ViewVC logotype

Diff of /code/trunk/ChangeLog

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

revision 61 by nigel, Sat Feb 24 21:39:58 2007 UTC revision 205 by ph10, Fri Aug 3 13:18:33 2007 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
4  Version 3.0 02-Jan-02  Version 7.3 05-Jul-07
5    ---------------------
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:
13          #include "pcre.h"
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.
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.
27     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
29     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
30        in debug output.
32     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
33        long printing in the pcrecpp unittest when running under MinGW.
35     6. ESC_K was left out of the EBCDIC table.
37     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
38        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
39        limit also applies to "virtual nesting" when a pattern is recursive, and in
40        this case 1000 isn't so big. I have been able to remove this limit at the
41        expense of backing off one optimization in certain circumstances. Normally,
42        when pcre_exec() would call its internal match() function recursively and
43        immediately return the result unconditionally, it uses a "tail recursion"
44        feature to save stack. However, when a subpattern that can match an empty
45        string has an unlimited repetition quantifier, it no longer makes this
46        optimization. That gives it a stack frame in which to save the data for
47        checking that an empty string has been matched. Previously this was taken
48        from the 1000-entry workspace that had been reserved. So now there is no
49        explicit limit, but more stack is used.
51     8. Applied Daniel's patches to solve problems with the import/export magic
52        syntax that is required for Windows, and which was going wrong for the
53        pcreposix and pcrecpp parts of the library. These were overlooked when this
54        problem was solved for the main library.
56     9. There were some crude static tests to avoid integer overflow when computing
57        the size of patterns that contain repeated groups with explicit upper
58        limits. As the maximum quantifier is 65535, the maximum group length was
59        set at 30,000 so that the product of these two numbers did not overflow a
60        32-bit integer. However, it turns out that people want to use groups that
61        are longer than 30,000 bytes (though not repeat them that many times).
62        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
63        made it possible to implement the integer overflow checks in a much more
64        dynamic way, which I have now done. The artificial limitation on group
65        length has been removed - we now have only the limit on the total length of
66        the compiled pattern, which depends on the LINK_SIZE setting.
68    10. Fixed a bug in the documentation for get/copy named substring when
69        duplicate names are permitted. If none of the named substrings are set, the
70        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
71        empty string.
73    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
74        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
75        because the ] is interpreted as the first data character and the
76        terminating ] is not found. PCRE has been made compatible with Perl in this
77        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
78        cause memory overwriting.
81    Version 7.2 19-Jun-07
82    ---------------------
84     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
85        which is apparently normally available under Windows.
87     2. Re-jig the pcregrep tests with different newline settings in an attempt
88        to make them independent of the local environment's newline setting.
90     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
92     4. Some of the "internals" tests were previously cut out when the link size
93        was not 2, because the output contained actual offsets. The recent new
94        "Z" feature of pcretest means that these can be cut out, making the tests
95        usable with all link sizes.
97     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
98        stack recursion. This gives a massive performance boost under BSD, but just
99        a small improvement under Linux. However, it saves one field in the frame
100        in all cases.
102     6. Added more features from the forthcoming Perl 5.10:
104        (a) (?-n) (where n is a string of digits) is a relative subroutine or
105            recursion call. It refers to the nth most recently opened parentheses.
107        (b) (?+n) is also a relative subroutine call; it refers to the nth next
108            to be opened parentheses.
110        (c) Conditions that refer to capturing parentheses can be specified
111            relatively, for example, (?(-2)... or (?(+3)...
113        (d) \K resets the start of the current match so that everything before
114            is not part of it.
116        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
118        (f) \g{name} is another synonym - part of Perl 5.10's unification of
119            reference syntax.
121        (g) (?| introduces a group in which the numbering of parentheses in each
122            alternative starts with the same number.
124        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
126     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
129     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
130        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
131        for detecting groups that can match an empty string.
133     9. A pattern with a very large number of alternatives (more than several
134        hundred) was running out of internal workspace during the pre-compile
135        phase, where pcre_compile() figures out how much memory will be needed. A
136        bit of new cunning has reduced the workspace needed for groups with
137        alternatives. The 1000-alternative test pattern now uses 12 bytes of
138        workspace instead of running out of the 4096 that are available.
140    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
142    11. Applied patch from Google to remove an optimization that didn't quite work.
143        The report of the bug said:
145          pcrecpp::RE("a*").FullMatch("aaa") matches, while
146          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
147          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
149    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
150        it matched the wrong number of bytes.
153    Version 7.1 24-Apr-07
154    ---------------------
156     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
157        that is more "standard", making use of automake and other Autotools. There
158        is some re-arrangement of the files and adjustment of comments consequent
159        on this.
161     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
162        for recursive directory scanning broke on some systems because the files
163        are not scanned in any specific order and on different systems the order
164        was different. A call to "sort" has been inserted into RunGrepTest for the
165        approprate test as a short-term fix. In the longer term there may be an
166        alternative.
168     3. I had an email from Eric Raymond about problems translating some of PCRE's
169        man pages to HTML (despite the fact that I distribute HTML pages, some
170        people do their own conversions for various reasons). The problems
171        concerned the use of low-level troff macros .br and .in. I have therefore
172        removed all such uses from the man pages (some were redundant, some could
173        be replaced by .nf/.fi pairs). The 132html script that I use to generate
174        HTML has been updated to handle .nf/.fi and to complain if it encounters
175        .br or .in.
177     4. Updated comments in configure.ac that get placed in config.h.in and also
178        arranged for config.h to be included in the distribution, with the name
179        config.h.generic, for the benefit of those who have to compile without
180        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
182     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
183        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
184        makevp.bat for the current PCRE, using the additional files
185        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
187     6. A Windows user reported a minor discrepancy with test 2, which turned out
188        to be caused by a trailing space on an input line that had got lost in his
189        copy. The trailing space was an accident, so I've just removed it.
191     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
192        that is needed.
194     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
195        as "const" (a) because they are and (b) because it helps the PHP
196        maintainers who have recently made a script to detect big data structures
197        in the php code that should be moved to the .rodata section. I remembered
198        to update Builducptable as well, so it won't revert if ucptable.h is ever
199        re-created.
201     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
202        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
203        order to be able to cut out the UTF-8 tables in the latter when UTF-8
204        support is not required. This saves 1.5-2K of code, which is important in
205        some applications.
207        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
208        so as not to refer to the tables, even though these functions will never be
209        called when UTF-8 support is disabled. Otherwise there are problems with a
210        shared library.
212    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
214        (a) It was defining its arguments as char * instead of void *.
216        (b) It was assuming that all moves were upwards in memory; this was true
217            a long time ago when I wrote it, but is no longer the case.
219        The emulated memove() is provided for those environments that have neither
220        memmove() nor bcopy(). I didn't think anyone used it these days, but that
221        is clearly not the case, as these two bugs were recently reported.
223    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
224        and Detrail to create the HTML documentation, the .txt form of the man
225        pages, and it removes trailing spaces from listed files. It also creates
226        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
227        case, it wraps all the #defines with #ifndefs. This script should be run
228        before "make dist".
230    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
231        with Unicode property support.
233        (a) For a maximizing quantifier, if the two different cases of the
234            character were of different lengths in their UTF-8 codings (there are
235            some cases like this - I found 11), and the matching function had to
236            back up over a mixture of the two cases, it incorrectly assumed they
237            were both the same length.
239        (b) When PCRE was configured to use the heap rather than the stack for
240            recursion during matching, it was not correctly preserving the data for
241            the other case of a UTF-8 character when checking ahead for a match
242            while processing a minimizing repeat. If the check also involved
243            matching a wide character, but failed, corruption could cause an
244            erroneous result when trying to check for a repeat of the original
245            character.
247    13. Some tidying changes to the testing mechanism:
249        (a) The RunTest script now detects the internal link size and whether there
250            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
251            values substituted by "configure". (The RunGrepTest script already did
252            this for UTF-8.) The configure.ac script no longer substitutes the
253            relevant variables.
255        (b) The debugging options /B and /D in pcretest show the compiled bytecode
256            with length and offset values. This means that the output is different
257            for different internal link sizes. Test 2 is skipped for link sizes
258            other than 2 because of this, bypassing the problem. Unfortunately,
259            there was also a test in test 3 (the locale tests) that used /B and
260            failed for link sizes other than 2. Rather than cut the whole test out,
261            I have added a new /Z option to pcretest that replaces the length and
262            offset values with spaces. This is now used to make test 3 independent
263            of link size. (Test 2 will be tidied up later.)
265    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
266        segmentation fault instead of returning the appropriate error message.
268    15. In multiline mode when the newline sequence was set to "any", the pattern
269        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
270        This doesn't seem right; it now treats the CRLF combination as the line
271        ending, and so does not match in that case. It's only a pattern such as ^$
272        that would hit this one: something like ^ABC$ would have failed after \r
273        and then tried again after \r\n.
275    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
276        in an attempt to make files that differ only in their line terminators
277        compare equal. This works on Linux.
279    17. Under certain error circumstances pcregrep might try to free random memory
280        as it exited. This is now fixed, thanks to valgrind.
282    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
283        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
284        was because its rules for how to advance for /g after matching an empty
285        string at the end of a line did not allow for this case. They now check for
286        it specially.
288    20. pcretest is supposed to handle patterns and data of any length, by
289        extending its buffers when necessary. It was getting this wrong when the
290        buffer for a data line had to be extended.
292    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
293        CRLF as a newline sequence.
295    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
296        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
297        I have nevertheless tidied it up.
299    23. Added some casts to kill warnings from HP-UX ia64 compiler.
301    24. Added a man page for pcre-config.
304    Version 7.0 19-Dec-06
305    ---------------------
307     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
308        moving to gcc 4.1.1.
310     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
311        sys/time.h, which is documented as needed for this function. It doesn't
312        seem to matter on Linux, but it showed up on some releases of OS X.
314     3. It seems that there are systems where bytes whose values are greater than
315        127 match isprint() in the "C" locale. The "C" locale should be the
316        default when a C program starts up. In most systems, only ASCII printing
317        characters match isprint(). This difference caused the output from pcretest
318        to vary, making some of the tests fail. I have changed pcretest so that:
320        (a) When it is outputting text in the compiled version of a pattern, bytes
321            other than 32-126 are always shown as hex escapes.
323        (b) When it is outputting text that is a matched part of a subject string,
324            it does the same, unless a different locale has been set for the match
325            (using the /L modifier). In this case, it uses isprint() to decide.
327     4. Fixed a major bug that caused incorrect computation of the amount of memory
328        required for a compiled pattern when options that changed within the
329        pattern affected the logic of the preliminary scan that determines the
330        length. The relevant options are -x, and -i in UTF-8 mode. The result was
331        that the computed length was too small. The symptoms of this bug were
332        either the PCRE error "internal error: code overflow" from pcre_compile(),
333        or a glibc crash with a message such as "pcretest: free(): invalid next
334        size (fast)". Examples of patterns that provoked this bug (shown in
335        pcretest format) are:
337          /(?-x: )/x
338          /(?x)(?-x: \s*#\s*)/
339          /((?i)[\x{c0}])/8
340          /(?i:[\x{c0}])/8
342        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
343        is now done differently.
345     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
346        wrapper classes; (b) implement a new function in the C++ scanner that is
347        more efficient than the old way of doing things because it avoids levels of
348        recursion in the regex matching; (c) add a paragraph to the documentation
349        for the FullMatch() function.
351     6. The escape sequence \n was being treated as whatever was defined as
352        "newline". Not only was this contrary to the documentation, which states
353        that \n is character 10 (hex 0A), but it also went horribly wrong when
354        "newline" was defined as CRLF. This has been fixed.
356     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
357        was being set to -1 for the "end of line" case (supposedly a value that no
358        character can have). Though this value is never used (the check for end of
359        line is "zero bytes in current character"), it caused compiler complaints.
360        I've changed it to 0xffffffff.
362     8. In pcre_version.c, the version string was being built by a sequence of
363        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
364        string (as it is for production releases) called a macro with an empty
365        argument. The C standard says the result of this is undefined. The gcc
366        compiler treats it as an empty string (which was what was wanted) but it is
367        reported that Visual C gives an error. The source has been hacked around to
368        avoid this problem.
370     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
371        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
372        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
373        of them did).
375    10. Originally, pcretest opened its input and output without "b"; then I was
376        told that "b" was needed in some environments, so it was added for release
377        5.0 to both the input and output. (It makes no difference on Unix-like
378        systems.) Later I was told that it is wrong for the input on Windows. I've
379        now abstracted the modes into two macros, to make it easier to fiddle with
380        them, and removed "b" from the input mode under Windows.
382    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
384    12. Added -help and --help to pcretest as an official way of being reminded
385        of the options.
387    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
388        and pcrecpp.cc because they annoy compilers at high warning levels.
390    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
392    15. Fixed an occurrence of == in configure.ac that should have been = (shell
393        scripts are not C programs :-) and which was not noticed because it works
394        on Linux.
396    16. pcretest is supposed to handle any length of pattern and data line (as one
397        line or as a continued sequence of lines) by extending its input buffer if
398        necessary. This feature was broken for very long pattern lines, leading to
399        a string of junk being passed to pcre_compile() if the pattern was longer
400        than about 50K.
402    17. I have done a major re-factoring of the way pcre_compile() computes the
403        amount of memory needed for a compiled pattern. Previously, there was code
404        that made a preliminary scan of the pattern in order to do this. That was
405        OK when PCRE was new, but as the facilities have expanded, it has become
406        harder and harder to keep it in step with the real compile phase, and there
407        have been a number of bugs (see for example, 4 above). I have now found a
408        cunning way of running the real compile function in a "fake" mode that
409        enables it to compute how much memory it would need, while actually only
410        ever using a few hundred bytes of working memory and without too many
411        tests of the mode. This should make future maintenance and development
412        easier. A side effect of this work is that the limit of 200 on the nesting
413        depth of parentheses has been removed (though this was never a serious
414        limitation, I suspect). However, there is a downside: pcre_compile() now
415        runs more slowly than before (30% or more, depending on the pattern). I
416        hope this isn't a big issue. There is no effect on runtime performance.
418    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
419        newline (only possible for the last line of a file) and it was a
420        pattern that set a locale (followed by /Lsomething), pcretest crashed.
422    19. Added additional timing features to pcretest. (1) The -tm option now times
423        matching only, not compiling. (2) Both -t and -tm can be followed, as a
424        separate command line item, by a number that specifies the number of
425        repeats to use when timing. The default is 50000; this gives better
426        precision, but takes uncomfortably long for very large patterns.
428    20. Extended pcre_study() to be more clever in cases where a branch of a
429        subpattern has no definite first character. For example, (a*|b*)[cd] would
430        previously give no result from pcre_study(). Now it recognizes that the
431        first character must be a, b, c, or d.
433    21. There was an incorrect error "recursive call could loop indefinitely" if
434        a subpattern (or the entire pattern) that was being tested for matching an
435        empty string contained only one non-empty item after a nested subpattern.
436        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
437        incorrectly, because the \d was being skipped in the check.
439    22. The pcretest program now has a new pattern option /B and a command line
440        option -b, which is equivalent to adding /B to every pattern. This causes
441        it to show the compiled bytecode, without the additional information that
442        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
443        is the same as /B/I).
445    23. A new optimization is now able automatically to treat some sequences such
446        as a*b as a*+b. More specifically, if something simple (such as a character
447        or a simple class like \d) has an unlimited quantifier, and is followed by
448        something that cannot possibly match the quantified thing, the quantifier
449        is automatically "possessified".
451    24. A recursive reference to a subpattern whose number was greater than 39
452        went wrong under certain circumstances in UTF-8 mode. This bug could also
453        have affected the operation of pcre_study().
455    25. Realized that a little bit of performance could be had by replacing
456        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
458    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
460    27. Possessive quantifiers such as a++ were previously implemented by turning
461        them into atomic groups such as ($>a+). Now they have their own opcodes,
462        which improves performance. This includes the automatically created ones
463        from 23 above.
465    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
466        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
467        the first matched character to be a colon. This applied both to named and
468        numbered groups.
470    29. The ucpinternal.h header file was missing its idempotency #ifdef.
472    30. I was sent a "project" file called libpcre.a.dev which I understand makes
473        building PCRE on Windows easier, so I have included it in the distribution.
475    31. There is now a check in pcretest against a ridiculously large number being
476        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
477        loop, the loop is abandoned.
479    32. Forward references to subpatterns in conditions such as (?(2)...) where
480        subpattern 2 is defined later cause pcre_compile() to search forwards in
481        the pattern for the relevant set of parentheses. This search went wrong
482        when there were unescaped parentheses in a character class, parentheses
483        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
485    33. "Subroutine" calls and backreferences were previously restricted to
486        referencing subpatterns earlier in the regex. This restriction has now
487        been removed.
489    34. Added a number of extra features that are going to be in Perl 5.10. On the
490        whole, these are just syntactic alternatives for features that PCRE had
491        previously implemented using the Python syntax or my own invention. The
492        other formats are all retained for compatibility.
494        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
495            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
496            also .NET compatible.
498        (b) A recursion or subroutine call to a named group can now be defined as
499            (?&name) as well as (?P>name).
501        (c) A backreference to a named group can now be defined as \k<name> or
502            \k'name' as well as (?P=name). The new forms, as well as being in Perl
503            5.10, are also .NET compatible.
505        (d) A conditional reference to a named group can now use the syntax
506            (?(<name>) or (?('name') as well as (?(name).
508        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
509            groups (named and numbered) that are never evaluated inline, but can be
510            called as "subroutines" from elsewhere. In effect, the DEFINE condition
511            is always false. There may be only one alternative in such a group.
513        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
514            as the simple (?(R). The condition is true only if the most recent
515            recursion is that of the given number or name. It does not search out
516            through the entire recursion stack.
518        (g) The escape \gN or \g{N} has been added, where N is a positive or
519            negative number, specifying an absolute or relative reference.
521    35. Tidied to get rid of some further signed/unsigned compiler warnings and
522        some "unreachable code" warnings.
524    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
525        things, this adds five new scripts.
527    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
528        There were also incompatibilities regarding the handling of \Q..\E inside
529        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
530        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
532    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
533        matches an empty string, and forcibly breaks the loop. There were bugs in
534        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
535        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
536        separate and independent bugs (that affected different cases) have been
537        fixed.
539    39. Refactored the code to abolish the use of different opcodes for small
540        capturing bracket numbers. This is a tidy that I avoided doing when I
541        removed the limit on the number of capturing brackets for 3.5 back in 2001.
542        The new approach is not only tidier, it makes it possible to reduce the
543        memory needed to fix the previous bug (38).
545    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
546        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
547        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
548        mode.
550    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
551        report.
553    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
554        copy construction and assignment in the C++ wrapper.
556    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
557        couple of bugs that could have given wrong results in the "--newline=crlf"
558        case.
560    44. Added a number of casts and did some reorganization of signed/unsigned int
561        variables following suggestions from Dair Grant. Also renamed the variable
562        "this" as "item" because it is a C++ keyword.
564    45. Arranged for dftables to add
566          #include "pcre_internal.h"
568        to pcre_chartables.c because without it, gcc 4.x may remove the array
569        definition from the final binary if PCRE is built into a static library and
570        dead code stripping is activated.
572    46. For an unanchored pattern, if a match attempt fails at the start of a
573        newline sequence, and the newline setting is CRLF or ANY, and the next two
574        characters are CRLF, advance by two characters instead of one.
577    Version 6.7 04-Jul-06
578    ---------------------
580     1. In order to handle tests when input lines are enormously long, pcretest has
581        been re-factored so that it automatically extends its buffers when
582        necessary. The code is crude, but this _is_ just a test program. The
583        default size has been increased from 32K to 50K.
585     2. The code in pcre_study() was using the value of the re argument before
586        testing it for NULL. (Of course, in any sensible call of the function, it
587        won't be NULL.)
589     3. The memmove() emulation function in pcre_internal.h, which is used on
590        systems that lack both memmove() and bcopy() - that is, hardly ever -
591        was missing a "static" storage class specifier.
593     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
594        containing an extended class (one that cannot be represented by a bitmap
595        because it contains high-valued characters or Unicode property items, e.g.
596        [\pZ]). Almost always one would set UTF-8 mode when processing such a
597        pattern, but PCRE should not loop if you do not (it no longer does).
598        [Detail: two cases were found: (a) a repeated subpattern containing an
599        extended class; (b) a recursive reference to a subpattern that followed a
600        previous extended class. It wasn't skipping over the extended class
601        correctly when UTF-8 mode was not set.]
603     5. A negated single-character class was not being recognized as fixed-length
604        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
605        compile error "lookbehind assertion is not fixed length".
607     6. The RunPerlTest auxiliary script was showing an unexpected difference
608        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
609        write a Perl script that can interpret lines of an input file either as
610        byte characters or as UTF-8, which is what "perltest" was being required to
611        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
612        can't do is switch easily at run time between having the "use utf8;" pragma
613        or not. In the end, I fudged it by using the RunPerlTest script to insert
614        "use utf8;" explicitly for the UTF-8 tests.
616     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
617        the end of the subject string, contrary to the documentation and to what
618        Perl does. This was true of both matching functions. Now it matches only at
619        the start of the subject and immediately after *internal* newlines.
621     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
622        a pointer to an int instead of a pointer to an unsigned long int. This
623        caused problems on 64-bit systems.
625     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
626        instance of the 'standard' template library not being so standard".
628    10. There was no check on the number of named subpatterns nor the maximum
629        length of a subpattern name. The product of these values is used to compute
630        the size of the memory block for a compiled pattern. By supplying a very
631        long subpattern name and a large number of named subpatterns, the size
632        computation could be caused to overflow. This is now prevented by limiting
633        the length of names to 32 characters, and the number of named subpatterns
634        to 10,000.
636    11. Subpatterns that are repeated with specific counts have to be replicated in
637        the compiled pattern. The size of memory for this was computed from the
638        length of the subpattern and the repeat count. The latter is limited to
639        65535, but there was no limit on the former, meaning that integer overflow
640        could in principle occur. The compiled length of a repeated subpattern is
641        now limited to 30,000 bytes in order to prevent this.
643    12. Added the optional facility to have named substrings with the same name.
645    13. Added the ability to use a named substring as a condition, using the
646        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
647        are numbers (not recommended). Forward references are permitted.
649    14. Added forward references in named backreferences (if you see what I mean).
651    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
652        pattern could run off the end of the subject. For example, the pattern
653        "(?s)(.{1,5})"8 did this with the subject "ab".
655    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
656        PCRE_CASELESS was set when matching characters that were quantified with ?
657        or *.
659    17. A character class other than a single negated character that had a minimum
660        but no maximum quantifier - for example [ab]{6,} - was not handled
661        correctly by pce_dfa_exec(). It would match only one character.
663    18. A valid (though odd) pattern that looked like a POSIX character
664        class but used an invalid character after [ (for example [[,abc,]]) caused
665        pcre_compile() to give the error "Failed: internal error: code overflow" or
666        in some cases to crash with a glibc free() error. This could even happen if
667        the pattern terminated after [[ but there just happened to be a sequence of
668        letters, a binary zero, and a closing ] in the memory that followed.
670    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
671        over the years. Originally (before any Unicode support), just the bottom 8
672        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
673        output from "man perlunicode" includes this:
675          The regular expression compiler produces polymorphic opcodes.  That
676          is, the pattern adapts to the data and automatically switches to
677          the Unicode character scheme when presented with Unicode data--or
678          instead uses a traditional byte scheme when presented with byte
679          data.
681        Sadly, a wide octal escape does not cause a switch, and in a string with
682        no other multibyte characters, these octal escapes are treated as before.
683        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
684        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
685        Unicode string.
687        I have not perpetrated such confusion in PCRE. Up till now, it took just
688        the bottom 8 bits, as in old Perl. I have now made octal escapes with
689        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
690        translate to the appropriate multibyte character.
692    29. Applied some refactoring to reduce the number of warnings from Microsoft
693        and Borland compilers. This has included removing the fudge introduced
694        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
695        a warning about an unused variable.
697    21. PCRE has not included VT (character 0x0b) in the set of whitespace
698        characters since release 4.0, because Perl (from release 5.004) does not.
699        [Or at least, is documented not to: some releases seem to be in conflict
700        with the documentation.] However, when a pattern was studied with
701        pcre_study() and all its branches started with \s, PCRE still included VT
702        as a possible starting character. Of course, this did no harm; it just
703        caused an unnecessary match attempt.
705    22. Removed a now-redundant internal flag bit that recorded the fact that case
706        dependency changed within the pattern. This was once needed for "required
707        byte" processing, but is no longer used. This recovers a now-scarce options
708        bit. Also moved the least significant internal flag bit to the most-
709        significant bit of the word, which was not previously used (hangover from
710        the days when it was an int rather than a uint) to free up another bit for
711        the future.
713    23. Added support for CRLF line endings as well as CR and LF. As well as the
714        default being selectable at build time, it can now be changed at runtime
715        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
716        specify that it is scanning data with non-default line endings.
718    24. Changed the definition of CXXLINK to make it agree with the definition of
719        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
721    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
722        recursions. This makes a big different to stack usage for some patterns.
724    26. If a subpattern containing a named recursion or subroutine reference such
725        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
726        the space required for the compiled pattern went wrong and gave too small a
727        value. Depending on the environment, this could lead to "Failed: internal
728        error: code overflow at offset 49" or "glibc detected double free or
729        corruption" errors.
731    27. Applied patches from Google (a) to support the new newline modes and (b) to
732        advance over multibyte UTF-8 characters in GlobalReplace.
734    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
735        difference for some implementation of PCRE in some Windows version.
737    29. Added some extra testing facilities to pcretest:
739        \q<number>   in a data line sets the "match limit" value
740        \Q<number>   in a data line sets the "match recursion limt" value
741        -S <number>  sets the stack size, where <number> is in megabytes
743        The -S option isn't available for Windows.
746    Version 6.6 06-Feb-06
747    ---------------------
749     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
750        in pcreposix.h. I have copied the definition from pcre.h.
752     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
753        because pcre.h is no longer a built file.
755     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
756        not normally included in the compiled code.
759    Version 6.5 01-Feb-06
760    ---------------------
762     1. When using the partial match feature with pcre_dfa_exec(), it was not
763        anchoring the second and subsequent partial matches at the new starting
764        point. This could lead to incorrect results. For example, with the pattern
765        /1234/, partially matching against "123" and then "a4" gave a match.
767     2. Changes to pcregrep:
769        (a) All non-match returns from pcre_exec() were being treated as failures
770            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
771            error message is output. Some extra information is given for the
772            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
773            probably the only errors that are likely to be caused by users (by
774            specifying a regex that has nested indefinite repeats, for instance).
775            If there are more than 20 of these errors, pcregrep is abandoned.
777        (b) A binary zero was treated as data while matching, but terminated the
778            output line if it was written out. This has been fixed: binary zeroes
779            are now no different to any other data bytes.
781        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
782            used to set a locale for matching. The --locale=xxxx long option has
783            been added (no short equivalent) to specify a locale explicitly on the
784            pcregrep command, overriding the environment variables.
786        (d) When -B was used with -n, some line numbers in the output were one less
787            than they should have been.
789        (e) Added the -o (--only-matching) option.
791        (f) If -A or -C was used with -c (count only), some lines of context were
792            accidentally printed for the final match.
794        (g) Added the -H (--with-filename) option.
796        (h) The combination of options -rh failed to suppress file names for files
797            that were found from directory arguments.
799        (i) Added the -D (--devices) and -d (--directories) options.
801        (j) Added the -F (--fixed-strings) option.
803        (k) Allow "-" to be used as a file name for -f as well as for a data file.
805        (l) Added the --colo(u)r option.
807        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
808            is not present by default.
810     3. A nasty bug was discovered in the handling of recursive patterns, that is,
811        items such as (?R) or (?1), when the recursion could match a number of
812        alternatives. If it matched one of the alternatives, but subsequently,
813        outside the recursion, there was a failure, the code tried to back up into
814        the recursion. However, because of the way PCRE is implemented, this is not
815        possible, and the result was an incorrect result from the match.
817        In order to prevent this happening, the specification of recursion has
818        been changed so that all such subpatterns are automatically treated as
819        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
821     4. I had overlooked the fact that, in some locales, there are characters for
822        which isalpha() is true but neither isupper() nor islower() are true. In
823        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
824        and ordfeminine) are like this. This affected the treatment of \w and \W
825        when they appeared in character classes, but not when they appeared outside
826        a character class. The bit map for "word" characters is now created
827        separately from the results of isalnum() instead of just taking it from the
828        upper, lower, and digit maps. (Plus the underscore character, of course.)
830     5. The above bug also affected the handling of POSIX character classes such as
831        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
832        permanent tables. Instead, the bit maps for such a class were previously
833        created as the appropriate unions of the upper, lower, and digit bitmaps.
834        Now they are created by subtraction from the [[:word:]] class, which has
835        its own bitmap.
837     6. The [[:blank:]] character class matches horizontal, but not vertical space.
838        It is created by subtracting the vertical space characters (\x09, \x0a,
839        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
840        subtraction was done in the overall bitmap for a character class, meaning
841        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
842        be recognized. This bug has been fixed.
844     7. Patches from the folks at Google:
846          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
847          real life, but is still worth protecting against".
849          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
850          regular expressions".
852          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
853          have it.
855          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
856          "configure" and the latter not, in order to fix a problem somebody had
857          with compiling the Arg class on HP-UX.
859          (e) Improve the error-handling of the C++ wrapper a little bit.
861          (f) New tests for checking recursion limiting.
863     8. The pcre_memmove() function, which is used only if the environment does not
864        have a standard memmove() function (and is therefore rarely compiled),
865        contained two bugs: (a) use of int instead of size_t, and (b) it was not
866        returning a result (though PCRE never actually uses the result).
868     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
869        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
870        returned instead of calling malloc() with an overflowing number that would
871        most likely cause subsequent chaos.
873    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
875    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
876        with this option is matched, the nmatch and pmatch options of regexec() are
877        ignored.
879    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
880        provided in case anyone wants to the the POSIX interface with UTF-8
881        strings.
883    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
884        C++ linking (needed for some HP-UX environments).
886    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
887        (unused parameter) and in the pcre_printint() function (omitted "default"
888        switch label when the default is to do nothing).
890    15. Added some code to make it possible, when PCRE is compiled as a C++
891        library, to replace subject pointers for pcre_exec() with a smart pointer
892        class, thus making it possible to process discontinuous strings.
894    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
895        much the same function. They were added by different people who were trying
896        to make PCRE easy to compile on non-Unix systems. It has been suggested
897        that PCRE_EXPORT be abolished now that there is more automatic apparatus
898        for compiling on Windows systems. I have therefore replaced it with
899        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
900        defaults to "extern" for C or "extern C" for C++, which works fine on
901        Unix-like systems. It is now possible to override the value of PCRE_DATA_
902        SCOPE with something explicit in config.h. In addition:
904        (a) pcreposix.h still had just "extern" instead of either of these macros;
905            I have replaced it with PCRE_DATA_SCOPE.
907        (b) Functions such as _pcre_xclass(), which are internal to the library,
908            but external in the C sense, all had PCRE_EXPORT in their definitions.
909            This is apparently wrong for the Windows case, so I have removed it.
910            (It makes no difference on Unix-like systems.)
912    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
913        of recursive calls to match(). This is different to MATCH_LIMIT because
914        that limits the total number of calls to match(), not all of which increase
915        the depth of recursion. Limiting the recursion depth limits the amount of
916        stack (or heap if NO_RECURSE is set) that is used. The default can be set
917        when PCRE is compiled, and changed at run time. A patch from Google adds
918        this functionality to the C++ interface.
920    18. Changes to the handling of Unicode character properties:
922        (a) Updated the table to Unicode 4.1.0.
924        (b) Recognize characters that are not in the table as "Cn" (undefined).
926        (c) I revised the way the table is implemented to a much improved format
927            which includes recognition of ranges. It now supports the ranges that
928            are defined in UnicodeData.txt, and it also amalgamates other
929            characters into ranges. This has reduced the number of entries in the
930            table from around 16,000 to around 3,000, thus reducing its size
931            considerably. I realized I did not need to use a tree structure after
932            all - a binary chop search is just as efficient. Having reduced the
933            number of entries, I extended their size from 6 bytes to 8 bytes to
934            allow for more data.
936        (d) Added support for Unicode script names via properties such as \p{Han}.
938    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
939        matching that character.
941    20. When matching a repeated Unicode property with a minimum greater than zero,
942        (for example \pL{2,}), PCRE could look past the end of the subject if it
943        reached it while seeking the minimum number of characters. This could
944        happen only if some of the characters were more than one byte long, because
945        there is a check for at least the minimum number of bytes.
947    21. Refactored the implementation of \p and \P so as to be more general, to
948        allow for more different types of property in future. This has changed the
949        compiled form incompatibly. Anybody with saved compiled patterns that use
950        \p or \P will have to recompile them.
952    22. Added "Any" and "L&" to the supported property types.
954    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
955        but give a compile time error if the value is greater than 0xff.
957    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
958        accidentally not being installed or uninstalled.
960    25. The pcre.h file was built from pcre.h.in, but the only changes that were
961        made were to insert the current release number. This seemed silly, because
962        it made things harder for people building PCRE on systems that don't run
963        "configure". I have turned pcre.h into a distributed file, no longer built
964        by "configure", with the version identification directly included. There is
965        no longer a pcre.h.in file.
967        However, this change necessitated a change to the pcre-config script as
968        well. It is built from pcre-config.in, and one of the substitutions was the
969        release number. I have updated configure.ac so that ./configure now finds
970        the release number by grepping pcre.h.
972    26. Added the ability to run the tests under valgrind.
975    Version 6.4 05-Sep-05
976    ---------------------
978     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
979        "--" to be printed when multiple files were scanned, even when none of the
980        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
981        consider it to be a bug, and have restored the previous behaviour.
983     2. A couple of code tidies to get rid of compiler warnings.
985     3. The pcretest program used to cheat by referring to symbols in the library
986        whose names begin with _pcre_. These are internal symbols that are not
987        really supposed to be visible externally, and in some environments it is
988        possible to suppress them. The cheating is now confined to including
989        certain files from the library's source, which is a bit cleaner.
991     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
992        file's purpose clearer.
994     5. Reorganized pcre_ucp_findchar().
997    Version 6.3 15-Aug-05
998    ---------------------
1000     1. The file libpcre.pc.in did not have general read permission in the tarball.
1002     2. There were some problems when building without C++ support:
1004        (a) If C++ support was not built, "make install" and "make test" still
1005            tried to test it.
1007        (b) There were problems when the value of CXX was explicitly set. Some
1008            changes have been made to try to fix these, and ...
1010        (c) --disable-cpp can now be used to explicitly disable C++ support.
1012        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1013            backslash in a target when C++ was disabled. This confuses some
1014            versions of "make", apparently. Using an intermediate variable solves
1015            this. (Same for CPP_LOBJ.)
1017     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1018        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1019        necessary on certain architectures.
1021     4. Added a setting of -export-symbols-regex to the link command to remove
1022        those symbols that are exported in the C sense, but actually are local
1023        within the library, and not documented. Their names all begin with
1024        "_pcre_". This is not a perfect job, because (a) we have to except some
1025        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1026        available (and never for static libraries). I have made a note to try to
1027        find a way round (a) in the future.
1030    Version 6.2 01-Aug-05
1031    ---------------------
1033     1. There was no test for integer overflow of quantifier values. A construction
1034        such as {1111111111111111} would give undefined results. What is worse, if
1035        a minimum quantifier for a parenthesized subpattern overflowed and became
1036        negative, the calculation of the memory size went wrong. This could have
1037        led to memory overwriting.
1039     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1041     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1042        operating environments where this matters.
1044     4. Applied Giuseppe Maxia's patch to add additional features for controlling
1045        PCRE options from within the C++ wrapper.
1047     5. Named capturing subpatterns were not being correctly counted when a pattern
1048        was compiled. This caused two problems: (a) If there were more than 100
1049        such subpatterns, the calculation of the memory needed for the whole
1050        compiled pattern went wrong, leading to an overflow error. (b) Numerical
1051        back references of the form \12, where the number was greater than 9, were
1052        not recognized as back references, even though there were sufficient
1053        previous subpatterns.
1055     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1056        versions of gcc, e.g. 2.95.4.
1059    Version 6.1 21-Jun-05
1060    ---------------------
1062     1. There was one reference to the variable "posix" in pcretest.c that was not
1063        surrounded by "#if !defined NOPOSIX".
1065     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1066        the cross-check on the old pcre_info() function, for the benefit of the
1067        cut-down version of PCRE that is currently imported into Exim.
1069     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1070        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1071        patterns that start (?i) but I don't think that matters. The use of (?i) is
1072        just an example; this all applies to the other options as well.
1074     4. Since libtool seems to echo the compile commands it is issuing, the output
1075        from "make" can be reduced a bit by putting "@" in front of each libtool
1076        compile command.
1078     5. Patch from the folks at Google for configure.in to be a bit more thorough
1079        in checking for a suitable C++ installation before trying to compile the
1080        C++ stuff. This should fix a reported problem when a compiler was present,
1081        but no suitable headers.
1083     6. The man pages all had just "PCRE" as their title. I have changed them to
1084        be the relevant file name. I have also arranged that these names are
1085        retained in the file doc/pcre.txt, which is a concatenation in text format
1086        of all the man pages except the little individual ones for each function.
1088     7. The NON-UNIX-USE file had not been updated for the different set of source
1089        files that come with release 6. I also added a few comments about the C++
1090        wrapper.
1093    Version 6.0 07-Jun-05
1094    ---------------------
1096     1. Some minor internal re-organization to help with my DFA experiments.
1098     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1099        didn't matter for the library itself when fully configured, but did matter
1100        when compiling without UCP support, or within Exim, where the ucp files are
1101        not imported.
1103     3. Refactoring of the library code to split up the various functions into
1104        different source modules. The addition of the new DFA matching code (see
1105        below) to a single monolithic source would have made it really too
1106        unwieldy, quite apart from causing all the code to be include in a
1107        statically linked application, when only some functions are used. This is
1108        relevant even without the DFA addition now that patterns can be compiled in
1109        one application and matched in another.
1111        The downside of splitting up is that there have to be some external
1112        functions and data tables that are used internally in different modules of
1113        the library but which are not part of the API. These have all had their
1114        names changed to start with "_pcre_" so that they are unlikely to clash
1115        with other external names.
1117     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1118        a different (DFA) algorithm. Although it is slower than the original
1119        function, it does have some advantages for certain types of matching
1120        problem.
1122     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1123        including restarting after a partial match.
1125     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1126        defined when compiling for Windows was sent to me. I have put it into the
1127        code, though I have no means of testing or verifying it.
1129     7. Added the pcre_refcount() auxiliary function.
1131     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1132        match before or at the first newline in the subject string. In pcretest,
1133        the /f option on a pattern can be used to set this.
1135     9. A repeated \w when used in UTF-8 mode with characters greater than 256
1136        would behave wrongly. This has been present in PCRE since release 4.0.
1138    10. A number of changes to the pcregrep command:
1140        (a) Refactored how -x works; insert ^(...)$ instead of setting
1141            PCRE_ANCHORED and checking the length, in preparation for adding
1142            something similar for -w.
1144        (b) Added the -w (match as a word) option.
1146        (c) Refactored the way lines are read and buffered so as to have more
1147            than one at a time available.
1149        (d) Implemented a pcregrep test script.
1151        (e) Added the -M (multiline match) option. This allows patterns to match
1152            over several lines of the subject. The buffering ensures that at least
1153            8K, or the rest of the document (whichever is the shorter) is available
1154            for matching (and similarly the previous 8K for lookbehind assertions).
1156        (f) Changed the --help output so that it now says
1158              -w, --word-regex(p)
1160            instead of two lines, one with "regex" and the other with "regexp"
1161            because that confused at least one person since the short forms are the
1162            same. (This required a bit of code, as the output is generated
1163            automatically from a table. It wasn't just a text change.)
1165        (g) -- can be used to terminate pcregrep options if the next thing isn't an
1166            option but starts with a hyphen. Could be a pattern or a path name
1167            starting with a hyphen, for instance.
1169        (h) "-" can be given as a file name to represent stdin.
1171        (i) When file names are being printed, "(standard input)" is used for
1172            the standard input, for compatibility with GNU grep. Previously
1173            "<stdin>" was used.
1175        (j) The option --label=xxx can be used to supply a name to be used for
1176            stdin when file names are being printed. There is no short form.
1178        (k) Re-factored the options decoding logic because we are going to add
1179            two more options that take data. Such options can now be given in four
1180            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1182        (l) Added the -A, -B, and -C options for requesting that lines of context
1183            around matches be printed.
1185        (m) Added the -L option to print the names of files that do not contain
1186            any matching lines, that is, the complement of -l.
1188        (n) The return code is 2 if any file cannot be opened, but pcregrep does
1189            continue to scan other files.
1191        (o) The -s option was incorrectly implemented. For compatibility with other
1192            greps, it now suppresses the error message for a non-existent or non-
1193            accessible file (but not the return code). There is a new option called
1194            -q that suppresses the output of matching lines, which was what -s was
1195            previously doing.
1197        (p) Added --include and --exclude options to specify files for inclusion
1198            and exclusion when recursing.
1200    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1201        Hopefully, it now does.
1203    12. Missing cast in pcre_study().
1205    13. Added an "uninstall" target to the makefile.
1207    14. Replaced "extern" in the function prototypes in Makefile.in with
1208        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1209        world, but is set differently for Windows.
1211    15. Added a second compiling function called pcre_compile2(). The only
1212        difference is that it has an extra argument, which is a pointer to an
1213        integer error code. When there is a compile-time failure, this is set
1214        non-zero, in addition to the error test pointer being set to point to an
1215        error message. The new argument may be NULL if no error number is required
1216        (but then you may as well call pcre_compile(), which is now just a
1217        wrapper). This facility is provided because some applications need a
1218        numeric error indication, but it has also enabled me to tidy up the way
1219        compile-time errors are handled in the POSIX wrapper.
1221    16. Added VPATH=.libs to the makefile; this should help when building with one
1222        prefix path and installing with another. (Or so I'm told by someone who
1223        knows more about this stuff than I do.)
1225    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1226        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1227        match everything, including newlines. This is not POSIX-compatible, but
1228        somebody wanted the feature. From pcretest it can be activated by using
1229        both the P and the s flags.
1231    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1233    19. libpcre.pc was being incorrectly installed as executable.
1235    20. A couple of places in pcretest check for end-of-line by looking for '\n';
1236        it now also looks for '\r' so that it will work unmodified on Windows.
1238    21. Added Google's contributed C++ wrapper to the distribution.
1240    22. Added some untidy missing memory free() calls in pcretest, to keep
1241        Electric Fence happy when testing.
1245    Version 5.0 13-Sep-04
1246    ---------------------
1248     1. Internal change: literal characters are no longer packed up into items
1249        containing multiple characters in a single byte-string. Each character
1250        is now matched using a separate opcode. However, there may be more than one
1251        byte in the character in UTF-8 mode.
1253     2. The pcre_callout_block structure has two new fields: pattern_position and
1254        next_item_length. These contain the offset in the pattern to the next match
1255        item, and its length, respectively.
1257     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1258        insertion of callouts before each pattern item. Added the /C option to
1259        pcretest to make use of this.
1261     4. On the advice of a Windows user, the lines
1263          #if defined(_WIN32) || defined(WIN32)
1264          _setmode( _fileno( stdout ), 0x8000 );
1265          #endif  /* defined(_WIN32) || defined(WIN32) */
1267        have been added to the source of pcretest. This apparently does useful
1268        magic in relation to line terminators.
1270     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1271        for the benefit of those environments where the "b" makes a difference.
1273     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1274        to know about it. I have put a hack into configure.in that adds in code
1275        to set GCC=yes if CC=icc. This seems to end up at a point in the
1276        generated configure script that is early enough to affect the setting of
1277        compiler options, which is what is needed, but I have no means of testing
1278        whether it really works. (The user who reported this had patched the
1279        generated configure script, which of course I cannot do.)
1281        LATER: After change 22 below (new libtool files), the configure script
1282        seems to know about icc (and also ecc). Therefore, I have commented out
1283        this hack in configure.in.
1285     7. Added support for pkg-config (2 patches were sent in).
1287     8. Negated POSIX character classes that used a combination of internal tables
1288        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1289        [[:^ascii]]. Typically, they would match almost any characters. The other
1290        POSIX classes were not broken in this way.
1292     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1293        to find the match, as PCRE was deluded into thinking that the match had to
1294        start at the start point or following a newline. The same bug applied to
1295        patterns with negative forward assertions or any backward assertions
1296        preceding ".*" at the start, unless the pattern required a fixed first
1297        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1299    10. In UTF-8 mode, when moving forwards in the subject after a failed match
1300        starting at the last subject character, bytes beyond the end of the subject
1301        string were read.
1303    11. Renamed the variable "class" as "classbits" to make life easier for C++
1304        users. (Previously there was a macro definition, but it apparently wasn't
1305        enough.)
1307    12. Added the new field "tables" to the extra data so that tables can be passed
1308        in at exec time, or the internal tables can be re-selected. This allows
1309        a compiled regex to be saved and re-used at a later time by a different
1310        program that might have everything at different addresses.
1312    13. Modified the pcre-config script so that, when run on Solaris, it shows a
1313        -R library as well as a -L library.
1315    14. The debugging options of pcretest (-d on the command line or D on a
1316        pattern) showed incorrect output for anything following an extended class
1317        that contained multibyte characters and which was followed by a quantifier.
1319    15. Added optional support for general category Unicode character properties
1320        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1321        support. It adds about 90K to the size of the library. The meanings of the
1322        inbuilt class escapes such as \d and \s have NOT been changed.
1324    16. Updated pcredemo.c to include calls to free() to release the memory for the
1325        compiled pattern.
1327    17. The generated file chartables.c was being created in the source directory
1328        instead of in the building directory. This caused the build to fail if the
1329        source directory was different from the building directory, and was
1330        read-only.
1332    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1333        file. No doubt somebody will tell me if they don't make sense... Also added
1334        Dan Mooney's comments about building on OpenVMS.
1336    19. Added support for partial matching via the PCRE_PARTIAL option for
1337        pcre_exec() and the \P data escape in pcretest.
1339    20. Extended pcretest with 3 new pattern features:
1341        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
1342              write the compiled pattern to the file whose name is "rest-of-line".
1343              This is a straight binary dump of the data, with the saved pointer to
1344              the character tables forced to be NULL. The study data, if any, is
1345              written too. After writing, pcretest reads a new pattern.
1347        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1348              compiled pattern from the given file. There must not be any
1349              occurrences of "<" in the file name (pretty unlikely); if there are,
1350              pcretest will instead treat the initial "<" as a pattern delimiter.
1351              After reading in the pattern, pcretest goes on to read data lines as
1352              usual.
1354        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1355              and 16-bit fields in a compiled pattern, to simulate a pattern that
1356              was compiled on a host of opposite endianness.
1358    21. The pcre-exec() function can now cope with patterns that were compiled on
1359        hosts of opposite endianness, with this restriction:
1361          As for any compiled expression that is saved and used later, the tables
1362          pointer field cannot be preserved; the extra_data field in the arguments
1363          to pcre_exec() should be used to pass in a tables address if a value
1364          other than the default internal tables were used at compile time.
1366    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1367        now diagnosed as an error. Previously, most of the time, a negative number
1368        would have been treated as zero, but if in addition "ovector" was passed as
1369        NULL, a crash could occur.
1371    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1372        new versions from the libtool 1.5 distribution (the last one is a copy of
1373        a file called libtool.m4). This seems to have fixed the need to patch
1374        "configure" to support Darwin 1.3 (which I used to do). However, I still
1375        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1376        workstation).
1378    24. Changed the PCRE licence to be the more standard "BSD" licence.
1381    Version 4.5 01-Dec-03
1382    ---------------------
1384     1. There has been some re-arrangement of the code for the match() function so
1385        that it can be compiled in a version that does not call itself recursively.
1386        Instead, it keeps those local variables that need separate instances for
1387        each "recursion" in a frame on the heap, and gets/frees frames whenever it
1388        needs to "recurse". Keeping track of where control must go is done by means
1389        of setjmp/longjmp. The whole thing is implemented by a set of macros that
1390        hide most of the details from the main code, and operates only if
1391        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1392        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1393        operating.
1395        To make it easier for callers to provide specially tailored get/free
1396        functions for this usage, two new functions, pcre_stack_malloc, and
1397        pcre_stack_free, are used. They are always called in strict stacking order,
1398        and the size of block requested is always the same.
1400        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1401        PCRE has been compiled to use the stack or the heap for recursion. The
1402        -C option of pcretest uses this to show which version is compiled.
1404        A new data escape \S, is added to pcretest; it causes the amounts of store
1405        obtained and freed by both kinds of malloc/free at match time to be added
1406        to the output.
1408     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1409        what's available on my current Linux desktop machine.
1411     3. When matching a UTF-8 string, the test for a valid string at the start has
1412        been extended. If start_offset is not zero, PCRE now checks that it points
1413        to a byte that is the start of a UTF-8 character. If not, it returns
1414        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1415        this is necessary because there may be backward assertions in the pattern.
1416        When matching the same subject several times, it may save resources to use
1417        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1419     4. The code for checking the validity of UTF-8 strings has been tightened so
1420        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1421        containing "overlong sequences".
1423     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1424        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1425        should have been "&", but it just so happened that all the cases this let
1426        through by mistake were picked up later in the function.
1428     6. I had used a variable called "isblank" - this is a C99 function, causing
1429        some compilers to warn. To avoid this, I renamed it (as "blankclass").
1431     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1432        prompting; (b) run "./pcretest /dev/null" at the start of the test script
1433        so the version is shown; (c) stop "make test" echoing "./RunTest".
1435     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1437     9. The prototype for memmove() for systems that don't have it was using
1438        size_t, but the inclusion of the header that defines size_t was later. I've
1439        moved the #includes for the C headers earlier to avoid this.
1441    10. Added some adjustments to the code to make it easier to compiler on certain
1442        special systems:
1444          (a) Some "const" qualifiers were missing.
1445          (b) Added the macro EXPORT before all exported functions; by default this
1446              is defined to be empty.
1447          (c) Changed the dftables auxiliary program (that builds chartables.c) so
1448              that it reads its output file name as an argument instead of writing
1449              to the standard output and assuming this can be redirected.
1451    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
1452        class containing characters with values greater than 255, PCRE compilation
1453        went into a loop.
1455    12. A recursive reference to a subpattern that was within another subpattern
1456        that had a minimum quantifier of zero caused PCRE to crash. For example,
1457        (x(y(?2))z)? provoked this bug with a subject that got as far as the
1458        recursion. If the recursively-called subpattern itself had a zero repeat,
1459        that was OK.
1461    13. In pcretest, the buffer for reading a data line was set at 30K, but the
1462        buffer into which it was copied (for escape processing) was still set at
1463        1024, so long lines caused crashes.
1465    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
1466        "internal error: code overflow...". This applied to any character class
1467        that was followed by a possessive quantifier.
1469    15. Modified the Makefile to add libpcre.la as a prerequisite for
1470        libpcreposix.la because I was told this is needed for a parallel build to
1471        work.
1473    16. If a pattern that contained .* following optional items at the start was
1474        studied, the wrong optimizing data was generated, leading to matching
1475        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
1476        matching string must start with a or b or c. The correct conclusion for
1477        this pattern is that a match can start with any character.
1480    Version 4.4 13-Aug-03
1481    ---------------------
1483     1. In UTF-8 mode, a character class containing characters with values between
1484        127 and 255 was not handled correctly if the compiled pattern was studied.
1485        In fixing this, I have also improved the studying algorithm for such
1486        classes (slightly).
1488     2. Three internal functions had redundant arguments passed to them. Removal
1489        might give a very teeny performance improvement.
1491     3. Documentation bug: the value of the capture_top field in a callout is *one
1492        more than* the number of the hightest numbered captured substring.
1494     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
1495        in incorrectly linking with a previously installed version. They now link
1496        explicitly with libpcre.la.
1498     5. configure.in no longer needs to recognize Cygwin specially.
1500     6. A problem in pcre.in for Windows platforms is fixed.
1502     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
1503        pcretest, it used to include the size of the study block as part of its
1504        output. Unfortunately, the structure contains a field that has a different
1505        size on different hardware architectures. This meant that the tests that
1506        showed this size failed. As the block is currently always of a fixed size,
1507        this information isn't actually particularly useful in pcretest output, so
1508        I have just removed it.
1510     8. Three pre-processor statements accidentally did not start in column 1.
1511        Sadly, there are *still* compilers around that complain, even though
1512        standard C has not required this for well over a decade. Sigh.
1514     9. In pcretest, the code for checking callouts passed small integers in the
1515        callout_data field, which is a void * field. However, some picky compilers
1516        complained about the casts involved for this on 64-bit systems. Now
1517        pcretest passes the address of the small integer instead, which should get
1518        rid of the warnings.
1520    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
1521        both compile and run time, and gives an error if an invalid UTF-8 sequence
1522        is found. There is a option for disabling this check in cases where the
1523        string is known to be correct and/or the maximum performance is wanted.
1525    11. In response to a bug report, I changed one line in Makefile.in from
1527            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
1528        to
1529            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
1531        to look similar to other lines, but I have no way of telling whether this
1532        is the right thing to do, as I do not use Windows. No doubt I'll get told
1533        if it's wrong...
1536    Version 4.3 21-May-03
1537    ---------------------
1539    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
1540       Makefile.
1542    2. Some refactoring to improve the quality of the code:
1544       (i)   The utf8_table... variables are now declared "const".
1546       (ii)  The code for \cx, which used the "case flipping" table to upper case
1547             lower case letters, now just substracts 32. This is ASCII-specific,
1548             but the whole concept of \cx is ASCII-specific, so it seems
1549             reasonable.
1551       (iii) PCRE was using its character types table to recognize decimal and
1552             hexadecimal digits in the pattern. This is silly, because it handles
1553             only 0-9, a-f, and A-F, but the character types table is locale-
1554             specific, which means strange things might happen. A private
1555             table is now used for this - though it costs 256 bytes, a table is
1556             much faster than multiple explicit tests. Of course, the standard
1557             character types table is still used for matching digits in subject
1558             strings against \d.
1560       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
1561             ending in _t are). So I've renamed it as ESC_tee.
1563    3. The first argument for regexec() in the POSIX wrapper should have been
1564       defined as "const".
1566    4. Changed pcretest to use malloc() for its buffers so that they can be
1567       Electric Fenced for debugging.
1569    5. There were several places in the code where, in UTF-8 mode, PCRE would try
1570       to read one or more bytes before the start of the subject string. Often this
1571       had no effect on PCRE's behaviour, but in some circumstances it could
1572       provoke a segmentation fault.
1574    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
1575       to try to read one or more bytes before the start of the subject string.
1577    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
1578       UTF-8 support could misbehave in various ways if the subject string
1579       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
1580       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
1581       back over UTF-8 characters.)
1584    Version 4.2 14-Apr-03
1585    ---------------------
1587    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
1589    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
1590         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
1591         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
1592         [WIN_PREFIX]: new variable, "cyg" for Cygwin
1593         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
1594           and BUILD_EXEEXT
1595         Note: automatic setting of the BUILD variables is not yet working
1596         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
1597           compile-time but not at link-time
1598         [LINK]: use for linking executables only
1599         make different versions for Windows and non-Windows
1600         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
1601           libraries
1602         [LINK_FOR_BUILD]: new variable
1603         [OBJEXT]: use throughout
1604         [EXEEXT]: use throughout
1605         <winshared>: new target
1606         <wininstall>: new target
1607         <dftables.o>: use native compiler
1608         <dftables>: use native linker
1609         <install>: handle Windows platform correctly
1610         <clean>: ditto
1611         <check>: ditto
1612         copy DLL to top builddir before testing
1614       As part of these changes, -no-undefined was removed again. This was reported
1615       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
1616       in any case.
1618    3. Some tidies to get rid of compiler warnings:
1620       . In the match_data structure, match_limit was an unsigned long int, whereas
1621         match_call_count was an int. I've made them both unsigned long ints.
1623       . In pcretest the fact that a const uschar * doesn't automatically cast to
1624         a void * provoked a warning.
1626       . Turning on some more compiler warnings threw up some "shadow" variables
1627         and a few more missing casts.
1629    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
1630       option, a class that contained a single character with a value between 128
1631       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
1633    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
1634       option, a class that contained several characters, but with at least one
1635       whose value was between 128 and 255 caused PCRE to crash.
1638    Version 4.1 12-Mar-03
1639    ---------------------
1641    1. Compiling with gcc -pedantic found a couple of places where casts were
1642    needed, and a string in dftables.c that was longer than standard compilers are
1643    required to support.
1645    2. Compiling with Sun's compiler found a few more places where the code could
1646    be tidied up in order to avoid warnings.
1648    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
1649    first of these names is deprecated in the latest Autoconf in favour of the name
1650    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
1651    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
1652    analogy I have changed it to CFLAGS_FOR_BUILD.
1654    4. Added -no-undefined to the linking command in the Makefile, because this is
1655    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
1656    linking step for the pcreposix library.
1658    5. PCRE was failing to diagnose the case of two named groups with the same
1659    name.
1661    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
1662    literal character that is needed in the subject for a match, and scans along to
1663    ensure that it is present before embarking on the full matching process. This
1664    saves time in cases of nested unlimited repeats that are never going to match.
1665    Problem: the scan can take a lot of time if the subject is very long (e.g.
1666    megabytes), thus penalizing straightforward matches. It is now done only if the
1667    amount of subject to be scanned is less than 1000 bytes.
1669    7. A lesser problem with the same optimization is that it was recording the
1670    first character of an anchored pattern as "needed", thus provoking a search
1671    right along the subject, even when the first match of the pattern was going to
1672    fail. The "needed" character is now not set for anchored patterns, unless it
1673    follows something in the pattern that is of non-fixed length. Thus, it still
1674    fulfils its original purpose of finding quick non-matches in cases of nested
1675    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
1678    Version 4.0 17-Feb-03
1679    ---------------------
1681    1. If a comment in an extended regex that started immediately after a meta-item
1682    extended to the end of string, PCRE compiled incorrect data. This could lead to
1683    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
1685    2. Moved to autoconf 2.53 and libtool 1.4.2.
1687    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
1688    the special perltest8 script is no longer needed - all the tests can be run
1689    from a single perltest script.
1691    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
1692    by \s. It has now been removed in PCRE. This means it isn't recognized as
1693    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
1694    class [:space:] *does* include VT, thereby creating a mess.
1696    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
1697    space and tab.
1699    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
1700    its new features into the main test script, reducing the number of scripts.
1702    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
1703    were backward compatible, and made the (?i) apply to the whole pattern, as if
1704    /i were given. Now it behaves more logically, and applies the option setting
1705    only to what follows. PCRE has been changed to follow suit. However, if it
1706    finds options settings right at the start of the pattern, it extracts them into
1707    the global options, as before. Thus, they show up in the info data.
1709    8. Added support for the \Q...\E escape sequence. Characters in between are
1710    treated as literals. This is slightly different from Perl in that $ and @ are
1711    also handled as literals inside the quotes. In Perl, they will cause variable
1712    interpolation. Note the following examples:
1714        Pattern            PCRE matches      Perl matches
1716        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
1717        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
1718        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
1720    For compatibility with Perl, \Q...\E sequences are recognized inside character
1721    classes as well as outside them.
1723    9. Re-organized 3 code statements in pcretest to avoid "overflow in
1724    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
1725    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
1726    signed/unsigned warnings.
1728    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
1729    option for pcretest, so I've replaced it by a simple function that does just
1730    that job.
1732    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
1733    "pcregrep -".
1735    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
1736    Java package. This provides some syntactic sugar for simple cases of what my
1737    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
1738    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
1739    item, you can use this simplified notation. Note that only makes sense with
1740    greedy quantifiers. Consequently, the use of the possessive quantifier forces
1741    greediness, whatever the setting of the PCRE_UNGREEDY option.
1743    13. A change of greediness default within a pattern was not taking effect at
1744    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
1745    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
1746    was abstracted outside.
1748    14. PCRE now supports the \G assertion. It is true when the current matching
1749    position is at the start point of the match. This differs from \A when the
1750    starting offset is non-zero. Used with the /g option of pcretest (or similar
1751    code), it works in the same way as it does for Perl's /g option. If all
1752    alternatives of a regex begin with \G, the expression is anchored to the start
1753    match position, and the "anchored" flag is set in the compiled expression.
1755    15. Some bugs concerning the handling of certain option changes within patterns
1756    have been fixed. These applied to options other than (?ims). For example,
1757    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
1758    been the other way round. Some of this was related to change 7 above.
1760    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
1761    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
1762    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
1763    POSIX classes only within a class (e.g. /[[:alpha:]]/).
1765    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
1766    mode. Unlike ".", it always matches newline, whatever the setting of
1767    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
1768    assertions. Perl allows it, but it doesn't (in general) work because it can't
1769    calculate the length of the lookbehind. At least, that's the case for Perl
1770    5.8.0 - I've been told they are going to document that it doesn't work in
1771    future.
1773    18. Added an error diagnosis for escapes that PCRE does not support: these are
1774    \L, \l, \N, \P, \p, \U, \u, and \X.
1776    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
1777    reading past the end of the pattern in cases such as /[abcd/.
1779    20. PCRE was getting more memory than necessary for patterns with classes that
1780    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
1782    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
1783    compiling PCRE for use with Virtual Pascal.
1785    22. Small fix to the Makefile to make it work properly if the build is done
1786    outside the source tree.
1788    23. Added a new extension: a condition to go with recursion. If a conditional
1789    subpattern starts with (?(R) the "true" branch is used if recursion has
1790    happened, whereas the "false" branch is used only at the top level.
1792    24. When there was a very long string of literal characters (over 255 bytes
1793    without UTF support, over 250 bytes with UTF support), the computation of how
1794    much memory was required could be incorrect, leading to segfaults or other
1795    strange effects.
1797    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
1798    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
1799    there was a subsequent back reference to those brackets. This meant that, for
1800    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
1801    possible to check for precisely this case. All we can do is abandon the
1802    optimization if .* occurs inside capturing brackets when there are any back
1803    references whatsoever. (See below for a better fix that came later.)
1805    26. The handling of the optimization for finding the first character of a
1806    non-anchored pattern, and for finding a character that is required later in the
1807    match were failing in some cases. This didn't break the matching; it just
1808    failed to optimize when it could. The way this is done has been re-implemented.
1810    27. Fixed typo in error message for invalid (?R item (it said "(?p").
1812    28. Added a new feature that provides some of the functionality that Perl
1813    provides with (?{...}). The facility is termed a "callout". The way it is done
1814    in PCRE is for the caller to provide an optional function, by setting
1815    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
1816    global variable. By default it is unset, which disables all calling out. To get
1817    the function called, the regex must include (?C) at appropriate points. This
1818    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
1819    This provides a means of identifying different callout points. When PCRE
1820    reaches such a point in the regex, if pcre_callout has been set, the external
1821    function is called. It is provided with data in a structure called
1822    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
1823    matching continues; if it returns a non-zero value, the match at the current
1824    point fails. However, backtracking will occur if possible. [This was changed
1825    later and other features added - see item 49 below.]
1827    29. pcretest is upgraded to test the callout functionality. It provides a
1828    callout function that displays information. By default, it shows the start of
1829    the match and the current position in the text. There are some new data escapes
1830    to vary what happens:
1832        \C+         in addition, show current contents of captured substrings
1833        \C-         do not supply a callout function
1834        \C!n        return 1 when callout number n is reached
1835        \C!n!m      return 1 when callout number n is reached for the mth time
1837    30. If pcregrep was called with the -l option and just a single file name, it
1838    output "<stdin>" if a match was found, instead of the file name.
1840    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
1841    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
1842    pcre_exec(). This saves a malloc/free per call. The default value of
1843    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
1844    when configuring.
1846    32. The default maximum size of a compiled pattern is 64K. There have been a
1847    few cases of people hitting this limit. The code now uses macros to handle the
1848    storing of links as offsets within the compiled pattern. It defaults to 2-byte
1849    links, but this can be changed to 3 or 4 bytes by --with-link-size when
1850    configuring. Tests 2 and 5 work only with 2-byte links because they output
1851    debugging information about compiled patterns.
1853    33. Internal code re-arrangements:
1855    (a) Moved the debugging function for printing out a compiled regex into
1856        its own source file (printint.c) and used #include to pull it into
1857        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
1858        separate copies.
1860    (b) Defined the list of op-code names for debugging as a macro in
1861        internal.h so that it is next to the definition of the opcodes.
1863    (c) Defined a table of op-code lengths for simpler skipping along compiled
1864        code. This is again a macro in internal.h so that it is next to the
1865        definition of the opcodes.
1867    34. Added support for recursive calls to individual subpatterns, along the
1868    lines of Robin Houston's patch (but implemented somewhat differently).
1870    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
1871    allow it to read and process whole directories in Win32. This code was
1872    contributed by Lionel Fourquaux; it has not been tested by me.
1874    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
1875    used to name a group. Names consist of alphanumerics and underscores, and must
1876    be unique. Back references use the syntax (?P=name) and recursive calls use
1877    (?P>name) which is a PCRE extension to the Python extension. Groups still have
1878    numbers. The function pcre_fullinfo() can be used after compilation to extract
1879    a name/number map. There are three relevant calls:
1881      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
1882      PCRE_INFO_NAMECOUNT            yields the number of entries
1883      PCRE_INFO_NAMETABLE            yields a pointer to the map.
1885    The map is a vector of fixed-size entries. The size of each entry depends on
1886    the length of the longest name used. The first two bytes of each entry are the
1887    group number, most significant byte first. There follows the corresponding
1888    name, zero terminated. The names are in alphabetical order.
1890    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
1891    case instead of 255. Making it the same both with and without UTF-8 support
1892    means that the same test output works with both.
1894    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
1895    calling malloc() with a zero argument.
1897    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
1898    optimization. I've improved things by keeping a bitmap of backreferences with
1899    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
1900    fact referenced, the optimization can be applied. It is unlikely that a
1901    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
1902    the match to follow \n) will appear inside brackets with a number greater than
1903    31, but if it does, any back reference > 31 suppresses the optimization.
1905    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
1906    of disabling numbered capturing parentheses. Any opening parenthesis that is
1907    not followed by ? behaves as if it were followed by ?: but named parentheses
1908    can still be used for capturing (and they will acquire numbers in the usual
1909    way).
1911    41. Redesigned the return codes from the match() function into yes/no/error so
1912    that errors can be passed back from deep inside the nested calls. A malloc
1913    failure while inside a recursive subpattern call now causes the
1914    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
1916    42. It is now possible to set a limit on the number of times the match()
1917    function is called in a call to pcre_exec(). This facility makes it possible to
1918    limit the amount of recursion and backtracking, though not in a directly
1919    obvious way, because the match() function is used in a number of different
1920    circumstances. The count starts from zero for each position in the subject
1921    string (for non-anchored patterns). The default limit is, for compatibility, a
1922    large number, namely 10 000 000. You can change this in two ways:
1924    (a) When configuring PCRE before making, you can use --with-match-limit=n
1925        to set a default value for the compiled library.
1927    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
1928        a different value is set. See 45 below.
1930    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
1932    43. Added a new function pcre_config(int, void *) to enable run-time extraction
1933    of things that can be changed at compile time. The first argument specifies
1934    what is wanted and the second points to where the information is to be placed.
1935    The current list of available information is:
1937      PCRE_CONFIG_UTF8
1939    The output is an integer that is set to one if UTF-8 support is available;
1940    otherwise it is set to zero.
1944    The output is an integer that it set to the value of the code that is used for
1945    newline. It is either LF (10) or CR (13).
1949    The output is an integer that contains the number of bytes used for internal
1950    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
1954    The output is an integer that contains the threshold above which the POSIX
1955    interface uses malloc() for output vectors. See item 31 above.
1959    The output is an unsigned integer that contains the default limit of the number
1960    of match() calls in a pcre_exec() execution. See 42 above.
1962    44. pcretest has been upgraded by the addition of the -C option. This causes it
1963    to extract all the available output from the new pcre_config() function, and to
1964    output it. The program then exits immediately.
1966    45. A need has arisen to pass over additional data with calls to pcre_exec() in
1967    order to support additional features. One way would have been to define
1968    pcre_exec2() (for example) with extra arguments, but this would not have been
1969    extensible, and would also have required all calls to the original function to
1970    be mapped to the new one. Instead, I have chosen to extend the mechanism that
1971    is used for passing in "extra" data from pcre_study().
1973    The pcre_extra structure is now exposed and defined in pcre.h. It currently
1974    contains the following fields:
1976      flags         a bitmap indicating which of the following fields are set
1977      study_data    opaque data from pcre_study()
1978      match_limit   a way of specifying a limit on match() calls for a specific
1979                      call to pcre_exec()
1980      callout_data  data for callouts (see 49 below)
1982    The flag bits are also defined in pcre.h, and are
1988    The pcre_study() function now returns one of these new pcre_extra blocks, with
1989    the actual study data pointed to by the study_data field, and the
1990    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
1991    before. That is, this change is entirely upwards-compatible and requires no
1992    change to existing code.
1994    If you want to pass in additional data to pcre_exec(), you can either place it
1995    in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
1996    block.
1998    46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
1999    data string contains the escape sequence \M, pcretest calls pcre_exec() several
2000    times with different match limits, until it finds the minimum value needed for
2001    pcre_exec() to complete. The value is then output. This can be instructive; for
2002    most simple matches the number is quite small, but for pathological cases it
2003    gets very large very quickly.
2005    47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
2006    returns the size of the data block pointed to by the study_data field in a
2007    pcre_extra block, that is, the value that was passed as the argument to
2008    pcre_malloc() when PCRE was getting memory in which to place the information
2009    created by pcre_study(). The fourth argument should point to a size_t variable.
2010    pcretest has been extended so that this information is shown after a successful
2011    pcre_study() call when information about the compiled regex is being displayed.
2013    48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
2014    because what follows is always an absolute path. (Later: it turns out that this
2015    is more than cosmetic for MinGW, because it doesn't like empty path
2016    components.)
2018    49. Some changes have been made to the callout feature (see 28 above):
2020    (i)  A callout function now has three choices for what it returns:
2022           0  =>  success, carry on matching
2023         > 0  =>  failure at this point, but backtrack if possible
2024         < 0  =>  serious error, return this value from pcre_exec()
2026         Negative values should normally be chosen from the set of PCRE_ERROR_xxx
2027         values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
2028         "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
2029         use by callout functions. It will never be used by PCRE itself.
2031    (ii) The pcre_extra structure (see 45 above) has a void * field called
2032         callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
2033         pcre_callout_block structure has a field of the same name. The contents of
2034         the field passed in the pcre_extra structure are passed to the callout
2035         function in the corresponding field in the callout block. This makes it
2036         easier to use the same callout-containing regex from multiple threads. For
2037         testing, the pcretest program has a new data escape
2039           \C*n        pass the number n (may be negative) as callout_data
2041         If the callout function in pcretest receives a non-zero value as
2042         callout_data, it returns that value.
2044    50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
2045    there were some redundant $(CFLAGS) in commands that are now specified as
2046    $(LINK), which already includes $(CFLAGS).
2048    51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
2049    has been compiled with UTF-8 support *and* pcre_compile() has been compiled
2050    with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
2051    one-byte characters throughout. Note that case-insensitive matching applies
2052    only to characters whose values are less than 256. PCRE doesn't support the
2053    notion of cases for higher-valued characters.
2055    (i)   A character class whose characters are all within 0-255 is handled as
2056          a bit map, and the map is inverted for negative classes. Previously, a
2057          character > 255 always failed to match such a class; however it should
2058          match if the class was a negative one (e.g. [^ab]). This has been fixed.
2060    (ii)  A negated character class with a single character < 255 is coded as
2061          "not this character" (OP_NOT). This wasn't working properly when the test
2062          character was multibyte, either singly or repeated.
2064    (iii) Repeats of multibyte characters are now handled correctly in UTF-8
2065          mode, for example: \x{100}{2,3}.
2067    (iv)  The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
2068          singly or repeated) now correctly test multibyte characters. However,
2069          PCRE doesn't recognize any characters with values greater than 255 as
2070          digits, spaces, or word characters. Such characters always match \D, \S,
2071          and \W, and never match \d, \s, or \w.
2073    (v)   Classes may now contain characters and character ranges with values
2074          greater than 255. For example: [ab\x{100}-\x{400}].
2076    (vi)  pcregrep now has a --utf-8 option (synonym -u) which makes it call
2077          PCRE in UTF-8 mode.
2079    52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
2080    PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
2081    retained for backwards compatibility. (Note that LASTLITERAL is also a byte
2082    value.)
2084    53. The single man page has become too large. I have therefore split it up into
2085    a number of separate man pages. These also give rise to individual HTML pages;
2086    these are now put in a separate directory, and there is an index.html page that
2087    lists them all. Some hyperlinking between the pages has been installed.
2089    54. Added convenience functions for handling named capturing parentheses.
2091    55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2092    aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2093    true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2094    are faulted.
2096    56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2097    calling configure. These values are used when compiling the dftables.c program
2098    which is run to generate the source of the default character tables. They
2099    default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2100    you will need to set these values.
2102    57. Updated the building process for Windows DLL, as provided by Fred Cox.
2105    Version 3.9 02-Jan-02
2106  ---------------------  ---------------------
2108  1. A bit of extraneous text had somehow crept into the pcregrep documentation.  1. A bit of extraneous text had somehow crept into the pcregrep documentation.

Removed from v.61  
changed lines
  Added in v.205

  ViewVC Help
Powered by ViewVC 1.1.5