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

Diff of /code/trunk/ChangeLog

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

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

Legend:
Removed from v.49  
changed lines
  Added in v.234

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12