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

Diff of /code/trunk/ChangeLog

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

revision 49 by nigel, Sat Feb 24 21:39:33 2007 UTC revision 244 by ph10, Fri Sep 14 09:14:24 2007 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4    Version 7.4 10-Sep-07
5    ---------------------
6    
7    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
8        means that a class such as [\s] counted as "explicit reference to CR or
9        LF". That isn't really right - the whole point of the change was to try to
10        help when there was an actual mention of one of the two characters. So now
11        the change happens only if \r or \n (or a literal CR or LF) character is
12        encountered.
13    
14    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
15        of both had grown to the point where there were only 3 bits left.
16        Fortunately, there was spare space in the data structure, and so I have
17        moved the internal flags into a new 16-bit field to free up more option
18        bits.
19    
20    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
21        but did not set the internal JCHANGED flag - either of these is enough to
22        control the way the "get" function works - but the PCRE_INFO_JCHANGED
23        facility is supposed to tell if (?J) was ever used, so now (?J) at the
24        start sets both bits.
25    
26    4.  Added options (at build time, compile time, exec time) to change \R from
27        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
28    
29    5.  doc/pcresyntax.html was missing from the distribution.
30    
31    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
32        compatibility, even though it is no longer used.
33    
34    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
35        strtoull to pcrecpp.cc to select the available functions in WIN32 (where
36        different names are used).
37    
38    8.  Changed all #include <config.h> to #include "config.h". There were also
39        some further <pcre.h> cases that I changed to "pcre.h".
40    
41    9.  When pcregrep was used with the --colour option, it missed the line ending
42        sequence off the lines that it output.
43    
44    10. It was pointed out to me that arrays of string pointers cause lots of
45        relocations when a shared library is dynamically loaded. A technique of
46        using a single long string with a table of offsets can drastically reduce
47        these. I have refactored PCRE in four places to do this. The result is
48        dramatic:
49    
50          Originally:                          290
51          After changing UCP table:            187
52          After changing error message table:   43
53          After changing table of "verbs"       36
54          After changing table of Posix names   22
55    
56        Thanks to the folks working on Gregex for glib for this insight.
57    
58    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
59        unicode-properties was also set.
60    
61    
62    Version 7.3 28-Aug-07
63    ---------------------
64    
65     1. In the rejigging of the build system that eventually resulted in 7.1, the
66        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
67        brackets there is not right, since it causes compilers to look for an
68        installed pcre.h, not the version that is in the source that is being
69        compiled (which of course may be different). I have changed it back to:
70    
71          #include "pcre.h"
72    
73        I have a vague recollection that the change was concerned with compiling in
74        different directories, but in the new build system, that is taken care of
75        by the VPATH setting the Makefile.
76    
77     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
78        when the subject happened to end in the byte 0x85 (e.g. if the last
79        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
80        characters but of course it shouldn't be taken as a newline when it is part
81        of another character. The bug was that, for an unlimited repeat of . in
82        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
83        characters when looking for a newline.
84    
85     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
86    
87     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
88        in debug output.
89    
90     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
91        long printing in the pcrecpp unittest when running under MinGW.
92    
93     6. ESC_K was left out of the EBCDIC table.
94    
95     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
96        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
97        limit also applies to "virtual nesting" when a pattern is recursive, and in
98        this case 1000 isn't so big. I have been able to remove this limit at the
99        expense of backing off one optimization in certain circumstances. Normally,
100        when pcre_exec() would call its internal match() function recursively and
101        immediately return the result unconditionally, it uses a "tail recursion"
102        feature to save stack. However, when a subpattern that can match an empty
103        string has an unlimited repetition quantifier, it no longer makes this
104        optimization. That gives it a stack frame in which to save the data for
105        checking that an empty string has been matched. Previously this was taken
106        from the 1000-entry workspace that had been reserved. So now there is no
107        explicit limit, but more stack is used.
108    
109     8. Applied Daniel's patches to solve problems with the import/export magic
110        syntax that is required for Windows, and which was going wrong for the
111        pcreposix and pcrecpp parts of the library. These were overlooked when this
112        problem was solved for the main library.
113    
114     9. There were some crude static tests to avoid integer overflow when computing
115        the size of patterns that contain repeated groups with explicit upper
116        limits. As the maximum quantifier is 65535, the maximum group length was
117        set at 30,000 so that the product of these two numbers did not overflow a
118        32-bit integer. However, it turns out that people want to use groups that
119        are longer than 30,000 bytes (though not repeat them that many times).
120        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
121        made it possible to implement the integer overflow checks in a much more
122        dynamic way, which I have now done. The artificial limitation on group
123        length has been removed - we now have only the limit on the total length of
124        the compiled pattern, which depends on the LINK_SIZE setting.
125    
126    10. Fixed a bug in the documentation for get/copy named substring when
127        duplicate names are permitted. If none of the named substrings are set, the
128        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
129        empty string.
130    
131    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
132        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
133        because the ] is interpreted as the first data character and the
134        terminating ] is not found. PCRE has been made compatible with Perl in this
135        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
136        cause memory overwriting.
137    
138    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
139        string has been matched (to stop an infinite loop). It was not recognizing
140        a conditional subpattern that could match an empty string if that
141        subpattern was within another subpattern. For example, it looped when
142        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
143        condition was not nested. This bug has been fixed.
144    
145    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
146        past the start of the subject in the presence of bytes with the top bit
147        set, for example "\x8aBCD".
148    
149    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
150        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
151    
152    14. Optimized (?!) to (*FAIL).
153    
154    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
155        This restricts code points to be within the range 0 to 0x10FFFF, excluding
156        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
157        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
158        does: it's just the validity check that is more restrictive.
159    
160    16. Inserted checks for integer overflows during escape sequence (backslash)
161        processing, and also fixed erroneous offset values for syntax errors during
162        backslash processing.
163    
164    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
165        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
166    
167    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
168        caused an overrun.
169    
170    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
171        something other than just ASCII characters) inside a group that had an
172        unlimited repeat caused a loop at compile time (while checking to see
173        whether the group could match an empty string).
174    
175    20. Debugging a pattern containing \p or \P could cause a crash. For example,
176        [\P{Any}] did so. (Error in the code for printing property names.)
177    
178    21. An orphan \E inside a character class could cause a crash.
179    
180    22. A repeated capturing bracket such as (A)? could cause a wild memory
181        reference during compilation.
182    
183    23. There are several functions in pcre_compile() that scan along a compiled
184        expression for various reasons (e.g. to see if it's fixed length for look
185        behind). There were bugs in these functions when a repeated \p or \P was
186        present in the pattern. These operators have additional parameters compared
187        with \d, etc, and these were not being taken into account when moving along
188        the compiled data. Specifically:
189    
190        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
191            length.
192    
193        (b) An item such as \pL+ within a repeated group could cause crashes or
194            loops.
195    
196        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
197            "reference to non-existent subpattern" error.
198    
199        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
200    
201    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
202        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
203    
204    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
205    
206    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
207        character were causing crashes (broken optimization).
208    
209    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
210        \p or \P) caused a compile-time loop.
211    
212    28. More problems have arisen in unanchored patterns when CRLF is a valid line
213        break. For example, the unstudied pattern [\r\n]A does not match the string
214        "\r\nA" because change 7.0/46 below moves the current point on by two
215        characters after failing to match at the start. However, the pattern \nA
216        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
217        the same is true. There doesn't seem any very clean way out of this, but
218        what I have chosen to do makes the common cases work: PCRE now takes note
219        of whether there can be an explicit match for \r or \n anywhere in the
220        pattern, and if so, 7.0/46 no longer applies. As part of this change,
221        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
222        pattern has explicit CR or LF references.
223    
224    29. Added (*CR) etc for changing newline setting at start of pattern.
225    
226    
227    Version 7.2 19-Jun-07
228    ---------------------
229    
230     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
231        which is apparently normally available under Windows.
232    
233     2. Re-jig the pcregrep tests with different newline settings in an attempt
234        to make them independent of the local environment's newline setting.
235    
236     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
237    
238     4. Some of the "internals" tests were previously cut out when the link size
239        was not 2, because the output contained actual offsets. The recent new
240        "Z" feature of pcretest means that these can be cut out, making the tests
241        usable with all link sizes.
242    
243     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
244        stack recursion. This gives a massive performance boost under BSD, but just
245        a small improvement under Linux. However, it saves one field in the frame
246        in all cases.
247    
248     6. Added more features from the forthcoming Perl 5.10:
249    
250        (a) (?-n) (where n is a string of digits) is a relative subroutine or
251            recursion call. It refers to the nth most recently opened parentheses.
252    
253        (b) (?+n) is also a relative subroutine call; it refers to the nth next
254            to be opened parentheses.
255    
256        (c) Conditions that refer to capturing parentheses can be specified
257            relatively, for example, (?(-2)... or (?(+3)...
258    
259        (d) \K resets the start of the current match so that everything before
260            is not part of it.
261    
262        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
263    
264        (f) \g{name} is another synonym - part of Perl 5.10's unification of
265            reference syntax.
266    
267        (g) (?| introduces a group in which the numbering of parentheses in each
268            alternative starts with the same number.
269    
270        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
271    
272     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
273        PCRE_INFO_JCHANGED.
274    
275     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
276        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
277        for detecting groups that can match an empty string.
278    
279     9. A pattern with a very large number of alternatives (more than several
280        hundred) was running out of internal workspace during the pre-compile
281        phase, where pcre_compile() figures out how much memory will be needed. A
282        bit of new cunning has reduced the workspace needed for groups with
283        alternatives. The 1000-alternative test pattern now uses 12 bytes of
284        workspace instead of running out of the 4096 that are available.
285    
286    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
287    
288    11. Applied patch from Google to remove an optimization that didn't quite work.
289        The report of the bug said:
290    
291          pcrecpp::RE("a*").FullMatch("aaa") matches, while
292          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
293          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
294    
295    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
296        it matched the wrong number of bytes.
297    
298    
299    Version 7.1 24-Apr-07
300    ---------------------
301    
302     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
303        that is more "standard", making use of automake and other Autotools. There
304        is some re-arrangement of the files and adjustment of comments consequent
305        on this.
306    
307     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
308        for recursive directory scanning broke on some systems because the files
309        are not scanned in any specific order and on different systems the order
310        was different. A call to "sort" has been inserted into RunGrepTest for the
311        approprate test as a short-term fix. In the longer term there may be an
312        alternative.
313    
314     3. I had an email from Eric Raymond about problems translating some of PCRE's
315        man pages to HTML (despite the fact that I distribute HTML pages, some
316        people do their own conversions for various reasons). The problems
317        concerned the use of low-level troff macros .br and .in. I have therefore
318        removed all such uses from the man pages (some were redundant, some could
319        be replaced by .nf/.fi pairs). The 132html script that I use to generate
320        HTML has been updated to handle .nf/.fi and to complain if it encounters
321        .br or .in.
322    
323     4. Updated comments in configure.ac that get placed in config.h.in and also
324        arranged for config.h to be included in the distribution, with the name
325        config.h.generic, for the benefit of those who have to compile without
326        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
327    
328     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
329        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
330        makevp.bat for the current PCRE, using the additional files
331        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
332    
333     6. A Windows user reported a minor discrepancy with test 2, which turned out
334        to be caused by a trailing space on an input line that had got lost in his
335        copy. The trailing space was an accident, so I've just removed it.
336    
337     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
338        that is needed.
339    
340     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
341        as "const" (a) because they are and (b) because it helps the PHP
342        maintainers who have recently made a script to detect big data structures
343        in the php code that should be moved to the .rodata section. I remembered
344        to update Builducptable as well, so it won't revert if ucptable.h is ever
345        re-created.
346    
347     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
348        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
349        order to be able to cut out the UTF-8 tables in the latter when UTF-8
350        support is not required. This saves 1.5-2K of code, which is important in
351        some applications.
352    
353        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
354        so as not to refer to the tables, even though these functions will never be
355        called when UTF-8 support is disabled. Otherwise there are problems with a
356        shared library.
357    
358    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
359    
360        (a) It was defining its arguments as char * instead of void *.
361    
362        (b) It was assuming that all moves were upwards in memory; this was true
363            a long time ago when I wrote it, but is no longer the case.
364    
365        The emulated memove() is provided for those environments that have neither
366        memmove() nor bcopy(). I didn't think anyone used it these days, but that
367        is clearly not the case, as these two bugs were recently reported.
368    
369    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
370        and Detrail to create the HTML documentation, the .txt form of the man
371        pages, and it removes trailing spaces from listed files. It also creates
372        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
373        case, it wraps all the #defines with #ifndefs. This script should be run
374        before "make dist".
375    
376    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
377        with Unicode property support.
378    
379        (a) For a maximizing quantifier, if the two different cases of the
380            character were of different lengths in their UTF-8 codings (there are
381            some cases like this - I found 11), and the matching function had to
382            back up over a mixture of the two cases, it incorrectly assumed they
383            were both the same length.
384    
385        (b) When PCRE was configured to use the heap rather than the stack for
386            recursion during matching, it was not correctly preserving the data for
387            the other case of a UTF-8 character when checking ahead for a match
388            while processing a minimizing repeat. If the check also involved
389            matching a wide character, but failed, corruption could cause an
390            erroneous result when trying to check for a repeat of the original
391            character.
392    
393    13. Some tidying changes to the testing mechanism:
394    
395        (a) The RunTest script now detects the internal link size and whether there
396            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
397            values substituted by "configure". (The RunGrepTest script already did
398            this for UTF-8.) The configure.ac script no longer substitutes the
399            relevant variables.
400    
401        (b) The debugging options /B and /D in pcretest show the compiled bytecode
402            with length and offset values. This means that the output is different
403            for different internal link sizes. Test 2 is skipped for link sizes
404            other than 2 because of this, bypassing the problem. Unfortunately,
405            there was also a test in test 3 (the locale tests) that used /B and
406            failed for link sizes other than 2. Rather than cut the whole test out,
407            I have added a new /Z option to pcretest that replaces the length and
408            offset values with spaces. This is now used to make test 3 independent
409            of link size. (Test 2 will be tidied up later.)
410    
411    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
412        segmentation fault instead of returning the appropriate error message.
413    
414    15. In multiline mode when the newline sequence was set to "any", the pattern
415        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
416        This doesn't seem right; it now treats the CRLF combination as the line
417        ending, and so does not match in that case. It's only a pattern such as ^$
418        that would hit this one: something like ^ABC$ would have failed after \r
419        and then tried again after \r\n.
420    
421    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
422        in an attempt to make files that differ only in their line terminators
423        compare equal. This works on Linux.
424    
425    17. Under certain error circumstances pcregrep might try to free random memory
426        as it exited. This is now fixed, thanks to valgrind.
427    
428    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
429        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
430        was because its rules for how to advance for /g after matching an empty
431        string at the end of a line did not allow for this case. They now check for
432        it specially.
433    
434    20. pcretest is supposed to handle patterns and data of any length, by
435        extending its buffers when necessary. It was getting this wrong when the
436        buffer for a data line had to be extended.
437    
438    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
439        CRLF as a newline sequence.
440    
441    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
442        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
443        I have nevertheless tidied it up.
444    
445    23. Added some casts to kill warnings from HP-UX ia64 compiler.
446    
447    24. Added a man page for pcre-config.
448    
449    
450    Version 7.0 19-Dec-06
451    ---------------------
452    
453     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
454        moving to gcc 4.1.1.
455    
456     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
457        sys/time.h, which is documented as needed for this function. It doesn't
458        seem to matter on Linux, but it showed up on some releases of OS X.
459    
460     3. It seems that there are systems where bytes whose values are greater than
461        127 match isprint() in the "C" locale. The "C" locale should be the
462        default when a C program starts up. In most systems, only ASCII printing
463        characters match isprint(). This difference caused the output from pcretest
464        to vary, making some of the tests fail. I have changed pcretest so that:
465    
466        (a) When it is outputting text in the compiled version of a pattern, bytes
467            other than 32-126 are always shown as hex escapes.
468    
469        (b) When it is outputting text that is a matched part of a subject string,
470            it does the same, unless a different locale has been set for the match
471            (using the /L modifier). In this case, it uses isprint() to decide.
472    
473     4. Fixed a major bug that caused incorrect computation of the amount of memory
474        required for a compiled pattern when options that changed within the
475        pattern affected the logic of the preliminary scan that determines the
476        length. The relevant options are -x, and -i in UTF-8 mode. The result was
477        that the computed length was too small. The symptoms of this bug were
478        either the PCRE error "internal error: code overflow" from pcre_compile(),
479        or a glibc crash with a message such as "pcretest: free(): invalid next
480        size (fast)". Examples of patterns that provoked this bug (shown in
481        pcretest format) are:
482    
483          /(?-x: )/x
484          /(?x)(?-x: \s*#\s*)/
485          /((?i)[\x{c0}])/8
486          /(?i:[\x{c0}])/8
487    
488        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
489        is now done differently.
490    
491     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
492        wrapper classes; (b) implement a new function in the C++ scanner that is
493        more efficient than the old way of doing things because it avoids levels of
494        recursion in the regex matching; (c) add a paragraph to the documentation
495        for the FullMatch() function.
496    
497     6. The escape sequence \n was being treated as whatever was defined as
498        "newline". Not only was this contrary to the documentation, which states
499        that \n is character 10 (hex 0A), but it also went horribly wrong when
500        "newline" was defined as CRLF. This has been fixed.
501    
502     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
503        was being set to -1 for the "end of line" case (supposedly a value that no
504        character can have). Though this value is never used (the check for end of
505        line is "zero bytes in current character"), it caused compiler complaints.
506        I've changed it to 0xffffffff.
507    
508     8. In pcre_version.c, the version string was being built by a sequence of
509        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
510        string (as it is for production releases) called a macro with an empty
511        argument. The C standard says the result of this is undefined. The gcc
512        compiler treats it as an empty string (which was what was wanted) but it is
513        reported that Visual C gives an error. The source has been hacked around to
514        avoid this problem.
515    
516     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
517        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
518        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
519        of them did).
520    
521    10. Originally, pcretest opened its input and output without "b"; then I was
522        told that "b" was needed in some environments, so it was added for release
523        5.0 to both the input and output. (It makes no difference on Unix-like
524        systems.) Later I was told that it is wrong for the input on Windows. I've
525        now abstracted the modes into two macros, to make it easier to fiddle with
526        them, and removed "b" from the input mode under Windows.
527    
528    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
529    
530    12. Added -help and --help to pcretest as an official way of being reminded
531        of the options.
532    
533    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
534        and pcrecpp.cc because they annoy compilers at high warning levels.
535    
536    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
537    
538    15. Fixed an occurrence of == in configure.ac that should have been = (shell
539        scripts are not C programs :-) and which was not noticed because it works
540        on Linux.
541    
542    16. pcretest is supposed to handle any length of pattern and data line (as one
543        line or as a continued sequence of lines) by extending its input buffer if
544        necessary. This feature was broken for very long pattern lines, leading to
545        a string of junk being passed to pcre_compile() if the pattern was longer
546        than about 50K.
547    
548    17. I have done a major re-factoring of the way pcre_compile() computes the
549        amount of memory needed for a compiled pattern. Previously, there was code
550        that made a preliminary scan of the pattern in order to do this. That was
551        OK when PCRE was new, but as the facilities have expanded, it has become
552        harder and harder to keep it in step with the real compile phase, and there
553        have been a number of bugs (see for example, 4 above). I have now found a
554        cunning way of running the real compile function in a "fake" mode that
555        enables it to compute how much memory it would need, while actually only
556        ever using a few hundred bytes of working memory and without too many
557        tests of the mode. This should make future maintenance and development
558        easier. A side effect of this work is that the limit of 200 on the nesting
559        depth of parentheses has been removed (though this was never a serious
560        limitation, I suspect). However, there is a downside: pcre_compile() now
561        runs more slowly than before (30% or more, depending on the pattern). I
562        hope this isn't a big issue. There is no effect on runtime performance.
563    
564    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
565        newline (only possible for the last line of a file) and it was a
566        pattern that set a locale (followed by /Lsomething), pcretest crashed.
567    
568    19. Added additional timing features to pcretest. (1) The -tm option now times
569        matching only, not compiling. (2) Both -t and -tm can be followed, as a
570        separate command line item, by a number that specifies the number of
571        repeats to use when timing. The default is 50000; this gives better
572        precision, but takes uncomfortably long for very large patterns.
573    
574    20. Extended pcre_study() to be more clever in cases where a branch of a
575        subpattern has no definite first character. For example, (a*|b*)[cd] would
576        previously give no result from pcre_study(). Now it recognizes that the
577        first character must be a, b, c, or d.
578    
579    21. There was an incorrect error "recursive call could loop indefinitely" if
580        a subpattern (or the entire pattern) that was being tested for matching an
581        empty string contained only one non-empty item after a nested subpattern.
582        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
583        incorrectly, because the \d was being skipped in the check.
584    
585    22. The pcretest program now has a new pattern option /B and a command line
586        option -b, which is equivalent to adding /B to every pattern. This causes
587        it to show the compiled bytecode, without the additional information that
588        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
589        is the same as /B/I).
590    
591    23. A new optimization is now able automatically to treat some sequences such
592        as a*b as a*+b. More specifically, if something simple (such as a character
593        or a simple class like \d) has an unlimited quantifier, and is followed by
594        something that cannot possibly match the quantified thing, the quantifier
595        is automatically "possessified".
596    
597    24. A recursive reference to a subpattern whose number was greater than 39
598        went wrong under certain circumstances in UTF-8 mode. This bug could also
599        have affected the operation of pcre_study().
600    
601    25. Realized that a little bit of performance could be had by replacing
602        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
603    
604    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
605    
606    27. Possessive quantifiers such as a++ were previously implemented by turning
607        them into atomic groups such as ($>a+). Now they have their own opcodes,
608        which improves performance. This includes the automatically created ones
609        from 23 above.
610    
611    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
612        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
613        the first matched character to be a colon. This applied both to named and
614        numbered groups.
615    
616    29. The ucpinternal.h header file was missing its idempotency #ifdef.
617    
618    30. I was sent a "project" file called libpcre.a.dev which I understand makes
619        building PCRE on Windows easier, so I have included it in the distribution.
620    
621    31. There is now a check in pcretest against a ridiculously large number being
622        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
623        loop, the loop is abandoned.
624    
625    32. Forward references to subpatterns in conditions such as (?(2)...) where
626        subpattern 2 is defined later cause pcre_compile() to search forwards in
627        the pattern for the relevant set of parentheses. This search went wrong
628        when there were unescaped parentheses in a character class, parentheses
629        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
630    
631    33. "Subroutine" calls and backreferences were previously restricted to
632        referencing subpatterns earlier in the regex. This restriction has now
633        been removed.
634    
635    34. Added a number of extra features that are going to be in Perl 5.10. On the
636        whole, these are just syntactic alternatives for features that PCRE had
637        previously implemented using the Python syntax or my own invention. The
638        other formats are all retained for compatibility.
639    
640        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
641            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
642            also .NET compatible.
643    
644        (b) A recursion or subroutine call to a named group can now be defined as
645            (?&name) as well as (?P>name).
646    
647        (c) A backreference to a named group can now be defined as \k<name> or
648            \k'name' as well as (?P=name). The new forms, as well as being in Perl
649            5.10, are also .NET compatible.
650    
651        (d) A conditional reference to a named group can now use the syntax
652            (?(<name>) or (?('name') as well as (?(name).
653    
654        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
655            groups (named and numbered) that are never evaluated inline, but can be
656            called as "subroutines" from elsewhere. In effect, the DEFINE condition
657            is always false. There may be only one alternative in such a group.
658    
659        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
660            as the simple (?(R). The condition is true only if the most recent
661            recursion is that of the given number or name. It does not search out
662            through the entire recursion stack.
663    
664        (g) The escape \gN or \g{N} has been added, where N is a positive or
665            negative number, specifying an absolute or relative reference.
666    
667    35. Tidied to get rid of some further signed/unsigned compiler warnings and
668        some "unreachable code" warnings.
669    
670    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
671        things, this adds five new scripts.
672    
673    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
674        There were also incompatibilities regarding the handling of \Q..\E inside
675        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
676        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
677    
678    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
679        matches an empty string, and forcibly breaks the loop. There were bugs in
680        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
681        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
682        separate and independent bugs (that affected different cases) have been
683        fixed.
684    
685    39. Refactored the code to abolish the use of different opcodes for small
686        capturing bracket numbers. This is a tidy that I avoided doing when I
687        removed the limit on the number of capturing brackets for 3.5 back in 2001.
688        The new approach is not only tidier, it makes it possible to reduce the
689        memory needed to fix the previous bug (38).
690    
691    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
692        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
693        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
694        mode.
695    
696    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
697        report.
698    
699    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
700        copy construction and assignment in the C++ wrapper.
701    
702    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
703        couple of bugs that could have given wrong results in the "--newline=crlf"
704        case.
705    
706    44. Added a number of casts and did some reorganization of signed/unsigned int
707        variables following suggestions from Dair Grant. Also renamed the variable
708        "this" as "item" because it is a C++ keyword.
709    
710    45. Arranged for dftables to add
711    
712          #include "pcre_internal.h"
713    
714        to pcre_chartables.c because without it, gcc 4.x may remove the array
715        definition from the final binary if PCRE is built into a static library and
716        dead code stripping is activated.
717    
718    46. For an unanchored pattern, if a match attempt fails at the start of a
719        newline sequence, and the newline setting is CRLF or ANY, and the next two
720        characters are CRLF, advance by two characters instead of one.
721    
722    
723    Version 6.7 04-Jul-06
724    ---------------------
725    
726     1. In order to handle tests when input lines are enormously long, pcretest has
727        been re-factored so that it automatically extends its buffers when
728        necessary. The code is crude, but this _is_ just a test program. The
729        default size has been increased from 32K to 50K.
730    
731     2. The code in pcre_study() was using the value of the re argument before
732        testing it for NULL. (Of course, in any sensible call of the function, it
733        won't be NULL.)
734    
735     3. The memmove() emulation function in pcre_internal.h, which is used on
736        systems that lack both memmove() and bcopy() - that is, hardly ever -
737        was missing a "static" storage class specifier.
738    
739     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
740        containing an extended class (one that cannot be represented by a bitmap
741        because it contains high-valued characters or Unicode property items, e.g.
742        [\pZ]). Almost always one would set UTF-8 mode when processing such a
743        pattern, but PCRE should not loop if you do not (it no longer does).
744        [Detail: two cases were found: (a) a repeated subpattern containing an
745        extended class; (b) a recursive reference to a subpattern that followed a
746        previous extended class. It wasn't skipping over the extended class
747        correctly when UTF-8 mode was not set.]
748    
749     5. A negated single-character class was not being recognized as fixed-length
750        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
751        compile error "lookbehind assertion is not fixed length".
752    
753     6. The RunPerlTest auxiliary script was showing an unexpected difference
754        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
755        write a Perl script that can interpret lines of an input file either as
756        byte characters or as UTF-8, which is what "perltest" was being required to
757        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
758        can't do is switch easily at run time between having the "use utf8;" pragma
759        or not. In the end, I fudged it by using the RunPerlTest script to insert
760        "use utf8;" explicitly for the UTF-8 tests.
761    
762     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
763        the end of the subject string, contrary to the documentation and to what
764        Perl does. This was true of both matching functions. Now it matches only at
765        the start of the subject and immediately after *internal* newlines.
766    
767     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
768        a pointer to an int instead of a pointer to an unsigned long int. This
769        caused problems on 64-bit systems.
770    
771     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
772        instance of the 'standard' template library not being so standard".
773    
774    10. There was no check on the number of named subpatterns nor the maximum
775        length of a subpattern name. The product of these values is used to compute
776        the size of the memory block for a compiled pattern. By supplying a very
777        long subpattern name and a large number of named subpatterns, the size
778        computation could be caused to overflow. This is now prevented by limiting
779        the length of names to 32 characters, and the number of named subpatterns
780        to 10,000.
781    
782    11. Subpatterns that are repeated with specific counts have to be replicated in
783        the compiled pattern. The size of memory for this was computed from the
784        length of the subpattern and the repeat count. The latter is limited to
785        65535, but there was no limit on the former, meaning that integer overflow
786        could in principle occur. The compiled length of a repeated subpattern is
787        now limited to 30,000 bytes in order to prevent this.
788    
789    12. Added the optional facility to have named substrings with the same name.
790    
791    13. Added the ability to use a named substring as a condition, using the
792        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
793        are numbers (not recommended). Forward references are permitted.
794    
795    14. Added forward references in named backreferences (if you see what I mean).
796    
797    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
798        pattern could run off the end of the subject. For example, the pattern
799        "(?s)(.{1,5})"8 did this with the subject "ab".
800    
801    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
802        PCRE_CASELESS was set when matching characters that were quantified with ?
803        or *.
804    
805    17. A character class other than a single negated character that had a minimum
806        but no maximum quantifier - for example [ab]{6,} - was not handled
807        correctly by pce_dfa_exec(). It would match only one character.
808    
809    18. A valid (though odd) pattern that looked like a POSIX character
810        class but used an invalid character after [ (for example [[,abc,]]) caused
811        pcre_compile() to give the error "Failed: internal error: code overflow" or
812        in some cases to crash with a glibc free() error. This could even happen if
813        the pattern terminated after [[ but there just happened to be a sequence of
814        letters, a binary zero, and a closing ] in the memory that followed.
815    
816    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
817        over the years. Originally (before any Unicode support), just the bottom 8
818        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
819        output from "man perlunicode" includes this:
820    
821          The regular expression compiler produces polymorphic opcodes.  That
822          is, the pattern adapts to the data and automatically switches to
823          the Unicode character scheme when presented with Unicode data--or
824          instead uses a traditional byte scheme when presented with byte
825          data.
826    
827        Sadly, a wide octal escape does not cause a switch, and in a string with
828        no other multibyte characters, these octal escapes are treated as before.
829        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
830        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
831        Unicode string.
832    
833        I have not perpetrated such confusion in PCRE. Up till now, it took just
834        the bottom 8 bits, as in old Perl. I have now made octal escapes with
835        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
836        translate to the appropriate multibyte character.
837    
838    29. Applied some refactoring to reduce the number of warnings from Microsoft
839        and Borland compilers. This has included removing the fudge introduced
840        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
841        a warning about an unused variable.
842    
843    21. PCRE has not included VT (character 0x0b) in the set of whitespace
844        characters since release 4.0, because Perl (from release 5.004) does not.
845        [Or at least, is documented not to: some releases seem to be in conflict
846        with the documentation.] However, when a pattern was studied with
847        pcre_study() and all its branches started with \s, PCRE still included VT
848        as a possible starting character. Of course, this did no harm; it just
849        caused an unnecessary match attempt.
850    
851    22. Removed a now-redundant internal flag bit that recorded the fact that case
852        dependency changed within the pattern. This was once needed for "required
853        byte" processing, but is no longer used. This recovers a now-scarce options
854        bit. Also moved the least significant internal flag bit to the most-
855        significant bit of the word, which was not previously used (hangover from
856        the days when it was an int rather than a uint) to free up another bit for
857        the future.
858    
859    23. Added support for CRLF line endings as well as CR and LF. As well as the
860        default being selectable at build time, it can now be changed at runtime
861        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
862        specify that it is scanning data with non-default line endings.
863    
864    24. Changed the definition of CXXLINK to make it agree with the definition of
865        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
866    
867    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
868        recursions. This makes a big different to stack usage for some patterns.
869    
870    26. If a subpattern containing a named recursion or subroutine reference such
871        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
872        the space required for the compiled pattern went wrong and gave too small a
873        value. Depending on the environment, this could lead to "Failed: internal
874        error: code overflow at offset 49" or "glibc detected double free or
875        corruption" errors.
876    
877    27. Applied patches from Google (a) to support the new newline modes and (b) to
878        advance over multibyte UTF-8 characters in GlobalReplace.
879    
880    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
881        difference for some implementation of PCRE in some Windows version.
882    
883    29. Added some extra testing facilities to pcretest:
884    
885        \q<number>   in a data line sets the "match limit" value
886        \Q<number>   in a data line sets the "match recursion limt" value
887        -S <number>  sets the stack size, where <number> is in megabytes
888    
889        The -S option isn't available for Windows.
890    
891    
892    Version 6.6 06-Feb-06
893    ---------------------
894    
895     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
896        in pcreposix.h. I have copied the definition from pcre.h.
897    
898     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
899        because pcre.h is no longer a built file.
900    
901     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
902        not normally included in the compiled code.
903    
904    
905    Version 6.5 01-Feb-06
906    ---------------------
907    
908     1. When using the partial match feature with pcre_dfa_exec(), it was not
909        anchoring the second and subsequent partial matches at the new starting
910        point. This could lead to incorrect results. For example, with the pattern
911        /1234/, partially matching against "123" and then "a4" gave a match.
912    
913     2. Changes to pcregrep:
914    
915        (a) All non-match returns from pcre_exec() were being treated as failures
916            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
917            error message is output. Some extra information is given for the
918            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
919            probably the only errors that are likely to be caused by users (by
920            specifying a regex that has nested indefinite repeats, for instance).
921            If there are more than 20 of these errors, pcregrep is abandoned.
922    
923        (b) A binary zero was treated as data while matching, but terminated the
924            output line if it was written out. This has been fixed: binary zeroes
925            are now no different to any other data bytes.
926    
927        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
928            used to set a locale for matching. The --locale=xxxx long option has
929            been added (no short equivalent) to specify a locale explicitly on the
930            pcregrep command, overriding the environment variables.
931    
932        (d) When -B was used with -n, some line numbers in the output were one less
933            than they should have been.
934    
935        (e) Added the -o (--only-matching) option.
936    
937        (f) If -A or -C was used with -c (count only), some lines of context were
938            accidentally printed for the final match.
939    
940        (g) Added the -H (--with-filename) option.
941    
942        (h) The combination of options -rh failed to suppress file names for files
943            that were found from directory arguments.
944    
945        (i) Added the -D (--devices) and -d (--directories) options.
946    
947        (j) Added the -F (--fixed-strings) option.
948    
949        (k) Allow "-" to be used as a file name for -f as well as for a data file.
950    
951        (l) Added the --colo(u)r option.
952    
953        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
954            is not present by default.
955    
956     3. A nasty bug was discovered in the handling of recursive patterns, that is,
957        items such as (?R) or (?1), when the recursion could match a number of
958        alternatives. If it matched one of the alternatives, but subsequently,
959        outside the recursion, there was a failure, the code tried to back up into
960        the recursion. However, because of the way PCRE is implemented, this is not
961        possible, and the result was an incorrect result from the match.
962    
963        In order to prevent this happening, the specification of recursion has
964        been changed so that all such subpatterns are automatically treated as
965        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
966    
967     4. I had overlooked the fact that, in some locales, there are characters for
968        which isalpha() is true but neither isupper() nor islower() are true. In
969        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
970        and ordfeminine) are like this. This affected the treatment of \w and \W
971        when they appeared in character classes, but not when they appeared outside
972        a character class. The bit map for "word" characters is now created
973        separately from the results of isalnum() instead of just taking it from the
974        upper, lower, and digit maps. (Plus the underscore character, of course.)
975    
976     5. The above bug also affected the handling of POSIX character classes such as
977        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
978        permanent tables. Instead, the bit maps for such a class were previously
979        created as the appropriate unions of the upper, lower, and digit bitmaps.
980        Now they are created by subtraction from the [[:word:]] class, which has
981        its own bitmap.
982    
983     6. The [[:blank:]] character class matches horizontal, but not vertical space.
984        It is created by subtracting the vertical space characters (\x09, \x0a,
985        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
986        subtraction was done in the overall bitmap for a character class, meaning
987        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
988        be recognized. This bug has been fixed.
989    
990     7. Patches from the folks at Google:
991    
992          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
993          real life, but is still worth protecting against".
994    
995          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
996          regular expressions".
997    
998          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
999          have it.
1000    
1001          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1002          "configure" and the latter not, in order to fix a problem somebody had
1003          with compiling the Arg class on HP-UX.
1004    
1005          (e) Improve the error-handling of the C++ wrapper a little bit.
1006    
1007          (f) New tests for checking recursion limiting.
1008    
1009     8. The pcre_memmove() function, which is used only if the environment does not
1010        have a standard memmove() function (and is therefore rarely compiled),
1011        contained two bugs: (a) use of int instead of size_t, and (b) it was not
1012        returning a result (though PCRE never actually uses the result).
1013    
1014     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1015        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1016        returned instead of calling malloc() with an overflowing number that would
1017        most likely cause subsequent chaos.
1018    
1019    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1020    
1021    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1022        with this option is matched, the nmatch and pmatch options of regexec() are
1023        ignored.
1024    
1025    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1026        provided in case anyone wants to the the POSIX interface with UTF-8
1027        strings.
1028    
1029    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1030        C++ linking (needed for some HP-UX environments).
1031    
1032    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1033        (unused parameter) and in the pcre_printint() function (omitted "default"
1034        switch label when the default is to do nothing).
1035    
1036    15. Added some code to make it possible, when PCRE is compiled as a C++
1037        library, to replace subject pointers for pcre_exec() with a smart pointer
1038        class, thus making it possible to process discontinuous strings.
1039    
1040    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1041        much the same function. They were added by different people who were trying
1042        to make PCRE easy to compile on non-Unix systems. It has been suggested
1043        that PCRE_EXPORT be abolished now that there is more automatic apparatus
1044        for compiling on Windows systems. I have therefore replaced it with
1045        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1046        defaults to "extern" for C or "extern C" for C++, which works fine on
1047        Unix-like systems. It is now possible to override the value of PCRE_DATA_
1048        SCOPE with something explicit in config.h. In addition:
1049    
1050        (a) pcreposix.h still had just "extern" instead of either of these macros;
1051            I have replaced it with PCRE_DATA_SCOPE.
1052    
1053        (b) Functions such as _pcre_xclass(), which are internal to the library,
1054            but external in the C sense, all had PCRE_EXPORT in their definitions.
1055            This is apparently wrong for the Windows case, so I have removed it.
1056            (It makes no difference on Unix-like systems.)
1057    
1058    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1059        of recursive calls to match(). This is different to MATCH_LIMIT because
1060        that limits the total number of calls to match(), not all of which increase
1061        the depth of recursion. Limiting the recursion depth limits the amount of
1062        stack (or heap if NO_RECURSE is set) that is used. The default can be set
1063        when PCRE is compiled, and changed at run time. A patch from Google adds
1064        this functionality to the C++ interface.
1065    
1066    18. Changes to the handling of Unicode character properties:
1067    
1068        (a) Updated the table to Unicode 4.1.0.
1069    
1070        (b) Recognize characters that are not in the table as "Cn" (undefined).
1071    
1072        (c) I revised the way the table is implemented to a much improved format
1073            which includes recognition of ranges. It now supports the ranges that
1074            are defined in UnicodeData.txt, and it also amalgamates other
1075            characters into ranges. This has reduced the number of entries in the
1076            table from around 16,000 to around 3,000, thus reducing its size
1077            considerably. I realized I did not need to use a tree structure after
1078            all - a binary chop search is just as efficient. Having reduced the
1079            number of entries, I extended their size from 6 bytes to 8 bytes to
1080            allow for more data.
1081    
1082        (d) Added support for Unicode script names via properties such as \p{Han}.
1083    
1084    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1085        matching that character.
1086    
1087    20. When matching a repeated Unicode property with a minimum greater than zero,
1088        (for example \pL{2,}), PCRE could look past the end of the subject if it
1089        reached it while seeking the minimum number of characters. This could
1090        happen only if some of the characters were more than one byte long, because
1091        there is a check for at least the minimum number of bytes.
1092    
1093    21. Refactored the implementation of \p and \P so as to be more general, to
1094        allow for more different types of property in future. This has changed the
1095        compiled form incompatibly. Anybody with saved compiled patterns that use
1096        \p or \P will have to recompile them.
1097    
1098    22. Added "Any" and "L&" to the supported property types.
1099    
1100    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1101        but give a compile time error if the value is greater than 0xff.
1102    
1103    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1104        accidentally not being installed or uninstalled.
1105    
1106    25. The pcre.h file was built from pcre.h.in, but the only changes that were
1107        made were to insert the current release number. This seemed silly, because
1108        it made things harder for people building PCRE on systems that don't run
1109        "configure". I have turned pcre.h into a distributed file, no longer built
1110        by "configure", with the version identification directly included. There is
1111        no longer a pcre.h.in file.
1112    
1113        However, this change necessitated a change to the pcre-config script as
1114        well. It is built from pcre-config.in, and one of the substitutions was the
1115        release number. I have updated configure.ac so that ./configure now finds
1116        the release number by grepping pcre.h.
1117    
1118    26. Added the ability to run the tests under valgrind.
1119    
1120    
1121    Version 6.4 05-Sep-05
1122    ---------------------
1123    
1124     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1125        "--" to be printed when multiple files were scanned, even when none of the
1126        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1127        consider it to be a bug, and have restored the previous behaviour.
1128    
1129     2. A couple of code tidies to get rid of compiler warnings.
1130    
1131     3. The pcretest program used to cheat by referring to symbols in the library
1132        whose names begin with _pcre_. These are internal symbols that are not
1133        really supposed to be visible externally, and in some environments it is
1134        possible to suppress them. The cheating is now confined to including
1135        certain files from the library's source, which is a bit cleaner.
1136    
1137     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1138        file's purpose clearer.
1139    
1140     5. Reorganized pcre_ucp_findchar().
1141    
1142    
1143    Version 6.3 15-Aug-05
1144    ---------------------
1145    
1146     1. The file libpcre.pc.in did not have general read permission in the tarball.
1147    
1148     2. There were some problems when building without C++ support:
1149    
1150        (a) If C++ support was not built, "make install" and "make test" still
1151            tried to test it.
1152    
1153        (b) There were problems when the value of CXX was explicitly set. Some
1154            changes have been made to try to fix these, and ...
1155    
1156        (c) --disable-cpp can now be used to explicitly disable C++ support.
1157    
1158        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1159            backslash in a target when C++ was disabled. This confuses some
1160            versions of "make", apparently. Using an intermediate variable solves
1161            this. (Same for CPP_LOBJ.)
1162    
1163     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1164        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1165        necessary on certain architectures.
1166    
1167     4. Added a setting of -export-symbols-regex to the link command to remove
1168        those symbols that are exported in the C sense, but actually are local
1169        within the library, and not documented. Their names all begin with
1170        "_pcre_". This is not a perfect job, because (a) we have to except some
1171        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1172        available (and never for static libraries). I have made a note to try to
1173        find a way round (a) in the future.
1174    
1175    
1176    Version 6.2 01-Aug-05
1177    ---------------------
1178    
1179     1. There was no test for integer overflow of quantifier values. A construction
1180        such as {1111111111111111} would give undefined results. What is worse, if
1181        a minimum quantifier for a parenthesized subpattern overflowed and became
1182        negative, the calculation of the memory size went wrong. This could have
1183        led to memory overwriting.
1184    
1185     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1186    
1187     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1188        operating environments where this matters.
1189    
1190     4. Applied Giuseppe Maxia's patch to add additional features for controlling
1191        PCRE options from within the C++ wrapper.
1192    
1193     5. Named capturing subpatterns were not being correctly counted when a pattern
1194        was compiled. This caused two problems: (a) If there were more than 100
1195        such subpatterns, the calculation of the memory needed for the whole
1196        compiled pattern went wrong, leading to an overflow error. (b) Numerical
1197        back references of the form \12, where the number was greater than 9, were
1198        not recognized as back references, even though there were sufficient
1199        previous subpatterns.
1200    
1201     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1202        versions of gcc, e.g. 2.95.4.
1203    
1204    
1205    Version 6.1 21-Jun-05
1206    ---------------------
1207    
1208     1. There was one reference to the variable "posix" in pcretest.c that was not
1209        surrounded by "#if !defined NOPOSIX".
1210    
1211     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1212        the cross-check on the old pcre_info() function, for the benefit of the
1213        cut-down version of PCRE that is currently imported into Exim.
1214    
1215     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1216        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1217        patterns that start (?i) but I don't think that matters. The use of (?i) is
1218        just an example; this all applies to the other options as well.
1219    
1220     4. Since libtool seems to echo the compile commands it is issuing, the output
1221        from "make" can be reduced a bit by putting "@" in front of each libtool
1222        compile command.
1223    
1224     5. Patch from the folks at Google for configure.in to be a bit more thorough
1225        in checking for a suitable C++ installation before trying to compile the
1226        C++ stuff. This should fix a reported problem when a compiler was present,
1227        but no suitable headers.
1228    
1229     6. The man pages all had just "PCRE" as their title. I have changed them to
1230        be the relevant file name. I have also arranged that these names are
1231        retained in the file doc/pcre.txt, which is a concatenation in text format
1232        of all the man pages except the little individual ones for each function.
1233    
1234     7. The NON-UNIX-USE file had not been updated for the different set of source
1235        files that come with release 6. I also added a few comments about the C++
1236        wrapper.
1237    
1238    
1239    Version 6.0 07-Jun-05
1240    ---------------------
1241    
1242     1. Some minor internal re-organization to help with my DFA experiments.
1243    
1244     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1245        didn't matter for the library itself when fully configured, but did matter
1246        when compiling without UCP support, or within Exim, where the ucp files are
1247        not imported.
1248    
1249     3. Refactoring of the library code to split up the various functions into
1250        different source modules. The addition of the new DFA matching code (see
1251        below) to a single monolithic source would have made it really too
1252        unwieldy, quite apart from causing all the code to be include in a
1253        statically linked application, when only some functions are used. This is
1254        relevant even without the DFA addition now that patterns can be compiled in
1255        one application and matched in another.
1256    
1257        The downside of splitting up is that there have to be some external
1258        functions and data tables that are used internally in different modules of
1259        the library but which are not part of the API. These have all had their
1260        names changed to start with "_pcre_" so that they are unlikely to clash
1261        with other external names.
1262    
1263     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1264        a different (DFA) algorithm. Although it is slower than the original
1265        function, it does have some advantages for certain types of matching
1266        problem.
1267    
1268     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1269        including restarting after a partial match.
1270    
1271     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1272        defined when compiling for Windows was sent to me. I have put it into the
1273        code, though I have no means of testing or verifying it.
1274    
1275     7. Added the pcre_refcount() auxiliary function.
1276    
1277     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1278        match before or at the first newline in the subject string. In pcretest,
1279        the /f option on a pattern can be used to set this.
1280    
1281     9. A repeated \w when used in UTF-8 mode with characters greater than 256
1282        would behave wrongly. This has been present in PCRE since release 4.0.
1283    
1284    10. A number of changes to the pcregrep command:
1285    
1286        (a) Refactored how -x works; insert ^(...)$ instead of setting
1287            PCRE_ANCHORED and checking the length, in preparation for adding
1288            something similar for -w.
1289    
1290        (b) Added the -w (match as a word) option.
1291    
1292        (c) Refactored the way lines are read and buffered so as to have more
1293            than one at a time available.
1294    
1295        (d) Implemented a pcregrep test script.
1296    
1297        (e) Added the -M (multiline match) option. This allows patterns to match
1298            over several lines of the subject. The buffering ensures that at least
1299            8K, or the rest of the document (whichever is the shorter) is available
1300            for matching (and similarly the previous 8K for lookbehind assertions).
1301    
1302        (f) Changed the --help output so that it now says
1303    
1304              -w, --word-regex(p)
1305    
1306            instead of two lines, one with "regex" and the other with "regexp"
1307            because that confused at least one person since the short forms are the
1308            same. (This required a bit of code, as the output is generated
1309            automatically from a table. It wasn't just a text change.)
1310    
1311        (g) -- can be used to terminate pcregrep options if the next thing isn't an
1312            option but starts with a hyphen. Could be a pattern or a path name
1313            starting with a hyphen, for instance.
1314    
1315        (h) "-" can be given as a file name to represent stdin.
1316    
1317        (i) When file names are being printed, "(standard input)" is used for
1318            the standard input, for compatibility with GNU grep. Previously
1319            "<stdin>" was used.
1320    
1321        (j) The option --label=xxx can be used to supply a name to be used for
1322            stdin when file names are being printed. There is no short form.
1323    
1324        (k) Re-factored the options decoding logic because we are going to add
1325            two more options that take data. Such options can now be given in four
1326            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1327    
1328        (l) Added the -A, -B, and -C options for requesting that lines of context
1329            around matches be printed.
1330    
1331        (m) Added the -L option to print the names of files that do not contain
1332            any matching lines, that is, the complement of -l.
1333    
1334        (n) The return code is 2 if any file cannot be opened, but pcregrep does
1335            continue to scan other files.
1336    
1337        (o) The -s option was incorrectly implemented. For compatibility with other
1338            greps, it now suppresses the error message for a non-existent or non-
1339            accessible file (but not the return code). There is a new option called
1340            -q that suppresses the output of matching lines, which was what -s was
1341            previously doing.
1342    
1343        (p) Added --include and --exclude options to specify files for inclusion
1344            and exclusion when recursing.
1345    
1346    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1347        Hopefully, it now does.
1348    
1349    12. Missing cast in pcre_study().
1350    
1351    13. Added an "uninstall" target to the makefile.
1352    
1353    14. Replaced "extern" in the function prototypes in Makefile.in with
1354        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1355        world, but is set differently for Windows.
1356    
1357    15. Added a second compiling function called pcre_compile2(). The only
1358        difference is that it has an extra argument, which is a pointer to an
1359        integer error code. When there is a compile-time failure, this is set
1360        non-zero, in addition to the error test pointer being set to point to an
1361        error message. The new argument may be NULL if no error number is required
1362        (but then you may as well call pcre_compile(), which is now just a
1363        wrapper). This facility is provided because some applications need a
1364        numeric error indication, but it has also enabled me to tidy up the way
1365        compile-time errors are handled in the POSIX wrapper.
1366    
1367    16. Added VPATH=.libs to the makefile; this should help when building with one
1368        prefix path and installing with another. (Or so I'm told by someone who
1369        knows more about this stuff than I do.)
1370    
1371    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1372        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1373        match everything, including newlines. This is not POSIX-compatible, but
1374        somebody wanted the feature. From pcretest it can be activated by using
1375        both the P and the s flags.
1376    
1377    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1378    
1379    19. libpcre.pc was being incorrectly installed as executable.
1380    
1381    20. A couple of places in pcretest check for end-of-line by looking for '\n';
1382        it now also looks for '\r' so that it will work unmodified on Windows.
1383    
1384    21. Added Google's contributed C++ wrapper to the distribution.
1385    
1386    22. Added some untidy missing memory free() calls in pcretest, to keep
1387        Electric Fence happy when testing.
1388    
1389    
1390    
1391    Version 5.0 13-Sep-04
1392    ---------------------
1393    
1394     1. Internal change: literal characters are no longer packed up into items
1395        containing multiple characters in a single byte-string. Each character
1396        is now matched using a separate opcode. However, there may be more than one
1397        byte in the character in UTF-8 mode.
1398    
1399     2. The pcre_callout_block structure has two new fields: pattern_position and
1400        next_item_length. These contain the offset in the pattern to the next match
1401        item, and its length, respectively.
1402    
1403     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1404        insertion of callouts before each pattern item. Added the /C option to
1405        pcretest to make use of this.
1406    
1407     4. On the advice of a Windows user, the lines
1408    
1409          #if defined(_WIN32) || defined(WIN32)
1410          _setmode( _fileno( stdout ), 0x8000 );
1411          #endif  /* defined(_WIN32) || defined(WIN32) */
1412    
1413        have been added to the source of pcretest. This apparently does useful
1414        magic in relation to line terminators.
1415    
1416     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1417        for the benefit of those environments where the "b" makes a difference.
1418    
1419     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1420        to know about it. I have put a hack into configure.in that adds in code
1421        to set GCC=yes if CC=icc. This seems to end up at a point in the
1422        generated configure script that is early enough to affect the setting of
1423        compiler options, which is what is needed, but I have no means of testing
1424        whether it really works. (The user who reported this had patched the
1425        generated configure script, which of course I cannot do.)
1426    
1427        LATER: After change 22 below (new libtool files), the configure script
1428        seems to know about icc (and also ecc). Therefore, I have commented out
1429        this hack in configure.in.
1430    
1431     7. Added support for pkg-config (2 patches were sent in).
1432    
1433     8. Negated POSIX character classes that used a combination of internal tables
1434        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1435        [[:^ascii]]. Typically, they would match almost any characters. The other
1436        POSIX classes were not broken in this way.
1437    
1438     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1439        to find the match, as PCRE was deluded into thinking that the match had to
1440        start at the start point or following a newline. The same bug applied to
1441        patterns with negative forward assertions or any backward assertions
1442        preceding ".*" at the start, unless the pattern required a fixed first
1443        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1444    
1445    10. In UTF-8 mode, when moving forwards in the subject after a failed match
1446        starting at the last subject character, bytes beyond the end of the subject
1447        string were read.
1448    
1449    11. Renamed the variable "class" as "classbits" to make life easier for C++
1450        users. (Previously there was a macro definition, but it apparently wasn't
1451        enough.)
1452    
1453    12. Added the new field "tables" to the extra data so that tables can be passed
1454        in at exec time, or the internal tables can be re-selected. This allows
1455        a compiled regex to be saved and re-used at a later time by a different
1456        program that might have everything at different addresses.
1457    
1458    13. Modified the pcre-config script so that, when run on Solaris, it shows a
1459        -R library as well as a -L library.
1460    
1461    14. The debugging options of pcretest (-d on the command line or D on a
1462        pattern) showed incorrect output for anything following an extended class
1463        that contained multibyte characters and which was followed by a quantifier.
1464    
1465    15. Added optional support for general category Unicode character properties
1466        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1467        support. It adds about 90K to the size of the library. The meanings of the
1468        inbuilt class escapes such as \d and \s have NOT been changed.
1469    
1470    16. Updated pcredemo.c to include calls to free() to release the memory for the
1471        compiled pattern.
1472    
1473    17. The generated file chartables.c was being created in the source directory
1474        instead of in the building directory. This caused the build to fail if the
1475        source directory was different from the building directory, and was
1476        read-only.
1477    
1478    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1479        file. No doubt somebody will tell me if they don't make sense... Also added
1480        Dan Mooney's comments about building on OpenVMS.
1481    
1482    19. Added support for partial matching via the PCRE_PARTIAL option for
1483        pcre_exec() and the \P data escape in pcretest.
1484    
1485    20. Extended pcretest with 3 new pattern features:
1486    
1487        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
1488              write the compiled pattern to the file whose name is "rest-of-line".
1489              This is a straight binary dump of the data, with the saved pointer to
1490              the character tables forced to be NULL. The study data, if any, is
1491              written too. After writing, pcretest reads a new pattern.
1492    
1493        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1494              compiled pattern from the given file. There must not be any
1495              occurrences of "<" in the file name (pretty unlikely); if there are,
1496              pcretest will instead treat the initial "<" as a pattern delimiter.
1497              After reading in the pattern, pcretest goes on to read data lines as
1498              usual.
1499    
1500        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1501              and 16-bit fields in a compiled pattern, to simulate a pattern that
1502              was compiled on a host of opposite endianness.
1503    
1504    21. The pcre-exec() function can now cope with patterns that were compiled on
1505        hosts of opposite endianness, with this restriction:
1506    
1507          As for any compiled expression that is saved and used later, the tables
1508          pointer field cannot be preserved; the extra_data field in the arguments
1509          to pcre_exec() should be used to pass in a tables address if a value
1510          other than the default internal tables were used at compile time.
1511    
1512    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1513        now diagnosed as an error. Previously, most of the time, a negative number
1514        would have been treated as zero, but if in addition "ovector" was passed as
1515        NULL, a crash could occur.
1516    
1517    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1518        new versions from the libtool 1.5 distribution (the last one is a copy of
1519        a file called libtool.m4). This seems to have fixed the need to patch
1520        "configure" to support Darwin 1.3 (which I used to do). However, I still
1521        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1522        workstation).
1523    
1524    24. Changed the PCRE licence to be the more standard "BSD" licence.
1525    
1526    
1527    Version 4.5 01-Dec-03
1528    ---------------------
1529    
1530     1. There has been some re-arrangement of the code for the match() function so
1531        that it can be compiled in a version that does not call itself recursively.
1532        Instead, it keeps those local variables that need separate instances for
1533        each "recursion" in a frame on the heap, and gets/frees frames whenever it
1534        needs to "recurse". Keeping track of where control must go is done by means
1535        of setjmp/longjmp. The whole thing is implemented by a set of macros that
1536        hide most of the details from the main code, and operates only if
1537        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1538        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1539        operating.
1540    
1541        To make it easier for callers to provide specially tailored get/free
1542        functions for this usage, two new functions, pcre_stack_malloc, and
1543        pcre_stack_free, are used. They are always called in strict stacking order,
1544        and the size of block requested is always the same.
1545    
1546        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1547        PCRE has been compiled to use the stack or the heap for recursion. The
1548        -C option of pcretest uses this to show which version is compiled.
1549    
1550        A new data escape \S, is added to pcretest; it causes the amounts of store
1551        obtained and freed by both kinds of malloc/free at match time to be added
1552        to the output.
1553    
1554     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1555        what's available on my current Linux desktop machine.
1556    
1557     3. When matching a UTF-8 string, the test for a valid string at the start has
1558        been extended. If start_offset is not zero, PCRE now checks that it points
1559        to a byte that is the start of a UTF-8 character. If not, it returns
1560        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1561        this is necessary because there may be backward assertions in the pattern.
1562        When matching the same subject several times, it may save resources to use
1563        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1564    
1565     4. The code for checking the validity of UTF-8 strings has been tightened so
1566        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1567        containing "overlong sequences".
1568    
1569     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1570        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1571        should have been "&", but it just so happened that all the cases this let
1572        through by mistake were picked up later in the function.
1573    
1574     6. I had used a variable called "isblank" - this is a C99 function, causing
1575        some compilers to warn. To avoid this, I renamed it (as "blankclass").
1576    
1577     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1578        prompting; (b) run "./pcretest /dev/null" at the start of the test script
1579        so the version is shown; (c) stop "make test" echoing "./RunTest".
1580    
1581     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1582    
1583     9. The prototype for memmove() for systems that don't have it was using
1584        size_t, but the inclusion of the header that defines size_t was later. I've
1585        moved the #includes for the C headers earlier to avoid this.
1586    
1587    10. Added some adjustments to the code to make it easier to compiler on certain
1588        special systems:
1589    
1590          (a) Some "const" qualifiers were missing.
1591          (b) Added the macro EXPORT before all exported functions; by default this
1592              is defined to be empty.
1593          (c) Changed the dftables auxiliary program (that builds chartables.c) so
1594              that it reads its output file name as an argument instead of writing
1595              to the standard output and assuming this can be redirected.
1596    
1597    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
1598        class containing characters with values greater than 255, PCRE compilation
1599        went into a loop.
1600    
1601    12. A recursive reference to a subpattern that was within another subpattern
1602        that had a minimum quantifier of zero caused PCRE to crash. For example,
1603        (x(y(?2))z)? provoked this bug with a subject that got as far as the
1604        recursion. If the recursively-called subpattern itself had a zero repeat,
1605        that was OK.
1606    
1607    13. In pcretest, the buffer for reading a data line was set at 30K, but the
1608        buffer into which it was copied (for escape processing) was still set at
1609        1024, so long lines caused crashes.
1610    
1611    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
1612        "internal error: code overflow...". This applied to any character class
1613        that was followed by a possessive quantifier.
1614    
1615    15. Modified the Makefile to add libpcre.la as a prerequisite for
1616        libpcreposix.la because I was told this is needed for a parallel build to
1617        work.
1618    
1619    16. If a pattern that contained .* following optional items at the start was
1620        studied, the wrong optimizing data was generated, leading to matching
1621        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
1622        matching string must start with a or b or c. The correct conclusion for
1623        this pattern is that a match can start with any character.
1624    
1625    
1626    Version 4.4 13-Aug-03
1627    ---------------------
1628    
1629     1. In UTF-8 mode, a character class containing characters with values between
1630        127 and 255 was not handled correctly if the compiled pattern was studied.
1631        In fixing this, I have also improved the studying algorithm for such
1632        classes (slightly).
1633    
1634     2. Three internal functions had redundant arguments passed to them. Removal
1635        might give a very teeny performance improvement.
1636    
1637     3. Documentation bug: the value of the capture_top field in a callout is *one
1638        more than* the number of the hightest numbered captured substring.
1639    
1640     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
1641        in incorrectly linking with a previously installed version. They now link
1642        explicitly with libpcre.la.
1643    
1644     5. configure.in no longer needs to recognize Cygwin specially.
1645    
1646     6. A problem in pcre.in for Windows platforms is fixed.
1647    
1648     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
1649        pcretest, it used to include the size of the study block as part of its
1650        output. Unfortunately, the structure contains a field that has a different
1651        size on different hardware architectures. This meant that the tests that
1652        showed this size failed. As the block is currently always of a fixed size,
1653        this information isn't actually particularly useful in pcretest output, so
1654        I have just removed it.
1655    
1656     8. Three pre-processor statements accidentally did not start in column 1.
1657        Sadly, there are *still* compilers around that complain, even though
1658        standard C has not required this for well over a decade. Sigh.
1659    
1660     9. In pcretest, the code for checking callouts passed small integers in the
1661        callout_data field, which is a void * field. However, some picky compilers
1662        complained about the casts involved for this on 64-bit systems. Now
1663        pcretest passes the address of the small integer instead, which should get
1664        rid of the warnings.
1665    
1666    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
1667        both compile and run time, and gives an error if an invalid UTF-8 sequence
1668        is found. There is a option for disabling this check in cases where the
1669        string is known to be correct and/or the maximum performance is wanted.
1670    
1671    11. In response to a bug report, I changed one line in Makefile.in from
1672    
1673            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
1674        to
1675            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
1676    
1677        to look similar to other lines, but I have no way of telling whether this
1678        is the right thing to do, as I do not use Windows. No doubt I'll get told
1679        if it's wrong...
1680    
1681    
1682    Version 4.3 21-May-03
1683    ---------------------
1684    
1685    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
1686       Makefile.
1687    
1688    2. Some refactoring to improve the quality of the code:
1689    
1690       (i)   The utf8_table... variables are now declared "const".
1691    
1692       (ii)  The code for \cx, which used the "case flipping" table to upper case
1693             lower case letters, now just substracts 32. This is ASCII-specific,
1694             but the whole concept of \cx is ASCII-specific, so it seems
1695             reasonable.
1696    
1697       (iii) PCRE was using its character types table to recognize decimal and
1698             hexadecimal digits in the pattern. This is silly, because it handles
1699             only 0-9, a-f, and A-F, but the character types table is locale-
1700             specific, which means strange things might happen. A private
1701             table is now used for this - though it costs 256 bytes, a table is
1702             much faster than multiple explicit tests. Of course, the standard
1703             character types table is still used for matching digits in subject
1704             strings against \d.
1705    
1706       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
1707             ending in _t are). So I've renamed it as ESC_tee.
1708    
1709    3. The first argument for regexec() in the POSIX wrapper should have been
1710       defined as "const".
1711    
1712    4. Changed pcretest to use malloc() for its buffers so that they can be
1713       Electric Fenced for debugging.
1714    
1715    5. There were several places in the code where, in UTF-8 mode, PCRE would try
1716       to read one or more bytes before the start of the subject string. Often this
1717       had no effect on PCRE's behaviour, but in some circumstances it could
1718       provoke a segmentation fault.
1719    
1720    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
1721       to try to read one or more bytes before the start of the subject string.
1722    
1723    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
1724       UTF-8 support could misbehave in various ways if the subject string
1725       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
1726       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
1727       back over UTF-8 characters.)
1728    
1729    
1730    Version 4.2 14-Apr-03
1731    ---------------------
1732    
1733    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
1734    
1735    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
1736         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
1737         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
1738         [WIN_PREFIX]: new variable, "cyg" for Cygwin
1739         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
1740           and BUILD_EXEEXT
1741         Note: automatic setting of the BUILD variables is not yet working
1742         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
1743           compile-time but not at link-time
1744         [LINK]: use for linking executables only
1745         make different versions for Windows and non-Windows
1746         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
1747           libraries
1748         [LINK_FOR_BUILD]: new variable
1749         [OBJEXT]: use throughout
1750         [EXEEXT]: use throughout
1751         <winshared>: new target
1752         <wininstall>: new target
1753         <dftables.o>: use native compiler
1754         <dftables>: use native linker
1755         <install>: handle Windows platform correctly
1756         <clean>: ditto
1757         <check>: ditto
1758         copy DLL to top builddir before testing
1759    
1760       As part of these changes, -no-undefined was removed again. This was reported
1761       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
1762       in any case.
1763    
1764    3. Some tidies to get rid of compiler warnings:
1765    
1766       . In the match_data structure, match_limit was an unsigned long int, whereas
1767         match_call_count was an int. I've made them both unsigned long ints.
1768    
1769       . In pcretest the fact that a const uschar * doesn't automatically cast to
1770         a void * provoked a warning.
1771    
1772       . Turning on some more compiler warnings threw up some "shadow" variables
1773         and a few more missing casts.
1774    
1775    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
1776       option, a class that contained a single character with a value between 128
1777       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
1778    
1779    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
1780       option, a class that contained several characters, but with at least one
1781       whose value was between 128 and 255 caused PCRE to crash.
1782    
1783    
1784    Version 4.1 12-Mar-03
1785    ---------------------
1786    
1787    1. Compiling with gcc -pedantic found a couple of places where casts were
1788    needed, and a string in dftables.c that was longer than standard compilers are
1789    required to support.
1790    
1791    2. Compiling with Sun's compiler found a few more places where the code could
1792    be tidied up in order to avoid warnings.
1793    
1794    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
1795    first of these names is deprecated in the latest Autoconf in favour of the name
1796    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
1797    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
1798    analogy I have changed it to CFLAGS_FOR_BUILD.
1799    
1800    4. Added -no-undefined to the linking command in the Makefile, because this is
1801    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
1802    linking step for the pcreposix library.
1803    
1804    5. PCRE was failing to diagnose the case of two named groups with the same
1805    name.
1806    
1807    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
1808    literal character that is needed in the subject for a match, and scans along to
1809    ensure that it is present before embarking on the full matching process. This
1810    saves time in cases of nested unlimited repeats that are never going to match.
1811    Problem: the scan can take a lot of time if the subject is very long (e.g.
1812    megabytes), thus penalizing straightforward matches. It is now done only if the
1813    amount of subject to be scanned is less than 1000 bytes.
1814    
1815    7. A lesser problem with the same optimization is that it was recording the
1816    first character of an anchored pattern as "needed", thus provoking a search
1817    right along the subject, even when the first match of the pattern was going to
1818    fail. The "needed" character is now not set for anchored patterns, unless it
1819    follows something in the pattern that is of non-fixed length. Thus, it still
1820    fulfils its original purpose of finding quick non-matches in cases of nested
1821    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
1822    
1823    
1824    Version 4.0 17-Feb-03
1825    ---------------------
1826    
1827    1. If a comment in an extended regex that started immediately after a meta-item
1828    extended to the end of string, PCRE compiled incorrect data. This could lead to
1829    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
1830    
1831    2. Moved to autoconf 2.53 and libtool 1.4.2.
1832    
1833    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
1834    the special perltest8 script is no longer needed - all the tests can be run
1835    from a single perltest script.
1836    
1837    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
1838    by \s. It has now been removed in PCRE. This means it isn't recognized as
1839    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
1840    class [:space:] *does* include VT, thereby creating a mess.
1841    
1842    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
1843    space and tab.
1844    
1845    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
1846    its new features into the main test script, reducing the number of scripts.
1847    
1848    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
1849    were backward compatible, and made the (?i) apply to the whole pattern, as if
1850    /i were given. Now it behaves more logically, and applies the option setting
1851    only to what follows. PCRE has been changed to follow suit. However, if it
1852    finds options settings right at the start of the pattern, it extracts them into
1853    the global options, as before. Thus, they show up in the info data.
1854    
1855    8. Added support for the \Q...\E escape sequence. Characters in between are
1856    treated as literals. This is slightly different from Perl in that $ and @ are
1857    also handled as literals inside the quotes. In Perl, they will cause variable
1858    interpolation. Note the following examples:
1859    
1860        Pattern            PCRE matches      Perl matches
1861    
1862        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
1863        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
1864        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
1865    
1866    For compatibility with Perl, \Q...\E sequences are recognized inside character
1867    classes as well as outside them.
1868    
1869    9. Re-organized 3 code statements in pcretest to avoid "overflow in
1870    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
1871    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
1872    signed/unsigned warnings.
1873    
1874    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
1875    option for pcretest, so I've replaced it by a simple function that does just
1876    that job.
1877    
1878    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
1879    "pcregrep -".
1880    
1881    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
1882    Java package. This provides some syntactic sugar for simple cases of what my
1883    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
1884    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
1885    item, you can use this simplified notation. Note that only makes sense with
1886    greedy quantifiers. Consequently, the use of the possessive quantifier forces
1887    greediness, whatever the setting of the PCRE_UNGREEDY option.
1888    
1889    13. A change of greediness default within a pattern was not taking effect at
1890    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
1891    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
1892    was abstracted outside.
1893    
1894    14. PCRE now supports the \G assertion. It is true when the current matching
1895    position is at the start point of the match. This differs from \A when the
1896    starting offset is non-zero. Used with the /g option of pcretest (or similar
1897    code), it works in the same way as it does for Perl's /g option. If all
1898    alternatives of a regex begin with \G, the expression is anchored to the start
1899    match position, and the "anchored" flag is set in the compiled expression.
1900    
1901    15. Some bugs concerning the handling of certain option changes within patterns
1902    have been fixed. These applied to options other than (?ims). For example,
1903    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
1904    been the other way round. Some of this was related to change 7 above.
1905    
1906    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
1907    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
1908    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
1909    POSIX classes only within a class (e.g. /[[:alpha:]]/).
1910    
1911    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
1912    mode. Unlike ".", it always matches newline, whatever the setting of
1913    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
1914    assertions. Perl allows it, but it doesn't (in general) work because it can't
1915    calculate the length of the lookbehind. At least, that's the case for Perl
1916    5.8.0 - I've been told they are going to document that it doesn't work in
1917    future.
1918    
1919    18. Added an error diagnosis for escapes that PCRE does not support: these are
1920    \L, \l, \N, \P, \p, \U, \u, and \X.
1921    
1922    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
1923    reading past the end of the pattern in cases such as /[abcd/.
1924    
1925    20. PCRE was getting more memory than necessary for patterns with classes that
1926    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
1927    
1928    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
1929    compiling PCRE for use with Virtual Pascal.
1930    
1931    22. Small fix to the Makefile to make it work properly if the build is done
1932    outside the source tree.
1933    
1934    23. Added a new extension: a condition to go with recursion. If a conditional
1935    subpattern starts with (?(R) the "true" branch is used if recursion has
1936    happened, whereas the "false" branch is used only at the top level.
1937    
1938    24. When there was a very long string of literal characters (over 255 bytes
1939    without UTF support, over 250 bytes with UTF support), the computation of how
1940    much memory was required could be incorrect, leading to segfaults or other
1941    strange effects.
1942    
1943    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
1944    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
1945    there was a subsequent back reference to those brackets. This meant that, for
1946    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
1947    possible to check for precisely this case. All we can do is abandon the
1948    optimization if .* occurs inside capturing brackets when there are any back
1949    references whatsoever. (See below for a better fix that came later.)
1950    
1951    26. The handling of the optimization for finding the first character of a
1952    non-anchored pattern, and for finding a character that is required later in the
1953    match were failing in some cases. This didn't break the matching; it just
1954    failed to optimize when it could. The way this is done has been re-implemented.
1955    
1956    27. Fixed typo in error message for invalid (?R item (it said "(?p").
1957    
1958    28. Added a new feature that provides some of the functionality that Perl
1959    provides with (?{...}). The facility is termed a "callout". The way it is done
1960    in PCRE is for the caller to provide an optional function, by setting
1961    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
1962    global variable. By default it is unset, which disables all calling out. To get
1963    the function called, the regex must include (?C) at appropriate points. This
1964    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
1965    This provides a means of identifying different callout points. When PCRE
1966    reaches such a point in the regex, if pcre_callout has been set, the external
1967    function is called. It is provided with data in a structure called
1968    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
1969    matching continues; if it returns a non-zero value, the match at the current
1970    point fails. However, backtracking will occur if possible. [This was changed
1971    later and other features added - see item 49 below.]
1972    
1973    29. pcretest is upgraded to test the callout functionality. It provides a
1974    callout function that displays information. By default, it shows the start of
1975    the match and the current position in the text. There are some new data escapes
1976    to vary what happens:
1977    
1978        \C+         in addition, show current contents of captured substrings
1979        \C-         do not supply a callout function
1980        \C!n        return 1 when callout number n is reached
1981        \C!n!m      return 1 when callout number n is reached for the mth time
1982    
1983    30. If pcregrep was called with the -l option and just a single file name, it
1984    output "<stdin>" if a match was found, instead of the file name.
1985    
1986    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
1987    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
1988    pcre_exec(). This saves a malloc/free per call. The default value of
1989    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
1990    when configuring.
1991    
1992    32. The default maximum size of a compiled pattern is 64K. There have been a
1993    few cases of people hitting this limit. The code now uses macros to handle the
1994    storing of links as offsets within the compiled pattern. It defaults to 2-byte
1995    links, but this can be changed to 3 or 4 bytes by --with-link-size when
1996    configuring. Tests 2 and 5 work only with 2-byte links because they output
1997    debugging information about compiled patterns.
1998    
1999    33. Internal code re-arrangements:
2000    
2001    (a) Moved the debugging function for printing out a compiled regex into
2002        its own source file (printint.c) and used #include to pull it into
2003        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
2004        separate copies.
2005    
2006    (b) Defined the list of op-code names for debugging as a macro in
2007        internal.h so that it is next to the definition of the opcodes.
2008    
2009    (c) Defined a table of op-code lengths for simpler skipping along compiled
2010        code. This is again a macro in internal.h so that it is next to the
2011        definition of the opcodes.
2012    
2013    34. Added support for recursive calls to individual subpatterns, along the
2014    lines of Robin Houston's patch (but implemented somewhat differently).
2015    
2016    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
2017    allow it to read and process whole directories in Win32. This code was
2018    contributed by Lionel Fourquaux; it has not been tested by me.
2019    
2020    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
2021    used to name a group. Names consist of alphanumerics and underscores, and must
2022    be unique. Back references use the syntax (?P=name) and recursive calls use
2023    (?P>name) which is a PCRE extension to the Python extension. Groups still have
2024    numbers. The function pcre_fullinfo() can be used after compilation to extract
2025    a name/number map. There are three relevant calls:
2026    
2027      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
2028      PCRE_INFO_NAMECOUNT            yields the number of entries
2029      PCRE_INFO_NAMETABLE            yields a pointer to the map.
2030    
2031    The map is a vector of fixed-size entries. The size of each entry depends on
2032    the length of the longest name used. The first two bytes of each entry are the
2033    group number, most significant byte first. There follows the corresponding
2034    name, zero terminated. The names are in alphabetical order.
2035    
2036    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
2037    case instead of 255. Making it the same both with and without UTF-8 support
2038    means that the same test output works with both.
2039    
2040    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
2041    calling malloc() with a zero argument.
2042    
2043    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
2044    optimization. I've improved things by keeping a bitmap of backreferences with
2045    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
2046    fact referenced, the optimization can be applied. It is unlikely that a
2047    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
2048    the match to follow \n) will appear inside brackets with a number greater than
2049    31, but if it does, any back reference > 31 suppresses the optimization.
2050    
2051    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
2052    of disabling numbered capturing parentheses. Any opening parenthesis that is
2053    not followed by ? behaves as if it were followed by ?: but named parentheses
2054    can still be used for capturing (and they will acquire numbers in the usual
2055    way).
2056    
2057    41. Redesigned the return codes from the match() function into yes/no/error so
2058    that errors can be passed back from deep inside the nested calls. A malloc
2059    failure while inside a recursive subpattern call now causes the
2060    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
2061    
2062    42. It is now possible to set a limit on the number of times the match()
2063    function is called in a call to pcre_exec(). This facility makes it possible to
2064    limit the amount of recursion and backtracking, though not in a directly
2065    obvious way, because the match() function is used in a number of different
2066    circumstances. The count starts from zero for each position in the subject
2067    string (for non-anchored patterns). The default limit is, for compatibility, a
2068    large number, namely 10 000 000. You can change this in two ways:
2069    
2070    (a) When configuring PCRE before making, you can use --with-match-limit=n
2071        to set a default value for the compiled library.
2072    
2073    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
2074        a different value is set. See 45 below.
2075    
2076    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
2077    
2078    43. Added a new function pcre_config(int, void *) to enable run-time extraction
2079    of things that can be changed at compile time. The first argument specifies
2080    what is wanted and the second points to where the information is to be placed.
2081    The current list of available information is:
2082    
2083      PCRE_CONFIG_UTF8
2084    
2085    The output is an integer that is set to one if UTF-8 support is available;
2086    otherwise it is set to zero.
2087    
2088      PCRE_CONFIG_NEWLINE
2089    
2090    The output is an integer that it set to the value of the code that is used for
2091    newline. It is either LF (10) or CR (13).
2092    
2093      PCRE_CONFIG_LINK_SIZE
2094    
2095    The output is an integer that contains the number of bytes used for internal
2096    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
2097    
2098      PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
2099    
2100    The output is an integer that contains the threshold above which the POSIX
2101    interface uses malloc() for output vectors. See item 31 above.
2102    
2103      PCRE_CONFIG_MATCH_LIMIT
2104    
2105    The output is an unsigned integer that contains the default limit of the number
2106    of match() calls in a pcre_exec() execution. See 42 above.
2107    
2108    44. pcretest has been upgraded by the addition of the -C option. This causes it
2109    to extract all the available output from the new pcre_config() function, and to
2110    output it. The program then exits immediately.
2111    
2112    45. A need has arisen to pass over additional data with calls to pcre_exec() in
2113    order to support additional features. One way would have been to define
2114    pcre_exec2() (for example) with extra arguments, but this would not have been
2115    extensible, and would also have required all calls to the original function to
2116    be mapped to the new one. Instead, I have chosen to extend the mechanism that
2117    is used for passing in "extra" data from pcre_study().
2118    
2119    The pcre_extra structure is now exposed and defined in pcre.h. It currently
2120    contains the following fields:
2121    
2122      flags         a bitmap indicating which of the following fields are set
2123      study_data    opaque data from pcre_study()
2124      match_limit   a way of specifying a limit on match() calls for a specific
2125                      call to pcre_exec()
2126      callout_data  data for callouts (see 49 below)
2127    
2128    The flag bits are also defined in pcre.h, and are
2129    
2130      PCRE_EXTRA_STUDY_DATA
2131      PCRE_EXTRA_MATCH_LIMIT
2132      PCRE_EXTRA_CALLOUT_DATA
2133    
2134    The pcre_study() function now returns one of these new pcre_extra blocks, with
2135    the actual study data pointed to by the study_data field, and the
2136    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
2137    before. That is, this change is entirely upwards-compatible and requires no
2138    change to existing code.
2139    
2140    If you want to pass in additional data to pcre_exec(), you can either place it
2141    in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
2142    block.
2143    
2144    46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
2145    data string contains the escape sequence \M, pcretest calls pcre_exec() several
2146    times with different match limits, until it finds the minimum value needed for
2147    pcre_exec() to complete. The value is then output. This can be instructive; for
2148    most simple matches the number is quite small, but for pathological cases it
2149    gets very large very quickly.
2150    
2151    47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
2152    returns the size of the data block pointed to by the study_data field in a
2153    pcre_extra block, that is, the value that was passed as the argument to
2154    pcre_malloc() when PCRE was getting memory in which to place the information
2155    created by pcre_study(). The fourth argument should point to a size_t variable.
2156    pcretest has been extended so that this information is shown after a successful
2157    pcre_study() call when information about the compiled regex is being displayed.
2158    
2159    48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
2160    because what follows is always an absolute path. (Later: it turns out that this
2161    is more than cosmetic for MinGW, because it doesn't like empty path
2162    components.)
2163    
2164    49. Some changes have been made to the callout feature (see 28 above):
2165    
2166    (i)  A callout function now has three choices for what it returns:
2167    
2168           0  =>  success, carry on matching
2169         > 0  =>  failure at this point, but backtrack if possible
2170         < 0  =>  serious error, return this value from pcre_exec()
2171    
2172         Negative values should normally be chosen from the set of PCRE_ERROR_xxx
2173         values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
2174         "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
2175         use by callout functions. It will never be used by PCRE itself.
2176    
2177    (ii) The pcre_extra structure (see 45 above) has a void * field called
2178         callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
2179         pcre_callout_block structure has a field of the same name. The contents of
2180         the field passed in the pcre_extra structure are passed to the callout
2181         function in the corresponding field in the callout block. This makes it
2182         easier to use the same callout-containing regex from multiple threads. For
2183         testing, the pcretest program has a new data escape
2184    
2185           \C*n        pass the number n (may be negative) as callout_data
2186    
2187         If the callout function in pcretest receives a non-zero value as
2188         callout_data, it returns that value.
2189    
2190    50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
2191    there were some redundant $(CFLAGS) in commands that are now specified as
2192    $(LINK), which already includes $(CFLAGS).
2193    
2194    51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
2195    has been compiled with UTF-8 support *and* pcre_compile() has been compiled
2196    with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
2197    one-byte characters throughout. Note that case-insensitive matching applies
2198    only to characters whose values are less than 256. PCRE doesn't support the
2199    notion of cases for higher-valued characters.
2200    
2201    (i)   A character class whose characters are all within 0-255 is handled as
2202          a bit map, and the map is inverted for negative classes. Previously, a
2203          character > 255 always failed to match such a class; however it should
2204          match if the class was a negative one (e.g. [^ab]). This has been fixed.
2205    
2206    (ii)  A negated character class with a single character < 255 is coded as
2207          "not this character" (OP_NOT). This wasn't working properly when the test
2208          character was multibyte, either singly or repeated.
2209    
2210    (iii) Repeats of multibyte characters are now handled correctly in UTF-8
2211          mode, for example: \x{100}{2,3}.
2212    
2213    (iv)  The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
2214          singly or repeated) now correctly test multibyte characters. However,
2215          PCRE doesn't recognize any characters with values greater than 255 as
2216          digits, spaces, or word characters. Such characters always match \D, \S,
2217          and \W, and never match \d, \s, or \w.
2218    
2219    (v)   Classes may now contain characters and character ranges with values
2220          greater than 255. For example: [ab\x{100}-\x{400}].
2221    
2222    (vi)  pcregrep now has a --utf-8 option (synonym -u) which makes it call
2223          PCRE in UTF-8 mode.
2224    
2225    52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
2226    PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
2227    retained for backwards compatibility. (Note that LASTLITERAL is also a byte
2228    value.)
2229    
2230    53. The single man page has become too large. I have therefore split it up into
2231    a number of separate man pages. These also give rise to individual HTML pages;
2232    these are now put in a separate directory, and there is an index.html page that
2233    lists them all. Some hyperlinking between the pages has been installed.
2234    
2235    54. Added convenience functions for handling named capturing parentheses.
2236    
2237    55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2238    aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2239    true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2240    are faulted.
2241    
2242    56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2243    calling configure. These values are used when compiling the dftables.c program
2244    which is run to generate the source of the default character tables. They
2245    default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2246    you will need to set these values.
2247    
2248    57. Updated the building process for Windows DLL, as provided by Fred Cox.
2249    
2250    
2251    Version 3.9 02-Jan-02
2252    ---------------------
2253    
2254    1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2255    
2256    2. If --disable-static was given, the building process failed when trying to
2257    build pcretest and pcregrep. (For some reason it was using libtool to compile
2258    them, which is not right, as they aren't part of the library.)
2259    
2260    
2261    Version 3.8 18-Dec-01
2262    ---------------------
2263    
2264    1. The experimental UTF-8 code was completely screwed up. It was packing the
2265    bytes in the wrong order. How dumb can you get?
2266    
2267    
2268    Version 3.7 29-Oct-01
2269    ---------------------
2270    
2271    1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2272    This caused pcretest, when used on the test data, to segfault. Unfortunately,
2273    this didn't happen under Solaris 8, where I normally test things.
2274    
2275    2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2276    doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2277    isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2278    this fix an hour or so after the initial 3.7 release.)
2279    
2280    
2281    Version 3.6 23-Oct-01
2282    ---------------------
2283    
2284    1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2285    offsets passed as NULL with zero offset count.
2286    
2287    2. The config.guess and config.sub files had not been updated when I moved to
2288    the latest autoconf.
2289    
2290    
2291    Version 3.5 15-Aug-01
2292    ---------------------
2293    
2294    1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2295    had been forgotten.
2296    
2297    2. By using declared but undefined structures, we can avoid using "void"
2298    definitions in pcre.h while keeping the internal definitions of the structures
2299    private.
2300    
2301    3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2302    user point of view, this means that both static and shared libraries are built
2303    by default, but this can be individually controlled. More of the work of
2304    handling this static/shared cases is now inside libtool instead of PCRE's make
2305    file.
2306    
2307    4. The pcretest utility is now installed along with pcregrep because it is
2308    useful for users (to test regexs) and by doing this, it automatically gets
2309    relinked by libtool. The documentation has been turned into a man page, so
2310    there are now .1, .txt, and .html versions in /doc.
2311    
2312    5. Upgrades to pcregrep:
2313       (i)   Added long-form option names like gnu grep.
2314       (ii)  Added --help to list all options with an explanatory phrase.
2315       (iii) Added -r, --recursive to recurse into sub-directories.
2316       (iv)  Added -f, --file to read patterns from a file.
2317    
2318    6. pcre_exec() was referring to its "code" argument before testing that
2319    argument for NULL (and giving an error if it was NULL).
2320    
2321    7. Upgraded Makefile.in to allow for compiling in a different directory from
2322    the source directory.
2323    
2324    8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2325    options bits, the pointer it was passed was to an int instead of to an unsigned
2326    long int. This mattered only on 64-bit systems.
2327    
2328    9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2329    generated) instead of pcre.in, which it its source. Also made the same change
2330    in several of the .c files.
2331    
2332    10. A new release of gcc defines printf() as a macro, which broke pcretest
2333    because it had an ifdef in the middle of a string argument for printf(). Fixed
2334    by using separate calls to printf().
2335    
2336    11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2337    script, to force use of CR or LF instead of \n in the source. On non-Unix
2338    systems, the value can be set in config.h.
2339    
2340    12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2341    absolute limit. Changed the text of the error message to make this clear, and
2342    likewise updated the man page.
2343    
2344    13. The limit of 99 on the number of capturing subpatterns has been removed.
2345    The new limit is 65535, which I hope will not be a "real" limit.
2346    
2347    
2348    Version 3.4 22-Aug-00
2349    ---------------------
2350    
2351    1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2352    
2353    2. Diagnose condition (?(0) as an error instead of crashing on matching.
2354    
2355    
2356  Version 3.3 01-Aug-00  Version 3.3 01-Aug-00
2357  ---------------------  ---------------------

Legend:
Removed from v.49  
changed lines
  Added in v.244

  ViewVC Help
Powered by ViewVC 1.1.5