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 286 by ph10, Mon Dec 17 14:46:11 2007 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
4    Version 7.5 12-Nov-07
5    ---------------------
7    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
8        values in parens when parsing an RE using the C++ wrapper."
10    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
11        Characters greater than 255 were excluded from the class instead of being
12        included.
14    3.  The same bug as (2) above applied to negated POSIX classes such as
15        [:^space:].
17    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
18        defined or documented. It seems to have been a typo for PCRE_STATIC, so
19        I have changed it.
21    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
22        first named subpattern) and a construct such as (?&a) would reference the
23        first named subpattern whose name started with "a" (in other words, the
24        length check was missing). Both these problems are fixed. "Subpattern name
25        expected" is now given for (?&) (a zero-length name), and this patch also
26        makes it give the same error for \k'' (previously it complained that that
27        was a reference to a non-existent subpattern).
29    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
30        this is right because (?- can be followed by option settings as well as by
31        digits. I have, however, made the messages clearer.
33    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
34        than the number used in the conditional) now cause a compile-time error.
35        This is actually not compatible with Perl, which accepts such patterns, but
36        treats the conditional as always being FALSE (as PCRE used to), but it
37        seems to me that giving a diagnostic is better.
39    8.  Change "alphameric" to the more common word "alphanumeric" in comments
40        and messages.
42    9.  Fix two occurrences of "backslash" in comments that should have been
43        "backspace".
45    10. Remove two redundant lines of code that can never be obeyed (their function
46        was moved elsewhere).
48    11. The program that makes PCRE's Unicode character property table had a bug
49        which caused it to generate incorrect table entries for sequences of
50        characters that have the same character type, but are in different scripts.
51        It amalgamated them into a single range, with the script of the first of
52        them. In other words, some characters were in the wrong script. There were
53        thirteen such cases, affecting characters in the following ranges:
55          U+002b0 - U+002c1
56          U+0060c - U+0060d
57          U+0061e - U+00612
58          U+0064b - U+0065e
59          U+0074d - U+0076d
60          U+01800 - U+01805
61          U+01d00 - U+01d77
62          U+01d9b - U+01dbf
63          U+0200b - U+0200f
64          U+030fc - U+030fe
65          U+03260 - U+0327f
66          U+0fb46 - U+0fbb1
67          U+10450 - U+1049d
69    12. The -o option (show only the matching part of a line) for pcregrep was not
70        compatible with GNU grep in that, if there was more than one match in a
71        line, it showed only the first of them. It now behaves in the same way as
72        GNU grep.
74    13. If the -o and -v options were combined for pcregrep, it printed a blank
75        line for every non-matching line. GNU grep prints nothing, and pcregrep now
76        does the same. The return code can be used to tell if there were any
77        non-matching lines.
79    14. The pattern (?=something)(?R) was not being diagnosed as a potentially
80        infinitely looping recursion. The bug was that positive lookaheads were not
81        being skipped when checking for a possible empty match (negative lookaheads
82        and both kinds of lookbehind were skipped).
84    15. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
85        inclusion of <windows.h> to before rather than after the definition of
86        INVALID_FILE_ATTRIBUTES (patch from David Byron).
88    16. Specifying a possessive quantifier with a specific limit for a Unicode
89        character property caused pcre_compile() to compile bad code, which led at
90        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
91        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
92        caused the error; without that there was no problem.
94    17. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
97    Version 7.4 21-Sep-07
98    ---------------------
100    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
101        means that a class such as [\s] counted as "explicit reference to CR or
102        LF". That isn't really right - the whole point of the change was to try to
103        help when there was an actual mention of one of the two characters. So now
104        the change happens only if \r or \n (or a literal CR or LF) character is
105        encountered.
107    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
108        of both had grown to the point where there were only 3 bits left.
109        Fortunately, there was spare space in the data structure, and so I have
110        moved the internal flags into a new 16-bit field to free up more option
111        bits.
113    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
114        but did not set the internal JCHANGED flag - either of these is enough to
115        control the way the "get" function works - but the PCRE_INFO_JCHANGED
116        facility is supposed to tell if (?J) was ever used, so now (?J) at the
117        start sets both bits.
119    4.  Added options (at build time, compile time, exec time) to change \R from
120        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
122    5.  doc/pcresyntax.html was missing from the distribution.
124    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
125        compatibility, even though it is no longer used.
127    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
128        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
129        windows.h file is present (where different names are used). [This was
130        reversed later after testing - see 16 below.]
132    8.  Changed all #include <config.h> to #include "config.h". There were also
133        some further <pcre.h> cases that I changed to "pcre.h".
135    9.  When pcregrep was used with the --colour option, it missed the line ending
136        sequence off the lines that it output.
138    10. It was pointed out to me that arrays of string pointers cause lots of
139        relocations when a shared library is dynamically loaded. A technique of
140        using a single long string with a table of offsets can drastically reduce
141        these. I have refactored PCRE in four places to do this. The result is
142        dramatic:
144          Originally:                          290
145          After changing UCP table:            187
146          After changing error message table:   43
147          After changing table of "verbs"       36
148          After changing table of Posix names   22
150        Thanks to the folks working on Gregex for glib for this insight.
152    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
153        unicode-properties was also set.
155    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
157    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
158        checked only for CRLF.
160    14. Added casts to pcretest.c to avoid compiler warnings.
162    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
164    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
165        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
166        entirely. This removes changes made in 7 above.
168    17. The CMake files have been updated, and there is now more information about
169        building with CMake in the NON-UNIX-USE document.
172    Version 7.3 28-Aug-07
173    ---------------------
175     1. In the rejigging of the build system that eventually resulted in 7.1, the
176        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
177        brackets there is not right, since it causes compilers to look for an
178        installed pcre.h, not the version that is in the source that is being
179        compiled (which of course may be different). I have changed it back to:
181          #include "pcre.h"
183        I have a vague recollection that the change was concerned with compiling in
184        different directories, but in the new build system, that is taken care of
185        by the VPATH setting the Makefile.
187     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
188        when the subject happened to end in the byte 0x85 (e.g. if the last
189        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
190        characters but of course it shouldn't be taken as a newline when it is part
191        of another character. The bug was that, for an unlimited repeat of . in
192        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
193        characters when looking for a newline.
195     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
197     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
198        in debug output.
200     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
201        long printing in the pcrecpp unittest when running under MinGW.
203     6. ESC_K was left out of the EBCDIC table.
205     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
206        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
207        limit also applies to "virtual nesting" when a pattern is recursive, and in
208        this case 1000 isn't so big. I have been able to remove this limit at the
209        expense of backing off one optimization in certain circumstances. Normally,
210        when pcre_exec() would call its internal match() function recursively and
211        immediately return the result unconditionally, it uses a "tail recursion"
212        feature to save stack. However, when a subpattern that can match an empty
213        string has an unlimited repetition quantifier, it no longer makes this
214        optimization. That gives it a stack frame in which to save the data for
215        checking that an empty string has been matched. Previously this was taken
216        from the 1000-entry workspace that had been reserved. So now there is no
217        explicit limit, but more stack is used.
219     8. Applied Daniel's patches to solve problems with the import/export magic
220        syntax that is required for Windows, and which was going wrong for the
221        pcreposix and pcrecpp parts of the library. These were overlooked when this
222        problem was solved for the main library.
224     9. There were some crude static tests to avoid integer overflow when computing
225        the size of patterns that contain repeated groups with explicit upper
226        limits. As the maximum quantifier is 65535, the maximum group length was
227        set at 30,000 so that the product of these two numbers did not overflow a
228        32-bit integer. However, it turns out that people want to use groups that
229        are longer than 30,000 bytes (though not repeat them that many times).
230        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
231        made it possible to implement the integer overflow checks in a much more
232        dynamic way, which I have now done. The artificial limitation on group
233        length has been removed - we now have only the limit on the total length of
234        the compiled pattern, which depends on the LINK_SIZE setting.
236    10. Fixed a bug in the documentation for get/copy named substring when
237        duplicate names are permitted. If none of the named substrings are set, the
238        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
239        empty string.
241    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
242        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
243        because the ] is interpreted as the first data character and the
244        terminating ] is not found. PCRE has been made compatible with Perl in this
245        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
246        cause memory overwriting.
248    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
249        string has been matched (to stop an infinite loop). It was not recognizing
250        a conditional subpattern that could match an empty string if that
251        subpattern was within another subpattern. For example, it looped when
252        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
253        condition was not nested. This bug has been fixed.
255    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
256        past the start of the subject in the presence of bytes with the top bit
257        set, for example "\x8aBCD".
259    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
260        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
262    14. Optimized (?!) to (*FAIL).
264    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
265        This restricts code points to be within the range 0 to 0x10FFFF, excluding
266        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
267        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
268        does: it's just the validity check that is more restrictive.
270    16. Inserted checks for integer overflows during escape sequence (backslash)
271        processing, and also fixed erroneous offset values for syntax errors during
272        backslash processing.
274    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
275        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
277    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
278        caused an overrun.
280    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
281        something other than just ASCII characters) inside a group that had an
282        unlimited repeat caused a loop at compile time (while checking to see
283        whether the group could match an empty string).
285    20. Debugging a pattern containing \p or \P could cause a crash. For example,
286        [\P{Any}] did so. (Error in the code for printing property names.)
288    21. An orphan \E inside a character class could cause a crash.
290    22. A repeated capturing bracket such as (A)? could cause a wild memory
291        reference during compilation.
293    23. There are several functions in pcre_compile() that scan along a compiled
294        expression for various reasons (e.g. to see if it's fixed length for look
295        behind). There were bugs in these functions when a repeated \p or \P was
296        present in the pattern. These operators have additional parameters compared
297        with \d, etc, and these were not being taken into account when moving along
298        the compiled data. Specifically:
300        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
301            length.
303        (b) An item such as \pL+ within a repeated group could cause crashes or
304            loops.
306        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
307            "reference to non-existent subpattern" error.
309        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
311    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
312        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
314    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
316    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
317        character were causing crashes (broken optimization).
319    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
320        \p or \P) caused a compile-time loop.
322    28. More problems have arisen in unanchored patterns when CRLF is a valid line
323        break. For example, the unstudied pattern [\r\n]A does not match the string
324        "\r\nA" because change 7.0/46 below moves the current point on by two
325        characters after failing to match at the start. However, the pattern \nA
326        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
327        the same is true. There doesn't seem any very clean way out of this, but
328        what I have chosen to do makes the common cases work: PCRE now takes note
329        of whether there can be an explicit match for \r or \n anywhere in the
330        pattern, and if so, 7.0/46 no longer applies. As part of this change,
331        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
332        pattern has explicit CR or LF references.
334    29. Added (*CR) etc for changing newline setting at start of pattern.
337    Version 7.2 19-Jun-07
338    ---------------------
340     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
341        which is apparently normally available under Windows.
343     2. Re-jig the pcregrep tests with different newline settings in an attempt
344        to make them independent of the local environment's newline setting.
346     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
348     4. Some of the "internals" tests were previously cut out when the link size
349        was not 2, because the output contained actual offsets. The recent new
350        "Z" feature of pcretest means that these can be cut out, making the tests
351        usable with all link sizes.
353     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
354        stack recursion. This gives a massive performance boost under BSD, but just
355        a small improvement under Linux. However, it saves one field in the frame
356        in all cases.
358     6. Added more features from the forthcoming Perl 5.10:
360        (a) (?-n) (where n is a string of digits) is a relative subroutine or
361            recursion call. It refers to the nth most recently opened parentheses.
363        (b) (?+n) is also a relative subroutine call; it refers to the nth next
364            to be opened parentheses.
366        (c) Conditions that refer to capturing parentheses can be specified
367            relatively, for example, (?(-2)... or (?(+3)...
369        (d) \K resets the start of the current match so that everything before
370            is not part of it.
372        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
374        (f) \g{name} is another synonym - part of Perl 5.10's unification of
375            reference syntax.
377        (g) (?| introduces a group in which the numbering of parentheses in each
378            alternative starts with the same number.
380        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
382     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
385     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
386        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
387        for detecting groups that can match an empty string.
389     9. A pattern with a very large number of alternatives (more than several
390        hundred) was running out of internal workspace during the pre-compile
391        phase, where pcre_compile() figures out how much memory will be needed. A
392        bit of new cunning has reduced the workspace needed for groups with
393        alternatives. The 1000-alternative test pattern now uses 12 bytes of
394        workspace instead of running out of the 4096 that are available.
396    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
398    11. Applied patch from Google to remove an optimization that didn't quite work.
399        The report of the bug said:
401          pcrecpp::RE("a*").FullMatch("aaa") matches, while
402          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
403          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
405    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
406        it matched the wrong number of bytes.
409    Version 7.1 24-Apr-07
410    ---------------------
412     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
413        that is more "standard", making use of automake and other Autotools. There
414        is some re-arrangement of the files and adjustment of comments consequent
415        on this.
417     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
418        for recursive directory scanning broke on some systems because the files
419        are not scanned in any specific order and on different systems the order
420        was different. A call to "sort" has been inserted into RunGrepTest for the
421        approprate test as a short-term fix. In the longer term there may be an
422        alternative.
424     3. I had an email from Eric Raymond about problems translating some of PCRE's
425        man pages to HTML (despite the fact that I distribute HTML pages, some
426        people do their own conversions for various reasons). The problems
427        concerned the use of low-level troff macros .br and .in. I have therefore
428        removed all such uses from the man pages (some were redundant, some could
429        be replaced by .nf/.fi pairs). The 132html script that I use to generate
430        HTML has been updated to handle .nf/.fi and to complain if it encounters
431        .br or .in.
433     4. Updated comments in configure.ac that get placed in config.h.in and also
434        arranged for config.h to be included in the distribution, with the name
435        config.h.generic, for the benefit of those who have to compile without
436        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
438     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
439        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
440        makevp.bat for the current PCRE, using the additional files
441        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
443     6. A Windows user reported a minor discrepancy with test 2, which turned out
444        to be caused by a trailing space on an input line that had got lost in his
445        copy. The trailing space was an accident, so I've just removed it.
447     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
448        that is needed.
450     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
451        as "const" (a) because they are and (b) because it helps the PHP
452        maintainers who have recently made a script to detect big data structures
453        in the php code that should be moved to the .rodata section. I remembered
454        to update Builducptable as well, so it won't revert if ucptable.h is ever
455        re-created.
457     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
458        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
459        order to be able to cut out the UTF-8 tables in the latter when UTF-8
460        support is not required. This saves 1.5-2K of code, which is important in
461        some applications.
463        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
464        so as not to refer to the tables, even though these functions will never be
465        called when UTF-8 support is disabled. Otherwise there are problems with a
466        shared library.
468    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
470        (a) It was defining its arguments as char * instead of void *.
472        (b) It was assuming that all moves were upwards in memory; this was true
473            a long time ago when I wrote it, but is no longer the case.
475        The emulated memove() is provided for those environments that have neither
476        memmove() nor bcopy(). I didn't think anyone used it these days, but that
477        is clearly not the case, as these two bugs were recently reported.
479    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
480        and Detrail to create the HTML documentation, the .txt form of the man
481        pages, and it removes trailing spaces from listed files. It also creates
482        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
483        case, it wraps all the #defines with #ifndefs. This script should be run
484        before "make dist".
486    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
487        with Unicode property support.
489        (a) For a maximizing quantifier, if the two different cases of the
490            character were of different lengths in their UTF-8 codings (there are
491            some cases like this - I found 11), and the matching function had to
492            back up over a mixture of the two cases, it incorrectly assumed they
493            were both the same length.
495        (b) When PCRE was configured to use the heap rather than the stack for
496            recursion during matching, it was not correctly preserving the data for
497            the other case of a UTF-8 character when checking ahead for a match
498            while processing a minimizing repeat. If the check also involved
499            matching a wide character, but failed, corruption could cause an
500            erroneous result when trying to check for a repeat of the original
501            character.
503    13. Some tidying changes to the testing mechanism:
505        (a) The RunTest script now detects the internal link size and whether there
506            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
507            values substituted by "configure". (The RunGrepTest script already did
508            this for UTF-8.) The configure.ac script no longer substitutes the
509            relevant variables.
511        (b) The debugging options /B and /D in pcretest show the compiled bytecode
512            with length and offset values. This means that the output is different
513            for different internal link sizes. Test 2 is skipped for link sizes
514            other than 2 because of this, bypassing the problem. Unfortunately,
515            there was also a test in test 3 (the locale tests) that used /B and
516            failed for link sizes other than 2. Rather than cut the whole test out,
517            I have added a new /Z option to pcretest that replaces the length and
518            offset values with spaces. This is now used to make test 3 independent
519            of link size. (Test 2 will be tidied up later.)
521    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
522        segmentation fault instead of returning the appropriate error message.
524    15. In multiline mode when the newline sequence was set to "any", the pattern
525        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
526        This doesn't seem right; it now treats the CRLF combination as the line
527        ending, and so does not match in that case. It's only a pattern such as ^$
528        that would hit this one: something like ^ABC$ would have failed after \r
529        and then tried again after \r\n.
531    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
532        in an attempt to make files that differ only in their line terminators
533        compare equal. This works on Linux.
535    17. Under certain error circumstances pcregrep might try to free random memory
536        as it exited. This is now fixed, thanks to valgrind.
538    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
539        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
540        was because its rules for how to advance for /g after matching an empty
541        string at the end of a line did not allow for this case. They now check for
542        it specially.
544    20. pcretest is supposed to handle patterns and data of any length, by
545        extending its buffers when necessary. It was getting this wrong when the
546        buffer for a data line had to be extended.
548    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
549        CRLF as a newline sequence.
551    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
552        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
553        I have nevertheless tidied it up.
555    23. Added some casts to kill warnings from HP-UX ia64 compiler.
557    24. Added a man page for pcre-config.
560    Version 7.0 19-Dec-06
561    ---------------------
563     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
564        moving to gcc 4.1.1.
566     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
567        sys/time.h, which is documented as needed for this function. It doesn't
568        seem to matter on Linux, but it showed up on some releases of OS X.
570     3. It seems that there are systems where bytes whose values are greater than
571        127 match isprint() in the "C" locale. The "C" locale should be the
572        default when a C program starts up. In most systems, only ASCII printing
573        characters match isprint(). This difference caused the output from pcretest
574        to vary, making some of the tests fail. I have changed pcretest so that:
576        (a) When it is outputting text in the compiled version of a pattern, bytes
577            other than 32-126 are always shown as hex escapes.
579        (b) When it is outputting text that is a matched part of a subject string,
580            it does the same, unless a different locale has been set for the match
581            (using the /L modifier). In this case, it uses isprint() to decide.
583     4. Fixed a major bug that caused incorrect computation of the amount of memory
584        required for a compiled pattern when options that changed within the
585        pattern affected the logic of the preliminary scan that determines the
586        length. The relevant options are -x, and -i in UTF-8 mode. The result was
587        that the computed length was too small. The symptoms of this bug were
588        either the PCRE error "internal error: code overflow" from pcre_compile(),
589        or a glibc crash with a message such as "pcretest: free(): invalid next
590        size (fast)". Examples of patterns that provoked this bug (shown in
591        pcretest format) are:
593          /(?-x: )/x
594          /(?x)(?-x: \s*#\s*)/
595          /((?i)[\x{c0}])/8
596          /(?i:[\x{c0}])/8
598        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
599        is now done differently.
601     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
602        wrapper classes; (b) implement a new function in the C++ scanner that is
603        more efficient than the old way of doing things because it avoids levels of
604        recursion in the regex matching; (c) add a paragraph to the documentation
605        for the FullMatch() function.
607     6. The escape sequence \n was being treated as whatever was defined as
608        "newline". Not only was this contrary to the documentation, which states
609        that \n is character 10 (hex 0A), but it also went horribly wrong when
610        "newline" was defined as CRLF. This has been fixed.
612     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
613        was being set to -1 for the "end of line" case (supposedly a value that no
614        character can have). Though this value is never used (the check for end of
615        line is "zero bytes in current character"), it caused compiler complaints.
616        I've changed it to 0xffffffff.
618     8. In pcre_version.c, the version string was being built by a sequence of
619        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
620        string (as it is for production releases) called a macro with an empty
621        argument. The C standard says the result of this is undefined. The gcc
622        compiler treats it as an empty string (which was what was wanted) but it is
623        reported that Visual C gives an error. The source has been hacked around to
624        avoid this problem.
626     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
627        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
628        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
629        of them did).
631    10. Originally, pcretest opened its input and output without "b"; then I was
632        told that "b" was needed in some environments, so it was added for release
633        5.0 to both the input and output. (It makes no difference on Unix-like
634        systems.) Later I was told that it is wrong for the input on Windows. I've
635        now abstracted the modes into two macros, to make it easier to fiddle with
636        them, and removed "b" from the input mode under Windows.
638    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
640    12. Added -help and --help to pcretest as an official way of being reminded
641        of the options.
643    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
644        and pcrecpp.cc because they annoy compilers at high warning levels.
646    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
648    15. Fixed an occurrence of == in configure.ac that should have been = (shell
649        scripts are not C programs :-) and which was not noticed because it works
650        on Linux.
652    16. pcretest is supposed to handle any length of pattern and data line (as one
653        line or as a continued sequence of lines) by extending its input buffer if
654        necessary. This feature was broken for very long pattern lines, leading to
655        a string of junk being passed to pcre_compile() if the pattern was longer
656        than about 50K.
658    17. I have done a major re-factoring of the way pcre_compile() computes the
659        amount of memory needed for a compiled pattern. Previously, there was code
660        that made a preliminary scan of the pattern in order to do this. That was
661        OK when PCRE was new, but as the facilities have expanded, it has become
662        harder and harder to keep it in step with the real compile phase, and there
663        have been a number of bugs (see for example, 4 above). I have now found a
664        cunning way of running the real compile function in a "fake" mode that
665        enables it to compute how much memory it would need, while actually only
666        ever using a few hundred bytes of working memory and without too many
667        tests of the mode. This should make future maintenance and development
668        easier. A side effect of this work is that the limit of 200 on the nesting
669        depth of parentheses has been removed (though this was never a serious
670        limitation, I suspect). However, there is a downside: pcre_compile() now
671        runs more slowly than before (30% or more, depending on the pattern). I
672        hope this isn't a big issue. There is no effect on runtime performance.
674    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
675        newline (only possible for the last line of a file) and it was a
676        pattern that set a locale (followed by /Lsomething), pcretest crashed.
678    19. Added additional timing features to pcretest. (1) The -tm option now times
679        matching only, not compiling. (2) Both -t and -tm can be followed, as a
680        separate command line item, by a number that specifies the number of
681        repeats to use when timing. The default is 50000; this gives better
682        precision, but takes uncomfortably long for very large patterns.
684    20. Extended pcre_study() to be more clever in cases where a branch of a
685        subpattern has no definite first character. For example, (a*|b*)[cd] would
686        previously give no result from pcre_study(). Now it recognizes that the
687        first character must be a, b, c, or d.
689    21. There was an incorrect error "recursive call could loop indefinitely" if
690        a subpattern (or the entire pattern) that was being tested for matching an
691        empty string contained only one non-empty item after a nested subpattern.
692        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
693        incorrectly, because the \d was being skipped in the check.
695    22. The pcretest program now has a new pattern option /B and a command line
696        option -b, which is equivalent to adding /B to every pattern. This causes
697        it to show the compiled bytecode, without the additional information that
698        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
699        is the same as /B/I).
701    23. A new optimization is now able automatically to treat some sequences such
702        as a*b as a*+b. More specifically, if something simple (such as a character
703        or a simple class like \d) has an unlimited quantifier, and is followed by
704        something that cannot possibly match the quantified thing, the quantifier
705        is automatically "possessified".
707    24. A recursive reference to a subpattern whose number was greater than 39
708        went wrong under certain circumstances in UTF-8 mode. This bug could also
709        have affected the operation of pcre_study().
711    25. Realized that a little bit of performance could be had by replacing
712        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
714    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
716    27. Possessive quantifiers such as a++ were previously implemented by turning
717        them into atomic groups such as ($>a+). Now they have their own opcodes,
718        which improves performance. This includes the automatically created ones
719        from 23 above.
721    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
722        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
723        the first matched character to be a colon. This applied both to named and
724        numbered groups.
726    29. The ucpinternal.h header file was missing its idempotency #ifdef.
728    30. I was sent a "project" file called libpcre.a.dev which I understand makes
729        building PCRE on Windows easier, so I have included it in the distribution.
731    31. There is now a check in pcretest against a ridiculously large number being
732        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
733        loop, the loop is abandoned.
735    32. Forward references to subpatterns in conditions such as (?(2)...) where
736        subpattern 2 is defined later cause pcre_compile() to search forwards in
737        the pattern for the relevant set of parentheses. This search went wrong
738        when there were unescaped parentheses in a character class, parentheses
739        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
741    33. "Subroutine" calls and backreferences were previously restricted to
742        referencing subpatterns earlier in the regex. This restriction has now
743        been removed.
745    34. Added a number of extra features that are going to be in Perl 5.10. On the
746        whole, these are just syntactic alternatives for features that PCRE had
747        previously implemented using the Python syntax or my own invention. The
748        other formats are all retained for compatibility.
750        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
751            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
752            also .NET compatible.
754        (b) A recursion or subroutine call to a named group can now be defined as
755            (?&name) as well as (?P>name).
757        (c) A backreference to a named group can now be defined as \k<name> or
758            \k'name' as well as (?P=name). The new forms, as well as being in Perl
759            5.10, are also .NET compatible.
761        (d) A conditional reference to a named group can now use the syntax
762            (?(<name>) or (?('name') as well as (?(name).
764        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
765            groups (named and numbered) that are never evaluated inline, but can be
766            called as "subroutines" from elsewhere. In effect, the DEFINE condition
767            is always false. There may be only one alternative in such a group.
769        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
770            as the simple (?(R). The condition is true only if the most recent
771            recursion is that of the given number or name. It does not search out
772            through the entire recursion stack.
774        (g) The escape \gN or \g{N} has been added, where N is a positive or
775            negative number, specifying an absolute or relative reference.
777    35. Tidied to get rid of some further signed/unsigned compiler warnings and
778        some "unreachable code" warnings.
780    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
781        things, this adds five new scripts.
783    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
784        There were also incompatibilities regarding the handling of \Q..\E inside
785        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
786        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
788    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
789        matches an empty string, and forcibly breaks the loop. There were bugs in
790        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
791        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
792        separate and independent bugs (that affected different cases) have been
793        fixed.
795    39. Refactored the code to abolish the use of different opcodes for small
796        capturing bracket numbers. This is a tidy that I avoided doing when I
797        removed the limit on the number of capturing brackets for 3.5 back in 2001.
798        The new approach is not only tidier, it makes it possible to reduce the
799        memory needed to fix the previous bug (38).
801    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
802        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
803        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
804        mode.
806    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
807        report.
809    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
810        copy construction and assignment in the C++ wrapper.
812    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
813        couple of bugs that could have given wrong results in the "--newline=crlf"
814        case.
816    44. Added a number of casts and did some reorganization of signed/unsigned int
817        variables following suggestions from Dair Grant. Also renamed the variable
818        "this" as "item" because it is a C++ keyword.
820    45. Arranged for dftables to add
822          #include "pcre_internal.h"
824        to pcre_chartables.c because without it, gcc 4.x may remove the array
825        definition from the final binary if PCRE is built into a static library and
826        dead code stripping is activated.
828    46. For an unanchored pattern, if a match attempt fails at the start of a
829        newline sequence, and the newline setting is CRLF or ANY, and the next two
830        characters are CRLF, advance by two characters instead of one.
833    Version 6.7 04-Jul-06
834    ---------------------
836     1. In order to handle tests when input lines are enormously long, pcretest has
837        been re-factored so that it automatically extends its buffers when
838        necessary. The code is crude, but this _is_ just a test program. The
839        default size has been increased from 32K to 50K.
841     2. The code in pcre_study() was using the value of the re argument before
842        testing it for NULL. (Of course, in any sensible call of the function, it
843        won't be NULL.)
845     3. The memmove() emulation function in pcre_internal.h, which is used on
846        systems that lack both memmove() and bcopy() - that is, hardly ever -
847        was missing a "static" storage class specifier.
849     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
850        containing an extended class (one that cannot be represented by a bitmap
851        because it contains high-valued characters or Unicode property items, e.g.
852        [\pZ]). Almost always one would set UTF-8 mode when processing such a
853        pattern, but PCRE should not loop if you do not (it no longer does).
854        [Detail: two cases were found: (a) a repeated subpattern containing an
855        extended class; (b) a recursive reference to a subpattern that followed a
856        previous extended class. It wasn't skipping over the extended class
857        correctly when UTF-8 mode was not set.]
859     5. A negated single-character class was not being recognized as fixed-length
860        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
861        compile error "lookbehind assertion is not fixed length".
863     6. The RunPerlTest auxiliary script was showing an unexpected difference
864        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
865        write a Perl script that can interpret lines of an input file either as
866        byte characters or as UTF-8, which is what "perltest" was being required to
867        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
868        can't do is switch easily at run time between having the "use utf8;" pragma
869        or not. In the end, I fudged it by using the RunPerlTest script to insert
870        "use utf8;" explicitly for the UTF-8 tests.
872     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
873        the end of the subject string, contrary to the documentation and to what
874        Perl does. This was true of both matching functions. Now it matches only at
875        the start of the subject and immediately after *internal* newlines.
877     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
878        a pointer to an int instead of a pointer to an unsigned long int. This
879        caused problems on 64-bit systems.
881     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
882        instance of the 'standard' template library not being so standard".
884    10. There was no check on the number of named subpatterns nor the maximum
885        length of a subpattern name. The product of these values is used to compute
886        the size of the memory block for a compiled pattern. By supplying a very
887        long subpattern name and a large number of named subpatterns, the size
888        computation could be caused to overflow. This is now prevented by limiting
889        the length of names to 32 characters, and the number of named subpatterns
890        to 10,000.
892    11. Subpatterns that are repeated with specific counts have to be replicated in
893        the compiled pattern. The size of memory for this was computed from the
894        length of the subpattern and the repeat count. The latter is limited to
895        65535, but there was no limit on the former, meaning that integer overflow
896        could in principle occur. The compiled length of a repeated subpattern is
897        now limited to 30,000 bytes in order to prevent this.
899    12. Added the optional facility to have named substrings with the same name.
901    13. Added the ability to use a named substring as a condition, using the
902        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
903        are numbers (not recommended). Forward references are permitted.
905    14. Added forward references in named backreferences (if you see what I mean).
907    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
908        pattern could run off the end of the subject. For example, the pattern
909        "(?s)(.{1,5})"8 did this with the subject "ab".
911    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
912        PCRE_CASELESS was set when matching characters that were quantified with ?
913        or *.
915    17. A character class other than a single negated character that had a minimum
916        but no maximum quantifier - for example [ab]{6,} - was not handled
917        correctly by pce_dfa_exec(). It would match only one character.
919    18. A valid (though odd) pattern that looked like a POSIX character
920        class but used an invalid character after [ (for example [[,abc,]]) caused
921        pcre_compile() to give the error "Failed: internal error: code overflow" or
922        in some cases to crash with a glibc free() error. This could even happen if
923        the pattern terminated after [[ but there just happened to be a sequence of
924        letters, a binary zero, and a closing ] in the memory that followed.
926    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
927        over the years. Originally (before any Unicode support), just the bottom 8
928        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
929        output from "man perlunicode" includes this:
931          The regular expression compiler produces polymorphic opcodes.  That
932          is, the pattern adapts to the data and automatically switches to
933          the Unicode character scheme when presented with Unicode data--or
934          instead uses a traditional byte scheme when presented with byte
935          data.
937        Sadly, a wide octal escape does not cause a switch, and in a string with
938        no other multibyte characters, these octal escapes are treated as before.
939        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
940        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
941        Unicode string.
943        I have not perpetrated such confusion in PCRE. Up till now, it took just
944        the bottom 8 bits, as in old Perl. I have now made octal escapes with
945        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
946        translate to the appropriate multibyte character.
948    29. Applied some refactoring to reduce the number of warnings from Microsoft
949        and Borland compilers. This has included removing the fudge introduced
950        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
951        a warning about an unused variable.
953    21. PCRE has not included VT (character 0x0b) in the set of whitespace
954        characters since release 4.0, because Perl (from release 5.004) does not.
955        [Or at least, is documented not to: some releases seem to be in conflict
956        with the documentation.] However, when a pattern was studied with
957        pcre_study() and all its branches started with \s, PCRE still included VT
958        as a possible starting character. Of course, this did no harm; it just
959        caused an unnecessary match attempt.
961    22. Removed a now-redundant internal flag bit that recorded the fact that case
962        dependency changed within the pattern. This was once needed for "required
963        byte" processing, but is no longer used. This recovers a now-scarce options
964        bit. Also moved the least significant internal flag bit to the most-
965        significant bit of the word, which was not previously used (hangover from
966        the days when it was an int rather than a uint) to free up another bit for
967        the future.
969    23. Added support for CRLF line endings as well as CR and LF. As well as the
970        default being selectable at build time, it can now be changed at runtime
971        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
972        specify that it is scanning data with non-default line endings.
974    24. Changed the definition of CXXLINK to make it agree with the definition of
975        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
977    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
978        recursions. This makes a big different to stack usage for some patterns.
980    26. If a subpattern containing a named recursion or subroutine reference such
981        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
982        the space required for the compiled pattern went wrong and gave too small a
983        value. Depending on the environment, this could lead to "Failed: internal
984        error: code overflow at offset 49" or "glibc detected double free or
985        corruption" errors.
987    27. Applied patches from Google (a) to support the new newline modes and (b) to
988        advance over multibyte UTF-8 characters in GlobalReplace.
990    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
991        difference for some implementation of PCRE in some Windows version.
993    29. Added some extra testing facilities to pcretest:
995        \q<number>   in a data line sets the "match limit" value
996        \Q<number>   in a data line sets the "match recursion limt" value
997        -S <number>  sets the stack size, where <number> is in megabytes
999        The -S option isn't available for Windows.
1002    Version 6.6 06-Feb-06
1003    ---------------------
1005     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1006        in pcreposix.h. I have copied the definition from pcre.h.
1008     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1009        because pcre.h is no longer a built file.
1011     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1012        not normally included in the compiled code.
1015    Version 6.5 01-Feb-06
1016    ---------------------
1018     1. When using the partial match feature with pcre_dfa_exec(), it was not
1019        anchoring the second and subsequent partial matches at the new starting
1020        point. This could lead to incorrect results. For example, with the pattern
1021        /1234/, partially matching against "123" and then "a4" gave a match.
1023     2. Changes to pcregrep:
1025        (a) All non-match returns from pcre_exec() were being treated as failures
1026            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1027            error message is output. Some extra information is given for the
1028            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
1029            probably the only errors that are likely to be caused by users (by
1030            specifying a regex that has nested indefinite repeats, for instance).
1031            If there are more than 20 of these errors, pcregrep is abandoned.
1033        (b) A binary zero was treated as data while matching, but terminated the
1034            output line if it was written out. This has been fixed: binary zeroes
1035            are now no different to any other data bytes.
1037        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1038            used to set a locale for matching. The --locale=xxxx long option has
1039            been added (no short equivalent) to specify a locale explicitly on the
1040            pcregrep command, overriding the environment variables.
1042        (d) When -B was used with -n, some line numbers in the output were one less
1043            than they should have been.
1045        (e) Added the -o (--only-matching) option.
1047        (f) If -A or -C was used with -c (count only), some lines of context were
1048            accidentally printed for the final match.
1050        (g) Added the -H (--with-filename) option.
1052        (h) The combination of options -rh failed to suppress file names for files
1053            that were found from directory arguments.
1055        (i) Added the -D (--devices) and -d (--directories) options.
1057        (j) Added the -F (--fixed-strings) option.
1059        (k) Allow "-" to be used as a file name for -f as well as for a data file.
1061        (l) Added the --colo(u)r option.
1063        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1064            is not present by default.
1066     3. A nasty bug was discovered in the handling of recursive patterns, that is,
1067        items such as (?R) or (?1), when the recursion could match a number of
1068        alternatives. If it matched one of the alternatives, but subsequently,
1069        outside the recursion, there was a failure, the code tried to back up into
1070        the recursion. However, because of the way PCRE is implemented, this is not
1071        possible, and the result was an incorrect result from the match.
1073        In order to prevent this happening, the specification of recursion has
1074        been changed so that all such subpatterns are automatically treated as
1075        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1077     4. I had overlooked the fact that, in some locales, there are characters for
1078        which isalpha() is true but neither isupper() nor islower() are true. In
1079        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1080        and ordfeminine) are like this. This affected the treatment of \w and \W
1081        when they appeared in character classes, but not when they appeared outside
1082        a character class. The bit map for "word" characters is now created
1083        separately from the results of isalnum() instead of just taking it from the
1084        upper, lower, and digit maps. (Plus the underscore character, of course.)
1086     5. The above bug also affected the handling of POSIX character classes such as
1087        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1088        permanent tables. Instead, the bit maps for such a class were previously
1089        created as the appropriate unions of the upper, lower, and digit bitmaps.
1090        Now they are created by subtraction from the [[:word:]] class, which has
1091        its own bitmap.
1093     6. The [[:blank:]] character class matches horizontal, but not vertical space.
1094        It is created by subtracting the vertical space characters (\x09, \x0a,
1095        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1096        subtraction was done in the overall bitmap for a character class, meaning
1097        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1098        be recognized. This bug has been fixed.
1100     7. Patches from the folks at Google:
1102          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1103          real life, but is still worth protecting against".
1105          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1106          regular expressions".
1108          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1109          have it.
1111          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1112          "configure" and the latter not, in order to fix a problem somebody had
1113          with compiling the Arg class on HP-UX.
1115          (e) Improve the error-handling of the C++ wrapper a little bit.
1117          (f) New tests for checking recursion limiting.
1119     8. The pcre_memmove() function, which is used only if the environment does not
1120        have a standard memmove() function (and is therefore rarely compiled),
1121        contained two bugs: (a) use of int instead of size_t, and (b) it was not
1122        returning a result (though PCRE never actually uses the result).
1124     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1125        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1126        returned instead of calling malloc() with an overflowing number that would
1127        most likely cause subsequent chaos.
1129    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1131    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1132        with this option is matched, the nmatch and pmatch options of regexec() are
1133        ignored.
1135    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1136        provided in case anyone wants to the the POSIX interface with UTF-8
1137        strings.
1139    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1140        C++ linking (needed for some HP-UX environments).
1142    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1143        (unused parameter) and in the pcre_printint() function (omitted "default"
1144        switch label when the default is to do nothing).
1146    15. Added some code to make it possible, when PCRE is compiled as a C++
1147        library, to replace subject pointers for pcre_exec() with a smart pointer
1148        class, thus making it possible to process discontinuous strings.
1150    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1151        much the same function. They were added by different people who were trying
1152        to make PCRE easy to compile on non-Unix systems. It has been suggested
1153        that PCRE_EXPORT be abolished now that there is more automatic apparatus
1154        for compiling on Windows systems. I have therefore replaced it with
1155        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1156        defaults to "extern" for C or "extern C" for C++, which works fine on
1157        Unix-like systems. It is now possible to override the value of PCRE_DATA_
1158        SCOPE with something explicit in config.h. In addition:
1160        (a) pcreposix.h still had just "extern" instead of either of these macros;
1161            I have replaced it with PCRE_DATA_SCOPE.
1163        (b) Functions such as _pcre_xclass(), which are internal to the library,
1164            but external in the C sense, all had PCRE_EXPORT in their definitions.
1165            This is apparently wrong for the Windows case, so I have removed it.
1166            (It makes no difference on Unix-like systems.)
1168    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1169        of recursive calls to match(). This is different to MATCH_LIMIT because
1170        that limits the total number of calls to match(), not all of which increase
1171        the depth of recursion. Limiting the recursion depth limits the amount of
1172        stack (or heap if NO_RECURSE is set) that is used. The default can be set
1173        when PCRE is compiled, and changed at run time. A patch from Google adds
1174        this functionality to the C++ interface.
1176    18. Changes to the handling of Unicode character properties:
1178        (a) Updated the table to Unicode 4.1.0.
1180        (b) Recognize characters that are not in the table as "Cn" (undefined).
1182        (c) I revised the way the table is implemented to a much improved format
1183            which includes recognition of ranges. It now supports the ranges that
1184            are defined in UnicodeData.txt, and it also amalgamates other
1185            characters into ranges. This has reduced the number of entries in the
1186            table from around 16,000 to around 3,000, thus reducing its size
1187            considerably. I realized I did not need to use a tree structure after
1188            all - a binary chop search is just as efficient. Having reduced the
1189            number of entries, I extended their size from 6 bytes to 8 bytes to
1190            allow for more data.
1192        (d) Added support for Unicode script names via properties such as \p{Han}.
1194    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1195        matching that character.
1197    20. When matching a repeated Unicode property with a minimum greater than zero,
1198        (for example \pL{2,}), PCRE could look past the end of the subject if it
1199        reached it while seeking the minimum number of characters. This could
1200        happen only if some of the characters were more than one byte long, because
1201        there is a check for at least the minimum number of bytes.
1203    21. Refactored the implementation of \p and \P so as to be more general, to
1204        allow for more different types of property in future. This has changed the
1205        compiled form incompatibly. Anybody with saved compiled patterns that use
1206        \p or \P will have to recompile them.
1208    22. Added "Any" and "L&" to the supported property types.
1210    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1211        but give a compile time error if the value is greater than 0xff.
1213    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1214        accidentally not being installed or uninstalled.
1216    25. The pcre.h file was built from pcre.h.in, but the only changes that were
1217        made were to insert the current release number. This seemed silly, because
1218        it made things harder for people building PCRE on systems that don't run
1219        "configure". I have turned pcre.h into a distributed file, no longer built
1220        by "configure", with the version identification directly included. There is
1221        no longer a pcre.h.in file.
1223        However, this change necessitated a change to the pcre-config script as
1224        well. It is built from pcre-config.in, and one of the substitutions was the
1225        release number. I have updated configure.ac so that ./configure now finds
1226        the release number by grepping pcre.h.
1228    26. Added the ability to run the tests under valgrind.
1231    Version 6.4 05-Sep-05
1232    ---------------------
1234     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1235        "--" to be printed when multiple files were scanned, even when none of the
1236        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1237        consider it to be a bug, and have restored the previous behaviour.
1239     2. A couple of code tidies to get rid of compiler warnings.
1241     3. The pcretest program used to cheat by referring to symbols in the library
1242        whose names begin with _pcre_. These are internal symbols that are not
1243        really supposed to be visible externally, and in some environments it is
1244        possible to suppress them. The cheating is now confined to including
1245        certain files from the library's source, which is a bit cleaner.
1247     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1248        file's purpose clearer.
1250     5. Reorganized pcre_ucp_findchar().
1253    Version 6.3 15-Aug-05
1254    ---------------------
1256     1. The file libpcre.pc.in did not have general read permission in the tarball.
1258     2. There were some problems when building without C++ support:
1260        (a) If C++ support was not built, "make install" and "make test" still
1261            tried to test it.
1263        (b) There were problems when the value of CXX was explicitly set. Some
1264            changes have been made to try to fix these, and ...
1266        (c) --disable-cpp can now be used to explicitly disable C++ support.
1268        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1269            backslash in a target when C++ was disabled. This confuses some
1270            versions of "make", apparently. Using an intermediate variable solves
1271            this. (Same for CPP_LOBJ.)
1273     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1274        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1275        necessary on certain architectures.
1277     4. Added a setting of -export-symbols-regex to the link command to remove
1278        those symbols that are exported in the C sense, but actually are local
1279        within the library, and not documented. Their names all begin with
1280        "_pcre_". This is not a perfect job, because (a) we have to except some
1281        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1282        available (and never for static libraries). I have made a note to try to
1283        find a way round (a) in the future.
1286    Version 6.2 01-Aug-05
1287    ---------------------
1289     1. There was no test for integer overflow of quantifier values. A construction
1290        such as {1111111111111111} would give undefined results. What is worse, if
1291        a minimum quantifier for a parenthesized subpattern overflowed and became
1292        negative, the calculation of the memory size went wrong. This could have
1293        led to memory overwriting.
1295     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1297     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1298        operating environments where this matters.
1300     4. Applied Giuseppe Maxia's patch to add additional features for controlling
1301        PCRE options from within the C++ wrapper.
1303     5. Named capturing subpatterns were not being correctly counted when a pattern
1304        was compiled. This caused two problems: (a) If there were more than 100
1305        such subpatterns, the calculation of the memory needed for the whole
1306        compiled pattern went wrong, leading to an overflow error. (b) Numerical
1307        back references of the form \12, where the number was greater than 9, were
1308        not recognized as back references, even though there were sufficient
1309        previous subpatterns.
1311     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1312        versions of gcc, e.g. 2.95.4.
1315    Version 6.1 21-Jun-05
1316    ---------------------
1318     1. There was one reference to the variable "posix" in pcretest.c that was not
1319        surrounded by "#if !defined NOPOSIX".
1321     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1322        the cross-check on the old pcre_info() function, for the benefit of the
1323        cut-down version of PCRE that is currently imported into Exim.
1325     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1326        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1327        patterns that start (?i) but I don't think that matters. The use of (?i) is
1328        just an example; this all applies to the other options as well.
1330     4. Since libtool seems to echo the compile commands it is issuing, the output
1331        from "make" can be reduced a bit by putting "@" in front of each libtool
1332        compile command.
1334     5. Patch from the folks at Google for configure.in to be a bit more thorough
1335        in checking for a suitable C++ installation before trying to compile the
1336        C++ stuff. This should fix a reported problem when a compiler was present,
1337        but no suitable headers.
1339     6. The man pages all had just "PCRE" as their title. I have changed them to
1340        be the relevant file name. I have also arranged that these names are
1341        retained in the file doc/pcre.txt, which is a concatenation in text format
1342        of all the man pages except the little individual ones for each function.
1344     7. The NON-UNIX-USE file had not been updated for the different set of source
1345        files that come with release 6. I also added a few comments about the C++
1346        wrapper.
1349    Version 6.0 07-Jun-05
1350    ---------------------
1352     1. Some minor internal re-organization to help with my DFA experiments.
1354     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1355        didn't matter for the library itself when fully configured, but did matter
1356        when compiling without UCP support, or within Exim, where the ucp files are
1357        not imported.
1359     3. Refactoring of the library code to split up the various functions into
1360        different source modules. The addition of the new DFA matching code (see
1361        below) to a single monolithic source would have made it really too
1362        unwieldy, quite apart from causing all the code to be include in a
1363        statically linked application, when only some functions are used. This is
1364        relevant even without the DFA addition now that patterns can be compiled in
1365        one application and matched in another.
1367        The downside of splitting up is that there have to be some external
1368        functions and data tables that are used internally in different modules of
1369        the library but which are not part of the API. These have all had their
1370        names changed to start with "_pcre_" so that they are unlikely to clash
1371        with other external names.
1373     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1374        a different (DFA) algorithm. Although it is slower than the original
1375        function, it does have some advantages for certain types of matching
1376        problem.
1378     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1379        including restarting after a partial match.
1381     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1382        defined when compiling for Windows was sent to me. I have put it into the
1383        code, though I have no means of testing or verifying it.
1385     7. Added the pcre_refcount() auxiliary function.
1387     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1388        match before or at the first newline in the subject string. In pcretest,
1389        the /f option on a pattern can be used to set this.
1391     9. A repeated \w when used in UTF-8 mode with characters greater than 256
1392        would behave wrongly. This has been present in PCRE since release 4.0.
1394    10. A number of changes to the pcregrep command:
1396        (a) Refactored how -x works; insert ^(...)$ instead of setting
1397            PCRE_ANCHORED and checking the length, in preparation for adding
1398            something similar for -w.
1400        (b) Added the -w (match as a word) option.
1402        (c) Refactored the way lines are read and buffered so as to have more
1403            than one at a time available.
1405        (d) Implemented a pcregrep test script.
1407        (e) Added the -M (multiline match) option. This allows patterns to match
1408            over several lines of the subject. The buffering ensures that at least
1409            8K, or the rest of the document (whichever is the shorter) is available
1410            for matching (and similarly the previous 8K for lookbehind assertions).
1412        (f) Changed the --help output so that it now says
1414              -w, --word-regex(p)
1416            instead of two lines, one with "regex" and the other with "regexp"
1417            because that confused at least one person since the short forms are the
1418            same. (This required a bit of code, as the output is generated
1419            automatically from a table. It wasn't just a text change.)
1421        (g) -- can be used to terminate pcregrep options if the next thing isn't an
1422            option but starts with a hyphen. Could be a pattern or a path name
1423            starting with a hyphen, for instance.
1425        (h) "-" can be given as a file name to represent stdin.
1427        (i) When file names are being printed, "(standard input)" is used for
1428            the standard input, for compatibility with GNU grep. Previously
1429            "<stdin>" was used.
1431        (j) The option --label=xxx can be used to supply a name to be used for
1432            stdin when file names are being printed. There is no short form.
1434        (k) Re-factored the options decoding logic because we are going to add
1435            two more options that take data. Such options can now be given in four
1436            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1438        (l) Added the -A, -B, and -C options for requesting that lines of context
1439            around matches be printed.
1441        (m) Added the -L option to print the names of files that do not contain
1442            any matching lines, that is, the complement of -l.
1444        (n) The return code is 2 if any file cannot be opened, but pcregrep does
1445            continue to scan other files.
1447        (o) The -s option was incorrectly implemented. For compatibility with other
1448            greps, it now suppresses the error message for a non-existent or non-
1449            accessible file (but not the return code). There is a new option called
1450            -q that suppresses the output of matching lines, which was what -s was
1451            previously doing.
1453        (p) Added --include and --exclude options to specify files for inclusion
1454            and exclusion when recursing.
1456    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1457        Hopefully, it now does.
1459    12. Missing cast in pcre_study().
1461    13. Added an "uninstall" target to the makefile.
1463    14. Replaced "extern" in the function prototypes in Makefile.in with
1464        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1465        world, but is set differently for Windows.
1467    15. Added a second compiling function called pcre_compile2(). The only
1468        difference is that it has an extra argument, which is a pointer to an
1469        integer error code. When there is a compile-time failure, this is set
1470        non-zero, in addition to the error test pointer being set to point to an
1471        error message. The new argument may be NULL if no error number is required
1472        (but then you may as well call pcre_compile(), which is now just a
1473        wrapper). This facility is provided because some applications need a
1474        numeric error indication, but it has also enabled me to tidy up the way
1475        compile-time errors are handled in the POSIX wrapper.
1477    16. Added VPATH=.libs to the makefile; this should help when building with one
1478        prefix path and installing with another. (Or so I'm told by someone who
1479        knows more about this stuff than I do.)
1481    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1482        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1483        match everything, including newlines. This is not POSIX-compatible, but
1484        somebody wanted the feature. From pcretest it can be activated by using
1485        both the P and the s flags.
1487    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1489    19. libpcre.pc was being incorrectly installed as executable.
1491    20. A couple of places in pcretest check for end-of-line by looking for '\n';
1492        it now also looks for '\r' so that it will work unmodified on Windows.
1494    21. Added Google's contributed C++ wrapper to the distribution.
1496    22. Added some untidy missing memory free() calls in pcretest, to keep
1497        Electric Fence happy when testing.
1501    Version 5.0 13-Sep-04
1502    ---------------------
1504     1. Internal change: literal characters are no longer packed up into items
1505        containing multiple characters in a single byte-string. Each character
1506        is now matched using a separate opcode. However, there may be more than one
1507        byte in the character in UTF-8 mode.
1509     2. The pcre_callout_block structure has two new fields: pattern_position and
1510        next_item_length. These contain the offset in the pattern to the next match
1511        item, and its length, respectively.
1513     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1514        insertion of callouts before each pattern item. Added the /C option to
1515        pcretest to make use of this.
1517     4. On the advice of a Windows user, the lines
1519          #if defined(_WIN32) || defined(WIN32)
1520          _setmode( _fileno( stdout ), 0x8000 );
1521          #endif  /* defined(_WIN32) || defined(WIN32) */
1523        have been added to the source of pcretest. This apparently does useful
1524        magic in relation to line terminators.
1526     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1527        for the benefit of those environments where the "b" makes a difference.
1529     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1530        to know about it. I have put a hack into configure.in that adds in code
1531        to set GCC=yes if CC=icc. This seems to end up at a point in the
1532        generated configure script that is early enough to affect the setting of
1533        compiler options, which is what is needed, but I have no means of testing
1534        whether it really works. (The user who reported this had patched the
1535        generated configure script, which of course I cannot do.)
1537        LATER: After change 22 below (new libtool files), the configure script
1538        seems to know about icc (and also ecc). Therefore, I have commented out
1539        this hack in configure.in.
1541     7. Added support for pkg-config (2 patches were sent in).
1543     8. Negated POSIX character classes that used a combination of internal tables
1544        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1545        [[:^ascii]]. Typically, they would match almost any characters. The other
1546        POSIX classes were not broken in this way.
1548     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1549        to find the match, as PCRE was deluded into thinking that the match had to
1550        start at the start point or following a newline. The same bug applied to
1551        patterns with negative forward assertions or any backward assertions
1552        preceding ".*" at the start, unless the pattern required a fixed first
1553        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1555    10. In UTF-8 mode, when moving forwards in the subject after a failed match
1556        starting at the last subject character, bytes beyond the end of the subject
1557        string were read.
1559    11. Renamed the variable "class" as "classbits" to make life easier for C++
1560        users. (Previously there was a macro definition, but it apparently wasn't
1561        enough.)
1563    12. Added the new field "tables" to the extra data so that tables can be passed
1564        in at exec time, or the internal tables can be re-selected. This allows
1565        a compiled regex to be saved and re-used at a later time by a different
1566        program that might have everything at different addresses.
1568    13. Modified the pcre-config script so that, when run on Solaris, it shows a
1569        -R library as well as a -L library.
1571    14. The debugging options of pcretest (-d on the command line or D on a
1572        pattern) showed incorrect output for anything following an extended class
1573        that contained multibyte characters and which was followed by a quantifier.
1575    15. Added optional support for general category Unicode character properties
1576        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1577        support. It adds about 90K to the size of the library. The meanings of the
1578        inbuilt class escapes such as \d and \s have NOT been changed.
1580    16. Updated pcredemo.c to include calls to free() to release the memory for the
1581        compiled pattern.
1583    17. The generated file chartables.c was being created in the source directory
1584        instead of in the building directory. This caused the build to fail if the
1585        source directory was different from the building directory, and was
1586        read-only.
1588    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1589        file. No doubt somebody will tell me if they don't make sense... Also added
1590        Dan Mooney's comments about building on OpenVMS.
1592    19. Added support for partial matching via the PCRE_PARTIAL option for
1593        pcre_exec() and the \P data escape in pcretest.
1595    20. Extended pcretest with 3 new pattern features:
1597        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
1598              write the compiled pattern to the file whose name is "rest-of-line".
1599              This is a straight binary dump of the data, with the saved pointer to
1600              the character tables forced to be NULL. The study data, if any, is
1601              written too. After writing, pcretest reads a new pattern.
1603        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1604              compiled pattern from the given file. There must not be any
1605              occurrences of "<" in the file name (pretty unlikely); if there are,
1606              pcretest will instead treat the initial "<" as a pattern delimiter.
1607              After reading in the pattern, pcretest goes on to read data lines as
1608              usual.
1610        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1611              and 16-bit fields in a compiled pattern, to simulate a pattern that
1612              was compiled on a host of opposite endianness.
1614    21. The pcre-exec() function can now cope with patterns that were compiled on
1615        hosts of opposite endianness, with this restriction:
1617          As for any compiled expression that is saved and used later, the tables
1618          pointer field cannot be preserved; the extra_data field in the arguments
1619          to pcre_exec() should be used to pass in a tables address if a value
1620          other than the default internal tables were used at compile time.
1622    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1623        now diagnosed as an error. Previously, most of the time, a negative number
1624        would have been treated as zero, but if in addition "ovector" was passed as
1625        NULL, a crash could occur.
1627    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1628        new versions from the libtool 1.5 distribution (the last one is a copy of
1629        a file called libtool.m4). This seems to have fixed the need to patch
1630        "configure" to support Darwin 1.3 (which I used to do). However, I still
1631        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1632        workstation).
1634    24. Changed the PCRE licence to be the more standard "BSD" licence.
1637    Version 4.5 01-Dec-03
1638    ---------------------
1640     1. There has been some re-arrangement of the code for the match() function so
1641        that it can be compiled in a version that does not call itself recursively.
1642        Instead, it keeps those local variables that need separate instances for
1643        each "recursion" in a frame on the heap, and gets/frees frames whenever it
1644        needs to "recurse". Keeping track of where control must go is done by means
1645        of setjmp/longjmp. The whole thing is implemented by a set of macros that
1646        hide most of the details from the main code, and operates only if
1647        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1648        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1649        operating.
1651        To make it easier for callers to provide specially tailored get/free
1652        functions for this usage, two new functions, pcre_stack_malloc, and
1653        pcre_stack_free, are used. They are always called in strict stacking order,
1654        and the size of block requested is always the same.
1656        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1657        PCRE has been compiled to use the stack or the heap for recursion. The
1658        -C option of pcretest uses this to show which version is compiled.
1660        A new data escape \S, is added to pcretest; it causes the amounts of store
1661        obtained and freed by both kinds of malloc/free at match time to be added
1662        to the output.
1664     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1665        what's available on my current Linux desktop machine.
1667     3. When matching a UTF-8 string, the test for a valid string at the start has
1668        been extended. If start_offset is not zero, PCRE now checks that it points
1669        to a byte that is the start of a UTF-8 character. If not, it returns
1670        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1671        this is necessary because there may be backward assertions in the pattern.
1672        When matching the same subject several times, it may save resources to use
1673        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1675     4. The code for checking the validity of UTF-8 strings has been tightened so
1676        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1677        containing "overlong sequences".
1679     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1680        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1681        should have been "&", but it just so happened that all the cases this let
1682        through by mistake were picked up later in the function.
1684     6. I had used a variable called "isblank" - this is a C99 function, causing
1685        some compilers to warn. To avoid this, I renamed it (as "blankclass").
1687     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1688        prompting; (b) run "./pcretest /dev/null" at the start of the test script
1689        so the version is shown; (c) stop "make test" echoing "./RunTest".
1691     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1693     9. The prototype for memmove() for systems that don't have it was using
1694        size_t, but the inclusion of the header that defines size_t was later. I've
1695        moved the #includes for the C headers earlier to avoid this.
1697    10. Added some adjustments to the code to make it easier to compiler on certain
1698        special systems:
1700          (a) Some "const" qualifiers were missing.
1701          (b) Added the macro EXPORT before all exported functions; by default this
1702              is defined to be empty.
1703          (c) Changed the dftables auxiliary program (that builds chartables.c) so
1704              that it reads its output file name as an argument instead of writing
1705              to the standard output and assuming this can be redirected.
1707    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
1708        class containing characters with values greater than 255, PCRE compilation
1709        went into a loop.
1711    12. A recursive reference to a subpattern that was within another subpattern
1712        that had a minimum quantifier of zero caused PCRE to crash. For example,
1713        (x(y(?2))z)? provoked this bug with a subject that got as far as the
1714        recursion. If the recursively-called subpattern itself had a zero repeat,
1715        that was OK.
1717    13. In pcretest, the buffer for reading a data line was set at 30K, but the
1718        buffer into which it was copied (for escape processing) was still set at
1719        1024, so long lines caused crashes.
1721    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
1722        "internal error: code overflow...". This applied to any character class
1723        that was followed by a possessive quantifier.
1725    15. Modified the Makefile to add libpcre.la as a prerequisite for
1726        libpcreposix.la because I was told this is needed for a parallel build to
1727        work.
1729    16. If a pattern that contained .* following optional items at the start was
1730        studied, the wrong optimizing data was generated, leading to matching
1731        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
1732        matching string must start with a or b or c. The correct conclusion for
1733        this pattern is that a match can start with any character.
1736    Version 4.4 13-Aug-03
1737    ---------------------
1739     1. In UTF-8 mode, a character class containing characters with values between
1740        127 and 255 was not handled correctly if the compiled pattern was studied.
1741        In fixing this, I have also improved the studying algorithm for such
1742        classes (slightly).
1744     2. Three internal functions had redundant arguments passed to them. Removal
1745        might give a very teeny performance improvement.
1747     3. Documentation bug: the value of the capture_top field in a callout is *one
1748        more than* the number of the hightest numbered captured substring.
1750     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
1751        in incorrectly linking with a previously installed version. They now link
1752        explicitly with libpcre.la.
1754     5. configure.in no longer needs to recognize Cygwin specially.
1756     6. A problem in pcre.in for Windows platforms is fixed.
1758     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
1759        pcretest, it used to include the size of the study block as part of its
1760        output. Unfortunately, the structure contains a field that has a different
1761        size on different hardware architectures. This meant that the tests that
1762        showed this size failed. As the block is currently always of a fixed size,
1763        this information isn't actually particularly useful in pcretest output, so
1764        I have just removed it.
1766     8. Three pre-processor statements accidentally did not start in column 1.
1767        Sadly, there are *still* compilers around that complain, even though
1768        standard C has not required this for well over a decade. Sigh.
1770     9. In pcretest, the code for checking callouts passed small integers in the
1771        callout_data field, which is a void * field. However, some picky compilers
1772        complained about the casts involved for this on 64-bit systems. Now
1773        pcretest passes the address of the small integer instead, which should get
1774        rid of the warnings.
1776    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
1777        both compile and run time, and gives an error if an invalid UTF-8 sequence
1778        is found. There is a option for disabling this check in cases where the
1779        string is known to be correct and/or the maximum performance is wanted.
1781    11. In response to a bug report, I changed one line in Makefile.in from
1783            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
1784        to
1785            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
1787        to look similar to other lines, but I have no way of telling whether this
1788        is the right thing to do, as I do not use Windows. No doubt I'll get told
1789        if it's wrong...
1792    Version 4.3 21-May-03
1793    ---------------------
1795    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
1796       Makefile.
1798    2. Some refactoring to improve the quality of the code:
1800       (i)   The utf8_table... variables are now declared "const".
1802       (ii)  The code for \cx, which used the "case flipping" table to upper case
1803             lower case letters, now just substracts 32. This is ASCII-specific,
1804             but the whole concept of \cx is ASCII-specific, so it seems
1805             reasonable.
1807       (iii) PCRE was using its character types table to recognize decimal and
1808             hexadecimal digits in the pattern. This is silly, because it handles
1809             only 0-9, a-f, and A-F, but the character types table is locale-
1810             specific, which means strange things might happen. A private
1811             table is now used for this - though it costs 256 bytes, a table is
1812             much faster than multiple explicit tests. Of course, the standard
1813             character types table is still used for matching digits in subject
1814             strings against \d.
1816       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
1817             ending in _t are). So I've renamed it as ESC_tee.
1819    3. The first argument for regexec() in the POSIX wrapper should have been
1820       defined as "const".
1822    4. Changed pcretest to use malloc() for its buffers so that they can be
1823       Electric Fenced for debugging.
1825    5. There were several places in the code where, in UTF-8 mode, PCRE would try
1826       to read one or more bytes before the start of the subject string. Often this
1827       had no effect on PCRE's behaviour, but in some circumstances it could
1828       provoke a segmentation fault.
1830    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
1831       to try to read one or more bytes before the start of the subject string.
1833    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
1834       UTF-8 support could misbehave in various ways if the subject string
1835       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
1836       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
1837       back over UTF-8 characters.)
1840    Version 4.2 14-Apr-03
1841    ---------------------
1843    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
1845    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
1846         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
1847         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
1848         [WIN_PREFIX]: new variable, "cyg" for Cygwin
1849         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
1850           and BUILD_EXEEXT
1851         Note: automatic setting of the BUILD variables is not yet working
1852         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
1853           compile-time but not at link-time
1854         [LINK]: use for linking executables only
1855         make different versions for Windows and non-Windows
1856         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
1857           libraries
1858         [LINK_FOR_BUILD]: new variable
1859         [OBJEXT]: use throughout
1860         [EXEEXT]: use throughout
1861         <winshared>: new target
1862         <wininstall>: new target
1863         <dftables.o>: use native compiler
1864         <dftables>: use native linker
1865         <install>: handle Windows platform correctly
1866         <clean>: ditto
1867         <check>: ditto
1868         copy DLL to top builddir before testing
1870       As part of these changes, -no-undefined was removed again. This was reported
1871       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
1872       in any case.
1874    3. Some tidies to get rid of compiler warnings:
1876       . In the match_data structure, match_limit was an unsigned long int, whereas
1877         match_call_count was an int. I've made them both unsigned long ints.
1879       . In pcretest the fact that a const uschar * doesn't automatically cast to
1880         a void * provoked a warning.
1882       . Turning on some more compiler warnings threw up some "shadow" variables
1883         and a few more missing casts.
1885    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
1886       option, a class that contained a single character with a value between 128
1887       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
1889    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
1890       option, a class that contained several characters, but with at least one
1891       whose value was between 128 and 255 caused PCRE to crash.
1894    Version 4.1 12-Mar-03
1895    ---------------------
1897    1. Compiling with gcc -pedantic found a couple of places where casts were
1898    needed, and a string in dftables.c that was longer than standard compilers are
1899    required to support.
1901    2. Compiling with Sun's compiler found a few more places where the code could
1902    be tidied up in order to avoid warnings.
1904    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
1905    first of these names is deprecated in the latest Autoconf in favour of the name
1906    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
1907    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
1908    analogy I have changed it to CFLAGS_FOR_BUILD.
1910    4. Added -no-undefined to the linking command in the Makefile, because this is
1911    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
1912    linking step for the pcreposix library.
1914    5. PCRE was failing to diagnose the case of two named groups with the same
1915    name.
1917    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
1918    literal character that is needed in the subject for a match, and scans along to
1919    ensure that it is present before embarking on the full matching process. This
1920    saves time in cases of nested unlimited repeats that are never going to match.
1921    Problem: the scan can take a lot of time if the subject is very long (e.g.
1922    megabytes), thus penalizing straightforward matches. It is now done only if the
1923    amount of subject to be scanned is less than 1000 bytes.
1925    7. A lesser problem with the same optimization is that it was recording the
1926    first character of an anchored pattern as "needed", thus provoking a search
1927    right along the subject, even when the first match of the pattern was going to
1928    fail. The "needed" character is now not set for anchored patterns, unless it
1929    follows something in the pattern that is of non-fixed length. Thus, it still
1930    fulfils its original purpose of finding quick non-matches in cases of nested
1931    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
1934    Version 4.0 17-Feb-03
1935    ---------------------
1937    1. If a comment in an extended regex that started immediately after a meta-item
1938    extended to the end of string, PCRE compiled incorrect data. This could lead to
1939    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
1941    2. Moved to autoconf 2.53 and libtool 1.4.2.
1943    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
1944    the special perltest8 script is no longer needed - all the tests can be run
1945    from a single perltest script.
1947    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
1948    by \s. It has now been removed in PCRE. This means it isn't recognized as
1949    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
1950    class [:space:] *does* include VT, thereby creating a mess.
1952    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
1953    space and tab.
1955    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
1956    its new features into the main test script, reducing the number of scripts.
1958    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
1959    were backward compatible, and made the (?i) apply to the whole pattern, as if
1960    /i were given. Now it behaves more logically, and applies the option setting
1961    only to what follows. PCRE has been changed to follow suit. However, if it
1962    finds options settings right at the start of the pattern, it extracts them into
1963    the global options, as before. Thus, they show up in the info data.
1965    8. Added support for the \Q...\E escape sequence. Characters in between are
1966    treated as literals. This is slightly different from Perl in that $ and @ are
1967    also handled as literals inside the quotes. In Perl, they will cause variable
1968    interpolation. Note the following examples:
1970        Pattern            PCRE matches      Perl matches
1972        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
1973        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
1974        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
1976    For compatibility with Perl, \Q...\E sequences are recognized inside character
1977    classes as well as outside them.
1979    9. Re-organized 3 code statements in pcretest to avoid "overflow in
1980    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
1981    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
1982    signed/unsigned warnings.
1984    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
1985    option for pcretest, so I've replaced it by a simple function that does just
1986    that job.
1988    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
1989    "pcregrep -".
1991    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
1992    Java package. This provides some syntactic sugar for simple cases of what my
1993    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
1994    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
1995    item, you can use this simplified notation. Note that only makes sense with
1996    greedy quantifiers. Consequently, the use of the possessive quantifier forces
1997    greediness, whatever the setting of the PCRE_UNGREEDY option.
1999    13. A change of greediness default within a pattern was not taking effect at
2000    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
2001    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
2002    was abstracted outside.
2004    14. PCRE now supports the \G assertion. It is true when the current matching
2005    position is at the start point of the match. This differs from \A when the
2006    starting offset is non-zero. Used with the /g option of pcretest (or similar
2007    code), it works in the same way as it does for Perl's /g option. If all
2008    alternatives of a regex begin with \G, the expression is anchored to the start
2009    match position, and the "anchored" flag is set in the compiled expression.
2011    15. Some bugs concerning the handling of certain option changes within patterns
2012    have been fixed. These applied to options other than (?ims). For example,
2013    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
2014    been the other way round. Some of this was related to change 7 above.
2016    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
2017    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
2018    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
2019    POSIX classes only within a class (e.g. /[[:alpha:]]/).
2021    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
2022    mode. Unlike ".", it always matches newline, whatever the setting of
2023    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
2024    assertions. Perl allows it, but it doesn't (in general) work because it can't
2025    calculate the length of the lookbehind. At least, that's the case for Perl
2026    5.8.0 - I've been told they are going to document that it doesn't work in
2027    future.
2029    18. Added an error diagnosis for escapes that PCRE does not support: these are
2030    \L, \l, \N, \P, \p, \U, \u, and \X.
2032    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
2033    reading past the end of the pattern in cases such as /[abcd/.
2035    20. PCRE was getting more memory than necessary for patterns with classes that
2036    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
2038    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
2039    compiling PCRE for use with Virtual Pascal.
2041    22. Small fix to the Makefile to make it work properly if the build is done
2042    outside the source tree.
2044    23. Added a new extension: a condition to go with recursion. If a conditional
2045    subpattern starts with (?(R) the "true" branch is used if recursion has
2046    happened, whereas the "false" branch is used only at the top level.
2048    24. When there was a very long string of literal characters (over 255 bytes
2049    without UTF support, over 250 bytes with UTF support), the computation of how
2050    much memory was required could be incorrect, leading to segfaults or other
2051    strange effects.
2053    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
2054    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
2055    there was a subsequent back reference to those brackets. This meant that, for
2056    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
2057    possible to check for precisely this case. All we can do is abandon the
2058    optimization if .* occurs inside capturing brackets when there are any back
2059    references whatsoever. (See below for a better fix that came later.)
2061    26. The handling of the optimization for finding the first character of a
2062    non-anchored pattern, and for finding a character that is required later in the
2063    match were failing in some cases. This didn't break the matching; it just
2064    failed to optimize when it could. The way this is done has been re-implemented.
2066    27. Fixed typo in error message for invalid (?R item (it said "(?p").
2068    28. Added a new feature that provides some of the functionality that Perl
2069    provides with (?{...}). The facility is termed a "callout". The way it is done
2070    in PCRE is for the caller to provide an optional function, by setting
2071    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
2072    global variable. By default it is unset, which disables all calling out. To get
2073    the function called, the regex must include (?C) at appropriate points. This
2074    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
2075    This provides a means of identifying different callout points. When PCRE
2076    reaches such a point in the regex, if pcre_callout has been set, the external
2077    function is called. It is provided with data in a structure called
2078    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
2079    matching continues; if it returns a non-zero value, the match at the current
2080    point fails. However, backtracking will occur if possible. [This was changed
2081    later and other features added - see item 49 below.]
2083    29. pcretest is upgraded to test the callout functionality. It provides a
2084    callout function that displays information. By default, it shows the start of
2085    the match and the current position in the text. There are some new data escapes
2086    to vary what happens:
2088        \C+         in addition, show current contents of captured substrings
2089        \C-         do not supply a callout function
2090        \C!n        return 1 when callout number n is reached
2091        \C!n!m      return 1 when callout number n is reached for the mth time
2093    30. If pcregrep was called with the -l option and just a single file name, it
2094    output "<stdin>" if a match was found, instead of the file name.
2096    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
2097    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
2098    pcre_exec(). This saves a malloc/free per call. The default value of
2099    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
2100    when configuring.
2102    32. The default maximum size of a compiled pattern is 64K. There have been a
2103    few cases of people hitting this limit. The code now uses macros to handle the
2104    storing of links as offsets within the compiled pattern. It defaults to 2-byte
2105    links, but this can be changed to 3 or 4 bytes by --with-link-size when
2106    configuring. Tests 2 and 5 work only with 2-byte links because they output
2107    debugging information about compiled patterns.
2109    33. Internal code re-arrangements:
2111    (a) Moved the debugging function for printing out a compiled regex into
2112        its own source file (printint.c) and used #include to pull it into
2113        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
2114        separate copies.
2116    (b) Defined the list of op-code names for debugging as a macro in
2117        internal.h so that it is next to the definition of the opcodes.
2119    (c) Defined a table of op-code lengths for simpler skipping along compiled
2120        code. This is again a macro in internal.h so that it is next to the
2121        definition of the opcodes.
2123    34. Added support for recursive calls to individual subpatterns, along the
2124    lines of Robin Houston's patch (but implemented somewhat differently).
2126    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
2127    allow it to read and process whole directories in Win32. This code was
2128    contributed by Lionel Fourquaux; it has not been tested by me.
2130    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
2131    used to name a group. Names consist of alphanumerics and underscores, and must
2132    be unique. Back references use the syntax (?P=name) and recursive calls use
2133    (?P>name) which is a PCRE extension to the Python extension. Groups still have
2134    numbers. The function pcre_fullinfo() can be used after compilation to extract
2135    a name/number map. There are three relevant calls:
2137      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
2138      PCRE_INFO_NAMECOUNT            yields the number of entries
2139      PCRE_INFO_NAMETABLE            yields a pointer to the map.
2141    The map is a vector of fixed-size entries. The size of each entry depends on
2142    the length of the longest name used. The first two bytes of each entry are the
2143    group number, most significant byte first. There follows the corresponding
2144    name, zero terminated. The names are in alphabetical order.
2146    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
2147    case instead of 255. Making it the same both with and without UTF-8 support
2148    means that the same test output works with both.
2150    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
2151    calling malloc() with a zero argument.
2153    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
2154    optimization. I've improved things by keeping a bitmap of backreferences with
2155    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
2156    fact referenced, the optimization can be applied. It is unlikely that a
2157    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
2158    the match to follow \n) will appear inside brackets with a number greater than
2159    31, but if it does, any back reference > 31 suppresses the optimization.
2161    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
2162    of disabling numbered capturing parentheses. Any opening parenthesis that is
2163    not followed by ? behaves as if it were followed by ?: but named parentheses
2164    can still be used for capturing (and they will acquire numbers in the usual
2165    way).
2167    41. Redesigned the return codes from the match() function into yes/no/error so
2168    that errors can be passed back from deep inside the nested calls. A malloc
2169    failure while inside a recursive subpattern call now causes the
2170    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
2172    42. It is now possible to set a limit on the number of times the match()
2173    function is called in a call to pcre_exec(). This facility makes it possible to
2174    limit the amount of recursion and backtracking, though not in a directly
2175    obvious way, because the match() function is used in a number of different
2176    circumstances. The count starts from zero for each position in the subject
2177    string (for non-anchored patterns). The default limit is, for compatibility, a
2178    large number, namely 10 000 000. You can change this in two ways:
2180    (a) When configuring PCRE before making, you can use --with-match-limit=n
2181        to set a default value for the compiled library.
2183    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
2184        a different value is set. See 45 below.
2186    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
2188    43. Added a new function pcre_config(int, void *) to enable run-time extraction
2189    of things that can be changed at compile time. The first argument specifies
2190    what is wanted and the second points to where the information is to be placed.
2191    The current list of available information is:
2193      PCRE_CONFIG_UTF8
2195    The output is an integer that is set to one if UTF-8 support is available;
2196    otherwise it is set to zero.
2200    The output is an integer that it set to the value of the code that is used for
2201    newline. It is either LF (10) or CR (13).
2205    The output is an integer that contains the number of bytes used for internal
2206    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
2210    The output is an integer that contains the threshold above which the POSIX
2211    interface uses malloc() for output vectors. See item 31 above.
2215    The output is an unsigned integer that contains the default limit of the number
2216    of match() calls in a pcre_exec() execution. See 42 above.
2218    44. pcretest has been upgraded by the addition of the -C option. This causes it
2219    to extract all the available output from the new pcre_config() function, and to
2220    output it. The program then exits immediately.
2222    45. A need has arisen to pass over additional data with calls to pcre_exec() in
2223    order to support additional features. One way would have been to define
2224    pcre_exec2() (for example) with extra arguments, but this would not have been
2225    extensible, and would also have required all calls to the original function to
2226    be mapped to the new one. Instead, I have chosen to extend the mechanism that
2227    is used for passing in "extra" data from pcre_study().
2229    The pcre_extra structure is now exposed and defined in pcre.h. It currently
2230    contains the following fields:
2232      flags         a bitmap indicating which of the following fields are set
2233      study_data    opaque data from pcre_study()
2234      match_limit   a way of specifying a limit on match() calls for a specific
2235                      call to pcre_exec()
2236      callout_data  data for callouts (see 49 below)
2238    The flag bits are also defined in pcre.h, and are
2244    The pcre_study() function now returns one of these new pcre_extra blocks, with
2245    the actual study data pointed to by the study_data field, and the
2246    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
2247    before. That is, this change is entirely upwards-compatible and requires no
2248    change to existing code.
2250    If you want to pass in additional data to pcre_exec(), you can either place it
2251    in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
2252    block.
2254    46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
2255    data string contains the escape sequence \M, pcretest calls pcre_exec() several
2256    times with different match limits, until it finds the minimum value needed for
2257    pcre_exec() to complete. The value is then output. This can be instructive; for
2258    most simple matches the number is quite small, but for pathological cases it
2259    gets very large very quickly.
2261    47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
2262    returns the size of the data block pointed to by the study_data field in a
2263    pcre_extra block, that is, the value that was passed as the argument to
2264    pcre_malloc() when PCRE was getting memory in which to place the information
2265    created by pcre_study(). The fourth argument should point to a size_t variable.
2266    pcretest has been extended so that this information is shown after a successful
2267    pcre_study() call when information about the compiled regex is being displayed.
2269    48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
2270    because what follows is always an absolute path. (Later: it turns out that this
2271    is more than cosmetic for MinGW, because it doesn't like empty path
2272    components.)
2274    49. Some changes have been made to the callout feature (see 28 above):
2276    (i)  A callout function now has three choices for what it returns:
2278           0  =>  success, carry on matching
2279         > 0  =>  failure at this point, but backtrack if possible
2280         < 0  =>  serious error, return this value from pcre_exec()
2282         Negative values should normally be chosen from the set of PCRE_ERROR_xxx
2283         values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
2284         "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
2285         use by callout functions. It will never be used by PCRE itself.
2287    (ii) The pcre_extra structure (see 45 above) has a void * field called
2288         callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
2289         pcre_callout_block structure has a field of the same name. The contents of
2290         the field passed in the pcre_extra structure are passed to the callout
2291         function in the corresponding field in the callout block. This makes it
2292         easier to use the same callout-containing regex from multiple threads. For
2293         testing, the pcretest program has a new data escape
2295           \C*n        pass the number n (may be negative) as callout_data
2297         If the callout function in pcretest receives a non-zero value as
2298         callout_data, it returns that value.
2300    50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
2301    there were some redundant $(CFLAGS) in commands that are now specified as
2302    $(LINK), which already includes $(CFLAGS).
2304    51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
2305    has been compiled with UTF-8 support *and* pcre_compile() has been compiled
2306    with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
2307    one-byte characters throughout. Note that case-insensitive matching applies
2308    only to characters whose values are less than 256. PCRE doesn't support the
2309    notion of cases for higher-valued characters.
2311    (i)   A character class whose characters are all within 0-255 is handled as
2312          a bit map, and the map is inverted for negative classes. Previously, a
2313          character > 255 always failed to match such a class; however it should
2314          match if the class was a negative one (e.g. [^ab]). This has been fixed.
2316    (ii)  A negated character class with a single character < 255 is coded as
2317          "not this character" (OP_NOT). This wasn't working properly when the test
2318          character was multibyte, either singly or repeated.
2320    (iii) Repeats of multibyte characters are now handled correctly in UTF-8
2321          mode, for example: \x{100}{2,3}.
2323    (iv)  The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
2324          singly or repeated) now correctly test multibyte characters. However,
2325          PCRE doesn't recognize any characters with values greater than 255 as
2326          digits, spaces, or word characters. Such characters always match \D, \S,
2327          and \W, and never match \d, \s, or \w.
2329    (v)   Classes may now contain characters and character ranges with values
2330          greater than 255. For example: [ab\x{100}-\x{400}].
2332    (vi)  pcregrep now has a --utf-8 option (synonym -u) which makes it call
2333          PCRE in UTF-8 mode.
2335    52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
2336    PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
2337    retained for backwards compatibility. (Note that LASTLITERAL is also a byte
2338    value.)
2340    53. The single man page has become too large. I have therefore split it up into
2341    a number of separate man pages. These also give rise to individual HTML pages;
2342    these are now put in a separate directory, and there is an index.html page that
2343    lists them all. Some hyperlinking between the pages has been installed.
2345    54. Added convenience functions for handling named capturing parentheses.
2347    55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2348    aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2349    true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2350    are faulted.
2352    56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2353    calling configure. These values are used when compiling the dftables.c program
2354    which is run to generate the source of the default character tables. They
2355    default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2356    you will need to set these values.
2358    57. Updated the building process for Windows DLL, as provided by Fred Cox.
2361    Version 3.9 02-Jan-02
2362    ---------------------
2364    1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2366    2. If --disable-static was given, the building process failed when trying to
2367    build pcretest and pcregrep. (For some reason it was using libtool to compile
2368    them, which is not right, as they aren't part of the library.)
2371    Version 3.8 18-Dec-01
2372    ---------------------
2374    1. The experimental UTF-8 code was completely screwed up. It was packing the
2375    bytes in the wrong order. How dumb can you get?
2378    Version 3.7 29-Oct-01
2379    ---------------------
2381    1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2382    This caused pcretest, when used on the test data, to segfault. Unfortunately,
2383    this didn't happen under Solaris 8, where I normally test things.
2385    2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2386    doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2387    isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2388    this fix an hour or so after the initial 3.7 release.)
2391    Version 3.6 23-Oct-01
2392    ---------------------
2394    1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2395    offsets passed as NULL with zero offset count.
2397    2. The config.guess and config.sub files had not been updated when I moved to
2398    the latest autoconf.
2401    Version 3.5 15-Aug-01
2402    ---------------------
2404    1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2405    had been forgotten.
2407    2. By using declared but undefined structures, we can avoid using "void"
2408    definitions in pcre.h while keeping the internal definitions of the structures
2409    private.
2411    3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2412    user point of view, this means that both static and shared libraries are built
2413    by default, but this can be individually controlled. More of the work of
2414    handling this static/shared cases is now inside libtool instead of PCRE's make
2415    file.
2417    4. The pcretest utility is now installed along with pcregrep because it is
2418    useful for users (to test regexs) and by doing this, it automatically gets
2419    relinked by libtool. The documentation has been turned into a man page, so
2420    there are now .1, .txt, and .html versions in /doc.
2422    5. Upgrades to pcregrep:
2423       (i)   Added long-form option names like gnu grep.
2424       (ii)  Added --help to list all options with an explanatory phrase.
2425       (iii) Added -r, --recursive to recurse into sub-directories.
2426       (iv)  Added -f, --file to read patterns from a file.
2428    6. pcre_exec() was referring to its "code" argument before testing that
2429    argument for NULL (and giving an error if it was NULL).
2431    7. Upgraded Makefile.in to allow for compiling in a different directory from
2432    the source directory.
2434    8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2435    options bits, the pointer it was passed was to an int instead of to an unsigned
2436    long int. This mattered only on 64-bit systems.
2438    9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2439    generated) instead of pcre.in, which it its source. Also made the same change
2440    in several of the .c files.
2442    10. A new release of gcc defines printf() as a macro, which broke pcretest
2443    because it had an ifdef in the middle of a string argument for printf(). Fixed
2444    by using separate calls to printf().
2446    11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2447    script, to force use of CR or LF instead of \n in the source. On non-Unix
2448    systems, the value can be set in config.h.
2450    12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2451    absolute limit. Changed the text of the error message to make this clear, and
2452    likewise updated the man page.
2454    13. The limit of 99 on the number of capturing subpatterns has been removed.
2455    The new limit is 65535, which I hope will not be a "real" limit.
2458    Version 3.4 22-Aug-00
2459    ---------------------
2461    1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2463    2. Diagnose condition (?(0) as an error instead of crashing on matching.
2466    Version 3.3 01-Aug-00
2467    ---------------------
2469    1. If an octal character was given, but the value was greater than \377, it
2470    was not getting masked to the least significant bits, as documented. This could
2471    lead to crashes in some systems.
2473    2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
2474    the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
2476    3. Added the functions pcre_free_substring() and pcre_free_substring_list().
2477    These just pass their arguments on to (pcre_free)(), but they are provided
2478    because some uses of PCRE bind it to non-C systems that can call its functions,
2479    but cannot call free() or pcre_free() directly.
2481    4. Add "make test" as a synonym for "make check". Corrected some comments in
2482    the Makefile.
2484    5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
2485    Makefile.
2487    6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
2488    command called pgrep for grepping around the active processes.
2490    7. Added the beginnings of support for UTF-8 character strings.
2492    8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
2493    RANLIB to ./ltconfig so that they are used by libtool. I think these are all
2494    the relevant ones. (AR is not passed because ./ltconfig does its own figuring
2495    out for the ar command.)
2498    Version 3.2 12-May-00
2499    ---------------------
2501    This is purely a bug fixing release.
2503    1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
2504    of ZA. This was just one example of several cases that could provoke this bug,
2505    which was introduced by change 9 of version 2.00. The code for breaking
2506    infinite loops after an iteration that matches an empty string was't working
2507    correctly.
2509    2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
2510    when matched against abbab (for example). After matching an empty string, it
2511    wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
2512    caused it to match further down the string than it should.
2514    3. The code contained an inclusion of sys/types.h. It isn't clear why this
2515    was there because it doesn't seem to be needed, and it causes trouble on some
2516    systems, as it is not a Standard C header. It has been removed.
2518    4. Made 4 silly changes to the source to avoid stupid compiler warnings that
2519    were reported on the Macintosh. The changes were from
2521      while ((c = *(++ptr)) != 0 && c != '\n');
2522    to
2523      while ((c = *(++ptr)) != 0 && c != '\n') ;
2525    Totally extraordinary, but if that's what it takes...
2527    5. PCRE is being used in one environment where neither memmove() nor bcopy() is
2528    available. Added HAVE_BCOPY and an autoconf test for it; if neither
2529    HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
2530    assumes the way PCRE uses memmove() (always moving upwards).
2532    6. PCRE is being used in one environment where strchr() is not available. There
2533    was only one use in pcre.c, and writing it out to avoid strchr() probably gives
2534    faster code anyway.
2537    Version 3.1 09-Feb-00
2538    ---------------------
2540    The only change in this release is the fixing of some bugs in Makefile.in for
2541    the "install" target:
2543    (1) It was failing to install pcreposix.h.
2545    (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
2548    Version 3.0 01-Feb-00
2549    ---------------------
2551    1. Add support for the /+ modifier to perltest (to output $` like it does in
2552    pcretest).
2554    2. Add support for the /g modifier to perltest.
2556    3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
2557    matches null strings.
2559    4. Fix perltest so that it doesn't do unwanted things when fed an empty
2560    pattern. Perl treats empty patterns specially - it reuses the most recent
2561    pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
2562    effect.
2564    5. The POSIX interface was broken in that it was just handing over the POSIX
2565    captured string vector to pcre_exec(), but (since release 2.00) PCRE has
2566    required a bigger vector, with some working space on the end. This means that
2567    the POSIX wrapper now has to get and free some memory, and copy the results.
2569    6. Added some simple autoconf support, placing the test data and the
2570    documentation in separate directories, re-organizing some of the
2571    information files, and making it build pcre-config (a GNU standard). Also added
2572    libtool support for building PCRE as a shared library, which is now the
2573    default.
2575    7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
2576    09 are not valid octal constants. Single digits will be used for minor values
2577    less than 10.
2579    8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
2580    existing programs that set these in the POSIX interface can use PCRE without
2581    modification.
2583    9. Added a new function, pcre_fullinfo() with an extensible interface. It can
2584    return all that pcre_info() returns, plus additional data. The pcre_info()
2585    function is retained for compatibility, but is considered to be obsolete.
2587    10. Added experimental recursion feature (?R) to handle one common case that
2588    Perl 5.6 will be able to do with (?p{...}).
2590    11. Added support for POSIX character classes like [:alpha:], which Perl is
2591    adopting.
2594    Version 2.08 31-Aug-99
2595    ----------------------
2597    1. When startoffset was not zero and the pattern began with ".*", PCRE was not
2598    trying to match at the startoffset position, but instead was moving forward to
2599    the next newline as if a previous match had failed.
2601    2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
2602    and could get into a loop if a null string was matched other than at the start
2603    of the subject.
2605    3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
2606    be distinguished at compile time, and for completeness also added PCRE_DATE.
2608    5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
2609    in GnuWin32 environments.
2612    Version 2.07 29-Jul-99
2613    ----------------------
2615    1. The documentation is now supplied in plain text form and HTML as well as in
2616    the form of man page sources.
2618    2. C++ compilers don't like assigning (void *) values to other pointer types.
2619    In particular this affects malloc(). Although there is no problem in Standard
2620    C, I've put in casts to keep C++ compilers happy.
2622    3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
2623    should be (const char *).
2625    4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
2626    be useful for non-Unix systems who don't want to bother with the POSIX stuff.
2627    However, I haven't made this a standard facility. The documentation doesn't
2628    mention it, and the Makefile doesn't support it.
2630    5. The Makefile now contains an "install" target, with editable destinations at
2631    the top of the file. The pcretest program is not installed.
2633    6. pgrep -V now gives the PCRE version number and date.
2635    7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
2636    causing the entire string to be ignored, instead of just the last character.
2638    8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
2639    non-matching string, it can take a very, very long time, even for strings of
2640    quite modest length, because of the nested recursion. PCRE now does better in
2641    some of these cases. It does this by remembering the last required literal
2642    character in the pattern, and pre-searching the subject to ensure it is present
2643    before running the real match. In other words, it applies a heuristic to detect
2644    some types of certain failure quickly, and in the above example, if presented
2645    with a string that has no trailing " it gives "no match" very quickly.
2647    9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
2648    other alternatives are tried instead.
2651    Version 2.06 09-Jun-99
2652    ----------------------
2654    1. Change pcretest's output for amount of store used to show just the code
2655    space, because the remainder (the data block) varies in size between 32-bit and
2656    64-bit systems.
2658    2. Added an extra argument to pcre_exec() to supply an offset in the subject to
2659    start matching at. This allows lookbehinds to work when searching for multiple
2660    occurrences in a string.
2662    3. Added additional options to pcretest for testing multiple occurrences:
2664       /+   outputs the rest of the string that follows a match
2665       /g   loops for multiple occurrences, using the new startoffset argument
2666       /G   loops for multiple occurrences by passing an incremented pointer
2668    4. PCRE wasn't doing the "first character" optimization for patterns starting
2669    with \b or \B, though it was doing it for other lookbehind assertions. That is,
2670    it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
2671    the letter 'x'. On long subject strings, this gives a significant speed-up.
2674    Version 2.05 21-Apr-99
2675    ----------------------
2677    1. Changed the type of magic_number from int to long int so that it works
2678    properly on 16-bit systems.
2680    2. Fixed a bug which caused patterns starting with .* not to work correctly
2681    when the subject string contained newline characters. PCRE was assuming
2682    anchoring for such patterns in all cases, which is not correct because .* will
2683    not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
2684    DOTALL is set at top level; otherwise it knows that patterns starting with .*
2685    must be retried after every newline in the subject.
2688    Version 2.04 18-Feb-99
2689    ----------------------
2691    1. For parenthesized subpatterns with repeats whose minimum was zero, the
2692    computation of the store needed to hold the pattern was incorrect (too large).
2693    If such patterns were nested a few deep, this could multiply and become a real
2694    problem.
2696    2. Added /M option to pcretest to show the memory requirement of a specific
2697    pattern. Made -m a synonym of -s (which does this globally) for compatibility.
2699    3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
2700    compiled in such a way that the backtracking after subsequent failure was
2701    pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
2702    ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
2705    Version 2.03 02-Feb-99
2706    ----------------------
2708    1. Fixed typo and small mistake in man page.
2710    2. Added 4th condition (GPL supersedes if conflict) and created separate
2711    LICENCE file containing the conditions.
2713    3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
2714    Perl, that is the internal \ allows the delimiter to be included in the
2715    pattern. Locked out the use of \ as a delimiter. If \ immediately follows
2716    the final delimiter, add \ to the end of the pattern (to test the error).
2718    4. Added the convenience functions for extracting substrings after a successful
2719    match. Updated pcretest to make it able to test these functions.
2722  Version 2.02 14-Jan-99  Version 2.02 14-Jan-99
2723  ----------------------  ----------------------

Removed from v.27  
changed lines
  Added in v.286

  ViewVC Help
Powered by ViewVC 1.1.5