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

Diff of /code/trunk/ChangeLog

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

revision 47 by nigel, Sat Feb 24 21:39:29 2007 UTC revision 268 by ph10, Thu Nov 15 10:28:09 2007 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4    Version 7.5 12-Nov-07
5    ---------------------
6    
7    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
8        values in parens when parsing an RE using the C++ wrapper."
9    
10    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
11        Characters greater than 255 were excluded from the class instead of being
12        included.
13    
14    3.  The same bug as (2) above applied to negated POSIX classes such as
15        [:^space:].
16    
17    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
18        defined or documented. It seems to have been a typo for PCRE_STATIC, so
19        I have changed it.
20    
21    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
22        first named subpattern) and a construct such as (?&a) would reference the
23        first named subpattern whose name started with "a" (in other words, the
24        length check was missing).
25    
26    
27    Version 7.4 21-Sep-07
28    ---------------------
29    
30    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
31        means that a class such as [\s] counted as "explicit reference to CR or
32        LF". That isn't really right - the whole point of the change was to try to
33        help when there was an actual mention of one of the two characters. So now
34        the change happens only if \r or \n (or a literal CR or LF) character is
35        encountered.
36    
37    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
38        of both had grown to the point where there were only 3 bits left.
39        Fortunately, there was spare space in the data structure, and so I have
40        moved the internal flags into a new 16-bit field to free up more option
41        bits.
42    
43    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
44        but did not set the internal JCHANGED flag - either of these is enough to
45        control the way the "get" function works - but the PCRE_INFO_JCHANGED
46        facility is supposed to tell if (?J) was ever used, so now (?J) at the
47        start sets both bits.
48    
49    4.  Added options (at build time, compile time, exec time) to change \R from
50        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
51    
52    5.  doc/pcresyntax.html was missing from the distribution.
53    
54    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
55        compatibility, even though it is no longer used.
56    
57    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
58        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
59        windows.h file is present (where different names are used). [This was
60        reversed later after testing - see 16 below.]
61    
62    8.  Changed all #include <config.h> to #include "config.h". There were also
63        some further <pcre.h> cases that I changed to "pcre.h".
64    
65    9.  When pcregrep was used with the --colour option, it missed the line ending
66        sequence off the lines that it output.
67    
68    10. It was pointed out to me that arrays of string pointers cause lots of
69        relocations when a shared library is dynamically loaded. A technique of
70        using a single long string with a table of offsets can drastically reduce
71        these. I have refactored PCRE in four places to do this. The result is
72        dramatic:
73    
74          Originally:                          290
75          After changing UCP table:            187
76          After changing error message table:   43
77          After changing table of "verbs"       36
78          After changing table of Posix names   22
79    
80        Thanks to the folks working on Gregex for glib for this insight.
81    
82    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
83        unicode-properties was also set.
84    
85    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
86    
87    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
88        checked only for CRLF.
89    
90    14. Added casts to pcretest.c to avoid compiler warnings.
91    
92    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
93    
94    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
95        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
96        entirely. This removes changes made in 7 above.
97    
98    17. The CMake files have been updated, and there is now more information about
99        building with CMake in the NON-UNIX-USE document.
100    
101    
102    Version 7.3 28-Aug-07
103    ---------------------
104    
105     1. In the rejigging of the build system that eventually resulted in 7.1, the
106        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
107        brackets there is not right, since it causes compilers to look for an
108        installed pcre.h, not the version that is in the source that is being
109        compiled (which of course may be different). I have changed it back to:
110    
111          #include "pcre.h"
112    
113        I have a vague recollection that the change was concerned with compiling in
114        different directories, but in the new build system, that is taken care of
115        by the VPATH setting the Makefile.
116    
117     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
118        when the subject happened to end in the byte 0x85 (e.g. if the last
119        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
120        characters but of course it shouldn't be taken as a newline when it is part
121        of another character. The bug was that, for an unlimited repeat of . in
122        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
123        characters when looking for a newline.
124    
125     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
126    
127     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
128        in debug output.
129    
130     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
131        long printing in the pcrecpp unittest when running under MinGW.
132    
133     6. ESC_K was left out of the EBCDIC table.
134    
135     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
136        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
137        limit also applies to "virtual nesting" when a pattern is recursive, and in
138        this case 1000 isn't so big. I have been able to remove this limit at the
139        expense of backing off one optimization in certain circumstances. Normally,
140        when pcre_exec() would call its internal match() function recursively and
141        immediately return the result unconditionally, it uses a "tail recursion"
142        feature to save stack. However, when a subpattern that can match an empty
143        string has an unlimited repetition quantifier, it no longer makes this
144        optimization. That gives it a stack frame in which to save the data for
145        checking that an empty string has been matched. Previously this was taken
146        from the 1000-entry workspace that had been reserved. So now there is no
147        explicit limit, but more stack is used.
148    
149     8. Applied Daniel's patches to solve problems with the import/export magic
150        syntax that is required for Windows, and which was going wrong for the
151        pcreposix and pcrecpp parts of the library. These were overlooked when this
152        problem was solved for the main library.
153    
154     9. There were some crude static tests to avoid integer overflow when computing
155        the size of patterns that contain repeated groups with explicit upper
156        limits. As the maximum quantifier is 65535, the maximum group length was
157        set at 30,000 so that the product of these two numbers did not overflow a
158        32-bit integer. However, it turns out that people want to use groups that
159        are longer than 30,000 bytes (though not repeat them that many times).
160        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
161        made it possible to implement the integer overflow checks in a much more
162        dynamic way, which I have now done. The artificial limitation on group
163        length has been removed - we now have only the limit on the total length of
164        the compiled pattern, which depends on the LINK_SIZE setting.
165    
166    10. Fixed a bug in the documentation for get/copy named substring when
167        duplicate names are permitted. If none of the named substrings are set, the
168        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
169        empty string.
170    
171    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
172        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
173        because the ] is interpreted as the first data character and the
174        terminating ] is not found. PCRE has been made compatible with Perl in this
175        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
176        cause memory overwriting.
177    
178    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
179        string has been matched (to stop an infinite loop). It was not recognizing
180        a conditional subpattern that could match an empty string if that
181        subpattern was within another subpattern. For example, it looped when
182        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
183        condition was not nested. This bug has been fixed.
184    
185    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
186        past the start of the subject in the presence of bytes with the top bit
187        set, for example "\x8aBCD".
188    
189    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
190        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
191    
192    14. Optimized (?!) to (*FAIL).
193    
194    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
195        This restricts code points to be within the range 0 to 0x10FFFF, excluding
196        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
197        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
198        does: it's just the validity check that is more restrictive.
199    
200    16. Inserted checks for integer overflows during escape sequence (backslash)
201        processing, and also fixed erroneous offset values for syntax errors during
202        backslash processing.
203    
204    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
205        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
206    
207    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
208        caused an overrun.
209    
210    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
211        something other than just ASCII characters) inside a group that had an
212        unlimited repeat caused a loop at compile time (while checking to see
213        whether the group could match an empty string).
214    
215    20. Debugging a pattern containing \p or \P could cause a crash. For example,
216        [\P{Any}] did so. (Error in the code for printing property names.)
217    
218    21. An orphan \E inside a character class could cause a crash.
219    
220    22. A repeated capturing bracket such as (A)? could cause a wild memory
221        reference during compilation.
222    
223    23. There are several functions in pcre_compile() that scan along a compiled
224        expression for various reasons (e.g. to see if it's fixed length for look
225        behind). There were bugs in these functions when a repeated \p or \P was
226        present in the pattern. These operators have additional parameters compared
227        with \d, etc, and these were not being taken into account when moving along
228        the compiled data. Specifically:
229    
230        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
231            length.
232    
233        (b) An item such as \pL+ within a repeated group could cause crashes or
234            loops.
235    
236        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
237            "reference to non-existent subpattern" error.
238    
239        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
240    
241    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
242        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
243    
244    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
245    
246    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
247        character were causing crashes (broken optimization).
248    
249    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
250        \p or \P) caused a compile-time loop.
251    
252    28. More problems have arisen in unanchored patterns when CRLF is a valid line
253        break. For example, the unstudied pattern [\r\n]A does not match the string
254        "\r\nA" because change 7.0/46 below moves the current point on by two
255        characters after failing to match at the start. However, the pattern \nA
256        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
257        the same is true. There doesn't seem any very clean way out of this, but
258        what I have chosen to do makes the common cases work: PCRE now takes note
259        of whether there can be an explicit match for \r or \n anywhere in the
260        pattern, and if so, 7.0/46 no longer applies. As part of this change,
261        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
262        pattern has explicit CR or LF references.
263    
264    29. Added (*CR) etc for changing newline setting at start of pattern.
265    
266    
267    Version 7.2 19-Jun-07
268    ---------------------
269    
270     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
271        which is apparently normally available under Windows.
272    
273     2. Re-jig the pcregrep tests with different newline settings in an attempt
274        to make them independent of the local environment's newline setting.
275    
276     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
277    
278     4. Some of the "internals" tests were previously cut out when the link size
279        was not 2, because the output contained actual offsets. The recent new
280        "Z" feature of pcretest means that these can be cut out, making the tests
281        usable with all link sizes.
282    
283     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
284        stack recursion. This gives a massive performance boost under BSD, but just
285        a small improvement under Linux. However, it saves one field in the frame
286        in all cases.
287    
288     6. Added more features from the forthcoming Perl 5.10:
289    
290        (a) (?-n) (where n is a string of digits) is a relative subroutine or
291            recursion call. It refers to the nth most recently opened parentheses.
292    
293        (b) (?+n) is also a relative subroutine call; it refers to the nth next
294            to be opened parentheses.
295    
296        (c) Conditions that refer to capturing parentheses can be specified
297            relatively, for example, (?(-2)... or (?(+3)...
298    
299        (d) \K resets the start of the current match so that everything before
300            is not part of it.
301    
302        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
303    
304        (f) \g{name} is another synonym - part of Perl 5.10's unification of
305            reference syntax.
306    
307        (g) (?| introduces a group in which the numbering of parentheses in each
308            alternative starts with the same number.
309    
310        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
311    
312     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
313        PCRE_INFO_JCHANGED.
314    
315     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
316        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
317        for detecting groups that can match an empty string.
318    
319     9. A pattern with a very large number of alternatives (more than several
320        hundred) was running out of internal workspace during the pre-compile
321        phase, where pcre_compile() figures out how much memory will be needed. A
322        bit of new cunning has reduced the workspace needed for groups with
323        alternatives. The 1000-alternative test pattern now uses 12 bytes of
324        workspace instead of running out of the 4096 that are available.
325    
326    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
327    
328    11. Applied patch from Google to remove an optimization that didn't quite work.
329        The report of the bug said:
330    
331          pcrecpp::RE("a*").FullMatch("aaa") matches, while
332          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
333          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
334    
335    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
336        it matched the wrong number of bytes.
337    
338    
339    Version 7.1 24-Apr-07
340    ---------------------
341    
342     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
343        that is more "standard", making use of automake and other Autotools. There
344        is some re-arrangement of the files and adjustment of comments consequent
345        on this.
346    
347     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
348        for recursive directory scanning broke on some systems because the files
349        are not scanned in any specific order and on different systems the order
350        was different. A call to "sort" has been inserted into RunGrepTest for the
351        approprate test as a short-term fix. In the longer term there may be an
352        alternative.
353    
354     3. I had an email from Eric Raymond about problems translating some of PCRE's
355        man pages to HTML (despite the fact that I distribute HTML pages, some
356        people do their own conversions for various reasons). The problems
357        concerned the use of low-level troff macros .br and .in. I have therefore
358        removed all such uses from the man pages (some were redundant, some could
359        be replaced by .nf/.fi pairs). The 132html script that I use to generate
360        HTML has been updated to handle .nf/.fi and to complain if it encounters
361        .br or .in.
362    
363     4. Updated comments in configure.ac that get placed in config.h.in and also
364        arranged for config.h to be included in the distribution, with the name
365        config.h.generic, for the benefit of those who have to compile without
366        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
367    
368     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
369        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
370        makevp.bat for the current PCRE, using the additional files
371        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
372    
373     6. A Windows user reported a minor discrepancy with test 2, which turned out
374        to be caused by a trailing space on an input line that had got lost in his
375        copy. The trailing space was an accident, so I've just removed it.
376    
377     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
378        that is needed.
379    
380     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
381        as "const" (a) because they are and (b) because it helps the PHP
382        maintainers who have recently made a script to detect big data structures
383        in the php code that should be moved to the .rodata section. I remembered
384        to update Builducptable as well, so it won't revert if ucptable.h is ever
385        re-created.
386    
387     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
388        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
389        order to be able to cut out the UTF-8 tables in the latter when UTF-8
390        support is not required. This saves 1.5-2K of code, which is important in
391        some applications.
392    
393        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
394        so as not to refer to the tables, even though these functions will never be
395        called when UTF-8 support is disabled. Otherwise there are problems with a
396        shared library.
397    
398    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
399    
400        (a) It was defining its arguments as char * instead of void *.
401    
402        (b) It was assuming that all moves were upwards in memory; this was true
403            a long time ago when I wrote it, but is no longer the case.
404    
405        The emulated memove() is provided for those environments that have neither
406        memmove() nor bcopy(). I didn't think anyone used it these days, but that
407        is clearly not the case, as these two bugs were recently reported.
408    
409    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
410        and Detrail to create the HTML documentation, the .txt form of the man
411        pages, and it removes trailing spaces from listed files. It also creates
412        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
413        case, it wraps all the #defines with #ifndefs. This script should be run
414        before "make dist".
415    
416    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
417        with Unicode property support.
418    
419        (a) For a maximizing quantifier, if the two different cases of the
420            character were of different lengths in their UTF-8 codings (there are
421            some cases like this - I found 11), and the matching function had to
422            back up over a mixture of the two cases, it incorrectly assumed they
423            were both the same length.
424    
425        (b) When PCRE was configured to use the heap rather than the stack for
426            recursion during matching, it was not correctly preserving the data for
427            the other case of a UTF-8 character when checking ahead for a match
428            while processing a minimizing repeat. If the check also involved
429            matching a wide character, but failed, corruption could cause an
430            erroneous result when trying to check for a repeat of the original
431            character.
432    
433    13. Some tidying changes to the testing mechanism:
434    
435        (a) The RunTest script now detects the internal link size and whether there
436            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
437            values substituted by "configure". (The RunGrepTest script already did
438            this for UTF-8.) The configure.ac script no longer substitutes the
439            relevant variables.
440    
441        (b) The debugging options /B and /D in pcretest show the compiled bytecode
442            with length and offset values. This means that the output is different
443            for different internal link sizes. Test 2 is skipped for link sizes
444            other than 2 because of this, bypassing the problem. Unfortunately,
445            there was also a test in test 3 (the locale tests) that used /B and
446            failed for link sizes other than 2. Rather than cut the whole test out,
447            I have added a new /Z option to pcretest that replaces the length and
448            offset values with spaces. This is now used to make test 3 independent
449            of link size. (Test 2 will be tidied up later.)
450    
451    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
452        segmentation fault instead of returning the appropriate error message.
453    
454    15. In multiline mode when the newline sequence was set to "any", the pattern
455        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
456        This doesn't seem right; it now treats the CRLF combination as the line
457        ending, and so does not match in that case. It's only a pattern such as ^$
458        that would hit this one: something like ^ABC$ would have failed after \r
459        and then tried again after \r\n.
460    
461    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
462        in an attempt to make files that differ only in their line terminators
463        compare equal. This works on Linux.
464    
465    17. Under certain error circumstances pcregrep might try to free random memory
466        as it exited. This is now fixed, thanks to valgrind.
467    
468    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
469        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
470        was because its rules for how to advance for /g after matching an empty
471        string at the end of a line did not allow for this case. They now check for
472        it specially.
473    
474    20. pcretest is supposed to handle patterns and data of any length, by
475        extending its buffers when necessary. It was getting this wrong when the
476        buffer for a data line had to be extended.
477    
478    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
479        CRLF as a newline sequence.
480    
481    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
482        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
483        I have nevertheless tidied it up.
484    
485    23. Added some casts to kill warnings from HP-UX ia64 compiler.
486    
487    24. Added a man page for pcre-config.
488    
489    
490    Version 7.0 19-Dec-06
491    ---------------------
492    
493     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
494        moving to gcc 4.1.1.
495    
496     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
497        sys/time.h, which is documented as needed for this function. It doesn't
498        seem to matter on Linux, but it showed up on some releases of OS X.
499    
500     3. It seems that there are systems where bytes whose values are greater than
501        127 match isprint() in the "C" locale. The "C" locale should be the
502        default when a C program starts up. In most systems, only ASCII printing
503        characters match isprint(). This difference caused the output from pcretest
504        to vary, making some of the tests fail. I have changed pcretest so that:
505    
506        (a) When it is outputting text in the compiled version of a pattern, bytes
507            other than 32-126 are always shown as hex escapes.
508    
509        (b) When it is outputting text that is a matched part of a subject string,
510            it does the same, unless a different locale has been set for the match
511            (using the /L modifier). In this case, it uses isprint() to decide.
512    
513     4. Fixed a major bug that caused incorrect computation of the amount of memory
514        required for a compiled pattern when options that changed within the
515        pattern affected the logic of the preliminary scan that determines the
516        length. The relevant options are -x, and -i in UTF-8 mode. The result was
517        that the computed length was too small. The symptoms of this bug were
518        either the PCRE error "internal error: code overflow" from pcre_compile(),
519        or a glibc crash with a message such as "pcretest: free(): invalid next
520        size (fast)". Examples of patterns that provoked this bug (shown in
521        pcretest format) are:
522    
523          /(?-x: )/x
524          /(?x)(?-x: \s*#\s*)/
525          /((?i)[\x{c0}])/8
526          /(?i:[\x{c0}])/8
527    
528        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
529        is now done differently.
530    
531     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
532        wrapper classes; (b) implement a new function in the C++ scanner that is
533        more efficient than the old way of doing things because it avoids levels of
534        recursion in the regex matching; (c) add a paragraph to the documentation
535        for the FullMatch() function.
536    
537     6. The escape sequence \n was being treated as whatever was defined as
538        "newline". Not only was this contrary to the documentation, which states
539        that \n is character 10 (hex 0A), but it also went horribly wrong when
540        "newline" was defined as CRLF. This has been fixed.
541    
542     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
543        was being set to -1 for the "end of line" case (supposedly a value that no
544        character can have). Though this value is never used (the check for end of
545        line is "zero bytes in current character"), it caused compiler complaints.
546        I've changed it to 0xffffffff.
547    
548     8. In pcre_version.c, the version string was being built by a sequence of
549        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
550        string (as it is for production releases) called a macro with an empty
551        argument. The C standard says the result of this is undefined. The gcc
552        compiler treats it as an empty string (which was what was wanted) but it is
553        reported that Visual C gives an error. The source has been hacked around to
554        avoid this problem.
555    
556     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
557        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
558        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
559        of them did).
560    
561    10. Originally, pcretest opened its input and output without "b"; then I was
562        told that "b" was needed in some environments, so it was added for release
563        5.0 to both the input and output. (It makes no difference on Unix-like
564        systems.) Later I was told that it is wrong for the input on Windows. I've
565        now abstracted the modes into two macros, to make it easier to fiddle with
566        them, and removed "b" from the input mode under Windows.
567    
568    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
569    
570    12. Added -help and --help to pcretest as an official way of being reminded
571        of the options.
572    
573    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
574        and pcrecpp.cc because they annoy compilers at high warning levels.
575    
576    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
577    
578    15. Fixed an occurrence of == in configure.ac that should have been = (shell
579        scripts are not C programs :-) and which was not noticed because it works
580        on Linux.
581    
582    16. pcretest is supposed to handle any length of pattern and data line (as one
583        line or as a continued sequence of lines) by extending its input buffer if
584        necessary. This feature was broken for very long pattern lines, leading to
585        a string of junk being passed to pcre_compile() if the pattern was longer
586        than about 50K.
587    
588    17. I have done a major re-factoring of the way pcre_compile() computes the
589        amount of memory needed for a compiled pattern. Previously, there was code
590        that made a preliminary scan of the pattern in order to do this. That was
591        OK when PCRE was new, but as the facilities have expanded, it has become
592        harder and harder to keep it in step with the real compile phase, and there
593        have been a number of bugs (see for example, 4 above). I have now found a
594        cunning way of running the real compile function in a "fake" mode that
595        enables it to compute how much memory it would need, while actually only
596        ever using a few hundred bytes of working memory and without too many
597        tests of the mode. This should make future maintenance and development
598        easier. A side effect of this work is that the limit of 200 on the nesting
599        depth of parentheses has been removed (though this was never a serious
600        limitation, I suspect). However, there is a downside: pcre_compile() now
601        runs more slowly than before (30% or more, depending on the pattern). I
602        hope this isn't a big issue. There is no effect on runtime performance.
603    
604    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
605        newline (only possible for the last line of a file) and it was a
606        pattern that set a locale (followed by /Lsomething), pcretest crashed.
607    
608    19. Added additional timing features to pcretest. (1) The -tm option now times
609        matching only, not compiling. (2) Both -t and -tm can be followed, as a
610        separate command line item, by a number that specifies the number of
611        repeats to use when timing. The default is 50000; this gives better
612        precision, but takes uncomfortably long for very large patterns.
613    
614    20. Extended pcre_study() to be more clever in cases where a branch of a
615        subpattern has no definite first character. For example, (a*|b*)[cd] would
616        previously give no result from pcre_study(). Now it recognizes that the
617        first character must be a, b, c, or d.
618    
619    21. There was an incorrect error "recursive call could loop indefinitely" if
620        a subpattern (or the entire pattern) that was being tested for matching an
621        empty string contained only one non-empty item after a nested subpattern.
622        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
623        incorrectly, because the \d was being skipped in the check.
624    
625    22. The pcretest program now has a new pattern option /B and a command line
626        option -b, which is equivalent to adding /B to every pattern. This causes
627        it to show the compiled bytecode, without the additional information that
628        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
629        is the same as /B/I).
630    
631    23. A new optimization is now able automatically to treat some sequences such
632        as a*b as a*+b. More specifically, if something simple (such as a character
633        or a simple class like \d) has an unlimited quantifier, and is followed by
634        something that cannot possibly match the quantified thing, the quantifier
635        is automatically "possessified".
636    
637    24. A recursive reference to a subpattern whose number was greater than 39
638        went wrong under certain circumstances in UTF-8 mode. This bug could also
639        have affected the operation of pcre_study().
640    
641    25. Realized that a little bit of performance could be had by replacing
642        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
643    
644    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
645    
646    27. Possessive quantifiers such as a++ were previously implemented by turning
647        them into atomic groups such as ($>a+). Now they have their own opcodes,
648        which improves performance. This includes the automatically created ones
649        from 23 above.
650    
651    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
652        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
653        the first matched character to be a colon. This applied both to named and
654        numbered groups.
655    
656    29. The ucpinternal.h header file was missing its idempotency #ifdef.
657    
658    30. I was sent a "project" file called libpcre.a.dev which I understand makes
659        building PCRE on Windows easier, so I have included it in the distribution.
660    
661    31. There is now a check in pcretest against a ridiculously large number being
662        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
663        loop, the loop is abandoned.
664    
665    32. Forward references to subpatterns in conditions such as (?(2)...) where
666        subpattern 2 is defined later cause pcre_compile() to search forwards in
667        the pattern for the relevant set of parentheses. This search went wrong
668        when there were unescaped parentheses in a character class, parentheses
669        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
670    
671    33. "Subroutine" calls and backreferences were previously restricted to
672        referencing subpatterns earlier in the regex. This restriction has now
673        been removed.
674    
675    34. Added a number of extra features that are going to be in Perl 5.10. On the
676        whole, these are just syntactic alternatives for features that PCRE had
677        previously implemented using the Python syntax or my own invention. The
678        other formats are all retained for compatibility.
679    
680        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
681            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
682            also .NET compatible.
683    
684        (b) A recursion or subroutine call to a named group can now be defined as
685            (?&name) as well as (?P>name).
686    
687        (c) A backreference to a named group can now be defined as \k<name> or
688            \k'name' as well as (?P=name). The new forms, as well as being in Perl
689            5.10, are also .NET compatible.
690    
691        (d) A conditional reference to a named group can now use the syntax
692            (?(<name>) or (?('name') as well as (?(name).
693    
694        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
695            groups (named and numbered) that are never evaluated inline, but can be
696            called as "subroutines" from elsewhere. In effect, the DEFINE condition
697            is always false. There may be only one alternative in such a group.
698    
699        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
700            as the simple (?(R). The condition is true only if the most recent
701            recursion is that of the given number or name. It does not search out
702            through the entire recursion stack.
703    
704        (g) The escape \gN or \g{N} has been added, where N is a positive or
705            negative number, specifying an absolute or relative reference.
706    
707    35. Tidied to get rid of some further signed/unsigned compiler warnings and
708        some "unreachable code" warnings.
709    
710    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
711        things, this adds five new scripts.
712    
713    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
714        There were also incompatibilities regarding the handling of \Q..\E inside
715        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
716        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
717    
718    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
719        matches an empty string, and forcibly breaks the loop. There were bugs in
720        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
721        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
722        separate and independent bugs (that affected different cases) have been
723        fixed.
724    
725    39. Refactored the code to abolish the use of different opcodes for small
726        capturing bracket numbers. This is a tidy that I avoided doing when I
727        removed the limit on the number of capturing brackets for 3.5 back in 2001.
728        The new approach is not only tidier, it makes it possible to reduce the
729        memory needed to fix the previous bug (38).
730    
731    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
732        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
733        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
734        mode.
735    
736    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
737        report.
738    
739    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
740        copy construction and assignment in the C++ wrapper.
741    
742    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
743        couple of bugs that could have given wrong results in the "--newline=crlf"
744        case.
745    
746    44. Added a number of casts and did some reorganization of signed/unsigned int
747        variables following suggestions from Dair Grant. Also renamed the variable
748        "this" as "item" because it is a C++ keyword.
749    
750    45. Arranged for dftables to add
751    
752          #include "pcre_internal.h"
753    
754        to pcre_chartables.c because without it, gcc 4.x may remove the array
755        definition from the final binary if PCRE is built into a static library and
756        dead code stripping is activated.
757    
758    46. For an unanchored pattern, if a match attempt fails at the start of a
759        newline sequence, and the newline setting is CRLF or ANY, and the next two
760        characters are CRLF, advance by two characters instead of one.
761    
762    
763    Version 6.7 04-Jul-06
764    ---------------------
765    
766     1. In order to handle tests when input lines are enormously long, pcretest has
767        been re-factored so that it automatically extends its buffers when
768        necessary. The code is crude, but this _is_ just a test program. The
769        default size has been increased from 32K to 50K.
770    
771     2. The code in pcre_study() was using the value of the re argument before
772        testing it for NULL. (Of course, in any sensible call of the function, it
773        won't be NULL.)
774    
775     3. The memmove() emulation function in pcre_internal.h, which is used on
776        systems that lack both memmove() and bcopy() - that is, hardly ever -
777        was missing a "static" storage class specifier.
778    
779     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
780        containing an extended class (one that cannot be represented by a bitmap
781        because it contains high-valued characters or Unicode property items, e.g.
782        [\pZ]). Almost always one would set UTF-8 mode when processing such a
783        pattern, but PCRE should not loop if you do not (it no longer does).
784        [Detail: two cases were found: (a) a repeated subpattern containing an
785        extended class; (b) a recursive reference to a subpattern that followed a
786        previous extended class. It wasn't skipping over the extended class
787        correctly when UTF-8 mode was not set.]
788    
789     5. A negated single-character class was not being recognized as fixed-length
790        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
791        compile error "lookbehind assertion is not fixed length".
792    
793     6. The RunPerlTest auxiliary script was showing an unexpected difference
794        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
795        write a Perl script that can interpret lines of an input file either as
796        byte characters or as UTF-8, which is what "perltest" was being required to
797        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
798        can't do is switch easily at run time between having the "use utf8;" pragma
799        or not. In the end, I fudged it by using the RunPerlTest script to insert
800        "use utf8;" explicitly for the UTF-8 tests.
801    
802     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
803        the end of the subject string, contrary to the documentation and to what
804        Perl does. This was true of both matching functions. Now it matches only at
805        the start of the subject and immediately after *internal* newlines.
806    
807     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
808        a pointer to an int instead of a pointer to an unsigned long int. This
809        caused problems on 64-bit systems.
810    
811     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
812        instance of the 'standard' template library not being so standard".
813    
814    10. There was no check on the number of named subpatterns nor the maximum
815        length of a subpattern name. The product of these values is used to compute
816        the size of the memory block for a compiled pattern. By supplying a very
817        long subpattern name and a large number of named subpatterns, the size
818        computation could be caused to overflow. This is now prevented by limiting
819        the length of names to 32 characters, and the number of named subpatterns
820        to 10,000.
821    
822    11. Subpatterns that are repeated with specific counts have to be replicated in
823        the compiled pattern. The size of memory for this was computed from the
824        length of the subpattern and the repeat count. The latter is limited to
825        65535, but there was no limit on the former, meaning that integer overflow
826        could in principle occur. The compiled length of a repeated subpattern is
827        now limited to 30,000 bytes in order to prevent this.
828    
829    12. Added the optional facility to have named substrings with the same name.
830    
831    13. Added the ability to use a named substring as a condition, using the
832        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
833        are numbers (not recommended). Forward references are permitted.
834    
835    14. Added forward references in named backreferences (if you see what I mean).
836    
837    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
838        pattern could run off the end of the subject. For example, the pattern
839        "(?s)(.{1,5})"8 did this with the subject "ab".
840    
841    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
842        PCRE_CASELESS was set when matching characters that were quantified with ?
843        or *.
844    
845    17. A character class other than a single negated character that had a minimum
846        but no maximum quantifier - for example [ab]{6,} - was not handled
847        correctly by pce_dfa_exec(). It would match only one character.
848    
849    18. A valid (though odd) pattern that looked like a POSIX character
850        class but used an invalid character after [ (for example [[,abc,]]) caused
851        pcre_compile() to give the error "Failed: internal error: code overflow" or
852        in some cases to crash with a glibc free() error. This could even happen if
853        the pattern terminated after [[ but there just happened to be a sequence of
854        letters, a binary zero, and a closing ] in the memory that followed.
855    
856    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
857        over the years. Originally (before any Unicode support), just the bottom 8
858        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
859        output from "man perlunicode" includes this:
860    
861          The regular expression compiler produces polymorphic opcodes.  That
862          is, the pattern adapts to the data and automatically switches to
863          the Unicode character scheme when presented with Unicode data--or
864          instead uses a traditional byte scheme when presented with byte
865          data.
866    
867        Sadly, a wide octal escape does not cause a switch, and in a string with
868        no other multibyte characters, these octal escapes are treated as before.
869        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
870        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
871        Unicode string.
872    
873        I have not perpetrated such confusion in PCRE. Up till now, it took just
874        the bottom 8 bits, as in old Perl. I have now made octal escapes with
875        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
876        translate to the appropriate multibyte character.
877    
878    29. Applied some refactoring to reduce the number of warnings from Microsoft
879        and Borland compilers. This has included removing the fudge introduced
880        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
881        a warning about an unused variable.
882    
883    21. PCRE has not included VT (character 0x0b) in the set of whitespace
884        characters since release 4.0, because Perl (from release 5.004) does not.
885        [Or at least, is documented not to: some releases seem to be in conflict
886        with the documentation.] However, when a pattern was studied with
887        pcre_study() and all its branches started with \s, PCRE still included VT
888        as a possible starting character. Of course, this did no harm; it just
889        caused an unnecessary match attempt.
890    
891    22. Removed a now-redundant internal flag bit that recorded the fact that case
892        dependency changed within the pattern. This was once needed for "required
893        byte" processing, but is no longer used. This recovers a now-scarce options
894        bit. Also moved the least significant internal flag bit to the most-
895        significant bit of the word, which was not previously used (hangover from
896        the days when it was an int rather than a uint) to free up another bit for
897        the future.
898    
899    23. Added support for CRLF line endings as well as CR and LF. As well as the
900        default being selectable at build time, it can now be changed at runtime
901        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
902        specify that it is scanning data with non-default line endings.
903    
904    24. Changed the definition of CXXLINK to make it agree with the definition of
905        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
906    
907    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
908        recursions. This makes a big different to stack usage for some patterns.
909    
910    26. If a subpattern containing a named recursion or subroutine reference such
911        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
912        the space required for the compiled pattern went wrong and gave too small a
913        value. Depending on the environment, this could lead to "Failed: internal
914        error: code overflow at offset 49" or "glibc detected double free or
915        corruption" errors.
916    
917    27. Applied patches from Google (a) to support the new newline modes and (b) to
918        advance over multibyte UTF-8 characters in GlobalReplace.
919    
920    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
921        difference for some implementation of PCRE in some Windows version.
922    
923    29. Added some extra testing facilities to pcretest:
924    
925        \q<number>   in a data line sets the "match limit" value
926        \Q<number>   in a data line sets the "match recursion limt" value
927        -S <number>  sets the stack size, where <number> is in megabytes
928    
929        The -S option isn't available for Windows.
930    
931    
932    Version 6.6 06-Feb-06
933    ---------------------
934    
935     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
936        in pcreposix.h. I have copied the definition from pcre.h.
937    
938     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
939        because pcre.h is no longer a built file.
940    
941     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
942        not normally included in the compiled code.
943    
944    
945    Version 6.5 01-Feb-06
946    ---------------------
947    
948     1. When using the partial match feature with pcre_dfa_exec(), it was not
949        anchoring the second and subsequent partial matches at the new starting
950        point. This could lead to incorrect results. For example, with the pattern
951        /1234/, partially matching against "123" and then "a4" gave a match.
952    
953     2. Changes to pcregrep:
954    
955        (a) All non-match returns from pcre_exec() were being treated as failures
956            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
957            error message is output. Some extra information is given for the
958            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
959            probably the only errors that are likely to be caused by users (by
960            specifying a regex that has nested indefinite repeats, for instance).
961            If there are more than 20 of these errors, pcregrep is abandoned.
962    
963        (b) A binary zero was treated as data while matching, but terminated the
964            output line if it was written out. This has been fixed: binary zeroes
965            are now no different to any other data bytes.
966    
967        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
968            used to set a locale for matching. The --locale=xxxx long option has
969            been added (no short equivalent) to specify a locale explicitly on the
970            pcregrep command, overriding the environment variables.
971    
972        (d) When -B was used with -n, some line numbers in the output were one less
973            than they should have been.
974    
975        (e) Added the -o (--only-matching) option.
976    
977        (f) If -A or -C was used with -c (count only), some lines of context were
978            accidentally printed for the final match.
979    
980        (g) Added the -H (--with-filename) option.
981    
982        (h) The combination of options -rh failed to suppress file names for files
983            that were found from directory arguments.
984    
985        (i) Added the -D (--devices) and -d (--directories) options.
986    
987        (j) Added the -F (--fixed-strings) option.
988    
989        (k) Allow "-" to be used as a file name for -f as well as for a data file.
990    
991        (l) Added the --colo(u)r option.
992    
993        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
994            is not present by default.
995    
996     3. A nasty bug was discovered in the handling of recursive patterns, that is,
997        items such as (?R) or (?1), when the recursion could match a number of
998        alternatives. If it matched one of the alternatives, but subsequently,
999        outside the recursion, there was a failure, the code tried to back up into
1000        the recursion. However, because of the way PCRE is implemented, this is not
1001        possible, and the result was an incorrect result from the match.
1002    
1003        In order to prevent this happening, the specification of recursion has
1004        been changed so that all such subpatterns are automatically treated as
1005        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1006    
1007     4. I had overlooked the fact that, in some locales, there are characters for
1008        which isalpha() is true but neither isupper() nor islower() are true. In
1009        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1010        and ordfeminine) are like this. This affected the treatment of \w and \W
1011        when they appeared in character classes, but not when they appeared outside
1012        a character class. The bit map for "word" characters is now created
1013        separately from the results of isalnum() instead of just taking it from the
1014        upper, lower, and digit maps. (Plus the underscore character, of course.)
1015    
1016     5. The above bug also affected the handling of POSIX character classes such as
1017        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1018        permanent tables. Instead, the bit maps for such a class were previously
1019        created as the appropriate unions of the upper, lower, and digit bitmaps.
1020        Now they are created by subtraction from the [[:word:]] class, which has
1021        its own bitmap.
1022    
1023     6. The [[:blank:]] character class matches horizontal, but not vertical space.
1024        It is created by subtracting the vertical space characters (\x09, \x0a,
1025        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1026        subtraction was done in the overall bitmap for a character class, meaning
1027        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1028        be recognized. This bug has been fixed.
1029    
1030     7. Patches from the folks at Google:
1031    
1032          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1033          real life, but is still worth protecting against".
1034    
1035          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1036          regular expressions".
1037    
1038          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1039          have it.
1040    
1041          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1042          "configure" and the latter not, in order to fix a problem somebody had
1043          with compiling the Arg class on HP-UX.
1044    
1045          (e) Improve the error-handling of the C++ wrapper a little bit.
1046    
1047          (f) New tests for checking recursion limiting.
1048    
1049     8. The pcre_memmove() function, which is used only if the environment does not
1050        have a standard memmove() function (and is therefore rarely compiled),
1051        contained two bugs: (a) use of int instead of size_t, and (b) it was not
1052        returning a result (though PCRE never actually uses the result).
1053    
1054     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1055        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1056        returned instead of calling malloc() with an overflowing number that would
1057        most likely cause subsequent chaos.
1058    
1059    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1060    
1061    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1062        with this option is matched, the nmatch and pmatch options of regexec() are
1063        ignored.
1064    
1065    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1066        provided in case anyone wants to the the POSIX interface with UTF-8
1067        strings.
1068    
1069    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1070        C++ linking (needed for some HP-UX environments).
1071    
1072    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1073        (unused parameter) and in the pcre_printint() function (omitted "default"
1074        switch label when the default is to do nothing).
1075    
1076    15. Added some code to make it possible, when PCRE is compiled as a C++
1077        library, to replace subject pointers for pcre_exec() with a smart pointer
1078        class, thus making it possible to process discontinuous strings.
1079    
1080    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1081        much the same function. They were added by different people who were trying
1082        to make PCRE easy to compile on non-Unix systems. It has been suggested
1083        that PCRE_EXPORT be abolished now that there is more automatic apparatus
1084        for compiling on Windows systems. I have therefore replaced it with
1085        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1086        defaults to "extern" for C or "extern C" for C++, which works fine on
1087        Unix-like systems. It is now possible to override the value of PCRE_DATA_
1088        SCOPE with something explicit in config.h. In addition:
1089    
1090        (a) pcreposix.h still had just "extern" instead of either of these macros;
1091            I have replaced it with PCRE_DATA_SCOPE.
1092    
1093        (b) Functions such as _pcre_xclass(), which are internal to the library,
1094            but external in the C sense, all had PCRE_EXPORT in their definitions.
1095            This is apparently wrong for the Windows case, so I have removed it.
1096            (It makes no difference on Unix-like systems.)
1097    
1098    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1099        of recursive calls to match(). This is different to MATCH_LIMIT because
1100        that limits the total number of calls to match(), not all of which increase
1101        the depth of recursion. Limiting the recursion depth limits the amount of
1102        stack (or heap if NO_RECURSE is set) that is used. The default can be set
1103        when PCRE is compiled, and changed at run time. A patch from Google adds
1104        this functionality to the C++ interface.
1105    
1106    18. Changes to the handling of Unicode character properties:
1107    
1108        (a) Updated the table to Unicode 4.1.0.
1109    
1110        (b) Recognize characters that are not in the table as "Cn" (undefined).
1111    
1112        (c) I revised the way the table is implemented to a much improved format
1113            which includes recognition of ranges. It now supports the ranges that
1114            are defined in UnicodeData.txt, and it also amalgamates other
1115            characters into ranges. This has reduced the number of entries in the
1116            table from around 16,000 to around 3,000, thus reducing its size
1117            considerably. I realized I did not need to use a tree structure after
1118            all - a binary chop search is just as efficient. Having reduced the
1119            number of entries, I extended their size from 6 bytes to 8 bytes to
1120            allow for more data.
1121    
1122        (d) Added support for Unicode script names via properties such as \p{Han}.
1123    
1124    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1125        matching that character.
1126    
1127    20. When matching a repeated Unicode property with a minimum greater than zero,
1128        (for example \pL{2,}), PCRE could look past the end of the subject if it
1129        reached it while seeking the minimum number of characters. This could
1130        happen only if some of the characters were more than one byte long, because
1131        there is a check for at least the minimum number of bytes.
1132    
1133    21. Refactored the implementation of \p and \P so as to be more general, to
1134        allow for more different types of property in future. This has changed the
1135        compiled form incompatibly. Anybody with saved compiled patterns that use
1136        \p or \P will have to recompile them.
1137    
1138    22. Added "Any" and "L&" to the supported property types.
1139    
1140    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1141        but give a compile time error if the value is greater than 0xff.
1142    
1143    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1144        accidentally not being installed or uninstalled.
1145    
1146    25. The pcre.h file was built from pcre.h.in, but the only changes that were
1147        made were to insert the current release number. This seemed silly, because
1148        it made things harder for people building PCRE on systems that don't run
1149        "configure". I have turned pcre.h into a distributed file, no longer built
1150        by "configure", with the version identification directly included. There is
1151        no longer a pcre.h.in file.
1152    
1153        However, this change necessitated a change to the pcre-config script as
1154        well. It is built from pcre-config.in, and one of the substitutions was the
1155        release number. I have updated configure.ac so that ./configure now finds
1156        the release number by grepping pcre.h.
1157    
1158    26. Added the ability to run the tests under valgrind.
1159    
1160    
1161    Version 6.4 05-Sep-05
1162    ---------------------
1163    
1164     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1165        "--" to be printed when multiple files were scanned, even when none of the
1166        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1167        consider it to be a bug, and have restored the previous behaviour.
1168    
1169     2. A couple of code tidies to get rid of compiler warnings.
1170    
1171     3. The pcretest program used to cheat by referring to symbols in the library
1172        whose names begin with _pcre_. These are internal symbols that are not
1173        really supposed to be visible externally, and in some environments it is
1174        possible to suppress them. The cheating is now confined to including
1175        certain files from the library's source, which is a bit cleaner.
1176    
1177     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1178        file's purpose clearer.
1179    
1180     5. Reorganized pcre_ucp_findchar().
1181    
1182    
1183    Version 6.3 15-Aug-05
1184    ---------------------
1185    
1186     1. The file libpcre.pc.in did not have general read permission in the tarball.
1187    
1188     2. There were some problems when building without C++ support:
1189    
1190        (a) If C++ support was not built, "make install" and "make test" still
1191            tried to test it.
1192    
1193        (b) There were problems when the value of CXX was explicitly set. Some
1194            changes have been made to try to fix these, and ...
1195    
1196        (c) --disable-cpp can now be used to explicitly disable C++ support.
1197    
1198        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1199            backslash in a target when C++ was disabled. This confuses some
1200            versions of "make", apparently. Using an intermediate variable solves
1201            this. (Same for CPP_LOBJ.)
1202    
1203     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1204        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1205        necessary on certain architectures.
1206    
1207     4. Added a setting of -export-symbols-regex to the link command to remove
1208        those symbols that are exported in the C sense, but actually are local
1209        within the library, and not documented. Their names all begin with
1210        "_pcre_". This is not a perfect job, because (a) we have to except some
1211        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1212        available (and never for static libraries). I have made a note to try to
1213        find a way round (a) in the future.
1214    
1215    
1216    Version 6.2 01-Aug-05
1217    ---------------------
1218    
1219     1. There was no test for integer overflow of quantifier values. A construction
1220        such as {1111111111111111} would give undefined results. What is worse, if
1221        a minimum quantifier for a parenthesized subpattern overflowed and became
1222        negative, the calculation of the memory size went wrong. This could have
1223        led to memory overwriting.
1224    
1225     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1226    
1227     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1228        operating environments where this matters.
1229    
1230     4. Applied Giuseppe Maxia's patch to add additional features for controlling
1231        PCRE options from within the C++ wrapper.
1232    
1233     5. Named capturing subpatterns were not being correctly counted when a pattern
1234        was compiled. This caused two problems: (a) If there were more than 100
1235        such subpatterns, the calculation of the memory needed for the whole
1236        compiled pattern went wrong, leading to an overflow error. (b) Numerical
1237        back references of the form \12, where the number was greater than 9, were
1238        not recognized as back references, even though there were sufficient
1239        previous subpatterns.
1240    
1241     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1242        versions of gcc, e.g. 2.95.4.
1243    
1244    
1245    Version 6.1 21-Jun-05
1246    ---------------------
1247    
1248     1. There was one reference to the variable "posix" in pcretest.c that was not
1249        surrounded by "#if !defined NOPOSIX".
1250    
1251     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1252        the cross-check on the old pcre_info() function, for the benefit of the
1253        cut-down version of PCRE that is currently imported into Exim.
1254    
1255     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1256        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1257        patterns that start (?i) but I don't think that matters. The use of (?i) is
1258        just an example; this all applies to the other options as well.
1259    
1260     4. Since libtool seems to echo the compile commands it is issuing, the output
1261        from "make" can be reduced a bit by putting "@" in front of each libtool
1262        compile command.
1263    
1264     5. Patch from the folks at Google for configure.in to be a bit more thorough
1265        in checking for a suitable C++ installation before trying to compile the
1266        C++ stuff. This should fix a reported problem when a compiler was present,
1267        but no suitable headers.
1268    
1269     6. The man pages all had just "PCRE" as their title. I have changed them to
1270        be the relevant file name. I have also arranged that these names are
1271        retained in the file doc/pcre.txt, which is a concatenation in text format
1272        of all the man pages except the little individual ones for each function.
1273    
1274     7. The NON-UNIX-USE file had not been updated for the different set of source
1275        files that come with release 6. I also added a few comments about the C++
1276        wrapper.
1277    
1278    
1279    Version 6.0 07-Jun-05
1280    ---------------------
1281    
1282     1. Some minor internal re-organization to help with my DFA experiments.
1283    
1284     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1285        didn't matter for the library itself when fully configured, but did matter
1286        when compiling without UCP support, or within Exim, where the ucp files are
1287        not imported.
1288    
1289     3. Refactoring of the library code to split up the various functions into
1290        different source modules. The addition of the new DFA matching code (see
1291        below) to a single monolithic source would have made it really too
1292        unwieldy, quite apart from causing all the code to be include in a
1293        statically linked application, when only some functions are used. This is
1294        relevant even without the DFA addition now that patterns can be compiled in
1295        one application and matched in another.
1296    
1297        The downside of splitting up is that there have to be some external
1298        functions and data tables that are used internally in different modules of
1299        the library but which are not part of the API. These have all had their
1300        names changed to start with "_pcre_" so that they are unlikely to clash
1301        with other external names.
1302    
1303     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1304        a different (DFA) algorithm. Although it is slower than the original
1305        function, it does have some advantages for certain types of matching
1306        problem.
1307    
1308     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1309        including restarting after a partial match.
1310    
1311     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1312        defined when compiling for Windows was sent to me. I have put it into the
1313        code, though I have no means of testing or verifying it.
1314    
1315     7. Added the pcre_refcount() auxiliary function.
1316    
1317     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1318        match before or at the first newline in the subject string. In pcretest,
1319        the /f option on a pattern can be used to set this.
1320    
1321     9. A repeated \w when used in UTF-8 mode with characters greater than 256
1322        would behave wrongly. This has been present in PCRE since release 4.0.
1323    
1324    10. A number of changes to the pcregrep command:
1325    
1326        (a) Refactored how -x works; insert ^(...)$ instead of setting
1327            PCRE_ANCHORED and checking the length, in preparation for adding
1328            something similar for -w.
1329    
1330        (b) Added the -w (match as a word) option.
1331    
1332        (c) Refactored the way lines are read and buffered so as to have more
1333            than one at a time available.
1334    
1335        (d) Implemented a pcregrep test script.
1336    
1337        (e) Added the -M (multiline match) option. This allows patterns to match
1338            over several lines of the subject. The buffering ensures that at least
1339            8K, or the rest of the document (whichever is the shorter) is available
1340            for matching (and similarly the previous 8K for lookbehind assertions).
1341    
1342        (f) Changed the --help output so that it now says
1343    
1344              -w, --word-regex(p)
1345    
1346            instead of two lines, one with "regex" and the other with "regexp"
1347            because that confused at least one person since the short forms are the
1348            same. (This required a bit of code, as the output is generated
1349            automatically from a table. It wasn't just a text change.)
1350    
1351        (g) -- can be used to terminate pcregrep options if the next thing isn't an
1352            option but starts with a hyphen. Could be a pattern or a path name
1353            starting with a hyphen, for instance.
1354    
1355        (h) "-" can be given as a file name to represent stdin.
1356    
1357        (i) When file names are being printed, "(standard input)" is used for
1358            the standard input, for compatibility with GNU grep. Previously
1359            "<stdin>" was used.
1360    
1361        (j) The option --label=xxx can be used to supply a name to be used for
1362            stdin when file names are being printed. There is no short form.
1363    
1364        (k) Re-factored the options decoding logic because we are going to add
1365            two more options that take data. Such options can now be given in four
1366            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1367    
1368        (l) Added the -A, -B, and -C options for requesting that lines of context
1369            around matches be printed.
1370    
1371        (m) Added the -L option to print the names of files that do not contain
1372            any matching lines, that is, the complement of -l.
1373    
1374        (n) The return code is 2 if any file cannot be opened, but pcregrep does
1375            continue to scan other files.
1376    
1377        (o) The -s option was incorrectly implemented. For compatibility with other
1378            greps, it now suppresses the error message for a non-existent or non-
1379            accessible file (but not the return code). There is a new option called
1380            -q that suppresses the output of matching lines, which was what -s was
1381            previously doing.
1382    
1383        (p) Added --include and --exclude options to specify files for inclusion
1384            and exclusion when recursing.
1385    
1386    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1387        Hopefully, it now does.
1388    
1389    12. Missing cast in pcre_study().
1390    
1391    13. Added an "uninstall" target to the makefile.
1392    
1393    14. Replaced "extern" in the function prototypes in Makefile.in with
1394        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1395        world, but is set differently for Windows.
1396    
1397    15. Added a second compiling function called pcre_compile2(). The only
1398        difference is that it has an extra argument, which is a pointer to an
1399        integer error code. When there is a compile-time failure, this is set
1400        non-zero, in addition to the error test pointer being set to point to an
1401        error message. The new argument may be NULL if no error number is required
1402        (but then you may as well call pcre_compile(), which is now just a
1403        wrapper). This facility is provided because some applications need a
1404        numeric error indication, but it has also enabled me to tidy up the way
1405        compile-time errors are handled in the POSIX wrapper.
1406    
1407    16. Added VPATH=.libs to the makefile; this should help when building with one
1408        prefix path and installing with another. (Or so I'm told by someone who
1409        knows more about this stuff than I do.)
1410    
1411    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1412        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1413        match everything, including newlines. This is not POSIX-compatible, but
1414        somebody wanted the feature. From pcretest it can be activated by using
1415        both the P and the s flags.
1416    
1417    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1418    
1419    19. libpcre.pc was being incorrectly installed as executable.
1420    
1421    20. A couple of places in pcretest check for end-of-line by looking for '\n';
1422        it now also looks for '\r' so that it will work unmodified on Windows.
1423    
1424    21. Added Google's contributed C++ wrapper to the distribution.
1425    
1426    22. Added some untidy missing memory free() calls in pcretest, to keep
1427        Electric Fence happy when testing.
1428    
1429    
1430    
1431    Version 5.0 13-Sep-04
1432    ---------------------
1433    
1434     1. Internal change: literal characters are no longer packed up into items
1435        containing multiple characters in a single byte-string. Each character
1436        is now matched using a separate opcode. However, there may be more than one
1437        byte in the character in UTF-8 mode.
1438    
1439     2. The pcre_callout_block structure has two new fields: pattern_position and
1440        next_item_length. These contain the offset in the pattern to the next match
1441        item, and its length, respectively.
1442    
1443     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1444        insertion of callouts before each pattern item. Added the /C option to
1445        pcretest to make use of this.
1446    
1447     4. On the advice of a Windows user, the lines
1448    
1449          #if defined(_WIN32) || defined(WIN32)
1450          _setmode( _fileno( stdout ), 0x8000 );
1451          #endif  /* defined(_WIN32) || defined(WIN32) */
1452    
1453        have been added to the source of pcretest. This apparently does useful
1454        magic in relation to line terminators.
1455    
1456     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1457        for the benefit of those environments where the "b" makes a difference.
1458    
1459     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1460        to know about it. I have put a hack into configure.in that adds in code
1461        to set GCC=yes if CC=icc. This seems to end up at a point in the
1462        generated configure script that is early enough to affect the setting of
1463        compiler options, which is what is needed, but I have no means of testing
1464        whether it really works. (The user who reported this had patched the
1465        generated configure script, which of course I cannot do.)
1466    
1467        LATER: After change 22 below (new libtool files), the configure script
1468        seems to know about icc (and also ecc). Therefore, I have commented out
1469        this hack in configure.in.
1470    
1471     7. Added support for pkg-config (2 patches were sent in).
1472    
1473     8. Negated POSIX character classes that used a combination of internal tables
1474        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1475        [[:^ascii]]. Typically, they would match almost any characters. The other
1476        POSIX classes were not broken in this way.
1477    
1478     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1479        to find the match, as PCRE was deluded into thinking that the match had to
1480        start at the start point or following a newline. The same bug applied to
1481        patterns with negative forward assertions or any backward assertions
1482        preceding ".*" at the start, unless the pattern required a fixed first
1483        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1484    
1485    10. In UTF-8 mode, when moving forwards in the subject after a failed match
1486        starting at the last subject character, bytes beyond the end of the subject
1487        string were read.
1488    
1489    11. Renamed the variable "class" as "classbits" to make life easier for C++
1490        users. (Previously there was a macro definition, but it apparently wasn't
1491        enough.)
1492    
1493    12. Added the new field "tables" to the extra data so that tables can be passed
1494        in at exec time, or the internal tables can be re-selected. This allows
1495        a compiled regex to be saved and re-used at a later time by a different
1496        program that might have everything at different addresses.
1497    
1498    13. Modified the pcre-config script so that, when run on Solaris, it shows a
1499        -R library as well as a -L library.
1500    
1501    14. The debugging options of pcretest (-d on the command line or D on a
1502        pattern) showed incorrect output for anything following an extended class
1503        that contained multibyte characters and which was followed by a quantifier.
1504    
1505    15. Added optional support for general category Unicode character properties
1506        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1507        support. It adds about 90K to the size of the library. The meanings of the
1508        inbuilt class escapes such as \d and \s have NOT been changed.
1509    
1510    16. Updated pcredemo.c to include calls to free() to release the memory for the
1511        compiled pattern.
1512    
1513    17. The generated file chartables.c was being created in the source directory
1514        instead of in the building directory. This caused the build to fail if the
1515        source directory was different from the building directory, and was
1516        read-only.
1517    
1518    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1519        file. No doubt somebody will tell me if they don't make sense... Also added
1520        Dan Mooney's comments about building on OpenVMS.
1521    
1522    19. Added support for partial matching via the PCRE_PARTIAL option for
1523        pcre_exec() and the \P data escape in pcretest.
1524    
1525    20. Extended pcretest with 3 new pattern features:
1526    
1527        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
1528              write the compiled pattern to the file whose name is "rest-of-line".
1529              This is a straight binary dump of the data, with the saved pointer to
1530              the character tables forced to be NULL. The study data, if any, is
1531              written too. After writing, pcretest reads a new pattern.
1532    
1533        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1534              compiled pattern from the given file. There must not be any
1535              occurrences of "<" in the file name (pretty unlikely); if there are,
1536              pcretest will instead treat the initial "<" as a pattern delimiter.
1537              After reading in the pattern, pcretest goes on to read data lines as
1538              usual.
1539    
1540        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1541              and 16-bit fields in a compiled pattern, to simulate a pattern that
1542              was compiled on a host of opposite endianness.
1543    
1544    21. The pcre-exec() function can now cope with patterns that were compiled on
1545        hosts of opposite endianness, with this restriction:
1546    
1547          As for any compiled expression that is saved and used later, the tables
1548          pointer field cannot be preserved; the extra_data field in the arguments
1549          to pcre_exec() should be used to pass in a tables address if a value
1550          other than the default internal tables were used at compile time.
1551    
1552    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1553        now diagnosed as an error. Previously, most of the time, a negative number
1554        would have been treated as zero, but if in addition "ovector" was passed as
1555        NULL, a crash could occur.
1556    
1557    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1558        new versions from the libtool 1.5 distribution (the last one is a copy of
1559        a file called libtool.m4). This seems to have fixed the need to patch
1560        "configure" to support Darwin 1.3 (which I used to do). However, I still
1561        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1562        workstation).
1563    
1564    24. Changed the PCRE licence to be the more standard "BSD" licence.
1565    
1566    
1567    Version 4.5 01-Dec-03
1568    ---------------------
1569    
1570     1. There has been some re-arrangement of the code for the match() function so
1571        that it can be compiled in a version that does not call itself recursively.
1572        Instead, it keeps those local variables that need separate instances for
1573        each "recursion" in a frame on the heap, and gets/frees frames whenever it
1574        needs to "recurse". Keeping track of where control must go is done by means
1575        of setjmp/longjmp. The whole thing is implemented by a set of macros that
1576        hide most of the details from the main code, and operates only if
1577        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1578        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1579        operating.
1580    
1581        To make it easier for callers to provide specially tailored get/free
1582        functions for this usage, two new functions, pcre_stack_malloc, and
1583        pcre_stack_free, are used. They are always called in strict stacking order,
1584        and the size of block requested is always the same.
1585    
1586        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1587        PCRE has been compiled to use the stack or the heap for recursion. The
1588        -C option of pcretest uses this to show which version is compiled.
1589    
1590        A new data escape \S, is added to pcretest; it causes the amounts of store
1591        obtained and freed by both kinds of malloc/free at match time to be added
1592        to the output.
1593    
1594     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1595        what's available on my current Linux desktop machine.
1596    
1597     3. When matching a UTF-8 string, the test for a valid string at the start has
1598        been extended. If start_offset is not zero, PCRE now checks that it points
1599        to a byte that is the start of a UTF-8 character. If not, it returns
1600        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1601        this is necessary because there may be backward assertions in the pattern.
1602        When matching the same subject several times, it may save resources to use
1603        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1604    
1605     4. The code for checking the validity of UTF-8 strings has been tightened so
1606        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1607        containing "overlong sequences".
1608    
1609     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1610        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1611        should have been "&", but it just so happened that all the cases this let
1612        through by mistake were picked up later in the function.
1613    
1614     6. I had used a variable called "isblank" - this is a C99 function, causing
1615        some compilers to warn. To avoid this, I renamed it (as "blankclass").
1616    
1617     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1618        prompting; (b) run "./pcretest /dev/null" at the start of the test script
1619        so the version is shown; (c) stop "make test" echoing "./RunTest".
1620    
1621     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1622    
1623     9. The prototype for memmove() for systems that don't have it was using
1624        size_t, but the inclusion of the header that defines size_t was later. I've
1625        moved the #includes for the C headers earlier to avoid this.
1626    
1627    10. Added some adjustments to the code to make it easier to compiler on certain
1628        special systems:
1629    
1630          (a) Some "const" qualifiers were missing.
1631          (b) Added the macro EXPORT before all exported functions; by default this
1632              is defined to be empty.
1633          (c) Changed the dftables auxiliary program (that builds chartables.c) so
1634              that it reads its output file name as an argument instead of writing
1635              to the standard output and assuming this can be redirected.
1636    
1637    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
1638        class containing characters with values greater than 255, PCRE compilation
1639        went into a loop.
1640    
1641    12. A recursive reference to a subpattern that was within another subpattern
1642        that had a minimum quantifier of zero caused PCRE to crash. For example,
1643        (x(y(?2))z)? provoked this bug with a subject that got as far as the
1644        recursion. If the recursively-called subpattern itself had a zero repeat,
1645        that was OK.
1646    
1647    13. In pcretest, the buffer for reading a data line was set at 30K, but the
1648        buffer into which it was copied (for escape processing) was still set at
1649        1024, so long lines caused crashes.
1650    
1651    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
1652        "internal error: code overflow...". This applied to any character class
1653        that was followed by a possessive quantifier.
1654    
1655    15. Modified the Makefile to add libpcre.la as a prerequisite for
1656        libpcreposix.la because I was told this is needed for a parallel build to
1657        work.
1658    
1659    16. If a pattern that contained .* following optional items at the start was
1660        studied, the wrong optimizing data was generated, leading to matching
1661        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
1662        matching string must start with a or b or c. The correct conclusion for
1663        this pattern is that a match can start with any character.
1664    
1665    
1666    Version 4.4 13-Aug-03
1667    ---------------------
1668    
1669     1. In UTF-8 mode, a character class containing characters with values between
1670        127 and 255 was not handled correctly if the compiled pattern was studied.
1671        In fixing this, I have also improved the studying algorithm for such
1672        classes (slightly).
1673    
1674     2. Three internal functions had redundant arguments passed to them. Removal
1675        might give a very teeny performance improvement.
1676    
1677     3. Documentation bug: the value of the capture_top field in a callout is *one
1678        more than* the number of the hightest numbered captured substring.
1679    
1680     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
1681        in incorrectly linking with a previously installed version. They now link
1682        explicitly with libpcre.la.
1683    
1684     5. configure.in no longer needs to recognize Cygwin specially.
1685    
1686     6. A problem in pcre.in for Windows platforms is fixed.
1687    
1688     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
1689        pcretest, it used to include the size of the study block as part of its
1690        output. Unfortunately, the structure contains a field that has a different
1691        size on different hardware architectures. This meant that the tests that
1692        showed this size failed. As the block is currently always of a fixed size,
1693        this information isn't actually particularly useful in pcretest output, so
1694        I have just removed it.
1695    
1696     8. Three pre-processor statements accidentally did not start in column 1.
1697        Sadly, there are *still* compilers around that complain, even though
1698        standard C has not required this for well over a decade. Sigh.
1699    
1700     9. In pcretest, the code for checking callouts passed small integers in the
1701        callout_data field, which is a void * field. However, some picky compilers
1702        complained about the casts involved for this on 64-bit systems. Now
1703        pcretest passes the address of the small integer instead, which should get
1704        rid of the warnings.
1705    
1706    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
1707        both compile and run time, and gives an error if an invalid UTF-8 sequence
1708        is found. There is a option for disabling this check in cases where the
1709        string is known to be correct and/or the maximum performance is wanted.
1710    
1711    11. In response to a bug report, I changed one line in Makefile.in from
1712    
1713            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
1714        to
1715            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
1716    
1717        to look similar to other lines, but I have no way of telling whether this
1718        is the right thing to do, as I do not use Windows. No doubt I'll get told
1719        if it's wrong...
1720    
1721    
1722    Version 4.3 21-May-03
1723    ---------------------
1724    
1725    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
1726       Makefile.
1727    
1728    2. Some refactoring to improve the quality of the code:
1729    
1730       (i)   The utf8_table... variables are now declared "const".
1731    
1732       (ii)  The code for \cx, which used the "case flipping" table to upper case
1733             lower case letters, now just substracts 32. This is ASCII-specific,
1734             but the whole concept of \cx is ASCII-specific, so it seems
1735             reasonable.
1736    
1737       (iii) PCRE was using its character types table to recognize decimal and
1738             hexadecimal digits in the pattern. This is silly, because it handles
1739             only 0-9, a-f, and A-F, but the character types table is locale-
1740             specific, which means strange things might happen. A private
1741             table is now used for this - though it costs 256 bytes, a table is
1742             much faster than multiple explicit tests. Of course, the standard
1743             character types table is still used for matching digits in subject
1744             strings against \d.
1745    
1746       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
1747             ending in _t are). So I've renamed it as ESC_tee.
1748    
1749    3. The first argument for regexec() in the POSIX wrapper should have been
1750       defined as "const".
1751    
1752    4. Changed pcretest to use malloc() for its buffers so that they can be
1753       Electric Fenced for debugging.
1754    
1755    5. There were several places in the code where, in UTF-8 mode, PCRE would try
1756       to read one or more bytes before the start of the subject string. Often this
1757       had no effect on PCRE's behaviour, but in some circumstances it could
1758       provoke a segmentation fault.
1759    
1760    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
1761       to try to read one or more bytes before the start of the subject string.
1762    
1763    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
1764       UTF-8 support could misbehave in various ways if the subject string
1765       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
1766       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
1767       back over UTF-8 characters.)
1768    
1769    
1770    Version 4.2 14-Apr-03
1771    ---------------------
1772    
1773    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
1774    
1775    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
1776         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
1777         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
1778         [WIN_PREFIX]: new variable, "cyg" for Cygwin
1779         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
1780           and BUILD_EXEEXT
1781         Note: automatic setting of the BUILD variables is not yet working
1782         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
1783           compile-time but not at link-time
1784         [LINK]: use for linking executables only
1785         make different versions for Windows and non-Windows
1786         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
1787           libraries
1788         [LINK_FOR_BUILD]: new variable
1789         [OBJEXT]: use throughout
1790         [EXEEXT]: use throughout
1791         <winshared>: new target
1792         <wininstall>: new target
1793         <dftables.o>: use native compiler
1794         <dftables>: use native linker
1795         <install>: handle Windows platform correctly
1796         <clean>: ditto
1797         <check>: ditto
1798         copy DLL to top builddir before testing
1799    
1800       As part of these changes, -no-undefined was removed again. This was reported
1801       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
1802       in any case.
1803    
1804    3. Some tidies to get rid of compiler warnings:
1805    
1806       . In the match_data structure, match_limit was an unsigned long int, whereas
1807         match_call_count was an int. I've made them both unsigned long ints.
1808    
1809       . In pcretest the fact that a const uschar * doesn't automatically cast to
1810         a void * provoked a warning.
1811    
1812       . Turning on some more compiler warnings threw up some "shadow" variables
1813         and a few more missing casts.
1814    
1815    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
1816       option, a class that contained a single character with a value between 128
1817       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
1818    
1819    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
1820       option, a class that contained several characters, but with at least one
1821       whose value was between 128 and 255 caused PCRE to crash.
1822    
1823    
1824    Version 4.1 12-Mar-03
1825    ---------------------
1826    
1827    1. Compiling with gcc -pedantic found a couple of places where casts were
1828    needed, and a string in dftables.c that was longer than standard compilers are
1829    required to support.
1830    
1831    2. Compiling with Sun's compiler found a few more places where the code could
1832    be tidied up in order to avoid warnings.
1833    
1834    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
1835    first of these names is deprecated in the latest Autoconf in favour of the name
1836    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
1837    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
1838    analogy I have changed it to CFLAGS_FOR_BUILD.
1839    
1840    4. Added -no-undefined to the linking command in the Makefile, because this is
1841    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
1842    linking step for the pcreposix library.
1843    
1844    5. PCRE was failing to diagnose the case of two named groups with the same
1845    name.
1846    
1847    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
1848    literal character that is needed in the subject for a match, and scans along to
1849    ensure that it is present before embarking on the full matching process. This
1850    saves time in cases of nested unlimited repeats that are never going to match.
1851    Problem: the scan can take a lot of time if the subject is very long (e.g.
1852    megabytes), thus penalizing straightforward matches. It is now done only if the
1853    amount of subject to be scanned is less than 1000 bytes.
1854    
1855    7. A lesser problem with the same optimization is that it was recording the
1856    first character of an anchored pattern as "needed", thus provoking a search
1857    right along the subject, even when the first match of the pattern was going to
1858    fail. The "needed" character is now not set for anchored patterns, unless it
1859    follows something in the pattern that is of non-fixed length. Thus, it still
1860    fulfils its original purpose of finding quick non-matches in cases of nested
1861    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
1862    
1863    
1864    Version 4.0 17-Feb-03
1865    ---------------------
1866    
1867    1. If a comment in an extended regex that started immediately after a meta-item
1868    extended to the end of string, PCRE compiled incorrect data. This could lead to
1869    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
1870    
1871    2. Moved to autoconf 2.53 and libtool 1.4.2.
1872    
1873    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
1874    the special perltest8 script is no longer needed - all the tests can be run
1875    from a single perltest script.
1876    
1877    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
1878    by \s. It has now been removed in PCRE. This means it isn't recognized as
1879    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
1880    class [:space:] *does* include VT, thereby creating a mess.
1881    
1882    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
1883    space and tab.
1884    
1885    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
1886    its new features into the main test script, reducing the number of scripts.
1887    
1888    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
1889    were backward compatible, and made the (?i) apply to the whole pattern, as if
1890    /i were given. Now it behaves more logically, and applies the option setting
1891    only to what follows. PCRE has been changed to follow suit. However, if it
1892    finds options settings right at the start of the pattern, it extracts them into
1893    the global options, as before. Thus, they show up in the info data.
1894    
1895    8. Added support for the \Q...\E escape sequence. Characters in between are
1896    treated as literals. This is slightly different from Perl in that $ and @ are
1897    also handled as literals inside the quotes. In Perl, they will cause variable
1898    interpolation. Note the following examples:
1899    
1900        Pattern            PCRE matches      Perl matches
1901    
1902        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
1903        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
1904        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
1905    
1906    For compatibility with Perl, \Q...\E sequences are recognized inside character
1907    classes as well as outside them.
1908    
1909    9. Re-organized 3 code statements in pcretest to avoid "overflow in
1910    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
1911    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
1912    signed/unsigned warnings.
1913    
1914    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
1915    option for pcretest, so I've replaced it by a simple function that does just
1916    that job.
1917    
1918    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
1919    "pcregrep -".
1920    
1921    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
1922    Java package. This provides some syntactic sugar for simple cases of what my
1923    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
1924    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
1925    item, you can use this simplified notation. Note that only makes sense with
1926    greedy quantifiers. Consequently, the use of the possessive quantifier forces
1927    greediness, whatever the setting of the PCRE_UNGREEDY option.
1928    
1929    13. A change of greediness default within a pattern was not taking effect at
1930    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
1931    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
1932    was abstracted outside.
1933    
1934    14. PCRE now supports the \G assertion. It is true when the current matching
1935    position is at the start point of the match. This differs from \A when the
1936    starting offset is non-zero. Used with the /g option of pcretest (or similar
1937    code), it works in the same way as it does for Perl's /g option. If all
1938    alternatives of a regex begin with \G, the expression is anchored to the start
1939    match position, and the "anchored" flag is set in the compiled expression.
1940    
1941    15. Some bugs concerning the handling of certain option changes within patterns
1942    have been fixed. These applied to options other than (?ims). For example,
1943    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
1944    been the other way round. Some of this was related to change 7 above.
1945    
1946    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
1947    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
1948    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
1949    POSIX classes only within a class (e.g. /[[:alpha:]]/).
1950    
1951    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
1952    mode. Unlike ".", it always matches newline, whatever the setting of
1953    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
1954    assertions. Perl allows it, but it doesn't (in general) work because it can't
1955    calculate the length of the lookbehind. At least, that's the case for Perl
1956    5.8.0 - I've been told they are going to document that it doesn't work in
1957    future.
1958    
1959    18. Added an error diagnosis for escapes that PCRE does not support: these are
1960    \L, \l, \N, \P, \p, \U, \u, and \X.
1961    
1962    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
1963    reading past the end of the pattern in cases such as /[abcd/.
1964    
1965    20. PCRE was getting more memory than necessary for patterns with classes that
1966    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
1967    
1968    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
1969    compiling PCRE for use with Virtual Pascal.
1970    
1971    22. Small fix to the Makefile to make it work properly if the build is done
1972    outside the source tree.
1973    
1974    23. Added a new extension: a condition to go with recursion. If a conditional
1975    subpattern starts with (?(R) the "true" branch is used if recursion has
1976    happened, whereas the "false" branch is used only at the top level.
1977    
1978    24. When there was a very long string of literal characters (over 255 bytes
1979    without UTF support, over 250 bytes with UTF support), the computation of how
1980    much memory was required could be incorrect, leading to segfaults or other
1981    strange effects.
1982    
1983    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
1984    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
1985    there was a subsequent back reference to those brackets. This meant that, for
1986    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
1987    possible to check for precisely this case. All we can do is abandon the
1988    optimization if .* occurs inside capturing brackets when there are any back
1989    references whatsoever. (See below for a better fix that came later.)
1990    
1991    26. The handling of the optimization for finding the first character of a
1992    non-anchored pattern, and for finding a character that is required later in the
1993    match were failing in some cases. This didn't break the matching; it just
1994    failed to optimize when it could. The way this is done has been re-implemented.
1995    
1996    27. Fixed typo in error message for invalid (?R item (it said "(?p").
1997    
1998    28. Added a new feature that provides some of the functionality that Perl
1999    provides with (?{...}). The facility is termed a "callout". The way it is done
2000    in PCRE is for the caller to provide an optional function, by setting
2001    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
2002    global variable. By default it is unset, which disables all calling out. To get
2003    the function called, the regex must include (?C) at appropriate points. This
2004    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
2005    This provides a means of identifying different callout points. When PCRE
2006    reaches such a point in the regex, if pcre_callout has been set, the external
2007    function is called. It is provided with data in a structure called
2008    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
2009    matching continues; if it returns a non-zero value, the match at the current
2010    point fails. However, backtracking will occur if possible. [This was changed
2011    later and other features added - see item 49 below.]
2012    
2013    29. pcretest is upgraded to test the callout functionality. It provides a
2014    callout function that displays information. By default, it shows the start of
2015    the match and the current position in the text. There are some new data escapes
2016    to vary what happens:
2017    
2018        \C+         in addition, show current contents of captured substrings
2019        \C-         do not supply a callout function
2020        \C!n        return 1 when callout number n is reached
2021        \C!n!m      return 1 when callout number n is reached for the mth time
2022    
2023    30. If pcregrep was called with the -l option and just a single file name, it
2024    output "<stdin>" if a match was found, instead of the file name.
2025    
2026    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
2027    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
2028    pcre_exec(). This saves a malloc/free per call. The default value of
2029    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
2030    when configuring.
2031    
2032    32. The default maximum size of a compiled pattern is 64K. There have been a
2033    few cases of people hitting this limit. The code now uses macros to handle the
2034    storing of links as offsets within the compiled pattern. It defaults to 2-byte
2035    links, but this can be changed to 3 or 4 bytes by --with-link-size when
2036    configuring. Tests 2 and 5 work only with 2-byte links because they output
2037    debugging information about compiled patterns.
2038    
2039    33. Internal code re-arrangements:
2040    
2041    (a) Moved the debugging function for printing out a compiled regex into
2042        its own source file (printint.c) and used #include to pull it into
2043        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
2044        separate copies.
2045    
2046    (b) Defined the list of op-code names for debugging as a macro in
2047        internal.h so that it is next to the definition of the opcodes.
2048    
2049    (c) Defined a table of op-code lengths for simpler skipping along compiled
2050        code. This is again a macro in internal.h so that it is next to the
2051        definition of the opcodes.
2052    
2053    34. Added support for recursive calls to individual subpatterns, along the
2054    lines of Robin Houston's patch (but implemented somewhat differently).
2055    
2056    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
2057    allow it to read and process whole directories in Win32. This code was
2058    contributed by Lionel Fourquaux; it has not been tested by me.
2059    
2060    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
2061    used to name a group. Names consist of alphanumerics and underscores, and must
2062    be unique. Back references use the syntax (?P=name) and recursive calls use
2063    (?P>name) which is a PCRE extension to the Python extension. Groups still have
2064    numbers. The function pcre_fullinfo() can be used after compilation to extract
2065    a name/number map. There are three relevant calls:
2066    
2067      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
2068      PCRE_INFO_NAMECOUNT            yields the number of entries
2069      PCRE_INFO_NAMETABLE            yields a pointer to the map.
2070    
2071    The map is a vector of fixed-size entries. The size of each entry depends on
2072    the length of the longest name used. The first two bytes of each entry are the
2073    group number, most significant byte first. There follows the corresponding
2074    name, zero terminated. The names are in alphabetical order.
2075    
2076    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
2077    case instead of 255. Making it the same both with and without UTF-8 support
2078    means that the same test output works with both.
2079    
2080    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
2081    calling malloc() with a zero argument.
2082    
2083    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
2084    optimization. I've improved things by keeping a bitmap of backreferences with
2085    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
2086    fact referenced, the optimization can be applied. It is unlikely that a
2087    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
2088    the match to follow \n) will appear inside brackets with a number greater than
2089    31, but if it does, any back reference > 31 suppresses the optimization.
2090    
2091    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
2092    of disabling numbered capturing parentheses. Any opening parenthesis that is
2093    not followed by ? behaves as if it were followed by ?: but named parentheses
2094    can still be used for capturing (and they will acquire numbers in the usual
2095    way).
2096    
2097    41. Redesigned the return codes from the match() function into yes/no/error so
2098    that errors can be passed back from deep inside the nested calls. A malloc
2099    failure while inside a recursive subpattern call now causes the
2100    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
2101    
2102    42. It is now possible to set a limit on the number of times the match()
2103    function is called in a call to pcre_exec(). This facility makes it possible to
2104    limit the amount of recursion and backtracking, though not in a directly
2105    obvious way, because the match() function is used in a number of different
2106    circumstances. The count starts from zero for each position in the subject
2107    string (for non-anchored patterns). The default limit is, for compatibility, a
2108    large number, namely 10 000 000. You can change this in two ways:
2109    
2110    (a) When configuring PCRE before making, you can use --with-match-limit=n
2111        to set a default value for the compiled library.
2112    
2113    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
2114        a different value is set. See 45 below.
2115    
2116    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
2117    
2118    43. Added a new function pcre_config(int, void *) to enable run-time extraction
2119    of things that can be changed at compile time. The first argument specifies
2120    what is wanted and the second points to where the information is to be placed.
2121    The current list of available information is:
2122    
2123      PCRE_CONFIG_UTF8
2124    
2125    The output is an integer that is set to one if UTF-8 support is available;
2126    otherwise it is set to zero.
2127    
2128      PCRE_CONFIG_NEWLINE
2129    
2130    The output is an integer that it set to the value of the code that is used for
2131    newline. It is either LF (10) or CR (13).
2132    
2133      PCRE_CONFIG_LINK_SIZE
2134    
2135    The output is an integer that contains the number of bytes used for internal
2136    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
2137    
2138      PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
2139    
2140    The output is an integer that contains the threshold above which the POSIX
2141    interface uses malloc() for output vectors. See item 31 above.
2142    
2143      PCRE_CONFIG_MATCH_LIMIT
2144    
2145    The output is an unsigned integer that contains the default limit of the number
2146    of match() calls in a pcre_exec() execution. See 42 above.
2147    
2148    44. pcretest has been upgraded by the addition of the -C option. This causes it
2149    to extract all the available output from the new pcre_config() function, and to
2150    output it. The program then exits immediately.
2151    
2152    45. A need has arisen to pass over additional data with calls to pcre_exec() in
2153    order to support additional features. One way would have been to define
2154    pcre_exec2() (for example) with extra arguments, but this would not have been
2155    extensible, and would also have required all calls to the original function to
2156    be mapped to the new one. Instead, I have chosen to extend the mechanism that
2157    is used for passing in "extra" data from pcre_study().
2158    
2159    The pcre_extra structure is now exposed and defined in pcre.h. It currently
2160    contains the following fields:
2161    
2162      flags         a bitmap indicating which of the following fields are set
2163      study_data    opaque data from pcre_study()
2164      match_limit   a way of specifying a limit on match() calls for a specific
2165                      call to pcre_exec()
2166      callout_data  data for callouts (see 49 below)
2167    
2168    The flag bits are also defined in pcre.h, and are
2169    
2170      PCRE_EXTRA_STUDY_DATA
2171      PCRE_EXTRA_MATCH_LIMIT
2172      PCRE_EXTRA_CALLOUT_DATA
2173    
2174    The pcre_study() function now returns one of these new pcre_extra blocks, with
2175    the actual study data pointed to by the study_data field, and the
2176    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
2177    before. That is, this change is entirely upwards-compatible and requires no
2178    change to existing code.
2179    
2180    If you want to pass in additional data to pcre_exec(), you can either place it
2181    in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
2182    block.
2183    
2184    46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
2185    data string contains the escape sequence \M, pcretest calls pcre_exec() several
2186    times with different match limits, until it finds the minimum value needed for
2187    pcre_exec() to complete. The value is then output. This can be instructive; for
2188    most simple matches the number is quite small, but for pathological cases it
2189    gets very large very quickly.
2190    
2191    47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
2192    returns the size of the data block pointed to by the study_data field in a
2193    pcre_extra block, that is, the value that was passed as the argument to
2194    pcre_malloc() when PCRE was getting memory in which to place the information
2195    created by pcre_study(). The fourth argument should point to a size_t variable.
2196    pcretest has been extended so that this information is shown after a successful
2197    pcre_study() call when information about the compiled regex is being displayed.
2198    
2199    48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
2200    because what follows is always an absolute path. (Later: it turns out that this
2201    is more than cosmetic for MinGW, because it doesn't like empty path
2202    components.)
2203    
2204    49. Some changes have been made to the callout feature (see 28 above):
2205    
2206    (i)  A callout function now has three choices for what it returns:
2207    
2208           0  =>  success, carry on matching
2209         > 0  =>  failure at this point, but backtrack if possible
2210         < 0  =>  serious error, return this value from pcre_exec()
2211    
2212         Negative values should normally be chosen from the set of PCRE_ERROR_xxx
2213         values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
2214         "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
2215         use by callout functions. It will never be used by PCRE itself.
2216    
2217    (ii) The pcre_extra structure (see 45 above) has a void * field called
2218         callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
2219         pcre_callout_block structure has a field of the same name. The contents of
2220         the field passed in the pcre_extra structure are passed to the callout
2221         function in the corresponding field in the callout block. This makes it
2222         easier to use the same callout-containing regex from multiple threads. For
2223         testing, the pcretest program has a new data escape
2224    
2225           \C*n        pass the number n (may be negative) as callout_data
2226    
2227         If the callout function in pcretest receives a non-zero value as
2228         callout_data, it returns that value.
2229    
2230    50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
2231    there were some redundant $(CFLAGS) in commands that are now specified as
2232    $(LINK), which already includes $(CFLAGS).
2233    
2234    51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
2235    has been compiled with UTF-8 support *and* pcre_compile() has been compiled
2236    with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
2237    one-byte characters throughout. Note that case-insensitive matching applies
2238    only to characters whose values are less than 256. PCRE doesn't support the
2239    notion of cases for higher-valued characters.
2240    
2241    (i)   A character class whose characters are all within 0-255 is handled as
2242          a bit map, and the map is inverted for negative classes. Previously, a
2243          character > 255 always failed to match such a class; however it should
2244          match if the class was a negative one (e.g. [^ab]). This has been fixed.
2245    
2246    (ii)  A negated character class with a single character < 255 is coded as
2247          "not this character" (OP_NOT). This wasn't working properly when the test
2248          character was multibyte, either singly or repeated.
2249    
2250    (iii) Repeats of multibyte characters are now handled correctly in UTF-8
2251          mode, for example: \x{100}{2,3}.
2252    
2253    (iv)  The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
2254          singly or repeated) now correctly test multibyte characters. However,
2255          PCRE doesn't recognize any characters with values greater than 255 as
2256          digits, spaces, or word characters. Such characters always match \D, \S,
2257          and \W, and never match \d, \s, or \w.
2258    
2259    (v)   Classes may now contain characters and character ranges with values
2260          greater than 255. For example: [ab\x{100}-\x{400}].
2261    
2262    (vi)  pcregrep now has a --utf-8 option (synonym -u) which makes it call
2263          PCRE in UTF-8 mode.
2264    
2265    52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
2266    PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
2267    retained for backwards compatibility. (Note that LASTLITERAL is also a byte
2268    value.)
2269    
2270    53. The single man page has become too large. I have therefore split it up into
2271    a number of separate man pages. These also give rise to individual HTML pages;
2272    these are now put in a separate directory, and there is an index.html page that
2273    lists them all. Some hyperlinking between the pages has been installed.
2274    
2275    54. Added convenience functions for handling named capturing parentheses.
2276    
2277    55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2278    aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2279    true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2280    are faulted.
2281    
2282    56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2283    calling configure. These values are used when compiling the dftables.c program
2284    which is run to generate the source of the default character tables. They
2285    default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2286    you will need to set these values.
2287    
2288    57. Updated the building process for Windows DLL, as provided by Fred Cox.
2289    
2290    
2291    Version 3.9 02-Jan-02
2292    ---------------------
2293    
2294    1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2295    
2296    2. If --disable-static was given, the building process failed when trying to
2297    build pcretest and pcregrep. (For some reason it was using libtool to compile
2298    them, which is not right, as they aren't part of the library.)
2299    
2300    
2301    Version 3.8 18-Dec-01
2302    ---------------------
2303    
2304    1. The experimental UTF-8 code was completely screwed up. It was packing the
2305    bytes in the wrong order. How dumb can you get?
2306    
2307    
2308    Version 3.7 29-Oct-01
2309    ---------------------
2310    
2311    1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2312    This caused pcretest, when used on the test data, to segfault. Unfortunately,
2313    this didn't happen under Solaris 8, where I normally test things.
2314    
2315    2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2316    doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2317    isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2318    this fix an hour or so after the initial 3.7 release.)
2319    
2320    
2321    Version 3.6 23-Oct-01
2322    ---------------------
2323    
2324    1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2325    offsets passed as NULL with zero offset count.
2326    
2327    2. The config.guess and config.sub files had not been updated when I moved to
2328    the latest autoconf.
2329    
2330    
2331    Version 3.5 15-Aug-01
2332    ---------------------
2333    
2334    1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2335    had been forgotten.
2336    
2337    2. By using declared but undefined structures, we can avoid using "void"
2338    definitions in pcre.h while keeping the internal definitions of the structures
2339    private.
2340    
2341    3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2342    user point of view, this means that both static and shared libraries are built
2343    by default, but this can be individually controlled. More of the work of
2344    handling this static/shared cases is now inside libtool instead of PCRE's make
2345    file.
2346    
2347    4. The pcretest utility is now installed along with pcregrep because it is
2348    useful for users (to test regexs) and by doing this, it automatically gets
2349    relinked by libtool. The documentation has been turned into a man page, so
2350    there are now .1, .txt, and .html versions in /doc.
2351    
2352    5. Upgrades to pcregrep:
2353       (i)   Added long-form option names like gnu grep.
2354       (ii)  Added --help to list all options with an explanatory phrase.
2355       (iii) Added -r, --recursive to recurse into sub-directories.
2356       (iv)  Added -f, --file to read patterns from a file.
2357    
2358    6. pcre_exec() was referring to its "code" argument before testing that
2359    argument for NULL (and giving an error if it was NULL).
2360    
2361    7. Upgraded Makefile.in to allow for compiling in a different directory from
2362    the source directory.
2363    
2364    8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2365    options bits, the pointer it was passed was to an int instead of to an unsigned
2366    long int. This mattered only on 64-bit systems.
2367    
2368    9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2369    generated) instead of pcre.in, which it its source. Also made the same change
2370    in several of the .c files.
2371    
2372    10. A new release of gcc defines printf() as a macro, which broke pcretest
2373    because it had an ifdef in the middle of a string argument for printf(). Fixed
2374    by using separate calls to printf().
2375    
2376    11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2377    script, to force use of CR or LF instead of \n in the source. On non-Unix
2378    systems, the value can be set in config.h.
2379    
2380    12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2381    absolute limit. Changed the text of the error message to make this clear, and
2382    likewise updated the man page.
2383    
2384    13. The limit of 99 on the number of capturing subpatterns has been removed.
2385    The new limit is 65535, which I hope will not be a "real" limit.
2386    
2387    
2388    Version 3.4 22-Aug-00
2389    ---------------------
2390    
2391    1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2392    
2393    2. Diagnose condition (?(0) as an error instead of crashing on matching.
2394    
2395    
2396    Version 3.3 01-Aug-00
2397    ---------------------
2398    
2399    1. If an octal character was given, but the value was greater than \377, it
2400    was not getting masked to the least significant bits, as documented. This could
2401    lead to crashes in some systems.
2402    
2403    2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
2404    the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
2405    
2406    3. Added the functions pcre_free_substring() and pcre_free_substring_list().
2407    These just pass their arguments on to (pcre_free)(), but they are provided
2408    because some uses of PCRE bind it to non-C systems that can call its functions,
2409    but cannot call free() or pcre_free() directly.
2410    
2411    4. Add "make test" as a synonym for "make check". Corrected some comments in
2412    the Makefile.
2413    
2414    5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
2415    Makefile.
2416    
2417    6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
2418    command called pgrep for grepping around the active processes.
2419    
2420    7. Added the beginnings of support for UTF-8 character strings.
2421    
2422    8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
2423    RANLIB to ./ltconfig so that they are used by libtool. I think these are all
2424    the relevant ones. (AR is not passed because ./ltconfig does its own figuring
2425    out for the ar command.)
2426    
2427    
2428  Version 3.2 12-May-00  Version 3.2 12-May-00
2429  ---------------------  ---------------------

Legend:
Removed from v.47  
changed lines
  Added in v.268

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12