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

Diff of /code/trunk/ChangeLog

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

revision 73 by nigel, Sat Feb 24 21:40:30 2007 UTC revision 264 by ph10, Tue Nov 13 11:07:16 2007 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4    Version 7.5 12-Nov-07
5    ---------------------
6    
7    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
8        values in parens when parsing an RE using the C++ wrapper."
9    
10    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
11        Characters greater than 255 were excluded from the class instead of being
12        included.
13    
14    
15    Version 7.4 21-Sep-07
16    ---------------------
17    
18    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
19        means that a class such as [\s] counted as "explicit reference to CR or
20        LF". That isn't really right - the whole point of the change was to try to
21        help when there was an actual mention of one of the two characters. So now
22        the change happens only if \r or \n (or a literal CR or LF) character is
23        encountered.
24    
25    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
26        of both had grown to the point where there were only 3 bits left.
27        Fortunately, there was spare space in the data structure, and so I have
28        moved the internal flags into a new 16-bit field to free up more option
29        bits.
30    
31    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
32        but did not set the internal JCHANGED flag - either of these is enough to
33        control the way the "get" function works - but the PCRE_INFO_JCHANGED
34        facility is supposed to tell if (?J) was ever used, so now (?J) at the
35        start sets both bits.
36    
37    4.  Added options (at build time, compile time, exec time) to change \R from
38        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
39    
40    5.  doc/pcresyntax.html was missing from the distribution.
41    
42    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
43        compatibility, even though it is no longer used.
44    
45    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
46        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
47        windows.h file is present (where different names are used). [This was
48        reversed later after testing - see 16 below.]
49    
50    8.  Changed all #include <config.h> to #include "config.h". There were also
51        some further <pcre.h> cases that I changed to "pcre.h".
52    
53    9.  When pcregrep was used with the --colour option, it missed the line ending
54        sequence off the lines that it output.
55    
56    10. It was pointed out to me that arrays of string pointers cause lots of
57        relocations when a shared library is dynamically loaded. A technique of
58        using a single long string with a table of offsets can drastically reduce
59        these. I have refactored PCRE in four places to do this. The result is
60        dramatic:
61    
62          Originally:                          290
63          After changing UCP table:            187
64          After changing error message table:   43
65          After changing table of "verbs"       36
66          After changing table of Posix names   22
67    
68        Thanks to the folks working on Gregex for glib for this insight.
69    
70    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
71        unicode-properties was also set.
72    
73    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
74    
75    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
76        checked only for CRLF.
77    
78    14. Added casts to pcretest.c to avoid compiler warnings.
79    
80    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
81    
82    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
83        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
84        entirely. This removes changes made in 7 above.
85    
86    17. The CMake files have been updated, and there is now more information about
87        building with CMake in the NON-UNIX-USE document.
88    
89    
90    Version 7.3 28-Aug-07
91    ---------------------
92    
93     1. In the rejigging of the build system that eventually resulted in 7.1, the
94        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
95        brackets there is not right, since it causes compilers to look for an
96        installed pcre.h, not the version that is in the source that is being
97        compiled (which of course may be different). I have changed it back to:
98    
99          #include "pcre.h"
100    
101        I have a vague recollection that the change was concerned with compiling in
102        different directories, but in the new build system, that is taken care of
103        by the VPATH setting the Makefile.
104    
105     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
106        when the subject happened to end in the byte 0x85 (e.g. if the last
107        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
108        characters but of course it shouldn't be taken as a newline when it is part
109        of another character. The bug was that, for an unlimited repeat of . in
110        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
111        characters when looking for a newline.
112    
113     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
114    
115     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
116        in debug output.
117    
118     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
119        long printing in the pcrecpp unittest when running under MinGW.
120    
121     6. ESC_K was left out of the EBCDIC table.
122    
123     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
124        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
125        limit also applies to "virtual nesting" when a pattern is recursive, and in
126        this case 1000 isn't so big. I have been able to remove this limit at the
127        expense of backing off one optimization in certain circumstances. Normally,
128        when pcre_exec() would call its internal match() function recursively and
129        immediately return the result unconditionally, it uses a "tail recursion"
130        feature to save stack. However, when a subpattern that can match an empty
131        string has an unlimited repetition quantifier, it no longer makes this
132        optimization. That gives it a stack frame in which to save the data for
133        checking that an empty string has been matched. Previously this was taken
134        from the 1000-entry workspace that had been reserved. So now there is no
135        explicit limit, but more stack is used.
136    
137     8. Applied Daniel's patches to solve problems with the import/export magic
138        syntax that is required for Windows, and which was going wrong for the
139        pcreposix and pcrecpp parts of the library. These were overlooked when this
140        problem was solved for the main library.
141    
142     9. There were some crude static tests to avoid integer overflow when computing
143        the size of patterns that contain repeated groups with explicit upper
144        limits. As the maximum quantifier is 65535, the maximum group length was
145        set at 30,000 so that the product of these two numbers did not overflow a
146        32-bit integer. However, it turns out that people want to use groups that
147        are longer than 30,000 bytes (though not repeat them that many times).
148        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
149        made it possible to implement the integer overflow checks in a much more
150        dynamic way, which I have now done. The artificial limitation on group
151        length has been removed - we now have only the limit on the total length of
152        the compiled pattern, which depends on the LINK_SIZE setting.
153    
154    10. Fixed a bug in the documentation for get/copy named substring when
155        duplicate names are permitted. If none of the named substrings are set, the
156        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
157        empty string.
158    
159    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
160        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
161        because the ] is interpreted as the first data character and the
162        terminating ] is not found. PCRE has been made compatible with Perl in this
163        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
164        cause memory overwriting.
165    
166    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
167        string has been matched (to stop an infinite loop). It was not recognizing
168        a conditional subpattern that could match an empty string if that
169        subpattern was within another subpattern. For example, it looped when
170        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
171        condition was not nested. This bug has been fixed.
172    
173    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
174        past the start of the subject in the presence of bytes with the top bit
175        set, for example "\x8aBCD".
176    
177    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
178        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
179    
180    14. Optimized (?!) to (*FAIL).
181    
182    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
183        This restricts code points to be within the range 0 to 0x10FFFF, excluding
184        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
185        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
186        does: it's just the validity check that is more restrictive.
187    
188    16. Inserted checks for integer overflows during escape sequence (backslash)
189        processing, and also fixed erroneous offset values for syntax errors during
190        backslash processing.
191    
192    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
193        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
194    
195    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
196        caused an overrun.
197    
198    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
199        something other than just ASCII characters) inside a group that had an
200        unlimited repeat caused a loop at compile time (while checking to see
201        whether the group could match an empty string).
202    
203    20. Debugging a pattern containing \p or \P could cause a crash. For example,
204        [\P{Any}] did so. (Error in the code for printing property names.)
205    
206    21. An orphan \E inside a character class could cause a crash.
207    
208    22. A repeated capturing bracket such as (A)? could cause a wild memory
209        reference during compilation.
210    
211    23. There are several functions in pcre_compile() that scan along a compiled
212        expression for various reasons (e.g. to see if it's fixed length for look
213        behind). There were bugs in these functions when a repeated \p or \P was
214        present in the pattern. These operators have additional parameters compared
215        with \d, etc, and these were not being taken into account when moving along
216        the compiled data. Specifically:
217    
218        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
219            length.
220    
221        (b) An item such as \pL+ within a repeated group could cause crashes or
222            loops.
223    
224        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
225            "reference to non-existent subpattern" error.
226    
227        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
228    
229    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
230        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
231    
232    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
233    
234    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
235        character were causing crashes (broken optimization).
236    
237    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
238        \p or \P) caused a compile-time loop.
239    
240    28. More problems have arisen in unanchored patterns when CRLF is a valid line
241        break. For example, the unstudied pattern [\r\n]A does not match the string
242        "\r\nA" because change 7.0/46 below moves the current point on by two
243        characters after failing to match at the start. However, the pattern \nA
244        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
245        the same is true. There doesn't seem any very clean way out of this, but
246        what I have chosen to do makes the common cases work: PCRE now takes note
247        of whether there can be an explicit match for \r or \n anywhere in the
248        pattern, and if so, 7.0/46 no longer applies. As part of this change,
249        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
250        pattern has explicit CR or LF references.
251    
252    29. Added (*CR) etc for changing newline setting at start of pattern.
253    
254    
255    Version 7.2 19-Jun-07
256    ---------------------
257    
258     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
259        which is apparently normally available under Windows.
260    
261     2. Re-jig the pcregrep tests with different newline settings in an attempt
262        to make them independent of the local environment's newline setting.
263    
264     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
265    
266     4. Some of the "internals" tests were previously cut out when the link size
267        was not 2, because the output contained actual offsets. The recent new
268        "Z" feature of pcretest means that these can be cut out, making the tests
269        usable with all link sizes.
270    
271     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
272        stack recursion. This gives a massive performance boost under BSD, but just
273        a small improvement under Linux. However, it saves one field in the frame
274        in all cases.
275    
276     6. Added more features from the forthcoming Perl 5.10:
277    
278        (a) (?-n) (where n is a string of digits) is a relative subroutine or
279            recursion call. It refers to the nth most recently opened parentheses.
280    
281        (b) (?+n) is also a relative subroutine call; it refers to the nth next
282            to be opened parentheses.
283    
284        (c) Conditions that refer to capturing parentheses can be specified
285            relatively, for example, (?(-2)... or (?(+3)...
286    
287        (d) \K resets the start of the current match so that everything before
288            is not part of it.
289    
290        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
291    
292        (f) \g{name} is another synonym - part of Perl 5.10's unification of
293            reference syntax.
294    
295        (g) (?| introduces a group in which the numbering of parentheses in each
296            alternative starts with the same number.
297    
298        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
299    
300     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
301        PCRE_INFO_JCHANGED.
302    
303     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
304        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
305        for detecting groups that can match an empty string.
306    
307     9. A pattern with a very large number of alternatives (more than several
308        hundred) was running out of internal workspace during the pre-compile
309        phase, where pcre_compile() figures out how much memory will be needed. A
310        bit of new cunning has reduced the workspace needed for groups with
311        alternatives. The 1000-alternative test pattern now uses 12 bytes of
312        workspace instead of running out of the 4096 that are available.
313    
314    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
315    
316    11. Applied patch from Google to remove an optimization that didn't quite work.
317        The report of the bug said:
318    
319          pcrecpp::RE("a*").FullMatch("aaa") matches, while
320          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
321          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
322    
323    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
324        it matched the wrong number of bytes.
325    
326    
327    Version 7.1 24-Apr-07
328    ---------------------
329    
330     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
331        that is more "standard", making use of automake and other Autotools. There
332        is some re-arrangement of the files and adjustment of comments consequent
333        on this.
334    
335     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
336        for recursive directory scanning broke on some systems because the files
337        are not scanned in any specific order and on different systems the order
338        was different. A call to "sort" has been inserted into RunGrepTest for the
339        approprate test as a short-term fix. In the longer term there may be an
340        alternative.
341    
342     3. I had an email from Eric Raymond about problems translating some of PCRE's
343        man pages to HTML (despite the fact that I distribute HTML pages, some
344        people do their own conversions for various reasons). The problems
345        concerned the use of low-level troff macros .br and .in. I have therefore
346        removed all such uses from the man pages (some were redundant, some could
347        be replaced by .nf/.fi pairs). The 132html script that I use to generate
348        HTML has been updated to handle .nf/.fi and to complain if it encounters
349        .br or .in.
350    
351     4. Updated comments in configure.ac that get placed in config.h.in and also
352        arranged for config.h to be included in the distribution, with the name
353        config.h.generic, for the benefit of those who have to compile without
354        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
355    
356     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
357        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
358        makevp.bat for the current PCRE, using the additional files
359        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
360    
361     6. A Windows user reported a minor discrepancy with test 2, which turned out
362        to be caused by a trailing space on an input line that had got lost in his
363        copy. The trailing space was an accident, so I've just removed it.
364    
365     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
366        that is needed.
367    
368     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
369        as "const" (a) because they are and (b) because it helps the PHP
370        maintainers who have recently made a script to detect big data structures
371        in the php code that should be moved to the .rodata section. I remembered
372        to update Builducptable as well, so it won't revert if ucptable.h is ever
373        re-created.
374    
375     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
376        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
377        order to be able to cut out the UTF-8 tables in the latter when UTF-8
378        support is not required. This saves 1.5-2K of code, which is important in
379        some applications.
380    
381        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
382        so as not to refer to the tables, even though these functions will never be
383        called when UTF-8 support is disabled. Otherwise there are problems with a
384        shared library.
385    
386    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
387    
388        (a) It was defining its arguments as char * instead of void *.
389    
390        (b) It was assuming that all moves were upwards in memory; this was true
391            a long time ago when I wrote it, but is no longer the case.
392    
393        The emulated memove() is provided for those environments that have neither
394        memmove() nor bcopy(). I didn't think anyone used it these days, but that
395        is clearly not the case, as these two bugs were recently reported.
396    
397    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
398        and Detrail to create the HTML documentation, the .txt form of the man
399        pages, and it removes trailing spaces from listed files. It also creates
400        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
401        case, it wraps all the #defines with #ifndefs. This script should be run
402        before "make dist".
403    
404    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
405        with Unicode property support.
406    
407        (a) For a maximizing quantifier, if the two different cases of the
408            character were of different lengths in their UTF-8 codings (there are
409            some cases like this - I found 11), and the matching function had to
410            back up over a mixture of the two cases, it incorrectly assumed they
411            were both the same length.
412    
413        (b) When PCRE was configured to use the heap rather than the stack for
414            recursion during matching, it was not correctly preserving the data for
415            the other case of a UTF-8 character when checking ahead for a match
416            while processing a minimizing repeat. If the check also involved
417            matching a wide character, but failed, corruption could cause an
418            erroneous result when trying to check for a repeat of the original
419            character.
420    
421    13. Some tidying changes to the testing mechanism:
422    
423        (a) The RunTest script now detects the internal link size and whether there
424            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
425            values substituted by "configure". (The RunGrepTest script already did
426            this for UTF-8.) The configure.ac script no longer substitutes the
427            relevant variables.
428    
429        (b) The debugging options /B and /D in pcretest show the compiled bytecode
430            with length and offset values. This means that the output is different
431            for different internal link sizes. Test 2 is skipped for link sizes
432            other than 2 because of this, bypassing the problem. Unfortunately,
433            there was also a test in test 3 (the locale tests) that used /B and
434            failed for link sizes other than 2. Rather than cut the whole test out,
435            I have added a new /Z option to pcretest that replaces the length and
436            offset values with spaces. This is now used to make test 3 independent
437            of link size. (Test 2 will be tidied up later.)
438    
439    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
440        segmentation fault instead of returning the appropriate error message.
441    
442    15. In multiline mode when the newline sequence was set to "any", the pattern
443        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
444        This doesn't seem right; it now treats the CRLF combination as the line
445        ending, and so does not match in that case. It's only a pattern such as ^$
446        that would hit this one: something like ^ABC$ would have failed after \r
447        and then tried again after \r\n.
448    
449    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
450        in an attempt to make files that differ only in their line terminators
451        compare equal. This works on Linux.
452    
453    17. Under certain error circumstances pcregrep might try to free random memory
454        as it exited. This is now fixed, thanks to valgrind.
455    
456    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
457        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
458        was because its rules for how to advance for /g after matching an empty
459        string at the end of a line did not allow for this case. They now check for
460        it specially.
461    
462    20. pcretest is supposed to handle patterns and data of any length, by
463        extending its buffers when necessary. It was getting this wrong when the
464        buffer for a data line had to be extended.
465    
466    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
467        CRLF as a newline sequence.
468    
469    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
470        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
471        I have nevertheless tidied it up.
472    
473    23. Added some casts to kill warnings from HP-UX ia64 compiler.
474    
475    24. Added a man page for pcre-config.
476    
477    
478    Version 7.0 19-Dec-06
479    ---------------------
480    
481     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
482        moving to gcc 4.1.1.
483    
484     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
485        sys/time.h, which is documented as needed for this function. It doesn't
486        seem to matter on Linux, but it showed up on some releases of OS X.
487    
488     3. It seems that there are systems where bytes whose values are greater than
489        127 match isprint() in the "C" locale. The "C" locale should be the
490        default when a C program starts up. In most systems, only ASCII printing
491        characters match isprint(). This difference caused the output from pcretest
492        to vary, making some of the tests fail. I have changed pcretest so that:
493    
494        (a) When it is outputting text in the compiled version of a pattern, bytes
495            other than 32-126 are always shown as hex escapes.
496    
497        (b) When it is outputting text that is a matched part of a subject string,
498            it does the same, unless a different locale has been set for the match
499            (using the /L modifier). In this case, it uses isprint() to decide.
500    
501     4. Fixed a major bug that caused incorrect computation of the amount of memory
502        required for a compiled pattern when options that changed within the
503        pattern affected the logic of the preliminary scan that determines the
504        length. The relevant options are -x, and -i in UTF-8 mode. The result was
505        that the computed length was too small. The symptoms of this bug were
506        either the PCRE error "internal error: code overflow" from pcre_compile(),
507        or a glibc crash with a message such as "pcretest: free(): invalid next
508        size (fast)". Examples of patterns that provoked this bug (shown in
509        pcretest format) are:
510    
511          /(?-x: )/x
512          /(?x)(?-x: \s*#\s*)/
513          /((?i)[\x{c0}])/8
514          /(?i:[\x{c0}])/8
515    
516        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
517        is now done differently.
518    
519     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
520        wrapper classes; (b) implement a new function in the C++ scanner that is
521        more efficient than the old way of doing things because it avoids levels of
522        recursion in the regex matching; (c) add a paragraph to the documentation
523        for the FullMatch() function.
524    
525     6. The escape sequence \n was being treated as whatever was defined as
526        "newline". Not only was this contrary to the documentation, which states
527        that \n is character 10 (hex 0A), but it also went horribly wrong when
528        "newline" was defined as CRLF. This has been fixed.
529    
530     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
531        was being set to -1 for the "end of line" case (supposedly a value that no
532        character can have). Though this value is never used (the check for end of
533        line is "zero bytes in current character"), it caused compiler complaints.
534        I've changed it to 0xffffffff.
535    
536     8. In pcre_version.c, the version string was being built by a sequence of
537        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
538        string (as it is for production releases) called a macro with an empty
539        argument. The C standard says the result of this is undefined. The gcc
540        compiler treats it as an empty string (which was what was wanted) but it is
541        reported that Visual C gives an error. The source has been hacked around to
542        avoid this problem.
543    
544     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
545        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
546        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
547        of them did).
548    
549    10. Originally, pcretest opened its input and output without "b"; then I was
550        told that "b" was needed in some environments, so it was added for release
551        5.0 to both the input and output. (It makes no difference on Unix-like
552        systems.) Later I was told that it is wrong for the input on Windows. I've
553        now abstracted the modes into two macros, to make it easier to fiddle with
554        them, and removed "b" from the input mode under Windows.
555    
556    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
557    
558    12. Added -help and --help to pcretest as an official way of being reminded
559        of the options.
560    
561    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
562        and pcrecpp.cc because they annoy compilers at high warning levels.
563    
564    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
565    
566    15. Fixed an occurrence of == in configure.ac that should have been = (shell
567        scripts are not C programs :-) and which was not noticed because it works
568        on Linux.
569    
570    16. pcretest is supposed to handle any length of pattern and data line (as one
571        line or as a continued sequence of lines) by extending its input buffer if
572        necessary. This feature was broken for very long pattern lines, leading to
573        a string of junk being passed to pcre_compile() if the pattern was longer
574        than about 50K.
575    
576    17. I have done a major re-factoring of the way pcre_compile() computes the
577        amount of memory needed for a compiled pattern. Previously, there was code
578        that made a preliminary scan of the pattern in order to do this. That was
579        OK when PCRE was new, but as the facilities have expanded, it has become
580        harder and harder to keep it in step with the real compile phase, and there
581        have been a number of bugs (see for example, 4 above). I have now found a
582        cunning way of running the real compile function in a "fake" mode that
583        enables it to compute how much memory it would need, while actually only
584        ever using a few hundred bytes of working memory and without too many
585        tests of the mode. This should make future maintenance and development
586        easier. A side effect of this work is that the limit of 200 on the nesting
587        depth of parentheses has been removed (though this was never a serious
588        limitation, I suspect). However, there is a downside: pcre_compile() now
589        runs more slowly than before (30% or more, depending on the pattern). I
590        hope this isn't a big issue. There is no effect on runtime performance.
591    
592    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
593        newline (only possible for the last line of a file) and it was a
594        pattern that set a locale (followed by /Lsomething), pcretest crashed.
595    
596    19. Added additional timing features to pcretest. (1) The -tm option now times
597        matching only, not compiling. (2) Both -t and -tm can be followed, as a
598        separate command line item, by a number that specifies the number of
599        repeats to use when timing. The default is 50000; this gives better
600        precision, but takes uncomfortably long for very large patterns.
601    
602    20. Extended pcre_study() to be more clever in cases where a branch of a
603        subpattern has no definite first character. For example, (a*|b*)[cd] would
604        previously give no result from pcre_study(). Now it recognizes that the
605        first character must be a, b, c, or d.
606    
607    21. There was an incorrect error "recursive call could loop indefinitely" if
608        a subpattern (or the entire pattern) that was being tested for matching an
609        empty string contained only one non-empty item after a nested subpattern.
610        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
611        incorrectly, because the \d was being skipped in the check.
612    
613    22. The pcretest program now has a new pattern option /B and a command line
614        option -b, which is equivalent to adding /B to every pattern. This causes
615        it to show the compiled bytecode, without the additional information that
616        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
617        is the same as /B/I).
618    
619    23. A new optimization is now able automatically to treat some sequences such
620        as a*b as a*+b. More specifically, if something simple (such as a character
621        or a simple class like \d) has an unlimited quantifier, and is followed by
622        something that cannot possibly match the quantified thing, the quantifier
623        is automatically "possessified".
624    
625    24. A recursive reference to a subpattern whose number was greater than 39
626        went wrong under certain circumstances in UTF-8 mode. This bug could also
627        have affected the operation of pcre_study().
628    
629    25. Realized that a little bit of performance could be had by replacing
630        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
631    
632    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
633    
634    27. Possessive quantifiers such as a++ were previously implemented by turning
635        them into atomic groups such as ($>a+). Now they have their own opcodes,
636        which improves performance. This includes the automatically created ones
637        from 23 above.
638    
639    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
640        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
641        the first matched character to be a colon. This applied both to named and
642        numbered groups.
643    
644    29. The ucpinternal.h header file was missing its idempotency #ifdef.
645    
646    30. I was sent a "project" file called libpcre.a.dev which I understand makes
647        building PCRE on Windows easier, so I have included it in the distribution.
648    
649    31. There is now a check in pcretest against a ridiculously large number being
650        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
651        loop, the loop is abandoned.
652    
653    32. Forward references to subpatterns in conditions such as (?(2)...) where
654        subpattern 2 is defined later cause pcre_compile() to search forwards in
655        the pattern for the relevant set of parentheses. This search went wrong
656        when there were unescaped parentheses in a character class, parentheses
657        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
658    
659    33. "Subroutine" calls and backreferences were previously restricted to
660        referencing subpatterns earlier in the regex. This restriction has now
661        been removed.
662    
663    34. Added a number of extra features that are going to be in Perl 5.10. On the
664        whole, these are just syntactic alternatives for features that PCRE had
665        previously implemented using the Python syntax or my own invention. The
666        other formats are all retained for compatibility.
667    
668        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
669            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
670            also .NET compatible.
671    
672        (b) A recursion or subroutine call to a named group can now be defined as
673            (?&name) as well as (?P>name).
674    
675        (c) A backreference to a named group can now be defined as \k<name> or
676            \k'name' as well as (?P=name). The new forms, as well as being in Perl
677            5.10, are also .NET compatible.
678    
679        (d) A conditional reference to a named group can now use the syntax
680            (?(<name>) or (?('name') as well as (?(name).
681    
682        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
683            groups (named and numbered) that are never evaluated inline, but can be
684            called as "subroutines" from elsewhere. In effect, the DEFINE condition
685            is always false. There may be only one alternative in such a group.
686    
687        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
688            as the simple (?(R). The condition is true only if the most recent
689            recursion is that of the given number or name. It does not search out
690            through the entire recursion stack.
691    
692        (g) The escape \gN or \g{N} has been added, where N is a positive or
693            negative number, specifying an absolute or relative reference.
694    
695    35. Tidied to get rid of some further signed/unsigned compiler warnings and
696        some "unreachable code" warnings.
697    
698    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
699        things, this adds five new scripts.
700    
701    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
702        There were also incompatibilities regarding the handling of \Q..\E inside
703        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
704        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
705    
706    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
707        matches an empty string, and forcibly breaks the loop. There were bugs in
708        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
709        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
710        separate and independent bugs (that affected different cases) have been
711        fixed.
712    
713    39. Refactored the code to abolish the use of different opcodes for small
714        capturing bracket numbers. This is a tidy that I avoided doing when I
715        removed the limit on the number of capturing brackets for 3.5 back in 2001.
716        The new approach is not only tidier, it makes it possible to reduce the
717        memory needed to fix the previous bug (38).
718    
719    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
720        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
721        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
722        mode.
723    
724    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
725        report.
726    
727    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
728        copy construction and assignment in the C++ wrapper.
729    
730    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
731        couple of bugs that could have given wrong results in the "--newline=crlf"
732        case.
733    
734    44. Added a number of casts and did some reorganization of signed/unsigned int
735        variables following suggestions from Dair Grant. Also renamed the variable
736        "this" as "item" because it is a C++ keyword.
737    
738    45. Arranged for dftables to add
739    
740          #include "pcre_internal.h"
741    
742        to pcre_chartables.c because without it, gcc 4.x may remove the array
743        definition from the final binary if PCRE is built into a static library and
744        dead code stripping is activated.
745    
746    46. For an unanchored pattern, if a match attempt fails at the start of a
747        newline sequence, and the newline setting is CRLF or ANY, and the next two
748        characters are CRLF, advance by two characters instead of one.
749    
750    
751    Version 6.7 04-Jul-06
752    ---------------------
753    
754     1. In order to handle tests when input lines are enormously long, pcretest has
755        been re-factored so that it automatically extends its buffers when
756        necessary. The code is crude, but this _is_ just a test program. The
757        default size has been increased from 32K to 50K.
758    
759     2. The code in pcre_study() was using the value of the re argument before
760        testing it for NULL. (Of course, in any sensible call of the function, it
761        won't be NULL.)
762    
763     3. The memmove() emulation function in pcre_internal.h, which is used on
764        systems that lack both memmove() and bcopy() - that is, hardly ever -
765        was missing a "static" storage class specifier.
766    
767     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
768        containing an extended class (one that cannot be represented by a bitmap
769        because it contains high-valued characters or Unicode property items, e.g.
770        [\pZ]). Almost always one would set UTF-8 mode when processing such a
771        pattern, but PCRE should not loop if you do not (it no longer does).
772        [Detail: two cases were found: (a) a repeated subpattern containing an
773        extended class; (b) a recursive reference to a subpattern that followed a
774        previous extended class. It wasn't skipping over the extended class
775        correctly when UTF-8 mode was not set.]
776    
777     5. A negated single-character class was not being recognized as fixed-length
778        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
779        compile error "lookbehind assertion is not fixed length".
780    
781     6. The RunPerlTest auxiliary script was showing an unexpected difference
782        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
783        write a Perl script that can interpret lines of an input file either as
784        byte characters or as UTF-8, which is what "perltest" was being required to
785        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
786        can't do is switch easily at run time between having the "use utf8;" pragma
787        or not. In the end, I fudged it by using the RunPerlTest script to insert
788        "use utf8;" explicitly for the UTF-8 tests.
789    
790     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
791        the end of the subject string, contrary to the documentation and to what
792        Perl does. This was true of both matching functions. Now it matches only at
793        the start of the subject and immediately after *internal* newlines.
794    
795     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
796        a pointer to an int instead of a pointer to an unsigned long int. This
797        caused problems on 64-bit systems.
798    
799     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
800        instance of the 'standard' template library not being so standard".
801    
802    10. There was no check on the number of named subpatterns nor the maximum
803        length of a subpattern name. The product of these values is used to compute
804        the size of the memory block for a compiled pattern. By supplying a very
805        long subpattern name and a large number of named subpatterns, the size
806        computation could be caused to overflow. This is now prevented by limiting
807        the length of names to 32 characters, and the number of named subpatterns
808        to 10,000.
809    
810    11. Subpatterns that are repeated with specific counts have to be replicated in
811        the compiled pattern. The size of memory for this was computed from the
812        length of the subpattern and the repeat count. The latter is limited to
813        65535, but there was no limit on the former, meaning that integer overflow
814        could in principle occur. The compiled length of a repeated subpattern is
815        now limited to 30,000 bytes in order to prevent this.
816    
817    12. Added the optional facility to have named substrings with the same name.
818    
819    13. Added the ability to use a named substring as a condition, using the
820        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
821        are numbers (not recommended). Forward references are permitted.
822    
823    14. Added forward references in named backreferences (if you see what I mean).
824    
825    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
826        pattern could run off the end of the subject. For example, the pattern
827        "(?s)(.{1,5})"8 did this with the subject "ab".
828    
829    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
830        PCRE_CASELESS was set when matching characters that were quantified with ?
831        or *.
832    
833    17. A character class other than a single negated character that had a minimum
834        but no maximum quantifier - for example [ab]{6,} - was not handled
835        correctly by pce_dfa_exec(). It would match only one character.
836    
837    18. A valid (though odd) pattern that looked like a POSIX character
838        class but used an invalid character after [ (for example [[,abc,]]) caused
839        pcre_compile() to give the error "Failed: internal error: code overflow" or
840        in some cases to crash with a glibc free() error. This could even happen if
841        the pattern terminated after [[ but there just happened to be a sequence of
842        letters, a binary zero, and a closing ] in the memory that followed.
843    
844    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
845        over the years. Originally (before any Unicode support), just the bottom 8
846        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
847        output from "man perlunicode" includes this:
848    
849          The regular expression compiler produces polymorphic opcodes.  That
850          is, the pattern adapts to the data and automatically switches to
851          the Unicode character scheme when presented with Unicode data--or
852          instead uses a traditional byte scheme when presented with byte
853          data.
854    
855        Sadly, a wide octal escape does not cause a switch, and in a string with
856        no other multibyte characters, these octal escapes are treated as before.
857        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
858        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
859        Unicode string.
860    
861        I have not perpetrated such confusion in PCRE. Up till now, it took just
862        the bottom 8 bits, as in old Perl. I have now made octal escapes with
863        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
864        translate to the appropriate multibyte character.
865    
866    29. Applied some refactoring to reduce the number of warnings from Microsoft
867        and Borland compilers. This has included removing the fudge introduced
868        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
869        a warning about an unused variable.
870    
871    21. PCRE has not included VT (character 0x0b) in the set of whitespace
872        characters since release 4.0, because Perl (from release 5.004) does not.
873        [Or at least, is documented not to: some releases seem to be in conflict
874        with the documentation.] However, when a pattern was studied with
875        pcre_study() and all its branches started with \s, PCRE still included VT
876        as a possible starting character. Of course, this did no harm; it just
877        caused an unnecessary match attempt.
878    
879    22. Removed a now-redundant internal flag bit that recorded the fact that case
880        dependency changed within the pattern. This was once needed for "required
881        byte" processing, but is no longer used. This recovers a now-scarce options
882        bit. Also moved the least significant internal flag bit to the most-
883        significant bit of the word, which was not previously used (hangover from
884        the days when it was an int rather than a uint) to free up another bit for
885        the future.
886    
887    23. Added support for CRLF line endings as well as CR and LF. As well as the
888        default being selectable at build time, it can now be changed at runtime
889        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
890        specify that it is scanning data with non-default line endings.
891    
892    24. Changed the definition of CXXLINK to make it agree with the definition of
893        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
894    
895    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
896        recursions. This makes a big different to stack usage for some patterns.
897    
898    26. If a subpattern containing a named recursion or subroutine reference such
899        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
900        the space required for the compiled pattern went wrong and gave too small a
901        value. Depending on the environment, this could lead to "Failed: internal
902        error: code overflow at offset 49" or "glibc detected double free or
903        corruption" errors.
904    
905    27. Applied patches from Google (a) to support the new newline modes and (b) to
906        advance over multibyte UTF-8 characters in GlobalReplace.
907    
908    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
909        difference for some implementation of PCRE in some Windows version.
910    
911    29. Added some extra testing facilities to pcretest:
912    
913        \q<number>   in a data line sets the "match limit" value
914        \Q<number>   in a data line sets the "match recursion limt" value
915        -S <number>  sets the stack size, where <number> is in megabytes
916    
917        The -S option isn't available for Windows.
918    
919    
920    Version 6.6 06-Feb-06
921    ---------------------
922    
923     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
924        in pcreposix.h. I have copied the definition from pcre.h.
925    
926     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
927        because pcre.h is no longer a built file.
928    
929     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
930        not normally included in the compiled code.
931    
932    
933    Version 6.5 01-Feb-06
934    ---------------------
935    
936     1. When using the partial match feature with pcre_dfa_exec(), it was not
937        anchoring the second and subsequent partial matches at the new starting
938        point. This could lead to incorrect results. For example, with the pattern
939        /1234/, partially matching against "123" and then "a4" gave a match.
940    
941     2. Changes to pcregrep:
942    
943        (a) All non-match returns from pcre_exec() were being treated as failures
944            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
945            error message is output. Some extra information is given for the
946            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
947            probably the only errors that are likely to be caused by users (by
948            specifying a regex that has nested indefinite repeats, for instance).
949            If there are more than 20 of these errors, pcregrep is abandoned.
950    
951        (b) A binary zero was treated as data while matching, but terminated the
952            output line if it was written out. This has been fixed: binary zeroes
953            are now no different to any other data bytes.
954    
955        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
956            used to set a locale for matching. The --locale=xxxx long option has
957            been added (no short equivalent) to specify a locale explicitly on the
958            pcregrep command, overriding the environment variables.
959    
960        (d) When -B was used with -n, some line numbers in the output were one less
961            than they should have been.
962    
963        (e) Added the -o (--only-matching) option.
964    
965        (f) If -A or -C was used with -c (count only), some lines of context were
966            accidentally printed for the final match.
967    
968        (g) Added the -H (--with-filename) option.
969    
970        (h) The combination of options -rh failed to suppress file names for files
971            that were found from directory arguments.
972    
973        (i) Added the -D (--devices) and -d (--directories) options.
974    
975        (j) Added the -F (--fixed-strings) option.
976    
977        (k) Allow "-" to be used as a file name for -f as well as for a data file.
978    
979        (l) Added the --colo(u)r option.
980    
981        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
982            is not present by default.
983    
984     3. A nasty bug was discovered in the handling of recursive patterns, that is,
985        items such as (?R) or (?1), when the recursion could match a number of
986        alternatives. If it matched one of the alternatives, but subsequently,
987        outside the recursion, there was a failure, the code tried to back up into
988        the recursion. However, because of the way PCRE is implemented, this is not
989        possible, and the result was an incorrect result from the match.
990    
991        In order to prevent this happening, the specification of recursion has
992        been changed so that all such subpatterns are automatically treated as
993        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
994    
995     4. I had overlooked the fact that, in some locales, there are characters for
996        which isalpha() is true but neither isupper() nor islower() are true. In
997        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
998        and ordfeminine) are like this. This affected the treatment of \w and \W
999        when they appeared in character classes, but not when they appeared outside
1000        a character class. The bit map for "word" characters is now created
1001        separately from the results of isalnum() instead of just taking it from the
1002        upper, lower, and digit maps. (Plus the underscore character, of course.)
1003    
1004     5. The above bug also affected the handling of POSIX character classes such as
1005        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1006        permanent tables. Instead, the bit maps for such a class were previously
1007        created as the appropriate unions of the upper, lower, and digit bitmaps.
1008        Now they are created by subtraction from the [[:word:]] class, which has
1009        its own bitmap.
1010    
1011     6. The [[:blank:]] character class matches horizontal, but not vertical space.
1012        It is created by subtracting the vertical space characters (\x09, \x0a,
1013        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1014        subtraction was done in the overall bitmap for a character class, meaning
1015        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1016        be recognized. This bug has been fixed.
1017    
1018     7. Patches from the folks at Google:
1019    
1020          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1021          real life, but is still worth protecting against".
1022    
1023          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1024          regular expressions".
1025    
1026          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1027          have it.
1028    
1029          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1030          "configure" and the latter not, in order to fix a problem somebody had
1031          with compiling the Arg class on HP-UX.
1032    
1033          (e) Improve the error-handling of the C++ wrapper a little bit.
1034    
1035          (f) New tests for checking recursion limiting.
1036    
1037     8. The pcre_memmove() function, which is used only if the environment does not
1038        have a standard memmove() function (and is therefore rarely compiled),
1039        contained two bugs: (a) use of int instead of size_t, and (b) it was not
1040        returning a result (though PCRE never actually uses the result).
1041    
1042     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1043        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1044        returned instead of calling malloc() with an overflowing number that would
1045        most likely cause subsequent chaos.
1046    
1047    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1048    
1049    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1050        with this option is matched, the nmatch and pmatch options of regexec() are
1051        ignored.
1052    
1053    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1054        provided in case anyone wants to the the POSIX interface with UTF-8
1055        strings.
1056    
1057    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1058        C++ linking (needed for some HP-UX environments).
1059    
1060    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1061        (unused parameter) and in the pcre_printint() function (omitted "default"
1062        switch label when the default is to do nothing).
1063    
1064    15. Added some code to make it possible, when PCRE is compiled as a C++
1065        library, to replace subject pointers for pcre_exec() with a smart pointer
1066        class, thus making it possible to process discontinuous strings.
1067    
1068    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1069        much the same function. They were added by different people who were trying
1070        to make PCRE easy to compile on non-Unix systems. It has been suggested
1071        that PCRE_EXPORT be abolished now that there is more automatic apparatus
1072        for compiling on Windows systems. I have therefore replaced it with
1073        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1074        defaults to "extern" for C or "extern C" for C++, which works fine on
1075        Unix-like systems. It is now possible to override the value of PCRE_DATA_
1076        SCOPE with something explicit in config.h. In addition:
1077    
1078        (a) pcreposix.h still had just "extern" instead of either of these macros;
1079            I have replaced it with PCRE_DATA_SCOPE.
1080    
1081        (b) Functions such as _pcre_xclass(), which are internal to the library,
1082            but external in the C sense, all had PCRE_EXPORT in their definitions.
1083            This is apparently wrong for the Windows case, so I have removed it.
1084            (It makes no difference on Unix-like systems.)
1085    
1086    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1087        of recursive calls to match(). This is different to MATCH_LIMIT because
1088        that limits the total number of calls to match(), not all of which increase
1089        the depth of recursion. Limiting the recursion depth limits the amount of
1090        stack (or heap if NO_RECURSE is set) that is used. The default can be set
1091        when PCRE is compiled, and changed at run time. A patch from Google adds
1092        this functionality to the C++ interface.
1093    
1094    18. Changes to the handling of Unicode character properties:
1095    
1096        (a) Updated the table to Unicode 4.1.0.
1097    
1098        (b) Recognize characters that are not in the table as "Cn" (undefined).
1099    
1100        (c) I revised the way the table is implemented to a much improved format
1101            which includes recognition of ranges. It now supports the ranges that
1102            are defined in UnicodeData.txt, and it also amalgamates other
1103            characters into ranges. This has reduced the number of entries in the
1104            table from around 16,000 to around 3,000, thus reducing its size
1105            considerably. I realized I did not need to use a tree structure after
1106            all - a binary chop search is just as efficient. Having reduced the
1107            number of entries, I extended their size from 6 bytes to 8 bytes to
1108            allow for more data.
1109    
1110        (d) Added support for Unicode script names via properties such as \p{Han}.
1111    
1112    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1113        matching that character.
1114    
1115    20. When matching a repeated Unicode property with a minimum greater than zero,
1116        (for example \pL{2,}), PCRE could look past the end of the subject if it
1117        reached it while seeking the minimum number of characters. This could
1118        happen only if some of the characters were more than one byte long, because
1119        there is a check for at least the minimum number of bytes.
1120    
1121    21. Refactored the implementation of \p and \P so as to be more general, to
1122        allow for more different types of property in future. This has changed the
1123        compiled form incompatibly. Anybody with saved compiled patterns that use
1124        \p or \P will have to recompile them.
1125    
1126    22. Added "Any" and "L&" to the supported property types.
1127    
1128    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1129        but give a compile time error if the value is greater than 0xff.
1130    
1131    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1132        accidentally not being installed or uninstalled.
1133    
1134    25. The pcre.h file was built from pcre.h.in, but the only changes that were
1135        made were to insert the current release number. This seemed silly, because
1136        it made things harder for people building PCRE on systems that don't run
1137        "configure". I have turned pcre.h into a distributed file, no longer built
1138        by "configure", with the version identification directly included. There is
1139        no longer a pcre.h.in file.
1140    
1141        However, this change necessitated a change to the pcre-config script as
1142        well. It is built from pcre-config.in, and one of the substitutions was the
1143        release number. I have updated configure.ac so that ./configure now finds
1144        the release number by grepping pcre.h.
1145    
1146    26. Added the ability to run the tests under valgrind.
1147    
1148    
1149    Version 6.4 05-Sep-05
1150    ---------------------
1151    
1152     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1153        "--" to be printed when multiple files were scanned, even when none of the
1154        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1155        consider it to be a bug, and have restored the previous behaviour.
1156    
1157     2. A couple of code tidies to get rid of compiler warnings.
1158    
1159     3. The pcretest program used to cheat by referring to symbols in the library
1160        whose names begin with _pcre_. These are internal symbols that are not
1161        really supposed to be visible externally, and in some environments it is
1162        possible to suppress them. The cheating is now confined to including
1163        certain files from the library's source, which is a bit cleaner.
1164    
1165     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1166        file's purpose clearer.
1167    
1168     5. Reorganized pcre_ucp_findchar().
1169    
1170    
1171    Version 6.3 15-Aug-05
1172    ---------------------
1173    
1174     1. The file libpcre.pc.in did not have general read permission in the tarball.
1175    
1176     2. There were some problems when building without C++ support:
1177    
1178        (a) If C++ support was not built, "make install" and "make test" still
1179            tried to test it.
1180    
1181        (b) There were problems when the value of CXX was explicitly set. Some
1182            changes have been made to try to fix these, and ...
1183    
1184        (c) --disable-cpp can now be used to explicitly disable C++ support.
1185    
1186        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1187            backslash in a target when C++ was disabled. This confuses some
1188            versions of "make", apparently. Using an intermediate variable solves
1189            this. (Same for CPP_LOBJ.)
1190    
1191     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1192        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1193        necessary on certain architectures.
1194    
1195     4. Added a setting of -export-symbols-regex to the link command to remove
1196        those symbols that are exported in the C sense, but actually are local
1197        within the library, and not documented. Their names all begin with
1198        "_pcre_". This is not a perfect job, because (a) we have to except some
1199        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1200        available (and never for static libraries). I have made a note to try to
1201        find a way round (a) in the future.
1202    
1203    
1204    Version 6.2 01-Aug-05
1205    ---------------------
1206    
1207     1. There was no test for integer overflow of quantifier values. A construction
1208        such as {1111111111111111} would give undefined results. What is worse, if
1209        a minimum quantifier for a parenthesized subpattern overflowed and became
1210        negative, the calculation of the memory size went wrong. This could have
1211        led to memory overwriting.
1212    
1213     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1214    
1215     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1216        operating environments where this matters.
1217    
1218     4. Applied Giuseppe Maxia's patch to add additional features for controlling
1219        PCRE options from within the C++ wrapper.
1220    
1221     5. Named capturing subpatterns were not being correctly counted when a pattern
1222        was compiled. This caused two problems: (a) If there were more than 100
1223        such subpatterns, the calculation of the memory needed for the whole
1224        compiled pattern went wrong, leading to an overflow error. (b) Numerical
1225        back references of the form \12, where the number was greater than 9, were
1226        not recognized as back references, even though there were sufficient
1227        previous subpatterns.
1228    
1229     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1230        versions of gcc, e.g. 2.95.4.
1231    
1232    
1233    Version 6.1 21-Jun-05
1234    ---------------------
1235    
1236     1. There was one reference to the variable "posix" in pcretest.c that was not
1237        surrounded by "#if !defined NOPOSIX".
1238    
1239     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1240        the cross-check on the old pcre_info() function, for the benefit of the
1241        cut-down version of PCRE that is currently imported into Exim.
1242    
1243     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1244        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1245        patterns that start (?i) but I don't think that matters. The use of (?i) is
1246        just an example; this all applies to the other options as well.
1247    
1248     4. Since libtool seems to echo the compile commands it is issuing, the output
1249        from "make" can be reduced a bit by putting "@" in front of each libtool
1250        compile command.
1251    
1252     5. Patch from the folks at Google for configure.in to be a bit more thorough
1253        in checking for a suitable C++ installation before trying to compile the
1254        C++ stuff. This should fix a reported problem when a compiler was present,
1255        but no suitable headers.
1256    
1257     6. The man pages all had just "PCRE" as their title. I have changed them to
1258        be the relevant file name. I have also arranged that these names are
1259        retained in the file doc/pcre.txt, which is a concatenation in text format
1260        of all the man pages except the little individual ones for each function.
1261    
1262     7. The NON-UNIX-USE file had not been updated for the different set of source
1263        files that come with release 6. I also added a few comments about the C++
1264        wrapper.
1265    
1266    
1267    Version 6.0 07-Jun-05
1268    ---------------------
1269    
1270     1. Some minor internal re-organization to help with my DFA experiments.
1271    
1272     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1273        didn't matter for the library itself when fully configured, but did matter
1274        when compiling without UCP support, or within Exim, where the ucp files are
1275        not imported.
1276    
1277     3. Refactoring of the library code to split up the various functions into
1278        different source modules. The addition of the new DFA matching code (see
1279        below) to a single monolithic source would have made it really too
1280        unwieldy, quite apart from causing all the code to be include in a
1281        statically linked application, when only some functions are used. This is
1282        relevant even without the DFA addition now that patterns can be compiled in
1283        one application and matched in another.
1284    
1285        The downside of splitting up is that there have to be some external
1286        functions and data tables that are used internally in different modules of
1287        the library but which are not part of the API. These have all had their
1288        names changed to start with "_pcre_" so that they are unlikely to clash
1289        with other external names.
1290    
1291     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1292        a different (DFA) algorithm. Although it is slower than the original
1293        function, it does have some advantages for certain types of matching
1294        problem.
1295    
1296     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1297        including restarting after a partial match.
1298    
1299     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1300        defined when compiling for Windows was sent to me. I have put it into the
1301        code, though I have no means of testing or verifying it.
1302    
1303     7. Added the pcre_refcount() auxiliary function.
1304    
1305     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1306        match before or at the first newline in the subject string. In pcretest,
1307        the /f option on a pattern can be used to set this.
1308    
1309     9. A repeated \w when used in UTF-8 mode with characters greater than 256
1310        would behave wrongly. This has been present in PCRE since release 4.0.
1311    
1312    10. A number of changes to the pcregrep command:
1313    
1314        (a) Refactored how -x works; insert ^(...)$ instead of setting
1315            PCRE_ANCHORED and checking the length, in preparation for adding
1316            something similar for -w.
1317    
1318        (b) Added the -w (match as a word) option.
1319    
1320        (c) Refactored the way lines are read and buffered so as to have more
1321            than one at a time available.
1322    
1323        (d) Implemented a pcregrep test script.
1324    
1325        (e) Added the -M (multiline match) option. This allows patterns to match
1326            over several lines of the subject. The buffering ensures that at least
1327            8K, or the rest of the document (whichever is the shorter) is available
1328            for matching (and similarly the previous 8K for lookbehind assertions).
1329    
1330        (f) Changed the --help output so that it now says
1331    
1332              -w, --word-regex(p)
1333    
1334            instead of two lines, one with "regex" and the other with "regexp"
1335            because that confused at least one person since the short forms are the
1336            same. (This required a bit of code, as the output is generated
1337            automatically from a table. It wasn't just a text change.)
1338    
1339        (g) -- can be used to terminate pcregrep options if the next thing isn't an
1340            option but starts with a hyphen. Could be a pattern or a path name
1341            starting with a hyphen, for instance.
1342    
1343        (h) "-" can be given as a file name to represent stdin.
1344    
1345        (i) When file names are being printed, "(standard input)" is used for
1346            the standard input, for compatibility with GNU grep. Previously
1347            "<stdin>" was used.
1348    
1349        (j) The option --label=xxx can be used to supply a name to be used for
1350            stdin when file names are being printed. There is no short form.
1351    
1352        (k) Re-factored the options decoding logic because we are going to add
1353            two more options that take data. Such options can now be given in four
1354            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1355    
1356        (l) Added the -A, -B, and -C options for requesting that lines of context
1357            around matches be printed.
1358    
1359        (m) Added the -L option to print the names of files that do not contain
1360            any matching lines, that is, the complement of -l.
1361    
1362        (n) The return code is 2 if any file cannot be opened, but pcregrep does
1363            continue to scan other files.
1364    
1365        (o) The -s option was incorrectly implemented. For compatibility with other
1366            greps, it now suppresses the error message for a non-existent or non-
1367            accessible file (but not the return code). There is a new option called
1368            -q that suppresses the output of matching lines, which was what -s was
1369            previously doing.
1370    
1371        (p) Added --include and --exclude options to specify files for inclusion
1372            and exclusion when recursing.
1373    
1374    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1375        Hopefully, it now does.
1376    
1377    12. Missing cast in pcre_study().
1378    
1379    13. Added an "uninstall" target to the makefile.
1380    
1381    14. Replaced "extern" in the function prototypes in Makefile.in with
1382        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1383        world, but is set differently for Windows.
1384    
1385    15. Added a second compiling function called pcre_compile2(). The only
1386        difference is that it has an extra argument, which is a pointer to an
1387        integer error code. When there is a compile-time failure, this is set
1388        non-zero, in addition to the error test pointer being set to point to an
1389        error message. The new argument may be NULL if no error number is required
1390        (but then you may as well call pcre_compile(), which is now just a
1391        wrapper). This facility is provided because some applications need a
1392        numeric error indication, but it has also enabled me to tidy up the way
1393        compile-time errors are handled in the POSIX wrapper.
1394    
1395    16. Added VPATH=.libs to the makefile; this should help when building with one
1396        prefix path and installing with another. (Or so I'm told by someone who
1397        knows more about this stuff than I do.)
1398    
1399    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1400        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1401        match everything, including newlines. This is not POSIX-compatible, but
1402        somebody wanted the feature. From pcretest it can be activated by using
1403        both the P and the s flags.
1404    
1405    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1406    
1407    19. libpcre.pc was being incorrectly installed as executable.
1408    
1409    20. A couple of places in pcretest check for end-of-line by looking for '\n';
1410        it now also looks for '\r' so that it will work unmodified on Windows.
1411    
1412    21. Added Google's contributed C++ wrapper to the distribution.
1413    
1414    22. Added some untidy missing memory free() calls in pcretest, to keep
1415        Electric Fence happy when testing.
1416    
1417    
1418    
1419    Version 5.0 13-Sep-04
1420    ---------------------
1421    
1422     1. Internal change: literal characters are no longer packed up into items
1423        containing multiple characters in a single byte-string. Each character
1424        is now matched using a separate opcode. However, there may be more than one
1425        byte in the character in UTF-8 mode.
1426    
1427     2. The pcre_callout_block structure has two new fields: pattern_position and
1428        next_item_length. These contain the offset in the pattern to the next match
1429        item, and its length, respectively.
1430    
1431     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1432        insertion of callouts before each pattern item. Added the /C option to
1433        pcretest to make use of this.
1434    
1435     4. On the advice of a Windows user, the lines
1436    
1437          #if defined(_WIN32) || defined(WIN32)
1438          _setmode( _fileno( stdout ), 0x8000 );
1439          #endif  /* defined(_WIN32) || defined(WIN32) */
1440    
1441        have been added to the source of pcretest. This apparently does useful
1442        magic in relation to line terminators.
1443    
1444     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1445        for the benefit of those environments where the "b" makes a difference.
1446    
1447     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1448        to know about it. I have put a hack into configure.in that adds in code
1449        to set GCC=yes if CC=icc. This seems to end up at a point in the
1450        generated configure script that is early enough to affect the setting of
1451        compiler options, which is what is needed, but I have no means of testing
1452        whether it really works. (The user who reported this had patched the
1453        generated configure script, which of course I cannot do.)
1454    
1455        LATER: After change 22 below (new libtool files), the configure script
1456        seems to know about icc (and also ecc). Therefore, I have commented out
1457        this hack in configure.in.
1458    
1459     7. Added support for pkg-config (2 patches were sent in).
1460    
1461     8. Negated POSIX character classes that used a combination of internal tables
1462        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1463        [[:^ascii]]. Typically, they would match almost any characters. The other
1464        POSIX classes were not broken in this way.
1465    
1466     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1467        to find the match, as PCRE was deluded into thinking that the match had to
1468        start at the start point or following a newline. The same bug applied to
1469        patterns with negative forward assertions or any backward assertions
1470        preceding ".*" at the start, unless the pattern required a fixed first
1471        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1472    
1473    10. In UTF-8 mode, when moving forwards in the subject after a failed match
1474        starting at the last subject character, bytes beyond the end of the subject
1475        string were read.
1476    
1477    11. Renamed the variable "class" as "classbits" to make life easier for C++
1478        users. (Previously there was a macro definition, but it apparently wasn't
1479        enough.)
1480    
1481    12. Added the new field "tables" to the extra data so that tables can be passed
1482        in at exec time, or the internal tables can be re-selected. This allows
1483        a compiled regex to be saved and re-used at a later time by a different
1484        program that might have everything at different addresses.
1485    
1486    13. Modified the pcre-config script so that, when run on Solaris, it shows a
1487        -R library as well as a -L library.
1488    
1489    14. The debugging options of pcretest (-d on the command line or D on a
1490        pattern) showed incorrect output for anything following an extended class
1491        that contained multibyte characters and which was followed by a quantifier.
1492    
1493    15. Added optional support for general category Unicode character properties
1494        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1495        support. It adds about 90K to the size of the library. The meanings of the
1496        inbuilt class escapes such as \d and \s have NOT been changed.
1497    
1498    16. Updated pcredemo.c to include calls to free() to release the memory for the
1499        compiled pattern.
1500    
1501    17. The generated file chartables.c was being created in the source directory
1502        instead of in the building directory. This caused the build to fail if the
1503        source directory was different from the building directory, and was
1504        read-only.
1505    
1506    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1507        file. No doubt somebody will tell me if they don't make sense... Also added
1508        Dan Mooney's comments about building on OpenVMS.
1509    
1510    19. Added support for partial matching via the PCRE_PARTIAL option for
1511        pcre_exec() and the \P data escape in pcretest.
1512    
1513    20. Extended pcretest with 3 new pattern features:
1514    
1515        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
1516              write the compiled pattern to the file whose name is "rest-of-line".
1517              This is a straight binary dump of the data, with the saved pointer to
1518              the character tables forced to be NULL. The study data, if any, is
1519              written too. After writing, pcretest reads a new pattern.
1520    
1521        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1522              compiled pattern from the given file. There must not be any
1523              occurrences of "<" in the file name (pretty unlikely); if there are,
1524              pcretest will instead treat the initial "<" as a pattern delimiter.
1525              After reading in the pattern, pcretest goes on to read data lines as
1526              usual.
1527    
1528        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1529              and 16-bit fields in a compiled pattern, to simulate a pattern that
1530              was compiled on a host of opposite endianness.
1531    
1532    21. The pcre-exec() function can now cope with patterns that were compiled on
1533        hosts of opposite endianness, with this restriction:
1534    
1535          As for any compiled expression that is saved and used later, the tables
1536          pointer field cannot be preserved; the extra_data field in the arguments
1537          to pcre_exec() should be used to pass in a tables address if a value
1538          other than the default internal tables were used at compile time.
1539    
1540    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1541        now diagnosed as an error. Previously, most of the time, a negative number
1542        would have been treated as zero, but if in addition "ovector" was passed as
1543        NULL, a crash could occur.
1544    
1545    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1546        new versions from the libtool 1.5 distribution (the last one is a copy of
1547        a file called libtool.m4). This seems to have fixed the need to patch
1548        "configure" to support Darwin 1.3 (which I used to do). However, I still
1549        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1550        workstation).
1551    
1552    24. Changed the PCRE licence to be the more standard "BSD" licence.
1553    
1554    
1555  Version 4.5 01-Dec-03  Version 4.5 01-Dec-03
1556  ---------------------  ---------------------
1557    

Legend:
Removed from v.73  
changed lines
  Added in v.264

  ViewVC Help
Powered by ViewVC 1.1.5