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

Diff of /code/trunk/ChangeLog

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

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

Legend:
Removed from v.55  
changed lines
  Added in v.179

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12