956    code), it works in the same way as it does for Perl's /g option. If all
1957    alternatives of a regex begin with \G, the expression is anchored to the start
1958    match position, and the "anchored" flag is set in the compiled expression.
1959    
1960    15. Some bugs concerning the handling of certain option changes within patterns
1961    have been fixed. These applied to options other than (?ims). For example,
1962    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
1963    been the other way round. Some of this was related to change 7 above.
1964    
1965    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
1966    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
1967    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
1968    POSIX classes only within a class (e.g. /[[:alpha:]]/).
1969    
1970    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
1971    mode. Unlike ".", it always matches newline, whatever the setting of
1972    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
1973    assertions. Perl allows it, but it doesn't (in general) work because it can't
1974    calculate the length of the lookbehind. At least, that's the case for Perl
1975    5.8.0 - I've been told they are going to document that it doesn't work in
1976    future.
1977    
1978    18. Added an error diagnosis for escapes that PCRE does not support: these are
1979    \L, \l, \N, \P, \p, \U, \u, and \X.
1980    
1981    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
1982    reading past the end of the pattern in cases such as /[abcd/.
1983    
1984    20. PCRE was getting more memory than necessary for patterns with classes that
1985    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
1986    
1987    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
1988    compiling PCRE for use with Virtual Pascal.
1989    
1990    22. Small fix to the Makefile to make it work properly if the build is done
1991    outside the source tree.
1992    
1993    23. Added a new extension: a condition to go with recursion. If a conditional
1994    subpattern starts with (?(R) the "true" branch is used if recursion has
1995    happened, whereas the "false" branch is used only at the top level.
1996    
1997    24. When there was a very long string of literal characters (over 255 bytes
1998    without UTF support, over 250 bytes with UTF support), the computation of how
1999    much memory was required could be incorrect, leading to segfaults or other
2000    strange effects.
2001    
2002    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
2003    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
2004    there was a subsequent back reference to those brackets. This meant that, for
2005    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
2006    possible to check for precisely this case. All we can do is abandon the
2007    optimization if .* occurs inside capturing brackets when there are any back
2008    references whatsoever. (See below for a better fix that came later.)
2009    
2010    26. The handling of the optimization for finding the first character of a
2011    non-anchored pattern, and for finding a character that is required later in the
2012    match were failing in some cases. This didn't break the matching; it just
2013    failed to optimize when it could. The way this is done has been re-implemented.
2014    
2015    27. Fixed typo in error message for invalid (?R item (it said "(?p").
2016    
2017    28. Added a new feature that provides some of the functionality that Perl
2018    provides with (?{...}). The facility is termed a "callout". The way it is done
2019    in PCRE is for the caller to provide an optional function, by setting
2020    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
2021    global variable. By default it is unset, which disables all calling out. To get
2022    the function called, the regex must include (?C) at appropriate points. This
2023    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
2024    This provides a means of identifying different callout points. When PCRE
2025    reaches such a point in the regex, if pcre_callout has been set, the external
2026    function is called. It is provided with data in a structure called
2027    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
2028    matching continues; if it returns a non-zero value, the match at the current
2029    point fails. However, backtracking will occur if possible. [This was changed
2030    later and other features added - see item 49 below.]
2031    
2032    29. pcretest is upgraded to test the callout functionality. It provides a
2033    callout function that displays information. By default, it shows the start of
2034    the match and the current position in the text. There are some new data escapes
2035    to vary what happens:
2036    
2037        \C+         in addition, show current contents of captured substrings
2038        \C-         do not supply a callout function
2039        \C!n        return 1 when callout number n is reached
2040        \C!n!m      return 1 when callout number n is reached for the mth time
2041    
2042    30. If pcregrep was called with the -l option and just a single file name, it
2043    output "<stdin>" if a match was found, instead of the file name.
2044    
2045    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
2046    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
2047    pcre_exec(). This saves a malloc/free per call. The default value of
2048    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
2049    when configuring.
2050    
2051    32. The default maximum size of a compiled pattern is 64K. There have been a
2052    few cases of people hitting this limit. The code now uses macros to handle the
2053    storing of links as offsets within the compiled pattern. It defaults to 2-byte
2054    links, but this can be changed to 3 or 4 bytes by --with-link-size when
2055    configuring. Tests 2 and 5 work only with 2-byte links because they output
2056    debugging information about compiled patterns.
2057    
2058    33. Internal code re-arrangements:
2059    
2060    (a) Moved the debugging function for printing out a compiled regex into
2061        its own source file (printint.c) and used #include to pull it into
2062        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
2063        separate copies.
2064    
2065    (b) Defined the list of op-code names for debugging as a macro in
2066        internal.h so that it is next to the definition of the opcodes.
2067    
2068    (c) Defined a table of op-code lengths for simpler skipping along compiled
2069        code. This is again a macro in internal.h so that it is next to the
2070        definition of the opcodes.
2071    
2072    34. Added support for recursive calls to individual subpatterns, along the
2073    lines of Robin Houston's patch (but implemented somewhat differently).
2074    
2075    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
2076    allow it to read and process whole directories in Win32. This code was
2077    contributed by Lionel Fourquaux; it has not been tested by me.
2078    
2079    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
2080    used to name a group. Names consist of alphanumerics and underscores, and must
2081    be unique. Back references use the syntax (?P=name) and recursive calls use
2082    (?P>name) which is a PCRE extension to the Python extension. Groups still have
2083    numbers. The function pcre_fullinfo() can be used after compilation to extract
2084    a name/number map. There are three relevant calls:
2085    
2086      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
2087      PCRE_INFO_NAMECOUNT            yields the number of entries
2088      PCRE_INFO_NAMETABLE            yields a pointer to the map.
2089    
2090    The map is a vector of fixed-size entries. The size of each entry depends on
2091    the length of the longest name used. The first two bytes of each entry are the
2092    group number, most significant byte first. There follows the corresponding
2093    name, zero terminated. The names are in alphabetical order.
2094    
2095    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
2096    case instead of 255. Making it the same both with and without UTF-8 support
2097    means that the same test output works with both.
2098    
2099    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
2100    calling malloc() with a zero argument.
2101    
2102    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
2103    optimization. I've improved things by keeping a bitmap of backreferences with
2104    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
2105    fact referenced, the optimization can be applied. It is unlikely that a
2106    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
2107    the match to follow \n) will appear inside brackets with a number greater than
2108    31, but if it does, any back reference > 31 suppresses the optimization.
2109    
2110    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
2111    of disabling numbered capturing parentheses. Any opening parenthesis that is
2112    not followed by ? behaves as if it were followed by ?: but named parentheses
2113    can still be used for capturing (and they will acquire numbers in the usual
2114    way).
2115    
2116    41. Redesigned the return codes from the match() function into yes/no/error so
2117    that errors can be passed back from deep inside the nested calls. A malloc
2118    failure while inside a recursive subpattern call now causes the
2119    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
2120    
2121    42. It is now possible to set a limit on the number of times the match()
2122    function is called in a call to pcre_exec(). This facility makes it possible to
2123    limit the amount of recursion and backtracking, though not in a directly
2124    obvious way, because the match() function is used in a number of different
2125    circumstances. The count starts from zero for each position in the subject
2126    string (for non-anchored patterns). The default limit is, for compatibility, a
2127    large number, namely 10 000 000. You can change this in two ways:
2128    
2129    (a) When configuring PCRE before making, you can use --with-match-limit=n
2130        to set a default value for the compiled library.
2131    
2132    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
2133        a different value is set. See 45 below.
2134    
2135    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
2136    
2137    43. Added a new function pcre_config(int, void *) to enable run-time extraction
2138    of things that can be changed at compile time. The first argument specifies
2139    what is wanted and the second points to where the information is to be placed.
2140    The current list of available information is:
2141    
2142      PCRE_CONFIG_UTF8
2143    
2144    The output is an integer that is set to one if UTF-8 support is available;
2145    otherwise it is set to zero.
2146    
2147      PCRE_CONFIG_NEWLINE
2148    
2149    The output is an integer that it set to the value of the code that is used for
2150    newline. It is either LF (10) or CR (13).
2151    
2152      PCRE_CONFIG_LINK_SIZE
2153    
2154    The output is an integer that contains the number of bytes used for internal
2155    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
2156    
2157      PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
2158    
2159    The output is an integer that contains the threshold above which the POSIX
2160    interface uses malloc() for output vectors. See item 31 above.
2161    
2162      PCRE_CONFIG_MATCH_LIMIT
2163    
2164    The output is an unsigned integer that contains the default limit of the number
2165    of match() calls in a pcre_exec() execution. See 42 above.
2166    
2167    44. pcretest has been upgraded by the addition of the -C option. This causes it
2168    to extract all the available output from the new pcre_config() function, and to
2169    output it. The program then exits immediately.
2170    
2171    45. A need has arisen to pass over additional data with calls to pcre_exec() in
2172    order to support additional features. One way would have been to define
2173    pcre_exec2() (for example) with extra arguments, but this would not have been
2174    extensible, and would also have required all calls to the original function to
2175    be mapped to the new one. Instead, I have chosen to extend the mechanism that
2176    is used for passing in "extra" data from pcre_study().
2177    
2178    The pcre_extra structure is now exposed and defined in pcre.h. It currently
2179    contains the following fields:
2180    
2181      flags         a bitmap indicating which of the following fields are set
2182      study_data    opaque data from pcre_study()
2183      match_limit   a way of specifying a limit on match() calls for a specific
2184                      call to pcre_exec()
2185      callout_data  data for callouts (see 49 below)
2186    
2187    The flag bits are also defined in pcre.h, and are
2188    
2189      PCRE_EXTRA_STUDY_DATA
2190      PCRE_EXTRA_MATCH_LIMIT
2191      PCRE_EXTRA_CALLOUT_DATA
2192    
2193    The pcre_study() function now returns one of these new pcre_extra blocks, with
2194    the actual study data pointed to by the study_data field, and the
2195    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
2196    before. That is, this change is entirely upwards-compatible and requires no
2197    change to existing code.
2198    
2199    If you want to pass in additional data to pcre_exec(), you can either place it
2200    in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
2201    block.
2202    
2203    46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
2204    data string contains the escape sequence \M, pcretest calls pcre_exec() several
2205    times with different match limits, until it finds the minimum value needed for
2206    pcre_exec() to complete. The value is then output. This can be instructive; for
2207    most simple matches the number is quite small, but for pathological cases it
2208    gets very large very quickly.
2209    
2210    47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
2211    returns the size of the data block pointed to by the study_data field in a
2212    pcre_extra block, that is, the value that was passed as the argument to
2213    pcre_malloc() when PCRE was getting memory in which to place the information
2214    created by pcre_study(). The fourth argument should point to a size_t variable.
2215    pcretest has been extended so that this information is shown after a successful
2216    pcre_study() call when information about the compiled regex is being displayed.
2217    
2218    48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
2219    because what follows is always an absolute path. (Later: it turns out that this
2220    is more than cosmetic for MinGW, because it doesn't like empty path
2221    components.)
2222    
2223    49. Some changes have been made to the callout feature (see 28 above):
2224    
2225    (i)  A callout function now has three choices for what it returns:
2226    
2227           0  =>  success, carry on matching
2228         > 0  =>  failure at this point, but backtrack if possible
2229         < 0  =>  serious error, return this value from pcre_exec()
2230    
2231         Negative values should normally be chosen from the set of PCRE_ERROR_xxx
2232         values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
2233         "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
2234         use by callout functions. It will never be used by PCRE itself.
2235    
2236    (ii) The pcre_extra structure (see 45 above) has a void * field called
2237         callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
2238         pcre_callout_block structure has a field of the same name. The contents of
2239         the field passed in the pcre_extra structure are passed to the callout
2240         function in the corresponding field in the callout block. This makes it
2241         easier to use the same callout-containing regex from multiple threads. For
2242         testing, the pcretest program has a new data escape
2243    
2244           \C*n        pass the number n (may be negative) as callout_data
2245    
2246         If the callout function in pcretest receives a non-zero value as
2247         callout_data, it returns that value.
2248    
2249    50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
2250    there were some redundant $(CFLAGS) in commands that are now specified as
2251    $(LINK), which already includes $(CFLAGS).
2252    
2253    51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
2254    has been compiled with UTF-8 support *and* pcre_compile() has been compiled
2255    with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
2256    one-byte characters throughout. Note that case-insensitive matching applies
2257    only to characters whose values are less than 256. PCRE doesn't support the
2258    notion of cases for higher-valued characters.
2259    
2260    (i)   A character class whose characters are all within 0-255 is handled as
2261          a bit map, and the map is inverted for negative classes. Previously, a
2262          character > 255 always failed to match such a class; however it should
2263          match if the class was a negative one (e.g. [^ab]). This has been fixed.
2264    
2265    (ii)  A negated character class with a single character < 255 is coded as
2266          "not this character" (OP_NOT). This wasn't working properly when the test
2267          character was multibyte, either singly or repeated.
2268    
2269    (iii) Repeats of multibyte characters are now handled correctly in UTF-8
2270          mode, for example: \x{100}{2,3}.
2271    
2272    (iv)  The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
2273          singly or repeated) now correctly test multibyte characters. However,
2274          PCRE doesn't recognize any characters with values greater than 255 as
2275          digits, spaces, or word characters. Such characters always match \D, \S,
2276          and \W, and never match \d, \s, or \w.
2277    
2278    (v)   Classes may now contain characters and character ranges with values
2279          greater than 255. For example: [ab\x{100}-\x{400}].
2280    
2281    (vi)  pcregrep now has a --utf-8 option (synonym -u) which makes it call
2282          PCRE in UTF-8 mode.
2283    
2284    52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
2285    PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
2286    retained for backwards compatibility. (Note that LASTLITERAL is also a byte
2287    value.)
2288    
2289    53. The single man page has become too large. I have therefore split it up into
2290    a number of separate man pages. These also give rise to individual HTML pages;
2291    these are now put in a separate directory, and there is an index.html page that
2292    lists them all. Some hyperlinking between the pages has been installed.
2293    
2294    54. Added convenience functions for handling named capturing parentheses.
2295    
2296    55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2297    aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2298    true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2299    are faulted.
2300    
2301    56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2302    calling configure. These values are used when compiling the dftables.c program
2303    which is run to generate the source of the default character tables. They
2304    default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2305    you will need to set these values.
2306    
2307    57. Updated the building process for Windows DLL, as provided by Fred Cox.
2308    
2309    
2310    Version 3.9 02-Jan-02
2311    ---------------------
2312    
2313    1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2314    
2315    2. If --disable-static was given, the building process failed when trying to
2316    build pcretest and pcregrep. (For some reason it was using libtool to compile
2317    them, which is not right, as they aren't part of the library.)
2318    
2319    
2320    Version 3.8 18-Dec-01
2321    ---------------------
2322    
2323    1. The experimental UTF-8 code was completely screwed up. It was packing the
2324    bytes in the wrong order. How dumb can you get?
2325    
2326    
2327    Version 3.7 29-Oct-01
2328    ---------------------
2329    
2330    1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2331    This caused pcretest, when used on the test data, to segfault. Unfortunately,
2332    this didn't happen under Solaris 8, where I normally test things.
2333    
2334    2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2335    doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2336    isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2337    this fix an hour or so after the initial 3.7 release.)
2338    
2339    
2340    Version 3.6 23-Oct-01
2341    ---------------------
2342    
2343    1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2344    offsets passed as NULL with zero offset count.
2345    
2346    2. The config.guess and config.sub files had not been updated when I moved to
2347    the latest autoconf.
2348    
2349    
2350    Version 3.5 15-Aug-01
2351    ---------------------
2352    
2353    1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2354    had been forgotten.
2355    
2356    2. By using declared but undefined structures, we can avoid using "void"
2357    definitions in pcre.h while keeping the internal definitions of the structures
2358    private.
2359    
2360    3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2361    user point of view, this means that both static and shared libraries are built
2362    by default, but this can be individually controlled. More of the work of
2363    handling this static/shared cases is now inside libtool instead of PCRE's make
2364    file.
2365    
2366    4. The pcretest utility is now installed along with pcregrep because it is
2367    useful for users (to test regexs) and by doing this, it automatically gets
2368    relinked by libtool. The documentation has been turned into a man page, so
2369    there are now .1, .txt, and .html versions in /doc.
2370    
2371    5. Upgrades to pcregrep:
2372       (i)   Added long-form option names like gnu grep.
2373       (ii)  Added --help to list all options with an explanatory phrase.
2374       (iii) Added -r, --recursive to recurse into sub-directories.
2375       (iv)  Added -f, --file to read patterns from a file.
2376    
2377    6. pcre_exec() was referring to its "code" argument before testing that
2378    argument for NULL (and giving an error if it was NULL).
2379    
2380    7. Upgraded Makefile.in to allow for compiling in a different directory from
2381    the source directory.
2382    
2383    8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2384    options bits, the pointer it was passed was to an int instead of to an unsigned
2385    long int. This mattered only on 64-bit systems.
2386    
2387    9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2388    generated) instead of pcre.in, which it its source. Also made the same change
2389    in several of the .c files.
2390    
2391    10. A new release of gcc defines printf() as a macro, which broke pcretest
2392    because it had an ifdef in the middle of a string argument for printf(). Fixed
2393    by using separate calls to printf().
2394    
2395    11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2396    script, to force use of CR or LF instead of \n in the source. On non-Unix
2397    systems, the value can be set in config.h.
2398    
2399    12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2400    absolute limit. Changed the text of the error message to make this clear, and
2401    likewise updated the man page.
2402    
2403    13. The limit of 99 on the number of capturing subpatterns has been removed.
2404    The new limit is 65535, which I hope will not be a "real" limit.
2405    
2406    
2407    Version 3.4 22-Aug-00
2408    ---------------------
2409    
2410    1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2411    
2412    2. Diagnose condition (?(0) as an error instead of crashing on matching.
2413    
2414    
2415    Version 3.3 01-Aug-00
2416    ---------------------
2417    
2418    1. If an octal character was given, but the value was greater than \377, it
2419    was not getting masked to the least significant bits, as documented. This could
2420    lead to crashes in some systems.
2421    
2422    2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
2423    the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
2424    
2425    3. Added the functions pcre_free_substring() and pcre_free_substring_list().
2426    These just pass their arguments on to (pcre_free)(), but they are provided
2427    because some uses of PCRE bind it to non-C systems that can call its functions,
2428    but cannot call free() or pcre_free() directly.
2429    
2430    4. Add "make test" as a synonym for "make check". Corrected some comments in
2431    the Makefile.
2432    
2433    5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
2434    Makefile.
2435    
2436    6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
2437    command called pgrep for grepping around the active processes.
2438    
2439    7. Added the beginnings of support for UTF-8 character strings.
2440    
2441    8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
2442    RANLIB to ./ltconfig so that they are used by libtool. I think these are all
2443    the relevant ones. (AR is not passed because ./ltconfig does its own figuring
2444    out for the ar command.)
2445    
2446    
2447    Version 3.2 12-May-00
2448    ---------------------
2449    
2450    This is purely a bug fixing release.
2451    
2452    1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
2453    of ZA. This was just one example of several cases that could provoke this bug,
2454    which was introduced by change 9 of version 2.00. The code for breaking
2455    infinite loops after an iteration that matches an empty string was't working
2456    correctly.
2457    
2458    2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
2459    when matched against abbab (for example). After matching an empty string, it
2460    wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
2461    caused it to match further down the string than it should.
2462    
2463    3. The code contained an inclusion of sys/types.h. It isn't clear why this
2464    was there because it doesn't seem to be needed, and it causes trouble on some
2465    systems, as it is not a Standard C header. It has been removed.
2466    
2467    4. Made 4 silly changes to the source to avoid stupid compiler warnings that
2468    were reported on the Macintosh. The changes were from
2469    
2470      while ((c = *(++ptr)) != 0 && c != '\n');
2471    to
2472      while ((c = *(++ptr)) != 0 && c != '\n') ;
2473    
2474    Totally extraordinary, but if that's what it takes...
2475    
2476    5. PCRE is being used in one environment where neither memmove() nor bcopy() is
2477    available. Added HAVE_BCOPY and an autoconf test for it; if neither
2478    HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
2479    assumes the way PCRE uses memmove() (always moving upwards).
2480    
2481    6. PCRE is being used in one environment where strchr() is not available. There
2482    was only one use in pcre.c, and writing it out to avoid strchr() probably gives
2483    faster code anyway.
2484    
2485    
2486    Version 3.1 09-Feb-00
2487    ---------------------
2488    
2489    The only change in this release is the fixing of some bugs in Makefile.in for
2490    the "install" target:
2491    
2492    (1) It was failing to install pcreposix.h.
2493    
2494    (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
2495    
2496    
2497    Version 3.0 01-Feb-00
2498    ---------------------
2499    
2500    1. Add support for the /+ modifier to perltest (to output $` like it does in
2501    pcretest).
2502    
2503    2. Add support for the /g modifier to perltest.
2504    
2505    3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
2506    matches null strings.
2507    
2508    4. Fix perltest so that it doesn't do unwanted things when fed an empty
2509    pattern. Perl treats empty patterns specially - it reuses the most recent
2510    pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
2511    effect.
2512    
2513    5. The POSIX interface was broken in that it was just handing over the POSIX
2514    captured string vector to pcre_exec(), but (since release 2.00) PCRE has
2515    required a bigger vector, with some working space on the end. This means that
2516    the POSIX wrapper now has to get and free some memory, and copy the results.
2517    
2518    6. Added some simple autoconf support, placing the test data and the
2519    documentation in separate directories, re-organizing some of the
2520    information files, and making it build pcre-config (a GNU standard). Also added
2521    libtool support for building PCRE as a shared library, which is now the
2522    default.
2523    
2524    7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
2525    09 are not valid octal constants. Single digits will be used for minor values
2526    less than 10.
2527    
2528    8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
2529    existing programs that set these in the POSIX interface can use PCRE without
2530    modification.
2531    
2532    9. Added a new function, pcre_fullinfo() with an extensible interface. It can
2533    return all that pcre_info() returns, plus additional data. The pcre_info()
2534    function is retained for compatibility, but is considered to be obsolete.
2535    
2536    10. Added experimental recursion feature (?R) to handle one common case that
2537    Perl 5.6 will be able to do with (?p{...}).
2538    
2539    11. Added support for POSIX character classes like [:alpha:], which Perl is
2540    adopting.
2541    
2542    
2543    Version 2.08 31-Aug-99
2544    ----------------------
2545    
2546    1. When startoffset was not zero and the pattern began with ".*", PCRE was not
2547    trying to match at the startoffset position, but instead was moving forward to
2548    the next newline as if a previous match had failed.
2549    
2550    2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
2551    and could get into a loop if a null string was matched other than at the start
2552    of the subject.
2553    
2554    3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
2555    be distinguished at compile time, and for completeness also added PCRE_DATE.
2556    
2557    5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
2558    in GnuWin32 environments.
2559    
2560    
2561    Version 2.07 29-Jul-99
2562    ----------------------
2563    
2564    1. The documentation is now supplied in plain text form and HTML as well as in
2565    the form of man page sources.
2566    
2567    2. C++ compilers don't like assigning (void *) values to other pointer types.
2568    In particular this affects malloc(). Although there is no problem in Standard
2569    C, I've put in casts to keep C++ compilers happy.
2570    
2571    3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
2572    should be (const char *).
2573    
2574    4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
2575    be useful for non-Unix systems who don't want to bother with the POSIX stuff.
2576    However, I haven't made this a standard facility. The documentation doesn't
2577    mention it, and the Makefile doesn't support it.
2578    
2579    5. The Makefile now contains an "install" target, with editable destinations at
2580    the top of the file. The pcretest program is not installed.
2581    
2582    6. pgrep -V now gives the PCRE version number and date.
2583    
2584    7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
2585    causing the entire string to be ignored, instead of just the last character.
2586    
2587    8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
2588    non-matching string, it can take a very, very long time, even for strings of
2589    quite modest length, because of the nested recursion. PCRE now does better in
2590    some of these cases. It does this by remembering the last required literal
2591    character in the pattern, and pre-searching the subject to ensure it is present
2592    before running the real match. In other words, it applies a heuristic to detect
2593    some types of certain failure quickly, and in the above example, if presented
2594    with a string that has no trailing " it gives "no match" very quickly.
2595    
2596    9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
2597    other alternatives are tried instead.
2598    
2599    
2600    Version 2.06 09-Jun-99
2601    ----------------------
2602    
2603    1. Change pcretest's output for amount of store used to show just the code
2604    space, because the remainder (the data block) varies in size between 32-bit and
2605    64-bit systems.
2606    
2607    2. Added an extra argument to pcre_exec() to supply an offset in the subject to
2608    start matching at. This allows lookbehinds to work when searching for multiple
2609    occurrences in a string.
2610    
2611    3. Added additional options to pcretest for testing multiple occurrences:
2612    
2613       /+   outputs the rest of the string that follows a match
2614       /g   loops for multiple occurrences, using the new startoffset argument
2615       /G   loops for multiple occurrences by passing an incremented pointer
2616    
2617    4. PCRE wasn't doing the "first character" optimization for patterns starting
2618    with \b or \B, though it was doing it for other lookbehind assertions. That is,
2619    it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
2620    the letter 'x'. On long subject strings, this gives a significant speed-up.
2621    
2622    
2623    Version 2.05 21-Apr-99
2624    ----------------------
2625    
2626    1. Changed the type of magic_number from int to long int so that it works
2627    properly on 16-bit systems.
2628    
2629    2. Fixed a bug which caused patterns starting with .* not to work correctly
2630    when the subject string contained newline characters. PCRE was assuming
2631    anchoring for such patterns in all cases, which is not correct because .* will
2632    not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
2633    DOTALL is set at top level; otherwise it knows that patterns starting with .*
2634    must be retried after every newline in the subject.
2635    
2636    
2637    Version 2.04 18-Feb-99
2638    ----------------------
2639    
2640    1. For parenthesized subpatterns with repeats whose minimum was zero, the
2641    computation of the store needed to hold the pattern was incorrect (too large).
2642    If such patterns were nested a few deep, this could multiply and become a real
2643    problem.
2644    
2645    2. Added /M option to pcretest to show the memory requirement of a specific
2646    pattern. Made -m a synonym of -s (which does this globally) for compatibility.
2647    
2648    3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
2649    compiled in such a way that the backtracking after subsequent failure was
2650    pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
2651    ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
2652    
2653    
2654    Version 2.03 02-Feb-99
2655    ----------------------
2656    
2657    1. Fixed typo and small mistake in man page.
2658    
2659    2. Added 4th condition (GPL supersedes if conflict) and created separate
2660    LICENCE file containing the conditions.
2661    
2662    3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
2663    Perl, that is the internal \ allows the delimiter to be included in the
2664    pattern. Locked out the use of \ as a delimiter. If \ immediately follows
2665    the final delimiter, add \ to the end of the pattern (to test the error).
2666    
2667    4. Added the convenience functions for extracting substrings after a successful
2668    match. Updated pcretest to make it able to test these functions.
2669    
2670    
2671    Version 2.02 14-Jan-99
2672    ----------------------
2673    
2674    1. Initialized the working variables associated with each extraction so that
2675    their saving and restoring doesn't refer to uninitialized store.
2676    
2677    2. Put dummy code into study.c in order to trick the optimizer of the IBM C
2678    compiler for OS/2 into generating correct code. Apparently IBM isn't going to
2679    fix the problem.
2680    
2681    3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
2682    calls, and wasn't printing the correct value for compiling calls. Increased the
2683    default value of LOOPREPEAT, and the number of significant figures in the
2684    times.
2685    
2686    4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
2687    
2688    5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
2689    a building problem on Windows NT with a FAT file system.
2690    
2691    
2692    Version 2.01 21-Oct-98
2693    ----------------------
2694    
2695    1. Changed the API for pcre_compile() to allow for the provision of a pointer
2696    to character tables built by pcre_maketables() in the current locale. If NULL
2697    is passed, the default tables are used.
2698    
2699    
2700    Version 2.00 24-Sep-98
2701    ----------------------
2702    
2703    1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
2704    it any more.
2705    
2706    2. Allow quantification of (?>) groups, and make it work correctly.
2707    
2708    3. The first character computation wasn't working for (?>) groups.
2709    
2710    4. Correct the implementation of \Z (it is permitted to match on the \n at the
2711    end of the subject) and add 5.005's \z, which really does match only at the
2712    very end of the subject.
2713    
2714    5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
2715    
2716    6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
2717    DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
2718    localized options. All options to pcre_study() were also removed.
2719    
2720    7. Add other new features from 5.005:
2721    
2722       $(?<=           positive lookbehind
2723       $(?<!           negative lookbehind
2724       (?imsx-imsx)    added the unsetting capability
2725                       such a setting is global if at outer level; local otherwise
2726       (?imsx-imsx:)   non-capturing groups with option setting
2727       (?(cond)re|re)  conditional pattern matching
2728    
2729       A backreference to itself in a repeated group matches the previous
2730       captured string.
2731    
2732    8. General tidying up of studying (both automatic and via "study")
2733    consequential on the addition of new assertions.
2734    
2735    9. As in 5.005, unlimited repeated groups that could match an empty substring
2736    are no longer faulted at compile time. Instead, the loop is forcibly broken at
2737    runtime if any iteration does actually match an empty substring.
2738    
2739    10. Include the RunTest script in the distribution.
2740    
2741    11. Added tests from the Perl 5.005_02 distribution. This showed up a few
2742    discrepancies, some of which were old and were also with respect to 5.004. They
2743    have now been fixed.
2744    
2745    
2746    Version 1.09 28-Apr-98
2747    ----------------------
2748    
2749    1. A negated single character class followed by a quantifier with a minimum
2750    value of one (e.g.  [^x]{1,6}  ) was not compiled correctly. This could lead to
2751    program crashes, or just wrong answers. This did not apply to negated classes
2752    containing more than one character, or to minima other than one.
2753    
2754    
2755    Version 1.08 27-Mar-98
2756    ----------------------
2757    
2758    1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
2759    
2760    2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
2761    latter must appear before anything that relies on it in the pattern.
2762    
2763    
2764    Version 1.07 16-Feb-98
2765    ----------------------
2766    
2767    1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
2768    repeat of a potentially empty string).
2769    
2770    
2771    Version 1.06 23-Jan-98
2772    ----------------------
2773    
2774    1. Added Markus Oberhumer's little patches for C++.
2775    
2776    2. Literal strings longer than 255 characters were broken.
2777    
2778    
2779    Version 1.05 23-Dec-97
2780    ----------------------
2781    
2782    1. Negated character classes containing more than one character were failing if
2783    PCRE_CASELESS was set at run time.
2784    
2785    
2786    Version 1.04 19-Dec-97
2787    ----------------------
2788    
2789    1. Corrected the man page, where some "const" qualifiers had been omitted.
2790    
2791    2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
2792    input syntax.
2793    
2794    3. Fixed memory leak which occurred when a regex with back references was
2795    matched with an offsets vector that wasn't big enough. The temporary memory
2796    that is used in this case wasn't being freed if the match failed.
2797    
2798    4. Tidied pcretest to ensure it frees memory that it gets.
2799    
2800    5. Temporary memory was being obtained in the case where the passed offsets
2801    vector was exactly big enough.
2802    
2803    6. Corrected definition of offsetof() from change 5 below.
2804    
2805    7. I had screwed up change 6 below and broken the rules for the use of
2806    setjmp(). Now fixed.
2807    
2808    
2809    Version 1.03 18-Dec-97
2810    ----------------------
2811    
2812    1. A erroneous regex with a missing opening parenthesis was correctly
2813    diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
2814    on some systems.
2815    
2816    2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
2817    it was reported that one broken compiler failed on the former because "code" is
2818    also an independent variable.
2819    
2820    3. The erroneous regex a[]b caused an array overrun reference.
2821    
2822    4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
2823    fail on data ending with that character. (It was going on too far, and checking
2824    the next character, typically a binary zero.) This was specific to the
2825    optimized code for single-character negative classes.
2826    
2827    5. Added a contributed patch from the TIN world which does the following:
2828    
2829      + Add an undef for memmove, in case the the system defines a macro for it.
2830    
2831      + Add a definition of offsetof(), in case there isn't one. (I don't know
2832        the reason behind this - offsetof() is part of the ANSI standard - but
2833        it does no harm).
2834    
2835      + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
2836        most of the places where whitespace preceded '#'. I have given up and
2837        allowed the remaining 2 cases to be at the margin.
2838    
2839      + Rename some variables in pcre to eliminate shadowing. This seems very
2840        pedantic, but does no harm, of course.
2841    
2842    6. Moved the call to setjmp() into its own function, to get rid of warnings
2843    from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
2844    
2845    7. Constructs such as \d{8,} were compiling into the equivalent of
2846    \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
2847    outcome, but in this particular case used more store than had been allocated,
2848    which caused the bug to be discovered because it threw up an internal error.
2849    
2850    8. The debugging code in both pcre and pcretest for outputting the compiled
2851    form of a regex was going wrong in the case of back references followed by
2852    curly-bracketed repeats.
2853    
2854    
2855  Version 1.02 12-Dec-97  Version 1.02 12-Dec-97
2856  ----------------------  ----------------------

Legend:
Removed from v.7  
changed lines
  Added in v.275

  ViewVC Help
Powered by ViewVC 1.1.5