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

Diff of /code/trunk/ChangeLog

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

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

Legend:
Removed from v.65  
changed lines
  Added in v.190

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12