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

Diff of /code/trunk/ChangeLog

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

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

Legend:
Removed from v.53  
changed lines
  Added in v.269

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12