ViewVC logotype

Diff of /code/trunk/ChangeLog

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

revision 27 by nigel, Sat Feb 24 21:38:49 2007 UTC revision 319 by ph10, Mon Jan 28 15:27:45 2008 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
4    Version 7.6 28-Jan-08
5    ---------------------
7    1.  A character class containing a very large number of characters with
8        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
9        overflow.
11    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
12        HAVE_LONG_LONG is not defined.
14    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
15        bring them up to date and include new features. This patch includes:
17        - Fixed PH's badly added libz and libbz2 support.
18        - Fixed a problem with static linking.
19        - Added pcredemo. [But later removed - see 7 below.]
20        - Fixed dftables problem and added an option.
21        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
22            HAVE_LONG_LONG.
23        - Added readline support for pcretest.
24        - Added an listing of the option settings after cmake has run.
26    4.  A user submitted a patch to Makefile that makes it easy to create
27        "pcre.dll" under mingw when using Configure/Make. I added stuff to
28        Makefile.am that cause it to include this special target, without
29        affecting anything else. Note that the same mingw target plus all
30        the other distribution libraries and programs are now supported
31        when configuring with CMake (see 6 below) instead of with
32        Configure/Make.
34    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
35        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
36        exported in the Windows port". It has not yet been confirmed that the patch
37        solves the problem, but it does no harm.
39    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
40        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
41        with CMake, and also correct the comment about stack recursion.
43    7.  Remove the automatic building of pcredemo from the ./configure system and
44        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
45        of a program that users should build themselves after PCRE is installed, so
46        building it automatically is not really right. What is more, it gave
47        trouble in some build environments.
49    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
52    Version 7.5 10-Jan-08
53    ---------------------
55    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
56        values in parens when parsing an RE using the C++ wrapper."
58    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
59        Characters greater than 255 were excluded from the class instead of being
60        included.
62    3.  The same bug as (2) above applied to negated POSIX classes such as
63        [:^space:].
65    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
66        defined or documented. It seems to have been a typo for PCRE_STATIC, so
67        I have changed it.
69    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
70        first named subpattern) and a construct such as (?&a) would reference the
71        first named subpattern whose name started with "a" (in other words, the
72        length check was missing). Both these problems are fixed. "Subpattern name
73        expected" is now given for (?&) (a zero-length name), and this patch also
74        makes it give the same error for \k'' (previously it complained that that
75        was a reference to a non-existent subpattern).
77    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
78        this is right because (?- can be followed by option settings as well as by
79        digits. I have, however, made the messages clearer.
81    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
82        than the number used in the conditional) now cause a compile-time error.
83        This is actually not compatible with Perl, which accepts such patterns, but
84        treats the conditional as always being FALSE (as PCRE used to), but it
85        seems to me that giving a diagnostic is better.
87    8.  Change "alphameric" to the more common word "alphanumeric" in comments
88        and messages.
90    9.  Fix two occurrences of "backslash" in comments that should have been
91        "backspace".
93    10. Remove two redundant lines of code that can never be obeyed (their function
94        was moved elsewhere).
96    11. The program that makes PCRE's Unicode character property table had a bug
97        which caused it to generate incorrect table entries for sequences of
98        characters that have the same character type, but are in different scripts.
99        It amalgamated them into a single range, with the script of the first of
100        them. In other words, some characters were in the wrong script. There were
101        thirteen such cases, affecting characters in the following ranges:
103          U+002b0 - U+002c1
104          U+0060c - U+0060d
105          U+0061e - U+00612
106          U+0064b - U+0065e
107          U+0074d - U+0076d
108          U+01800 - U+01805
109          U+01d00 - U+01d77
110          U+01d9b - U+01dbf
111          U+0200b - U+0200f
112          U+030fc - U+030fe
113          U+03260 - U+0327f
114          U+0fb46 - U+0fbb1
115          U+10450 - U+1049d
117    12. The -o option (show only the matching part of a line) for pcregrep was not
118        compatible with GNU grep in that, if there was more than one match in a
119        line, it showed only the first of them. It now behaves in the same way as
120        GNU grep.
122    13. If the -o and -v options were combined for pcregrep, it printed a blank
123        line for every non-matching line. GNU grep prints nothing, and pcregrep now
124        does the same. The return code can be used to tell if there were any
125        non-matching lines.
127    14. Added --file-offsets and --line-offsets to pcregrep.
129    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
130        infinitely looping recursion. The bug was that positive lookaheads were not
131        being skipped when checking for a possible empty match (negative lookaheads
132        and both kinds of lookbehind were skipped).
134    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
135        inclusion of <windows.h> to before rather than after the definition of
136        INVALID_FILE_ATTRIBUTES (patch from David Byron).
138    17. Specifying a possessive quantifier with a specific limit for a Unicode
139        character property caused pcre_compile() to compile bad code, which led at
140        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
141        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
142        caused the error; without that there was no problem.
144    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
146    19. Added --enable-pcretest-libreadline.
148    20. In pcrecpp.cc, the variable 'count' was incremented twice in
149        RE::GlobalReplace(). As a result, the number of replacements returned was
150        double what it should be. I removed one of the increments, but Craig sent a
151        later patch that removed the other one (the right fix) and added unit tests
152        that check the return values (which was not done before).
154    21. Several CMake things:
156        (1) Arranged that, when cmake is used on Unix, the libraries end up with
157            the names libpcre and libpcreposix, not just pcre and pcreposix.
159        (2) The above change means that pcretest and pcregrep are now correctly
160            linked with the newly-built libraries, not previously installed ones.
164    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
165        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
166        UTF-8 newline character). The key issue is that the pattern starts .*;
167        this means that the match must be either at the beginning, or after a
168        newline. The bug was in the code for advancing after a failed match and
169        checking that the new position followed a newline. It was not taking
170        account of UTF-8 characters correctly.
172    23. PCRE was behaving differently from Perl in the way it recognized POSIX
173        character classes. PCRE was not treating the sequence [:...:] as a
174        character class unless the ... were all letters. Perl, however, seems to
175        allow any characters between [: and :], though of course it rejects as
176        unknown any "names" that contain non-letters, because all the known class
177        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
178        for example, whereas PCRE did not - it did not recognize a POSIX character
179        class. This seemed a bit dangerous, so the code has been changed to be
180        closer to Perl. The behaviour is not identical to Perl, because PCRE will
181        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
182        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
183        Perl does, and where it didn't before.
185    24. Rewrite so as to remove the single use of %n from pcregrep because in some
186        Windows environments %n is disabled by default.
189    Version 7.4 21-Sep-07
190    ---------------------
192    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
193        means that a class such as [\s] counted as "explicit reference to CR or
194        LF". That isn't really right - the whole point of the change was to try to
195        help when there was an actual mention of one of the two characters. So now
196        the change happens only if \r or \n (or a literal CR or LF) character is
197        encountered.
199    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
200        of both had grown to the point where there were only 3 bits left.
201        Fortunately, there was spare space in the data structure, and so I have
202        moved the internal flags into a new 16-bit field to free up more option
203        bits.
205    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
206        but did not set the internal JCHANGED flag - either of these is enough to
207        control the way the "get" function works - but the PCRE_INFO_JCHANGED
208        facility is supposed to tell if (?J) was ever used, so now (?J) at the
209        start sets both bits.
211    4.  Added options (at build time, compile time, exec time) to change \R from
212        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
214    5.  doc/pcresyntax.html was missing from the distribution.
216    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
217        compatibility, even though it is no longer used.
219    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
220        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
221        windows.h file is present (where different names are used). [This was
222        reversed later after testing - see 16 below.]
224    8.  Changed all #include <config.h> to #include "config.h". There were also
225        some further <pcre.h> cases that I changed to "pcre.h".
227    9.  When pcregrep was used with the --colour option, it missed the line ending
228        sequence off the lines that it output.
230    10. It was pointed out to me that arrays of string pointers cause lots of
231        relocations when a shared library is dynamically loaded. A technique of
232        using a single long string with a table of offsets can drastically reduce
233        these. I have refactored PCRE in four places to do this. The result is
234        dramatic:
236          Originally:                          290
237          After changing UCP table:            187
238          After changing error message table:   43
239          After changing table of "verbs"       36
240          After changing table of Posix names   22
242        Thanks to the folks working on Gregex for glib for this insight.
244    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
245        unicode-properties was also set.
247    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
249    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
250        checked only for CRLF.
252    14. Added casts to pcretest.c to avoid compiler warnings.
254    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
256    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
257        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
258        entirely. This removes changes made in 7 above.
260    17. The CMake files have been updated, and there is now more information about
261        building with CMake in the NON-UNIX-USE document.
264    Version 7.3 28-Aug-07
265    ---------------------
267     1. In the rejigging of the build system that eventually resulted in 7.1, the
268        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
269        brackets there is not right, since it causes compilers to look for an
270        installed pcre.h, not the version that is in the source that is being
271        compiled (which of course may be different). I have changed it back to:
273          #include "pcre.h"
275        I have a vague recollection that the change was concerned with compiling in
276        different directories, but in the new build system, that is taken care of
277        by the VPATH setting the Makefile.
279     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
280        when the subject happened to end in the byte 0x85 (e.g. if the last
281        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
282        characters but of course it shouldn't be taken as a newline when it is part
283        of another character. The bug was that, for an unlimited repeat of . in
284        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
285        characters when looking for a newline.
287     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
289     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
290        in debug output.
292     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
293        long printing in the pcrecpp unittest when running under MinGW.
295     6. ESC_K was left out of the EBCDIC table.
297     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
298        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
299        limit also applies to "virtual nesting" when a pattern is recursive, and in
300        this case 1000 isn't so big. I have been able to remove this limit at the
301        expense of backing off one optimization in certain circumstances. Normally,
302        when pcre_exec() would call its internal match() function recursively and
303        immediately return the result unconditionally, it uses a "tail recursion"
304        feature to save stack. However, when a subpattern that can match an empty
305        string has an unlimited repetition quantifier, it no longer makes this
306        optimization. That gives it a stack frame in which to save the data for
307        checking that an empty string has been matched. Previously this was taken
308        from the 1000-entry workspace that had been reserved. So now there is no
309        explicit limit, but more stack is used.
311     8. Applied Daniel's patches to solve problems with the import/export magic
312        syntax that is required for Windows, and which was going wrong for the
313        pcreposix and pcrecpp parts of the library. These were overlooked when this
314        problem was solved for the main library.
316     9. There were some crude static tests to avoid integer overflow when computing
317        the size of patterns that contain repeated groups with explicit upper
318        limits. As the maximum quantifier is 65535, the maximum group length was
319        set at 30,000 so that the product of these two numbers did not overflow a
320        32-bit integer. However, it turns out that people want to use groups that
321        are longer than 30,000 bytes (though not repeat them that many times).
322        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
323        made it possible to implement the integer overflow checks in a much more
324        dynamic way, which I have now done. The artificial limitation on group
325        length has been removed - we now have only the limit on the total length of
326        the compiled pattern, which depends on the LINK_SIZE setting.
328    10. Fixed a bug in the documentation for get/copy named substring when
329        duplicate names are permitted. If none of the named substrings are set, the
330        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
331        empty string.
333    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
334        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
335        because the ] is interpreted as the first data character and the
336        terminating ] is not found. PCRE has been made compatible with Perl in this
337        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
338        cause memory overwriting.
340    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
341        string has been matched (to stop an infinite loop). It was not recognizing
342        a conditional subpattern that could match an empty string if that
343        subpattern was within another subpattern. For example, it looped when
344        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
345        condition was not nested. This bug has been fixed.
347    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
348        past the start of the subject in the presence of bytes with the top bit
349        set, for example "\x8aBCD".
351    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
352        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
354    14. Optimized (?!) to (*FAIL).
356    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
357        This restricts code points to be within the range 0 to 0x10FFFF, excluding
358        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
359        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
360        does: it's just the validity check that is more restrictive.
362    16. Inserted checks for integer overflows during escape sequence (backslash)
363        processing, and also fixed erroneous offset values for syntax errors during
364        backslash processing.
366    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
367        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
369    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
370        caused an overrun.
372    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
373        something other than just ASCII characters) inside a group that had an
374        unlimited repeat caused a loop at compile time (while checking to see
375        whether the group could match an empty string).
377    20. Debugging a pattern containing \p or \P could cause a crash. For example,
378        [\P{Any}] did so. (Error in the code for printing property names.)
380    21. An orphan \E inside a character class could cause a crash.
382    22. A repeated capturing bracket such as (A)? could cause a wild memory
383        reference during compilation.
385    23. There are several functions in pcre_compile() that scan along a compiled
386        expression for various reasons (e.g. to see if it's fixed length for look
387        behind). There were bugs in these functions when a repeated \p or \P was
388        present in the pattern. These operators have additional parameters compared
389        with \d, etc, and these were not being taken into account when moving along
390        the compiled data. Specifically:
392        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
393            length.
395        (b) An item such as \pL+ within a repeated group could cause crashes or
396            loops.
398        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
399            "reference to non-existent subpattern" error.
401        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
403    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
404        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
406    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
408    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
409        character were causing crashes (broken optimization).
411    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
412        \p or \P) caused a compile-time loop.
414    28. More problems have arisen in unanchored patterns when CRLF is a valid line
415        break. For example, the unstudied pattern [\r\n]A does not match the string
416        "\r\nA" because change 7.0/46 below moves the current point on by two
417        characters after failing to match at the start. However, the pattern \nA
418        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
419        the same is true. There doesn't seem any very clean way out of this, but
420        what I have chosen to do makes the common cases work: PCRE now takes note
421        of whether there can be an explicit match for \r or \n anywhere in the
422        pattern, and if so, 7.0/46 no longer applies. As part of this change,
423        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
424        pattern has explicit CR or LF references.
426    29. Added (*CR) etc for changing newline setting at start of pattern.
429    Version 7.2 19-Jun-07
430    ---------------------
432     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
433        which is apparently normally available under Windows.
435     2. Re-jig the pcregrep tests with different newline settings in an attempt
436        to make them independent of the local environment's newline setting.
438     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
440     4. Some of the "internals" tests were previously cut out when the link size
441        was not 2, because the output contained actual offsets. The recent new
442        "Z" feature of pcretest means that these can be cut out, making the tests
443        usable with all link sizes.
445     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
446        stack recursion. This gives a massive performance boost under BSD, but just
447        a small improvement under Linux. However, it saves one field in the frame
448        in all cases.
450     6. Added more features from the forthcoming Perl 5.10:
452        (a) (?-n) (where n is a string of digits) is a relative subroutine or
453            recursion call. It refers to the nth most recently opened parentheses.
455        (b) (?+n) is also a relative subroutine call; it refers to the nth next
456            to be opened parentheses.
458        (c) Conditions that refer to capturing parentheses can be specified
459            relatively, for example, (?(-2)... or (?(+3)...
461        (d) \K resets the start of the current match so that everything before
462            is not part of it.
464        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
466        (f) \g{name} is another synonym - part of Perl 5.10's unification of
467            reference syntax.
469        (g) (?| introduces a group in which the numbering of parentheses in each
470            alternative starts with the same number.
472        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
474     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
477     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
478        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
479        for detecting groups that can match an empty string.
481     9. A pattern with a very large number of alternatives (more than several
482        hundred) was running out of internal workspace during the pre-compile
483        phase, where pcre_compile() figures out how much memory will be needed. A
484        bit of new cunning has reduced the workspace needed for groups with
485        alternatives. The 1000-alternative test pattern now uses 12 bytes of
486        workspace instead of running out of the 4096 that are available.
488    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
490    11. Applied patch from Google to remove an optimization that didn't quite work.
491        The report of the bug said:
493          pcrecpp::RE("a*").FullMatch("aaa") matches, while
494          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
495          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
497    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
498        it matched the wrong number of bytes.
501    Version 7.1 24-Apr-07
502    ---------------------
504     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
505        that is more "standard", making use of automake and other Autotools. There
506        is some re-arrangement of the files and adjustment of comments consequent
507        on this.
509     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
510        for recursive directory scanning broke on some systems because the files
511        are not scanned in any specific order and on different systems the order
512        was different. A call to "sort" has been inserted into RunGrepTest for the
513        approprate test as a short-term fix. In the longer term there may be an
514        alternative.
516     3. I had an email from Eric Raymond about problems translating some of PCRE's
517        man pages to HTML (despite the fact that I distribute HTML pages, some
518        people do their own conversions for various reasons). The problems
519        concerned the use of low-level troff macros .br and .in. I have therefore
520        removed all such uses from the man pages (some were redundant, some could
521        be replaced by .nf/.fi pairs). The 132html script that I use to generate
522        HTML has been updated to handle .nf/.fi and to complain if it encounters
523        .br or .in.
525     4. Updated comments in configure.ac that get placed in config.h.in and also
526        arranged for config.h to be included in the distribution, with the name
527        config.h.generic, for the benefit of those who have to compile without
528        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
530     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
531        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
532        makevp.bat for the current PCRE, using the additional files
533        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
535     6. A Windows user reported a minor discrepancy with test 2, which turned out
536        to be caused by a trailing space on an input line that had got lost in his
537        copy. The trailing space was an accident, so I've just removed it.
539     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
540        that is needed.
542     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
543        as "const" (a) because they are and (b) because it helps the PHP
544        maintainers who have recently made a script to detect big data structures
545        in the php code that should be moved to the .rodata section. I remembered
546        to update Builducptable as well, so it won't revert if ucptable.h is ever
547        re-created.
549     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
550        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
551        order to be able to cut out the UTF-8 tables in the latter when UTF-8
552        support is not required. This saves 1.5-2K of code, which is important in
553        some applications.
555        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
556        so as not to refer to the tables, even though these functions will never be
557        called when UTF-8 support is disabled. Otherwise there are problems with a
558        shared library.
560    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
562        (a) It was defining its arguments as char * instead of void *.
564        (b) It was assuming that all moves were upwards in memory; this was true
565            a long time ago when I wrote it, but is no longer the case.
567        The emulated memove() is provided for those environments that have neither
568        memmove() nor bcopy(). I didn't think anyone used it these days, but that
569        is clearly not the case, as these two bugs were recently reported.
571    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
572        and Detrail to create the HTML documentation, the .txt form of the man
573        pages, and it removes trailing spaces from listed files. It also creates
574        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
575        case, it wraps all the #defines with #ifndefs. This script should be run
576        before "make dist".
578    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
579        with Unicode property support.
581        (a) For a maximizing quantifier, if the two different cases of the
582            character were of different lengths in their UTF-8 codings (there are
583            some cases like this - I found 11), and the matching function had to
584            back up over a mixture of the two cases, it incorrectly assumed they
585            were both the same length.
587        (b) When PCRE was configured to use the heap rather than the stack for
588            recursion during matching, it was not correctly preserving the data for
589            the other case of a UTF-8 character when checking ahead for a match
590            while processing a minimizing repeat. If the check also involved
591            matching a wide character, but failed, corruption could cause an
592            erroneous result when trying to check for a repeat of the original
593            character.
595    13. Some tidying changes to the testing mechanism:
597        (a) The RunTest script now detects the internal link size and whether there
598            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
599            values substituted by "configure". (The RunGrepTest script already did
600            this for UTF-8.) The configure.ac script no longer substitutes the
601            relevant variables.
603        (b) The debugging options /B and /D in pcretest show the compiled bytecode
604            with length and offset values. This means that the output is different
605            for different internal link sizes. Test 2 is skipped for link sizes
606            other than 2 because of this, bypassing the problem. Unfortunately,
607            there was also a test in test 3 (the locale tests) that used /B and
608            failed for link sizes other than 2. Rather than cut the whole test out,
609            I have added a new /Z option to pcretest that replaces the length and
610            offset values with spaces. This is now used to make test 3 independent
611            of link size. (Test 2 will be tidied up later.)
613    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
614        segmentation fault instead of returning the appropriate error message.
616    15. In multiline mode when the newline sequence was set to "any", the pattern
617        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
618        This doesn't seem right; it now treats the CRLF combination as the line
619        ending, and so does not match in that case. It's only a pattern such as ^$
620        that would hit this one: something like ^ABC$ would have failed after \r
621        and then tried again after \r\n.
623    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
624        in an attempt to make files that differ only in their line terminators
625        compare equal. This works on Linux.
627    17. Under certain error circumstances pcregrep might try to free random memory
628        as it exited. This is now fixed, thanks to valgrind.
630    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
631        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
632        was because its rules for how to advance for /g after matching an empty
633        string at the end of a line did not allow for this case. They now check for
634        it specially.
636    20. pcretest is supposed to handle patterns and data of any length, by
637        extending its buffers when necessary. It was getting this wrong when the
638        buffer for a data line had to be extended.
640    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
641        CRLF as a newline sequence.
643    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
644        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
645        I have nevertheless tidied it up.
647    23. Added some casts to kill warnings from HP-UX ia64 compiler.
649    24. Added a man page for pcre-config.
652    Version 7.0 19-Dec-06
653    ---------------------
655     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
656        moving to gcc 4.1.1.
658     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
659        sys/time.h, which is documented as needed for this function. It doesn't
660        seem to matter on Linux, but it showed up on some releases of OS X.
662     3. It seems that there are systems where bytes whose values are greater than
663        127 match isprint() in the "C" locale. The "C" locale should be the
664        default when a C program starts up. In most systems, only ASCII printing
665        characters match isprint(). This difference caused the output from pcretest
666        to vary, making some of the tests fail. I have changed pcretest so that:
668        (a) When it is outputting text in the compiled version of a pattern, bytes
669            other than 32-126 are always shown as hex escapes.
671        (b) When it is outputting text that is a matched part of a subject string,
672            it does the same, unless a different locale has been set for the match
673            (using the /L modifier). In this case, it uses isprint() to decide.
675     4. Fixed a major bug that caused incorrect computation of the amount of memory
676        required for a compiled pattern when options that changed within the
677        pattern affected the logic of the preliminary scan that determines the
678        length. The relevant options are -x, and -i in UTF-8 mode. The result was
679        that the computed length was too small. The symptoms of this bug were
680        either the PCRE error "internal error: code overflow" from pcre_compile(),
681        or a glibc crash with a message such as "pcretest: free(): invalid next
682        size (fast)". Examples of patterns that provoked this bug (shown in
683        pcretest format) are:
685          /(?-x: )/x
686          /(?x)(?-x: \s*#\s*)/
687          /((?i)[\x{c0}])/8
688          /(?i:[\x{c0}])/8
690        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
691        is now done differently.
693     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
694        wrapper classes; (b) implement a new function in the C++ scanner that is
695        more efficient than the old way of doing things because it avoids levels of
696        recursion in the regex matching; (c) add a paragraph to the documentation
697        for the FullMatch() function.
699     6. The escape sequence \n was being treated as whatever was defined as
700        "newline". Not only was this contrary to the documentation, which states
701        that \n is character 10 (hex 0A), but it also went horribly wrong when
702        "newline" was defined as CRLF. This has been fixed.
704     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
705        was being set to -1 for the "end of line" case (supposedly a value that no
706        character can have). Though this value is never used (the check for end of
707        line is "zero bytes in current character"), it caused compiler complaints.
708        I've changed it to 0xffffffff.
710     8. In pcre_version.c, the version string was being built by a sequence of
711        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
712        string (as it is for production releases) called a macro with an empty
713        argument. The C standard says the result of this is undefined. The gcc
714        compiler treats it as an empty string (which was what was wanted) but it is
715        reported that Visual C gives an error. The source has been hacked around to
716        avoid this problem.
718     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
719        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
720        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
721        of them did).
723    10. Originally, pcretest opened its input and output without "b"; then I was
724        told that "b" was needed in some environments, so it was added for release
725        5.0 to both the input and output. (It makes no difference on Unix-like
726        systems.) Later I was told that it is wrong for the input on Windows. I've
727        now abstracted the modes into two macros, to make it easier to fiddle with
728        them, and removed "b" from the input mode under Windows.
730    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
732    12. Added -help and --help to pcretest as an official way of being reminded
733        of the options.
735    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
736        and pcrecpp.cc because they annoy compilers at high warning levels.
738    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
740    15. Fixed an occurrence of == in configure.ac that should have been = (shell
741        scripts are not C programs :-) and which was not noticed because it works
742        on Linux.
744    16. pcretest is supposed to handle any length of pattern and data line (as one
745        line or as a continued sequence of lines) by extending its input buffer if
746        necessary. This feature was broken for very long pattern lines, leading to
747        a string of junk being passed to pcre_compile() if the pattern was longer
748        than about 50K.
750    17. I have done a major re-factoring of the way pcre_compile() computes the
751        amount of memory needed for a compiled pattern. Previously, there was code
752        that made a preliminary scan of the pattern in order to do this. That was
753        OK when PCRE was new, but as the facilities have expanded, it has become
754        harder and harder to keep it in step with the real compile phase, and there
755        have been a number of bugs (see for example, 4 above). I have now found a
756        cunning way of running the real compile function in a "fake" mode that
757        enables it to compute how much memory it would need, while actually only
758        ever using a few hundred bytes of working memory and without too many
759        tests of the mode. This should make future maintenance and development
760        easier. A side effect of this work is that the limit of 200 on the nesting
761        depth of parentheses has been removed (though this was never a serious
762        limitation, I suspect). However, there is a downside: pcre_compile() now
763        runs more slowly than before (30% or more, depending on the pattern). I
764        hope this isn't a big issue. There is no effect on runtime performance.
766    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
767        newline (only possible for the last line of a file) and it was a
768        pattern that set a locale (followed by /Lsomething), pcretest crashed.
770    19. Added additional timing features to pcretest. (1) The -tm option now times
771        matching only, not compiling. (2) Both -t and -tm can be followed, as a
772        separate command line item, by a number that specifies the number of
773        repeats to use when timing. The default is 50000; this gives better
774        precision, but takes uncomfortably long for very large patterns.
776    20. Extended pcre_study() to be more clever in cases where a branch of a
777        subpattern has no definite first character. For example, (a*|b*)[cd] would
778        previously give no result from pcre_study(). Now it recognizes that the
779        first character must be a, b, c, or d.
781    21. There was an incorrect error "recursive call could loop indefinitely" if
782        a subpattern (or the entire pattern) that was being tested for matching an
783        empty string contained only one non-empty item after a nested subpattern.
784        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
785        incorrectly, because the \d was being skipped in the check.
787    22. The pcretest program now has a new pattern option /B and a command line
788        option -b, which is equivalent to adding /B to every pattern. This causes
789        it to show the compiled bytecode, without the additional information that
790        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
791        is the same as /B/I).
793    23. A new optimization is now able automatically to treat some sequences such
794        as a*b as a*+b. More specifically, if something simple (such as a character
795        or a simple class like \d) has an unlimited quantifier, and is followed by
796        something that cannot possibly match the quantified thing, the quantifier
797        is automatically "possessified".
799    24. A recursive reference to a subpattern whose number was greater than 39
800        went wrong under certain circumstances in UTF-8 mode. This bug could also
801        have affected the operation of pcre_study().
803    25. Realized that a little bit of performance could be had by replacing
804        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
806    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
808    27. Possessive quantifiers such as a++ were previously implemented by turning
809        them into atomic groups such as ($>a+). Now they have their own opcodes,
810        which improves performance. This includes the automatically created ones
811        from 23 above.
813    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
814        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
815        the first matched character to be a colon. This applied both to named and
816        numbered groups.
818    29. The ucpinternal.h header file was missing its idempotency #ifdef.
820    30. I was sent a "project" file called libpcre.a.dev which I understand makes
821        building PCRE on Windows easier, so I have included it in the distribution.
823    31. There is now a check in pcretest against a ridiculously large number being
824        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
825        loop, the loop is abandoned.
827    32. Forward references to subpatterns in conditions such as (?(2)...) where
828        subpattern 2 is defined later cause pcre_compile() to search forwards in
829        the pattern for the relevant set of parentheses. This search went wrong
830        when there were unescaped parentheses in a character class, parentheses
831        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
833    33. "Subroutine" calls and backreferences were previously restricted to
834        referencing subpatterns earlier in the regex. This restriction has now
835        been removed.
837    34. Added a number of extra features that are going to be in Perl 5.10. On the
838        whole, these are just syntactic alternatives for features that PCRE had
839        previously implemented using the Python syntax or my own invention. The
840        other formats are all retained for compatibility.
842        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
843            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
844            also .NET compatible.
846        (b) A recursion or subroutine call to a named group can now be defined as
847            (?&name) as well as (?P>name).
849        (c) A backreference to a named group can now be defined as \k<name> or
850            \k'name' as well as (?P=name). The new forms, as well as being in Perl
851            5.10, are also .NET compatible.
853        (d) A conditional reference to a named group can now use the syntax
854            (?(<name>) or (?('name') as well as (?(name).
856        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
857            groups (named and numbered) that are never evaluated inline, but can be
858            called as "subroutines" from elsewhere. In effect, the DEFINE condition
859            is always false. There may be only one alternative in such a group.
861        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
862            as the simple (?(R). The condition is true only if the most recent
863            recursion is that of the given number or name. It does not search out
864            through the entire recursion stack.
866        (g) The escape \gN or \g{N} has been added, where N is a positive or
867            negative number, specifying an absolute or relative reference.
869    35. Tidied to get rid of some further signed/unsigned compiler warnings and
870        some "unreachable code" warnings.
872    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
873        things, this adds five new scripts.
875    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
876        There were also incompatibilities regarding the handling of \Q..\E inside
877        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
878        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
880    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
881        matches an empty string, and forcibly breaks the loop. There were bugs in
882        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
883        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
884        separate and independent bugs (that affected different cases) have been
885        fixed.
887    39. Refactored the code to abolish the use of different opcodes for small
888        capturing bracket numbers. This is a tidy that I avoided doing when I
889        removed the limit on the number of capturing brackets for 3.5 back in 2001.
890        The new approach is not only tidier, it makes it possible to reduce the
891        memory needed to fix the previous bug (38).
893    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
894        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
895        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
896        mode.
898    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
899        report.
901    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
902        copy construction and assignment in the C++ wrapper.
904    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
905        couple of bugs that could have given wrong results in the "--newline=crlf"
906        case.
908    44. Added a number of casts and did some reorganization of signed/unsigned int
909        variables following suggestions from Dair Grant. Also renamed the variable
910        "this" as "item" because it is a C++ keyword.
912    45. Arranged for dftables to add
914          #include "pcre_internal.h"
916        to pcre_chartables.c because without it, gcc 4.x may remove the array
917        definition from the final binary if PCRE is built into a static library and
918        dead code stripping is activated.
920    46. For an unanchored pattern, if a match attempt fails at the start of a
921        newline sequence, and the newline setting is CRLF or ANY, and the next two
922        characters are CRLF, advance by two characters instead of one.
925    Version 6.7 04-Jul-06
926    ---------------------
928     1. In order to handle tests when input lines are enormously long, pcretest has
929        been re-factored so that it automatically extends its buffers when
930        necessary. The code is crude, but this _is_ just a test program. The
931        default size has been increased from 32K to 50K.
933     2. The code in pcre_study() was using the value of the re argument before
934        testing it for NULL. (Of course, in any sensible call of the function, it
935        won't be NULL.)
937     3. The memmove() emulation function in pcre_internal.h, which is used on
938        systems that lack both memmove() and bcopy() - that is, hardly ever -
939        was missing a "static" storage class specifier.
941     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
942        containing an extended class (one that cannot be represented by a bitmap
943        because it contains high-valued characters or Unicode property items, e.g.
944        [\pZ]). Almost always one would set UTF-8 mode when processing such a
945        pattern, but PCRE should not loop if you do not (it no longer does).
946        [Detail: two cases were found: (a) a repeated subpattern containing an
947        extended class; (b) a recursive reference to a subpattern that followed a
948        previous extended class. It wasn't skipping over the extended class
949        correctly when UTF-8 mode was not set.]
951     5. A negated single-character class was not being recognized as fixed-length
952        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
953        compile error "lookbehind assertion is not fixed length".
955     6. The RunPerlTest auxiliary script was showing an unexpected difference
956        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
957        write a Perl script that can interpret lines of an input file either as
958        byte characters or as UTF-8, which is what "perltest" was being required to
959        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
960        can't do is switch easily at run time between having the "use utf8;" pragma
961        or not. In the end, I fudged it by using the RunPerlTest script to insert
962        "use utf8;" explicitly for the UTF-8 tests.
964     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
965        the end of the subject string, contrary to the documentation and to what
966        Perl does. This was true of both matching functions. Now it matches only at
967        the start of the subject and immediately after *internal* newlines.
969     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
970        a pointer to an int instead of a pointer to an unsigned long int. This
971        caused problems on 64-bit systems.
973     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
974        instance of the 'standard' template library not being so standard".
976    10. There was no check on the number of named subpatterns nor the maximum
977        length of a subpattern name. The product of these values is used to compute
978        the size of the memory block for a compiled pattern. By supplying a very
979        long subpattern name and a large number of named subpatterns, the size
980        computation could be caused to overflow. This is now prevented by limiting
981        the length of names to 32 characters, and the number of named subpatterns
982        to 10,000.
984    11. Subpatterns that are repeated with specific counts have to be replicated in
985        the compiled pattern. The size of memory for this was computed from the
986        length of the subpattern and the repeat count. The latter is limited to
987        65535, but there was no limit on the former, meaning that integer overflow
988        could in principle occur. The compiled length of a repeated subpattern is
989        now limited to 30,000 bytes in order to prevent this.
991    12. Added the optional facility to have named substrings with the same name.
993    13. Added the ability to use a named substring as a condition, using the
994        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
995        are numbers (not recommended). Forward references are permitted.
997    14. Added forward references in named backreferences (if you see what I mean).
999    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1000        pattern could run off the end of the subject. For example, the pattern
1001        "(?s)(.{1,5})"8 did this with the subject "ab".
1003    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1004        PCRE_CASELESS was set when matching characters that were quantified with ?
1005        or *.
1007    17. A character class other than a single negated character that had a minimum
1008        but no maximum quantifier - for example [ab]{6,} - was not handled
1009        correctly by pce_dfa_exec(). It would match only one character.
1011    18. A valid (though odd) pattern that looked like a POSIX character
1012        class but used an invalid character after [ (for example [[,abc,]]) caused
1013        pcre_compile() to give the error "Failed: internal error: code overflow" or
1014        in some cases to crash with a glibc free() error. This could even happen if
1015        the pattern terminated after [[ but there just happened to be a sequence of
1016        letters, a binary zero, and a closing ] in the memory that followed.
1018    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1019        over the years. Originally (before any Unicode support), just the bottom 8
1020        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1021        output from "man perlunicode" includes this:
1023          The regular expression compiler produces polymorphic opcodes.  That
1024          is, the pattern adapts to the data and automatically switches to
1025          the Unicode character scheme when presented with Unicode data--or
1026          instead uses a traditional byte scheme when presented with byte
1027          data.
1029        Sadly, a wide octal escape does not cause a switch, and in a string with
1030        no other multibyte characters, these octal escapes are treated as before.
1031        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
1032        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1033        Unicode string.
1035        I have not perpetrated such confusion in PCRE. Up till now, it took just
1036        the bottom 8 bits, as in old Perl. I have now made octal escapes with
1037        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1038        translate to the appropriate multibyte character.
1040    29. Applied some refactoring to reduce the number of warnings from Microsoft
1041        and Borland compilers. This has included removing the fudge introduced
1042        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1043        a warning about an unused variable.
1045    21. PCRE has not included VT (character 0x0b) in the set of whitespace
1046        characters since release 4.0, because Perl (from release 5.004) does not.
1047        [Or at least, is documented not to: some releases seem to be in conflict
1048        with the documentation.] However, when a pattern was studied with
1049        pcre_study() and all its branches started with \s, PCRE still included VT
1050        as a possible starting character. Of course, this did no harm; it just
1051        caused an unnecessary match attempt.
1053    22. Removed a now-redundant internal flag bit that recorded the fact that case
1054        dependency changed within the pattern. This was once needed for "required
1055        byte" processing, but is no longer used. This recovers a now-scarce options
1056        bit. Also moved the least significant internal flag bit to the most-
1057        significant bit of the word, which was not previously used (hangover from
1058        the days when it was an int rather than a uint) to free up another bit for
1059        the future.
1061    23. Added support for CRLF line endings as well as CR and LF. As well as the
1062        default being selectable at build time, it can now be changed at runtime
1063        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1064        specify that it is scanning data with non-default line endings.
1066    24. Changed the definition of CXXLINK to make it agree with the definition of
1067        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1069    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1070        recursions. This makes a big different to stack usage for some patterns.
1072    26. If a subpattern containing a named recursion or subroutine reference such
1073        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1074        the space required for the compiled pattern went wrong and gave too small a
1075        value. Depending on the environment, this could lead to "Failed: internal
1076        error: code overflow at offset 49" or "glibc detected double free or
1077        corruption" errors.
1079    27. Applied patches from Google (a) to support the new newline modes and (b) to
1080        advance over multibyte UTF-8 characters in GlobalReplace.
1082    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1083        difference for some implementation of PCRE in some Windows version.
1085    29. Added some extra testing facilities to pcretest:
1087        \q<number>   in a data line sets the "match limit" value
1088        \Q<number>   in a data line sets the "match recursion limt" value
1089        -S <number>  sets the stack size, where <number> is in megabytes
1091        The -S option isn't available for Windows.
1094    Version 6.6 06-Feb-06
1095    ---------------------
1097     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1098        in pcreposix.h. I have copied the definition from pcre.h.
1100     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1101        because pcre.h is no longer a built file.
1103     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1104        not normally included in the compiled code.
1107    Version 6.5 01-Feb-06
1108    ---------------------
1110     1. When using the partial match feature with pcre_dfa_exec(), it was not
1111        anchoring the second and subsequent partial matches at the new starting
1112        point. This could lead to incorrect results. For example, with the pattern
1113        /1234/, partially matching against "123" and then "a4" gave a match.
1115     2. Changes to pcregrep:
1117        (a) All non-match returns from pcre_exec() were being treated as failures
1118            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1119            error message is output. Some extra information is given for the
1120            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
1121            probably the only errors that are likely to be caused by users (by
1122            specifying a regex that has nested indefinite repeats, for instance).
1123            If there are more than 20 of these errors, pcregrep is abandoned.
1125        (b) A binary zero was treated as data while matching, but terminated the
1126            output line if it was written out. This has been fixed: binary zeroes
1127            are now no different to any other data bytes.
1129        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1130            used to set a locale for matching. The --locale=xxxx long option has
1131            been added (no short equivalent) to specify a locale explicitly on the
1132            pcregrep command, overriding the environment variables.
1134        (d) When -B was used with -n, some line numbers in the output were one less
1135            than they should have been.
1137        (e) Added the -o (--only-matching) option.
1139        (f) If -A or -C was used with -c (count only), some lines of context were
1140            accidentally printed for the final match.
1142        (g) Added the -H (--with-filename) option.
1144        (h) The combination of options -rh failed to suppress file names for files
1145            that were found from directory arguments.
1147        (i) Added the -D (--devices) and -d (--directories) options.
1149        (j) Added the -F (--fixed-strings) option.
1151        (k) Allow "-" to be used as a file name for -f as well as for a data file.
1153        (l) Added the --colo(u)r option.
1155        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1156            is not present by default.
1158     3. A nasty bug was discovered in the handling of recursive patterns, that is,
1159        items such as (?R) or (?1), when the recursion could match a number of
1160        alternatives. If it matched one of the alternatives, but subsequently,
1161        outside the recursion, there was a failure, the code tried to back up into
1162        the recursion. However, because of the way PCRE is implemented, this is not
1163        possible, and the result was an incorrect result from the match.
1165        In order to prevent this happening, the specification of recursion has
1166        been changed so that all such subpatterns are automatically treated as
1167        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1169     4. I had overlooked the fact that, in some locales, there are characters for
1170        which isalpha() is true but neither isupper() nor islower() are true. In
1171        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1172        and ordfeminine) are like this. This affected the treatment of \w and \W
1173        when they appeared in character classes, but not when they appeared outside
1174        a character class. The bit map for "word" characters is now created
1175        separately from the results of isalnum() instead of just taking it from the
1176        upper, lower, and digit maps. (Plus the underscore character, of course.)
1178     5. The above bug also affected the handling of POSIX character classes such as
1179        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1180        permanent tables. Instead, the bit maps for such a class were previously
1181        created as the appropriate unions of the upper, lower, and digit bitmaps.
1182        Now they are created by subtraction from the [[:word:]] class, which has
1183        its own bitmap.
1185     6. The [[:blank:]] character class matches horizontal, but not vertical space.
1186        It is created by subtracting the vertical space characters (\x09, \x0a,
1187        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1188        subtraction was done in the overall bitmap for a character class, meaning
1189        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1190        be recognized. This bug has been fixed.
1192     7. Patches from the folks at Google:
1194          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1195          real life, but is still worth protecting against".
1197          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1198          regular expressions".
1200          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1201          have it.
1203          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1204          "configure" and the latter not, in order to fix a problem somebody had
1205          with compiling the Arg class on HP-UX.
1207          (e) Improve the error-handling of the C++ wrapper a little bit.
1209          (f) New tests for checking recursion limiting.
1211     8. The pcre_memmove() function, which is used only if the environment does not
1212        have a standard memmove() function (and is therefore rarely compiled),
1213        contained two bugs: (a) use of int instead of size_t, and (b) it was not
1214        returning a result (though PCRE never actually uses the result).
1216     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1217        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1218        returned instead of calling malloc() with an overflowing number that would
1219        most likely cause subsequent chaos.
1221    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1223    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1224        with this option is matched, the nmatch and pmatch options of regexec() are
1225        ignored.
1227    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1228        provided in case anyone wants to the the POSIX interface with UTF-8
1229        strings.
1231    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1232        C++ linking (needed for some HP-UX environments).
1234    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1235        (unused parameter) and in the pcre_printint() function (omitted "default"
1236        switch label when the default is to do nothing).
1238    15. Added some code to make it possible, when PCRE is compiled as a C++
1239        library, to replace subject pointers for pcre_exec() with a smart pointer
1240        class, thus making it possible to process discontinuous strings.
1242    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1243        much the same function. They were added by different people who were trying
1244        to make PCRE easy to compile on non-Unix systems. It has been suggested
1245        that PCRE_EXPORT be abolished now that there is more automatic apparatus
1246        for compiling on Windows systems. I have therefore replaced it with
1247        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1248        defaults to "extern" for C or "extern C" for C++, which works fine on
1249        Unix-like systems. It is now possible to override the value of PCRE_DATA_
1250        SCOPE with something explicit in config.h. In addition:
1252        (a) pcreposix.h still had just "extern" instead of either of these macros;
1253            I have replaced it with PCRE_DATA_SCOPE.
1255        (b) Functions such as _pcre_xclass(), which are internal to the library,
1256            but external in the C sense, all had PCRE_EXPORT in their definitions.
1257            This is apparently wrong for the Windows case, so I have removed it.
1258            (It makes no difference on Unix-like systems.)
1260    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1261        of recursive calls to match(). This is different to MATCH_LIMIT because
1262        that limits the total number of calls to match(), not all of which increase
1263        the depth of recursion. Limiting the recursion depth limits the amount of
1264        stack (or heap if NO_RECURSE is set) that is used. The default can be set
1265        when PCRE is compiled, and changed at run time. A patch from Google adds
1266        this functionality to the C++ interface.
1268    18. Changes to the handling of Unicode character properties:
1270        (a) Updated the table to Unicode 4.1.0.
1272        (b) Recognize characters that are not in the table as "Cn" (undefined).
1274        (c) I revised the way the table is implemented to a much improved format
1275            which includes recognition of ranges. It now supports the ranges that
1276            are defined in UnicodeData.txt, and it also amalgamates other
1277            characters into ranges. This has reduced the number of entries in the
1278            table from around 16,000 to around 3,000, thus reducing its size
1279            considerably. I realized I did not need to use a tree structure after
1280            all - a binary chop search is just as efficient. Having reduced the
1281            number of entries, I extended their size from 6 bytes to 8 bytes to
1282            allow for more data.
1284        (d) Added support for Unicode script names via properties such as \p{Han}.
1286    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1287        matching that character.
1289    20. When matching a repeated Unicode property with a minimum greater than zero,
1290        (for example \pL{2,}), PCRE could look past the end of the subject if it
1291        reached it while seeking the minimum number of characters. This could
1292        happen only if some of the characters were more than one byte long, because
1293        there is a check for at least the minimum number of bytes.
1295    21. Refactored the implementation of \p and \P so as to be more general, to
1296        allow for more different types of property in future. This has changed the
1297        compiled form incompatibly. Anybody with saved compiled patterns that use
1298        \p or \P will have to recompile them.
1300    22. Added "Any" and "L&" to the supported property types.
1302    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1303        but give a compile time error if the value is greater than 0xff.
1305    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1306        accidentally not being installed or uninstalled.
1308    25. The pcre.h file was built from pcre.h.in, but the only changes that were
1309        made were to insert the current release number. This seemed silly, because
1310        it made things harder for people building PCRE on systems that don't run
1311        "configure". I have turned pcre.h into a distributed file, no longer built
1312        by "configure", with the version identification directly included. There is
1313        no longer a pcre.h.in file.
1315        However, this change necessitated a change to the pcre-config script as
1316        well. It is built from pcre-config.in, and one of the substitutions was the
1317        release number. I have updated configure.ac so that ./configure now finds
1318        the release number by grepping pcre.h.
1320    26. Added the ability to run the tests under valgrind.
1323    Version 6.4 05-Sep-05
1324    ---------------------
1326     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1327        "--" to be printed when multiple files were scanned, even when none of the
1328        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1329        consider it to be a bug, and have restored the previous behaviour.
1331     2. A couple of code tidies to get rid of compiler warnings.
1333     3. The pcretest program used to cheat by referring to symbols in the library
1334        whose names begin with _pcre_. These are internal symbols that are not
1335        really supposed to be visible externally, and in some environments it is
1336        possible to suppress them. The cheating is now confined to including
1337        certain files from the library's source, which is a bit cleaner.
1339     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1340        file's purpose clearer.
1342     5. Reorganized pcre_ucp_findchar().
1345    Version 6.3 15-Aug-05
1346    ---------------------
1348     1. The file libpcre.pc.in did not have general read permission in the tarball.
1350     2. There were some problems when building without C++ support:
1352        (a) If C++ support was not built, "make install" and "make test" still
1353            tried to test it.
1355        (b) There were problems when the value of CXX was explicitly set. Some
1356            changes have been made to try to fix these, and ...
1358        (c) --disable-cpp can now be used to explicitly disable C++ support.
1360        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1361            backslash in a target when C++ was disabled. This confuses some
1362            versions of "make", apparently. Using an intermediate variable solves
1363            this. (Same for CPP_LOBJ.)
1365     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1366        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1367        necessary on certain architectures.
1369     4. Added a setting of -export-symbols-regex to the link command to remove
1370        those symbols that are exported in the C sense, but actually are local
1371        within the library, and not documented. Their names all begin with
1372        "_pcre_". This is not a perfect job, because (a) we have to except some
1373        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1374        available (and never for static libraries). I have made a note to try to
1375        find a way round (a) in the future.
1378    Version 6.2 01-Aug-05
1379    ---------------------
1381     1. There was no test for integer overflow of quantifier values. A construction
1382        such as {1111111111111111} would give undefined results. What is worse, if
1383        a minimum quantifier for a parenthesized subpattern overflowed and became
1384        negative, the calculation of the memory size went wrong. This could have
1385        led to memory overwriting.
1387     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1389     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1390        operating environments where this matters.
1392     4. Applied Giuseppe Maxia's patch to add additional features for controlling
1393        PCRE options from within the C++ wrapper.
1395     5. Named capturing subpatterns were not being correctly counted when a pattern
1396        was compiled. This caused two problems: (a) If there were more than 100
1397        such subpatterns, the calculation of the memory needed for the whole
1398        compiled pattern went wrong, leading to an overflow error. (b) Numerical
1399        back references of the form \12, where the number was greater than 9, were
1400        not recognized as back references, even though there were sufficient
1401        previous subpatterns.
1403     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1404        versions of gcc, e.g. 2.95.4.
1407    Version 6.1 21-Jun-05
1408    ---------------------
1410     1. There was one reference to the variable "posix" in pcretest.c that was not
1411        surrounded by "#if !defined NOPOSIX".
1413     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1414        the cross-check on the old pcre_info() function, for the benefit of the
1415        cut-down version of PCRE that is currently imported into Exim.
1417     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1418        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1419        patterns that start (?i) but I don't think that matters. The use of (?i) is
1420        just an example; this all applies to the other options as well.
1422     4. Since libtool seems to echo the compile commands it is issuing, the output
1423        from "make" can be reduced a bit by putting "@" in front of each libtool
1424        compile command.
1426     5. Patch from the folks at Google for configure.in to be a bit more thorough
1427        in checking for a suitable C++ installation before trying to compile the
1428        C++ stuff. This should fix a reported problem when a compiler was present,
1429        but no suitable headers.
1431     6. The man pages all had just "PCRE" as their title. I have changed them to
1432        be the relevant file name. I have also arranged that these names are
1433        retained in the file doc/pcre.txt, which is a concatenation in text format
1434        of all the man pages except the little individual ones for each function.
1436     7. The NON-UNIX-USE file had not been updated for the different set of source
1437        files that come with release 6. I also added a few comments about the C++
1438        wrapper.
1441    Version 6.0 07-Jun-05
1442    ---------------------
1444     1. Some minor internal re-organization to help with my DFA experiments.
1446     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1447        didn't matter for the library itself when fully configured, but did matter
1448        when compiling without UCP support, or within Exim, where the ucp files are
1449        not imported.
1451     3. Refactoring of the library code to split up the various functions into
1452        different source modules. The addition of the new DFA matching code (see
1453        below) to a single monolithic source would have made it really too
1454        unwieldy, quite apart from causing all the code to be include in a
1455        statically linked application, when only some functions are used. This is
1456        relevant even without the DFA addition now that patterns can be compiled in
1457        one application and matched in another.
1459        The downside of splitting up is that there have to be some external
1460        functions and data tables that are used internally in different modules of
1461        the library but which are not part of the API. These have all had their
1462        names changed to start with "_pcre_" so that they are unlikely to clash
1463        with other external names.
1465     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1466        a different (DFA) algorithm. Although it is slower than the original
1467        function, it does have some advantages for certain types of matching
1468        problem.
1470     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1471        including restarting after a partial match.
1473     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1474        defined when compiling for Windows was sent to me. I have put it into the
1475        code, though I have no means of testing or verifying it.
1477     7. Added the pcre_refcount() auxiliary function.
1479     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1480        match before or at the first newline in the subject string. In pcretest,
1481        the /f option on a pattern can be used to set this.
1483     9. A repeated \w when used in UTF-8 mode with characters greater than 256
1484        would behave wrongly. This has been present in PCRE since release 4.0.
1486    10. A number of changes to the pcregrep command:
1488        (a) Refactored how -x works; insert ^(...)$ instead of setting
1489            PCRE_ANCHORED and checking the length, in preparation for adding
1490            something similar for -w.
1492        (b) Added the -w (match as a word) option.
1494        (c) Refactored the way lines are read and buffered so as to have more
1495            than one at a time available.
1497        (d) Implemented a pcregrep test script.
1499        (e) Added the -M (multiline match) option. This allows patterns to match
1500            over several lines of the subject. The buffering ensures that at least
1501            8K, or the rest of the document (whichever is the shorter) is available
1502            for matching (and similarly the previous 8K for lookbehind assertions).
1504        (f) Changed the --help output so that it now says
1506              -w, --word-regex(p)
1508            instead of two lines, one with "regex" and the other with "regexp"
1509            because that confused at least one person since the short forms are the
1510            same. (This required a bit of code, as the output is generated
1511            automatically from a table. It wasn't just a text change.)
1513        (g) -- can be used to terminate pcregrep options if the next thing isn't an
1514            option but starts with a hyphen. Could be a pattern or a path name
1515            starting with a hyphen, for instance.
1517        (h) "-" can be given as a file name to represent stdin.
1519        (i) When file names are being printed, "(standard input)" is used for
1520            the standard input, for compatibility with GNU grep. Previously
1521            "<stdin>" was used.
1523        (j) The option --label=xxx can be used to supply a name to be used for
1524            stdin when file names are being printed. There is no short form.
1526        (k) Re-factored the options decoding logic because we are going to add
1527            two more options that take data. Such options can now be given in four
1528            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1530        (l) Added the -A, -B, and -C options for requesting that lines of context
1531            around matches be printed.
1533        (m) Added the -L option to print the names of files that do not contain
1534            any matching lines, that is, the complement of -l.
1536        (n) The return code is 2 if any file cannot be opened, but pcregrep does
1537            continue to scan other files.
1539        (o) The -s option was incorrectly implemented. For compatibility with other
1540            greps, it now suppresses the error message for a non-existent or non-
1541            accessible file (but not the return code). There is a new option called
1542            -q that suppresses the output of matching lines, which was what -s was
1543            previously doing.
1545        (p) Added --include and --exclude options to specify files for inclusion
1546            and exclusion when recursing.
1548    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1549        Hopefully, it now does.
1551    12. Missing cast in pcre_study().
1553    13. Added an "uninstall" target to the makefile.
1555    14. Replaced "extern" in the function prototypes in Makefile.in with
1556        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1557        world, but is set differently for Windows.
1559    15. Added a second compiling function called pcre_compile2(). The only
1560        difference is that it has an extra argument, which is a pointer to an
1561        integer error code. When there is a compile-time failure, this is set
1562        non-zero, in addition to the error test pointer being set to point to an
1563        error message. The new argument may be NULL if no error number is required
1564        (but then you may as well call pcre_compile(), which is now just a
1565        wrapper). This facility is provided because some applications need a
1566        numeric error indication, but it has also enabled me to tidy up the way
1567        compile-time errors are handled in the POSIX wrapper.
1569    16. Added VPATH=.libs to the makefile; this should help when building with one
1570        prefix path and installing with another. (Or so I'm told by someone who
1571        knows more about this stuff than I do.)
1573    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1574        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1575        match everything, including newlines. This is not POSIX-compatible, but
1576        somebody wanted the feature. From pcretest it can be activated by using
1577        both the P and the s flags.
1579    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1581    19. libpcre.pc was being incorrectly installed as executable.
1583    20. A couple of places in pcretest check for end-of-line by looking for '\n';
1584        it now also looks for '\r' so that it will work unmodified on Windows.
1586    21. Added Google's contributed C++ wrapper to the distribution.
1588    22. Added some untidy missing memory free() calls in pcretest, to keep
1589        Electric Fence happy when testing.
1593    Version 5.0 13-Sep-04
1594    ---------------------
1596     1. Internal change: literal characters are no longer packed up into items
1597        containing multiple characters in a single byte-string. Each character
1598        is now matched using a separate opcode. However, there may be more than one
1599        byte in the character in UTF-8 mode.
1601     2. The pcre_callout_block structure has two new fields: pattern_position and
1602        next_item_length. These contain the offset in the pattern to the next match
1603        item, and its length, respectively.
1605     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1606        insertion of callouts before each pattern item. Added the /C option to
1607        pcretest to make use of this.
1609     4. On the advice of a Windows user, the lines
1611          #if defined(_WIN32) || defined(WIN32)
1612          _setmode( _fileno( stdout ), 0x8000 );
1613          #endif  /* defined(_WIN32) || defined(WIN32) */
1615        have been added to the source of pcretest. This apparently does useful
1616        magic in relation to line terminators.
1618     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1619        for the benefit of those environments where the "b" makes a difference.
1621     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1622        to know about it. I have put a hack into configure.in that adds in code
1623        to set GCC=yes if CC=icc. This seems to end up at a point in the
1624        generated configure script that is early enough to affect the setting of
1625        compiler options, which is what is needed, but I have no means of testing
1626        whether it really works. (The user who reported this had patched the
1627        generated configure script, which of course I cannot do.)
1629        LATER: After change 22 below (new libtool files), the configure script
1630        seems to know about icc (and also ecc). Therefore, I have commented out
1631        this hack in configure.in.
1633     7. Added support for pkg-config (2 patches were sent in).
1635     8. Negated POSIX character classes that used a combination of internal tables
1636        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1637        [[:^ascii]]. Typically, they would match almost any characters. The other
1638        POSIX classes were not broken in this way.
1640     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1641        to find the match, as PCRE was deluded into thinking that the match had to
1642        start at the start point or following a newline. The same bug applied to
1643        patterns with negative forward assertions or any backward assertions
1644        preceding ".*" at the start, unless the pattern required a fixed first
1645        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1647    10. In UTF-8 mode, when moving forwards in the subject after a failed match
1648        starting at the last subject character, bytes beyond the end of the subject
1649        string were read.
1651    11. Renamed the variable "class" as "classbits" to make life easier for C++
1652        users. (Previously there was a macro definition, but it apparently wasn't
1653        enough.)
1655    12. Added the new field "tables" to the extra data so that tables can be passed
1656        in at exec time, or the internal tables can be re-selected. This allows
1657        a compiled regex to be saved and re-used at a later time by a different
1658        program that might have everything at different addresses.
1660    13. Modified the pcre-config script so that, when run on Solaris, it shows a
1661        -R library as well as a -L library.
1663    14. The debugging options of pcretest (-d on the command line or D on a
1664        pattern) showed incorrect output for anything following an extended class
1665        that contained multibyte characters and which was followed by a quantifier.
1667    15. Added optional support for general category Unicode character properties
1668        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1669        support. It adds about 90K to the size of the library. The meanings of the
1670        inbuilt class escapes such as \d and \s have NOT been changed.
1672    16. Updated pcredemo.c to include calls to free() to release the memory for the
1673        compiled pattern.
1675    17. The generated file chartables.c was being created in the source directory
1676        instead of in the building directory. This caused the build to fail if the
1677        source directory was different from the building directory, and was
1678        read-only.
1680    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1681        file. No doubt somebody will tell me if they don't make sense... Also added
1682        Dan Mooney's comments about building on OpenVMS.
1684    19. Added support for partial matching via the PCRE_PARTIAL option for
1685        pcre_exec() and the \P data escape in pcretest.
1687    20. Extended pcretest with 3 new pattern features:
1689        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
1690              write the compiled pattern to the file whose name is "rest-of-line".
1691              This is a straight binary dump of the data, with the saved pointer to
1692              the character tables forced to be NULL. The study data, if any, is
1693              written too. After writing, pcretest reads a new pattern.
1695        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1696              compiled pattern from the given file. There must not be any
1697              occurrences of "<" in the file name (pretty unlikely); if there are,
1698              pcretest will instead treat the initial "<" as a pattern delimiter.
1699              After reading in the pattern, pcretest goes on to read data lines as
1700              usual.
1702        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1703              and 16-bit fields in a compiled pattern, to simulate a pattern that
1704              was compiled on a host of opposite endianness.
1706    21. The pcre-exec() function can now cope with patterns that were compiled on
1707        hosts of opposite endianness, with this restriction:
1709          As for any compiled expression that is saved and used later, the tables
1710          pointer field cannot be preserved; the extra_data field in the arguments
1711          to pcre_exec() should be used to pass in a tables address if a value
1712          other than the default internal tables were used at compile time.
1714    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1715        now diagnosed as an error. Previously, most of the time, a negative number
1716        would have been treated as zero, but if in addition "ovector" was passed as
1717        NULL, a crash could occur.
1719    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1720        new versions from the libtool 1.5 distribution (the last one is a copy of
1721        a file called libtool.m4). This seems to have fixed the need to patch
1722        "configure" to support Darwin 1.3 (which I used to do). However, I still
1723        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1724        workstation).
1726    24. Changed the PCRE licence to be the more standard "BSD" licence.
1729    Version 4.5 01-Dec-03
1730    ---------------------
1732     1. There has been some re-arrangement of the code for the match() function so
1733        that it can be compiled in a version that does not call itself recursively.
1734        Instead, it keeps those local variables that need separate instances for
1735        each "recursion" in a frame on the heap, and gets/frees frames whenever it
1736        needs to "recurse". Keeping track of where control must go is done by means
1737        of setjmp/longjmp. The whole thing is implemented by a set of macros that
1738        hide most of the details from the main code, and operates only if
1739        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1740        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1741        operating.
1743        To make it easier for callers to provide specially tailored get/free
1744        functions for this usage, two new functions, pcre_stack_malloc, and
1745        pcre_stack_free, are used. They are always called in strict stacking order,
1746        and the size of block requested is always the same.
1748        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1749        PCRE has been compiled to use the stack or the heap for recursion. The
1750        -C option of pcretest uses this to show which version is compiled.
1752        A new data escape \S, is added to pcretest; it causes the amounts of store
1753        obtained and freed by both kinds of malloc/free at match time to be added
1754        to the output.
1756     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1757        what's available on my current Linux desktop machine.
1759     3. When matching a UTF-8 string, the test for a valid string at the start has
1760        been extended. If start_offset is not zero, PCRE now checks that it points
1761        to a byte that is the start of a UTF-8 character. If not, it returns
1762        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1763        this is necessary because there may be backward assertions in the pattern.
1764        When matching the same subject several times, it may save resources to use
1765        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1767     4. The code for checking the validity of UTF-8 strings has been tightened so
1768        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1769        containing "overlong sequences".
1771     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1772        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1773        should have been "&", but it just so happened that all the cases this let
1774        through by mistake were picked up later in the function.
1776     6. I had used a variable called "isblank" - this is a C99 function, causing
1777        some compilers to warn. To avoid this, I renamed it (as "blankclass").
1779     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1780        prompting; (b) run "./pcretest /dev/null" at the start of the test script
1781        so the version is shown; (c) stop "make test" echoing "./RunTest".
1783     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1785     9. The prototype for memmove() for systems that don't have it was using
1786        size_t, but the inclusion of the header that defines size_t was later. I've
1787        moved the #includes for the C headers earlier to avoid this.
1789    10. Added some adjustments to the code to make it easier to compiler on certain
1790        special systems:
1792          (a) Some "const" qualifiers were missing.
1793          (b) Added the macro EXPORT before all exported functions; by default this
1794              is defined to be empty.
1795          (c) Changed the dftables auxiliary program (that builds chartables.c) so
1796              that it reads its output file name as an argument instead of writing
1797              to the standard output and assuming this can be redirected.
1799    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
1800        class containing characters with values greater than 255, PCRE compilation
1801        went into a loop.
1803    12. A recursive reference to a subpattern that was within another subpattern
1804        that had a minimum quantifier of zero caused PCRE to crash. For example,
1805        (x(y(?2))z)? provoked this bug with a subject that got as far as the
1806        recursion. If the recursively-called subpattern itself had a zero repeat,
1807        that was OK.
1809    13. In pcretest, the buffer for reading a data line was set at 30K, but the
1810        buffer into which it was copied (for escape processing) was still set at
1811        1024, so long lines caused crashes.
1813    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
1814        "internal error: code overflow...". This applied to any character class
1815        that was followed by a possessive quantifier.
1817    15. Modified the Makefile to add libpcre.la as a prerequisite for
1818        libpcreposix.la because I was told this is needed for a parallel build to
1819        work.
1821    16. If a pattern that contained .* following optional items at the start was
1822        studied, the wrong optimizing data was generated, leading to matching
1823        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
1824        matching string must start with a or b or c. The correct conclusion for
1825        this pattern is that a match can start with any character.
1828    Version 4.4 13-Aug-03
1829    ---------------------
1831     1. In UTF-8 mode, a character class containing characters with values between
1832        127 and 255 was not handled correctly if the compiled pattern was studied.
1833        In fixing this, I have also improved the studying algorithm for such
1834        classes (slightly).
1836     2. Three internal functions had redundant arguments passed to them. Removal
1837        might give a very teeny performance improvement.
1839     3. Documentation bug: the value of the capture_top field in a callout is *one
1840        more than* the number of the hightest numbered captured substring.
1842     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
1843        in incorrectly linking with a previously installed version. They now link
1844        explicitly with libpcre.la.
1846     5. configure.in no longer needs to recognize Cygwin specially.
1848     6. A problem in pcre.in for Windows platforms is fixed.
1850     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
1851        pcretest, it used to include the size of the study block as part of its
1852        output. Unfortunately, the structure contains a field that has a different
1853        size on different hardware architectures. This meant that the tests that
1854        showed this size failed. As the block is currently always of a fixed size,
1855        this information isn't actually particularly useful in pcretest output, so
1856        I have just removed it.
1858     8. Three pre-processor statements accidentally did not start in column 1.
1859        Sadly, there are *still* compilers around that complain, even though
1860        standard C has not required this for well over a decade. Sigh.
1862     9. In pcretest, the code for checking callouts passed small integers in the
1863        callout_data field, which is a void * field. However, some picky compilers
1864        complained about the casts involved for this on 64-bit systems. Now
1865        pcretest passes the address of the small integer instead, which should get
1866        rid of the warnings.
1868    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
1869        both compile and run time, and gives an error if an invalid UTF-8 sequence
1870        is found. There is a option for disabling this check in cases where the
1871        string is known to be correct and/or the maximum performance is wanted.
1873    11. In response to a bug report, I changed one line in Makefile.in from
1875            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
1876        to
1877            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
1879        to look similar to other lines, but I have no way of telling whether this
1880        is the right thing to do, as I do not use Windows. No doubt I'll get told
1881        if it's wrong...
1884    Version 4.3 21-May-03
1885    ---------------------
1887    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
1888       Makefile.
1890    2. Some refactoring to improve the quality of the code:
1892       (i)   The utf8_table... variables are now declared "const".
1894       (ii)  The code for \cx, which used the "case flipping" table to upper case
1895             lower case letters, now just substracts 32. This is ASCII-specific,
1896             but the whole concept of \cx is ASCII-specific, so it seems
1897             reasonable.
1899       (iii) PCRE was using its character types table to recognize decimal and
1900             hexadecimal digits in the pattern. This is silly, because it handles
1901             only 0-9, a-f, and A-F, but the character types table is locale-
1902             specific, which means strange things might happen. A private
1903             table is now used for this - though it costs 256 bytes, a table is
1904             much faster than multiple explicit tests. Of course, the standard
1905             character types table is still used for matching digits in subject
1906             strings against \d.
1908       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
1909             ending in _t are). So I've renamed it as ESC_tee.
1911    3. The first argument for regexec() in the POSIX wrapper should have been
1912       defined as "const".
1914    4. Changed pcretest to use malloc() for its buffers so that they can be
1915       Electric Fenced for debugging.
1917    5. There were several places in the code where, in UTF-8 mode, PCRE would try
1918       to read one or more bytes before the start of the subject string. Often this
1919       had no effect on PCRE's behaviour, but in some circumstances it could
1920       provoke a segmentation fault.
1922    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
1923       to try to read one or more bytes before the start of the subject string.
1925    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
1926       UTF-8 support could misbehave in various ways if the subject string
1927       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
1928       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
1929       back over UTF-8 characters.)
1932    Version 4.2 14-Apr-03
1933    ---------------------
1935    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
1937    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
1938         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
1939         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
1940         [WIN_PREFIX]: new variable, "cyg" for Cygwin
1941         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
1942           and BUILD_EXEEXT
1943         Note: automatic setting of the BUILD variables is not yet working
1944         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
1945           compile-time but not at link-time
1946         [LINK]: use for linking executables only
1947         make different versions for Windows and non-Windows
1948         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
1949           libraries
1950         [LINK_FOR_BUILD]: new variable
1951         [OBJEXT]: use throughout
1952         [EXEEXT]: use throughout
1953         <winshared>: new target
1954         <wininstall>: new target
1955         <dftables.o>: use native compiler
1956         <dftables>: use native linker
1957         <install>: handle Windows platform correctly
1958         <clean>: ditto
1959         <check>: ditto
1960         copy DLL to top builddir before testing
1962       As part of these changes, -no-undefined was removed again. This was reported
1963       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
1964       in any case.
1966    3. Some tidies to get rid of compiler warnings:
1968       . In the match_data structure, match_limit was an unsigned long int, whereas
1969         match_call_count was an int. I've made them both unsigned long ints.
1971       . In pcretest the fact that a const uschar * doesn't automatically cast to
1972         a void * provoked a warning.
1974       . Turning on some more compiler warnings threw up some "shadow" variables
1975         and a few more missing casts.
1977    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
1978       option, a class that contained a single character with a value between 128
1979       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
1981    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
1982       option, a class that contained several characters, but with at least one
1983       whose value was between 128 and 255 caused PCRE to crash.
1986    Version 4.1 12-Mar-03
1987    ---------------------
1989    1. Compiling with gcc -pedantic found a couple of places where casts were
1990    needed, and a string in dftables.c that was longer than standard compilers are
1991    required to support.
1993    2. Compiling with Sun's compiler found a few more places where the code could
1994    be tidied up in order to avoid warnings.
1996    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
1997    first of these names is deprecated in the latest Autoconf in favour of the name
1998    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
1999    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
2000    analogy I have changed it to CFLAGS_FOR_BUILD.
2002    4. Added -no-undefined to the linking command in the Makefile, because this is
2003    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
2004    linking step for the pcreposix library.
2006    5. PCRE was failing to diagnose the case of two named groups with the same
2007    name.
2009    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
2010    literal character that is needed in the subject for a match, and scans along to
2011    ensure that it is present before embarking on the full matching process. This
2012    saves time in cases of nested unlimited repeats that are never going to match.
2013    Problem: the scan can take a lot of time if the subject is very long (e.g.
2014    megabytes), thus penalizing straightforward matches. It is now done only if the
2015    amount of subject to be scanned is less than 1000 bytes.
2017    7. A lesser problem with the same optimization is that it was recording the
2018    first character of an anchored pattern as "needed", thus provoking a search
2019    right along the subject, even when the first match of the pattern was going to
2020    fail. The "needed" character is now not set for anchored patterns, unless it
2021    follows something in the pattern that is of non-fixed length. Thus, it still
2022    fulfils its original purpose of finding quick non-matches in cases of nested
2023    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
2026    Version 4.0 17-Feb-03
2027    ---------------------
2029    1. If a comment in an extended regex that started immediately after a meta-item
2030    extended to the end of string, PCRE compiled incorrect data. This could lead to
2031    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
2033    2. Moved to autoconf 2.53 and libtool 1.4.2.
2035    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
2036    the special perltest8 script is no longer needed - all the tests can be run
2037    from a single perltest script.
2039    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
2040    by \s. It has now been removed in PCRE. This means it isn't recognized as
2041    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
2042    class [:space:] *does* include VT, thereby creating a mess.
2044    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
2045    space and tab.
2047    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
2048    its new features into the main test script, reducing the number of scripts.
2050    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
2051    were backward compatible, and made the (?i) apply to the whole pattern, as if
2052    /i were given. Now it behaves more logically, and applies the option setting
2053    only to what follows. PCRE has been changed to follow suit. However, if it
2054    finds options settings right at the start of the pattern, it extracts them into
2055    the global options, as before. Thus, they show up in the info data.
2057    8. Added support for the \Q...\E escape sequence. Characters in between are
2058    treated as literals. This is slightly different from Perl in that $ and @ are
2059    also handled as literals inside the quotes. In Perl, they will cause variable
2060    interpolation. Note the following examples:
2062        Pattern            PCRE matches      Perl matches
2064        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
2065        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
2066        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
2068    For compatibility with Perl, \Q...\E sequences are recognized inside character
2069    classes as well as outside them.
2071    9. Re-organized 3 code statements in pcretest to avoid "overflow in
2072    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
2073    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
2074    signed/unsigned warnings.
2076    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
2077    option for pcretest, so I've replaced it by a simple function that does just
2078    that job.
2080    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
2081    "pcregrep -".
2083    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
2084    Java package. This provides some syntactic sugar for simple cases of what my
2085    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
2086    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
2087    item, you can use this simplified notation. Note that only makes sense with
2088    greedy quantifiers. Consequently, the use of the possessive quantifier forces
2089    greediness, whatever the setting of the PCRE_UNGREEDY option.
2091    13. A change of greediness default within a pattern was not taking effect at
2092    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
2093    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
2094    was abstracted outside.
2096    14. PCRE now supports the \G assertion. It is true when the current matching
2097    position is at the start point of the match. This differs from \A when the
2098    starting offset is non-zero. Used with the /g option of pcretest (or similar
2099    code), it works in the same way as it does for Perl's /g option. If all
2100    alternatives of a regex begin with \G, the expression is anchored to the start
2101    match position, and the "anchored" flag is set in the compiled expression.
2103    15. Some bugs concerning the handling of certain option changes within patterns
2104    have been fixed. These applied to options other than (?ims). For example,
2105    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
2106    been the other way round. Some of this was related to change 7 above.
2108    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
2109    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
2110    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
2111    POSIX classes only within a class (e.g. /[[:alpha:]]/).
2113    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
2114    mode. Unlike ".", it always matches newline, whatever the setting of
2115    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
2116    assertions. Perl allows it, but it doesn't (in general) work because it can't
2117    calculate the length of the lookbehind. At least, that's the case for Perl
2118    5.8.0 - I've been told they are going to document that it doesn't work in
2119    future.
2121    18. Added an error diagnosis for escapes that PCRE does not support: these are
2122    \L, \l, \N, \P, \p, \U, \u, and \X.
2124    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
2125    reading past the end of the pattern in cases such as /[abcd/.
2127    20. PCRE was getting more memory than necessary for patterns with classes that
2128    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
2130    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
2131    compiling PCRE for use with Virtual Pascal.
2133    22. Small fix to the Makefile to make it work properly if the build is done
2134    outside the source tree.
2136    23. Added a new extension: a condition to go with recursion. If a conditional
2137    subpattern starts with (?(R) the "true" branch is used if recursion has
2138    happened, whereas the "false" branch is used only at the top level.
2140    24. When there was a very long string of literal characters (over 255 bytes
2141    without UTF support, over 250 bytes with UTF support), the computation of how
2142    much memory was required could be incorrect, leading to segfaults or other
2143    strange effects.
2145    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
2146    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
2147    there was a subsequent back reference to those brackets. This meant that, for
2148    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
2149    possible to check for precisely this case. All we can do is abandon the
2150    optimization if .* occurs inside capturing brackets when there are any back
2151    references whatsoever. (See below for a better fix that came later.)
2153    26. The handling of the optimization for finding the first character of a
2154    non-anchored pattern, and for finding a character that is required later in the
2155    match were failing in some cases. This didn't break the matching; it just
2156    failed to optimize when it could. The way this is done has been re-implemented.
2158    27. Fixed typo in error message for invalid (?R item (it said "(?p").
2160    28. Added a new feature that provides some of the functionality that Perl
2161    provides with (?{...}). The facility is termed a "callout". The way it is done
2162    in PCRE is for the caller to provide an optional function, by setting
2163    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
2164    global variable. By default it is unset, which disables all calling out. To get
2165    the function called, the regex must include (?C) at appropriate points. This
2166    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
2167    This provides a means of identifying different callout points. When PCRE
2168    reaches such a point in the regex, if pcre_callout has been set, the external
2169    function is called. It is provided with data in a structure called
2170    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
2171    matching continues; if it returns a non-zero value, the match at the current
2172    point fails. However, backtracking will occur if possible. [This was changed
2173    later and other features added - see item 49 below.]
2175    29. pcretest is upgraded to test the callout functionality. It provides a
2176    callout function that displays information. By default, it shows the start of
2177    the match and the current position in the text. There are some new data escapes
2178    to vary what happens:
2180        \C+         in addition, show current contents of captured substrings
2181        \C-         do not supply a callout function
2182        \C!n        return 1 when callout number n is reached
2183        \C!n!m      return 1 when callout number n is reached for the mth time
2185    30. If pcregrep was called with the -l option and just a single file name, it
2186    output "<stdin>" if a match was found, instead of the file name.
2188    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
2189    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
2190    pcre_exec(). This saves a malloc/free per call. The default value of
2191    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
2192    when configuring.
2194    32. The default maximum size of a compiled pattern is 64K. There have been a
2195    few cases of people hitting this limit. The code now uses macros to handle the
2196    storing of links as offsets within the compiled pattern. It defaults to 2-byte
2197    links, but this can be changed to 3 or 4 bytes by --with-link-size when
2198    configuring. Tests 2 and 5 work only with 2-byte links because they output
2199    debugging information about compiled patterns.
2201    33. Internal code re-arrangements:
2203    (a) Moved the debugging function for printing out a compiled regex into
2204        its own source file (printint.c) and used #include to pull it into
2205        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
2206        separate copies.
2208    (b) Defined the list of op-code names for debugging as a macro in
2209        internal.h so that it is next to the definition of the opcodes.
2211    (c) Defined a table of op-code lengths for simpler skipping along compiled
2212        code. This is again a macro in internal.h so that it is next to the
2213        definition of the opcodes.
2215    34. Added support for recursive calls to individual subpatterns, along the
2216    lines of Robin Houston's patch (but implemented somewhat differently).
2218    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
2219    allow it to read and process whole directories in Win32. This code was
2220    contributed by Lionel Fourquaux; it has not been tested by me.
2222    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
2223    used to name a group. Names consist of alphanumerics and underscores, and must
2224    be unique. Back references use the syntax (?P=name) and recursive calls use
2225    (?P>name) which is a PCRE extension to the Python extension. Groups still have
2226    numbers. The function pcre_fullinfo() can be used after compilation to extract
2227    a name/number map. There are three relevant calls:
2229      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
2230      PCRE_INFO_NAMECOUNT            yields the number of entries
2231      PCRE_INFO_NAMETABLE            yields a pointer to the map.
2233    The map is a vector of fixed-size entries. The size of each entry depends on
2234    the length of the longest name used. The first two bytes of each entry are the
2235    group number, most significant byte first. There follows the corresponding
2236    name, zero terminated. The names are in alphabetical order.
2238    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
2239    case instead of 255. Making it the same both with and without UTF-8 support
2240    means that the same test output works with both.
2242    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
2243    calling malloc() with a zero argument.
2245    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
2246    optimization. I've improved things by keeping a bitmap of backreferences with
2247    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
2248    fact referenced, the optimization can be applied. It is unlikely that a
2249    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
2250    the match to follow \n) will appear inside brackets with a number greater than
2251    31, but if it does, any back reference > 31 suppresses the optimization.
2253    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
2254    of disabling numbered capturing parentheses. Any opening parenthesis that is
2255    not followed by ? behaves as if it were followed by ?: but named parentheses
2256    can still be used for capturing (and they will acquire numbers in the usual
2257    way).
2259    41. Redesigned the return codes from the match() function into yes/no/error so
2260    that errors can be passed back from deep inside the nested calls. A malloc
2261    failure while inside a recursive subpattern call now causes the
2262    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
2264    42. It is now possible to set a limit on the number of times the match()
2265    function is called in a call to pcre_exec(). This facility makes it possible to
2266    limit the amount of recursion and backtracking, though not in a directly
2267    obvious way, because the match() function is used in a number of different
2268    circumstances. The count starts from zero for each position in the subject
2269    string (for non-anchored patterns). The default limit is, for compatibility, a
2270    large number, namely 10 000 000. You can change this in two ways:
2272    (a) When configuring PCRE before making, you can use --with-match-limit=n
2273        to set a default value for the compiled library.
2275    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
2276        a different value is set. See 45 below.
2278    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
2280    43. Added a new function pcre_config(int, void *) to enable run-time extraction
2281    of things that can be changed at compile time. The first argument specifies
2282    what is wanted and the second points to where the information is to be placed.
2283    The current list of available information is:
2285      PCRE_CONFIG_UTF8
2287    The output is an integer that is set to one if UTF-8 support is available;
2288    otherwise it is set to zero.
2292    The output is an integer that it set to the value of the code that is used for
2293    newline. It is either LF (10) or CR (13).
2297    The output is an integer that contains the number of bytes used for internal
2298    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
2302    The output is an integer that contains the threshold above which the POSIX
2303    interface uses malloc() for output vectors. See item 31 above.
2307    The output is an unsigned integer that contains the default limit of the number
2308    of match() calls in a pcre_exec() execution. See 42 above.
2310    44. pcretest has been upgraded by the addition of the -C option. This causes it
2311    to extract all the available output from the new pcre_config() function, and to
2312    output it. The program then exits immediately.
2314    45. A need has arisen to pass over additional data with calls to pcre_exec() in
2315    order to support additional features. One way would have been to define
2316    pcre_exec2() (for example) with extra arguments, but this would not have been
2317    extensible, and would also have required all calls to the original function to
2318    be mapped to the new one. Instead, I have chosen to extend the mechanism that
2319    is used for passing in "extra" data from pcre_study().
2321    The pcre_extra structure is now exposed and defined in pcre.h. It currently
2322    contains the following fields:
2324      flags         a bitmap indicating which of the following fields are set
2325      study_data    opaque data from pcre_study()
2326      match_limit   a way of specifying a limit on match() calls for a specific
2327                      call to pcre_exec()
2328      callout_data  data for callouts (see 49 below)
2330    The flag bits are also defined in pcre.h, and are
2336    The pcre_study() function now returns one of these new pcre_extra blocks, with
2337    the actual study data pointed to by the study_data field, and the
2338    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
2339    before. That is, this change is entirely upwards-compatible and requires no
2340    change to existing code.
2342    If you want to pass in additional data to pcre_exec(), you can either place it
2343    in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
2344    block.
2346    46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
2347    data string contains the escape sequence \M, pcretest calls pcre_exec() several
2348    times with different match limits, until it finds the minimum value needed for
2349    pcre_exec() to complete. The value is then output. This can be instructive; for
2350    most simple matches the number is quite small, but for pathological cases it
2351    gets very large very quickly.
2353    47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
2354    returns the size of the data block pointed to by the study_data field in a
2355    pcre_extra block, that is, the value that was passed as the argument to
2356    pcre_malloc() when PCRE was getting memory in which to place the information
2357    created by pcre_study(). The fourth argument should point to a size_t variable.
2358    pcretest has been extended so that this information is shown after a successful
2359    pcre_study() call when information about the compiled regex is being displayed.
2361    48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
2362    because what follows is always an absolute path. (Later: it turns out that this
2363    is more than cosmetic for MinGW, because it doesn't like empty path
2364    components.)
2366    49. Some changes have been made to the callout feature (see 28 above):
2368    (i)  A callout function now has three choices for what it returns:
2370           0  =>  success, carry on matching
2371         > 0  =>  failure at this point, but backtrack if possible
2372         < 0  =>  serious error, return this value from pcre_exec()
2374         Negative values should normally be chosen from the set of PCRE_ERROR_xxx
2375         values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
2376         "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
2377         use by callout functions. It will never be used by PCRE itself.
2379    (ii) The pcre_extra structure (see 45 above) has a void * field called
2380         callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
2381         pcre_callout_block structure has a field of the same name. The contents of
2382         the field passed in the pcre_extra structure are passed to the callout
2383         function in the corresponding field in the callout block. This makes it
2384         easier to use the same callout-containing regex from multiple threads. For
2385         testing, the pcretest program has a new data escape
2387           \C*n        pass the number n (may be negative) as callout_data
2389         If the callout function in pcretest receives a non-zero value as
2390         callout_data, it returns that value.
2392    50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
2393    there were some redundant $(CFLAGS) in commands that are now specified as
2394    $(LINK), which already includes $(CFLAGS).
2396    51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
2397    has been compiled with UTF-8 support *and* pcre_compile() has been compiled
2398    with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
2399    one-byte characters throughout. Note that case-insensitive matching applies
2400    only to characters whose values are less than 256. PCRE doesn't support the
2401    notion of cases for higher-valued characters.
2403    (i)   A character class whose characters are all within 0-255 is handled as
2404          a bit map, and the map is inverted for negative classes. Previously, a
2405          character > 255 always failed to match such a class; however it should
2406          match if the class was a negative one (e.g. [^ab]). This has been fixed.
2408    (ii)  A negated character class with a single character < 255 is coded as
2409          "not this character" (OP_NOT). This wasn't working properly when the test
2410          character was multibyte, either singly or repeated.
2412    (iii) Repeats of multibyte characters are now handled correctly in UTF-8
2413          mode, for example: \x{100}{2,3}.
2415    (iv)  The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
2416          singly or repeated) now correctly test multibyte characters. However,
2417          PCRE doesn't recognize any characters with values greater than 255 as
2418          digits, spaces, or word characters. Such characters always match \D, \S,
2419          and \W, and never match \d, \s, or \w.
2421    (v)   Classes may now contain characters and character ranges with values
2422          greater than 255. For example: [ab\x{100}-\x{400}].
2424    (vi)  pcregrep now has a --utf-8 option (synonym -u) which makes it call
2425          PCRE in UTF-8 mode.
2427    52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
2428    PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
2429    retained for backwards compatibility. (Note that LASTLITERAL is also a byte
2430    value.)
2432    53. The single man page has become too large. I have therefore split it up into
2433    a number of separate man pages. These also give rise to individual HTML pages;
2434    these are now put in a separate directory, and there is an index.html page that
2435    lists them all. Some hyperlinking between the pages has been installed.
2437    54. Added convenience functions for handling named capturing parentheses.
2439    55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2440    aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2441    true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2442    are faulted.
2444    56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2445    calling configure. These values are used when compiling the dftables.c program
2446    which is run to generate the source of the default character tables. They
2447    default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2448    you will need to set these values.
2450    57. Updated the building process for Windows DLL, as provided by Fred Cox.
2453    Version 3.9 02-Jan-02
2454    ---------------------
2456    1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2458    2. If --disable-static was given, the building process failed when trying to
2459    build pcretest and pcregrep. (For some reason it was using libtool to compile
2460    them, which is not right, as they aren't part of the library.)
2463    Version 3.8 18-Dec-01
2464    ---------------------
2466    1. The experimental UTF-8 code was completely screwed up. It was packing the
2467    bytes in the wrong order. How dumb can you get?
2470    Version 3.7 29-Oct-01
2471    ---------------------
2473    1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2474    This caused pcretest, when used on the test data, to segfault. Unfortunately,
2475    this didn't happen under Solaris 8, where I normally test things.
2477    2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2478    doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2479    isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2480    this fix an hour or so after the initial 3.7 release.)
2483    Version 3.6 23-Oct-01
2484    ---------------------
2486    1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2487    offsets passed as NULL with zero offset count.
2489    2. The config.guess and config.sub files had not been updated when I moved to
2490    the latest autoconf.
2493    Version 3.5 15-Aug-01
2494    ---------------------
2496    1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2497    had been forgotten.
2499    2. By using declared but undefined structures, we can avoid using "void"
2500    definitions in pcre.h while keeping the internal definitions of the structures
2501    private.
2503    3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2504    user point of view, this means that both static and shared libraries are built
2505    by default, but this can be individually controlled. More of the work of
2506    handling this static/shared cases is now inside libtool instead of PCRE's make
2507    file.
2509    4. The pcretest utility is now installed along with pcregrep because it is
2510    useful for users (to test regexs) and by doing this, it automatically gets
2511    relinked by libtool. The documentation has been turned into a man page, so
2512    there are now .1, .txt, and .html versions in /doc.
2514    5. Upgrades to pcregrep:
2515       (i)   Added long-form option names like gnu grep.
2516       (ii)  Added --help to list all options with an explanatory phrase.
2517       (iii) Added -r, --recursive to recurse into sub-directories.
2518       (iv)  Added -f, --file to read patterns from a file.
2520    6. pcre_exec() was referring to its "code" argument before testing that
2521    argument for NULL (and giving an error if it was NULL).
2523    7. Upgraded Makefile.in to allow for compiling in a different directory from
2524    the source directory.
2526    8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2527    options bits, the pointer it was passed was to an int instead of to an unsigned
2528    long int. This mattered only on 64-bit systems.
2530    9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2531    generated) instead of pcre.in, which it its source. Also made the same change
2532    in several of the .c files.
2534    10. A new release of gcc defines printf() as a macro, which broke pcretest
2535    because it had an ifdef in the middle of a string argument for printf(). Fixed
2536    by using separate calls to printf().
2538    11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2539    script, to force use of CR or LF instead of \n in the source. On non-Unix
2540    systems, the value can be set in config.h.
2542    12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2543    absolute limit. Changed the text of the error message to make this clear, and
2544    likewise updated the man page.
2546    13. The limit of 99 on the number of capturing subpatterns has been removed.
2547    The new limit is 65535, which I hope will not be a "real" limit.
2550    Version 3.4 22-Aug-00
2551    ---------------------
2553    1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2555    2. Diagnose condition (?(0) as an error instead of crashing on matching.
2558    Version 3.3 01-Aug-00
2559    ---------------------
2561    1. If an octal character was given, but the value was greater than \377, it
2562    was not getting masked to the least significant bits, as documented. This could
2563    lead to crashes in some systems.
2565    2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
2566    the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
2568    3. Added the functions pcre_free_substring() and pcre_free_substring_list().
2569    These just pass their arguments on to (pcre_free)(), but they are provided
2570    because some uses of PCRE bind it to non-C systems that can call its functions,
2571    but cannot call free() or pcre_free() directly.
2573    4. Add "make test" as a synonym for "make check". Corrected some comments in
2574    the Makefile.
2576    5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
2577    Makefile.
2579    6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
2580    command called pgrep for grepping around the active processes.
2582    7. Added the beginnings of support for UTF-8 character strings.
2584    8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
2585    RANLIB to ./ltconfig so that they are used by libtool. I think these are all
2586    the relevant ones. (AR is not passed because ./ltconfig does its own figuring
2587    out for the ar command.)
2590    Version 3.2 12-May-00
2591    ---------------------
2593    This is purely a bug fixing release.
2595    1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
2596    of ZA. This was just one example of several cases that could provoke this bug,
2597    which was introduced by change 9 of version 2.00. The code for breaking
2598    infinite loops after an iteration that matches an empty string was't working
2599    correctly.
2601    2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
2602    when matched against abbab (for example). After matching an empty string, it
2603    wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
2604    caused it to match further down the string than it should.
2606    3. The code contained an inclusion of sys/types.h. It isn't clear why this
2607    was there because it doesn't seem to be needed, and it causes trouble on some
2608    systems, as it is not a Standard C header. It has been removed.
2610    4. Made 4 silly changes to the source to avoid stupid compiler warnings that
2611    were reported on the Macintosh. The changes were from
2613      while ((c = *(++ptr)) != 0 && c != '\n');
2614    to
2615      while ((c = *(++ptr)) != 0 && c != '\n') ;
2617    Totally extraordinary, but if that's what it takes...
2619    5. PCRE is being used in one environment where neither memmove() nor bcopy() is
2620    available. Added HAVE_BCOPY and an autoconf test for it; if neither
2621    HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
2622    assumes the way PCRE uses memmove() (always moving upwards).
2624    6. PCRE is being used in one environment where strchr() is not available. There
2625    was only one use in pcre.c, and writing it out to avoid strchr() probably gives
2626    faster code anyway.
2629    Version 3.1 09-Feb-00
2630    ---------------------
2632    The only change in this release is the fixing of some bugs in Makefile.in for
2633    the "install" target:
2635    (1) It was failing to install pcreposix.h.
2637    (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
2640    Version 3.0 01-Feb-00
2641    ---------------------
2643    1. Add support for the /+ modifier to perltest (to output $` like it does in
2644    pcretest).
2646    2. Add support for the /g modifier to perltest.
2648    3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
2649    matches null strings.
2651    4. Fix perltest so that it doesn't do unwanted things when fed an empty
2652    pattern. Perl treats empty patterns specially - it reuses the most recent
2653    pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
2654    effect.
2656    5. The POSIX interface was broken in that it was just handing over the POSIX
2657    captured string vector to pcre_exec(), but (since release 2.00) PCRE has
2658    required a bigger vector, with some working space on the end. This means that
2659    the POSIX wrapper now has to get and free some memory, and copy the results.
2661    6. Added some simple autoconf support, placing the test data and the
2662    documentation in separate directories, re-organizing some of the
2663    information files, and making it build pcre-config (a GNU standard). Also added
2664    libtool support for building PCRE as a shared library, which is now the
2665    default.
2667    7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
2668    09 are not valid octal constants. Single digits will be used for minor values
2669    less than 10.
2671    8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
2672    existing programs that set these in the POSIX interface can use PCRE without
2673    modification.
2675    9. Added a new function, pcre_fullinfo() with an extensible interface. It can
2676    return all that pcre_info() returns, plus additional data. The pcre_info()
2677    function is retained for compatibility, but is considered to be obsolete.
2679    10. Added experimental recursion feature (?R) to handle one common case that
2680    Perl 5.6 will be able to do with (?p{...}).
2682    11. Added support for POSIX character classes like [:alpha:], which Perl is
2683    adopting.
2686    Version 2.08 31-Aug-99
2687    ----------------------
2689    1. When startoffset was not zero and the pattern began with ".*", PCRE was not
2690    trying to match at the startoffset position, but instead was moving forward to
2691    the next newline as if a previous match had failed.
2693    2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
2694    and could get into a loop if a null string was matched other than at the start
2695    of the subject.
2697    3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
2698    be distinguished at compile time, and for completeness also added PCRE_DATE.
2700    5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
2701    in GnuWin32 environments.
2704    Version 2.07 29-Jul-99
2705    ----------------------
2707    1. The documentation is now supplied in plain text form and HTML as well as in
2708    the form of man page sources.
2710    2. C++ compilers don't like assigning (void *) values to other pointer types.
2711    In particular this affects malloc(). Although there is no problem in Standard
2712    C, I've put in casts to keep C++ compilers happy.
2714    3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
2715    should be (const char *).
2717    4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
2718    be useful for non-Unix systems who don't want to bother with the POSIX stuff.
2719    However, I haven't made this a standard facility. The documentation doesn't
2720    mention it, and the Makefile doesn't support it.
2722    5. The Makefile now contains an "install" target, with editable destinations at
2723    the top of the file. The pcretest program is not installed.
2725    6. pgrep -V now gives the PCRE version number and date.
2727    7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
2728    causing the entire string to be ignored, instead of just the last character.
2730    8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
2731    non-matching string, it can take a very, very long time, even for strings of
2732    quite modest length, because of the nested recursion. PCRE now does better in
2733    some of these cases. It does this by remembering the last required literal
2734    character in the pattern, and pre-searching the subject to ensure it is present
2735    before running the real match. In other words, it applies a heuristic to detect
2736    some types of certain failure quickly, and in the above example, if presented
2737    with a string that has no trailing " it gives "no match" very quickly.
2739    9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
2740    other alternatives are tried instead.
2743    Version 2.06 09-Jun-99
2744    ----------------------
2746    1. Change pcretest's output for amount of store used to show just the code
2747    space, because the remainder (the data block) varies in size between 32-bit and
2748    64-bit systems.
2750    2. Added an extra argument to pcre_exec() to supply an offset in the subject to
2751    start matching at. This allows lookbehinds to work when searching for multiple
2752    occurrences in a string.
2754    3. Added additional options to pcretest for testing multiple occurrences:
2756       /+   outputs the rest of the string that follows a match
2757       /g   loops for multiple occurrences, using the new startoffset argument
2758       /G   loops for multiple occurrences by passing an incremented pointer
2760    4. PCRE wasn't doing the "first character" optimization for patterns starting
2761    with \b or \B, though it was doing it for other lookbehind assertions. That is,
2762    it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
2763    the letter 'x'. On long subject strings, this gives a significant speed-up.
2766    Version 2.05 21-Apr-99
2767    ----------------------
2769    1. Changed the type of magic_number from int to long int so that it works
2770    properly on 16-bit systems.
2772    2. Fixed a bug which caused patterns starting with .* not to work correctly
2773    when the subject string contained newline characters. PCRE was assuming
2774    anchoring for such patterns in all cases, which is not correct because .* will
2775    not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
2776    DOTALL is set at top level; otherwise it knows that patterns starting with .*
2777    must be retried after every newline in the subject.
2780    Version 2.04 18-Feb-99
2781    ----------------------
2783    1. For parenthesized subpatterns with repeats whose minimum was zero, the
2784    computation of the store needed to hold the pattern was incorrect (too large).
2785    If such patterns were nested a few deep, this could multiply and become a real
2786    problem.
2788    2. Added /M option to pcretest to show the memory requirement of a specific
2789    pattern. Made -m a synonym of -s (which does this globally) for compatibility.
2791    3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
2792    compiled in such a way that the backtracking after subsequent failure was
2793    pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
2794    ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
2797    Version 2.03 02-Feb-99
2798    ----------------------
2800    1. Fixed typo and small mistake in man page.
2802    2. Added 4th condition (GPL supersedes if conflict) and created separate
2803    LICENCE file containing the conditions.
2805    3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
2806    Perl, that is the internal \ allows the delimiter to be included in the
2807    pattern. Locked out the use of \ as a delimiter. If \ immediately follows
2808    the final delimiter, add \ to the end of the pattern (to test the error).
2810    4. Added the convenience functions for extracting substrings after a successful
2811    match. Updated pcretest to make it able to test these functions.
2814  Version 2.02 14-Jan-99  Version 2.02 14-Jan-99
2815  ----------------------  ----------------------

Removed from v.27  
changed lines
  Added in v.319

  ViewVC Help
Powered by ViewVC 1.1.5