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

Diff of /code/trunk/ChangeLog

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

revision 55 by nigel, Sat Feb 24 21:39:46 2007 UTC revision 277 by ph10, Mon Nov 26 17:03:45 2007 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4    Version 7.5 12-Nov-07
5    ---------------------
6    
7    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
8        values in parens when parsing an RE using the C++ wrapper."
9    
10    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
11        Characters greater than 255 were excluded from the class instead of being
12        included.
13    
14    3.  The same bug as (2) above applied to negated POSIX classes such as
15        [:^space:].
16    
17    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
18        defined or documented. It seems to have been a typo for PCRE_STATIC, so
19        I have changed it.
20    
21    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
22        first named subpattern) and a construct such as (?&a) would reference the
23        first named subpattern whose name started with "a" (in other words, the
24        length check was missing). Both these problems are fixed. "Subpattern name
25        expected" is now given for (?&) (a zero-length name), and this patch also
26        makes it give the same error for \k'' (previously it complained that that
27        was a reference to a non-existent subpattern).
28    
29    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
30        this is right because (?- can be followed by option settings as well as by
31        digits. I have, however, made the messages clearer.
32    
33    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
34        than the number used in the conditional) now cause a compile-time error.
35        This is actually not compatible with Perl, which accepts such patterns, but
36        treats the conditional as always being FALSE (as PCRE used to), but it
37        seems to me that giving a diagnostic is better.
38    
39    8.  Change "alphameric" to the more common word "alphanumeric" in comments
40        and messages.
41    
42    9.  Fix two occurrences of "backslash" in comments that should have been
43        "backspace".
44    
45    10. Remove two redundant lines of code that can never be obeyed (their function
46        was moved elsewhere).
47    
48    11. The program that makes PCRE's Unicode character property table had a bug
49        which caused it to generate incorrect table entries for sequences of
50        characters that have the same character type, but are in different scripts.
51        It amalgamated them into a single range, with the script of the first of
52        them. In other words, some characters were in the wrong script. There were
53        thirteen such cases, affecting characters in the following ranges:
54    
55          U+002b0 - U+002c1
56          U+0060c - U+0060d
57          U+0061e - U+00612
58          U+0064b - U+0065e
59          U+0074d - U+0076d
60          U+01800 - U+01805
61          U+01d00 - U+01d77
62          U+01d9b - U+01dbf
63          U+0200b - U+0200f
64          U+030fc - U+030fe
65          U+03260 - U+0327f
66          U+0fb46 - U+0fbb1
67          U+10450 - U+1049d
68    
69    
70    Version 7.4 21-Sep-07
71    ---------------------
72    
73    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
74        means that a class such as [\s] counted as "explicit reference to CR or
75        LF". That isn't really right - the whole point of the change was to try to
76        help when there was an actual mention of one of the two characters. So now
77        the change happens only if \r or \n (or a literal CR or LF) character is
78        encountered.
79    
80    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
81        of both had grown to the point where there were only 3 bits left.
82        Fortunately, there was spare space in the data structure, and so I have
83        moved the internal flags into a new 16-bit field to free up more option
84        bits.
85    
86    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
87        but did not set the internal JCHANGED flag - either of these is enough to
88        control the way the "get" function works - but the PCRE_INFO_JCHANGED
89        facility is supposed to tell if (?J) was ever used, so now (?J) at the
90        start sets both bits.
91    
92    4.  Added options (at build time, compile time, exec time) to change \R from
93        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
94    
95    5.  doc/pcresyntax.html was missing from the distribution.
96    
97    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
98        compatibility, even though it is no longer used.
99    
100    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
101        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
102        windows.h file is present (where different names are used). [This was
103        reversed later after testing - see 16 below.]
104    
105    8.  Changed all #include <config.h> to #include "config.h". There were also
106        some further <pcre.h> cases that I changed to "pcre.h".
107    
108    9.  When pcregrep was used with the --colour option, it missed the line ending
109        sequence off the lines that it output.
110    
111    10. It was pointed out to me that arrays of string pointers cause lots of
112        relocations when a shared library is dynamically loaded. A technique of
113        using a single long string with a table of offsets can drastically reduce
114        these. I have refactored PCRE in four places to do this. The result is
115        dramatic:
116    
117          Originally:                          290
118          After changing UCP table:            187
119          After changing error message table:   43
120          After changing table of "verbs"       36
121          After changing table of Posix names   22
122    
123        Thanks to the folks working on Gregex for glib for this insight.
124    
125    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
126        unicode-properties was also set.
127    
128    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
129    
130    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
131        checked only for CRLF.
132    
133    14. Added casts to pcretest.c to avoid compiler warnings.
134    
135    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
136    
137    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
138        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
139        entirely. This removes changes made in 7 above.
140    
141    17. The CMake files have been updated, and there is now more information about
142        building with CMake in the NON-UNIX-USE document.
143    
144    
145    Version 7.3 28-Aug-07
146    ---------------------
147    
148     1. In the rejigging of the build system that eventually resulted in 7.1, the
149        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
150        brackets there is not right, since it causes compilers to look for an
151        installed pcre.h, not the version that is in the source that is being
152        compiled (which of course may be different). I have changed it back to:
153    
154          #include "pcre.h"
155    
156        I have a vague recollection that the change was concerned with compiling in
157        different directories, but in the new build system, that is taken care of
158        by the VPATH setting the Makefile.
159    
160     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
161        when the subject happened to end in the byte 0x85 (e.g. if the last
162        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
163        characters but of course it shouldn't be taken as a newline when it is part
164        of another character. The bug was that, for an unlimited repeat of . in
165        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
166        characters when looking for a newline.
167    
168     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
169    
170     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
171        in debug output.
172    
173     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
174        long printing in the pcrecpp unittest when running under MinGW.
175    
176     6. ESC_K was left out of the EBCDIC table.
177    
178     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
179        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
180        limit also applies to "virtual nesting" when a pattern is recursive, and in
181        this case 1000 isn't so big. I have been able to remove this limit at the
182        expense of backing off one optimization in certain circumstances. Normally,
183        when pcre_exec() would call its internal match() function recursively and
184        immediately return the result unconditionally, it uses a "tail recursion"
185        feature to save stack. However, when a subpattern that can match an empty
186        string has an unlimited repetition quantifier, it no longer makes this
187        optimization. That gives it a stack frame in which to save the data for
188        checking that an empty string has been matched. Previously this was taken
189        from the 1000-entry workspace that had been reserved. So now there is no
190        explicit limit, but more stack is used.
191    
192     8. Applied Daniel's patches to solve problems with the import/export magic
193        syntax that is required for Windows, and which was going wrong for the
194        pcreposix and pcrecpp parts of the library. These were overlooked when this
195        problem was solved for the main library.
196    
197     9. There were some crude static tests to avoid integer overflow when computing
198        the size of patterns that contain repeated groups with explicit upper
199        limits. As the maximum quantifier is 65535, the maximum group length was
200        set at 30,000 so that the product of these two numbers did not overflow a
201        32-bit integer. However, it turns out that people want to use groups that
202        are longer than 30,000 bytes (though not repeat them that many times).
203        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
204        made it possible to implement the integer overflow checks in a much more
205        dynamic way, which I have now done. The artificial limitation on group
206        length has been removed - we now have only the limit on the total length of
207        the compiled pattern, which depends on the LINK_SIZE setting.
208    
209    10. Fixed a bug in the documentation for get/copy named substring when
210        duplicate names are permitted. If none of the named substrings are set, the
211        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
212        empty string.
213    
214    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
215        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
216        because the ] is interpreted as the first data character and the
217        terminating ] is not found. PCRE has been made compatible with Perl in this
218        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
219        cause memory overwriting.
220    
221    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
222        string has been matched (to stop an infinite loop). It was not recognizing
223        a conditional subpattern that could match an empty string if that
224        subpattern was within another subpattern. For example, it looped when
225        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
226        condition was not nested. This bug has been fixed.
227    
228    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
229        past the start of the subject in the presence of bytes with the top bit
230        set, for example "\x8aBCD".
231    
232    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
233        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
234    
235    14. Optimized (?!) to (*FAIL).
236    
237    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
238        This restricts code points to be within the range 0 to 0x10FFFF, excluding
239        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
240        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
241        does: it's just the validity check that is more restrictive.
242    
243    16. Inserted checks for integer overflows during escape sequence (backslash)
244        processing, and also fixed erroneous offset values for syntax errors during
245        backslash processing.
246    
247    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
248        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
249    
250    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
251        caused an overrun.
252    
253    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
254        something other than just ASCII characters) inside a group that had an
255        unlimited repeat caused a loop at compile time (while checking to see
256        whether the group could match an empty string).
257    
258    20. Debugging a pattern containing \p or \P could cause a crash. For example,
259        [\P{Any}] did so. (Error in the code for printing property names.)
260    
261    21. An orphan \E inside a character class could cause a crash.
262    
263    22. A repeated capturing bracket such as (A)? could cause a wild memory
264        reference during compilation.
265    
266    23. There are several functions in pcre_compile() that scan along a compiled
267        expression for various reasons (e.g. to see if it's fixed length for look
268        behind). There were bugs in these functions when a repeated \p or \P was
269        present in the pattern. These operators have additional parameters compared
270        with \d, etc, and these were not being taken into account when moving along
271        the compiled data. Specifically:
272    
273        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
274            length.
275    
276        (b) An item such as \pL+ within a repeated group could cause crashes or
277            loops.
278    
279        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
280            "reference to non-existent subpattern" error.
281    
282        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
283    
284    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
285        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
286    
287    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
288    
289    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
290        character were causing crashes (broken optimization).
291    
292    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
293        \p or \P) caused a compile-time loop.
294    
295    28. More problems have arisen in unanchored patterns when CRLF is a valid line
296        break. For example, the unstudied pattern [\r\n]A does not match the string
297        "\r\nA" because change 7.0/46 below moves the current point on by two
298        characters after failing to match at the start. However, the pattern \nA
299        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
300        the same is true. There doesn't seem any very clean way out of this, but
301        what I have chosen to do makes the common cases work: PCRE now takes note
302        of whether there can be an explicit match for \r or \n anywhere in the
303        pattern, and if so, 7.0/46 no longer applies. As part of this change,
304        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
305        pattern has explicit CR or LF references.
306    
307    29. Added (*CR) etc for changing newline setting at start of pattern.
308    
309    
310    Version 7.2 19-Jun-07
311    ---------------------
312    
313     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
314        which is apparently normally available under Windows.
315    
316     2. Re-jig the pcregrep tests with different newline settings in an attempt
317        to make them independent of the local environment's newline setting.
318    
319     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
320    
321     4. Some of the "internals" tests were previously cut out when the link size
322        was not 2, because the output contained actual offsets. The recent new
323        "Z" feature of pcretest means that these can be cut out, making the tests
324        usable with all link sizes.
325    
326     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
327        stack recursion. This gives a massive performance boost under BSD, but just
328        a small improvement under Linux. However, it saves one field in the frame
329        in all cases.
330    
331     6. Added more features from the forthcoming Perl 5.10:
332    
333        (a) (?-n) (where n is a string of digits) is a relative subroutine or
334            recursion call. It refers to the nth most recently opened parentheses.
335    
336        (b) (?+n) is also a relative subroutine call; it refers to the nth next
337            to be opened parentheses.
338    
339        (c) Conditions that refer to capturing parentheses can be specified
340            relatively, for example, (?(-2)... or (?(+3)...
341    
342        (d) \K resets the start of the current match so that everything before
343            is not part of it.
344    
345        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
346    
347        (f) \g{name} is another synonym - part of Perl 5.10's unification of
348            reference syntax.
349    
350        (g) (?| introduces a group in which the numbering of parentheses in each
351            alternative starts with the same number.
352    
353        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
354    
355     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
356        PCRE_INFO_JCHANGED.
357    
358     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
359        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
360        for detecting groups that can match an empty string.
361    
362     9. A pattern with a very large number of alternatives (more than several
363        hundred) was running out of internal workspace during the pre-compile
364        phase, where pcre_compile() figures out how much memory will be needed. A
365        bit of new cunning has reduced the workspace needed for groups with
366        alternatives. The 1000-alternative test pattern now uses 12 bytes of
367        workspace instead of running out of the 4096 that are available.
368    
369    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
370    
371    11. Applied patch from Google to remove an optimization that didn't quite work.
372        The report of the bug said:
373    
374          pcrecpp::RE("a*").FullMatch("aaa") matches, while
375          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
376          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
377    
378    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
379        it matched the wrong number of bytes.
380    
381    
382    Version 7.1 24-Apr-07
383    ---------------------
384    
385     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
386        that is more "standard", making use of automake and other Autotools. There
387        is some re-arrangement of the files and adjustment of comments consequent
388        on this.
389    
390     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
391        for recursive directory scanning broke on some systems because the files
392        are not scanned in any specific order and on different systems the order
393        was different. A call to "sort" has been inserted into RunGrepTest for the
394        approprate test as a short-term fix. In the longer term there may be an
395        alternative.
396    
397     3. I had an email from Eric Raymond about problems translating some of PCRE's
398        man pages to HTML (despite the fact that I distribute HTML pages, some
399        people do their own conversions for various reasons). The problems
400        concerned the use of low-level troff macros .br and .in. I have therefore
401        removed all such uses from the man pages (some were redundant, some could
402        be replaced by .nf/.fi pairs). The 132html script that I use to generate
403        HTML has been updated to handle .nf/.fi and to complain if it encounters
404        .br or .in.
405    
406     4. Updated comments in configure.ac that get placed in config.h.in and also
407        arranged for config.h to be included in the distribution, with the name
408        config.h.generic, for the benefit of those who have to compile without
409        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
410    
411     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
412        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
413        makevp.bat for the current PCRE, using the additional files
414        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
415    
416     6. A Windows user reported a minor discrepancy with test 2, which turned out
417        to be caused by a trailing space on an input line that had got lost in his
418        copy. The trailing space was an accident, so I've just removed it.
419    
420     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
421        that is needed.
422    
423     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
424        as "const" (a) because they are and (b) because it helps the PHP
425        maintainers who have recently made a script to detect big data structures
426        in the php code that should be moved to the .rodata section. I remembered
427        to update Builducptable as well, so it won't revert if ucptable.h is ever
428        re-created.
429    
430     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
431        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
432        order to be able to cut out the UTF-8 tables in the latter when UTF-8
433        support is not required. This saves 1.5-2K of code, which is important in
434        some applications.
435    
436        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
437        so as not to refer to the tables, even though these functions will never be
438        called when UTF-8 support is disabled. Otherwise there are problems with a
439        shared library.
440    
441    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
442    
443        (a) It was defining its arguments as char * instead of void *.
444    
445        (b) It was assuming that all moves were upwards in memory; this was true
446            a long time ago when I wrote it, but is no longer the case.
447    
448        The emulated memove() is provided for those environments that have neither
449        memmove() nor bcopy(). I didn't think anyone used it these days, but that
450        is clearly not the case, as these two bugs were recently reported.
451    
452    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
453        and Detrail to create the HTML documentation, the .txt form of the man
454        pages, and it removes trailing spaces from listed files. It also creates
455        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
456        case, it wraps all the #defines with #ifndefs. This script should be run
457        before "make dist".
458    
459    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
460        with Unicode property support.
461    
462        (a) For a maximizing quantifier, if the two different cases of the
463            character were of different lengths in their UTF-8 codings (there are
464            some cases like this - I found 11), and the matching function had to
465            back up over a mixture of the two cases, it incorrectly assumed they
466            were both the same length.
467    
468        (b) When PCRE was configured to use the heap rather than the stack for
469            recursion during matching, it was not correctly preserving the data for
470            the other case of a UTF-8 character when checking ahead for a match
471            while processing a minimizing repeat. If the check also involved
472            matching a wide character, but failed, corruption could cause an
473            erroneous result when trying to check for a repeat of the original
474            character.
475    
476    13. Some tidying changes to the testing mechanism:
477    
478        (a) The RunTest script now detects the internal link size and whether there
479            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
480            values substituted by "configure". (The RunGrepTest script already did
481            this for UTF-8.) The configure.ac script no longer substitutes the
482            relevant variables.
483    
484        (b) The debugging options /B and /D in pcretest show the compiled bytecode
485            with length and offset values. This means that the output is different
486            for different internal link sizes. Test 2 is skipped for link sizes
487            other than 2 because of this, bypassing the problem. Unfortunately,
488            there was also a test in test 3 (the locale tests) that used /B and
489            failed for link sizes other than 2. Rather than cut the whole test out,
490            I have added a new /Z option to pcretest that replaces the length and
491            offset values with spaces. This is now used to make test 3 independent
492            of link size. (Test 2 will be tidied up later.)
493    
494    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
495        segmentation fault instead of returning the appropriate error message.
496    
497    15. In multiline mode when the newline sequence was set to "any", the pattern
498        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
499        This doesn't seem right; it now treats the CRLF combination as the line
500        ending, and so does not match in that case. It's only a pattern such as ^$
501        that would hit this one: something like ^ABC$ would have failed after \r
502        and then tried again after \r\n.
503    
504    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
505        in an attempt to make files that differ only in their line terminators
506        compare equal. This works on Linux.
507    
508    17. Under certain error circumstances pcregrep might try to free random memory
509        as it exited. This is now fixed, thanks to valgrind.
510    
511    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
512        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
513        was because its rules for how to advance for /g after matching an empty
514        string at the end of a line did not allow for this case. They now check for
515        it specially.
516    
517    20. pcretest is supposed to handle patterns and data of any length, by
518        extending its buffers when necessary. It was getting this wrong when the
519        buffer for a data line had to be extended.
520    
521    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
522        CRLF as a newline sequence.
523    
524    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
525        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
526        I have nevertheless tidied it up.
527    
528    23. Added some casts to kill warnings from HP-UX ia64 compiler.
529    
530    24. Added a man page for pcre-config.
531    
532    
533    Version 7.0 19-Dec-06
534    ---------------------
535    
536     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
537        moving to gcc 4.1.1.
538    
539     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
540        sys/time.h, which is documented as needed for this function. It doesn't
541        seem to matter on Linux, but it showed up on some releases of OS X.
542    
543     3. It seems that there are systems where bytes whose values are greater than
544        127 match isprint() in the "C" locale. The "C" locale should be the
545        default when a C program starts up. In most systems, only ASCII printing
546        characters match isprint(). This difference caused the output from pcretest
547        to vary, making some of the tests fail. I have changed pcretest so that:
548    
549        (a) When it is outputting text in the compiled version of a pattern, bytes
550            other than 32-126 are always shown as hex escapes.
551    
552        (b) When it is outputting text that is a matched part of a subject string,
553            it does the same, unless a different locale has been set for the match
554            (using the /L modifier). In this case, it uses isprint() to decide.
555    
556     4. Fixed a major bug that caused incorrect computation of the amount of memory
557        required for a compiled pattern when options that changed within the
558        pattern affected the logic of the preliminary scan that determines the
559        length. The relevant options are -x, and -i in UTF-8 mode. The result was
560        that the computed length was too small. The symptoms of this bug were
561        either the PCRE error "internal error: code overflow" from pcre_compile(),
562        or a glibc crash with a message such as "pcretest: free(): invalid next
563        size (fast)". Examples of patterns that provoked this bug (shown in
564        pcretest format) are:
565    
566          /(?-x: )/x
567          /(?x)(?-x: \s*#\s*)/
568          /((?i)[\x{c0}])/8
569          /(?i:[\x{c0}])/8
570    
571        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
572        is now done differently.
573    
574     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
575        wrapper classes; (b) implement a new function in the C++ scanner that is
576        more efficient than the old way of doing things because it avoids levels of
577        recursion in the regex matching; (c) add a paragraph to the documentation
578        for the FullMatch() function.
579    
580     6. The escape sequence \n was being treated as whatever was defined as
581        "newline". Not only was this contrary to the documentation, which states
582        that \n is character 10 (hex 0A), but it also went horribly wrong when
583        "newline" was defined as CRLF. This has been fixed.
584    
585     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
586        was being set to -1 for the "end of line" case (supposedly a value that no
587        character can have). Though this value is never used (the check for end of
588        line is "zero bytes in current character"), it caused compiler complaints.
589        I've changed it to 0xffffffff.
590    
591     8. In pcre_version.c, the version string was being built by a sequence of
592        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
593        string (as it is for production releases) called a macro with an empty
594        argument. The C standard says the result of this is undefined. The gcc
595        compiler treats it as an empty string (which was what was wanted) but it is
596        reported that Visual C gives an error. The source has been hacked around to
597        avoid this problem.
598    
599     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
600        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
601        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
602        of them did).
603    
604    10. Originally, pcretest opened its input and output without "b"; then I was
605        told that "b" was needed in some environments, so it was added for release
606        5.0 to both the input and output. (It makes no difference on Unix-like
607        systems.) Later I was told that it is wrong for the input on Windows. I've
608        now abstracted the modes into two macros, to make it easier to fiddle with
609        them, and removed "b" from the input mode under Windows.
610    
611    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
612    
613    12. Added -help and --help to pcretest as an official way of being reminded
614        of the options.
615    
616    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
617        and pcrecpp.cc because they annoy compilers at high warning levels.
618    
619    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
620    
621    15. Fixed an occurrence of == in configure.ac that should have been = (shell
622        scripts are not C programs :-) and which was not noticed because it works
623        on Linux.
624    
625    16. pcretest is supposed to handle any length of pattern and data line (as one
626        line or as a continued sequence of lines) by extending its input buffer if
627        necessary. This feature was broken for very long pattern lines, leading to
628        a string of junk being passed to pcre_compile() if the pattern was longer
629        than about 50K.
630    
631    17. I have done a major re-factoring of the way pcre_compile() computes the
632        amount of memory needed for a compiled pattern. Previously, there was code
633        that made a preliminary scan of the pattern in order to do this. That was
634        OK when PCRE was new, but as the facilities have expanded, it has become
635        harder and harder to keep it in step with the real compile phase, and there
636        have been a number of bugs (see for example, 4 above). I have now found a
637        cunning way of running the real compile function in a "fake" mode that
638        enables it to compute how much memory it would need, while actually only
639        ever using a few hundred bytes of working memory and without too many
640        tests of the mode. This should make future maintenance and development
641        easier. A side effect of this work is that the limit of 200 on the nesting
642        depth of parentheses has been removed (though this was never a serious
643        limitation, I suspect). However, there is a downside: pcre_compile() now
644        runs more slowly than before (30% or more, depending on the pattern). I
645        hope this isn't a big issue. There is no effect on runtime performance.
646    
647    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
648        newline (only possible for the last line of a file) and it was a
649        pattern that set a locale (followed by /Lsomething), pcretest crashed.
650    
651    19. Added additional timing features to pcretest. (1) The -tm option now times
652        matching only, not compiling. (2) Both -t and -tm can be followed, as a
653        separate command line item, by a number that specifies the number of
654        repeats to use when timing. The default is 50000; this gives better
655        precision, but takes uncomfortably long for very large patterns.
656    
657    20. Extended pcre_study() to be more clever in cases where a branch of a
658        subpattern has no definite first character. For example, (a*|b*)[cd] would
659        previously give no result from pcre_study(). Now it recognizes that the
660        first character must be a, b, c, or d.
661    
662    21. There was an incorrect error "recursive call could loop indefinitely" if
663        a subpattern (or the entire pattern) that was being tested for matching an
664        empty string contained only one non-empty item after a nested subpattern.
665        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
666        incorrectly, because the \d was being skipped in the check.
667    
668    22. The pcretest program now has a new pattern option /B and a command line
669        option -b, which is equivalent to adding /B to every pattern. This causes
670        it to show the compiled bytecode, without the additional information that
671        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
672        is the same as /B/I).
673    
674    23. A new optimization is now able automatically to treat some sequences such
675        as a*b as a*+b. More specifically, if something simple (such as a character
676        or a simple class like \d) has an unlimited quantifier, and is followed by
677        something that cannot possibly match the quantified thing, the quantifier
678        is automatically "possessified".
679    
680    24. A recursive reference to a subpattern whose number was greater than 39
681        went wrong under certain circumstances in UTF-8 mode. This bug could also
682        have affected the operation of pcre_study().
683    
684    25. Realized that a little bit of performance could be had by replacing
685        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
686    
687    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
688    
689    27. Possessive quantifiers such as a++ were previously implemented by turning
690        them into atomic groups such as ($>a+). Now they have their own opcodes,
691        which improves performance. This includes the automatically created ones
692        from 23 above.
693    
694    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
695        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
696        the first matched character to be a colon. This applied both to named and
697        numbered groups.
698    
699    29. The ucpinternal.h header file was missing its idempotency #ifdef.
700    
701    30. I was sent a "project" file called libpcre.a.dev which I understand makes
702        building PCRE on Windows easier, so I have included it in the distribution.
703    
704    31. There is now a check in pcretest against a ridiculously large number being
705        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
706        loop, the loop is abandoned.
707    
708    32. Forward references to subpatterns in conditions such as (?(2)...) where
709        subpattern 2 is defined later cause pcre_compile() to search forwards in
710        the pattern for the relevant set of parentheses. This search went wrong
711        when there were unescaped parentheses in a character class, parentheses
712        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
713    
714    33. "Subroutine" calls and backreferences were previously restricted to
715        referencing subpatterns earlier in the regex. This restriction has now
716        been removed.
717    
718    34. Added a number of extra features that are going to be in Perl 5.10. On the
719        whole, these are just syntactic alternatives for features that PCRE had
720        previously implemented using the Python syntax or my own invention. The
721        other formats are all retained for compatibility.
722    
723        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
724            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
725            also .NET compatible.
726    
727        (b) A recursion or subroutine call to a named group can now be defined as
728            (?&name) as well as (?P>name).
729    
730        (c) A backreference to a named group can now be defined as \k<name> or
731            \k'name' as well as (?P=name). The new forms, as well as being in Perl
732            5.10, are also .NET compatible.
733    
734        (d) A conditional reference to a named group can now use the syntax
735            (?(<name>) or (?('name') as well as (?(name).
736    
737        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
738            groups (named and numbered) that are never evaluated inline, but can be
739            called as "subroutines" from elsewhere. In effect, the DEFINE condition
740            is always false. There may be only one alternative in such a group.
741    
742        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
743            as the simple (?(R). The condition is true only if the most recent
744            recursion is that of the given number or name. It does not search out
745            through the entire recursion stack.
746    
747        (g) The escape \gN or \g{N} has been added, where N is a positive or
748            negative number, specifying an absolute or relative reference.
749    
750    35. Tidied to get rid of some further signed/unsigned compiler warnings and
751        some "unreachable code" warnings.
752    
753    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
754        things, this adds five new scripts.
755    
756    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
757        There were also incompatibilities regarding the handling of \Q..\E inside
758        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
759        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
760    
761    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
762        matches an empty string, and forcibly breaks the loop. There were bugs in
763        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
764        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
765        separate and independent bugs (that affected different cases) have been
766        fixed.
767    
768    39. Refactored the code to abolish the use of different opcodes for small
769        capturing bracket numbers. This is a tidy that I avoided doing when I
770        removed the limit on the number of capturing brackets for 3.5 back in 2001.
771        The new approach is not only tidier, it makes it possible to reduce the
772        memory needed to fix the previous bug (38).
773    
774    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
775        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
776        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
777        mode.
778    
779    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
780        report.
781    
782    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
783        copy construction and assignment in the C++ wrapper.
784    
785    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
786        couple of bugs that could have given wrong results in the "--newline=crlf"
787        case.
788    
789    44. Added a number of casts and did some reorganization of signed/unsigned int
790        variables following suggestions from Dair Grant. Also renamed the variable
791        "this" as "item" because it is a C++ keyword.
792    
793    45. Arranged for dftables to add
794    
795          #include "pcre_internal.h"
796    
797        to pcre_chartables.c because without it, gcc 4.x may remove the array
798        definition from the final binary if PCRE is built into a static library and
799        dead code stripping is activated.
800    
801    46. For an unanchored pattern, if a match attempt fails at the start of a
802        newline sequence, and the newline setting is CRLF or ANY, and the next two
803        characters are CRLF, advance by two characters instead of one.
804    
805    
806    Version 6.7 04-Jul-06
807    ---------------------
808    
809     1. In order to handle tests when input lines are enormously long, pcretest has
810        been re-factored so that it automatically extends its buffers when
811        necessary. The code is crude, but this _is_ just a test program. The
812        default size has been increased from 32K to 50K.
813    
814     2. The code in pcre_study() was using the value of the re argument before
815        testing it for NULL. (Of course, in any sensible call of the function, it
816        won't be NULL.)
817    
818     3. The memmove() emulation function in pcre_internal.h, which is used on
819        systems that lack both memmove() and bcopy() - that is, hardly ever -
820        was missing a "static" storage class specifier.
821    
822     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
823        containing an extended class (one that cannot be represented by a bitmap
824        because it contains high-valued characters or Unicode property items, e.g.
825        [\pZ]). Almost always one would set UTF-8 mode when processing such a
826        pattern, but PCRE should not loop if you do not (it no longer does).
827        [Detail: two cases were found: (a) a repeated subpattern containing an
828        extended class; (b) a recursive reference to a subpattern that followed a
829        previous extended class. It wasn't skipping over the extended class
830        correctly when UTF-8 mode was not set.]
831    
832     5. A negated single-character class was not being recognized as fixed-length
833        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
834        compile error "lookbehind assertion is not fixed length".
835    
836     6. The RunPerlTest auxiliary script was showing an unexpected difference
837        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
838        write a Perl script that can interpret lines of an input file either as
839        byte characters or as UTF-8, which is what "perltest" was being required to
840        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
841        can't do is switch easily at run time between having the "use utf8;" pragma
842        or not. In the end, I fudged it by using the RunPerlTest script to insert
843        "use utf8;" explicitly for the UTF-8 tests.
844    
845     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
846        the end of the subject string, contrary to the documentation and to what
847        Perl does. This was true of both matching functions. Now it matches only at
848        the start of the subject and immediately after *internal* newlines.
849    
850     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
851        a pointer to an int instead of a pointer to an unsigned long int. This
852        caused problems on 64-bit systems.
853    
854     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
855        instance of the 'standard' template library not being so standard".
856    
857    10. There was no check on the number of named subpatterns nor the maximum
858        length of a subpattern name. The product of these values is used to compute
859        the size of the memory block for a compiled pattern. By supplying a very
860        long subpattern name and a large number of named subpatterns, the size
861        computation could be caused to overflow. This is now prevented by limiting
862        the length of names to 32 characters, and the number of named subpatterns
863        to 10,000.
864    
865    11. Subpatterns that are repeated with specific counts have to be replicated in
866        the compiled pattern. The size of memory for this was computed from the
867        length of the subpattern and the repeat count. The latter is limited to
868        65535, but there was no limit on the former, meaning that integer overflow
869        could in principle occur. The compiled length of a repeated subpattern is
870        now limited to 30,000 bytes in order to prevent this.
871    
872    12. Added the optional facility to have named substrings with the same name.
873    
874    13. Added the ability to use a named substring as a condition, using the
875        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
876        are numbers (not recommended). Forward references are permitted.
877    
878    14. Added forward references in named backreferences (if you see what I mean).
879    
880    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
881        pattern could run off the end of the subject. For example, the pattern
882        "(?s)(.{1,5})"8 did this with the subject "ab".
883    
884    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
885        PCRE_CASELESS was set when matching characters that were quantified with ?
886        or *.
887    
888    17. A character class other than a single negated character that had a minimum
889        but no maximum quantifier - for example [ab]{6,} - was not handled
890        correctly by pce_dfa_exec(). It would match only one character.
891    
892    18. A valid (though odd) pattern that looked like a POSIX character
893        class but used an invalid character after [ (for example [[,abc,]]) caused
894        pcre_compile() to give the error "Failed: internal error: code overflow" or
895        in some cases to crash with a glibc free() error. This could even happen if
896        the pattern terminated after [[ but there just happened to be a sequence of
897        letters, a binary zero, and a closing ] in the memory that followed.
898    
899    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
900        over the years. Originally (before any Unicode support), just the bottom 8
901        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
902        output from "man perlunicode" includes this:
903    
904          The regular expression compiler produces polymorphic opcodes.  That
905          is, the pattern adapts to the data and automatically switches to
906          the Unicode character scheme when presented with Unicode data--or
907          instead uses a traditional byte scheme when presented with byte
908          data.
909    
910        Sadly, a wide octal escape does not cause a switch, and in a string with
911        no other multibyte characters, these octal escapes are treated as before.
912        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
913        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
914        Unicode string.
915    
916        I have not perpetrated such confusion in PCRE. Up till now, it took just
917        the bottom 8 bits, as in old Perl. I have now made octal escapes with
918        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
919        translate to the appropriate multibyte character.
920    
921    29. Applied some refactoring to reduce the number of warnings from Microsoft
922        and Borland compilers. This has included removing the fudge introduced
923        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
924        a warning about an unused variable.
925    
926    21. PCRE has not included VT (character 0x0b) in the set of whitespace
927        characters since release 4.0, because Perl (from release 5.004) does not.
928        [Or at least, is documented not to: some releases seem to be in conflict
929        with the documentation.] However, when a pattern was studied with
930        pcre_study() and all its branches started with \s, PCRE still included VT
931        as a possible starting character. Of course, this did no harm; it just
932        caused an unnecessary match attempt.
933    
934    22. Removed a now-redundant internal flag bit that recorded the fact that case
935        dependency changed within the pattern. This was once needed for "required
936        byte" processing, but is no longer used. This recovers a now-scarce options
937        bit. Also moved the least significant internal flag bit to the most-
938        significant bit of the word, which was not previously used (hangover from
939        the days when it was an int rather than a uint) to free up another bit for
940        the future.
941    
942    23. Added support for CRLF line endings as well as CR and LF. As well as the
943        default being selectable at build time, it can now be changed at runtime
944        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
945        specify that it is scanning data with non-default line endings.
946    
947    24. Changed the definition of CXXLINK to make it agree with the definition of
948        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
949    
950    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
951        recursions. This makes a big different to stack usage for some patterns.
952    
953    26. If a subpattern containing a named recursion or subroutine reference such
954        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
955        the space required for the compiled pattern went wrong and gave too small a
956        value. Depending on the environment, this could lead to "Failed: internal
957        error: code overflow at offset 49" or "glibc detected double free or
958        corruption" errors.
959    
960    27. Applied patches from Google (a) to support the new newline modes and (b) to
961        advance over multibyte UTF-8 characters in GlobalReplace.
962    
963    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
964        difference for some implementation of PCRE in some Windows version.
965    
966    29. Added some extra testing facilities to pcretest:
967    
968        \q<number>   in a data line sets the "match limit" value
969        \Q<number>   in a data line sets the "match recursion limt" value
970        -S <number>  sets the stack size, where <number> is in megabytes
971    
972        The -S option isn't available for Windows.
973    
974    
975    Version 6.6 06-Feb-06
976    ---------------------
977    
978     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
979        in pcreposix.h. I have copied the definition from pcre.h.
980    
981     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
982        because pcre.h is no longer a built file.
983    
984     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
985        not normally included in the compiled code.
986    
987    
988    Version 6.5 01-Feb-06
989    ---------------------
990    
991     1. When using the partial match feature with pcre_dfa_exec(), it was not
992        anchoring the second and subsequent partial matches at the new starting
993        point. This could lead to incorrect results. For example, with the pattern
994        /1234/, partially matching against "123" and then "a4" gave a match.
995    
996     2. Changes to pcregrep:
997    
998        (a) All non-match returns from pcre_exec() were being treated as failures
999            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1000            error message is output. Some extra information is given for the
1001            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
1002            probably the only errors that are likely to be caused by users (by
1003            specifying a regex that has nested indefinite repeats, for instance).
1004            If there are more than 20 of these errors, pcregrep is abandoned.
1005    
1006        (b) A binary zero was treated as data while matching, but terminated the
1007            output line if it was written out. This has been fixed: binary zeroes
1008            are now no different to any other data bytes.
1009    
1010        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1011            used to set a locale for matching. The --locale=xxxx long option has
1012            been added (no short equivalent) to specify a locale explicitly on the
1013            pcregrep command, overriding the environment variables.
1014    
1015        (d) When -B was used with -n, some line numbers in the output were one less
1016            than they should have been.
1017    
1018        (e) Added the -o (--only-matching) option.
1019    
1020        (f) If -A or -C was used with -c (count only), some lines of context were
1021            accidentally printed for the final match.
1022    
1023        (g) Added the -H (--with-filename) option.
1024    
1025        (h) The combination of options -rh failed to suppress file names for files
1026            that were found from directory arguments.
1027    
1028        (i) Added the -D (--devices) and -d (--directories) options.
1029    
1030        (j) Added the -F (--fixed-strings) option.
1031    
1032        (k) Allow "-" to be used as a file name for -f as well as for a data file.
1033    
1034        (l) Added the --colo(u)r option.
1035    
1036        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1037            is not present by default.
1038    
1039     3. A nasty bug was discovered in the handling of recursive patterns, that is,
1040        items such as (?R) or (?1), when the recursion could match a number of
1041        alternatives. If it matched one of the alternatives, but subsequently,
1042        outside the recursion, there was a failure, the code tried to back up into
1043        the recursion. However, because of the way PCRE is implemented, this is not
1044        possible, and the result was an incorrect result from the match.
1045    
1046        In order to prevent this happening, the specification of recursion has
1047        been changed so that all such subpatterns are automatically treated as
1048        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1049    
1050     4. I had overlooked the fact that, in some locales, there are characters for
1051        which isalpha() is true but neither isupper() nor islower() are true. In
1052        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1053        and ordfeminine) are like this. This affected the treatment of \w and \W
1054        when they appeared in character classes, but not when they appeared outside
1055        a character class. The bit map for "word" characters is now created
1056        separately from the results of isalnum() instead of just taking it from the
1057        upper, lower, and digit maps. (Plus the underscore character, of course.)
1058    
1059     5. The above bug also affected the handling of POSIX character classes such as
1060        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1061        permanent tables. Instead, the bit maps for such a class were previously
1062        created as the appropriate unions of the upper, lower, and digit bitmaps.
1063        Now they are created by subtraction from the [[:word:]] class, which has
1064        its own bitmap.
1065    
1066     6. The [[:blank:]] character class matches horizontal, but not vertical space.
1067        It is created by subtracting the vertical space characters (\x09, \x0a,
1068        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1069        subtraction was done in the overall bitmap for a character class, meaning
1070        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1071        be recognized. This bug has been fixed.
1072    
1073     7. Patches from the folks at Google:
1074    
1075          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1076          real life, but is still worth protecting against".
1077    
1078          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1079          regular expressions".
1080    
1081          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1082          have it.
1083    
1084          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1085          "configure" and the latter not, in order to fix a problem somebody had
1086          with compiling the Arg class on HP-UX.
1087    
1088          (e) Improve the error-handling of the C++ wrapper a little bit.
1089    
1090          (f) New tests for checking recursion limiting.
1091    
1092     8. The pcre_memmove() function, which is used only if the environment does not
1093        have a standard memmove() function (and is therefore rarely compiled),
1094        contained two bugs: (a) use of int instead of size_t, and (b) it was not
1095        returning a result (though PCRE never actually uses the result).
1096    
1097     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1098        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1099        returned instead of calling malloc() with an overflowing number that would
1100        most likely cause subsequent chaos.
1101    
1102    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1103    
1104    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1105        with this option is matched, the nmatch and pmatch options of regexec() are
1106        ignored.
1107    
1108    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1109        provided in case anyone wants to the the POSIX interface with UTF-8
1110        strings.
1111    
1112    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1113        C++ linking (needed for some HP-UX environments).
1114    
1115    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1116        (unused parameter) and in the pcre_printint() function (omitted "default"
1117        switch label when the default is to do nothing).
1118    
1119    15. Added some code to make it possible, when PCRE is compiled as a C++
1120        library, to replace subject pointers for pcre_exec() with a smart pointer
1121        class, thus making it possible to process discontinuous strings.
1122    
1123    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1124        much the same function. They were added by different people who were trying
1125        to make PCRE easy to compile on non-Unix systems. It has been suggested
1126        that PCRE_EXPORT be abolished now that there is more automatic apparatus
1127        for compiling on Windows systems. I have therefore replaced it with
1128        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1129        defaults to "extern" for C or "extern C" for C++, which works fine on
1130        Unix-like systems. It is now possible to override the value of PCRE_DATA_
1131        SCOPE with something explicit in config.h. In addition:
1132    
1133        (a) pcreposix.h still had just "extern" instead of either of these macros;
1134            I have replaced it with PCRE_DATA_SCOPE.
1135    
1136        (b) Functions such as _pcre_xclass(), which are internal to the library,
1137            but external in the C sense, all had PCRE_EXPORT in their definitions.
1138            This is apparently wrong for the Windows case, so I have removed it.
1139            (It makes no difference on Unix-like systems.)
1140    
1141    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1142        of recursive calls to match(). This is different to MATCH_LIMIT because
1143        that limits the total number of calls to match(), not all of which increase
1144        the depth of recursion. Limiting the recursion depth limits the amount of
1145        stack (or heap if NO_RECURSE is set) that is used. The default can be set
1146        when PCRE is compiled, and changed at run time. A patch from Google adds
1147        this functionality to the C++ interface.
1148    
1149    18. Changes to the handling of Unicode character properties:
1150    
1151        (a) Updated the table to Unicode 4.1.0.
1152    
1153        (b) Recognize characters that are not in the table as "Cn" (undefined).
1154    
1155        (c) I revised the way the table is implemented to a much improved format
1156            which includes recognition of ranges. It now supports the ranges that
1157            are defined in UnicodeData.txt, and it also amalgamates other
1158            characters into ranges. This has reduced the number of entries in the
1159            table from around 16,000 to around 3,000, thus reducing its size
1160            considerably. I realized I did not need to use a tree structure after
1161            all - a binary chop search is just as efficient. Having reduced the
1162            number of entries, I extended their size from 6 bytes to 8 bytes to
1163            allow for more data.
1164    
1165        (d) Added support for Unicode script names via properties such as \p{Han}.
1166    
1167    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1168        matching that character.
1169    
1170    20. When matching a repeated Unicode property with a minimum greater than zero,
1171        (for example \pL{2,}), PCRE could look past the end of the subject if it
1172        reached it while seeking the minimum number of characters. This could
1173        happen only if some of the characters were more than one byte long, because
1174        there is a check for at least the minimum number of bytes.
1175    
1176    21. Refactored the implementation of \p and \P so as to be more general, to
1177        allow for more different types of property in future. This has changed the
1178        compiled form incompatibly. Anybody with saved compiled patterns that use
1179        \p or \P will have to recompile them.
1180    
1181    22. Added "Any" and "L&" to the supported property types.
1182    
1183    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1184        but give a compile time error if the value is greater than 0xff.
1185    
1186    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1187        accidentally not being installed or uninstalled.
1188    
1189    25. The pcre.h file was built from pcre.h.in, but the only changes that were
1190        made were to insert the current release number. This seemed silly, because
1191        it made things harder for people building PCRE on systems that don't run
1192        "configure". I have turned pcre.h into a distributed file, no longer built
1193        by "configure", with the version identification directly included. There is
1194        no longer a pcre.h.in file.
1195    
1196        However, this change necessitated a change to the pcre-config script as
1197        well. It is built from pcre-config.in, and one of the substitutions was the
1198        release number. I have updated configure.ac so that ./configure now finds
1199        the release number by grepping pcre.h.
1200    
1201    26. Added the ability to run the tests under valgrind.
1202    
1203    
1204    Version 6.4 05-Sep-05
1205    ---------------------
1206    
1207     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1208        "--" to be printed when multiple files were scanned, even when none of the
1209        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1210        consider it to be a bug, and have restored the previous behaviour.
1211    
1212     2. A couple of code tidies to get rid of compiler warnings.
1213    
1214     3. The pcretest program used to cheat by referring to symbols in the library
1215        whose names begin with _pcre_. These are internal symbols that are not
1216        really supposed to be visible externally, and in some environments it is
1217        possible to suppress them. The cheating is now confined to including
1218        certain files from the library's source, which is a bit cleaner.
1219    
1220     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1221        file's purpose clearer.
1222    
1223     5. Reorganized pcre_ucp_findchar().
1224    
1225    
1226    Version 6.3 15-Aug-05
1227    ---------------------
1228    
1229     1. The file libpcre.pc.in did not have general read permission in the tarball.
1230    
1231     2. There were some problems when building without C++ support:
1232    
1233        (a) If C++ support was not built, "make install" and "make test" still
1234            tried to test it.
1235    
1236        (b) There were problems when the value of CXX was explicitly set. Some
1237            changes have been made to try to fix these, and ...
1238    
1239        (c) --disable-cpp can now be used to explicitly disable C++ support.
1240    
1241        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1242            backslash in a target when C++ was disabled. This confuses some
1243            versions of "make", apparently. Using an intermediate variable solves
1244            this. (Same for CPP_LOBJ.)
1245    
1246     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1247        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1248        necessary on certain architectures.
1249    
1250     4. Added a setting of -export-symbols-regex to the link command to remove
1251        those symbols that are exported in the C sense, but actually are local
1252        within the library, and not documented. Their names all begin with
1253        "_pcre_". This is not a perfect job, because (a) we have to except some
1254        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1255        available (and never for static libraries). I have made a note to try to
1256        find a way round (a) in the future.
1257    
1258    
1259    Version 6.2 01-Aug-05
1260    ---------------------
1261    
1262     1. There was no test for integer overflow of quantifier values. A construction
1263        such as {1111111111111111} would give undefined results. What is worse, if
1264        a minimum quantifier for a parenthesized subpattern overflowed and became
1265        negative, the calculation of the memory size went wrong. This could have
1266        led to memory overwriting.
1267    
1268     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1269    
1270     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1271        operating environments where this matters.
1272    
1273     4. Applied Giuseppe Maxia's patch to add additional features for controlling
1274        PCRE options from within the C++ wrapper.
1275    
1276     5. Named capturing subpatterns were not being correctly counted when a pattern
1277        was compiled. This caused two problems: (a) If there were more than 100
1278        such subpatterns, the calculation of the memory needed for the whole
1279        compiled pattern went wrong, leading to an overflow error. (b) Numerical
1280        back references of the form \12, where the number was greater than 9, were
1281        not recognized as back references, even though there were sufficient
1282        previous subpatterns.
1283    
1284     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1285        versions of gcc, e.g. 2.95.4.
1286    
1287    
1288    Version 6.1 21-Jun-05
1289    ---------------------
1290    
1291     1. There was one reference to the variable "posix" in pcretest.c that was not
1292        surrounded by "#if !defined NOPOSIX".
1293    
1294     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1295        the cross-check on the old pcre_info() function, for the benefit of the
1296        cut-down version of PCRE that is currently imported into Exim.
1297    
1298     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1299        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1300        patterns that start (?i) but I don't think that matters. The use of (?i) is
1301        just an example; this all applies to the other options as well.
1302    
1303     4. Since libtool seems to echo the compile commands it is issuing, the output
1304        from "make" can be reduced a bit by putting "@" in front of each libtool
1305        compile command.
1306    
1307     5. Patch from the folks at Google for configure.in to be a bit more thorough
1308        in checking for a suitable C++ installation before trying to compile the
1309        C++ stuff. This should fix a reported problem when a compiler was present,
1310        but no suitable headers.
1311    
1312     6. The man pages all had just "PCRE" as their title. I have changed them to
1313        be the relevant file name. I have also arranged that these names are
1314        retained in the file doc/pcre.txt, which is a concatenation in text format
1315        of all the man pages except the little individual ones for each function.
1316    
1317     7. The NON-UNIX-USE file had not been updated for the different set of source
1318        files that come with release 6. I also added a few comments about the C++
1319        wrapper.
1320    
1321    
1322    Version 6.0 07-Jun-05
1323    ---------------------
1324    
1325     1. Some minor internal re-organization to help with my DFA experiments.
1326    
1327     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1328        didn't matter for the library itself when fully configured, but did matter
1329        when compiling without UCP support, or within Exim, where the ucp files are
1330        not imported.
1331    
1332     3. Refactoring of the library code to split up the various functions into
1333        different source modules. The addition of the new DFA matching code (see
1334        below) to a single monolithic source would have made it really too
1335        unwieldy, quite apart from causing all the code to be include in a
1336        statically linked application, when only some functions are used. This is
1337        relevant even without the DFA addition now that patterns can be compiled in
1338        one application and matched in another.
1339    
1340        The downside of splitting up is that there have to be some external
1341        functions and data tables that are used internally in different modules of
1342        the library but which are not part of the API. These have all had their
1343        names changed to start with "_pcre_" so that they are unlikely to clash
1344        with other external names.
1345    
1346     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1347        a different (DFA) algorithm. Although it is slower than the original
1348        function, it does have some advantages for certain types of matching
1349        problem.
1350    
1351     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1352        including restarting after a partial match.
1353    
1354     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1355        defined when compiling for Windows was sent to me. I have put it into the
1356        code, though I have no means of testing or verifying it.
1357    
1358     7. Added the pcre_refcount() auxiliary function.
1359    
1360     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1361        match before or at the first newline in the subject string. In pcretest,
1362        the /f option on a pattern can be used to set this.
1363    
1364     9. A repeated \w when used in UTF-8 mode with characters greater than 256
1365        would behave wrongly. This has been present in PCRE since release 4.0.
1366    
1367    10. A number of changes to the pcregrep command:
1368    
1369        (a) Refactored how -x works; insert ^(...)$ instead of setting
1370            PCRE_ANCHORED and checking the length, in preparation for adding
1371            something similar for -w.
1372    
1373        (b) Added the -w (match as a word) option.
1374    
1375        (c) Refactored the way lines are read and buffered so as to have more
1376            than one at a time available.
1377    
1378        (d) Implemented a pcregrep test script.
1379    
1380        (e) Added the -M (multiline match) option. This allows patterns to match
1381            over several lines of the subject. The buffering ensures that at least
1382            8K, or the rest of the document (whichever is the shorter) is available
1383            for matching (and similarly the previous 8K for lookbehind assertions).
1384    
1385        (f) Changed the --help output so that it now says
1386    
1387              -w, --word-regex(p)
1388    
1389            instead of two lines, one with "regex" and the other with "regexp"
1390            because that confused at least one person since the short forms are the
1391            same. (This required a bit of code, as the output is generated
1392            automatically from a table. It wasn't just a text change.)
1393    
1394        (g) -- can be used to terminate pcregrep options if the next thing isn't an
1395            option but starts with a hyphen. Could be a pattern or a path name
1396            starting with a hyphen, for instance.
1397    
1398        (h) "-" can be given as a file name to represent stdin.
1399    
1400        (i) When file names are being printed, "(standard input)" is used for
1401            the standard input, for compatibility with GNU grep. Previously
1402            "<stdin>" was used.
1403    
1404        (j) The option --label=xxx can be used to supply a name to be used for
1405            stdin when file names are being printed. There is no short form.
1406    
1407        (k) Re-factored the options decoding logic because we are going to add
1408            two more options that take data. Such options can now be given in four
1409            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1410    
1411        (l) Added the -A, -B, and -C options for requesting that lines of context
1412            around matches be printed.
1413    
1414        (m) Added the -L option to print the names of files that do not contain
1415            any matching lines, that is, the complement of -l.
1416    
1417        (n) The return code is 2 if any file cannot be opened, but pcregrep does
1418            continue to scan other files.
1419    
1420        (o) The -s option was incorrectly implemented. For compatibility with other
1421            greps, it now suppresses the error message for a non-existent or non-
1422            accessible file (but not the return code). There is a new option called
1423            -q that suppresses the output of matching lines, which was what -s was
1424            previously doing.
1425    
1426        (p) Added --include and --exclude options to specify files for inclusion
1427            and exclusion when recursing.
1428    
1429    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1430        Hopefully, it now does.
1431    
1432    12. Missing cast in pcre_study().
1433    
1434    13. Added an "uninstall" target to the makefile.
1435    
1436    14. Replaced "extern" in the function prototypes in Makefile.in with
1437        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1438        world, but is set differently for Windows.
1439    
1440    15. Added a second compiling function called pcre_compile2(). The only
1441        difference is that it has an extra argument, which is a pointer to an
1442        integer error code. When there is a compile-time failure, this is set
1443        non-zero, in addition to the error test pointer being set to point to an
1444        error message. The new argument may be NULL if no error number is required
1445        (but then you may as well call pcre_compile(), which is now just a
1446        wrapper). This facility is provided because some applications need a
1447        numeric error indication, but it has also enabled me to tidy up the way
1448        compile-time errors are handled in the POSIX wrapper.
1449    
1450    16. Added VPATH=.libs to the makefile; this should help when building with one
1451        prefix path and installing with another. (Or so I'm told by someone who
1452        knows more about this stuff than I do.)
1453    
1454    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1455        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1456        match everything, including newlines. This is not POSIX-compatible, but
1457        somebody wanted the feature. From pcretest it can be activated by using
1458        both the P and the s flags.
1459    
1460    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1461    
1462    19. libpcre.pc was being incorrectly installed as executable.
1463    
1464    20. A couple of places in pcretest check for end-of-line by looking for '\n';
1465        it now also looks for '\r' so that it will work unmodified on Windows.
1466    
1467    21. Added Google's contributed C++ wrapper to the distribution.
1468    
1469    22. Added some untidy missing memory free() calls in pcretest, to keep
1470        Electric Fence happy when testing.
1471    
1472    
1473    
1474    Version 5.0 13-Sep-04
1475    ---------------------
1476    
1477     1. Internal change: literal characters are no longer packed up into items
1478        containing multiple characters in a single byte-string. Each character
1479        is now matched using a separate opcode. However, there may be more than one
1480        byte in the character in UTF-8 mode.
1481    
1482     2. The pcre_callout_block structure has two new fields: pattern_position and
1483        next_item_length. These contain the offset in the pattern to the next match
1484        item, and its length, respectively.
1485    
1486     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1487        insertion of callouts before each pattern item. Added the /C option to
1488        pcretest to make use of this.
1489    
1490     4. On the advice of a Windows user, the lines
1491    
1492          #if defined(_WIN32) || defined(WIN32)
1493          _setmode( _fileno( stdout ), 0x8000 );
1494          #endif  /* defined(_WIN32) || defined(WIN32) */
1495    
1496        have been added to the source of pcretest. This apparently does useful
1497        magic in relation to line terminators.
1498    
1499     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1500        for the benefit of those environments where the "b" makes a difference.
1501    
1502     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1503        to know about it. I have put a hack into configure.in that adds in code
1504        to set GCC=yes if CC=icc. This seems to end up at a point in the
1505        generated configure script that is early enough to affect the setting of
1506        compiler options, which is what is needed, but I have no means of testing
1507        whether it really works. (The user who reported this had patched the
1508        generated configure script, which of course I cannot do.)
1509    
1510        LATER: After change 22 below (new libtool files), the configure script
1511        seems to know about icc (and also ecc). Therefore, I have commented out
1512        this hack in configure.in.
1513    
1514     7. Added support for pkg-config (2 patches were sent in).
1515    
1516     8. Negated POSIX character classes that used a combination of internal tables
1517        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1518        [[:^ascii]]. Typically, they would match almost any characters. The other
1519        POSIX classes were not broken in this way.
1520    
1521     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1522        to find the match, as PCRE was deluded into thinking that the match had to
1523        start at the start point or following a newline. The same bug applied to
1524        patterns with negative forward assertions or any backward assertions
1525        preceding ".*" at the start, unless the pattern required a fixed first
1526        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1527    
1528    10. In UTF-8 mode, when moving forwards in the subject after a failed match
1529        starting at the last subject character, bytes beyond the end of the subject
1530        string were read.
1531    
1532    11. Renamed the variable "class" as "classbits" to make life easier for C++
1533        users. (Previously there was a macro definition, but it apparently wasn't
1534        enough.)
1535    
1536    12. Added the new field "tables" to the extra data so that tables can be passed
1537        in at exec time, or the internal tables can be re-selected. This allows
1538        a compiled regex to be saved and re-used at a later time by a different
1539        program that might have everything at different addresses.
1540    
1541    13. Modified the pcre-config script so that, when run on Solaris, it shows a
1542        -R library as well as a -L library.
1543    
1544    14. The debugging options of pcretest (-d on the command line or D on a
1545        pattern) showed incorrect output for anything following an extended class
1546        that contained multibyte characters and which was followed by a quantifier.
1547    
1548    15. Added optional support for general category Unicode character properties
1549        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1550        support. It adds about 90K to the size of the library. The meanings of the
1551        inbuilt class escapes such as \d and \s have NOT been changed.
1552    
1553    16. Updated pcredemo.c to include calls to free() to release the memory for the
1554        compiled pattern.
1555    
1556    17. The generated file chartables.c was being created in the source directory
1557        instead of in the building directory. This caused the build to fail if the
1558        source directory was different from the building directory, and was
1559        read-only.
1560    
1561    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1562        file. No doubt somebody will tell me if they don't make sense... Also added
1563        Dan Mooney's comments about building on OpenVMS.
1564    
1565    19. Added support for partial matching via the PCRE_PARTIAL option for
1566        pcre_exec() and the \P data escape in pcretest.
1567    
1568    20. Extended pcretest with 3 new pattern features:
1569    
1570        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
1571              write the compiled pattern to the file whose name is "rest-of-line".
1572              This is a straight binary dump of the data, with the saved pointer to
1573              the character tables forced to be NULL. The study data, if any, is
1574              written too. After writing, pcretest reads a new pattern.
1575    
1576        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1577              compiled pattern from the given file. There must not be any
1578              occurrences of "<" in the file name (pretty unlikely); if there are,
1579              pcretest will instead treat the initial "<" as a pattern delimiter.
1580              After reading in the pattern, pcretest goes on to read data lines as
1581              usual.
1582    
1583        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1584              and 16-bit fields in a compiled pattern, to simulate a pattern that
1585              was compiled on a host of opposite endianness.
1586    
1587    21. The pcre-exec() function can now cope with patterns that were compiled on
1588        hosts of opposite endianness, with this restriction:
1589    
1590          As for any compiled expression that is saved and used later, the tables
1591          pointer field cannot be preserved; the extra_data field in the arguments
1592          to pcre_exec() should be used to pass in a tables address if a value
1593          other than the default internal tables were used at compile time.
1594    
1595    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1596        now diagnosed as an error. Previously, most of the time, a negative number
1597        would have been treated as zero, but if in addition "ovector" was passed as
1598        NULL, a crash could occur.
1599    
1600    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1601        new versions from the libtool 1.5 distribution (the last one is a copy of
1602        a file called libtool.m4). This seems to have fixed the need to patch
1603        "configure" to support Darwin 1.3 (which I used to do). However, I still
1604        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1605        workstation).
1606    
1607    24. Changed the PCRE licence to be the more standard "BSD" licence.
1608    
1609    
1610    Version 4.5 01-Dec-03
1611    ---------------------
1612    
1613     1. There has been some re-arrangement of the code for the match() function so
1614        that it can be compiled in a version that does not call itself recursively.
1615        Instead, it keeps those local variables that need separate instances for
1616        each "recursion" in a frame on the heap, and gets/frees frames whenever it
1617        needs to "recurse". Keeping track of where control must go is done by means
1618        of setjmp/longjmp. The whole thing is implemented by a set of macros that
1619        hide most of the details from the main code, and operates only if
1620        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1621        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1622        operating.
1623    
1624        To make it easier for callers to provide specially tailored get/free
1625        functions for this usage, two new functions, pcre_stack_malloc, and
1626        pcre_stack_free, are used. They are always called in strict stacking order,
1627        and the size of block requested is always the same.
1628    
1629        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1630        PCRE has been compiled to use the stack or the heap for recursion. The
1631        -C option of pcretest uses this to show which version is compiled.
1632    
1633        A new data escape \S, is added to pcretest; it causes the amounts of store
1634        obtained and freed by both kinds of malloc/free at match time to be added
1635        to the output.
1636    
1637     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1638        what's available on my current Linux desktop machine.
1639    
1640     3. When matching a UTF-8 string, the test for a valid string at the start has
1641        been extended. If start_offset is not zero, PCRE now checks that it points
1642        to a byte that is the start of a UTF-8 character. If not, it returns
1643        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1644        this is necessary because there may be backward assertions in the pattern.
1645        When matching the same subject several times, it may save resources to use
1646        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1647    
1648     4. The code for checking the validity of UTF-8 strings has been tightened so
1649        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1650        containing "overlong sequences".
1651    
1652     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1653        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1654        should have been "&", but it just so happened that all the cases this let
1655        through by mistake were picked up later in the function.
1656    
1657     6. I had used a variable called "isblank" - this is a C99 function, causing
1658        some compilers to warn. To avoid this, I renamed it (as "blankclass").
1659    
1660     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1661        prompting; (b) run "./pcretest /dev/null" at the start of the test script
1662        so the version is shown; (c) stop "make test" echoing "./RunTest".
1663    
1664     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1665    
1666     9. The prototype for memmove() for systems that don't have it was using
1667        size_t, but the inclusion of the header that defines size_t was later. I've
1668        moved the #includes for the C headers earlier to avoid this.
1669    
1670    10. Added some adjustments to the code to make it easier to compiler on certain
1671        special systems:
1672    
1673          (a) Some "const" qualifiers were missing.
1674          (b) Added the macro EXPORT before all exported functions; by default this
1675              is defined to be empty.
1676          (c) Changed the dftables auxiliary program (that builds chartables.c) so
1677              that it reads its output file name as an argument instead of writing
1678              to the standard output and assuming this can be redirected.
1679    
1680    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
1681        class containing characters with values greater than 255, PCRE compilation
1682        went into a loop.
1683    
1684    12. A recursive reference to a subpattern that was within another subpattern
1685        that had a minimum quantifier of zero caused PCRE to crash. For example,
1686        (x(y(?2))z)? provoked this bug with a subject that got as far as the
1687        recursion. If the recursively-called subpattern itself had a zero repeat,
1688        that was OK.
1689    
1690    13. In pcretest, the buffer for reading a data line was set at 30K, but the
1691        buffer into which it was copied (for escape processing) was still set at
1692        1024, so long lines caused crashes.
1693    
1694    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
1695        "internal error: code overflow...". This applied to any character class
1696        that was followed by a possessive quantifier.
1697    
1698    15. Modified the Makefile to add libpcre.la as a prerequisite for
1699        libpcreposix.la because I was told this is needed for a parallel build to
1700        work.
1701    
1702    16. If a pattern that contained .* following optional items at the start was
1703        studied, the wrong optimizing data was generated, leading to matching
1704        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
1705        matching string must start with a or b or c. The correct conclusion for
1706        this pattern is that a match can start with any character.
1707    
1708    
1709    Version 4.4 13-Aug-03
1710    ---------------------
1711    
1712     1. In UTF-8 mode, a character class containing characters with values between
1713        127 and 255 was not handled correctly if the compiled pattern was studied.
1714        In fixing this, I have also improved the studying algorithm for such
1715        classes (slightly).
1716    
1717     2. Three internal functions had redundant arguments passed to them. Removal
1718        might give a very teeny performance improvement.
1719    
1720     3. Documentation bug: the value of the capture_top field in a callout is *one
1721        more than* the number of the hightest numbered captured substring.
1722    
1723     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
1724        in incorrectly linking with a previously installed version. They now link
1725        explicitly with libpcre.la.
1726    
1727     5. configure.in no longer needs to recognize Cygwin specially.
1728    
1729     6. A problem in pcre.in for Windows platforms is fixed.
1730    
1731     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
1732        pcretest, it used to include the size of the study block as part of its
1733        output. Unfortunately, the structure contains a field that has a different
1734        size on different hardware architectures. This meant that the tests that
1735        showed this size failed. As the block is currently always of a fixed size,
1736        this information isn't actually particularly useful in pcretest output, so
1737        I have just removed it.
1738    
1739     8. Three pre-processor statements accidentally did not start in column 1.
1740        Sadly, there are *still* compilers around that complain, even though
1741        standard C has not required this for well over a decade. Sigh.
1742    
1743     9. In pcretest, the code for checking callouts passed small integers in the
1744        callout_data field, which is a void * field. However, some picky compilers
1745        complained about the casts involved for this on 64-bit systems. Now
1746        pcretest passes the address of the small integer instead, which should get
1747        rid of the warnings.
1748    
1749    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
1750        both compile and run time, and gives an error if an invalid UTF-8 sequence
1751        is found. There is a option for disabling this check in cases where the
1752        string is known to be correct and/or the maximum performance is wanted.
1753    
1754    11. In response to a bug report, I changed one line in Makefile.in from
1755    
1756            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
1757        to
1758            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
1759    
1760        to look similar to other lines, but I have no way of telling whether this
1761        is the right thing to do, as I do not use Windows. No doubt I'll get told
1762        if it's wrong...
1763    
1764    
1765    Version 4.3 21-May-03
1766    ---------------------
1767    
1768    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
1769       Makefile.
1770    
1771    2. Some refactoring to improve the quality of the code:
1772    
1773       (i)   The utf8_table... variables are now declared "const".
1774    
1775       (ii)  The code for \cx, which used the "case flipping" table to upper case
1776             lower case letters, now just substracts 32. This is ASCII-specific,
1777             but the whole concept of \cx is ASCII-specific, so it seems
1778             reasonable.
1779    
1780       (iii) PCRE was using its character types table to recognize decimal and
1781             hexadecimal digits in the pattern. This is silly, because it handles
1782             only 0-9, a-f, and A-F, but the character types table is locale-
1783             specific, which means strange things might happen. A private
1784             table is now used for this - though it costs 256 bytes, a table is
1785             much faster than multiple explicit tests. Of course, the standard
1786             character types table is still used for matching digits in subject
1787             strings against \d.
1788    
1789       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
1790             ending in _t are). So I've renamed it as ESC_tee.
1791    
1792    3. The first argument for regexec() in the POSIX wrapper should have been
1793       defined as "const".
1794    
1795    4. Changed pcretest to use malloc() for its buffers so that they can be
1796       Electric Fenced for debugging.
1797    
1798    5. There were several places in the code where, in UTF-8 mode, PCRE would try
1799       to read one or more bytes before the start of the subject string. Often this
1800       had no effect on PCRE's behaviour, but in some circumstances it could
1801       provoke a segmentation fault.
1802    
1803    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
1804       to try to read one or more bytes before the start of the subject string.
1805    
1806    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
1807       UTF-8 support could misbehave in various ways if the subject string
1808       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
1809       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
1810       back over UTF-8 characters.)
1811    
1812    
1813    Version 4.2 14-Apr-03
1814    ---------------------
1815    
1816    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
1817    
1818    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
1819         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
1820         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
1821         [WIN_PREFIX]: new variable, "cyg" for Cygwin
1822         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
1823           and BUILD_EXEEXT
1824         Note: automatic setting of the BUILD variables is not yet working
1825         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
1826           compile-time but not at link-time
1827         [LINK]: use for linking executables only
1828         make different versions for Windows and non-Windows
1829         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
1830           libraries
1831         [LINK_FOR_BUILD]: new variable
1832         [OBJEXT]: use throughout
1833         [EXEEXT]: use throughout
1834         <winshared>: new target
1835         <wininstall>: new target
1836         <dftables.o>: use native compiler
1837         <dftables>: use native linker
1838         <install>: handle Windows platform correctly
1839         <clean>: ditto
1840         <check>: ditto
1841         copy DLL to top builddir before testing
1842    
1843       As part of these changes, -no-undefined was removed again. This was reported
1844       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
1845       in any case.
1846    
1847    3. Some tidies to get rid of compiler warnings:
1848    
1849       . In the match_data structure, match_limit was an unsigned long int, whereas
1850         match_call_count was an int. I've made them both unsigned long ints.
1851    
1852       . In pcretest the fact that a const uschar * doesn't automatically cast to
1853         a void * provoked a warning.
1854    
1855       . Turning on some more compiler warnings threw up some "shadow" variables
1856         and a few more missing casts.
1857    
1858    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
1859       option, a class that contained a single character with a value between 128
1860       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
1861    
1862    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
1863       option, a class that contained several characters, but with at least one
1864       whose value was between 128 and 255 caused PCRE to crash.
1865    
1866    
1867    Version 4.1 12-Mar-03
1868    ---------------------
1869    
1870    1. Compiling with gcc -pedantic found a couple of places where casts were
1871    needed, and a string in dftables.c that was longer than standard compilers are
1872    required to support.
1873    
1874    2. Compiling with Sun's compiler found a few more places where the code could
1875    be tidied up in order to avoid warnings.
1876    
1877    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
1878    first of these names is deprecated in the latest Autoconf in favour of the name
1879    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
1880    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
1881    analogy I have changed it to CFLAGS_FOR_BUILD.
1882    
1883    4. Added -no-undefined to the linking command in the Makefile, because this is
1884    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
1885    linking step for the pcreposix library.
1886    
1887    5. PCRE was failing to diagnose the case of two named groups with the same
1888    name.
1889    
1890    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
1891    literal character that is needed in the subject for a match, and scans along to
1892    ensure that it is present before embarking on the full matching process. This
1893    saves time in cases of nested unlimited repeats that are never going to match.
1894    Problem: the scan can take a lot of time if the subject is very long (e.g.
1895    megabytes), thus penalizing straightforward matches. It is now done only if the
1896    amount of subject to be scanned is less than 1000 bytes.
1897    
1898    7. A lesser problem with the same optimization is that it was recording the
1899    first character of an anchored pattern as "needed", thus provoking a search
1900    right along the subject, even when the first match of the pattern was going to
1901    fail. The "needed" character is now not set for anchored patterns, unless it
1902    follows something in the pattern that is of non-fixed length. Thus, it still
1903    fulfils its original purpose of finding quick non-matches in cases of nested
1904    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
1905    
1906    
1907    Version 4.0 17-Feb-03
1908    ---------------------
1909    
1910    1. If a comment in an extended regex that started immediately after a meta-item
1911    extended to the end of string, PCRE compiled incorrect data. This could lead to
1912    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
1913    
1914    2. Moved to autoconf 2.53 and libtool 1.4.2.
1915    
1916    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
1917    the special perltest8 script is no longer needed - all the tests can be run
1918    from a single perltest script.
1919    
1920    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
1921    by \s. It has now been removed in PCRE. This means it isn't recognized as
1922    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
1923    class [:space:] *does* include VT, thereby creating a mess.
1924    
1925    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
1926    space and tab.
1927    
1928    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
1929    its new features into the main test script, reducing the number of scripts.
1930    
1931    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
1932    were backward compatible, and made the (?i) apply to the whole pattern, as if
1933    /i were given. Now it behaves more logically, and applies the option setting
1934    only to what follows. PCRE has been changed to follow suit. However, if it
1935    finds options settings right at the start of the pattern, it extracts them into
1936    the global options, as before. Thus, they show up in the info data.
1937    
1938    8. Added support for the \Q...\E escape sequence. Characters in between are
1939    treated as literals. This is slightly different from Perl in that $ and @ are
1940    also handled as literals inside the quotes. In Perl, they will cause variable
1941    interpolation. Note the following examples:
1942    
1943        Pattern            PCRE matches      Perl matches
1944    
1945        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
1946        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
1947        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
1948    
1949    For compatibility with Perl, \Q...\E sequences are recognized inside character
1950    classes as well as outside them.
1951    
1952    9. Re-organized 3 code statements in pcretest to avoid "overflow in
1953    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
1954    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
1955    signed/unsigned warnings.
1956    
1957    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
1958    option for pcretest, so I've replaced it by a simple function that does just
1959    that job.
1960    
1961    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
1962    "pcregrep -".
1963    
1964    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
1965    Java package. This provides some syntactic sugar for simple cases of what my
1966    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
1967    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
1968    item, you can use this simplified notation. Note that only makes sense with
1969    greedy quantifiers. Consequently, the use of the possessive quantifier forces
1970    greediness, whatever the setting of the PCRE_UNGREEDY option.
1971    
1972    13. A change of greediness default within a pattern was not taking effect at
1973    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
1974    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
1975    was abstracted outside.
1976    
1977    14. PCRE now supports the \G assertion. It is true when the current matching
1978    position is at the start point of the match. This differs from \A when the
1979    starting offset is non-zero. Used with the /g option of pcretest (or similar
1980    code), it works in the same way as it does for Perl's /g option. If all
1981    alternatives of a regex begin with \G, the expression is anchored to the start
1982    match position, and the "anchored" flag is set in the compiled expression.
1983    
1984    15. Some bugs concerning the handling of certain option changes within patterns
1985    have been fixed. These applied to options other than (?ims). For example,
1986    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
1987    been the other way round. Some of this was related to change 7 above.
1988    
1989    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
1990    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
1991    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
1992    POSIX classes only within a class (e.g. /[[:alpha:]]/).
1993    
1994    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
1995    mode. Unlike ".", it always matches newline, whatever the setting of
1996    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
1997    assertions. Perl allows it, but it doesn't (in general) work because it can't
1998    calculate the length of the lookbehind. At least, that's the case for Perl
1999    5.8.0 - I've been told they are going to document that it doesn't work in
2000    future.
2001    
2002    18. Added an error diagnosis for escapes that PCRE does not support: these are
2003    \L, \l, \N, \P, \p, \U, \u, and \X.
2004    
2005    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
2006    reading past the end of the pattern in cases such as /[abcd/.
2007    
2008    20. PCRE was getting more memory than necessary for patterns with classes that
2009    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
2010    
2011    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
2012    compiling PCRE for use with Virtual Pascal.
2013    
2014    22. Small fix to the Makefile to make it work properly if the build is done
2015    outside the source tree.
2016    
2017    23. Added a new extension: a condition to go with recursion. If a conditional
2018    subpattern starts with (?(R) the "true" branch is used if recursion has
2019    happened, whereas the "false" branch is used only at the top level.
2020    
2021    24. When there was a very long string of literal characters (over 255 bytes
2022    without UTF support, over 250 bytes with UTF support), the computation of how
2023    much memory was required could be incorrect, leading to segfaults or other
2024    strange effects.
2025    
2026    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
2027    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
2028    there was a subsequent back reference to those brackets. This meant that, for
2029    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
2030    possible to check for precisely this case. All we can do is abandon the
2031    optimization if .* occurs inside capturing brackets when there are any back
2032    references whatsoever. (See below for a better fix that came later.)
2033    
2034    26. The handling of the optimization for finding the first character of a
2035    non-anchored pattern, and for finding a character that is required later in the
2036    match were failing in some cases. This didn't break the matching; it just
2037    failed to optimize when it could. The way this is done has been re-implemented.
2038    
2039    27. Fixed typo in error message for invalid (?R item (it said "(?p").
2040    
2041    28. Added a new feature that provides some of the functionality that Perl
2042    provides with (?{...}). The facility is termed a "callout". The way it is done
2043    in PCRE is for the caller to provide an optional function, by setting
2044    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
2045    global variable. By default it is unset, which disables all calling out. To get
2046    the function called, the regex must include (?C) at appropriate points. This
2047    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
2048    This provides a means of identifying different callout points. When PCRE
2049    reaches such a point in the regex, if pcre_callout has been set, the external
2050    function is called. It is provided with data in a structure called
2051    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
2052    matching continues; if it returns a non-zero value, the match at the current
2053    point fails. However, backtracking will occur if possible. [This was changed
2054    later and other features added - see item 49 below.]
2055    
2056    29. pcretest is upgraded to test the callout functionality. It provides a
2057    callout function that displays information. By default, it shows the start of
2058    the match and the current position in the text. There are some new data escapes
2059    to vary what happens:
2060    
2061        \C+         in addition, show current contents of captured substrings
2062        \C-         do not supply a callout function
2063        \C!n        return 1 when callout number n is reached
2064        \C!n!m      return 1 when callout number n is reached for the mth time
2065    
2066    30. If pcregrep was called with the -l option and just a single file name, it
2067    output "<stdin>" if a match was found, instead of the file name.
2068    
2069    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
2070    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
2071    pcre_exec(). This saves a malloc/free per call. The default value of
2072    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
2073    when configuring.
2074    
2075    32. The default maximum size of a compiled pattern is 64K. There have been a
2076    few cases of people hitting this limit. The code now uses macros to handle the
2077    storing of links as offsets within the compiled pattern. It defaults to 2-byte
2078    links, but this can be changed to 3 or 4 bytes by --with-link-size when
2079    configuring. Tests 2 and 5 work only with 2-byte links because they output
2080    debugging information about compiled patterns.
2081    
2082    33. Internal code re-arrangements:
2083    
2084    (a) Moved the debugging function for printing out a compiled regex into
2085        its own source file (printint.c) and used #include to pull it into
2086        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
2087        separate copies.
2088    
2089    (b) Defined the list of op-code names for debugging as a macro in
2090        internal.h so that it is next to the definition of the opcodes.
2091    
2092    (c) Defined a table of op-code lengths for simpler skipping along compiled
2093        code. This is again a macro in internal.h so that it is next to the
2094        definition of the opcodes.
2095    
2096    34. Added support for recursive calls to individual subpatterns, along the
2097    lines of Robin Houston's patch (but implemented somewhat differently).
2098    
2099    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
2100    allow it to read and process whole directories in Win32. This code was
2101    contributed by Lionel Fourquaux; it has not been tested by me.
2102    
2103    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
2104    used to name a group. Names consist of alphanumerics and underscores, and must
2105    be unique. Back references use the syntax (?P=name) and recursive calls use
2106    (?P>name) which is a PCRE extension to the Python extension. Groups still have
2107    numbers. The function pcre_fullinfo() can be used after compilation to extract
2108    a name/number map. There are three relevant calls:
2109    
2110      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
2111      PCRE_INFO_NAMECOUNT            yields the number of entries
2112      PCRE_INFO_NAMETABLE            yields a pointer to the map.
2113    
2114    The map is a vector of fixed-size entries. The size of each entry depends on
2115    the length of the longest name used. The first two bytes of each entry are the
2116    group number, most significant byte first. There follows the corresponding
2117    name, zero terminated. The names are in alphabetical order.
2118    
2119    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
2120    case instead of 255. Making it the same both with and without UTF-8 support
2121    means that the same test output works with both.
2122    
2123    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
2124    calling malloc() with a zero argument.
2125    
2126    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
2127    optimization. I've improved things by keeping a bitmap of backreferences with
2128    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
2129    fact referenced, the optimization can be applied. It is unlikely that a
2130    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
2131    the match to follow \n) will appear inside brackets with a number greater than
2132    31, but if it does, any back reference > 31 suppresses the optimization.
2133    
2134    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
2135    of disabling numbered capturing parentheses. Any opening parenthesis that is
2136    not followed by ? behaves as if it were followed by ?: but named parentheses
2137    can still be used for capturing (and they will acquire numbers in the usual
2138    way).
2139    
2140    41. Redesigned the return codes from the match() function into yes/no/error so
2141    that errors can be passed back from deep inside the nested calls. A malloc
2142    failure while inside a recursive subpattern call now causes the
2143    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
2144    
2145    42. It is now possible to set a limit on the number of times the match()
2146    function is called in a call to pcre_exec(). This facility makes it possible to
2147    limit the amount of recursion and backtracking, though not in a directly
2148    obvious way, because the match() function is used in a number of different
2149    circumstances. The count starts from zero for each position in the subject
2150    string (for non-anchored patterns). The default limit is, for compatibility, a
2151    large number, namely 10 000 000. You can change this in two ways:
2152    
2153    (a) When configuring PCRE before making, you can use --with-match-limit=n
2154        to set a default value for the compiled library.
2155    
2156    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
2157        a different value is set. See 45 below.
2158    
2159    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
2160    
2161    43. Added a new function pcre_config(int, void *) to enable run-time extraction
2162    of things that can be changed at compile time. The first argument specifies
2163    what is wanted and the second points to where the information is to be placed.
2164    The current list of available information is:
2165    
2166      PCRE_CONFIG_UTF8
2167    
2168    The output is an integer that is set to one if UTF-8 support is available;
2169    otherwise it is set to zero.
2170    
2171      PCRE_CONFIG_NEWLINE
2172    
2173    The output is an integer that it set to the value of the code that is used for
2174    newline. It is either LF (10) or CR (13).
2175    
2176      PCRE_CONFIG_LINK_SIZE
2177    
2178    The output is an integer that contains the number of bytes used for internal
2179    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
2180    
2181      PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
2182    
2183    The output is an integer that contains the threshold above which the POSIX
2184    interface uses malloc() for output vectors. See item 31 above.
2185    
2186      PCRE_CONFIG_MATCH_LIMIT
2187    
2188    The output is an unsigned integer that contains the default limit of the number
2189    of match() calls in a pcre_exec() execution. See 42 above.
2190    
2191    44. pcretest has been upgraded by the addition of the -C option. This causes it
2192    to extract all the available output from the new pcre_config() function, and to
2193    output it. The program then exits immediately.
2194    
2195    45. A need has arisen to pass over additional data with calls to pcre_exec() in
2196    order to support additional features. One way would have been to define
2197    pcre_exec2() (for example) with extra arguments, but this would not have been
2198    extensible, and would also have required all calls to the original function to
2199    be mapped to the new one. Instead, I have chosen to extend the mechanism that
2200    is used for passing in "extra" data from pcre_study().
2201    
2202    The pcre_extra structure is now exposed and defined in pcre.h. It currently
2203    contains the following fields:
2204    
2205      flags         a bitmap indicating which of the following fields are set
2206      study_data    opaque data from pcre_study()
2207      match_limit   a way of specifying a limit on match() calls for a specific
2208                      call to pcre_exec()
2209      callout_data  data for callouts (see 49 below)
2210    
2211    The flag bits are also defined in pcre.h, and are
2212    
2213      PCRE_EXTRA_STUDY_DATA
2214      PCRE_EXTRA_MATCH_LIMIT
2215      PCRE_EXTRA_CALLOUT_DATA
2216    
2217    The pcre_study() function now returns one of these new pcre_extra blocks, with
2218    the actual study data pointed to by the study_data field, and the
2219    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
2220    before. That is, this change is entirely upwards-compatible and requires no
2221    change to existing code.
2222    
2223    If you want to pass in additional data to pcre_exec(), you can either place it
2224    in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
2225    block.
2226    
2227    46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
2228    data string contains the escape sequence \M, pcretest calls pcre_exec() several
2229    times with different match limits, until it finds the minimum value needed for
2230    pcre_exec() to complete. The value is then output. This can be instructive; for
2231    most simple matches the number is quite small, but for pathological cases it
2232    gets very large very quickly.
2233    
2234    47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
2235    returns the size of the data block pointed to by the study_data field in a
2236    pcre_extra block, that is, the value that was passed as the argument to
2237    pcre_malloc() when PCRE was getting memory in which to place the information
2238    created by pcre_study(). The fourth argument should point to a size_t variable.
2239    pcretest has been extended so that this information is shown after a successful
2240    pcre_study() call when information about the compiled regex is being displayed.
2241    
2242    48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
2243    because what follows is always an absolute path. (Later: it turns out that this
2244    is more than cosmetic for MinGW, because it doesn't like empty path
2245    components.)
2246    
2247    49. Some changes have been made to the callout feature (see 28 above):
2248    
2249    (i)  A callout function now has three choices for what it returns:
2250    
2251           0  =>  success, carry on matching
2252         > 0  =>  failure at this point, but backtrack if possible
2253         < 0  =>  serious error, return this value from pcre_exec()
2254    
2255         Negative values should normally be chosen from the set of PCRE_ERROR_xxx
2256         values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
2257         "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
2258         use by callout functions. It will never be used by PCRE itself.
2259    
2260    (ii) The pcre_extra structure (see 45 above) has a void * field called
2261         callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
2262         pcre_callout_block structure has a field of the same name. The contents of
2263         the field passed in the pcre_extra structure are passed to the callout
2264         function in the corresponding field in the callout block. This makes it
2265         easier to use the same callout-containing regex from multiple threads. For
2266         testing, the pcretest program has a new data escape
2267    
2268           \C*n        pass the number n (may be negative) as callout_data
2269    
2270         If the callout function in pcretest receives a non-zero value as
2271         callout_data, it returns that value.
2272    
2273    50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
2274    there were some redundant $(CFLAGS) in commands that are now specified as
2275    $(LINK), which already includes $(CFLAGS).
2276    
2277    51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
2278    has been compiled with UTF-8 support *and* pcre_compile() has been compiled
2279    with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
2280    one-byte characters throughout. Note that case-insensitive matching applies
2281    only to characters whose values are less than 256. PCRE doesn't support the
2282    notion of cases for higher-valued characters.
2283    
2284    (i)   A character class whose characters are all within 0-255 is handled as
2285          a bit map, and the map is inverted for negative classes. Previously, a
2286          character > 255 always failed to match such a class; however it should
2287          match if the class was a negative one (e.g. [^ab]). This has been fixed.
2288    
2289    (ii)  A negated character class with a single character < 255 is coded as
2290          "not this character" (OP_NOT). This wasn't working properly when the test
2291          character was multibyte, either singly or repeated.
2292    
2293    (iii) Repeats of multibyte characters are now handled correctly in UTF-8
2294          mode, for example: \x{100}{2,3}.
2295    
2296    (iv)  The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
2297          singly or repeated) now correctly test multibyte characters. However,
2298          PCRE doesn't recognize any characters with values greater than 255 as
2299          digits, spaces, or word characters. Such characters always match \D, \S,
2300          and \W, and never match \d, \s, or \w.
2301    
2302    (v)   Classes may now contain characters and character ranges with values
2303          greater than 255. For example: [ab\x{100}-\x{400}].
2304    
2305    (vi)  pcregrep now has a --utf-8 option (synonym -u) which makes it call
2306          PCRE in UTF-8 mode.
2307    
2308    52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
2309    PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
2310    retained for backwards compatibility. (Note that LASTLITERAL is also a byte
2311    value.)
2312    
2313    53. The single man page has become too large. I have therefore split it up into
2314    a number of separate man pages. These also give rise to individual HTML pages;
2315    these are now put in a separate directory, and there is an index.html page that
2316    lists them all. Some hyperlinking between the pages has been installed.
2317    
2318    54. Added convenience functions for handling named capturing parentheses.
2319    
2320    55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2321    aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2322    true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2323    are faulted.
2324    
2325    56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2326    calling configure. These values are used when compiling the dftables.c program
2327    which is run to generate the source of the default character tables. They
2328    default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2329    you will need to set these values.
2330    
2331    57. Updated the building process for Windows DLL, as provided by Fred Cox.
2332    
2333    
2334    Version 3.9 02-Jan-02
2335    ---------------------
2336    
2337    1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2338    
2339    2. If --disable-static was given, the building process failed when trying to
2340    build pcretest and pcregrep. (For some reason it was using libtool to compile
2341    them, which is not right, as they aren't part of the library.)
2342    
2343    
2344    Version 3.8 18-Dec-01
2345    ---------------------
2346    
2347    1. The experimental UTF-8 code was completely screwed up. It was packing the
2348    bytes in the wrong order. How dumb can you get?
2349    
2350    
2351    Version 3.7 29-Oct-01
2352    ---------------------
2353    
2354    1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2355    This caused pcretest, when used on the test data, to segfault. Unfortunately,
2356    this didn't happen under Solaris 8, where I normally test things.
2357    
2358    2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2359    doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2360    isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2361    this fix an hour or so after the initial 3.7 release.)
2362    
2363    
2364  Version 3.6 23-Oct-01  Version 3.6 23-Oct-01
2365  ---------------------  ---------------------
2366    

Legend:
Removed from v.55  
changed lines
  Added in v.277

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12