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

Diff of /code/trunk/ChangeLog

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

revision 71 by nigel, Sat Feb 24 21:40:24 2007 UTC revision 272 by ph10, Sat Nov 17 16:53:53 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    
40    Version 7.4 21-Sep-07
41    ---------------------
42    
43    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
44        means that a class such as [\s] counted as "explicit reference to CR or
45        LF". That isn't really right - the whole point of the change was to try to
46        help when there was an actual mention of one of the two characters. So now
47        the change happens only if \r or \n (or a literal CR or LF) character is
48        encountered.
49    
50    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
51        of both had grown to the point where there were only 3 bits left.
52        Fortunately, there was spare space in the data structure, and so I have
53        moved the internal flags into a new 16-bit field to free up more option
54        bits.
55    
56    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
57        but did not set the internal JCHANGED flag - either of these is enough to
58        control the way the "get" function works - but the PCRE_INFO_JCHANGED
59        facility is supposed to tell if (?J) was ever used, so now (?J) at the
60        start sets both bits.
61    
62    4.  Added options (at build time, compile time, exec time) to change \R from
63        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
64    
65    5.  doc/pcresyntax.html was missing from the distribution.
66    
67    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
68        compatibility, even though it is no longer used.
69    
70    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
71        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
72        windows.h file is present (where different names are used). [This was
73        reversed later after testing - see 16 below.]
74    
75    8.  Changed all #include <config.h> to #include "config.h". There were also
76        some further <pcre.h> cases that I changed to "pcre.h".
77    
78    9.  When pcregrep was used with the --colour option, it missed the line ending
79        sequence off the lines that it output.
80    
81    10. It was pointed out to me that arrays of string pointers cause lots of
82        relocations when a shared library is dynamically loaded. A technique of
83        using a single long string with a table of offsets can drastically reduce
84        these. I have refactored PCRE in four places to do this. The result is
85        dramatic:
86    
87          Originally:                          290
88          After changing UCP table:            187
89          After changing error message table:   43
90          After changing table of "verbs"       36
91          After changing table of Posix names   22
92    
93        Thanks to the folks working on Gregex for glib for this insight.
94    
95    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
96        unicode-properties was also set.
97    
98    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
99    
100    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
101        checked only for CRLF.
102    
103    14. Added casts to pcretest.c to avoid compiler warnings.
104    
105    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
106    
107    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
108        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
109        entirely. This removes changes made in 7 above.
110    
111    17. The CMake files have been updated, and there is now more information about
112        building with CMake in the NON-UNIX-USE document.
113    
114    
115    Version 7.3 28-Aug-07
116    ---------------------
117    
118     1. In the rejigging of the build system that eventually resulted in 7.1, the
119        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
120        brackets there is not right, since it causes compilers to look for an
121        installed pcre.h, not the version that is in the source that is being
122        compiled (which of course may be different). I have changed it back to:
123    
124          #include "pcre.h"
125    
126        I have a vague recollection that the change was concerned with compiling in
127        different directories, but in the new build system, that is taken care of
128        by the VPATH setting the Makefile.
129    
130     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
131        when the subject happened to end in the byte 0x85 (e.g. if the last
132        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
133        characters but of course it shouldn't be taken as a newline when it is part
134        of another character. The bug was that, for an unlimited repeat of . in
135        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
136        characters when looking for a newline.
137    
138     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
139    
140     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
141        in debug output.
142    
143     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
144        long printing in the pcrecpp unittest when running under MinGW.
145    
146     6. ESC_K was left out of the EBCDIC table.
147    
148     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
149        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
150        limit also applies to "virtual nesting" when a pattern is recursive, and in
151        this case 1000 isn't so big. I have been able to remove this limit at the
152        expense of backing off one optimization in certain circumstances. Normally,
153        when pcre_exec() would call its internal match() function recursively and
154        immediately return the result unconditionally, it uses a "tail recursion"
155        feature to save stack. However, when a subpattern that can match an empty
156        string has an unlimited repetition quantifier, it no longer makes this
157        optimization. That gives it a stack frame in which to save the data for
158        checking that an empty string has been matched. Previously this was taken
159        from the 1000-entry workspace that had been reserved. So now there is no
160        explicit limit, but more stack is used.
161    
162     8. Applied Daniel's patches to solve problems with the import/export magic
163        syntax that is required for Windows, and which was going wrong for the
164        pcreposix and pcrecpp parts of the library. These were overlooked when this
165        problem was solved for the main library.
166    
167     9. There were some crude static tests to avoid integer overflow when computing
168        the size of patterns that contain repeated groups with explicit upper
169        limits. As the maximum quantifier is 65535, the maximum group length was
170        set at 30,000 so that the product of these two numbers did not overflow a
171        32-bit integer. However, it turns out that people want to use groups that
172        are longer than 30,000 bytes (though not repeat them that many times).
173        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
174        made it possible to implement the integer overflow checks in a much more
175        dynamic way, which I have now done. The artificial limitation on group
176        length has been removed - we now have only the limit on the total length of
177        the compiled pattern, which depends on the LINK_SIZE setting.
178    
179    10. Fixed a bug in the documentation for get/copy named substring when
180        duplicate names are permitted. If none of the named substrings are set, the
181        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
182        empty string.
183    
184    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
185        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
186        because the ] is interpreted as the first data character and the
187        terminating ] is not found. PCRE has been made compatible with Perl in this
188        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
189        cause memory overwriting.
190    
191    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
192        string has been matched (to stop an infinite loop). It was not recognizing
193        a conditional subpattern that could match an empty string if that
194        subpattern was within another subpattern. For example, it looped when
195        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
196        condition was not nested. This bug has been fixed.
197    
198    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
199        past the start of the subject in the presence of bytes with the top bit
200        set, for example "\x8aBCD".
201    
202    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
203        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
204    
205    14. Optimized (?!) to (*FAIL).
206    
207    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
208        This restricts code points to be within the range 0 to 0x10FFFF, excluding
209        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
210        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
211        does: it's just the validity check that is more restrictive.
212    
213    16. Inserted checks for integer overflows during escape sequence (backslash)
214        processing, and also fixed erroneous offset values for syntax errors during
215        backslash processing.
216    
217    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
218        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
219    
220    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
221        caused an overrun.
222    
223    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
224        something other than just ASCII characters) inside a group that had an
225        unlimited repeat caused a loop at compile time (while checking to see
226        whether the group could match an empty string).
227    
228    20. Debugging a pattern containing \p or \P could cause a crash. For example,
229        [\P{Any}] did so. (Error in the code for printing property names.)
230    
231    21. An orphan \E inside a character class could cause a crash.
232    
233    22. A repeated capturing bracket such as (A)? could cause a wild memory
234        reference during compilation.
235    
236    23. There are several functions in pcre_compile() that scan along a compiled
237        expression for various reasons (e.g. to see if it's fixed length for look
238        behind). There were bugs in these functions when a repeated \p or \P was
239        present in the pattern. These operators have additional parameters compared
240        with \d, etc, and these were not being taken into account when moving along
241        the compiled data. Specifically:
242    
243        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
244            length.
245    
246        (b) An item such as \pL+ within a repeated group could cause crashes or
247            loops.
248    
249        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
250            "reference to non-existent subpattern" error.
251    
252        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
253    
254    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
255        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
256    
257    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
258    
259    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
260        character were causing crashes (broken optimization).
261    
262    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
263        \p or \P) caused a compile-time loop.
264    
265    28. More problems have arisen in unanchored patterns when CRLF is a valid line
266        break. For example, the unstudied pattern [\r\n]A does not match the string
267        "\r\nA" because change 7.0/46 below moves the current point on by two
268        characters after failing to match at the start. However, the pattern \nA
269        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
270        the same is true. There doesn't seem any very clean way out of this, but
271        what I have chosen to do makes the common cases work: PCRE now takes note
272        of whether there can be an explicit match for \r or \n anywhere in the
273        pattern, and if so, 7.0/46 no longer applies. As part of this change,
274        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
275        pattern has explicit CR or LF references.
276    
277    29. Added (*CR) etc for changing newline setting at start of pattern.
278    
279    
280    Version 7.2 19-Jun-07
281    ---------------------
282    
283     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
284        which is apparently normally available under Windows.
285    
286     2. Re-jig the pcregrep tests with different newline settings in an attempt
287        to make them independent of the local environment's newline setting.
288    
289     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
290    
291     4. Some of the "internals" tests were previously cut out when the link size
292        was not 2, because the output contained actual offsets. The recent new
293        "Z" feature of pcretest means that these can be cut out, making the tests
294        usable with all link sizes.
295    
296     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
297        stack recursion. This gives a massive performance boost under BSD, but just
298        a small improvement under Linux. However, it saves one field in the frame
299        in all cases.
300    
301     6. Added more features from the forthcoming Perl 5.10:
302    
303        (a) (?-n) (where n is a string of digits) is a relative subroutine or
304            recursion call. It refers to the nth most recently opened parentheses.
305    
306        (b) (?+n) is also a relative subroutine call; it refers to the nth next
307            to be opened parentheses.
308    
309        (c) Conditions that refer to capturing parentheses can be specified
310            relatively, for example, (?(-2)... or (?(+3)...
311    
312        (d) \K resets the start of the current match so that everything before
313            is not part of it.
314    
315        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
316    
317        (f) \g{name} is another synonym - part of Perl 5.10's unification of
318            reference syntax.
319    
320        (g) (?| introduces a group in which the numbering of parentheses in each
321            alternative starts with the same number.
322    
323        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
324    
325     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
326        PCRE_INFO_JCHANGED.
327    
328     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
329        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
330        for detecting groups that can match an empty string.
331    
332     9. A pattern with a very large number of alternatives (more than several
333        hundred) was running out of internal workspace during the pre-compile
334        phase, where pcre_compile() figures out how much memory will be needed. A
335        bit of new cunning has reduced the workspace needed for groups with
336        alternatives. The 1000-alternative test pattern now uses 12 bytes of
337        workspace instead of running out of the 4096 that are available.
338    
339    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
340    
341    11. Applied patch from Google to remove an optimization that didn't quite work.
342        The report of the bug said:
343    
344          pcrecpp::RE("a*").FullMatch("aaa") matches, while
345          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
346          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
347    
348    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
349        it matched the wrong number of bytes.
350    
351    
352    Version 7.1 24-Apr-07
353    ---------------------
354    
355     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
356        that is more "standard", making use of automake and other Autotools. There
357        is some re-arrangement of the files and adjustment of comments consequent
358        on this.
359    
360     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
361        for recursive directory scanning broke on some systems because the files
362        are not scanned in any specific order and on different systems the order
363        was different. A call to "sort" has been inserted into RunGrepTest for the
364        approprate test as a short-term fix. In the longer term there may be an
365        alternative.
366    
367     3. I had an email from Eric Raymond about problems translating some of PCRE's
368        man pages to HTML (despite the fact that I distribute HTML pages, some
369        people do their own conversions for various reasons). The problems
370        concerned the use of low-level troff macros .br and .in. I have therefore
371        removed all such uses from the man pages (some were redundant, some could
372        be replaced by .nf/.fi pairs). The 132html script that I use to generate
373        HTML has been updated to handle .nf/.fi and to complain if it encounters
374        .br or .in.
375    
376     4. Updated comments in configure.ac that get placed in config.h.in and also
377        arranged for config.h to be included in the distribution, with the name
378        config.h.generic, for the benefit of those who have to compile without
379        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
380    
381     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
382        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
383        makevp.bat for the current PCRE, using the additional files
384        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
385    
386     6. A Windows user reported a minor discrepancy with test 2, which turned out
387        to be caused by a trailing space on an input line that had got lost in his
388        copy. The trailing space was an accident, so I've just removed it.
389    
390     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
391        that is needed.
392    
393     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
394        as "const" (a) because they are and (b) because it helps the PHP
395        maintainers who have recently made a script to detect big data structures
396        in the php code that should be moved to the .rodata section. I remembered
397        to update Builducptable as well, so it won't revert if ucptable.h is ever
398        re-created.
399    
400     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
401        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
402        order to be able to cut out the UTF-8 tables in the latter when UTF-8
403        support is not required. This saves 1.5-2K of code, which is important in
404        some applications.
405    
406        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
407        so as not to refer to the tables, even though these functions will never be
408        called when UTF-8 support is disabled. Otherwise there are problems with a
409        shared library.
410    
411    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
412    
413        (a) It was defining its arguments as char * instead of void *.
414    
415        (b) It was assuming that all moves were upwards in memory; this was true
416            a long time ago when I wrote it, but is no longer the case.
417    
418        The emulated memove() is provided for those environments that have neither
419        memmove() nor bcopy(). I didn't think anyone used it these days, but that
420        is clearly not the case, as these two bugs were recently reported.
421    
422    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
423        and Detrail to create the HTML documentation, the .txt form of the man
424        pages, and it removes trailing spaces from listed files. It also creates
425        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
426        case, it wraps all the #defines with #ifndefs. This script should be run
427        before "make dist".
428    
429    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
430        with Unicode property support.
431    
432        (a) For a maximizing quantifier, if the two different cases of the
433            character were of different lengths in their UTF-8 codings (there are
434            some cases like this - I found 11), and the matching function had to
435            back up over a mixture of the two cases, it incorrectly assumed they
436            were both the same length.
437    
438        (b) When PCRE was configured to use the heap rather than the stack for
439            recursion during matching, it was not correctly preserving the data for
440            the other case of a UTF-8 character when checking ahead for a match
441            while processing a minimizing repeat. If the check also involved
442            matching a wide character, but failed, corruption could cause an
443            erroneous result when trying to check for a repeat of the original
444            character.
445    
446    13. Some tidying changes to the testing mechanism:
447    
448        (a) The RunTest script now detects the internal link size and whether there
449            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
450            values substituted by "configure". (The RunGrepTest script already did
451            this for UTF-8.) The configure.ac script no longer substitutes the
452            relevant variables.
453    
454        (b) The debugging options /B and /D in pcretest show the compiled bytecode
455            with length and offset values. This means that the output is different
456            for different internal link sizes. Test 2 is skipped for link sizes
457            other than 2 because of this, bypassing the problem. Unfortunately,
458            there was also a test in test 3 (the locale tests) that used /B and
459            failed for link sizes other than 2. Rather than cut the whole test out,
460            I have added a new /Z option to pcretest that replaces the length and
461            offset values with spaces. This is now used to make test 3 independent
462            of link size. (Test 2 will be tidied up later.)
463    
464    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
465        segmentation fault instead of returning the appropriate error message.
466    
467    15. In multiline mode when the newline sequence was set to "any", the pattern
468        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
469        This doesn't seem right; it now treats the CRLF combination as the line
470        ending, and so does not match in that case. It's only a pattern such as ^$
471        that would hit this one: something like ^ABC$ would have failed after \r
472        and then tried again after \r\n.
473    
474    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
475        in an attempt to make files that differ only in their line terminators
476        compare equal. This works on Linux.
477    
478    17. Under certain error circumstances pcregrep might try to free random memory
479        as it exited. This is now fixed, thanks to valgrind.
480    
481    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
482        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
483        was because its rules for how to advance for /g after matching an empty
484        string at the end of a line did not allow for this case. They now check for
485        it specially.
486    
487    20. pcretest is supposed to handle patterns and data of any length, by
488        extending its buffers when necessary. It was getting this wrong when the
489        buffer for a data line had to be extended.
490    
491    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
492        CRLF as a newline sequence.
493    
494    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
495        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
496        I have nevertheless tidied it up.
497    
498    23. Added some casts to kill warnings from HP-UX ia64 compiler.
499    
500    24. Added a man page for pcre-config.
501    
502    
503    Version 7.0 19-Dec-06
504    ---------------------
505    
506     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
507        moving to gcc 4.1.1.
508    
509     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
510        sys/time.h, which is documented as needed for this function. It doesn't
511        seem to matter on Linux, but it showed up on some releases of OS X.
512    
513     3. It seems that there are systems where bytes whose values are greater than
514        127 match isprint() in the "C" locale. The "C" locale should be the
515        default when a C program starts up. In most systems, only ASCII printing
516        characters match isprint(). This difference caused the output from pcretest
517        to vary, making some of the tests fail. I have changed pcretest so that:
518    
519        (a) When it is outputting text in the compiled version of a pattern, bytes
520            other than 32-126 are always shown as hex escapes.
521    
522        (b) When it is outputting text that is a matched part of a subject string,
523            it does the same, unless a different locale has been set for the match
524            (using the /L modifier). In this case, it uses isprint() to decide.
525    
526     4. Fixed a major bug that caused incorrect computation of the amount of memory
527        required for a compiled pattern when options that changed within the
528        pattern affected the logic of the preliminary scan that determines the
529        length. The relevant options are -x, and -i in UTF-8 mode. The result was
530        that the computed length was too small. The symptoms of this bug were
531        either the PCRE error "internal error: code overflow" from pcre_compile(),
532        or a glibc crash with a message such as "pcretest: free(): invalid next
533        size (fast)". Examples of patterns that provoked this bug (shown in
534        pcretest format) are:
535    
536          /(?-x: )/x
537          /(?x)(?-x: \s*#\s*)/
538          /((?i)[\x{c0}])/8
539          /(?i:[\x{c0}])/8
540    
541        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
542        is now done differently.
543    
544     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
545        wrapper classes; (b) implement a new function in the C++ scanner that is
546        more efficient than the old way of doing things because it avoids levels of
547        recursion in the regex matching; (c) add a paragraph to the documentation
548        for the FullMatch() function.
549    
550     6. The escape sequence \n was being treated as whatever was defined as
551        "newline". Not only was this contrary to the documentation, which states
552        that \n is character 10 (hex 0A), but it also went horribly wrong when
553        "newline" was defined as CRLF. This has been fixed.
554    
555     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
556        was being set to -1 for the "end of line" case (supposedly a value that no
557        character can have). Though this value is never used (the check for end of
558        line is "zero bytes in current character"), it caused compiler complaints.
559        I've changed it to 0xffffffff.
560    
561     8. In pcre_version.c, the version string was being built by a sequence of
562        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
563        string (as it is for production releases) called a macro with an empty
564        argument. The C standard says the result of this is undefined. The gcc
565        compiler treats it as an empty string (which was what was wanted) but it is
566        reported that Visual C gives an error. The source has been hacked around to
567        avoid this problem.
568    
569     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
570        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
571        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
572        of them did).
573    
574    10. Originally, pcretest opened its input and output without "b"; then I was
575        told that "b" was needed in some environments, so it was added for release
576        5.0 to both the input and output. (It makes no difference on Unix-like
577        systems.) Later I was told that it is wrong for the input on Windows. I've
578        now abstracted the modes into two macros, to make it easier to fiddle with
579        them, and removed "b" from the input mode under Windows.
580    
581    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
582    
583    12. Added -help and --help to pcretest as an official way of being reminded
584        of the options.
585    
586    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
587        and pcrecpp.cc because they annoy compilers at high warning levels.
588    
589    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
590    
591    15. Fixed an occurrence of == in configure.ac that should have been = (shell
592        scripts are not C programs :-) and which was not noticed because it works
593        on Linux.
594    
595    16. pcretest is supposed to handle any length of pattern and data line (as one
596        line or as a continued sequence of lines) by extending its input buffer if
597        necessary. This feature was broken for very long pattern lines, leading to
598        a string of junk being passed to pcre_compile() if the pattern was longer
599        than about 50K.
600    
601    17. I have done a major re-factoring of the way pcre_compile() computes the
602        amount of memory needed for a compiled pattern. Previously, there was code
603        that made a preliminary scan of the pattern in order to do this. That was
604        OK when PCRE was new, but as the facilities have expanded, it has become
605        harder and harder to keep it in step with the real compile phase, and there
606        have been a number of bugs (see for example, 4 above). I have now found a
607        cunning way of running the real compile function in a "fake" mode that
608        enables it to compute how much memory it would need, while actually only
609        ever using a few hundred bytes of working memory and without too many
610        tests of the mode. This should make future maintenance and development
611        easier. A side effect of this work is that the limit of 200 on the nesting
612        depth of parentheses has been removed (though this was never a serious
613        limitation, I suspect). However, there is a downside: pcre_compile() now
614        runs more slowly than before (30% or more, depending on the pattern). I
615        hope this isn't a big issue. There is no effect on runtime performance.
616    
617    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
618        newline (only possible for the last line of a file) and it was a
619        pattern that set a locale (followed by /Lsomething), pcretest crashed.
620    
621    19. Added additional timing features to pcretest. (1) The -tm option now times
622        matching only, not compiling. (2) Both -t and -tm can be followed, as a
623        separate command line item, by a number that specifies the number of
624        repeats to use when timing. The default is 50000; this gives better
625        precision, but takes uncomfortably long for very large patterns.
626    
627    20. Extended pcre_study() to be more clever in cases where a branch of a
628        subpattern has no definite first character. For example, (a*|b*)[cd] would
629        previously give no result from pcre_study(). Now it recognizes that the
630        first character must be a, b, c, or d.
631    
632    21. There was an incorrect error "recursive call could loop indefinitely" if
633        a subpattern (or the entire pattern) that was being tested for matching an
634        empty string contained only one non-empty item after a nested subpattern.
635        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
636        incorrectly, because the \d was being skipped in the check.
637    
638    22. The pcretest program now has a new pattern option /B and a command line
639        option -b, which is equivalent to adding /B to every pattern. This causes
640        it to show the compiled bytecode, without the additional information that
641        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
642        is the same as /B/I).
643    
644    23. A new optimization is now able automatically to treat some sequences such
645        as a*b as a*+b. More specifically, if something simple (such as a character
646        or a simple class like \d) has an unlimited quantifier, and is followed by
647        something that cannot possibly match the quantified thing, the quantifier
648        is automatically "possessified".
649    
650    24. A recursive reference to a subpattern whose number was greater than 39
651        went wrong under certain circumstances in UTF-8 mode. This bug could also
652        have affected the operation of pcre_study().
653    
654    25. Realized that a little bit of performance could be had by replacing
655        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
656    
657    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
658    
659    27. Possessive quantifiers such as a++ were previously implemented by turning
660        them into atomic groups such as ($>a+). Now they have their own opcodes,
661        which improves performance. This includes the automatically created ones
662        from 23 above.
663    
664    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
665        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
666        the first matched character to be a colon. This applied both to named and
667        numbered groups.
668    
669    29. The ucpinternal.h header file was missing its idempotency #ifdef.
670    
671    30. I was sent a "project" file called libpcre.a.dev which I understand makes
672        building PCRE on Windows easier, so I have included it in the distribution.
673    
674    31. There is now a check in pcretest against a ridiculously large number being
675        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
676        loop, the loop is abandoned.
677    
678    32. Forward references to subpatterns in conditions such as (?(2)...) where
679        subpattern 2 is defined later cause pcre_compile() to search forwards in
680        the pattern for the relevant set of parentheses. This search went wrong
681        when there were unescaped parentheses in a character class, parentheses
682        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
683    
684    33. "Subroutine" calls and backreferences were previously restricted to
685        referencing subpatterns earlier in the regex. This restriction has now
686        been removed.
687    
688    34. Added a number of extra features that are going to be in Perl 5.10. On the
689        whole, these are just syntactic alternatives for features that PCRE had
690        previously implemented using the Python syntax or my own invention. The
691        other formats are all retained for compatibility.
692    
693        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
694            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
695            also .NET compatible.
696    
697        (b) A recursion or subroutine call to a named group can now be defined as
698            (?&name) as well as (?P>name).
699    
700        (c) A backreference to a named group can now be defined as \k<name> or
701            \k'name' as well as (?P=name). The new forms, as well as being in Perl
702            5.10, are also .NET compatible.
703    
704        (d) A conditional reference to a named group can now use the syntax
705            (?(<name>) or (?('name') as well as (?(name).
706    
707        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
708            groups (named and numbered) that are never evaluated inline, but can be
709            called as "subroutines" from elsewhere. In effect, the DEFINE condition
710            is always false. There may be only one alternative in such a group.
711    
712        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
713            as the simple (?(R). The condition is true only if the most recent
714            recursion is that of the given number or name. It does not search out
715            through the entire recursion stack.
716    
717        (g) The escape \gN or \g{N} has been added, where N is a positive or
718            negative number, specifying an absolute or relative reference.
719    
720    35. Tidied to get rid of some further signed/unsigned compiler warnings and
721        some "unreachable code" warnings.
722    
723    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
724        things, this adds five new scripts.
725    
726    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
727        There were also incompatibilities regarding the handling of \Q..\E inside
728        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
729        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
730    
731    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
732        matches an empty string, and forcibly breaks the loop. There were bugs in
733        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
734        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
735        separate and independent bugs (that affected different cases) have been
736        fixed.
737    
738    39. Refactored the code to abolish the use of different opcodes for small
739        capturing bracket numbers. This is a tidy that I avoided doing when I
740        removed the limit on the number of capturing brackets for 3.5 back in 2001.
741        The new approach is not only tidier, it makes it possible to reduce the
742        memory needed to fix the previous bug (38).
743    
744    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
745        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
746        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
747        mode.
748    
749    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
750        report.
751    
752    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
753        copy construction and assignment in the C++ wrapper.
754    
755    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
756        couple of bugs that could have given wrong results in the "--newline=crlf"
757        case.
758    
759    44. Added a number of casts and did some reorganization of signed/unsigned int
760        variables following suggestions from Dair Grant. Also renamed the variable
761        "this" as "item" because it is a C++ keyword.
762    
763    45. Arranged for dftables to add
764    
765          #include "pcre_internal.h"
766    
767        to pcre_chartables.c because without it, gcc 4.x may remove the array
768        definition from the final binary if PCRE is built into a static library and
769        dead code stripping is activated.
770    
771    46. For an unanchored pattern, if a match attempt fails at the start of a
772        newline sequence, and the newline setting is CRLF or ANY, and the next two
773        characters are CRLF, advance by two characters instead of one.
774    
775    
776    Version 6.7 04-Jul-06
777    ---------------------
778    
779     1. In order to handle tests when input lines are enormously long, pcretest has
780        been re-factored so that it automatically extends its buffers when
781        necessary. The code is crude, but this _is_ just a test program. The
782        default size has been increased from 32K to 50K.
783    
784     2. The code in pcre_study() was using the value of the re argument before
785        testing it for NULL. (Of course, in any sensible call of the function, it
786        won't be NULL.)
787    
788     3. The memmove() emulation function in pcre_internal.h, which is used on
789        systems that lack both memmove() and bcopy() - that is, hardly ever -
790        was missing a "static" storage class specifier.
791    
792     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
793        containing an extended class (one that cannot be represented by a bitmap
794        because it contains high-valued characters or Unicode property items, e.g.
795        [\pZ]). Almost always one would set UTF-8 mode when processing such a
796        pattern, but PCRE should not loop if you do not (it no longer does).
797        [Detail: two cases were found: (a) a repeated subpattern containing an
798        extended class; (b) a recursive reference to a subpattern that followed a
799        previous extended class. It wasn't skipping over the extended class
800        correctly when UTF-8 mode was not set.]
801    
802     5. A negated single-character class was not being recognized as fixed-length
803        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
804        compile error "lookbehind assertion is not fixed length".
805    
806     6. The RunPerlTest auxiliary script was showing an unexpected difference
807        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
808        write a Perl script that can interpret lines of an input file either as
809        byte characters or as UTF-8, which is what "perltest" was being required to
810        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
811        can't do is switch easily at run time between having the "use utf8;" pragma
812        or not. In the end, I fudged it by using the RunPerlTest script to insert
813        "use utf8;" explicitly for the UTF-8 tests.
814    
815     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
816        the end of the subject string, contrary to the documentation and to what
817        Perl does. This was true of both matching functions. Now it matches only at
818        the start of the subject and immediately after *internal* newlines.
819    
820     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
821        a pointer to an int instead of a pointer to an unsigned long int. This
822        caused problems on 64-bit systems.
823    
824     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
825        instance of the 'standard' template library not being so standard".
826    
827    10. There was no check on the number of named subpatterns nor the maximum
828        length of a subpattern name. The product of these values is used to compute
829        the size of the memory block for a compiled pattern. By supplying a very
830        long subpattern name and a large number of named subpatterns, the size
831        computation could be caused to overflow. This is now prevented by limiting
832        the length of names to 32 characters, and the number of named subpatterns
833        to 10,000.
834    
835    11. Subpatterns that are repeated with specific counts have to be replicated in
836        the compiled pattern. The size of memory for this was computed from the
837        length of the subpattern and the repeat count. The latter is limited to
838        65535, but there was no limit on the former, meaning that integer overflow
839        could in principle occur. The compiled length of a repeated subpattern is
840        now limited to 30,000 bytes in order to prevent this.
841    
842    12. Added the optional facility to have named substrings with the same name.
843    
844    13. Added the ability to use a named substring as a condition, using the
845        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
846        are numbers (not recommended). Forward references are permitted.
847    
848    14. Added forward references in named backreferences (if you see what I mean).
849    
850    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
851        pattern could run off the end of the subject. For example, the pattern
852        "(?s)(.{1,5})"8 did this with the subject "ab".
853    
854    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
855        PCRE_CASELESS was set when matching characters that were quantified with ?
856        or *.
857    
858    17. A character class other than a single negated character that had a minimum
859        but no maximum quantifier - for example [ab]{6,} - was not handled
860        correctly by pce_dfa_exec(). It would match only one character.
861    
862    18. A valid (though odd) pattern that looked like a POSIX character
863        class but used an invalid character after [ (for example [[,abc,]]) caused
864        pcre_compile() to give the error "Failed: internal error: code overflow" or
865        in some cases to crash with a glibc free() error. This could even happen if
866        the pattern terminated after [[ but there just happened to be a sequence of
867        letters, a binary zero, and a closing ] in the memory that followed.
868    
869    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
870        over the years. Originally (before any Unicode support), just the bottom 8
871        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
872        output from "man perlunicode" includes this:
873    
874          The regular expression compiler produces polymorphic opcodes.  That
875          is, the pattern adapts to the data and automatically switches to
876          the Unicode character scheme when presented with Unicode data--or
877          instead uses a traditional byte scheme when presented with byte
878          data.
879    
880        Sadly, a wide octal escape does not cause a switch, and in a string with
881        no other multibyte characters, these octal escapes are treated as before.
882        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
883        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
884        Unicode string.
885    
886        I have not perpetrated such confusion in PCRE. Up till now, it took just
887        the bottom 8 bits, as in old Perl. I have now made octal escapes with
888        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
889        translate to the appropriate multibyte character.
890    
891    29. Applied some refactoring to reduce the number of warnings from Microsoft
892        and Borland compilers. This has included removing the fudge introduced
893        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
894        a warning about an unused variable.
895    
896    21. PCRE has not included VT (character 0x0b) in the set of whitespace
897        characters since release 4.0, because Perl (from release 5.004) does not.
898        [Or at least, is documented not to: some releases seem to be in conflict
899        with the documentation.] However, when a pattern was studied with
900        pcre_study() and all its branches started with \s, PCRE still included VT
901        as a possible starting character. Of course, this did no harm; it just
902        caused an unnecessary match attempt.
903    
904    22. Removed a now-redundant internal flag bit that recorded the fact that case
905        dependency changed within the pattern. This was once needed for "required
906        byte" processing, but is no longer used. This recovers a now-scarce options
907        bit. Also moved the least significant internal flag bit to the most-
908        significant bit of the word, which was not previously used (hangover from
909        the days when it was an int rather than a uint) to free up another bit for
910        the future.
911    
912    23. Added support for CRLF line endings as well as CR and LF. As well as the
913        default being selectable at build time, it can now be changed at runtime
914        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
915        specify that it is scanning data with non-default line endings.
916    
917    24. Changed the definition of CXXLINK to make it agree with the definition of
918        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
919    
920    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
921        recursions. This makes a big different to stack usage for some patterns.
922    
923    26. If a subpattern containing a named recursion or subroutine reference such
924        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
925        the space required for the compiled pattern went wrong and gave too small a
926        value. Depending on the environment, this could lead to "Failed: internal
927        error: code overflow at offset 49" or "glibc detected double free or
928        corruption" errors.
929    
930    27. Applied patches from Google (a) to support the new newline modes and (b) to
931        advance over multibyte UTF-8 characters in GlobalReplace.
932    
933    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
934        difference for some implementation of PCRE in some Windows version.
935    
936    29. Added some extra testing facilities to pcretest:
937    
938        \q<number>   in a data line sets the "match limit" value
939        \Q<number>   in a data line sets the "match recursion limt" value
940        -S <number>  sets the stack size, where <number> is in megabytes
941    
942        The -S option isn't available for Windows.
943    
944    
945    Version 6.6 06-Feb-06
946    ---------------------
947    
948     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
949        in pcreposix.h. I have copied the definition from pcre.h.
950    
951     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
952        because pcre.h is no longer a built file.
953    
954     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
955        not normally included in the compiled code.
956    
957    
958    Version 6.5 01-Feb-06
959    ---------------------
960    
961     1. When using the partial match feature with pcre_dfa_exec(), it was not
962        anchoring the second and subsequent partial matches at the new starting
963        point. This could lead to incorrect results. For example, with the pattern
964        /1234/, partially matching against "123" and then "a4" gave a match.
965    
966     2. Changes to pcregrep:
967    
968        (a) All non-match returns from pcre_exec() were being treated as failures
969            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
970            error message is output. Some extra information is given for the
971            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
972            probably the only errors that are likely to be caused by users (by
973            specifying a regex that has nested indefinite repeats, for instance).
974            If there are more than 20 of these errors, pcregrep is abandoned.
975    
976        (b) A binary zero was treated as data while matching, but terminated the
977            output line if it was written out. This has been fixed: binary zeroes
978            are now no different to any other data bytes.
979    
980        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
981            used to set a locale for matching. The --locale=xxxx long option has
982            been added (no short equivalent) to specify a locale explicitly on the
983            pcregrep command, overriding the environment variables.
984    
985        (d) When -B was used with -n, some line numbers in the output were one less
986            than they should have been.
987    
988        (e) Added the -o (--only-matching) option.
989    
990        (f) If -A or -C was used with -c (count only), some lines of context were
991            accidentally printed for the final match.
992    
993        (g) Added the -H (--with-filename) option.
994    
995        (h) The combination of options -rh failed to suppress file names for files
996            that were found from directory arguments.
997    
998        (i) Added the -D (--devices) and -d (--directories) options.
999    
1000        (j) Added the -F (--fixed-strings) option.
1001    
1002        (k) Allow "-" to be used as a file name for -f as well as for a data file.
1003    
1004        (l) Added the --colo(u)r option.
1005    
1006        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1007            is not present by default.
1008    
1009     3. A nasty bug was discovered in the handling of recursive patterns, that is,
1010        items such as (?R) or (?1), when the recursion could match a number of
1011        alternatives. If it matched one of the alternatives, but subsequently,
1012        outside the recursion, there was a failure, the code tried to back up into
1013        the recursion. However, because of the way PCRE is implemented, this is not
1014        possible, and the result was an incorrect result from the match.
1015    
1016        In order to prevent this happening, the specification of recursion has
1017        been changed so that all such subpatterns are automatically treated as
1018        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1019    
1020     4. I had overlooked the fact that, in some locales, there are characters for
1021        which isalpha() is true but neither isupper() nor islower() are true. In
1022        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1023        and ordfeminine) are like this. This affected the treatment of \w and \W
1024        when they appeared in character classes, but not when they appeared outside
1025        a character class. The bit map for "word" characters is now created
1026        separately from the results of isalnum() instead of just taking it from the
1027        upper, lower, and digit maps. (Plus the underscore character, of course.)
1028    
1029     5. The above bug also affected the handling of POSIX character classes such as
1030        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1031        permanent tables. Instead, the bit maps for such a class were previously
1032        created as the appropriate unions of the upper, lower, and digit bitmaps.
1033        Now they are created by subtraction from the [[:word:]] class, which has
1034        its own bitmap.
1035    
1036     6. The [[:blank:]] character class matches horizontal, but not vertical space.
1037        It is created by subtracting the vertical space characters (\x09, \x0a,
1038        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1039        subtraction was done in the overall bitmap for a character class, meaning
1040        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1041        be recognized. This bug has been fixed.
1042    
1043     7. Patches from the folks at Google:
1044    
1045          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1046          real life, but is still worth protecting against".
1047    
1048          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1049          regular expressions".
1050    
1051          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1052          have it.
1053    
1054          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1055          "configure" and the latter not, in order to fix a problem somebody had
1056          with compiling the Arg class on HP-UX.
1057    
1058          (e) Improve the error-handling of the C++ wrapper a little bit.
1059    
1060          (f) New tests for checking recursion limiting.
1061    
1062     8. The pcre_memmove() function, which is used only if the environment does not
1063        have a standard memmove() function (and is therefore rarely compiled),
1064        contained two bugs: (a) use of int instead of size_t, and (b) it was not
1065        returning a result (though PCRE never actually uses the result).
1066    
1067     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1068        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1069        returned instead of calling malloc() with an overflowing number that would
1070        most likely cause subsequent chaos.
1071    
1072    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1073    
1074    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1075        with this option is matched, the nmatch and pmatch options of regexec() are
1076        ignored.
1077    
1078    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1079        provided in case anyone wants to the the POSIX interface with UTF-8
1080        strings.
1081    
1082    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1083        C++ linking (needed for some HP-UX environments).
1084    
1085    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1086        (unused parameter) and in the pcre_printint() function (omitted "default"
1087        switch label when the default is to do nothing).
1088    
1089    15. Added some code to make it possible, when PCRE is compiled as a C++
1090        library, to replace subject pointers for pcre_exec() with a smart pointer
1091        class, thus making it possible to process discontinuous strings.
1092    
1093    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1094        much the same function. They were added by different people who were trying
1095        to make PCRE easy to compile on non-Unix systems. It has been suggested
1096        that PCRE_EXPORT be abolished now that there is more automatic apparatus
1097        for compiling on Windows systems. I have therefore replaced it with
1098        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1099        defaults to "extern" for C or "extern C" for C++, which works fine on
1100        Unix-like systems. It is now possible to override the value of PCRE_DATA_
1101        SCOPE with something explicit in config.h. In addition:
1102    
1103        (a) pcreposix.h still had just "extern" instead of either of these macros;
1104            I have replaced it with PCRE_DATA_SCOPE.
1105    
1106        (b) Functions such as _pcre_xclass(), which are internal to the library,
1107            but external in the C sense, all had PCRE_EXPORT in their definitions.
1108            This is apparently wrong for the Windows case, so I have removed it.
1109            (It makes no difference on Unix-like systems.)
1110    
1111    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1112        of recursive calls to match(). This is different to MATCH_LIMIT because
1113        that limits the total number of calls to match(), not all of which increase
1114        the depth of recursion. Limiting the recursion depth limits the amount of
1115        stack (or heap if NO_RECURSE is set) that is used. The default can be set
1116        when PCRE is compiled, and changed at run time. A patch from Google adds
1117        this functionality to the C++ interface.
1118    
1119    18. Changes to the handling of Unicode character properties:
1120    
1121        (a) Updated the table to Unicode 4.1.0.
1122    
1123        (b) Recognize characters that are not in the table as "Cn" (undefined).
1124    
1125        (c) I revised the way the table is implemented to a much improved format
1126            which includes recognition of ranges. It now supports the ranges that
1127            are defined in UnicodeData.txt, and it also amalgamates other
1128            characters into ranges. This has reduced the number of entries in the
1129            table from around 16,000 to around 3,000, thus reducing its size
1130            considerably. I realized I did not need to use a tree structure after
1131            all - a binary chop search is just as efficient. Having reduced the
1132            number of entries, I extended their size from 6 bytes to 8 bytes to
1133            allow for more data.
1134    
1135        (d) Added support for Unicode script names via properties such as \p{Han}.
1136    
1137    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1138        matching that character.
1139    
1140    20. When matching a repeated Unicode property with a minimum greater than zero,
1141        (for example \pL{2,}), PCRE could look past the end of the subject if it
1142        reached it while seeking the minimum number of characters. This could
1143        happen only if some of the characters were more than one byte long, because
1144        there is a check for at least the minimum number of bytes.
1145    
1146    21. Refactored the implementation of \p and \P so as to be more general, to
1147        allow for more different types of property in future. This has changed the
1148        compiled form incompatibly. Anybody with saved compiled patterns that use
1149        \p or \P will have to recompile them.
1150    
1151    22. Added "Any" and "L&" to the supported property types.
1152    
1153    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1154        but give a compile time error if the value is greater than 0xff.
1155    
1156    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1157        accidentally not being installed or uninstalled.
1158    
1159    25. The pcre.h file was built from pcre.h.in, but the only changes that were
1160        made were to insert the current release number. This seemed silly, because
1161        it made things harder for people building PCRE on systems that don't run
1162        "configure". I have turned pcre.h into a distributed file, no longer built
1163        by "configure", with the version identification directly included. There is
1164        no longer a pcre.h.in file.
1165    
1166        However, this change necessitated a change to the pcre-config script as
1167        well. It is built from pcre-config.in, and one of the substitutions was the
1168        release number. I have updated configure.ac so that ./configure now finds
1169        the release number by grepping pcre.h.
1170    
1171    26. Added the ability to run the tests under valgrind.
1172    
1173    
1174    Version 6.4 05-Sep-05
1175    ---------------------
1176    
1177     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1178        "--" to be printed when multiple files were scanned, even when none of the
1179        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1180        consider it to be a bug, and have restored the previous behaviour.
1181    
1182     2. A couple of code tidies to get rid of compiler warnings.
1183    
1184     3. The pcretest program used to cheat by referring to symbols in the library
1185        whose names begin with _pcre_. These are internal symbols that are not
1186        really supposed to be visible externally, and in some environments it is
1187        possible to suppress them. The cheating is now confined to including
1188        certain files from the library's source, which is a bit cleaner.
1189    
1190     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1191        file's purpose clearer.
1192    
1193     5. Reorganized pcre_ucp_findchar().
1194    
1195    
1196    Version 6.3 15-Aug-05
1197    ---------------------
1198    
1199     1. The file libpcre.pc.in did not have general read permission in the tarball.
1200    
1201     2. There were some problems when building without C++ support:
1202    
1203        (a) If C++ support was not built, "make install" and "make test" still
1204            tried to test it.
1205    
1206        (b) There were problems when the value of CXX was explicitly set. Some
1207            changes have been made to try to fix these, and ...
1208    
1209        (c) --disable-cpp can now be used to explicitly disable C++ support.
1210    
1211        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1212            backslash in a target when C++ was disabled. This confuses some
1213            versions of "make", apparently. Using an intermediate variable solves
1214            this. (Same for CPP_LOBJ.)
1215    
1216     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1217        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1218        necessary on certain architectures.
1219    
1220     4. Added a setting of -export-symbols-regex to the link command to remove
1221        those symbols that are exported in the C sense, but actually are local
1222        within the library, and not documented. Their names all begin with
1223        "_pcre_". This is not a perfect job, because (a) we have to except some
1224        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1225        available (and never for static libraries). I have made a note to try to
1226        find a way round (a) in the future.
1227    
1228    
1229    Version 6.2 01-Aug-05
1230    ---------------------
1231    
1232     1. There was no test for integer overflow of quantifier values. A construction
1233        such as {1111111111111111} would give undefined results. What is worse, if
1234        a minimum quantifier for a parenthesized subpattern overflowed and became
1235        negative, the calculation of the memory size went wrong. This could have
1236        led to memory overwriting.
1237    
1238     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1239    
1240     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1241        operating environments where this matters.
1242    
1243     4. Applied Giuseppe Maxia's patch to add additional features for controlling
1244        PCRE options from within the C++ wrapper.
1245    
1246     5. Named capturing subpatterns were not being correctly counted when a pattern
1247        was compiled. This caused two problems: (a) If there were more than 100
1248        such subpatterns, the calculation of the memory needed for the whole
1249        compiled pattern went wrong, leading to an overflow error. (b) Numerical
1250        back references of the form \12, where the number was greater than 9, were
1251        not recognized as back references, even though there were sufficient
1252        previous subpatterns.
1253    
1254     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1255        versions of gcc, e.g. 2.95.4.
1256    
1257    
1258    Version 6.1 21-Jun-05
1259    ---------------------
1260    
1261     1. There was one reference to the variable "posix" in pcretest.c that was not
1262        surrounded by "#if !defined NOPOSIX".
1263    
1264     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1265        the cross-check on the old pcre_info() function, for the benefit of the
1266        cut-down version of PCRE that is currently imported into Exim.
1267    
1268     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1269        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1270        patterns that start (?i) but I don't think that matters. The use of (?i) is
1271        just an example; this all applies to the other options as well.
1272    
1273     4. Since libtool seems to echo the compile commands it is issuing, the output
1274        from "make" can be reduced a bit by putting "@" in front of each libtool
1275        compile command.
1276    
1277     5. Patch from the folks at Google for configure.in to be a bit more thorough
1278        in checking for a suitable C++ installation before trying to compile the
1279        C++ stuff. This should fix a reported problem when a compiler was present,
1280        but no suitable headers.
1281    
1282     6. The man pages all had just "PCRE" as their title. I have changed them to
1283        be the relevant file name. I have also arranged that these names are
1284        retained in the file doc/pcre.txt, which is a concatenation in text format
1285        of all the man pages except the little individual ones for each function.
1286    
1287     7. The NON-UNIX-USE file had not been updated for the different set of source
1288        files that come with release 6. I also added a few comments about the C++
1289        wrapper.
1290    
1291    
1292    Version 6.0 07-Jun-05
1293    ---------------------
1294    
1295     1. Some minor internal re-organization to help with my DFA experiments.
1296    
1297     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1298        didn't matter for the library itself when fully configured, but did matter
1299        when compiling without UCP support, or within Exim, where the ucp files are
1300        not imported.
1301    
1302     3. Refactoring of the library code to split up the various functions into
1303        different source modules. The addition of the new DFA matching code (see
1304        below) to a single monolithic source would have made it really too
1305        unwieldy, quite apart from causing all the code to be include in a
1306        statically linked application, when only some functions are used. This is
1307        relevant even without the DFA addition now that patterns can be compiled in
1308        one application and matched in another.
1309    
1310        The downside of splitting up is that there have to be some external
1311        functions and data tables that are used internally in different modules of
1312        the library but which are not part of the API. These have all had their
1313        names changed to start with "_pcre_" so that they are unlikely to clash
1314        with other external names.
1315    
1316     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1317        a different (DFA) algorithm. Although it is slower than the original
1318        function, it does have some advantages for certain types of matching
1319        problem.
1320    
1321     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1322        including restarting after a partial match.
1323    
1324     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1325        defined when compiling for Windows was sent to me. I have put it into the
1326        code, though I have no means of testing or verifying it.
1327    
1328     7. Added the pcre_refcount() auxiliary function.
1329    
1330     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1331        match before or at the first newline in the subject string. In pcretest,
1332        the /f option on a pattern can be used to set this.
1333    
1334     9. A repeated \w when used in UTF-8 mode with characters greater than 256
1335        would behave wrongly. This has been present in PCRE since release 4.0.
1336    
1337    10. A number of changes to the pcregrep command:
1338    
1339        (a) Refactored how -x works; insert ^(...)$ instead of setting
1340            PCRE_ANCHORED and checking the length, in preparation for adding
1341            something similar for -w.
1342    
1343        (b) Added the -w (match as a word) option.
1344    
1345        (c) Refactored the way lines are read and buffered so as to have more
1346            than one at a time available.
1347    
1348        (d) Implemented a pcregrep test script.
1349    
1350        (e) Added the -M (multiline match) option. This allows patterns to match
1351            over several lines of the subject. The buffering ensures that at least
1352            8K, or the rest of the document (whichever is the shorter) is available
1353            for matching (and similarly the previous 8K for lookbehind assertions).
1354    
1355        (f) Changed the --help output so that it now says
1356    
1357              -w, --word-regex(p)
1358    
1359            instead of two lines, one with "regex" and the other with "regexp"
1360            because that confused at least one person since the short forms are the
1361            same. (This required a bit of code, as the output is generated
1362            automatically from a table. It wasn't just a text change.)
1363    
1364        (g) -- can be used to terminate pcregrep options if the next thing isn't an
1365            option but starts with a hyphen. Could be a pattern or a path name
1366            starting with a hyphen, for instance.
1367    
1368        (h) "-" can be given as a file name to represent stdin.
1369    
1370        (i) When file names are being printed, "(standard input)" is used for
1371            the standard input, for compatibility with GNU grep. Previously
1372            "<stdin>" was used.
1373    
1374        (j) The option --label=xxx can be used to supply a name to be used for
1375            stdin when file names are being printed. There is no short form.
1376    
1377        (k) Re-factored the options decoding logic because we are going to add
1378            two more options that take data. Such options can now be given in four
1379            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1380    
1381        (l) Added the -A, -B, and -C options for requesting that lines of context
1382            around matches be printed.
1383    
1384        (m) Added the -L option to print the names of files that do not contain
1385            any matching lines, that is, the complement of -l.
1386    
1387        (n) The return code is 2 if any file cannot be opened, but pcregrep does
1388            continue to scan other files.
1389    
1390        (o) The -s option was incorrectly implemented. For compatibility with other
1391            greps, it now suppresses the error message for a non-existent or non-
1392            accessible file (but not the return code). There is a new option called
1393            -q that suppresses the output of matching lines, which was what -s was
1394            previously doing.
1395    
1396        (p) Added --include and --exclude options to specify files for inclusion
1397            and exclusion when recursing.
1398    
1399    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1400        Hopefully, it now does.
1401    
1402    12. Missing cast in pcre_study().
1403    
1404    13. Added an "uninstall" target to the makefile.
1405    
1406    14. Replaced "extern" in the function prototypes in Makefile.in with
1407        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1408        world, but is set differently for Windows.
1409    
1410    15. Added a second compiling function called pcre_compile2(). The only
1411        difference is that it has an extra argument, which is a pointer to an
1412        integer error code. When there is a compile-time failure, this is set
1413        non-zero, in addition to the error test pointer being set to point to an
1414        error message. The new argument may be NULL if no error number is required
1415        (but then you may as well call pcre_compile(), which is now just a
1416        wrapper). This facility is provided because some applications need a
1417        numeric error indication, but it has also enabled me to tidy up the way
1418        compile-time errors are handled in the POSIX wrapper.
1419    
1420    16. Added VPATH=.libs to the makefile; this should help when building with one
1421        prefix path and installing with another. (Or so I'm told by someone who
1422        knows more about this stuff than I do.)
1423    
1424    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1425        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1426        match everything, including newlines. This is not POSIX-compatible, but
1427        somebody wanted the feature. From pcretest it can be activated by using
1428        both the P and the s flags.
1429    
1430    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1431    
1432    19. libpcre.pc was being incorrectly installed as executable.
1433    
1434    20. A couple of places in pcretest check for end-of-line by looking for '\n';
1435        it now also looks for '\r' so that it will work unmodified on Windows.
1436    
1437    21. Added Google's contributed C++ wrapper to the distribution.
1438    
1439    22. Added some untidy missing memory free() calls in pcretest, to keep
1440        Electric Fence happy when testing.
1441    
1442    
1443    
1444    Version 5.0 13-Sep-04
1445    ---------------------
1446    
1447     1. Internal change: literal characters are no longer packed up into items
1448        containing multiple characters in a single byte-string. Each character
1449        is now matched using a separate opcode. However, there may be more than one
1450        byte in the character in UTF-8 mode.
1451    
1452     2. The pcre_callout_block structure has two new fields: pattern_position and
1453        next_item_length. These contain the offset in the pattern to the next match
1454        item, and its length, respectively.
1455    
1456     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1457        insertion of callouts before each pattern item. Added the /C option to
1458        pcretest to make use of this.
1459    
1460     4. On the advice of a Windows user, the lines
1461    
1462          #if defined(_WIN32) || defined(WIN32)
1463          _setmode( _fileno( stdout ), 0x8000 );
1464          #endif  /* defined(_WIN32) || defined(WIN32) */
1465    
1466        have been added to the source of pcretest. This apparently does useful
1467        magic in relation to line terminators.
1468    
1469     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1470        for the benefit of those environments where the "b" makes a difference.
1471    
1472     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1473        to know about it. I have put a hack into configure.in that adds in code
1474        to set GCC=yes if CC=icc. This seems to end up at a point in the
1475        generated configure script that is early enough to affect the setting of
1476        compiler options, which is what is needed, but I have no means of testing
1477        whether it really works. (The user who reported this had patched the
1478        generated configure script, which of course I cannot do.)
1479    
1480        LATER: After change 22 below (new libtool files), the configure script
1481        seems to know about icc (and also ecc). Therefore, I have commented out
1482        this hack in configure.in.
1483    
1484     7. Added support for pkg-config (2 patches were sent in).
1485    
1486     8. Negated POSIX character classes that used a combination of internal tables
1487        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1488        [[:^ascii]]. Typically, they would match almost any characters. The other
1489        POSIX classes were not broken in this way.
1490    
1491     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1492        to find the match, as PCRE was deluded into thinking that the match had to
1493        start at the start point or following a newline. The same bug applied to
1494        patterns with negative forward assertions or any backward assertions
1495        preceding ".*" at the start, unless the pattern required a fixed first
1496        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1497    
1498    10. In UTF-8 mode, when moving forwards in the subject after a failed match
1499        starting at the last subject character, bytes beyond the end of the subject
1500        string were read.
1501    
1502    11. Renamed the variable "class" as "classbits" to make life easier for C++
1503        users. (Previously there was a macro definition, but it apparently wasn't
1504        enough.)
1505    
1506    12. Added the new field "tables" to the extra data so that tables can be passed
1507        in at exec time, or the internal tables can be re-selected. This allows
1508        a compiled regex to be saved and re-used at a later time by a different
1509        program that might have everything at different addresses.
1510    
1511    13. Modified the pcre-config script so that, when run on Solaris, it shows a
1512        -R library as well as a -L library.
1513    
1514    14. The debugging options of pcretest (-d on the command line or D on a
1515        pattern) showed incorrect output for anything following an extended class
1516        that contained multibyte characters and which was followed by a quantifier.
1517    
1518    15. Added optional support for general category Unicode character properties
1519        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1520        support. It adds about 90K to the size of the library. The meanings of the
1521        inbuilt class escapes such as \d and \s have NOT been changed.
1522    
1523    16. Updated pcredemo.c to include calls to free() to release the memory for the
1524        compiled pattern.
1525    
1526    17. The generated file chartables.c was being created in the source directory
1527        instead of in the building directory. This caused the build to fail if the
1528        source directory was different from the building directory, and was
1529        read-only.
1530    
1531    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1532        file. No doubt somebody will tell me if they don't make sense... Also added
1533        Dan Mooney's comments about building on OpenVMS.
1534    
1535    19. Added support for partial matching via the PCRE_PARTIAL option for
1536        pcre_exec() and the \P data escape in pcretest.
1537    
1538    20. Extended pcretest with 3 new pattern features:
1539    
1540        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
1541              write the compiled pattern to the file whose name is "rest-of-line".
1542              This is a straight binary dump of the data, with the saved pointer to
1543              the character tables forced to be NULL. The study data, if any, is
1544              written too. After writing, pcretest reads a new pattern.
1545    
1546        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1547              compiled pattern from the given file. There must not be any
1548              occurrences of "<" in the file name (pretty unlikely); if there are,
1549              pcretest will instead treat the initial "<" as a pattern delimiter.
1550              After reading in the pattern, pcretest goes on to read data lines as
1551              usual.
1552    
1553        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1554              and 16-bit fields in a compiled pattern, to simulate a pattern that
1555              was compiled on a host of opposite endianness.
1556    
1557    21. The pcre-exec() function can now cope with patterns that were compiled on
1558        hosts of opposite endianness, with this restriction:
1559    
1560          As for any compiled expression that is saved and used later, the tables
1561          pointer field cannot be preserved; the extra_data field in the arguments
1562          to pcre_exec() should be used to pass in a tables address if a value
1563          other than the default internal tables were used at compile time.
1564    
1565    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1566        now diagnosed as an error. Previously, most of the time, a negative number
1567        would have been treated as zero, but if in addition "ovector" was passed as
1568        NULL, a crash could occur.
1569    
1570    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1571        new versions from the libtool 1.5 distribution (the last one is a copy of
1572        a file called libtool.m4). This seems to have fixed the need to patch
1573        "configure" to support Darwin 1.3 (which I used to do). However, I still
1574        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1575        workstation).
1576    
1577    24. Changed the PCRE licence to be the more standard "BSD" licence.
1578    
1579    
1580    Version 4.5 01-Dec-03
1581    ---------------------
1582    
1583     1. There has been some re-arrangement of the code for the match() function so
1584        that it can be compiled in a version that does not call itself recursively.
1585        Instead, it keeps those local variables that need separate instances for
1586        each "recursion" in a frame on the heap, and gets/frees frames whenever it
1587        needs to "recurse". Keeping track of where control must go is done by means
1588        of setjmp/longjmp. The whole thing is implemented by a set of macros that
1589        hide most of the details from the main code, and operates only if
1590        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1591        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1592        operating.
1593    
1594        To make it easier for callers to provide specially tailored get/free
1595        functions for this usage, two new functions, pcre_stack_malloc, and
1596        pcre_stack_free, are used. They are always called in strict stacking order,
1597        and the size of block requested is always the same.
1598    
1599        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1600        PCRE has been compiled to use the stack or the heap for recursion. The
1601        -C option of pcretest uses this to show which version is compiled.
1602    
1603        A new data escape \S, is added to pcretest; it causes the amounts of store
1604        obtained and freed by both kinds of malloc/free at match time to be added
1605        to the output.
1606    
1607     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1608        what's available on my current Linux desktop machine.
1609    
1610     3. When matching a UTF-8 string, the test for a valid string at the start has
1611        been extended. If start_offset is not zero, PCRE now checks that it points
1612        to a byte that is the start of a UTF-8 character. If not, it returns
1613        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1614        this is necessary because there may be backward assertions in the pattern.
1615        When matching the same subject several times, it may save resources to use
1616        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1617    
1618     4. The code for checking the validity of UTF-8 strings has been tightened so
1619        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1620        containing "overlong sequences".
1621    
1622     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1623        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1624        should have been "&", but it just so happened that all the cases this let
1625        through by mistake were picked up later in the function.
1626    
1627     6. I had used a variable called "isblank" - this is a C99 function, causing
1628        some compilers to warn. To avoid this, I renamed it (as "blankclass").
1629    
1630     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1631        prompting; (b) run "./pcretest /dev/null" at the start of the test script
1632        so the version is shown; (c) stop "make test" echoing "./RunTest".
1633    
1634     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1635    
1636     9. The prototype for memmove() for systems that don't have it was using
1637        size_t, but the inclusion of the header that defines size_t was later. I've
1638        moved the #includes for the C headers earlier to avoid this.
1639    
1640    10. Added some adjustments to the code to make it easier to compiler on certain
1641        special systems:
1642    
1643          (a) Some "const" qualifiers were missing.
1644          (b) Added the macro EXPORT before all exported functions; by default this
1645              is defined to be empty.
1646          (c) Changed the dftables auxiliary program (that builds chartables.c) so
1647              that it reads its output file name as an argument instead of writing
1648              to the standard output and assuming this can be redirected.
1649    
1650    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
1651        class containing characters with values greater than 255, PCRE compilation
1652        went into a loop.
1653    
1654    12. A recursive reference to a subpattern that was within another subpattern
1655        that had a minimum quantifier of zero caused PCRE to crash. For example,
1656        (x(y(?2))z)? provoked this bug with a subject that got as far as the
1657        recursion. If the recursively-called subpattern itself had a zero repeat,
1658        that was OK.
1659    
1660    13. In pcretest, the buffer for reading a data line was set at 30K, but the
1661        buffer into which it was copied (for escape processing) was still set at
1662        1024, so long lines caused crashes.
1663    
1664    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
1665        "internal error: code overflow...". This applied to any character class
1666        that was followed by a possessive quantifier.
1667    
1668    15. Modified the Makefile to add libpcre.la as a prerequisite for
1669        libpcreposix.la because I was told this is needed for a parallel build to
1670        work.
1671    
1672    16. If a pattern that contained .* following optional items at the start was
1673        studied, the wrong optimizing data was generated, leading to matching
1674        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
1675        matching string must start with a or b or c. The correct conclusion for
1676        this pattern is that a match can start with any character.
1677    
1678    
1679  Version 4.4 13-Aug-03  Version 4.4 13-Aug-03
1680  ---------------------  ---------------------
1681    

Legend:
Removed from v.71  
changed lines
  Added in v.272

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12