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

Diff of /code/trunk/ChangeLog

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

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

Legend:
Removed from v.57  
changed lines
  Added in v.159

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12