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

Diff of /code/trunk/ChangeLog

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

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

Legend:
Removed from v.27  
changed lines
  Added in v.193

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12