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

Diff of /code/trunk/ChangeLog

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

revision 67 by nigel, Sat Feb 24 21:40:13 2007 UTC revision 734 by ph10, Wed Oct 12 16:35:29 2011 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4    Version 8.20 10-Oct-2011
5    ------------------------
6    
7    1.  Change 37 of 8.13 broke patterns like [:a]...[b:] because it thought it had
8        a POSIX class. After further experiments with Perl, which convinced me that
9        Perl has bugs and confusions, a closing square bracket is no longer allowed
10        in a POSIX name. This bug also affected patterns with classes that started
11        with full stops.
12    
13    2.  If a pattern such as /(a)b|ac/ is matched against "ac", there is no
14        captured substring, but while checking the failing first alternative,
15        substring 1 is temporarily captured. If the output vector supplied to
16        pcre_exec() was not big enough for this capture, the yield of the function
17        was still zero ("insufficient space for captured substrings"). This cannot
18        be totally fixed without adding another stack variable, which seems a lot
19        of expense for a edge case. However, I have improved the situation in cases
20        such as /(a)(b)x|abc/ matched against "abc", where the return code
21        indicates that fewer than the maximum number of slots in the ovector have
22        been set.
23    
24    3.  Related to (2) above: when there are more back references in a pattern than
25        slots in the output vector, pcre_exec() uses temporary memory during
26        matching, and copies in the captures as far as possible afterwards. It was
27        using the entire output vector, but this conflicts with the specification
28        that only 2/3 is used for passing back captured substrings. Now it uses
29        only the first 2/3, for compatibility. This is, of course, another edge
30        case.
31    
32    4.  Zoltan Herczeg's just-in-time compiler support has been integrated into the
33        main code base, and can be used by building with --enable-jit. When this is
34        done, pcregrep automatically uses it unless --disable-pcregrep-jit or the
35        runtime --no-jit option is given.
36    
37    5.  When the number of matches in a pcre_dfa_exec() run exactly filled the
38        ovector, the return from the function was zero, implying that there were
39        other matches that did not fit. The correct "exactly full" value is now
40        returned.
41    
42    6.  If a subpattern that was called recursively or as a subroutine contained
43        (*PRUNE) or any other control that caused it to give a non-standard return,
44        invalid errors such as "Error -26 (nested recursion at the same subject
45        position)" or even infinite loops could occur.
46    
47    7.  If a pattern such as /a(*SKIP)c|b(*ACCEPT)|/ was studied, it stopped
48        computing the minimum length on reaching *ACCEPT, and so ended up with the
49        wrong value of 1 rather than 0. Further investigation indicates that
50        computing a minimum subject length in the presence of *ACCEPT is difficult
51        (think back references, subroutine calls), and so I have changed the code
52        so that no minimum is registered for a pattern that contains *ACCEPT.
53    
54    8.  If (*THEN) was present in the first (true) branch of a conditional group,
55        it was not handled as intended. [But see 16 below.]
56    
57    9.  Replaced RunTest.bat and CMakeLists.txt with improved versions provided by
58        Sheri Pierce.
59    
60    10. A pathological pattern such as /(*ACCEPT)a/ was miscompiled, thinking that
61        the first byte in a match must be "a".
62    
63    11. Change 17 for 8.13 increased the recursion depth for patterns like
64        /a(?:.)*?a/ drastically. I've improved things by remembering whether a
65        pattern contains any instances of (*THEN). If it does not, the old
66        optimizations are restored. It would be nice to do this on a per-group
67        basis, but at the moment that is not feasible.
68    
69    12. In some environments, the output of pcretest -C is CRLF terminated. This
70        broke RunTest's code that checks for the link size. A single white space
71        after the value is now allowed for.
72    
73    13. RunTest now checks for the "fr" locale as well as for "fr_FR" and "french".
74        For "fr", it uses the Windows-specific input and output files.
75    
76    14. If (*THEN) appeared in a group that was called recursively or as a
77        subroutine, it did not work as intended. [But see next item.]
78    
79    15. Consider the pattern /A (B(*THEN)C) | D/ where A, B, C, and D are complex
80        pattern fragments (but not containing any | characters). If A and B are
81        matched, but there is a failure in C so that it backtracks to (*THEN), PCRE
82        was behaving differently to Perl. PCRE backtracked into A, but Perl goes to
83        D. In other words, Perl considers parentheses that do not contain any |
84        characters to be part of a surrounding alternative, whereas PCRE was
85        treading (B(*THEN)C) the same as (B(*THEN)C|(*FAIL)) -- which Perl handles
86        differently. PCRE now behaves in the same way as Perl, except in the case
87        of subroutine/recursion calls such as (?1) which have in any case always
88        been different (but PCRE had them first :-).
89    
90    16. Related to 15 above: Perl does not treat the | in a conditional group as
91        creating alternatives. Such a group is treated in the same way as an
92        ordinary group without any | characters when processing (*THEN). PCRE has
93        been changed to match Perl's behaviour.
94    
95    17. If a user had set PCREGREP_COLO(U)R to something other than 1:31, the
96        RunGrepTest script failed.
97    
98    18. Change 22 for version 13 caused atomic groups to use more stack. This is
99        inevitable for groups that contain captures, but it can lead to a lot of
100        stack use in large patterns. The old behaviour has been restored for atomic
101        groups that do not contain any capturing parentheses.
102    
103    19. If the PCRE_NO_START_OPTIMIZE option was set for pcre_compile(), it did not
104        suppress the check for a minimum subject length at run time. (If it was
105        given to pcre_exec() or pcre_dfa_exec() it did work.)
106    
107    20. Fixed an ASCII-dependent infelicity in pcretest that would have made it
108        fail to work when decoding hex characters in data strings in EBCDIC
109        environments.
110    
111    
112    Version 8.13 16-Aug-2011
113    ------------------------
114    
115    1.  The Unicode data tables have been updated to Unicode 6.0.0.
116    
117    2.  Two minor typos in pcre_internal.h have been fixed.
118    
119    3.  Added #include <string.h> to pcre_scanner_unittest.cc, pcrecpp.cc, and
120        pcrecpp_unittest.cc. They are needed for strcmp(), memset(), and strchr()
121        in some environments (e.g. Solaris 10/SPARC using Sun Studio 12U2).
122    
123    4.  There were a number of related bugs in the code for matching backrefences
124        caselessly in UTF-8 mode when codes for the characters concerned were
125        different numbers of bytes. For example, U+023A and U+2C65 are an upper
126        and lower case pair, using 2 and 3 bytes, respectively. The main bugs were:
127        (a) A reference to 3 copies of a 2-byte code matched only 2 of a 3-byte
128        code. (b) A reference to 2 copies of a 3-byte code would not match 2 of a
129        2-byte code at the end of the subject (it thought there wasn't enough data
130        left).
131    
132    5.  Comprehensive information about what went wrong is now returned by
133        pcre_exec() and pcre_dfa_exec() when the UTF-8 string check fails, as long
134        as the output vector has at least 2 elements. The offset of the start of
135        the failing character and a reason code are placed in the vector.
136    
137    6.  When the UTF-8 string check fails for pcre_compile(), the offset that is
138        now returned is for the first byte of the failing character, instead of the
139        last byte inspected. This is an incompatible change, but I hope it is small
140        enough not to be a problem. It makes the returned offset consistent with
141        pcre_exec() and pcre_dfa_exec().
142    
143    7.  pcretest now gives a text phrase as well as the error number when
144        pcre_exec() or pcre_dfa_exec() fails; if the error is a UTF-8 check
145        failure, the offset and reason code are output.
146    
147    8.  When \R was used with a maximizing quantifier it failed to skip backwards
148        over a \r\n pair if the subsequent match failed. Instead, it just skipped
149        back over a single character (\n). This seems wrong (because it treated the
150        two characters as a single entity when going forwards), conflicts with the
151        documentation that \R is equivalent to (?>\r\n|\n|...etc), and makes the
152        behaviour of \R* different to (\R)*, which also seems wrong. The behaviour
153        has been changed.
154    
155    9.  Some internal refactoring has changed the processing so that the handling
156        of the PCRE_CASELESS and PCRE_MULTILINE options is done entirely at compile
157        time (the PCRE_DOTALL option was changed this way some time ago: version
158        7.7 change 16). This has made it possible to abolish the OP_OPT op code,
159        which was always a bit of a fudge. It also means that there is one less
160        argument for the match() function, which reduces its stack requirements
161        slightly. This change also fixes an incompatibility with Perl: the pattern
162        (?i:([^b]))(?1) should not match "ab", but previously PCRE gave a match.
163    
164    10. More internal refactoring has drastically reduced the number of recursive
165        calls to match() for possessively repeated groups such as (abc)++ when
166        using pcre_exec().
167    
168    11. While implementing 10, a number of bugs in the handling of groups were
169        discovered and fixed:
170    
171        (?<=(a)+) was not diagnosed as invalid (non-fixed-length lookbehind).
172        (a|)*(?1) gave a compile-time internal error.
173        ((a|)+)+  did not notice that the outer group could match an empty string.
174        (^a|^)+   was not marked as anchored.
175        (.*a|.*)+ was not marked as matching at start or after a newline.
176    
177    12. Yet more internal refactoring has removed another argument from the match()
178        function. Special calls to this function are now indicated by setting a
179        value in a variable in the "match data" data block.
180    
181    13. Be more explicit in pcre_study() instead of relying on "default" for
182        opcodes that mean there is no starting character; this means that when new
183        ones are added and accidentally left out of pcre_study(), testing should
184        pick them up.
185    
186    14. The -s option of pcretest has been documented for ages as being an old
187        synonym of -m (show memory usage). I have changed it to mean "force study
188        for every regex", that is, assume /S for every regex. This is similar to -i
189        and -d etc. It's slightly incompatible, but I'm hoping nobody is still
190        using it. It makes it easier to run collections of tests with and without
191        study enabled, and thereby test pcre_study() more easily. All the standard
192        tests are now run with and without -s (but some patterns can be marked as
193        "never study" - see 20 below).
194    
195    15. When (*ACCEPT) was used in a subpattern that was called recursively, the
196        restoration of the capturing data to the outer values was not happening
197        correctly.
198    
199    16. If a recursively called subpattern ended with (*ACCEPT) and matched an
200        empty string, and PCRE_NOTEMPTY was set, pcre_exec() thought the whole
201        pattern had matched an empty string, and so incorrectly returned a no
202        match.
203    
204    17. There was optimizing code for the last branch of non-capturing parentheses,
205        and also for the obeyed branch of a conditional subexpression, which used
206        tail recursion to cut down on stack usage. Unfortunately, now that there is
207        the possibility of (*THEN) occurring in these branches, tail recursion is
208        no longer possible because the return has to be checked for (*THEN). These
209        two optimizations have therefore been removed. [But see 8.20/11 above.]
210    
211    18. If a pattern containing \R was studied, it was assumed that \R always
212        matched two bytes, thus causing the minimum subject length to be
213        incorrectly computed because \R can also match just one byte.
214    
215    19. If a pattern containing (*ACCEPT) was studied, the minimum subject length
216        was incorrectly computed.
217    
218    20. If /S is present twice on a test pattern in pcretest input, it now
219        *disables* studying, thereby overriding the use of -s on the command line
220        (see 14 above). This is necessary for one or two tests to keep the output
221        identical in both cases.
222    
223    21. When (*ACCEPT) was used in an assertion that matched an empty string and
224        PCRE_NOTEMPTY was set, PCRE applied the non-empty test to the assertion.
225    
226    22. When an atomic group that contained a capturing parenthesis was
227        successfully matched, but the branch in which it appeared failed, the
228        capturing was not being forgotten if a higher numbered group was later
229        captured. For example, /(?>(a))b|(a)c/ when matching "ac" set capturing
230        group 1 to "a", when in fact it should be unset. This applied to multi-
231        branched capturing and non-capturing groups, repeated or not, and also to
232        positive assertions (capturing in negative assertions does not happen
233        in PCRE) and also to nested atomic groups.
234    
235    23. Add the ++ qualifier feature to pcretest, to show the remainder of the
236        subject after a captured substring, to make it easier to tell which of a
237        number of identical substrings has been captured.
238    
239    24. The way atomic groups are processed by pcre_exec() has been changed so that
240        if they are repeated, backtracking one repetition now resets captured
241        values correctly. For example, if ((?>(a+)b)+aabab) is matched against
242        "aaaabaaabaabab" the value of captured group 2 is now correctly recorded as
243        "aaa". Previously, it would have been "a". As part of this code
244        refactoring, the way recursive calls are handled has also been changed.
245    
246    25. If an assertion condition captured any substrings, they were not passed
247        back unless some other capturing happened later. For example, if
248        (?(?=(a))a) was matched against "a", no capturing was returned.
249    
250    26. When studying a pattern that contained subroutine calls or assertions,
251        the code for finding the minimum length of a possible match was handling
252        direct recursions such as (xxx(?1)|yyy) but not mutual recursions (where
253        group 1 called group 2 while simultaneously a separate group 2 called group
254        1). A stack overflow occurred in this case. I have fixed this by limiting
255        the recursion depth to 10.
256    
257    27. Updated RunTest.bat in the distribution to the version supplied by Tom
258        Fortmann. This supports explicit test numbers on the command line, and has
259        argument validation and error reporting.
260    
261    28. An instance of \X with an unlimited repeat could fail if at any point the
262        first character it looked at was a mark character.
263    
264    29. Some minor code refactoring concerning Unicode properties and scripts
265        should reduce the stack requirement of match() slightly.
266    
267    30. Added the '=' option to pcretest to check the setting of unused capturing
268        slots at the end of the pattern, which are documented as being -1, but are
269        not included in the return count.
270    
271    31. If \k was not followed by a braced, angle-bracketed, or quoted name, PCRE
272        compiled something random. Now it gives a compile-time error (as does
273        Perl).
274    
275    32. A *MARK encountered during the processing of a positive assertion is now
276        recorded and passed back (compatible with Perl).
277    
278    33. If --only-matching or --colour was set on a pcregrep call whose pattern
279        had alternative anchored branches, the search for a second match in a line
280        was done as if at the line start. Thus, for example, /^01|^02/ incorrectly
281        matched the line "0102" twice. The same bug affected patterns that started
282        with a backwards assertion. For example /\b01|\b02/ also matched "0102"
283        twice.
284    
285    34. Previously, PCRE did not allow quantification of assertions. However, Perl
286        does, and because of capturing effects, quantifying parenthesized
287        assertions may at times be useful. Quantifiers are now allowed for
288        parenthesized assertions.
289    
290    35. A minor code tidy in pcre_compile() when checking options for \R usage.
291    
292    36. \g was being checked for fancy things in a character class, when it should
293        just be a literal "g".
294    
295    37. PCRE was rejecting [:a[:digit:]] whereas Perl was not. It seems that the
296        appearance of a nested POSIX class supersedes an apparent external class.
297        For example, [:a[:digit:]b:] matches "a", "b", ":", or a digit. Also,
298        unescaped square brackets may also appear as part of class names. For
299        example, [:a[:abc]b:] gives unknown class "[:abc]b:]". PCRE now behaves
300        more like Perl. (But see 8.20/1 above.)
301    
302    38. PCRE was giving an error for \N with a braced quantifier such as {1,} (this
303        was because it thought it was \N{name}, which is not supported).
304    
305    39. Add minix to OS list not supporting the -S option in pcretest.
306    
307    40. PCRE tries to detect cases of infinite recursion at compile time, but it
308        cannot analyze patterns in sufficient detail to catch mutual recursions
309        such as ((?1))((?2)). There is now a runtime test that gives an error if a
310        subgroup is called recursively as a subpattern for a second time at the
311        same position in the subject string. In previous releases this might have
312        been caught by the recursion limit, or it might have run out of stack.
313    
314    41. A pattern such as /(?(R)a+|(?R)b)/ is quite safe, as the recursion can
315        happen only once. PCRE was, however incorrectly giving a compile time error
316        "recursive call could loop indefinitely" because it cannot analyze the
317        pattern in sufficient detail. The compile time test no longer happens when
318        PCRE is compiling a conditional subpattern, but actual runaway loops are
319        now caught at runtime (see 40 above).
320    
321    42. It seems that Perl allows any characters other than a closing parenthesis
322        to be part of the NAME in (*MARK:NAME) and other backtracking verbs. PCRE
323        has been changed to be the same.
324    
325    43. Updated configure.ac to put in more quoting round AC_LANG_PROGRAM etc. so
326        as not to get warnings when autogen.sh is called. Also changed
327        AC_PROG_LIBTOOL (deprecated) to LT_INIT (the current macro).
328    
329    44. To help people who use pcregrep to scan files containing exceedingly long
330        lines, the following changes have been made:
331    
332        (a) The default value of the buffer size parameter has been increased from
333            8K to 20K. (The actual buffer used is three times this size.)
334    
335        (b) The default can be changed by ./configure --with-pcregrep-bufsize when
336            PCRE is built.
337    
338        (c) A --buffer-size=n option has been added to pcregrep, to allow the size
339            to be set at run time.
340    
341        (d) Numerical values in pcregrep options can be followed by K or M, for
342            example --buffer-size=50K.
343    
344        (e) If a line being scanned overflows pcregrep's buffer, an error is now
345            given and the return code is set to 2.
346    
347    45. Add a pointer to the latest mark to the callout data block.
348    
349    46. The pattern /.(*F)/, when applied to "abc" with PCRE_PARTIAL_HARD, gave a
350        partial match of an empty string instead of no match. This was specific to
351        the use of ".".
352    
353    47. The pattern /f.*/8s, when applied to "for" with PCRE_PARTIAL_HARD, gave a
354        complete match instead of a partial match. This bug was dependent on both
355        the PCRE_UTF8 and PCRE_DOTALL options being set.
356    
357    48. For a pattern such as /\babc|\bdef/ pcre_study() was failing to set up the
358        starting byte set, because \b was not being ignored.
359    
360    
361    Version 8.12 15-Jan-2011
362    ------------------------
363    
364    1.  Fixed some typos in the markup of the man pages, and wrote a script that
365        checks for such things as part of the documentation building process.
366    
367    2.  On a big-endian 64-bit system, pcregrep did not correctly process the
368        --match-limit and --recursion-limit options (added for 8.11). In
369        particular, this made one of the standard tests fail. (The integer value
370        went into the wrong half of a long int.)
371    
372    3.  If the --colour option was given to pcregrep with -v (invert match), it
373        did strange things, either producing crazy output, or crashing. It should,
374        of course, ignore a request for colour when reporting lines that do not
375        match.
376    
377    4.  Another pcregrep bug caused similar problems if --colour was specified with
378        -M (multiline) and the pattern match finished with a line ending.
379    
380    5.  In pcregrep, when a pattern that ended with a literal newline sequence was
381        matched in multiline mode, the following line was shown as part of the
382        match. This seems wrong, so I have changed it.
383    
384    6.  Another pcregrep bug in multiline mode, when --colour was specified, caused
385        the check for further matches in the same line (so they could be coloured)
386        to overrun the end of the current line. If another match was found, it was
387        incorrectly shown (and then shown again when found in the next line).
388    
389    7.  If pcregrep was compiled under Windows, there was a reference to the
390        function pcregrep_exit() before it was defined. I am assuming this was
391        the cause of the "error C2371: 'pcregrep_exit' : redefinition;" that was
392        reported by a user. I've moved the definition above the reference.
393    
394    
395    Version 8.11 10-Dec-2010
396    ------------------------
397    
398    1.  (*THEN) was not working properly if there were untried alternatives prior
399        to it in the current branch. For example, in ((a|b)(*THEN)(*F)|c..) it
400        backtracked to try for "b" instead of moving to the next alternative branch
401        at the same level (in this case, to look for "c"). The Perl documentation
402        is clear that when (*THEN) is backtracked onto, it goes to the "next
403        alternative in the innermost enclosing group".
404    
405    2.  (*COMMIT) was not overriding (*THEN), as it does in Perl. In a pattern
406        such as   (A(*COMMIT)B(*THEN)C|D)  any failure after matching A should
407        result in overall failure. Similarly, (*COMMIT) now overrides (*PRUNE) and
408        (*SKIP), (*SKIP) overrides (*PRUNE) and (*THEN), and (*PRUNE) overrides
409        (*THEN).
410    
411    3.  If \s appeared in a character class, it removed the VT character from
412        the class, even if it had been included by some previous item, for example
413        in [\x00-\xff\s]. (This was a bug related to the fact that VT is not part
414        of \s, but is part of the POSIX "space" class.)
415    
416    4.  A partial match never returns an empty string (because you can always
417        match an empty string at the end of the subject); however the checking for
418        an empty string was starting at the "start of match" point. This has been
419        changed to the "earliest inspected character" point, because the returned
420        data for a partial match starts at this character. This means that, for
421        example, /(?<=abc)def/ gives a partial match for the subject "abc"
422        (previously it gave "no match").
423    
424    5.  Changes have been made to the way PCRE_PARTIAL_HARD affects the matching
425        of $, \z, \Z, \b, and \B. If the match point is at the end of the string,
426        previously a full match would be given. However, setting PCRE_PARTIAL_HARD
427        has an implication that the given string is incomplete (because a partial
428        match is preferred over a full match). For this reason, these items now
429        give a partial match in this situation. [Aside: previously, the one case
430        /t\b/ matched against "cat" with PCRE_PARTIAL_HARD set did return a partial
431        match rather than a full match, which was wrong by the old rules, but is
432        now correct.]
433    
434    6.  There was a bug in the handling of #-introduced comments, recognized when
435        PCRE_EXTENDED is set, when PCRE_NEWLINE_ANY and PCRE_UTF8 were also set.
436        If a UTF-8 multi-byte character included the byte 0x85 (e.g. +U0445, whose
437        UTF-8 encoding is 0xd1,0x85), this was misinterpreted as a newline when
438        scanning for the end of the comment. (*Character* 0x85 is an "any" newline,
439        but *byte* 0x85 is not, in UTF-8 mode). This bug was present in several
440        places in pcre_compile().
441    
442    7.  Related to (6) above, when pcre_compile() was skipping #-introduced
443        comments when looking ahead for named forward references to subpatterns,
444        the only newline sequence it recognized was NL. It now handles newlines
445        according to the set newline convention.
446    
447    8.  SunOS4 doesn't have strerror() or strtoul(); pcregrep dealt with the
448        former, but used strtoul(), whereas pcretest avoided strtoul() but did not
449        cater for a lack of strerror(). These oversights have been fixed.
450    
451    9.  Added --match-limit and --recursion-limit to pcregrep.
452    
453    10. Added two casts needed to build with Visual Studio when NO_RECURSE is set.
454    
455    11. When the -o option was used, pcregrep was setting a return code of 1, even
456        when matches were found, and --line-buffered was not being honoured.
457    
458    12. Added an optional parentheses number to the -o and --only-matching options
459        of pcregrep.
460    
461    13. Imitating Perl's /g action for multiple matches is tricky when the pattern
462        can match an empty string. The code to do it in pcretest and pcredemo
463        needed fixing:
464    
465        (a) When the newline convention was "crlf", pcretest got it wrong, skipping
466            only one byte after an empty string match just before CRLF (this case
467            just got forgotten; "any" and "anycrlf" were OK).
468    
469        (b) The pcretest code also had a bug, causing it to loop forever in UTF-8
470            mode when an empty string match preceded an ASCII character followed by
471            a non-ASCII character. (The code for advancing by one character rather
472            than one byte was nonsense.)
473    
474        (c) The pcredemo.c sample program did not have any code at all to handle
475            the cases when CRLF is a valid newline sequence.
476    
477    14. Neither pcre_exec() nor pcre_dfa_exec() was checking that the value given
478        as a starting offset was within the subject string. There is now a new
479        error, PCRE_ERROR_BADOFFSET, which is returned if the starting offset is
480        negative or greater than the length of the string. In order to test this,
481        pcretest is extended to allow the setting of negative starting offsets.
482    
483    15. In both pcre_exec() and pcre_dfa_exec() the code for checking that the
484        starting offset points to the beginning of a UTF-8 character was
485        unnecessarily clumsy. I tidied it up.
486    
487    16. Added PCRE_ERROR_SHORTUTF8 to make it possible to distinguish between a
488        bad UTF-8 sequence and one that is incomplete when using PCRE_PARTIAL_HARD.
489    
490    17. Nobody had reported that the --include_dir option, which was added in
491        release 7.7 should have been called --include-dir (hyphen, not underscore)
492        for compatibility with GNU grep. I have changed it to --include-dir, but
493        left --include_dir as an undocumented synonym, and the same for
494        --exclude-dir, though that is not available in GNU grep, at least as of
495        release 2.5.4.
496    
497    18. At a user's suggestion, the macros GETCHAR and friends (which pick up UTF-8
498        characters from a string of bytes) have been redefined so as not to use
499        loops, in order to improve performance in some environments. At the same
500        time, I abstracted some of the common code into auxiliary macros to save
501        repetition (this should not affect the compiled code).
502    
503    19. If \c was followed by a multibyte UTF-8 character, bad things happened. A
504        compile-time error is now given if \c is not followed by an ASCII
505        character, that is, a byte less than 128. (In EBCDIC mode, the code is
506        different, and any byte value is allowed.)
507    
508    20. Recognize (*NO_START_OPT) at the start of a pattern to set the PCRE_NO_
509        START_OPTIMIZE option, which is now allowed at compile time - but just
510        passed through to pcre_exec() or pcre_dfa_exec(). This makes it available
511        to pcregrep and other applications that have no direct access to PCRE
512        options. The new /Y option in pcretest sets this option when calling
513        pcre_compile().
514    
515    21. Change 18 of release 8.01 broke the use of named subpatterns for recursive
516        back references. Groups containing recursive back references were forced to
517        be atomic by that change, but in the case of named groups, the amount of
518        memory required was incorrectly computed, leading to "Failed: internal
519        error: code overflow". This has been fixed.
520    
521    22. Some patches to pcre_stringpiece.h, pcre_stringpiece_unittest.cc, and
522        pcretest.c, to avoid build problems in some Borland environments.
523    
524    
525    Version 8.10 25-Jun-2010
526    ------------------------
527    
528    1.  Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
529        THEN.
530    
531    2.  (*ACCEPT) was not working when inside an atomic group.
532    
533    3.  Inside a character class, \B is treated as a literal by default, but
534        faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
535        causes the error). The code is unchanged, but I tidied the documentation.
536    
537    4.  Inside a character class, PCRE always treated \R and \X as literals,
538        whereas Perl faults them if its -w option is set. I have changed PCRE so
539        that it faults them when PCRE_EXTRA is set.
540    
541    5.  Added support for \N, which always matches any character other than
542        newline. (It is the same as "." when PCRE_DOTALL is not set.)
543    
544    6.  When compiling pcregrep with newer versions of gcc which may have
545        FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
546        declared with attribute warn_unused_result" were given. Just casting the
547        result to (void) does not stop the warnings; a more elaborate fudge is
548        needed. I've used a macro to implement this.
549    
550    7.  Minor change to pcretest.c to avoid a compiler warning.
551    
552    8.  Added four artifical Unicode properties to help with an option to make
553        \s etc use properties (see next item). The new properties are: Xan
554        (alphanumeric), Xsp (Perl space), Xps (POSIX space), and Xwd (word).
555    
556    9.  Added PCRE_UCP to make \b, \d, \s, \w, and certain POSIX character classes
557        use Unicode properties. (*UCP) at the start of a pattern can be used to set
558        this option. Modified pcretest to add /W to test this facility. Added
559        REG_UCP to make it available via the POSIX interface.
560    
561    10. Added --line-buffered to pcregrep.
562    
563    11. In UTF-8 mode, if a pattern that was compiled with PCRE_CASELESS was
564        studied, and the match started with a letter with a code point greater than
565        127 whose first byte was different to the first byte of the other case of
566        the letter, the other case of this starting letter was not recognized
567        (#976).
568    
569    12. If a pattern that was studied started with a repeated Unicode property
570        test, for example, \p{Nd}+, there was the theoretical possibility of
571        setting up an incorrect bitmap of starting bytes, but fortunately it could
572        not have actually happened in practice until change 8 above was made (it
573        added property types that matched character-matching opcodes).
574    
575    13. pcre_study() now recognizes \h, \v, and \R when constructing a bit map of
576        possible starting bytes for non-anchored patterns.
577    
578    14. Extended the "auto-possessify" feature of pcre_compile(). It now recognizes
579        \R, and also a number of cases that involve Unicode properties, both
580        explicit and implicit when PCRE_UCP is set.
581    
582    15. If a repeated Unicode property match (e.g. \p{Lu}*) was used with non-UTF-8
583        input, it could crash or give wrong results if characters with values
584        greater than 0xc0 were present in the subject string. (Detail: it assumed
585        UTF-8 input when processing these items.)
586    
587    16. Added a lot of (int) casts to avoid compiler warnings in systems where
588        size_t is 64-bit (#991).
589    
590    17. Added a check for running out of memory when PCRE is compiled with
591        --disable-stack-for-recursion (#990).
592    
593    18. If the last data line in a file for pcretest does not have a newline on
594        the end, a newline was missing in the output.
595    
596    19. The default pcre_chartables.c file recognizes only ASCII characters (values
597        less than 128) in its various bitmaps. However, there is a facility for
598        generating tables according to the current locale when PCRE is compiled. It
599        turns out that in some environments, 0x85 and 0xa0, which are Unicode space
600        characters, are recognized by isspace() and therefore were getting set in
601        these tables, and indeed these tables seem to approximate to ISO 8859. This
602        caused a problem in UTF-8 mode when pcre_study() was used to create a list
603        of bytes that can start a match. For \s, it was including 0x85 and 0xa0,
604        which of course cannot start UTF-8 characters. I have changed the code so
605        that only real ASCII characters (less than 128) and the correct starting
606        bytes for UTF-8 encodings are set for characters greater than 127 when in
607        UTF-8 mode. (When PCRE_UCP is set - see 9 above - the code is different
608        altogether.)
609    
610    20. Added the /T option to pcretest so as to be able to run tests with non-
611        standard character tables, thus making it possible to include the tests
612        used for 19 above in the standard set of tests.
613    
614    21. A pattern such as (?&t)(?#()(?(DEFINE)(?<t>a)) which has a forward
615        reference to a subpattern the other side of a comment that contains an
616        opening parenthesis caused either an internal compiling error, or a
617        reference to the wrong subpattern.
618    
619    
620    Version 8.02 19-Mar-2010
621    ------------------------
622    
623    1.  The Unicode data tables have been updated to Unicode 5.2.0.
624    
625    2.  Added the option --libs-cpp to pcre-config, but only when C++ support is
626        configured.
627    
628    3.  Updated the licensing terms in the pcregexp.pas file, as agreed with the
629        original author of that file, following a query about its status.
630    
631    4.  On systems that do not have stdint.h (e.g. Solaris), check for and include
632        inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
633    
634    5.  A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
635        quantifier applied to a forward-referencing subroutine call, could compile
636        incorrect code or give the error "internal error: previously-checked
637        referenced subpattern not found".
638    
639    6.  Both MS Visual Studio and Symbian OS have problems with initializing
640        variables to point to external functions. For these systems, therefore,
641        pcre_malloc etc. are now initialized to local functions that call the
642        relevant global functions.
643    
644    7.  There were two entries missing in the vectors called coptable and poptable
645        in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
646        I've fixed the data, and added a kludgy way of testing at compile time that
647        the lengths are correct (equal to the number of opcodes).
648    
649    8.  Following on from 7, I added a similar kludge to check the length of the
650        eint vector in pcreposix.c.
651    
652    9.  Error texts for pcre_compile() are held as one long string to avoid too
653        much relocation at load time. To find a text, the string is searched,
654        counting zeros. There was no check for running off the end of the string,
655        which could happen if a new error number was added without updating the
656        string.
657    
658    10. \K gave a compile-time error if it appeared in a lookbehind assersion.
659    
660    11. \K was not working if it appeared in an atomic group or in a group that
661        was called as a "subroutine", or in an assertion. Perl 5.11 documents that
662        \K is "not well defined" if used in an assertion. PCRE now accepts it if
663        the assertion is positive, but not if it is negative.
664    
665    12. Change 11 fortuitously reduced the size of the stack frame used in the
666        "match()" function of pcre_exec.c by one pointer. Forthcoming
667        implementation of support for (*MARK) will need an extra pointer on the
668        stack; I have reserved it now, so that the stack frame size does not
669        decrease.
670    
671    13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
672        item in branch that calls a recursion is a subroutine call - as in the
673        second branch in the above example - was incorrectly given the compile-
674        time error "recursive call could loop indefinitely" because pcre_compile()
675        was not correctly checking the subroutine for matching a non-empty string.
676    
677    14. The checks for overrunning compiling workspace could trigger after an
678        overrun had occurred. This is a "should never occur" error, but it can be
679        triggered by pathological patterns such as hundreds of nested parentheses.
680        The checks now trigger 100 bytes before the end of the workspace.
681    
682    15. Fix typo in configure.ac: "srtoq" should be "strtoq".
683    
684    
685    Version 8.01 19-Jan-2010
686    ------------------------
687    
688    1.  If a pattern contained a conditional subpattern with only one branch (in
689        particular, this includes all (*DEFINE) patterns), a call to pcre_study()
690        computed the wrong minimum data length (which is of course zero for such
691        subpatterns). This could cause incorrect "no match" results.
692    
693    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
694        the pattern is reset in the first branch, pcre_compile() failed with
695        "internal error: code overflow at offset...". This happened only when
696        the reset was to the original external option setting. (An optimization
697        abstracts leading options settings into an external setting, which was the
698        cause of this.)
699    
700    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
701        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
702        assertion pattern did not match (meaning that the assertion was true), it
703        was incorrectly treated as false if the SKIP had been reached during the
704        matching. This also applied to assertions used as conditions.
705    
706    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
707        assertion subpattern, including such a pattern used as a condition,
708        unpredictable results occurred, instead of the error return
709        PCRE_ERROR_DFA_UITEM.
710    
711    5.  The C++ GlobalReplace function was not working like Perl for the special
712        situation when an empty string is matched. It now does the fancy magic
713        stuff that is necessary.
714    
715    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
716        removed. (These were left over from very, very early versions of PCRE.)
717    
718    7.  Some cosmetic changes to the code to make life easier when compiling it
719        as part of something else:
720    
721        (a) Change DEBUG to PCRE_DEBUG.
722    
723        (b) In pcre_compile(), rename the member of the "branch_chain" structure
724            called "current" as "current_branch", to prevent a collision with the
725            Linux macro when compiled as a kernel module.
726    
727        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
728            prevent a collision with the Linux macro when compiled as a kernel
729            module.
730    
731    8.  In pcre_compile() there are some checks for integer overflows that used to
732        cast potentially large values to (double). This has been changed to that
733        when building, a check for int64_t is made, and if it is found, it is used
734        instead, thus avoiding the use of floating point arithmetic. (There is no
735        other use of FP in PCRE.) If int64_t is not found, the fallback is to
736        double.
737    
738    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
739        2005 (difference between two addresses compared to an unsigned value).
740    
741    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
742        custom one, because of the following reported problem in Windows:
743    
744          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
745              under Win32.
746          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
747              therefore missing the function definition.
748          - The compiler thus generates a "C" signature for the test function.
749          - The linker fails to find the "C" function.
750          - PCRE fails to configure if asked to do so against libbz2.
751    
752    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
753        messages were output:
754    
755          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
756          rerunning libtoolize, to keep the correct libtool macros in-tree.
757          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
758    
759        I have done both of these things.
760    
761    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
762        most of the time, it *can* run out if it is given a pattern that contains a
763        runaway infinite recursion. I updated the discussion in the pcrestack man
764        page.
765    
766    13. Now that we have gone to the x.xx style of version numbers, the minor
767        version may start with zero. Using 08 or 09 is a bad idea because users
768        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
769        interpreted as invalid octal numbers. I've updated the previous comment in
770        configure.ac, and also added a check that gives an error if 08 or 09 are
771        used.
772    
773    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
774        causing partial matching to fail when the end of the subject matched \W
775        in a UTF-8 pattern where \W was quantified with a minimum of 3.
776    
777    15. There were some discrepancies between the declarations in pcre_internal.h
778        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
779        their definitions. The declarations used "const uschar *" and the
780        definitions used USPTR. Even though USPTR is normally defined as "const
781        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
782        reported that: "This difference in casting confuses some C++ compilers, for
783        example, SunCC recognizes above declarations as different functions and
784        generates broken code for hbpcre." I have changed the declarations to use
785        USPTR.
786    
787    16. GNU libtool is named differently on some systems. The autogen.sh script now
788        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
789        (FreeBSD).
790    
791    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
792        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
793        comment: "Figure out how to create a longlong from a string: strtoll and
794        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
795        instance, but it only takes 2 args instead of 3!"
796    
797    18. A subtle bug concerned with back references has been fixed by a change of
798        specification, with a corresponding code fix. A pattern such as
799        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
800        refers, was giving matches when it shouldn't. For example, xa=xaaa would
801        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
802        same bug. Such groups have to be quantified to be useful, or contained
803        inside another quantified group. (If there's no repetition, the reference
804        can never match.) The problem arises because, having left the group and
805        moved on to the rest of the pattern, a later failure that backtracks into
806        the group uses the captured value from the final iteration of the group
807        rather than the correct earlier one. I have fixed this in PCRE by forcing
808        any group that contains a reference to itself to be an atomic group; that
809        is, there cannot be any backtracking into it once it has completed. This is
810        similar to recursive and subroutine calls.
811    
812    
813    Version 8.00 19-Oct-09
814    ----------------------
815    
816    1.  The table for translating pcre_compile() error codes into POSIX error codes
817        was out-of-date, and there was no check on the pcre_compile() error code
818        being within the table. This could lead to an OK return being given in
819        error.
820    
821    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
822        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
823        in a Windows environment.
824    
825    3.  The pcregrep --count option prints the count for each file even when it is
826        zero, as does GNU grep. However, pcregrep was also printing all files when
827        --files-with-matches was added. Now, when both options are given, it prints
828        counts only for those files that have at least one match. (GNU grep just
829        prints the file name in this circumstance, but including the count seems
830        more useful - otherwise, why use --count?) Also ensured that the
831        combination -clh just lists non-zero counts, with no names.
832    
833    4.  The long form of the pcregrep -F option was incorrectly implemented as
834        --fixed_strings instead of --fixed-strings. This is an incompatible change,
835        but it seems right to fix it, and I didn't think it was worth preserving
836        the old behaviour.
837    
838    5.  The command line items --regex=pattern and --regexp=pattern were not
839        recognized by pcregrep, which required --regex pattern or --regexp pattern
840        (with a space rather than an '='). The man page documented the '=' forms,
841        which are compatible with GNU grep; these now work.
842    
843    6.  No libpcreposix.pc file was created for pkg-config; there was just
844        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
845    
846    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
847        when UCP support is not needed, by modifying the Python script that
848        generates it from Unicode data files. This should not matter if the module
849        is correctly used as a library, but I received one complaint about 50K of
850        unwanted data. My guess is that the person linked everything into his
851        program rather than using a library. Anyway, it does no harm.
852    
853    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
854        was a minimum greater than 1 for a wide character in a possessive
855        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
856        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
857        character. Chaos in the form of incorrect output or a compiling loop could
858        result.
859    
860    9.  The restrictions on what a pattern can contain when partial matching is
861        requested for pcre_exec() have been removed. All patterns can now be
862        partially matched by this function. In addition, if there are at least two
863        slots in the offset vector, the offset of the earliest inspected character
864        for the match and the offset of the end of the subject are set in them when
865        PCRE_ERROR_PARTIAL is returned.
866    
867    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
868        synonymous with PCRE_PARTIAL, for backwards compatibility, and
869        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
870        and may be more useful for multi-segment matching.
871    
872    11. Partial matching with pcre_exec() is now more intuitive. A partial match
873        used to be given if ever the end of the subject was reached; now it is
874        given only if matching could not proceed because another character was
875        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
876        string "Z", which now yields "no match" instead of "partial match". In the
877        case of pcre_dfa_exec(), "no match" is given if every matching path for the
878        final character ended with (*FAIL).
879    
880    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
881        if the pattern had a "must contain" character that was already found in the
882        earlier partial match, unless partial matching was again requested. For
883        example, with the pattern /dog.(body)?/, the "must contain" character is
884        "g". If the first part-match was for the string "dog", restarting with
885        "sbody" failed. This bug has been fixed.
886    
887    13. The string returned by pcre_dfa_exec() after a partial match has been
888        changed so that it starts at the first inspected character rather than the
889        first character of the match. This makes a difference only if the pattern
890        starts with a lookbehind assertion or \b or \B (\K is not supported by
891        pcre_dfa_exec()). It's an incompatible change, but it makes the two
892        matching functions compatible, and I think it's the right thing to do.
893    
894    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
895        so that the demonstration program is easily available in environments where
896        PCRE has not been installed from source.
897    
898    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
899        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
900        library.
901    
902    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
903        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
904        is not the first non-POSIX option to be added. Clearly some people find
905        these options useful.
906    
907    17. If a caller to the POSIX matching function regexec() passes a non-zero
908        value for nmatch with a NULL value for pmatch, the value of
909        nmatch is forced to zero.
910    
911    18. RunGrepTest did not have a test for the availability of the -u option of
912        the diff command, as RunTest does. It now checks in the same way as
913        RunTest, and also checks for the -b option.
914    
915    19. If an odd number of negated classes containing just a single character
916        interposed, within parentheses, between a forward reference to a named
917        subpattern and the definition of the subpattern, compilation crashed with
918        an internal error, complaining that it could not find the referenced
919        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
920        [The bug was that it was starting one character too far in when skipping
921        over the character class, thus treating the ] as data rather than
922        terminating the class. This meant it could skip too much.]
923    
924    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
925        /g option in pcretest when the pattern contains \K, which makes it possible
926        to have an empty string match not at the start, even when the pattern is
927        anchored. Updated pcretest and pcredemo to use this option.
928    
929    21. If the maximum number of capturing subpatterns in a recursion was greater
930        than the maximum at the outer level, the higher number was returned, but
931        with unset values at the outer level. The correct (outer level) value is
932        now given.
933    
934    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
935        PCRE did not set those parentheses (unlike Perl). I have now found a way to
936        make it do so. The string so far is captured, making this feature
937        compatible with Perl.
938    
939    23. The tests have been re-organized, adding tests 11 and 12, to make it
940        possible to check the Perl 5.10 features against Perl 5.10.
941    
942    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
943        pattern matches a fixed length string. PCRE did not allow this; now it
944        does. Neither allows recursion.
945    
946    25. I finally figured out how to implement a request to provide the minimum
947        length of subject string that was needed in order to match a given pattern.
948        (It was back references and recursion that I had previously got hung up
949        on.) This code has now been added to pcre_study(); it finds a lower bound
950        to the length of subject needed. It is not necessarily the greatest lower
951        bound, but using it to avoid searching strings that are too short does give
952        some useful speed-ups. The value is available to calling programs via
953        pcre_fullinfo().
954    
955    26. While implementing 25, I discovered to my embarrassment that pcretest had
956        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
957        study optimizations had never been tested with that matching function.
958        Oops. What is worse, even when it was passed study data, there was a bug in
959        pcre_dfa_exec() that meant it never actually used it. Double oops. There
960        were also very few tests of studied patterns with pcre_dfa_exec().
961    
962    27. If (?| is used to create subpatterns with duplicate numbers, they are now
963        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
964        on the other side of the coin, they are no longer allowed to have different
965        names, because these cannot be distinguished in PCRE, and this has caused
966        confusion. (This is a difference from Perl.)
967    
968    28. When duplicate subpattern names are present (necessarily with different
969        numbers, as required by 27 above), and a test is made by name in a
970        conditional pattern, either for a subpattern having been matched, or for
971        recursion in such a pattern, all the associated numbered subpatterns are
972        tested, and the overall condition is true if the condition is true for any
973        one of them. This is the way Perl works, and is also more like the way
974        testing by number works.
975    
976    
977    Version 7.9 11-Apr-09
978    ---------------------
979    
980    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
981        (pcretest), all targets were linked against these libraries. This included
982        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
983        libraries. This caused unwanted dependencies to be created. This problem
984        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
985        pcretest is linked with readline.
986    
987    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
988        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
989        moved outside it again, because FALSE and TRUE are already defined in AIX,
990        but BOOL is not.
991    
992    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
993        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
994    
995    4.  The pcregrep documentation said spaces were inserted as well as colons (or
996        hyphens) following file names and line numbers when outputting matching
997        lines. This is not true; no spaces are inserted. I have also clarified the
998        wording for the --colour (or --color) option.
999    
1000    5.  In pcregrep, when --colour was used with -o, the list of matching strings
1001        was not coloured; this is different to GNU grep, so I have changed it to be
1002        the same.
1003    
1004    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
1005        each matching line was coloured. Now it goes on to look for further matches
1006        of any of the test patterns, which is the same behaviour as GNU grep.
1007    
1008    7.  A pattern that could match an empty string could cause pcregrep to loop; it
1009        doesn't make sense to accept an empty string match in pcregrep, so I have
1010        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
1011        seems to be how GNU grep behaves.
1012    
1013    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
1014        start or after a newline", because the conditional assertion was not being
1015        correctly handled. The rule now is that both the assertion and what follows
1016        in the first alternative must satisfy the test.
1017    
1018    9.  If auto-callout was enabled in a pattern with a conditional group whose
1019        condition was an assertion, PCRE could crash during matching, both with
1020        pcre_exec() and pcre_dfa_exec().
1021    
1022    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
1023        used for matching.
1024    
1025    11. Unicode property support in character classes was not working for
1026        characters (bytes) greater than 127 when not in UTF-8 mode.
1027    
1028    12. Added the -M command line option to pcretest.
1029    
1030    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
1031    
1032    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
1033    
1034    16. Added comments and documentation about mis-use of no_arg in the C++
1035        wrapper.
1036    
1037    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
1038        from Martin Jerabek that uses macro names for all relevant character and
1039        string constants.
1040    
1041    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
1042        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
1043        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
1044        these, but not everybody uses configure.
1045    
1046    19. A conditional group that had only one branch was not being correctly
1047        recognized as an item that could match an empty string. This meant that an
1048        enclosing group might also not be so recognized, causing infinite looping
1049        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
1050        with the subject "ab", where knowledge that the repeated group can match
1051        nothing is needed in order to break the loop.
1052    
1053    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
1054        exec(), but without supplying a callout function, matching went wrong.
1055    
1056    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
1057        leak if the size of the offset vector was greater than 30. When the vector
1058        is smaller, the saved offsets during recursion go onto a local stack
1059        vector, but for larger vectors malloc() is used. It was failing to free
1060        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
1061        error, in fact).
1062    
1063    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
1064        heapframe that is used only when UTF-8 support is enabled. This caused no
1065        problem, but was untidy.
1066    
1067    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
1068        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
1069        included within another project.
1070    
1071    24. Steven Van Ingelgem's patches to add more options to the CMake support,
1072        slightly modified by me:
1073    
1074          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
1075              not building pcregrep.
1076    
1077          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
1078              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
1079    
1080    25. Forward references, both numeric and by name, in patterns that made use of
1081        duplicate group numbers, could behave incorrectly or give incorrect errors,
1082        because when scanning forward to find the reference group, PCRE was not
1083        taking into account the duplicate group numbers. A pattern such as
1084        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
1085    
1086    26. Changed a few more instances of "const unsigned char *" to USPTR, making
1087        the feature of a custom pointer more persuasive (as requested by a user).
1088    
1089    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
1090        pcretest.c, inside #ifndefs, because it seems they are sometimes already
1091        pre-defined.
1092    
1093    28. Added support for (*UTF8) at the start of a pattern.
1094    
1095    29. Arrange for flags added by the "release type" setting in CMake to be shown
1096        in the configuration summary.
1097    
1098    
1099    Version 7.8 05-Sep-08
1100    ---------------------
1101    
1102    1.  Replaced UCP searching code with optimized version as implemented for Ad
1103        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
1104        stage table and inline lookup instead of a function, giving speed ups of 2
1105        to 5 times on some simple patterns that I tested. Permission was given to
1106        distribute the MultiStage2.py script that generates the tables (it's not in
1107        the tarball, but is in the Subversion repository).
1108    
1109    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
1110        scripts.
1111    
1112    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
1113        a group with a zero qualifier. The result of the study could be incorrect,
1114        or the function might crash, depending on the pattern.
1115    
1116    4.  Caseless matching was not working for non-ASCII characters in back
1117        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
1118        It now works when Unicode Property Support is available.
1119    
1120    5.  In pcretest, an escape such as \x{de} in the data was always generating
1121        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
1122        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
1123        truncation.
1124    
1125    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
1126    
1127    7.  Added two (int) casts to pcregrep when printing the difference of two
1128        pointers, in case they are 64-bit values.
1129    
1130    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
1131        test 2 if it fails.
1132    
1133    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
1134        and a #define of that name to empty if it is not externally set. This is to
1135        allow users of MSVC to set it if necessary.
1136    
1137    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
1138        the convenience functions in the pcre_get.c source file.
1139    
1140    11. An option change at the start of a pattern that had top-level alternatives
1141        could cause overwriting and/or a crash. This command provoked a crash in
1142        some environments:
1143    
1144          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
1145    
1146        This potential security problem was recorded as CVE-2008-2371.
1147    
1148    12. For a pattern where the match had to start at the beginning or immediately
1149        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
1150        pcre_dfa_exec() could read past the end of the passed subject if there was
1151        no match. To help with detecting such bugs (e.g. with valgrind), I modified
1152        pcretest so that it places the subject at the end of its malloc-ed buffer.
1153    
1154    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
1155        exec() might read past the end of the data buffer in UTF-8 mode.
1156    
1157    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
1158        the data contained the byte 0x85 as part of a UTF-8 character within its
1159        first line. This applied both to normal and DFA matching.
1160    
1161    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
1162        /^[^d]*?$/8 failed to match "abc".
1163    
1164    16. Added a missing copyright notice to pcrecpp_internal.h.
1165    
1166    17. Make it more clear in the documentation that values returned from
1167        pcre_exec() in ovector are byte offsets, not character counts.
1168    
1169    18. Tidied a few places to stop certain compilers from issuing warnings.
1170    
1171    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
1172        supplied by Stefan Weber. I made a further small update for 7.8 because
1173        there is a change of source arrangements: the pcre_searchfuncs.c module is
1174        replaced by pcre_ucd.c.
1175    
1176    
1177    Version 7.7 07-May-08
1178    ---------------------
1179    
1180    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
1181        a string to a long long, pretend we don't even have a long long." This is
1182        done by checking for the strtoq, strtoll, and _strtoi64 functions.
1183    
1184    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
1185        pre-7.6 versions, which defined a global no_arg variable instead of putting
1186        it in the RE class. (See also #8 below.)
1187    
1188    3.  Remove a line of dead code, identified by coverity and reported by Nuno
1189        Lopes.
1190    
1191    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
1192    
1193        (1) The include/exclude patterns were being applied to the whole pathnames
1194            of files, instead of just to the final components.
1195    
1196        (2) If there was more than one level of directory, the subdirectories were
1197            skipped unless they satisfied the include/exclude conditions. This is
1198            inconsistent with GNU grep (and could even be seen as contrary to the
1199            pcregrep specification - which I improved to make it absolutely clear).
1200            The action now is always to scan all levels of directory, and just
1201            apply the include/exclude patterns to regular files.
1202    
1203    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
1204        --exclude_dir in the tests to avoid scanning .svn directories.
1205    
1206    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
1207        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
1208        doesn't support NULs in patterns.
1209    
1210    7.  Added some missing "const"s to declarations of static tables in
1211        pcre_compile.c and pcre_dfa_exec.c.
1212    
1213    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
1214        caused by fix #2  above. (Subsequently also a second patch to fix the
1215        first patch. And a third patch - this was a messy problem.)
1216    
1217    9.  Applied Craig's patch to remove the use of push_back().
1218    
1219    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
1220        matching function regexec().
1221    
1222    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
1223        which, however, unlike Perl's \g{...}, are subroutine calls, not back
1224        references. PCRE supports relative numbers with this syntax (I don't think
1225        Oniguruma does).
1226    
1227    12. Previously, a group with a zero repeat such as (...){0} was completely
1228        omitted from the compiled regex. However, this means that if the group
1229        was called as a subroutine from elsewhere in the pattern, things went wrong
1230        (an internal error was given). Such groups are now left in the compiled
1231        pattern, with a new opcode that causes them to be skipped at execution
1232        time.
1233    
1234    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
1235        to the way PCRE behaves:
1236    
1237        (a) A lone ] character is dis-allowed (Perl treats it as data).
1238    
1239        (b) A back reference to an unmatched subpattern matches an empty string
1240            (Perl fails the current match path).
1241    
1242        (c) A data ] in a character class must be notated as \] because if the
1243            first data character in a class is ], it defines an empty class. (In
1244            Perl it is not possible to have an empty class.) The empty class []
1245            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
1246            The negative empty class [^] matches any one character, independently
1247            of the DOTALL setting.
1248    
1249    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
1250        non-existent subpattern following a character class starting with ']' and
1251        containing () gave an internal compiling error instead of "reference to
1252        non-existent subpattern". Fortunately, when the pattern did exist, the
1253        compiled code was correct. (When scanning forwards to check for the
1254        existencd of the subpattern, it was treating the data ']' as terminating
1255        the class, so got the count wrong. When actually compiling, the reference
1256        was subsequently set up correctly.)
1257    
1258    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
1259        it was being rejected as not supported by pcre_dfa_exec(), even though
1260        other assertions are supported. I have made pcre_dfa_exec() support
1261        (*FAIL).
1262    
1263    16. The implementation of 13c above involved the invention of a new opcode,
1264        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
1265        cannot be changed at match time, I realized I could make a small
1266        improvement to matching performance by compiling OP_ALLANY instead of
1267        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
1268        on the OP_ANY path.
1269    
1270    17. Compiling pcretest on Windows with readline support failed without the
1271        following two fixes: (1) Make the unistd.h include conditional on
1272        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
1273    
1274    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
1275        ncurses library to be included for pcretest when ReadLine support is
1276        requested, but also to allow for it to be overridden. This patch came from
1277        Daniel Bergström.
1278    
1279    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
1280        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
1281        any errors with the current Unicode tables. Thanks to Peter Kankowski for
1282        spotting this.
1283    
1284    
1285    Version 7.6 28-Jan-08
1286    ---------------------
1287    
1288    1.  A character class containing a very large number of characters with
1289        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
1290        overflow.
1291    
1292    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
1293        HAVE_LONG_LONG is not defined.
1294    
1295    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
1296        bring them up to date and include new features. This patch includes:
1297    
1298        - Fixed PH's badly added libz and libbz2 support.
1299        - Fixed a problem with static linking.
1300        - Added pcredemo. [But later removed - see 7 below.]
1301        - Fixed dftables problem and added an option.
1302        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
1303            HAVE_LONG_LONG.
1304        - Added readline support for pcretest.
1305        - Added an listing of the option settings after cmake has run.
1306    
1307    4.  A user submitted a patch to Makefile that makes it easy to create
1308        "pcre.dll" under mingw when using Configure/Make. I added stuff to
1309        Makefile.am that cause it to include this special target, without
1310        affecting anything else. Note that the same mingw target plus all
1311        the other distribution libraries and programs are now supported
1312        when configuring with CMake (see 6 below) instead of with
1313        Configure/Make.
1314    
1315    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
1316        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
1317        exported in the Windows port". It has not yet been confirmed that the patch
1318        solves the problem, but it does no harm.
1319    
1320    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
1321        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
1322        with CMake, and also correct the comment about stack recursion.
1323    
1324    7.  Remove the automatic building of pcredemo from the ./configure system and
1325        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
1326        of a program that users should build themselves after PCRE is installed, so
1327        building it automatically is not really right. What is more, it gave
1328        trouble in some build environments.
1329    
1330    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
1331    
1332    
1333    Version 7.5 10-Jan-08
1334    ---------------------
1335    
1336    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
1337        values in parens when parsing an RE using the C++ wrapper."
1338    
1339    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
1340        Characters greater than 255 were excluded from the class instead of being
1341        included.
1342    
1343    3.  The same bug as (2) above applied to negated POSIX classes such as
1344        [:^space:].
1345    
1346    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
1347        defined or documented. It seems to have been a typo for PCRE_STATIC, so
1348        I have changed it.
1349    
1350    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
1351        first named subpattern) and a construct such as (?&a) would reference the
1352        first named subpattern whose name started with "a" (in other words, the
1353        length check was missing). Both these problems are fixed. "Subpattern name
1354        expected" is now given for (?&) (a zero-length name), and this patch also
1355        makes it give the same error for \k'' (previously it complained that that
1356        was a reference to a non-existent subpattern).
1357    
1358    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
1359        this is right because (?- can be followed by option settings as well as by
1360        digits. I have, however, made the messages clearer.
1361    
1362    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
1363        than the number used in the conditional) now cause a compile-time error.
1364        This is actually not compatible with Perl, which accepts such patterns, but
1365        treats the conditional as always being FALSE (as PCRE used to), but it
1366        seems to me that giving a diagnostic is better.
1367    
1368    8.  Change "alphameric" to the more common word "alphanumeric" in comments
1369        and messages.
1370    
1371    9.  Fix two occurrences of "backslash" in comments that should have been
1372        "backspace".
1373    
1374    10. Remove two redundant lines of code that can never be obeyed (their function
1375        was moved elsewhere).
1376    
1377    11. The program that makes PCRE's Unicode character property table had a bug
1378        which caused it to generate incorrect table entries for sequences of
1379        characters that have the same character type, but are in different scripts.
1380        It amalgamated them into a single range, with the script of the first of
1381        them. In other words, some characters were in the wrong script. There were
1382        thirteen such cases, affecting characters in the following ranges:
1383    
1384          U+002b0 - U+002c1
1385          U+0060c - U+0060d
1386          U+0061e - U+00612
1387          U+0064b - U+0065e
1388          U+0074d - U+0076d
1389          U+01800 - U+01805
1390          U+01d00 - U+01d77
1391          U+01d9b - U+01dbf
1392          U+0200b - U+0200f
1393          U+030fc - U+030fe
1394          U+03260 - U+0327f
1395          U+0fb46 - U+0fbb1
1396          U+10450 - U+1049d
1397    
1398    12. The -o option (show only the matching part of a line) for pcregrep was not
1399        compatible with GNU grep in that, if there was more than one match in a
1400        line, it showed only the first of them. It now behaves in the same way as
1401        GNU grep.
1402    
1403    13. If the -o and -v options were combined for pcregrep, it printed a blank
1404        line for every non-matching line. GNU grep prints nothing, and pcregrep now
1405        does the same. The return code can be used to tell if there were any
1406        non-matching lines.
1407    
1408    14. Added --file-offsets and --line-offsets to pcregrep.
1409    
1410    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
1411        infinitely looping recursion. The bug was that positive lookaheads were not
1412        being skipped when checking for a possible empty match (negative lookaheads
1413        and both kinds of lookbehind were skipped).
1414    
1415    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
1416        inclusion of <windows.h> to before rather than after the definition of
1417        INVALID_FILE_ATTRIBUTES (patch from David Byron).
1418    
1419    17. Specifying a possessive quantifier with a specific limit for a Unicode
1420        character property caused pcre_compile() to compile bad code, which led at
1421        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
1422        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
1423        caused the error; without that there was no problem.
1424    
1425    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
1426    
1427    19. Added --enable-pcretest-libreadline.
1428    
1429    20. In pcrecpp.cc, the variable 'count' was incremented twice in
1430        RE::GlobalReplace(). As a result, the number of replacements returned was
1431        double what it should be. I removed one of the increments, but Craig sent a
1432        later patch that removed the other one (the right fix) and added unit tests
1433        that check the return values (which was not done before).
1434    
1435    21. Several CMake things:
1436    
1437        (1) Arranged that, when cmake is used on Unix, the libraries end up with
1438            the names libpcre and libpcreposix, not just pcre and pcreposix.
1439    
1440        (2) The above change means that pcretest and pcregrep are now correctly
1441            linked with the newly-built libraries, not previously installed ones.
1442    
1443        (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
1444    
1445    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
1446        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
1447        UTF-8 newline character). The key issue is that the pattern starts .*;
1448        this means that the match must be either at the beginning, or after a
1449        newline. The bug was in the code for advancing after a failed match and
1450        checking that the new position followed a newline. It was not taking
1451        account of UTF-8 characters correctly.
1452    
1453    23. PCRE was behaving differently from Perl in the way it recognized POSIX
1454        character classes. PCRE was not treating the sequence [:...:] as a
1455        character class unless the ... were all letters. Perl, however, seems to
1456        allow any characters between [: and :], though of course it rejects as
1457        unknown any "names" that contain non-letters, because all the known class
1458        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
1459        for example, whereas PCRE did not - it did not recognize a POSIX character
1460        class. This seemed a bit dangerous, so the code has been changed to be
1461        closer to Perl. The behaviour is not identical to Perl, because PCRE will
1462        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
1463        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
1464        Perl does, and where it didn't before.
1465    
1466    24. Rewrite so as to remove the single use of %n from pcregrep because in some
1467        Windows environments %n is disabled by default.
1468    
1469    
1470    Version 7.4 21-Sep-07
1471    ---------------------
1472    
1473    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
1474        means that a class such as [\s] counted as "explicit reference to CR or
1475        LF". That isn't really right - the whole point of the change was to try to
1476        help when there was an actual mention of one of the two characters. So now
1477        the change happens only if \r or \n (or a literal CR or LF) character is
1478        encountered.
1479    
1480    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
1481        of both had grown to the point where there were only 3 bits left.
1482        Fortunately, there was spare space in the data structure, and so I have
1483        moved the internal flags into a new 16-bit field to free up more option
1484        bits.
1485    
1486    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
1487        but did not set the internal JCHANGED flag - either of these is enough to
1488        control the way the "get" function works - but the PCRE_INFO_JCHANGED
1489        facility is supposed to tell if (?J) was ever used, so now (?J) at the
1490        start sets both bits.
1491    
1492    4.  Added options (at build time, compile time, exec time) to change \R from
1493        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
1494    
1495    5.  doc/pcresyntax.html was missing from the distribution.
1496    
1497    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
1498        compatibility, even though it is no longer used.
1499    
1500    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
1501        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
1502        windows.h file is present (where different names are used). [This was
1503        reversed later after testing - see 16 below.]
1504    
1505    8.  Changed all #include <config.h> to #include "config.h". There were also
1506        some further <pcre.h> cases that I changed to "pcre.h".
1507    
1508    9.  When pcregrep was used with the --colour option, it missed the line ending
1509        sequence off the lines that it output.
1510    
1511    10. It was pointed out to me that arrays of string pointers cause lots of
1512        relocations when a shared library is dynamically loaded. A technique of
1513        using a single long string with a table of offsets can drastically reduce
1514        these. I have refactored PCRE in four places to do this. The result is
1515        dramatic:
1516    
1517          Originally:                          290
1518          After changing UCP table:            187
1519          After changing error message table:   43
1520          After changing table of "verbs"       36
1521          After changing table of Posix names   22
1522    
1523        Thanks to the folks working on Gregex for glib for this insight.
1524    
1525    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
1526        unicode-properties was also set.
1527    
1528    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
1529    
1530    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
1531        checked only for CRLF.
1532    
1533    14. Added casts to pcretest.c to avoid compiler warnings.
1534    
1535    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
1536    
1537    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
1538        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
1539        entirely. This removes changes made in 7 above.
1540    
1541    17. The CMake files have been updated, and there is now more information about
1542        building with CMake in the NON-UNIX-USE document.
1543    
1544    
1545    Version 7.3 28-Aug-07
1546    ---------------------
1547    
1548     1. In the rejigging of the build system that eventually resulted in 7.1, the
1549        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
1550        brackets there is not right, since it causes compilers to look for an
1551        installed pcre.h, not the version that is in the source that is being
1552        compiled (which of course may be different). I have changed it back to:
1553    
1554          #include "pcre.h"
1555    
1556        I have a vague recollection that the change was concerned with compiling in
1557        different directories, but in the new build system, that is taken care of
1558        by the VPATH setting the Makefile.
1559    
1560     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
1561        when the subject happened to end in the byte 0x85 (e.g. if the last
1562        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
1563        characters but of course it shouldn't be taken as a newline when it is part
1564        of another character. The bug was that, for an unlimited repeat of . in
1565        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
1566        characters when looking for a newline.
1567    
1568     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
1569    
1570     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
1571        in debug output.
1572    
1573     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
1574        long printing in the pcrecpp unittest when running under MinGW.
1575    
1576     6. ESC_K was left out of the EBCDIC table.
1577    
1578     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
1579        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
1580        limit also applies to "virtual nesting" when a pattern is recursive, and in
1581        this case 1000 isn't so big. I have been able to remove this limit at the
1582        expense of backing off one optimization in certain circumstances. Normally,
1583        when pcre_exec() would call its internal match() function recursively and
1584        immediately return the result unconditionally, it uses a "tail recursion"
1585        feature to save stack. However, when a subpattern that can match an empty
1586        string has an unlimited repetition quantifier, it no longer makes this
1587        optimization. That gives it a stack frame in which to save the data for
1588        checking that an empty string has been matched. Previously this was taken
1589        from the 1000-entry workspace that had been reserved. So now there is no
1590        explicit limit, but more stack is used.
1591    
1592     8. Applied Daniel's patches to solve problems with the import/export magic
1593        syntax that is required for Windows, and which was going wrong for the
1594        pcreposix and pcrecpp parts of the library. These were overlooked when this
1595        problem was solved for the main library.
1596    
1597     9. There were some crude static tests to avoid integer overflow when computing
1598        the size of patterns that contain repeated groups with explicit upper
1599        limits. As the maximum quantifier is 65535, the maximum group length was
1600        set at 30,000 so that the product of these two numbers did not overflow a
1601        32-bit integer. However, it turns out that people want to use groups that
1602        are longer than 30,000 bytes (though not repeat them that many times).
1603        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
1604        made it possible to implement the integer overflow checks in a much more
1605        dynamic way, which I have now done. The artificial limitation on group
1606        length has been removed - we now have only the limit on the total length of
1607        the compiled pattern, which depends on the LINK_SIZE setting.
1608    
1609    10. Fixed a bug in the documentation for get/copy named substring when
1610        duplicate names are permitted. If none of the named substrings are set, the
1611        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
1612        empty string.
1613    
1614    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
1615        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
1616        because the ] is interpreted as the first data character and the
1617        terminating ] is not found. PCRE has been made compatible with Perl in this
1618        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
1619        cause memory overwriting.
1620    
1621    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
1622        string has been matched (to stop an infinite loop). It was not recognizing
1623        a conditional subpattern that could match an empty string if that
1624        subpattern was within another subpattern. For example, it looped when
1625        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
1626        condition was not nested. This bug has been fixed.
1627    
1628    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
1629        past the start of the subject in the presence of bytes with the top bit
1630        set, for example "\x8aBCD".
1631    
1632    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
1633        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
1634    
1635    14. Optimized (?!) to (*FAIL).
1636    
1637    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
1638        This restricts code points to be within the range 0 to 0x10FFFF, excluding
1639        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
1640        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
1641        does: it's just the validity check that is more restrictive.
1642    
1643    16. Inserted checks for integer overflows during escape sequence (backslash)
1644        processing, and also fixed erroneous offset values for syntax errors during
1645        backslash processing.
1646    
1647    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
1648        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
1649    
1650    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
1651        caused an overrun.
1652    
1653    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
1654        something other than just ASCII characters) inside a group that had an
1655        unlimited repeat caused a loop at compile time (while checking to see
1656        whether the group could match an empty string).
1657    
1658    20. Debugging a pattern containing \p or \P could cause a crash. For example,
1659        [\P{Any}] did so. (Error in the code for printing property names.)
1660    
1661    21. An orphan \E inside a character class could cause a crash.
1662    
1663    22. A repeated capturing bracket such as (A)? could cause a wild memory
1664        reference during compilation.
1665    
1666    23. There are several functions in pcre_compile() that scan along a compiled
1667        expression for various reasons (e.g. to see if it's fixed length for look
1668        behind). There were bugs in these functions when a repeated \p or \P was
1669        present in the pattern. These operators have additional parameters compared
1670        with \d, etc, and these were not being taken into account when moving along
1671        the compiled data. Specifically:
1672    
1673        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
1674            length.
1675    
1676        (b) An item such as \pL+ within a repeated group could cause crashes or
1677            loops.
1678    
1679        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
1680            "reference to non-existent subpattern" error.
1681    
1682        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
1683    
1684    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
1685        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
1686    
1687    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
1688    
1689    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
1690        character were causing crashes (broken optimization).
1691    
1692    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
1693        \p or \P) caused a compile-time loop.
1694    
1695    28. More problems have arisen in unanchored patterns when CRLF is a valid line
1696        break. For example, the unstudied pattern [\r\n]A does not match the string
1697        "\r\nA" because change 7.0/46 below moves the current point on by two
1698        characters after failing to match at the start. However, the pattern \nA
1699        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
1700        the same is true. There doesn't seem any very clean way out of this, but
1701        what I have chosen to do makes the common cases work: PCRE now takes note
1702        of whether there can be an explicit match for \r or \n anywhere in the
1703        pattern, and if so, 7.0/46 no longer applies. As part of this change,
1704        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
1705        pattern has explicit CR or LF references.
1706    
1707    29. Added (*CR) etc for changing newline setting at start of pattern.
1708    
1709    
1710    Version 7.2 19-Jun-07
1711    ---------------------
1712    
1713     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
1714        which is apparently normally available under Windows.
1715    
1716     2. Re-jig the pcregrep tests with different newline settings in an attempt
1717        to make them independent of the local environment's newline setting.
1718    
1719     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
1720    
1721     4. Some of the "internals" tests were previously cut out when the link size
1722        was not 2, because the output contained actual offsets. The recent new
1723        "Z" feature of pcretest means that these can be cut out, making the tests
1724        usable with all link sizes.
1725    
1726     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
1727        stack recursion. This gives a massive performance boost under BSD, but just
1728        a small improvement under Linux. However, it saves one field in the frame
1729        in all cases.
1730    
1731     6. Added more features from the forthcoming Perl 5.10:
1732    
1733        (a) (?-n) (where n is a string of digits) is a relative subroutine or
1734            recursion call. It refers to the nth most recently opened parentheses.
1735    
1736        (b) (?+n) is also a relative subroutine call; it refers to the nth next
1737            to be opened parentheses.
1738    
1739        (c) Conditions that refer to capturing parentheses can be specified
1740            relatively, for example, (?(-2)... or (?(+3)...
1741    
1742        (d) \K resets the start of the current match so that everything before
1743            is not part of it.
1744    
1745        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
1746    
1747        (f) \g{name} is another synonym - part of Perl 5.10's unification of
1748            reference syntax.
1749    
1750        (g) (?| introduces a group in which the numbering of parentheses in each
1751            alternative starts with the same number.
1752    
1753        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
1754    
1755     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
1756        PCRE_INFO_JCHANGED.
1757    
1758     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
1759        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
1760        for detecting groups that can match an empty string.
1761    
1762     9. A pattern with a very large number of alternatives (more than several
1763        hundred) was running out of internal workspace during the pre-compile
1764        phase, where pcre_compile() figures out how much memory will be needed. A
1765        bit of new cunning has reduced the workspace needed for groups with
1766        alternatives. The 1000-alternative test pattern now uses 12 bytes of
1767        workspace instead of running out of the 4096 that are available.
1768    
1769    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
1770    
1771    11. Applied patch from Google to remove an optimization that didn't quite work.
1772        The report of the bug said:
1773    
1774          pcrecpp::RE("a*").FullMatch("aaa") matches, while
1775          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
1776          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
1777    
1778    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
1779        it matched the wrong number of bytes.
1780    
1781    
1782    Version 7.1 24-Apr-07
1783    ---------------------
1784    
1785     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
1786        that is more "standard", making use of automake and other Autotools. There
1787        is some re-arrangement of the files and adjustment of comments consequent
1788        on this.
1789    
1790     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
1791        for recursive directory scanning broke on some systems because the files
1792        are not scanned in any specific order and on different systems the order
1793        was different. A call to "sort" has been inserted into RunGrepTest for the
1794        approprate test as a short-term fix. In the longer term there may be an
1795        alternative.
1796    
1797     3. I had an email from Eric Raymond about problems translating some of PCRE's
1798        man pages to HTML (despite the fact that I distribute HTML pages, some
1799        people do their own conversions for various reasons). The problems
1800        concerned the use of low-level troff macros .br and .in. I have therefore
1801        removed all such uses from the man pages (some were redundant, some could
1802        be replaced by .nf/.fi pairs). The 132html script that I use to generate
1803        HTML has been updated to handle .nf/.fi and to complain if it encounters
1804        .br or .in.
1805    
1806     4. Updated comments in configure.ac that get placed in config.h.in and also
1807        arranged for config.h to be included in the distribution, with the name
1808        config.h.generic, for the benefit of those who have to compile without
1809        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
1810    
1811     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
1812        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
1813        makevp.bat for the current PCRE, using the additional files
1814        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
1815    
1816     6. A Windows user reported a minor discrepancy with test 2, which turned out
1817        to be caused by a trailing space on an input line that had got lost in his
1818        copy. The trailing space was an accident, so I've just removed it.
1819    
1820     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
1821        that is needed.
1822    
1823     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
1824        as "const" (a) because they are and (b) because it helps the PHP
1825        maintainers who have recently made a script to detect big data structures
1826        in the php code that should be moved to the .rodata section. I remembered
1827        to update Builducptable as well, so it won't revert if ucptable.h is ever
1828        re-created.
1829    
1830     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
1831        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
1832        order to be able to cut out the UTF-8 tables in the latter when UTF-8
1833        support is not required. This saves 1.5-2K of code, which is important in
1834        some applications.
1835    
1836        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
1837        so as not to refer to the tables, even though these functions will never be
1838        called when UTF-8 support is disabled. Otherwise there are problems with a
1839        shared library.
1840    
1841    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
1842    
1843        (a) It was defining its arguments as char * instead of void *.
1844    
1845        (b) It was assuming that all moves were upwards in memory; this was true
1846            a long time ago when I wrote it, but is no longer the case.
1847    
1848        The emulated memove() is provided for those environments that have neither
1849        memmove() nor bcopy(). I didn't think anyone used it these days, but that
1850        is clearly not the case, as these two bugs were recently reported.
1851    
1852    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
1853        and Detrail to create the HTML documentation, the .txt form of the man
1854        pages, and it removes trailing spaces from listed files. It also creates
1855        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
1856        case, it wraps all the #defines with #ifndefs. This script should be run
1857        before "make dist".
1858    
1859    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
1860        with Unicode property support.
1861    
1862        (a) For a maximizing quantifier, if the two different cases of the
1863            character were of different lengths in their UTF-8 codings (there are
1864            some cases like this - I found 11), and the matching function had to
1865            back up over a mixture of the two cases, it incorrectly assumed they
1866            were both the same length.
1867    
1868        (b) When PCRE was configured to use the heap rather than the stack for
1869            recursion during matching, it was not correctly preserving the data for
1870            the other case of a UTF-8 character when checking ahead for a match
1871            while processing a minimizing repeat. If the check also involved
1872            matching a wide character, but failed, corruption could cause an
1873            erroneous result when trying to check for a repeat of the original
1874            character.
1875    
1876    13. Some tidying changes to the testing mechanism:
1877    
1878        (a) The RunTest script now detects the internal link size and whether there
1879            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
1880            values substituted by "configure". (The RunGrepTest script already did
1881            this for UTF-8.) The configure.ac script no longer substitutes the
1882            relevant variables.
1883    
1884        (b) The debugging options /B and /D in pcretest show the compiled bytecode
1885            with length and offset values. This means that the output is different
1886            for different internal link sizes. Test 2 is skipped for link sizes
1887            other than 2 because of this, bypassing the problem. Unfortunately,
1888            there was also a test in test 3 (the locale tests) that used /B and
1889            failed for link sizes other than 2. Rather than cut the whole test out,
1890            I have added a new /Z option to pcretest that replaces the length and
1891            offset values with spaces. This is now used to make test 3 independent
1892            of link size. (Test 2 will be tidied up later.)
1893    
1894    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
1895        segmentation fault instead of returning the appropriate error message.
1896    
1897    15. In multiline mode when the newline sequence was set to "any", the pattern
1898        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
1899        This doesn't seem right; it now treats the CRLF combination as the line
1900        ending, and so does not match in that case. It's only a pattern such as ^$
1901        that would hit this one: something like ^ABC$ would have failed after \r
1902        and then tried again after \r\n.
1903    
1904    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
1905        in an attempt to make files that differ only in their line terminators
1906        compare equal. This works on Linux.
1907    
1908    17. Under certain error circumstances pcregrep might try to free random memory
1909        as it exited. This is now fixed, thanks to valgrind.
1910    
1911    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
1912        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
1913        was because its rules for how to advance for /g after matching an empty
1914        string at the end of a line did not allow for this case. They now check for
1915        it specially.
1916    
1917    20. pcretest is supposed to handle patterns and data of any length, by
1918        extending its buffers when necessary. It was getting this wrong when the
1919        buffer for a data line had to be extended.
1920    
1921    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
1922        CRLF as a newline sequence.
1923    
1924    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
1925        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
1926        I have nevertheless tidied it up.
1927    
1928    23. Added some casts to kill warnings from HP-UX ia64 compiler.
1929    
1930    24. Added a man page for pcre-config.
1931    
1932    
1933    Version 7.0 19-Dec-06
1934    ---------------------
1935    
1936     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
1937        moving to gcc 4.1.1.
1938    
1939     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
1940        sys/time.h, which is documented as needed for this function. It doesn't
1941        seem to matter on Linux, but it showed up on some releases of OS X.
1942    
1943     3. It seems that there are systems where bytes whose values are greater than
1944        127 match isprint() in the "C" locale. The "C" locale should be the
1945        default when a C program starts up. In most systems, only ASCII printing
1946        characters match isprint(). This difference caused the output from pcretest
1947        to vary, making some of the tests fail. I have changed pcretest so that:
1948    
1949        (a) When it is outputting text in the compiled version of a pattern, bytes
1950            other than 32-126 are always shown as hex escapes.
1951    
1952        (b) When it is outputting text that is a matched part of a subject string,
1953            it does the same, unless a different locale has been set for the match
1954            (using the /L modifier). In this case, it uses isprint() to decide.
1955    
1956     4. Fixed a major bug that caused incorrect computation of the amount of memory
1957        required for a compiled pattern when options that changed within the
1958        pattern affected the logic of the preliminary scan that determines the
1959        length. The relevant options are -x, and -i in UTF-8 mode. The result was
1960        that the computed length was too small. The symptoms of this bug were
1961        either the PCRE error "internal error: code overflow" from pcre_compile(),
1962        or a glibc crash with a message such as "pcretest: free(): invalid next
1963        size (fast)". Examples of patterns that provoked this bug (shown in
1964        pcretest format) are:
1965    
1966          /(?-x: )/x
1967          /(?x)(?-x: \s*#\s*)/
1968          /((?i)[\x{c0}])/8
1969          /(?i:[\x{c0}])/8
1970    
1971        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
1972        is now done differently.
1973    
1974     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
1975        wrapper classes; (b) implement a new function in the C++ scanner that is
1976        more efficient than the old way of doing things because it avoids levels of
1977        recursion in the regex matching; (c) add a paragraph to the documentation
1978        for the FullMatch() function.
1979    
1980     6. The escape sequence \n was being treated as whatever was defined as
1981        "newline". Not only was this contrary to the documentation, which states
1982        that \n is character 10 (hex 0A), but it also went horribly wrong when
1983        "newline" was defined as CRLF. This has been fixed.
1984    
1985     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
1986        was being set to -1 for the "end of line" case (supposedly a value that no
1987        character can have). Though this value is never used (the check for end of
1988        line is "zero bytes in current character"), it caused compiler complaints.
1989        I've changed it to 0xffffffff.
1990    
1991     8. In pcre_version.c, the version string was being built by a sequence of
1992        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
1993        string (as it is for production releases) called a macro with an empty
1994        argument. The C standard says the result of this is undefined. The gcc
1995        compiler treats it as an empty string (which was what was wanted) but it is
1996        reported that Visual C gives an error. The source has been hacked around to
1997        avoid this problem.
1998    
1999     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
2000        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
2001        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
2002        of them did).
2003    
2004    10. Originally, pcretest opened its input and output without "b"; then I was
2005        told that "b" was needed in some environments, so it was added for release
2006        5.0 to both the input and output. (It makes no difference on Unix-like
2007        systems.) Later I was told that it is wrong for the input on Windows. I've
2008        now abstracted the modes into two macros, to make it easier to fiddle with
2009        them, and removed "b" from the input mode under Windows.
2010    
2011    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
2012    
2013    12. Added -help and --help to pcretest as an official way of being reminded
2014        of the options.
2015    
2016    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
2017        and pcrecpp.cc because they annoy compilers at high warning levels.
2018    
2019    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
2020    
2021    15. Fixed an occurrence of == in configure.ac that should have been = (shell
2022        scripts are not C programs :-) and which was not noticed because it works
2023        on Linux.
2024    
2025    16. pcretest is supposed to handle any length of pattern and data line (as one
2026        line or as a continued sequence of lines) by extending its input buffer if
2027        necessary. This feature was broken for very long pattern lines, leading to
2028        a string of junk being passed to pcre_compile() if the pattern was longer
2029        than about 50K.
2030    
2031    17. I have done a major re-factoring of the way pcre_compile() computes the
2032        amount of memory needed for a compiled pattern. Previously, there was code
2033        that made a preliminary scan of the pattern in order to do this. That was
2034        OK when PCRE was new, but as the facilities have expanded, it has become
2035        harder and harder to keep it in step with the real compile phase, and there
2036        have been a number of bugs (see for example, 4 above). I have now found a
2037        cunning way of running the real compile function in a "fake" mode that
2038        enables it to compute how much memory it would need, while actually only
2039        ever using a few hundred bytes of working memory and without too many
2040        tests of the mode. This should make future maintenance and development
2041        easier. A side effect of this work is that the limit of 200 on the nesting
2042        depth of parentheses has been removed (though this was never a serious
2043        limitation, I suspect). However, there is a downside: pcre_compile() now
2044        runs more slowly than before (30% or more, depending on the pattern). I
2045        hope this isn't a big issue. There is no effect on runtime performance.
2046    
2047    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
2048        newline (only possible for the last line of a file) and it was a
2049        pattern that set a locale (followed by /Lsomething), pcretest crashed.
2050    
2051    19. Added additional timing features to pcretest. (1) The -tm option now times
2052        matching only, not compiling. (2) Both -t and -tm can be followed, as a
2053        separate command line item, by a number that specifies the number of
2054        repeats to use when timing. The default is 50000; this gives better
2055        precision, but takes uncomfortably long for very large patterns.
2056    
2057    20. Extended pcre_study() to be more clever in cases where a branch of a
2058        subpattern has no definite first character. For example, (a*|b*)[cd] would
2059        previously give no result from pcre_study(). Now it recognizes that the
2060        first character must be a, b, c, or d.
2061    
2062    21. There was an incorrect error "recursive call could loop indefinitely" if
2063        a subpattern (or the entire pattern) that was being tested for matching an
2064        empty string contained only one non-empty item after a nested subpattern.
2065        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
2066        incorrectly, because the \d was being skipped in the check.
2067    
2068    22. The pcretest program now has a new pattern option /B and a command line
2069        option -b, which is equivalent to adding /B to every pattern. This causes
2070        it to show the compiled bytecode, without the additional information that
2071        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
2072        is the same as /B/I).
2073    
2074    23. A new optimization is now able automatically to treat some sequences such
2075        as a*b as a*+b. More specifically, if something simple (such as a character
2076        or a simple class like \d) has an unlimited quantifier, and is followed by
2077        something that cannot possibly match the quantified thing, the quantifier
2078        is automatically "possessified".
2079    
2080    24. A recursive reference to a subpattern whose number was greater than 39
2081        went wrong under certain circumstances in UTF-8 mode. This bug could also
2082        have affected the operation of pcre_study().
2083    
2084    25. Realized that a little bit of performance could be had by replacing
2085        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
2086    
2087    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
2088    
2089    27. Possessive quantifiers such as a++ were previously implemented by turning
2090        them into atomic groups such as ($>a+). Now they have their own opcodes,
2091        which improves performance. This includes the automatically created ones
2092        from 23 above.
2093    
2094    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
2095        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
2096        the first matched character to be a colon. This applied both to named and
2097        numbered groups.
2098    
2099    29. The ucpinternal.h header file was missing its idempotency #ifdef.
2100    
2101    30. I was sent a "project" file called libpcre.a.dev which I understand makes
2102        building PCRE on Windows easier, so I have included it in the distribution.
2103    
2104    31. There is now a check in pcretest against a ridiculously large number being
2105        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
2106        loop, the loop is abandoned.
2107    
2108    32. Forward references to subpatterns in conditions such as (?(2)...) where
2109        subpattern 2 is defined later cause pcre_compile() to search forwards in
2110        the pattern for the relevant set of parentheses. This search went wrong
2111        when there were unescaped parentheses in a character class, parentheses
2112        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
2113    
2114    33. "Subroutine" calls and backreferences were previously restricted to
2115        referencing subpatterns earlier in the regex. This restriction has now
2116        been removed.
2117    
2118    34. Added a number of extra features that are going to be in Perl 5.10. On the
2119        whole, these are just syntactic alternatives for features that PCRE had
2120        previously implemented using the Python syntax or my own invention. The
2121        other formats are all retained for compatibility.
2122    
2123        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
2124            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
2125            also .NET compatible.
2126    
2127        (b) A recursion or subroutine call to a named group can now be defined as
2128            (?&name) as well as (?P>name).
2129    
2130        (c) A backreference to a named group can now be defined as \k<name> or
2131            \k'name' as well as (?P=name). The new forms, as well as being in Perl
2132            5.10, are also .NET compatible.
2133    
2134        (d) A conditional reference to a named group can now use the syntax
2135            (?(<name>) or (?('name') as well as (?(name).
2136    
2137        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
2138            groups (named and numbered) that are never evaluated inline, but can be
2139            called as "subroutines" from elsewhere. In effect, the DEFINE condition
2140            is always false. There may be only one alternative in such a group.
2141    
2142        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
2143            as the simple (?(R). The condition is true only if the most recent
2144            recursion is that of the given number or name. It does not search out
2145            through the entire recursion stack.
2146    
2147        (g) The escape \gN or \g{N} has been added, where N is a positive or
2148            negative number, specifying an absolute or relative reference.
2149    
2150    35. Tidied to get rid of some further signed/unsigned compiler warnings and
2151        some "unreachable code" warnings.
2152    
2153    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
2154        things, this adds five new scripts.
2155    
2156    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
2157        There were also incompatibilities regarding the handling of \Q..\E inside
2158        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
2159        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
2160    
2161    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
2162        matches an empty string, and forcibly breaks the loop. There were bugs in
2163        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
2164        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
2165        separate and independent bugs (that affected different cases) have been
2166        fixed.
2167    
2168    39. Refactored the code to abolish the use of different opcodes for small
2169        capturing bracket numbers. This is a tidy that I avoided doing when I
2170        removed the limit on the number of capturing brackets for 3.5 back in 2001.
2171        The new approach is not only tidier, it makes it possible to reduce the
2172        memory needed to fix the previous bug (38).
2173    
2174    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
2175        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
2176        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
2177        mode.
2178    
2179    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
2180        report.
2181    
2182    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
2183        copy construction and assignment in the C++ wrapper.
2184    
2185    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
2186        couple of bugs that could have given wrong results in the "--newline=crlf"
2187        case.
2188    
2189    44. Added a number of casts and did some reorganization of signed/unsigned int
2190        variables following suggestions from Dair Grant. Also renamed the variable
2191        "this" as "item" because it is a C++ keyword.
2192    
2193    45. Arranged for dftables to add
2194    
2195          #include "pcre_internal.h"
2196    
2197        to pcre_chartables.c because without it, gcc 4.x may remove the array
2198        definition from the final binary if PCRE is built into a static library and
2199        dead code stripping is activated.
2200    
2201    46. For an unanchored pattern, if a match attempt fails at the start of a
2202        newline sequence, and the newline setting is CRLF or ANY, and the next two
2203        characters are CRLF, advance by two characters instead of one.
2204    
2205    
2206    Version 6.7 04-Jul-06
2207    ---------------------
2208    
2209     1. In order to handle tests when input lines are enormously long, pcretest has
2210        been re-factored so that it automatically extends its buffers when
2211        necessary. The code is crude, but this _is_ just a test program. The
2212        default size has been increased from 32K to 50K.
2213    
2214     2. The code in pcre_study() was using the value of the re argument before
2215        testing it for NULL. (Of course, in any sensible call of the function, it
2216        won't be NULL.)
2217    
2218     3. The memmove() emulation function in pcre_internal.h, which is used on
2219        systems that lack both memmove() and bcopy() - that is, hardly ever -
2220        was missing a "static" storage class specifier.
2221    
2222     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
2223        containing an extended class (one that cannot be represented by a bitmap
2224        because it contains high-valued characters or Unicode property items, e.g.
2225        [\pZ]). Almost always one would set UTF-8 mode when processing such a
2226        pattern, but PCRE should not loop if you do not (it no longer does).
2227        [Detail: two cases were found: (a) a repeated subpattern containing an
2228        extended class; (b) a recursive reference to a subpattern that followed a
2229        previous extended class. It wasn't skipping over the extended class
2230        correctly when UTF-8 mode was not set.]
2231    
2232     5. A negated single-character class was not being recognized as fixed-length
2233        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
2234        compile error "lookbehind assertion is not fixed length".
2235    
2236     6. The RunPerlTest auxiliary script was showing an unexpected difference
2237        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
2238        write a Perl script that can interpret lines of an input file either as
2239        byte characters or as UTF-8, which is what "perltest" was being required to
2240        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
2241        can't do is switch easily at run time between having the "use utf8;" pragma
2242        or not. In the end, I fudged it by using the RunPerlTest script to insert
2243        "use utf8;" explicitly for the UTF-8 tests.
2244    
2245     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
2246        the end of the subject string, contrary to the documentation and to what
2247        Perl does. This was true of both matching functions. Now it matches only at
2248        the start of the subject and immediately after *internal* newlines.
2249    
2250     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
2251        a pointer to an int instead of a pointer to an unsigned long int. This
2252        caused problems on 64-bit systems.
2253    
2254     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
2255        instance of the 'standard' template library not being so standard".
2256    
2257    10. There was no check on the number of named subpatterns nor the maximum
2258        length of a subpattern name. The product of these values is used to compute
2259        the size of the memory block for a compiled pattern. By supplying a very
2260        long subpattern name and a large number of named subpatterns, the size
2261        computation could be caused to overflow. This is now prevented by limiting
2262        the length of names to 32 characters, and the number of named subpatterns
2263        to 10,000.
2264    
2265    11. Subpatterns that are repeated with specific counts have to be replicated in
2266        the compiled pattern. The size of memory for this was computed from the
2267        length of the subpattern and the repeat count. The latter is limited to
2268        65535, but there was no limit on the former, meaning that integer overflow
2269        could in principle occur. The compiled length of a repeated subpattern is
2270        now limited to 30,000 bytes in order to prevent this.
2271    
2272    12. Added the optional facility to have named substrings with the same name.
2273    
2274    13. Added the ability to use a named substring as a condition, using the
2275        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
2276        are numbers (not recommended). Forward references are permitted.
2277    
2278    14. Added forward references in named backreferences (if you see what I mean).
2279    
2280    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
2281        pattern could run off the end of the subject. For example, the pattern
2282        "(?s)(.{1,5})"8 did this with the subject "ab".
2283    
2284    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
2285        PCRE_CASELESS was set when matching characters that were quantified with ?
2286        or *.
2287    
2288    17. A character class other than a single negated character that had a minimum
2289        but no maximum quantifier - for example [ab]{6,} - was not handled
2290        correctly by pce_dfa_exec(). It would match only one character.
2291    
2292    18. A valid (though odd) pattern that looked like a POSIX character
2293        class but used an invalid character after [ (for example [[,abc,]]) caused
2294        pcre_compile() to give the error "Failed: internal error: code overflow" or
2295        in some cases to crash with a glibc free() error. This could even happen if
2296        the pattern terminated after [[ but there just happened to be a sequence of
2297        letters, a binary zero, and a closing ] in the memory that followed.
2298    
2299    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
2300        over the years. Originally (before any Unicode support), just the bottom 8
2301        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
2302        output from "man perlunicode" includes this:
2303    
2304          The regular expression compiler produces polymorphic opcodes.  That
2305          is, the pattern adapts to the data and automatically switches to
2306          the Unicode character scheme when presented with Unicode data--or
2307          instead uses a traditional byte scheme when presented with byte
2308          data.
2309    
2310        Sadly, a wide octal escape does not cause a switch, and in a string with
2311        no other multibyte characters, these octal escapes are treated as before.
2312        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
2313        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
2314        Unicode string.
2315    
2316        I have not perpetrated such confusion in PCRE. Up till now, it took just
2317        the bottom 8 bits, as in old Perl. I have now made octal escapes with
2318        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
2319        translate to the appropriate multibyte character.
2320    
2321    29. Applied some refactoring to reduce the number of warnings from Microsoft
2322        and Borland compilers. This has included removing the fudge introduced
2323        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
2324        a warning about an unused variable.
2325    
2326    21. PCRE has not included VT (character 0x0b) in the set of whitespace
2327        characters since release 4.0, because Perl (from release 5.004) does not.
2328        [Or at least, is documented not to: some releases seem to be in conflict
2329        with the documentation.] However, when a pattern was studied with
2330        pcre_study() and all its branches started with \s, PCRE still included VT
2331        as a possible starting character. Of course, this did no harm; it just
2332        caused an unnecessary match attempt.
2333    
2334    22. Removed a now-redundant internal flag bit that recorded the fact that case
2335        dependency changed within the pattern. This was once needed for "required
2336        byte" processing, but is no longer used. This recovers a now-scarce options
2337        bit. Also moved the least significant internal flag bit to the most-
2338        significant bit of the word, which was not previously used (hangover from
2339        the days when it was an int rather than a uint) to free up another bit for
2340        the future.
2341    
2342    23. Added support for CRLF line endings as well as CR and LF. As well as the
2343        default being selectable at build time, it can now be changed at runtime
2344        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
2345        specify that it is scanning data with non-default line endings.
2346    
2347    24. Changed the definition of CXXLINK to make it agree with the definition of
2348        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
2349    
2350    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
2351        recursions. This makes a big different to stack usage for some patterns.
2352    
2353    26. If a subpattern containing a named recursion or subroutine reference such
2354        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
2355        the space required for the compiled pattern went wrong and gave too small a
2356        value. Depending on the environment, this could lead to "Failed: internal
2357        error: code overflow at offset 49" or "glibc detected double free or
2358        corruption" errors.
2359    
2360    27. Applied patches from Google (a) to support the new newline modes and (b) to
2361        advance over multibyte UTF-8 characters in GlobalReplace.
2362    
2363    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
2364        difference for some implementation of PCRE in some Windows version.
2365    
2366    29. Added some extra testing facilities to pcretest:
2367    
2368        \q<number>   in a data line sets the "match limit" value
2369        \Q<number>   in a data line sets the "match recursion limt" value
2370        -S <number>  sets the stack size, where <number> is in megabytes
2371    
2372        The -S option isn't available for Windows.
2373    
2374    
2375    Version 6.6 06-Feb-06
2376    ---------------------
2377    
2378     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
2379        in pcreposix.h. I have copied the definition from pcre.h.
2380    
2381     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
2382        because pcre.h is no longer a built file.
2383    
2384     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
2385        not normally included in the compiled code.
2386    
2387    
2388    Version 6.5 01-Feb-06
2389    ---------------------
2390    
2391     1. When using the partial match feature with pcre_dfa_exec(), it was not
2392        anchoring the second and subsequent partial matches at the new starting
2393        point. This could lead to incorrect results. For example, with the pattern
2394        /1234/, partially matching against "123" and then "a4" gave a match.
2395    
2396     2. Changes to pcregrep:
2397    
2398        (a) All non-match returns from pcre_exec() were being treated as failures
2399            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
2400            error message is output. Some extra information is given for the
2401            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
2402            probably the only errors that are likely to be caused by users (by
2403            specifying a regex that has nested indefinite repeats, for instance).
2404            If there are more than 20 of these errors, pcregrep is abandoned.
2405    
2406        (b) A binary zero was treated as data while matching, but terminated the
2407            output line if it was written out. This has been fixed: binary zeroes
2408            are now no different to any other data bytes.
2409    
2410        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
2411            used to set a locale for matching. The --locale=xxxx long option has
2412            been added (no short equivalent) to specify a locale explicitly on the
2413            pcregrep command, overriding the environment variables.
2414    
2415        (d) When -B was used with -n, some line numbers in the output were one less
2416            than they should have been.
2417    
2418        (e) Added the -o (--only-matching) option.
2419    
2420        (f) If -A or -C was used with -c (count only), some lines of context were
2421            accidentally printed for the final match.
2422    
2423        (g) Added the -H (--with-filename) option.
2424    
2425        (h) The combination of options -rh failed to suppress file names for files
2426            that were found from directory arguments.
2427    
2428        (i) Added the -D (--devices) and -d (--directories) options.
2429    
2430        (j) Added the -F (--fixed-strings) option.
2431    
2432        (k) Allow "-" to be used as a file name for -f as well as for a data file.
2433    
2434        (l) Added the --colo(u)r option.
2435    
2436        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
2437            is not present by default.
2438    
2439     3. A nasty bug was discovered in the handling of recursive patterns, that is,
2440        items such as (?R) or (?1), when the recursion could match a number of
2441        alternatives. If it matched one of the alternatives, but subsequently,
2442        outside the recursion, there was a failure, the code tried to back up into
2443        the recursion. However, because of the way PCRE is implemented, this is not
2444        possible, and the result was an incorrect result from the match.
2445    
2446        In order to prevent this happening, the specification of recursion has
2447        been changed so that all such subpatterns are automatically treated as
2448        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
2449    
2450     4. I had overlooked the fact that, in some locales, there are characters for
2451        which isalpha() is true but neither isupper() nor islower() are true. In
2452        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
2453        and ordfeminine) are like this. This affected the treatment of \w and \W
2454        when they appeared in character classes, but not when they appeared outside
2455        a character class. The bit map for "word" characters is now created
2456        separately from the results of isalnum() instead of just taking it from the
2457        upper, lower, and digit maps. (Plus the underscore character, of course.)
2458    
2459     5. The above bug also affected the handling of POSIX character classes such as
2460        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
2461        permanent tables. Instead, the bit maps for such a class were previously
2462        created as the appropriate unions of the upper, lower, and digit bitmaps.
2463        Now they are created by subtraction from the [[:word:]] class, which has
2464        its own bitmap.
2465    
2466     6. The [[:blank:]] character class matches horizontal, but not vertical space.
2467        It is created by subtracting the vertical space characters (\x09, \x0a,
2468        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
2469        subtraction was done in the overall bitmap for a character class, meaning
2470        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
2471        be recognized. This bug has been fixed.
2472    
2473     7. Patches from the folks at Google:
2474    
2475          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
2476          real life, but is still worth protecting against".
2477    
2478          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
2479          regular expressions".
2480    
2481          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
2482          have it.
2483    
2484          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
2485          "configure" and the latter not, in order to fix a problem somebody had
2486          with compiling the Arg class on HP-UX.
2487    
2488          (e) Improve the error-handling of the C++ wrapper a little bit.
2489    
2490          (f) New tests for checking recursion limiting.
2491    
2492     8. The pcre_memmove() function, which is used only if the environment does not
2493        have a standard memmove() function (and is therefore rarely compiled),
2494        contained two bugs: (a) use of int instead of size_t, and (b) it was not
2495        returning a result (though PCRE never actually uses the result).
2496    
2497     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
2498        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
2499        returned instead of calling malloc() with an overflowing number that would
2500        most likely cause subsequent chaos.
2501    
2502    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
2503    
2504    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
2505        with this option is matched, the nmatch and pmatch options of regexec() are
2506        ignored.
2507    
2508    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
2509        provided in case anyone wants to the the POSIX interface with UTF-8
2510        strings.
2511    
2512    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
2513        C++ linking (needed for some HP-UX environments).
2514    
2515    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
2516        (unused parameter) and in the pcre_printint() function (omitted "default"
2517        switch label when the default is to do nothing).
2518    
2519    15. Added some code to make it possible, when PCRE is compiled as a C++
2520        library, to replace subject pointers for pcre_exec() with a smart pointer
2521        class, thus making it possible to process discontinuous strings.
2522    
2523    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
2524        much the same function. They were added by different people who were trying
2525        to make PCRE easy to compile on non-Unix systems. It has been suggested
2526        that PCRE_EXPORT be abolished now that there is more automatic apparatus
2527        for compiling on Windows systems. I have therefore replaced it with
2528        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
2529        defaults to "extern" for C or "extern C" for C++, which works fine on
2530        Unix-like systems. It is now possible to override the value of PCRE_DATA_
2531        SCOPE with something explicit in config.h. In addition:
2532    
2533        (a) pcreposix.h still had just "extern" instead of either of these macros;
2534            I have replaced it with PCRE_DATA_SCOPE.
2535    
2536        (b) Functions such as _pcre_xclass(), which are internal to the library,
2537            but external in the C sense, all had PCRE_EXPORT in their definitions.
2538            This is apparently wrong for the Windows case, so I have removed it.
2539            (It makes no difference on Unix-like systems.)
2540    
2541    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
2542        of recursive calls to match(). This is different to MATCH_LIMIT because
2543        that limits the total number of calls to match(), not all of which increase
2544        the depth of recursion. Limiting the recursion depth limits the amount of
2545        stack (or heap if NO_RECURSE is set) that is used. The default can be set
2546        when PCRE is compiled, and changed at run time. A patch from Google adds
2547        this functionality to the C++ interface.
2548    
2549    18. Changes to the handling of Unicode character properties:
2550    
2551        (a) Updated the table to Unicode 4.1.0.
2552    
2553        (b) Recognize characters that are not in the table as "Cn" (undefined).
2554    
2555        (c) I revised the way the table is implemented to a much improved format
2556            which includes recognition of ranges. It now supports the ranges that
2557            are defined in UnicodeData.txt, and it also amalgamates other
2558            characters into ranges. This has reduced the number of entries in the
2559            table from around 16,000 to around 3,000, thus reducing its size
2560            considerably. I realized I did not need to use a tree structure after
2561            all - a binary chop search is just as efficient. Having reduced the
2562            number of entries, I extended their size from 6 bytes to 8 bytes to
2563            allow for more data.
2564    
2565        (d) Added support for Unicode script names via properties such as \p{Han}.
2566    
2567    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
2568        matching that character.
2569    
2570    20. When matching a repeated Unicode property with a minimum greater than zero,
2571        (for example \pL{2,}), PCRE could look past the end of the subject if it
2572        reached it while seeking the minimum number of characters. This could
2573        happen only if some of the characters were more than one byte long, because
2574        there is a check for at least the minimum number of bytes.
2575    
2576    21. Refactored the implementation of \p and \P so as to be more general, to
2577        allow for more different types of property in future. This has changed the
2578        compiled form incompatibly. Anybody with saved compiled patterns that use
2579        \p or \P will have to recompile them.
2580    
2581    22. Added "Any" and "L&" to the supported property types.
2582    
2583    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
2584        but give a compile time error if the value is greater than 0xff.
2585    
2586    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
2587        accidentally not being installed or uninstalled.
2588    
2589    25. The pcre.h file was built from pcre.h.in, but the only changes that were
2590        made were to insert the current release number. This seemed silly, because
2591        it made things harder for people building PCRE on systems that don't run
2592        "configure". I have turned pcre.h into a distributed file, no longer built
2593        by "configure", with the version identification directly included. There is
2594        no longer a pcre.h.in file.
2595    
2596        However, this change necessitated a change to the pcre-config script as
2597        well. It is built from pcre-config.in, and one of the substitutions was the
2598        release number. I have updated configure.ac so that ./configure now finds
2599        the release number by grepping pcre.h.
2600    
2601    26. Added the ability to run the tests under valgrind.
2602    
2603    
2604    Version 6.4 05-Sep-05
2605    ---------------------
2606    
2607     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
2608        "--" to be printed when multiple files were scanned, even when none of the
2609        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
2610        consider it to be a bug, and have restored the previous behaviour.
2611    
2612     2. A couple of code tidies to get rid of compiler warnings.
2613    
2614     3. The pcretest program used to cheat by referring to symbols in the library
2615        whose names begin with _pcre_. These are internal symbols that are not
2616        really supposed to be visible externally, and in some environments it is
2617        possible to suppress them. The cheating is now confined to including
2618        certain files from the library's source, which is a bit cleaner.
2619    
2620     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
2621        file's purpose clearer.
2622    
2623     5. Reorganized pcre_ucp_findchar().
2624    
2625    
2626    Version 6.3 15-Aug-05
2627    ---------------------
2628    
2629     1. The file libpcre.pc.in did not have general read permission in the tarball.
2630    
2631     2. There were some problems when building without C++ support:
2632    
2633        (a) If C++ support was not built, "make install" and "make test" still
2634            tried to test it.
2635    
2636        (b) There were problems when the value of CXX was explicitly set. Some
2637            changes have been made to try to fix these, and ...
2638    
2639        (c) --disable-cpp can now be used to explicitly disable C++ support.
2640    
2641        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
2642            backslash in a target when C++ was disabled. This confuses some
2643            versions of "make", apparently. Using an intermediate variable solves
2644            this. (Same for CPP_LOBJ.)
2645    
2646     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
2647        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
2648        necessary on certain architectures.
2649    
2650     4. Added a setting of -export-symbols-regex to the link command to remove
2651        those symbols that are exported in the C sense, but actually are local
2652        within the library, and not documented. Their names all begin with
2653        "_pcre_". This is not a perfect job, because (a) we have to except some
2654        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
2655        available (and never for static libraries). I have made a note to try to
2656        find a way round (a) in the future.
2657    
2658    
2659    Version 6.2 01-Aug-05
2660    ---------------------
2661    
2662     1. There was no test for integer overflow of quantifier values. A construction
2663        such as {1111111111111111} would give undefined results. What is worse, if
2664        a minimum quantifier for a parenthesized subpattern overflowed and became
2665        negative, the calculation of the memory size went wrong. This could have
2666        led to memory overwriting.
2667    
2668     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
2669    
2670     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
2671        operating environments where this matters.
2672    
2673     4. Applied Giuseppe Maxia's patch to add additional features for controlling
2674        PCRE options from within the C++ wrapper.
2675    
2676     5. Named capturing subpatterns were not being correctly counted when a pattern
2677        was compiled. This caused two problems: (a) If there were more than 100
2678        such subpatterns, the calculation of the memory needed for the whole
2679        compiled pattern went wrong, leading to an overflow error. (b) Numerical
2680        back references of the form \12, where the number was greater than 9, were
2681        not recognized as back references, even though there were sufficient
2682        previous subpatterns.
2683    
2684     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
2685        versions of gcc, e.g. 2.95.4.
2686    
2687    
2688    Version 6.1 21-Jun-05
2689    ---------------------
2690    
2691     1. There was one reference to the variable "posix" in pcretest.c that was not
2692        surrounded by "#if !defined NOPOSIX".
2693    
2694     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
2695        the cross-check on the old pcre_info() function, for the benefit of the
2696        cut-down version of PCRE that is currently imported into Exim.
2697    
2698     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
2699        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
2700        patterns that start (?i) but I don't think that matters. The use of (?i) is
2701        just an example; this all applies to the other options as well.
2702    
2703     4. Since libtool seems to echo the compile commands it is issuing, the output
2704        from "make" can be reduced a bit by putting "@" in front of each libtool
2705        compile command.
2706    
2707     5. Patch from the folks at Google for configure.in to be a bit more thorough
2708        in checking for a suitable C++ installation before trying to compile the
2709        C++ stuff. This should fix a reported problem when a compiler was present,
2710        but no suitable headers.
2711    
2712     6. The man pages all had just "PCRE" as their title. I have changed them to
2713        be the relevant file name. I have also arranged that these names are
2714        retained in the file doc/pcre.txt, which is a concatenation in text format
2715        of all the man pages except the little individual ones for each function.
2716    
2717     7. The NON-UNIX-USE file had not been updated for the different set of source
2718        files that come with release 6. I also added a few comments about the C++
2719        wrapper.
2720    
2721    
2722    Version 6.0 07-Jun-05
2723    ---------------------
2724    
2725     1. Some minor internal re-organization to help with my DFA experiments.
2726    
2727     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
2728        didn't matter for the library itself when fully configured, but did matter
2729        when compiling without UCP support, or within Exim, where the ucp files are
2730        not imported.
2731    
2732     3. Refactoring of the library code to split up the various functions into
2733        different source modules. The addition of the new DFA matching code (see
2734        below) to a single monolithic source would have made it really too
2735        unwieldy, quite apart from causing all the code to be include in a
2736        statically linked application, when only some functions are used. This is
2737        relevant even without the DFA addition now that patterns can be compiled in
2738        one application and matched in another.
2739    
2740        The downside of splitting up is that there have to be some external
2741        functions and data tables that are used internally in different modules of
2742        the library but which are not part of the API. These have all had their
2743        names changed to start with "_pcre_" so that they are unlikely to clash
2744        with other external names.
2745    
2746     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
2747        a different (DFA) algorithm. Although it is slower than the original
2748        function, it does have some advantages for certain types of matching
2749        problem.
2750    
2751     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
2752        including restarting after a partial match.
2753    
2754     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
2755        defined when compiling for Windows was sent to me. I have put it into the
2756        code, though I have no means of testing or verifying it.
2757    
2758     7. Added the pcre_refcount() auxiliary function.
2759    
2760     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
2761        match before or at the first newline in the subject string. In pcretest,
2762        the /f option on a pattern can be used to set this.
2763    
2764     9. A repeated \w when used in UTF-8 mode with characters greater than 256
2765        would behave wrongly. This has been present in PCRE since release 4.0.
2766    
2767    10. A number of changes to the pcregrep command:
2768    
2769        (a) Refactored how -x works; insert ^(...)$ instead of setting
2770            PCRE_ANCHORED and checking the length, in preparation for adding
2771            something similar for -w.
2772    
2773        (b) Added the -w (match as a word) option.
2774    
2775        (c) Refactored the way lines are read and buffered so as to have more
2776            than one at a time available.
2777    
2778        (d) Implemented a pcregrep test script.
2779    
2780        (e) Added the -M (multiline match) option. This allows patterns to match
2781            over several lines of the subject. The buffering ensures that at least
2782            8K, or the rest of the document (whichever is the shorter) is available
2783            for matching (and similarly the previous 8K for lookbehind assertions).
2784    
2785        (f) Changed the --help output so that it now says
2786    
2787              -w, --word-regex(p)
2788    
2789            instead of two lines, one with "regex" and the other with "regexp"
2790            because that confused at least one person since the short forms are the
2791            same. (This required a bit of code, as the output is generated
2792            automatically from a table. It wasn't just a text change.)
2793    
2794        (g) -- can be used to terminate pcregrep options if the next thing isn't an
2795            option but starts with a hyphen. Could be a pattern or a path name
2796            starting with a hyphen, for instance.
2797    
2798        (h) "-" can be given as a file name to represent stdin.
2799    
2800        (i) When file names are being printed, "(standard input)" is used for
2801            the standard input, for compatibility with GNU grep. Previously
2802            "<stdin>" was used.
2803    
2804        (j) The option --label=xxx can be used to supply a name to be used for
2805            stdin when file names are being printed. There is no short form.
2806    
2807        (k) Re-factored the options decoding logic because we are going to add
2808            two more options that take data. Such options can now be given in four
2809            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
2810    
2811        (l) Added the -A, -B, and -C options for requesting that lines of context
2812            around matches be printed.
2813    
2814        (m) Added the -L option to print the names of files that do not contain
2815            any matching lines, that is, the complement of -l.
2816    
2817        (n) The return code is 2 if any file cannot be opened, but pcregrep does
2818            continue to scan other files.
2819    
2820        (o) The -s option was incorrectly implemented. For compatibility with other
2821            greps, it now suppresses the error message for a non-existent or non-
2822            accessible file (but not the return code). There is a new option called
2823            -q that suppresses the output of matching lines, which was what -s was
2824            previously doing.
2825    
2826        (p) Added --include and --exclude options to specify files for inclusion
2827            and exclusion when recursing.
2828    
2829    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
2830        Hopefully, it now does.
2831    
2832    12. Missing cast in pcre_study().
2833    
2834    13. Added an "uninstall" target to the makefile.
2835    
2836    14. Replaced "extern" in the function prototypes in Makefile.in with
2837        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
2838        world, but is set differently for Windows.
2839    
2840    15. Added a second compiling function called pcre_compile2(). The only
2841        difference is that it has an extra argument, which is a pointer to an
2842        integer error code. When there is a compile-time failure, this is set
2843        non-zero, in addition to the error test pointer being set to point to an
2844        error message. The new argument may be NULL if no error number is required
2845        (but then you may as well call pcre_compile(), which is now just a
2846        wrapper). This facility is provided because some applications need a
2847        numeric error indication, but it has also enabled me to tidy up the way
2848        compile-time errors are handled in the POSIX wrapper.
2849    
2850    16. Added VPATH=.libs to the makefile; this should help when building with one
2851        prefix path and installing with another. (Or so I'm told by someone who
2852        knows more about this stuff than I do.)
2853    
2854    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
2855        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
2856        match everything, including newlines. This is not POSIX-compatible, but
2857        somebody wanted the feature. From pcretest it can be activated by using
2858        both the P and the s flags.
2859    
2860    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
2861    
2862    19. libpcre.pc was being incorrectly installed as executable.
2863    
2864    20. A couple of places in pcretest check for end-of-line by looking for '\n';
2865        it now also looks for '\r' so that it will work unmodified on Windows.
2866    
2867    21. Added Google's contributed C++ wrapper to the distribution.
2868    
2869    22. Added some untidy missing memory free() calls in pcretest, to keep
2870        Electric Fence happy when testing.
2871    
2872    
2873    
2874    Version 5.0 13-Sep-04
2875    ---------------------
2876    
2877     1. Internal change: literal characters are no longer packed up into items
2878        containing multiple characters in a single byte-string. Each character
2879        is now matched using a separate opcode. However, there may be more than one
2880        byte in the character in UTF-8 mode.
2881    
2882     2. The pcre_callout_block structure has two new fields: pattern_position and
2883        next_item_length. These contain the offset in the pattern to the next match
2884        item, and its length, respectively.
2885    
2886     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
2887        insertion of callouts before each pattern item. Added the /C option to
2888        pcretest to make use of this.
2889    
2890     4. On the advice of a Windows user, the lines
2891    
2892          #if defined(_WIN32) || defined(WIN32)
2893          _setmode( _fileno( stdout ), 0x8000 );
2894          #endif  /* defined(_WIN32) || defined(WIN32) */
2895    
2896        have been added to the source of pcretest. This apparently does useful
2897        magic in relation to line terminators.
2898    
2899     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
2900        for the benefit of those environments where the "b" makes a difference.
2901    
2902     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
2903        to know about it. I have put a hack into configure.in that adds in code
2904        to set GCC=yes if CC=icc. This seems to end up at a point in the
2905        generated configure script that is early enough to affect the setting of
2906        compiler options, which is what is needed, but I have no means of testing
2907        whether it really works. (The user who reported this had patched the
2908        generated configure script, which of course I cannot do.)
2909    
2910        LATER: After change 22 below (new libtool files), the configure script
2911        seems to know about icc (and also ecc). Therefore, I have commented out
2912        this hack in configure.in.
2913    
2914     7. Added support for pkg-config (2 patches were sent in).
2915    
2916     8. Negated POSIX character classes that used a combination of internal tables
2917        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
2918        [[:^ascii]]. Typically, they would match almost any characters. The other
2919        POSIX classes were not broken in this way.
2920    
2921     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
2922        to find the match, as PCRE was deluded into thinking that the match had to
2923        start at the start point or following a newline. The same bug applied to
2924        patterns with negative forward assertions or any backward assertions
2925        preceding ".*" at the start, unless the pattern required a fixed first
2926        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
2927    
2928    10. In UTF-8 mode, when moving forwards in the subject after a failed match
2929        starting at the last subject character, bytes beyond the end of the subject
2930        string were read.
2931    
2932    11. Renamed the variable "class" as "classbits" to make life easier for C++
2933        users. (Previously there was a macro definition, but it apparently wasn't
2934        enough.)
2935    
2936    12. Added the new field "tables" to the extra data so that tables can be passed
2937        in at exec time, or the internal tables can be re-selected. This allows
2938        a compiled regex to be saved and re-used at a later time by a different
2939        program that might have everything at different addresses.
2940    
2941    13. Modified the pcre-config script so that, when run on Solaris, it shows a
2942        -R library as well as a -L library.
2943    
2944    14. The debugging options of pcretest (-d on the command line or D on a
2945        pattern) showed incorrect output for anything following an extended class
2946        that contained multibyte characters and which was followed by a quantifier.
2947    
2948    15. Added optional support for general category Unicode character properties
2949        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
2950        support. It adds about 90K to the size of the library. The meanings of the
2951        inbuilt class escapes such as \d and \s have NOT been changed.
2952    
2953    16. Updated pcredemo.c to include calls to free() to release the memory for the
2954        compiled pattern.
2955    
2956    17. The generated file chartables.c was being created in the source directory
2957        instead of in the building directory. This caused the build to fail if the
2958        source directory was different from the building directory, and was
2959        read-only.
2960    
2961    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
2962        file. No doubt somebody will tell me if they don't make sense... Also added
2963        Dan Mooney's comments about building on OpenVMS.
2964    
2965    19. Added support for partial matching via the PCRE_PARTIAL option for
2966        pcre_exec() and the \P data escape in pcretest.
2967    
2968    20. Extended pcretest with 3 new pattern features:
2969    
2970        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
2971              write the compiled pattern to the file whose name is "rest-of-line".
2972              This is a straight binary dump of the data, with the saved pointer to
2973              the character tables forced to be NULL. The study data, if any, is
2974              written too. After writing, pcretest reads a new pattern.
2975    
2976        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
2977              compiled pattern from the given file. There must not be any
2978              occurrences of "<" in the file name (pretty unlikely); if there are,
2979              pcretest will instead treat the initial "<" as a pattern delimiter.
2980              After reading in the pattern, pcretest goes on to read data lines as
2981              usual.
2982    
2983        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
2984              and 16-bit fields in a compiled pattern, to simulate a pattern that
2985              was compiled on a host of opposite endianness.
2986    
2987    21. The pcre-exec() function can now cope with patterns that were compiled on
2988        hosts of opposite endianness, with this restriction:
2989    
2990          As for any compiled expression that is saved and used later, the tables
2991          pointer field cannot be preserved; the extra_data field in the arguments
2992          to pcre_exec() should be used to pass in a tables address if a value
2993          other than the default internal tables were used at compile time.
2994    
2995    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
2996        now diagnosed as an error. Previously, most of the time, a negative number
2997        would have been treated as zero, but if in addition "ovector" was passed as
2998        NULL, a crash could occur.
2999    
3000    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
3001        new versions from the libtool 1.5 distribution (the last one is a copy of
3002        a file called libtool.m4). This seems to have fixed the need to patch
3003        "configure" to support Darwin 1.3 (which I used to do). However, I still
3004        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
3005        workstation).
3006    
3007    24. Changed the PCRE licence to be the more standard "BSD" licence.
3008    
3009    
3010    Version 4.5 01-Dec-03
3011    ---------------------
3012    
3013     1. There has been some re-arrangement of the code for the match() function so
3014        that it can be compiled in a version that does not call itself recursively.
3015        Instead, it keeps those local variables that need separate instances for
3016        each "recursion" in a frame on the heap, and gets/frees frames whenever it
3017        needs to "recurse". Keeping track of where control must go is done by means
3018        of setjmp/longjmp. The whole thing is implemented by a set of macros that
3019        hide most of the details from the main code, and operates only if
3020        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
3021        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
3022        operating.
3023    
3024        To make it easier for callers to provide specially tailored get/free
3025        functions for this usage, two new functions, pcre_stack_malloc, and
3026        pcre_stack_free, are used. They are always called in strict stacking order,
3027        and the size of block requested is always the same.
3028    
3029        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
3030        PCRE has been compiled to use the stack or the heap for recursion. The
3031        -C option of pcretest uses this to show which version is compiled.
3032    
3033        A new data escape \S, is added to pcretest; it causes the amounts of store
3034        obtained and freed by both kinds of malloc/free at match time to be added
3035        to the output.
3036    
3037     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
3038        what's available on my current Linux desktop machine.
3039    
3040     3. When matching a UTF-8 string, the test for a valid string at the start has
3041        been extended. If start_offset is not zero, PCRE now checks that it points
3042        to a byte that is the start of a UTF-8 character. If not, it returns
3043        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
3044        this is necessary because there may be backward assertions in the pattern.
3045        When matching the same subject several times, it may save resources to use
3046        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
3047    
3048     4. The code for checking the validity of UTF-8 strings has been tightened so
3049        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
3050        containing "overlong sequences".
3051    
3052     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
3053        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
3054        should have been "&", but it just so happened that all the cases this let
3055        through by mistake were picked up later in the function.
3056    
3057     6. I had used a variable called "isblank" - this is a C99 function, causing
3058        some compilers to warn. To avoid this, I renamed it (as "blankclass").
3059    
3060     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
3061        prompting; (b) run "./pcretest /dev/null" at the start of the test script
3062        so the version is shown; (c) stop "make test" echoing "./RunTest".
3063    
3064     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
3065    
3066     9. The prototype for memmove() for systems that don't have it was using
3067        size_t, but the inclusion of the header that defines size_t was later. I've
3068        moved the #includes for the C headers earlier to avoid this.
3069    
3070    10. Added some adjustments to the code to make it easier to compiler on certain
3071        special systems:
3072    
3073          (a) Some "const" qualifiers were missing.
3074          (b) Added the macro EXPORT before all exported functions; by default this
3075              is defined to be empty.
3076          (c) Changed the dftables auxiliary program (that builds chartables.c) so
3077              that it reads its output file name as an argument instead of writing
3078              to the standard output and assuming this can be redirected.
3079    
3080    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
3081        class containing characters with values greater than 255, PCRE compilation
3082        went into a loop.
3083    
3084    12. A recursive reference to a subpattern that was within another subpattern
3085        that had a minimum quantifier of zero caused PCRE to crash. For example,
3086        (x(y(?2))z)? provoked this bug with a subject that got as far as the
3087        recursion. If the recursively-called subpattern itself had a zero repeat,
3088        that was OK.
3089    
3090    13. In pcretest, the buffer for reading a data line was set at 30K, but the
3091        buffer into which it was copied (for escape processing) was still set at
3092        1024, so long lines caused crashes.
3093    
3094    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
3095        "internal error: code overflow...". This applied to any character class
3096        that was followed by a possessive quantifier.
3097    
3098    15. Modified the Makefile to add libpcre.la as a prerequisite for
3099        libpcreposix.la because I was told this is needed for a parallel build to
3100        work.
3101    
3102    16. If a pattern that contained .* following optional items at the start was
3103        studied, the wrong optimizing data was generated, leading to matching
3104        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
3105        matching string must start with a or b or c. The correct conclusion for
3106        this pattern is that a match can start with any character.
3107    
3108    
3109    Version 4.4 13-Aug-03
3110    ---------------------
3111    
3112     1. In UTF-8 mode, a character class containing characters with values between
3113        127 and 255 was not handled correctly if the compiled pattern was studied.
3114        In fixing this, I have also improved the studying algorithm for such
3115        classes (slightly).
3116    
3117     2. Three internal functions had redundant arguments passed to them. Removal
3118        might give a very teeny performance improvement.
3119    
3120     3. Documentation bug: the value of the capture_top field in a callout is *one
3121        more than* the number of the hightest numbered captured substring.
3122    
3123     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
3124        in incorrectly linking with a previously installed version. They now link
3125        explicitly with libpcre.la.
3126    
3127     5. configure.in no longer needs to recognize Cygwin specially.
3128    
3129     6. A problem in pcre.in for Windows platforms is fixed.
3130    
3131     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
3132        pcretest, it used to include the size of the study block as part of its
3133        output. Unfortunately, the structure contains a field that has a different
3134        size on different hardware architectures. This meant that the tests that
3135        showed this size failed. As the block is currently always of a fixed size,
3136        this information isn't actually particularly useful in pcretest output, so
3137        I have just removed it.
3138    
3139     8. Three pre-processor statements accidentally did not start in column 1.
3140        Sadly, there are *still* compilers around that complain, even though
3141        standard C has not required this for well over a decade. Sigh.
3142    
3143     9. In pcretest, the code for checking callouts passed small integers in the
3144        callout_data field, which is a void * field. However, some picky compilers
3145        complained about the casts involved for this on 64-bit systems. Now
3146        pcretest passes the address of the small integer instead, which should get
3147        rid of the warnings.
3148    
3149    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
3150        both compile and run time, and gives an error if an invalid UTF-8 sequence
3151        is found. There is a option for disabling this check in cases where the
3152        string is known to be correct and/or the maximum performance is wanted.
3153    
3154    11. In response to a bug report, I changed one line in Makefile.in from
3155    
3156            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
3157        to
3158            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
3159    
3160        to look similar to other lines, but I have no way of telling whether this
3161        is the right thing to do, as I do not use Windows. No doubt I'll get told
3162        if it's wrong...
3163    
3164    
3165    Version 4.3 21-May-03
3166    ---------------------
3167    
3168    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
3169       Makefile.
3170    
3171    2. Some refactoring to improve the quality of the code:
3172    
3173       (i)   The utf8_table... variables are now declared "const".
3174    
3175       (ii)  The code for \cx, which used the "case flipping" table to upper case
3176             lower case letters, now just substracts 32. This is ASCII-specific,
3177             but the whole concept of \cx is ASCII-specific, so it seems
3178             reasonable.
3179    
3180       (iii) PCRE was using its character types table to recognize decimal and
3181             hexadecimal digits in the pattern. This is silly, because it handles
3182             only 0-9, a-f, and A-F, but the character types table is locale-
3183             specific, which means strange things might happen. A private
3184             table is now used for this - though it costs 256 bytes, a table is
3185             much faster than multiple explicit tests. Of course, the standard
3186             character types table is still used for matching digits in subject
3187             strings against \d.
3188    
3189       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
3190             ending in _t are). So I've renamed it as ESC_tee.
3191    
3192    3. The first argument for regexec() in the POSIX wrapper should have been
3193       defined as "const".
3194    
3195    4. Changed pcretest to use malloc() for its buffers so that they can be
3196       Electric Fenced for debugging.
3197    
3198    5. There were several places in the code where, in UTF-8 mode, PCRE would try
3199       to read one or more bytes before the start of the subject string. Often this
3200       had no effect on PCRE's behaviour, but in some circumstances it could
3201       provoke a segmentation fault.
3202    
3203    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
3204       to try to read one or more bytes before the start of the subject string.
3205    
3206    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
3207       UTF-8 support could misbehave in various ways if the subject string
3208       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
3209       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
3210       back over UTF-8 characters.)
3211    
3212    
3213  Version 4.2 14-Apr-03  Version 4.2 14-Apr-03
3214  ---------------------  ---------------------
3215    

Legend:
Removed from v.67  
changed lines
  Added in v.734

  ViewVC Help
Powered by ViewVC 1.1.5