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

Diff of /code/trunk/ChangeLog

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

revision 45 by nigel, Sat Feb 24 21:39:25 2007 UTC revision 239 by ph10, Tue Sep 11 13:46:30 2007 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4    Version 7.4 10-Sep-07
5    ---------------------
6    
7    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
8        means that a class such as [\s] counted as "explicit reference to CR or
9        LF". That isn't really right - the whole point of the change was to try to
10        help when there was an actual mention of one of the two characters. So now
11        the change happens only if \r or \n (or a literal CR or LF) character is
12        encountered.
13    
14    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
15        of both had grown to the point where there were only 3 bits left.
16        Fortunately, there was spare space in the data structure, and so I have
17        moved the internal flags into a new 16-bit field to free up more option
18        bits.
19    
20    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
21        but did not set the internal JCHANGED flag - either of these is enough to
22        control the way the "get" function works - but the PCRE_INFO_JCHANGED
23        facility is supposed to tell if (?J) was ever used, so now (?J) at the
24        start sets both bits.
25    
26    4.  Added options (at build time, compile time, exec time) to change \R from
27        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
28    
29    5.  doc/pcresyntax.html was missing from the distribution.
30    
31    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
32        compatibility, even though it is no longer used.
33    
34    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
35        strtoull to pcrecpp.cc to select the available functions in WIN32 (where
36        different names are used).
37    
38    8.  Changed all #include <config.h> to #include "config.h". There were also
39        some further <pcre.h> cases that I changed to "pcre.h".
40    
41    9.  When pcregrep was used with the --colour option, it missed the line ending
42        sequence off the lines that it output.
43    
44    
45    Version 7.3 28-Aug-07
46    ---------------------
47    
48     1. In the rejigging of the build system that eventually resulted in 7.1, the
49        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
50        brackets there is not right, since it causes compilers to look for an
51        installed pcre.h, not the version that is in the source that is being
52        compiled (which of course may be different). I have changed it back to:
53    
54          #include "pcre.h"
55    
56        I have a vague recollection that the change was concerned with compiling in
57        different directories, but in the new build system, that is taken care of
58        by the VPATH setting the Makefile.
59    
60     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
61        when the subject happened to end in the byte 0x85 (e.g. if the last
62        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
63        characters but of course it shouldn't be taken as a newline when it is part
64        of another character. The bug was that, for an unlimited repeat of . in
65        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
66        characters when looking for a newline.
67    
68     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
69    
70     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
71        in debug output.
72    
73     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
74        long printing in the pcrecpp unittest when running under MinGW.
75    
76     6. ESC_K was left out of the EBCDIC table.
77    
78     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
79        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
80        limit also applies to "virtual nesting" when a pattern is recursive, and in
81        this case 1000 isn't so big. I have been able to remove this limit at the
82        expense of backing off one optimization in certain circumstances. Normally,
83        when pcre_exec() would call its internal match() function recursively and
84        immediately return the result unconditionally, it uses a "tail recursion"
85        feature to save stack. However, when a subpattern that can match an empty
86        string has an unlimited repetition quantifier, it no longer makes this
87        optimization. That gives it a stack frame in which to save the data for
88        checking that an empty string has been matched. Previously this was taken
89        from the 1000-entry workspace that had been reserved. So now there is no
90        explicit limit, but more stack is used.
91    
92     8. Applied Daniel's patches to solve problems with the import/export magic
93        syntax that is required for Windows, and which was going wrong for the
94        pcreposix and pcrecpp parts of the library. These were overlooked when this
95        problem was solved for the main library.
96    
97     9. There were some crude static tests to avoid integer overflow when computing
98        the size of patterns that contain repeated groups with explicit upper
99        limits. As the maximum quantifier is 65535, the maximum group length was
100        set at 30,000 so that the product of these two numbers did not overflow a
101        32-bit integer. However, it turns out that people want to use groups that
102        are longer than 30,000 bytes (though not repeat them that many times).
103        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
104        made it possible to implement the integer overflow checks in a much more
105        dynamic way, which I have now done. The artificial limitation on group
106        length has been removed - we now have only the limit on the total length of
107        the compiled pattern, which depends on the LINK_SIZE setting.
108    
109    10. Fixed a bug in the documentation for get/copy named substring when
110        duplicate names are permitted. If none of the named substrings are set, the
111        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
112        empty string.
113    
114    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
115        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
116        because the ] is interpreted as the first data character and the
117        terminating ] is not found. PCRE has been made compatible with Perl in this
118        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
119        cause memory overwriting.
120    
121    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
122        string has been matched (to stop an infinite loop). It was not recognizing
123        a conditional subpattern that could match an empty string if that
124        subpattern was within another subpattern. For example, it looped when
125        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
126        condition was not nested. This bug has been fixed.
127    
128    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
129        past the start of the subject in the presence of bytes with the top bit
130        set, for example "\x8aBCD".
131    
132    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
133        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
134    
135    14. Optimized (?!) to (*FAIL).
136    
137    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
138        This restricts code points to be within the range 0 to 0x10FFFF, excluding
139        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
140        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
141        does: it's just the validity check that is more restrictive.
142    
143    16. Inserted checks for integer overflows during escape sequence (backslash)
144        processing, and also fixed erroneous offset values for syntax errors during
145        backslash processing.
146    
147    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
148        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
149    
150    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
151        caused an overrun.
152    
153    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
154        something other than just ASCII characters) inside a group that had an
155        unlimited repeat caused a loop at compile time (while checking to see
156        whether the group could match an empty string).
157    
158    20. Debugging a pattern containing \p or \P could cause a crash. For example,
159        [\P{Any}] did so. (Error in the code for printing property names.)
160    
161    21. An orphan \E inside a character class could cause a crash.
162    
163    22. A repeated capturing bracket such as (A)? could cause a wild memory
164        reference during compilation.
165    
166    23. There are several functions in pcre_compile() that scan along a compiled
167        expression for various reasons (e.g. to see if it's fixed length for look
168        behind). There were bugs in these functions when a repeated \p or \P was
169        present in the pattern. These operators have additional parameters compared
170        with \d, etc, and these were not being taken into account when moving along
171        the compiled data. Specifically:
172    
173        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
174            length.
175    
176        (b) An item such as \pL+ within a repeated group could cause crashes or
177            loops.
178    
179        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
180            "reference to non-existent subpattern" error.
181    
182        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
183    
184    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
185        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
186    
187    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
188    
189    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
190        character were causing crashes (broken optimization).
191    
192    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
193        \p or \P) caused a compile-time loop.
194    
195    28. More problems have arisen in unanchored patterns when CRLF is a valid line
196        break. For example, the unstudied pattern [\r\n]A does not match the string
197        "\r\nA" because change 7.0/46 below moves the current point on by two
198        characters after failing to match at the start. However, the pattern \nA
199        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
200        the same is true. There doesn't seem any very clean way out of this, but
201        what I have chosen to do makes the common cases work: PCRE now takes note
202        of whether there can be an explicit match for \r or \n anywhere in the
203        pattern, and if so, 7.0/46 no longer applies. As part of this change,
204        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
205        pattern has explicit CR or LF references.
206    
207    29. Added (*CR) etc for changing newline setting at start of pattern.
208    
209    
210    Version 7.2 19-Jun-07
211    ---------------------
212    
213     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
214        which is apparently normally available under Windows.
215    
216     2. Re-jig the pcregrep tests with different newline settings in an attempt
217        to make them independent of the local environment's newline setting.
218    
219     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
220    
221     4. Some of the "internals" tests were previously cut out when the link size
222        was not 2, because the output contained actual offsets. The recent new
223        "Z" feature of pcretest means that these can be cut out, making the tests
224        usable with all link sizes.
225    
226     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
227        stack recursion. This gives a massive performance boost under BSD, but just
228        a small improvement under Linux. However, it saves one field in the frame
229        in all cases.
230    
231     6. Added more features from the forthcoming Perl 5.10:
232    
233        (a) (?-n) (where n is a string of digits) is a relative subroutine or
234            recursion call. It refers to the nth most recently opened parentheses.
235    
236        (b) (?+n) is also a relative subroutine call; it refers to the nth next
237            to be opened parentheses.
238    
239        (c) Conditions that refer to capturing parentheses can be specified
240            relatively, for example, (?(-2)... or (?(+3)...
241    
242        (d) \K resets the start of the current match so that everything before
243            is not part of it.
244    
245        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
246    
247        (f) \g{name} is another synonym - part of Perl 5.10's unification of
248            reference syntax.
249    
250        (g) (?| introduces a group in which the numbering of parentheses in each
251            alternative starts with the same number.
252    
253        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
254    
255     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
256        PCRE_INFO_JCHANGED.
257    
258     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
259        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
260        for detecting groups that can match an empty string.
261    
262     9. A pattern with a very large number of alternatives (more than several
263        hundred) was running out of internal workspace during the pre-compile
264        phase, where pcre_compile() figures out how much memory will be needed. A
265        bit of new cunning has reduced the workspace needed for groups with
266        alternatives. The 1000-alternative test pattern now uses 12 bytes of
267        workspace instead of running out of the 4096 that are available.
268    
269    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
270    
271    11. Applied patch from Google to remove an optimization that didn't quite work.
272        The report of the bug said:
273    
274          pcrecpp::RE("a*").FullMatch("aaa") matches, while
275          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
276          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
277    
278    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
279        it matched the wrong number of bytes.
280    
281    
282    Version 7.1 24-Apr-07
283    ---------------------
284    
285     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
286        that is more "standard", making use of automake and other Autotools. There
287        is some re-arrangement of the files and adjustment of comments consequent
288        on this.
289    
290     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
291        for recursive directory scanning broke on some systems because the files
292        are not scanned in any specific order and on different systems the order
293        was different. A call to "sort" has been inserted into RunGrepTest for the
294        approprate test as a short-term fix. In the longer term there may be an
295        alternative.
296    
297     3. I had an email from Eric Raymond about problems translating some of PCRE's
298        man pages to HTML (despite the fact that I distribute HTML pages, some
299        people do their own conversions for various reasons). The problems
300        concerned the use of low-level troff macros .br and .in. I have therefore
301        removed all such uses from the man pages (some were redundant, some could
302        be replaced by .nf/.fi pairs). The 132html script that I use to generate
303        HTML has been updated to handle .nf/.fi and to complain if it encounters
304        .br or .in.
305    
306     4. Updated comments in configure.ac that get placed in config.h.in and also
307        arranged for config.h to be included in the distribution, with the name
308        config.h.generic, for the benefit of those who have to compile without
309        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
310    
311     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
312        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
313        makevp.bat for the current PCRE, using the additional files
314        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
315    
316     6. A Windows user reported a minor discrepancy with test 2, which turned out
317        to be caused by a trailing space on an input line that had got lost in his
318        copy. The trailing space was an accident, so I've just removed it.
319    
320     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
321        that is needed.
322    
323     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
324        as "const" (a) because they are and (b) because it helps the PHP
325        maintainers who have recently made a script to detect big data structures
326        in the php code that should be moved to the .rodata section. I remembered
327        to update Builducptable as well, so it won't revert if ucptable.h is ever
328        re-created.
329    
330     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
331        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
332        order to be able to cut out the UTF-8 tables in the latter when UTF-8
333        support is not required. This saves 1.5-2K of code, which is important in
334        some applications.
335    
336        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
337        so as not to refer to the tables, even though these functions will never be
338        called when UTF-8 support is disabled. Otherwise there are problems with a
339        shared library.
340    
341    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
342    
343        (a) It was defining its arguments as char * instead of void *.
344    
345        (b) It was assuming that all moves were upwards in memory; this was true
346            a long time ago when I wrote it, but is no longer the case.
347    
348        The emulated memove() is provided for those environments that have neither
349        memmove() nor bcopy(). I didn't think anyone used it these days, but that
350        is clearly not the case, as these two bugs were recently reported.
351    
352    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
353        and Detrail to create the HTML documentation, the .txt form of the man
354        pages, and it removes trailing spaces from listed files. It also creates
355        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
356        case, it wraps all the #defines with #ifndefs. This script should be run
357        before "make dist".
358    
359    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
360        with Unicode property support.
361    
362        (a) For a maximizing quantifier, if the two different cases of the
363            character were of different lengths in their UTF-8 codings (there are
364            some cases like this - I found 11), and the matching function had to
365            back up over a mixture of the two cases, it incorrectly assumed they
366            were both the same length.
367    
368        (b) When PCRE was configured to use the heap rather than the stack for
369            recursion during matching, it was not correctly preserving the data for
370            the other case of a UTF-8 character when checking ahead for a match
371            while processing a minimizing repeat. If the check also involved
372            matching a wide character, but failed, corruption could cause an
373            erroneous result when trying to check for a repeat of the original
374            character.
375    
376    13. Some tidying changes to the testing mechanism:
377    
378        (a) The RunTest script now detects the internal link size and whether there
379            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
380            values substituted by "configure". (The RunGrepTest script already did
381            this for UTF-8.) The configure.ac script no longer substitutes the
382            relevant variables.
383    
384        (b) The debugging options /B and /D in pcretest show the compiled bytecode
385            with length and offset values. This means that the output is different
386            for different internal link sizes. Test 2 is skipped for link sizes
387            other than 2 because of this, bypassing the problem. Unfortunately,
388            there was also a test in test 3 (the locale tests) that used /B and
389            failed for link sizes other than 2. Rather than cut the whole test out,
390            I have added a new /Z option to pcretest that replaces the length and
391            offset values with spaces. This is now used to make test 3 independent
392            of link size. (Test 2 will be tidied up later.)
393    
394    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
395        segmentation fault instead of returning the appropriate error message.
396    
397    15. In multiline mode when the newline sequence was set to "any", the pattern
398        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
399        This doesn't seem right; it now treats the CRLF combination as the line
400        ending, and so does not match in that case. It's only a pattern such as ^$
401        that would hit this one: something like ^ABC$ would have failed after \r
402        and then tried again after \r\n.
403    
404    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
405        in an attempt to make files that differ only in their line terminators
406        compare equal. This works on Linux.
407    
408    17. Under certain error circumstances pcregrep might try to free random memory
409        as it exited. This is now fixed, thanks to valgrind.
410    
411    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
412        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
413        was because its rules for how to advance for /g after matching an empty
414        string at the end of a line did not allow for this case. They now check for
415        it specially.
416    
417    20. pcretest is supposed to handle patterns and data of any length, by
418        extending its buffers when necessary. It was getting this wrong when the
419        buffer for a data line had to be extended.
420    
421    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
422        CRLF as a newline sequence.
423    
424    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
425        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
426        I have nevertheless tidied it up.
427    
428    23. Added some casts to kill warnings from HP-UX ia64 compiler.
429    
430    24. Added a man page for pcre-config.
431    
432    
433    Version 7.0 19-Dec-06
434    ---------------------
435    
436     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
437        moving to gcc 4.1.1.
438    
439     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
440        sys/time.h, which is documented as needed for this function. It doesn't
441        seem to matter on Linux, but it showed up on some releases of OS X.
442    
443     3. It seems that there are systems where bytes whose values are greater than
444        127 match isprint() in the "C" locale. The "C" locale should be the
445        default when a C program starts up. In most systems, only ASCII printing
446        characters match isprint(). This difference caused the output from pcretest
447        to vary, making some of the tests fail. I have changed pcretest so that:
448    
449        (a) When it is outputting text in the compiled version of a pattern, bytes
450            other than 32-126 are always shown as hex escapes.
451    
452        (b) When it is outputting text that is a matched part of a subject string,
453            it does the same, unless a different locale has been set for the match
454            (using the /L modifier). In this case, it uses isprint() to decide.
455    
456     4. Fixed a major bug that caused incorrect computation of the amount of memory
457        required for a compiled pattern when options that changed within the
458        pattern affected the logic of the preliminary scan that determines the
459        length. The relevant options are -x, and -i in UTF-8 mode. The result was
460        that the computed length was too small. The symptoms of this bug were
461        either the PCRE error "internal error: code overflow" from pcre_compile(),
462        or a glibc crash with a message such as "pcretest: free(): invalid next
463        size (fast)". Examples of patterns that provoked this bug (shown in
464        pcretest format) are:
465    
466          /(?-x: )/x
467          /(?x)(?-x: \s*#\s*)/
468          /((?i)[\x{c0}])/8
469          /(?i:[\x{c0}])/8
470    
471        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
472        is now done differently.
473    
474     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
475        wrapper classes; (b) implement a new function in the C++ scanner that is
476        more efficient than the old way of doing things because it avoids levels of
477        recursion in the regex matching; (c) add a paragraph to the documentation
478        for the FullMatch() function.
479    
480     6. The escape sequence \n was being treated as whatever was defined as
481        "newline". Not only was this contrary to the documentation, which states
482        that \n is character 10 (hex 0A), but it also went horribly wrong when
483        "newline" was defined as CRLF. This has been fixed.
484    
485     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
486        was being set to -1 for the "end of line" case (supposedly a value that no
487        character can have). Though this value is never used (the check for end of
488        line is "zero bytes in current character"), it caused compiler complaints.
489        I've changed it to 0xffffffff.
490    
491     8. In pcre_version.c, the version string was being built by a sequence of
492        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
493        string (as it is for production releases) called a macro with an empty
494        argument. The C standard says the result of this is undefined. The gcc
495        compiler treats it as an empty string (which was what was wanted) but it is
496        reported that Visual C gives an error. The source has been hacked around to
497        avoid this problem.
498    
499     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
500        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
501        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
502        of them did).
503    
504    10. Originally, pcretest opened its input and output without "b"; then I was
505        told that "b" was needed in some environments, so it was added for release
506        5.0 to both the input and output. (It makes no difference on Unix-like
507        systems.) Later I was told that it is wrong for the input on Windows. I've
508        now abstracted the modes into two macros, to make it easier to fiddle with
509        them, and removed "b" from the input mode under Windows.
510    
511    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
512    
513    12. Added -help and --help to pcretest as an official way of being reminded
514        of the options.
515    
516    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
517        and pcrecpp.cc because they annoy compilers at high warning levels.
518    
519    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
520    
521    15. Fixed an occurrence of == in configure.ac that should have been = (shell
522        scripts are not C programs :-) and which was not noticed because it works
523        on Linux.
524    
525    16. pcretest is supposed to handle any length of pattern and data line (as one
526        line or as a continued sequence of lines) by extending its input buffer if
527        necessary. This feature was broken for very long pattern lines, leading to
528        a string of junk being passed to pcre_compile() if the pattern was longer
529        than about 50K.
530    
531    17. I have done a major re-factoring of the way pcre_compile() computes the
532        amount of memory needed for a compiled pattern. Previously, there was code
533        that made a preliminary scan of the pattern in order to do this. That was
534        OK when PCRE was new, but as the facilities have expanded, it has become
535        harder and harder to keep it in step with the real compile phase, and there
536        have been a number of bugs (see for example, 4 above). I have now found a
537        cunning way of running the real compile function in a "fake" mode that
538        enables it to compute how much memory it would need, while actually only
539        ever using a few hundred bytes of working memory and without too many
540        tests of the mode. This should make future maintenance and development
541        easier. A side effect of this work is that the limit of 200 on the nesting
542        depth of parentheses has been removed (though this was never a serious
543        limitation, I suspect). However, there is a downside: pcre_compile() now
544        runs more slowly than before (30% or more, depending on the pattern). I
545        hope this isn't a big issue. There is no effect on runtime performance.
546    
547    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
548        newline (only possible for the last line of a file) and it was a
549        pattern that set a locale (followed by /Lsomething), pcretest crashed.
550    
551    19. Added additional timing features to pcretest. (1) The -tm option now times
552        matching only, not compiling. (2) Both -t and -tm can be followed, as a
553        separate command line item, by a number that specifies the number of
554        repeats to use when timing. The default is 50000; this gives better
555        precision, but takes uncomfortably long for very large patterns.
556    
557    20. Extended pcre_study() to be more clever in cases where a branch of a
558        subpattern has no definite first character. For example, (a*|b*)[cd] would
559        previously give no result from pcre_study(). Now it recognizes that the
560        first character must be a, b, c, or d.
561    
562    21. There was an incorrect error "recursive call could loop indefinitely" if
563        a subpattern (or the entire pattern) that was being tested for matching an
564        empty string contained only one non-empty item after a nested subpattern.
565        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
566        incorrectly, because the \d was being skipped in the check.
567    
568    22. The pcretest program now has a new pattern option /B and a command line
569        option -b, which is equivalent to adding /B to every pattern. This causes
570        it to show the compiled bytecode, without the additional information that
571        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
572        is the same as /B/I).
573    
574    23. A new optimization is now able automatically to treat some sequences such
575        as a*b as a*+b. More specifically, if something simple (such as a character
576        or a simple class like \d) has an unlimited quantifier, and is followed by
577        something that cannot possibly match the quantified thing, the quantifier
578        is automatically "possessified".
579    
580    24. A recursive reference to a subpattern whose number was greater than 39
581        went wrong under certain circumstances in UTF-8 mode. This bug could also
582        have affected the operation of pcre_study().
583    
584    25. Realized that a little bit of performance could be had by replacing
585        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
586    
587    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
588    
589    27. Possessive quantifiers such as a++ were previously implemented by turning
590        them into atomic groups such as ($>a+). Now they have their own opcodes,
591        which improves performance. This includes the automatically created ones
592        from 23 above.
593    
594    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
595        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
596        the first matched character to be a colon. This applied both to named and
597        numbered groups.
598    
599    29. The ucpinternal.h header file was missing its idempotency #ifdef.
600    
601    30. I was sent a "project" file called libpcre.a.dev which I understand makes
602        building PCRE on Windows easier, so I have included it in the distribution.
603    
604    31. There is now a check in pcretest against a ridiculously large number being
605        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
606        loop, the loop is abandoned.
607    
608    32. Forward references to subpatterns in conditions such as (?(2)...) where
609        subpattern 2 is defined later cause pcre_compile() to search forwards in
610        the pattern for the relevant set of parentheses. This search went wrong
611        when there were unescaped parentheses in a character class, parentheses
612        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
613    
614    33. "Subroutine" calls and backreferences were previously restricted to
615        referencing subpatterns earlier in the regex. This restriction has now
616        been removed.
617    
618    34. Added a number of extra features that are going to be in Perl 5.10. On the
619        whole, these are just syntactic alternatives for features that PCRE had
620        previously implemented using the Python syntax or my own invention. The
621        other formats are all retained for compatibility.
622    
623        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
624            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
625            also .NET compatible.
626    
627        (b) A recursion or subroutine call to a named group can now be defined as
628            (?&name) as well as (?P>name).
629    
630        (c) A backreference to a named group can now be defined as \k<name> or
631            \k'name' as well as (?P=name). The new forms, as well as being in Perl
632            5.10, are also .NET compatible.
633    
634        (d) A conditional reference to a named group can now use the syntax
635            (?(<name>) or (?('name') as well as (?(name).
636    
637        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
638            groups (named and numbered) that are never evaluated inline, but can be
639            called as "subroutines" from elsewhere. In effect, the DEFINE condition
640            is always false. There may be only one alternative in such a group.
641    
642        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
643            as the simple (?(R). The condition is true only if the most recent
644            recursion is that of the given number or name. It does not search out
645            through the entire recursion stack.
646    
647        (g) The escape \gN or \g{N} has been added, where N is a positive or
648            negative number, specifying an absolute or relative reference.
649    
650    35. Tidied to get rid of some further signed/unsigned compiler warnings and
651        some "unreachable code" warnings.
652    
653    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
654        things, this adds five new scripts.
655    
656    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
657        There were also incompatibilities regarding the handling of \Q..\E inside
658        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
659        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
660    
661    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
662        matches an empty string, and forcibly breaks the loop. There were bugs in
663        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
664        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
665        separate and independent bugs (that affected different cases) have been
666        fixed.
667    
668    39. Refactored the code to abolish the use of different opcodes for small
669        capturing bracket numbers. This is a tidy that I avoided doing when I
670        removed the limit on the number of capturing brackets for 3.5 back in 2001.
671        The new approach is not only tidier, it makes it possible to reduce the
672        memory needed to fix the previous bug (38).
673    
674    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
675        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
676        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
677        mode.
678    
679    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
680        report.
681    
682    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
683        copy construction and assignment in the C++ wrapper.
684    
685    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
686        couple of bugs that could have given wrong results in the "--newline=crlf"
687        case.
688    
689    44. Added a number of casts and did some reorganization of signed/unsigned int
690        variables following suggestions from Dair Grant. Also renamed the variable
691        "this" as "item" because it is a C++ keyword.
692    
693    45. Arranged for dftables to add
694    
695          #include "pcre_internal.h"
696    
697        to pcre_chartables.c because without it, gcc 4.x may remove the array
698        definition from the final binary if PCRE is built into a static library and
699        dead code stripping is activated.
700    
701    46. For an unanchored pattern, if a match attempt fails at the start of a
702        newline sequence, and the newline setting is CRLF or ANY, and the next two
703        characters are CRLF, advance by two characters instead of one.
704    
705    
706    Version 6.7 04-Jul-06
707    ---------------------
708    
709     1. In order to handle tests when input lines are enormously long, pcretest has
710        been re-factored so that it automatically extends its buffers when
711        necessary. The code is crude, but this _is_ just a test program. The
712        default size has been increased from 32K to 50K.
713    
714     2. The code in pcre_study() was using the value of the re argument before
715        testing it for NULL. (Of course, in any sensible call of the function, it
716        won't be NULL.)
717    
718     3. The memmove() emulation function in pcre_internal.h, which is used on
719        systems that lack both memmove() and bcopy() - that is, hardly ever -
720        was missing a "static" storage class specifier.
721    
722     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
723        containing an extended class (one that cannot be represented by a bitmap
724        because it contains high-valued characters or Unicode property items, e.g.
725        [\pZ]). Almost always one would set UTF-8 mode when processing such a
726        pattern, but PCRE should not loop if you do not (it no longer does).
727        [Detail: two cases were found: (a) a repeated subpattern containing an
728        extended class; (b) a recursive reference to a subpattern that followed a
729        previous extended class. It wasn't skipping over the extended class
730        correctly when UTF-8 mode was not set.]
731    
732     5. A negated single-character class was not being recognized as fixed-length
733        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
734        compile error "lookbehind assertion is not fixed length".
735    
736     6. The RunPerlTest auxiliary script was showing an unexpected difference
737        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
738        write a Perl script that can interpret lines of an input file either as
739        byte characters or as UTF-8, which is what "perltest" was being required to
740        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
741        can't do is switch easily at run time between having the "use utf8;" pragma
742        or not. In the end, I fudged it by using the RunPerlTest script to insert
743        "use utf8;" explicitly for the UTF-8 tests.
744    
745     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
746        the end of the subject string, contrary to the documentation and to what
747        Perl does. This was true of both matching functions. Now it matches only at
748        the start of the subject and immediately after *internal* newlines.
749    
750     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
751        a pointer to an int instead of a pointer to an unsigned long int. This
752        caused problems on 64-bit systems.
753    
754     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
755        instance of the 'standard' template library not being so standard".
756    
757    10. There was no check on the number of named subpatterns nor the maximum
758        length of a subpattern name. The product of these values is used to compute
759        the size of the memory block for a compiled pattern. By supplying a very
760        long subpattern name and a large number of named subpatterns, the size
761        computation could be caused to overflow. This is now prevented by limiting
762        the length of names to 32 characters, and the number of named subpatterns
763        to 10,000.
764    
765    11. Subpatterns that are repeated with specific counts have to be replicated in
766        the compiled pattern. The size of memory for this was computed from the
767        length of the subpattern and the repeat count. The latter is limited to
768        65535, but there was no limit on the former, meaning that integer overflow
769        could in principle occur. The compiled length of a repeated subpattern is
770        now limited to 30,000 bytes in order to prevent this.
771    
772    12. Added the optional facility to have named substrings with the same name.
773    
774    13. Added the ability to use a named substring as a condition, using the
775        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
776        are numbers (not recommended). Forward references are permitted.
777    
778    14. Added forward references in named backreferences (if you see what I mean).
779    
780    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
781        pattern could run off the end of the subject. For example, the pattern
782        "(?s)(.{1,5})"8 did this with the subject "ab".
783    
784    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
785        PCRE_CASELESS was set when matching characters that were quantified with ?
786        or *.
787    
788    17. A character class other than a single negated character that had a minimum
789        but no maximum quantifier - for example [ab]{6,} - was not handled
790        correctly by pce_dfa_exec(). It would match only one character.
791    
792    18. A valid (though odd) pattern that looked like a POSIX character
793        class but used an invalid character after [ (for example [[,abc,]]) caused
794        pcre_compile() to give the error "Failed: internal error: code overflow" or
795        in some cases to crash with a glibc free() error. This could even happen if
796        the pattern terminated after [[ but there just happened to be a sequence of
797        letters, a binary zero, and a closing ] in the memory that followed.
798    
799    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
800        over the years. Originally (before any Unicode support), just the bottom 8
801        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
802        output from "man perlunicode" includes this:
803    
804          The regular expression compiler produces polymorphic opcodes.  That
805          is, the pattern adapts to the data and automatically switches to
806          the Unicode character scheme when presented with Unicode data--or
807          instead uses a traditional byte scheme when presented with byte
808          data.
809    
810        Sadly, a wide octal escape does not cause a switch, and in a string with
811        no other multibyte characters, these octal escapes are treated as before.
812        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
813        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
814        Unicode string.
815    
816        I have not perpetrated such confusion in PCRE. Up till now, it took just
817        the bottom 8 bits, as in old Perl. I have now made octal escapes with
818        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
819        translate to the appropriate multibyte character.
820    
821    29. Applied some refactoring to reduce the number of warnings from Microsoft
822        and Borland compilers. This has included removing the fudge introduced
823        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
824        a warning about an unused variable.
825    
826    21. PCRE has not included VT (character 0x0b) in the set of whitespace
827        characters since release 4.0, because Perl (from release 5.004) does not.
828        [Or at least, is documented not to: some releases seem to be in conflict
829        with the documentation.] However, when a pattern was studied with
830        pcre_study() and all its branches started with \s, PCRE still included VT
831        as a possible starting character. Of course, this did no harm; it just
832        caused an unnecessary match attempt.
833    
834    22. Removed a now-redundant internal flag bit that recorded the fact that case
835        dependency changed within the pattern. This was once needed for "required
836        byte" processing, but is no longer used. This recovers a now-scarce options
837        bit. Also moved the least significant internal flag bit to the most-
838        significant bit of the word, which was not previously used (hangover from
839        the days when it was an int rather than a uint) to free up another bit for
840        the future.
841    
842    23. Added support for CRLF line endings as well as CR and LF. As well as the
843        default being selectable at build time, it can now be changed at runtime
844        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
845        specify that it is scanning data with non-default line endings.
846    
847    24. Changed the definition of CXXLINK to make it agree with the definition of
848        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
849    
850    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
851        recursions. This makes a big different to stack usage for some patterns.
852    
853    26. If a subpattern containing a named recursion or subroutine reference such
854        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
855        the space required for the compiled pattern went wrong and gave too small a
856        value. Depending on the environment, this could lead to "Failed: internal
857        error: code overflow at offset 49" or "glibc detected double free or
858        corruption" errors.
859    
860    27. Applied patches from Google (a) to support the new newline modes and (b) to
861        advance over multibyte UTF-8 characters in GlobalReplace.
862    
863    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
864        difference for some implementation of PCRE in some Windows version.
865    
866    29. Added some extra testing facilities to pcretest:
867    
868        \q<number>   in a data line sets the "match limit" value
869        \Q<number>   in a data line sets the "match recursion limt" value
870        -S <number>  sets the stack size, where <number> is in megabytes
871    
872        The -S option isn't available for Windows.
873    
874    
875    Version 6.6 06-Feb-06
876    ---------------------
877    
878     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
879        in pcreposix.h. I have copied the definition from pcre.h.
880    
881     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
882        because pcre.h is no longer a built file.
883    
884     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
885        not normally included in the compiled code.
886    
887    
888    Version 6.5 01-Feb-06
889    ---------------------
890    
891     1. When using the partial match feature with pcre_dfa_exec(), it was not
892        anchoring the second and subsequent partial matches at the new starting
893        point. This could lead to incorrect results. For example, with the pattern
894        /1234/, partially matching against "123" and then "a4" gave a match.
895    
896     2. Changes to pcregrep:
897    
898        (a) All non-match returns from pcre_exec() were being treated as failures
899            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
900            error message is output. Some extra information is given for the
901            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
902            probably the only errors that are likely to be caused by users (by
903            specifying a regex that has nested indefinite repeats, for instance).
904            If there are more than 20 of these errors, pcregrep is abandoned.
905    
906        (b) A binary zero was treated as data while matching, but terminated the
907            output line if it was written out. This has been fixed: binary zeroes
908            are now no different to any other data bytes.
909    
910        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
911            used to set a locale for matching. The --locale=xxxx long option has
912            been added (no short equivalent) to specify a locale explicitly on the
913            pcregrep command, overriding the environment variables.
914    
915        (d) When -B was used with -n, some line numbers in the output were one less
916            than they should have been.
917    
918        (e) Added the -o (--only-matching) option.
919    
920        (f) If -A or -C was used with -c (count only), some lines of context were
921            accidentally printed for the final match.
922    
923        (g) Added the -H (--with-filename) option.
924    
925        (h) The combination of options -rh failed to suppress file names for files
926            that were found from directory arguments.
927    
928        (i) Added the -D (--devices) and -d (--directories) options.
929    
930        (j) Added the -F (--fixed-strings) option.
931    
932        (k) Allow "-" to be used as a file name for -f as well as for a data file.
933    
934        (l) Added the --colo(u)r option.
935    
936        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
937            is not present by default.
938    
939     3. A nasty bug was discovered in the handling of recursive patterns, that is,
940        items such as (?R) or (?1), when the recursion could match a number of
941        alternatives. If it matched one of the alternatives, but subsequently,
942        outside the recursion, there was a failure, the code tried to back up into
943        the recursion. However, because of the way PCRE is implemented, this is not
944        possible, and the result was an incorrect result from the match.
945    
946        In order to prevent this happening, the specification of recursion has
947        been changed so that all such subpatterns are automatically treated as
948        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
949    
950     4. I had overlooked the fact that, in some locales, there are characters for
951        which isalpha() is true but neither isupper() nor islower() are true. In
952        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
953        and ordfeminine) are like this. This affected the treatment of \w and \W
954        when they appeared in character classes, but not when they appeared outside
955        a character class. The bit map for "word" characters is now created
956        separately from the results of isalnum() instead of just taking it from the
957        upper, lower, and digit maps. (Plus the underscore character, of course.)
958    
959     5. The above bug also affected the handling of POSIX character classes such as
960        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
961        permanent tables. Instead, the bit maps for such a class were previously
962        created as the appropriate unions of the upper, lower, and digit bitmaps.
963        Now they are created by subtraction from the [[:word:]] class, which has
964        its own bitmap.
965    
966     6. The [[:blank:]] character class matches horizontal, but not vertical space.
967        It is created by subtracting the vertical space characters (\x09, \x0a,
968        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
969        subtraction was done in the overall bitmap for a character class, meaning
970        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
971        be recognized. This bug has been fixed.
972    
973     7. Patches from the folks at Google:
974    
975          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
976          real life, but is still worth protecting against".
977    
978          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
979          regular expressions".
980    
981          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
982          have it.
983    
984          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
985          "configure" and the latter not, in order to fix a problem somebody had
986          with compiling the Arg class on HP-UX.
987    
988          (e) Improve the error-handling of the C++ wrapper a little bit.
989    
990          (f) New tests for checking recursion limiting.
991    
992     8. The pcre_memmove() function, which is used only if the environment does not
993        have a standard memmove() function (and is therefore rarely compiled),
994        contained two bugs: (a) use of int instead of size_t, and (b) it was not
995        returning a result (though PCRE never actually uses the result).
996    
997     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
998        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
999        returned instead of calling malloc() with an overflowing number that would
1000        most likely cause subsequent chaos.
1001    
1002    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1003    
1004    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1005        with this option is matched, the nmatch and pmatch options of regexec() are
1006        ignored.
1007    
1008    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1009        provided in case anyone wants to the the POSIX interface with UTF-8
1010        strings.
1011    
1012    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1013        C++ linking (needed for some HP-UX environments).
1014    
1015    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1016        (unused parameter) and in the pcre_printint() function (omitted "default"
1017        switch label when the default is to do nothing).
1018    
1019    15. Added some code to make it possible, when PCRE is compiled as a C++
1020        library, to replace subject pointers for pcre_exec() with a smart pointer
1021        class, thus making it possible to process discontinuous strings.
1022    
1023    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1024        much the same function. They were added by different people who were trying
1025        to make PCRE easy to compile on non-Unix systems. It has been suggested
1026        that PCRE_EXPORT be abolished now that there is more automatic apparatus
1027        for compiling on Windows systems. I have therefore replaced it with
1028        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1029        defaults to "extern" for C or "extern C" for C++, which works fine on
1030        Unix-like systems. It is now possible to override the value of PCRE_DATA_
1031        SCOPE with something explicit in config.h. In addition:
1032    
1033        (a) pcreposix.h still had just "extern" instead of either of these macros;
1034            I have replaced it with PCRE_DATA_SCOPE.
1035    
1036        (b) Functions such as _pcre_xclass(), which are internal to the library,
1037            but external in the C sense, all had PCRE_EXPORT in their definitions.
1038            This is apparently wrong for the Windows case, so I have removed it.
1039            (It makes no difference on Unix-like systems.)
1040    
1041    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1042        of recursive calls to match(). This is different to MATCH_LIMIT because
1043        that limits the total number of calls to match(), not all of which increase
1044        the depth of recursion. Limiting the recursion depth limits the amount of
1045        stack (or heap if NO_RECURSE is set) that is used. The default can be set
1046        when PCRE is compiled, and changed at run time. A patch from Google adds
1047        this functionality to the C++ interface.
1048    
1049    18. Changes to the handling of Unicode character properties:
1050    
1051        (a) Updated the table to Unicode 4.1.0.
1052    
1053        (b) Recognize characters that are not in the table as "Cn" (undefined).
1054    
1055        (c) I revised the way the table is implemented to a much improved format
1056            which includes recognition of ranges. It now supports the ranges that
1057            are defined in UnicodeData.txt, and it also amalgamates other
1058            characters into ranges. This has reduced the number of entries in the
1059            table from around 16,000 to around 3,000, thus reducing its size
1060            considerably. I realized I did not need to use a tree structure after
1061            all - a binary chop search is just as efficient. Having reduced the
1062            number of entries, I extended their size from 6 bytes to 8 bytes to
1063            allow for more data.
1064    
1065        (d) Added support for Unicode script names via properties such as \p{Han}.
1066    
1067    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1068        matching that character.
1069    
1070    20. When matching a repeated Unicode property with a minimum greater than zero,
1071        (for example \pL{2,}), PCRE could look past the end of the subject if it
1072        reached it while seeking the minimum number of characters. This could
1073        happen only if some of the characters were more than one byte long, because
1074        there is a check for at least the minimum number of bytes.
1075    
1076    21. Refactored the implementation of \p and \P so as to be more general, to
1077        allow for more different types of property in future. This has changed the
1078        compiled form incompatibly. Anybody with saved compiled patterns that use
1079        \p or \P will have to recompile them.
1080    
1081    22. Added "Any" and "L&" to the supported property types.
1082    
1083    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1084        but give a compile time error if the value is greater than 0xff.
1085    
1086    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1087        accidentally not being installed or uninstalled.
1088    
1089    25. The pcre.h file was built from pcre.h.in, but the only changes that were
1090        made were to insert the current release number. This seemed silly, because
1091        it made things harder for people building PCRE on systems that don't run
1092        "configure". I have turned pcre.h into a distributed file, no longer built
1093        by "configure", with the version identification directly included. There is
1094        no longer a pcre.h.in file.
1095    
1096        However, this change necessitated a change to the pcre-config script as
1097        well. It is built from pcre-config.in, and one of the substitutions was the
1098        release number. I have updated configure.ac so that ./configure now finds
1099        the release number by grepping pcre.h.
1100    
1101    26. Added the ability to run the tests under valgrind.
1102    
1103    
1104    Version 6.4 05-Sep-05
1105    ---------------------
1106    
1107     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1108        "--" to be printed when multiple files were scanned, even when none of the
1109        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1110        consider it to be a bug, and have restored the previous behaviour.
1111    
1112     2. A couple of code tidies to get rid of compiler warnings.
1113    
1114     3. The pcretest program used to cheat by referring to symbols in the library
1115        whose names begin with _pcre_. These are internal symbols that are not
1116        really supposed to be visible externally, and in some environments it is
1117        possible to suppress them. The cheating is now confined to including
1118        certain files from the library's source, which is a bit cleaner.
1119    
1120     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1121        file's purpose clearer.
1122    
1123     5. Reorganized pcre_ucp_findchar().
1124    
1125    
1126    Version 6.3 15-Aug-05
1127    ---------------------
1128    
1129     1. The file libpcre.pc.in did not have general read permission in the tarball.
1130    
1131     2. There were some problems when building without C++ support:
1132    
1133        (a) If C++ support was not built, "make install" and "make test" still
1134            tried to test it.
1135    
1136        (b) There were problems when the value of CXX was explicitly set. Some
1137            changes have been made to try to fix these, and ...
1138    
1139        (c) --disable-cpp can now be used to explicitly disable C++ support.
1140    
1141        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1142            backslash in a target when C++ was disabled. This confuses some
1143            versions of "make", apparently. Using an intermediate variable solves
1144            this. (Same for CPP_LOBJ.)
1145    
1146     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1147        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1148        necessary on certain architectures.
1149    
1150     4. Added a setting of -export-symbols-regex to the link command to remove
1151        those symbols that are exported in the C sense, but actually are local
1152        within the library, and not documented. Their names all begin with
1153        "_pcre_". This is not a perfect job, because (a) we have to except some
1154        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1155        available (and never for static libraries). I have made a note to try to
1156        find a way round (a) in the future.
1157    
1158    
1159    Version 6.2 01-Aug-05
1160    ---------------------
1161    
1162     1. There was no test for integer overflow of quantifier values. A construction
1163        such as {1111111111111111} would give undefined results. What is worse, if
1164        a minimum quantifier for a parenthesized subpattern overflowed and became
1165        negative, the calculation of the memory size went wrong. This could have
1166        led to memory overwriting.
1167    
1168     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1169    
1170     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1171        operating environments where this matters.
1172    
1173     4. Applied Giuseppe Maxia's patch to add additional features for controlling
1174        PCRE options from within the C++ wrapper.
1175    
1176     5. Named capturing subpatterns were not being correctly counted when a pattern
1177        was compiled. This caused two problems: (a) If there were more than 100
1178        such subpatterns, the calculation of the memory needed for the whole
1179        compiled pattern went wrong, leading to an overflow error. (b) Numerical
1180        back references of the form \12, where the number was greater than 9, were
1181        not recognized as back references, even though there were sufficient
1182        previous subpatterns.
1183    
1184     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1185        versions of gcc, e.g. 2.95.4.
1186    
1187    
1188    Version 6.1 21-Jun-05
1189    ---------------------
1190    
1191     1. There was one reference to the variable "posix" in pcretest.c that was not
1192        surrounded by "#if !defined NOPOSIX".
1193    
1194     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1195        the cross-check on the old pcre_info() function, for the benefit of the
1196        cut-down version of PCRE that is currently imported into Exim.
1197    
1198     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1199        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1200        patterns that start (?i) but I don't think that matters. The use of (?i) is
1201        just an example; this all applies to the other options as well.
1202    
1203     4. Since libtool seems to echo the compile commands it is issuing, the output
1204        from "make" can be reduced a bit by putting "@" in front of each libtool
1205        compile command.
1206    
1207     5. Patch from the folks at Google for configure.in to be a bit more thorough
1208        in checking for a suitable C++ installation before trying to compile the
1209        C++ stuff. This should fix a reported problem when a compiler was present,
1210        but no suitable headers.
1211    
1212     6. The man pages all had just "PCRE" as their title. I have changed them to
1213        be the relevant file name. I have also arranged that these names are
1214        retained in the file doc/pcre.txt, which is a concatenation in text format
1215        of all the man pages except the little individual ones for each function.
1216    
1217     7. The NON-UNIX-USE file had not been updated for the different set of source
1218        files that come with release 6. I also added a few comments about the C++
1219        wrapper.
1220    
1221    
1222    Version 6.0 07-Jun-05
1223    ---------------------
1224    
1225     1. Some minor internal re-organization to help with my DFA experiments.
1226    
1227     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1228        didn't matter for the library itself when fully configured, but did matter
1229        when compiling without UCP support, or within Exim, where the ucp files are
1230        not imported.
1231    
1232     3. Refactoring of the library code to split up the various functions into
1233        different source modules. The addition of the new DFA matching code (see
1234        below) to a single monolithic source would have made it really too
1235        unwieldy, quite apart from causing all the code to be include in a
1236        statically linked application, when only some functions are used. This is
1237        relevant even without the DFA addition now that patterns can be compiled in
1238        one application and matched in another.
1239    
1240        The downside of splitting up is that there have to be some external
1241        functions and data tables that are used internally in different modules of
1242        the library but which are not part of the API. These have all had their
1243        names changed to start with "_pcre_" so that they are unlikely to clash
1244        with other external names.
1245    
1246     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1247        a different (DFA) algorithm. Although it is slower than the original
1248        function, it does have some advantages for certain types of matching
1249        problem.
1250    
1251     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1252        including restarting after a partial match.
1253    
1254     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1255        defined when compiling for Windows was sent to me. I have put it into the
1256        code, though I have no means of testing or verifying it.
1257    
1258     7. Added the pcre_refcount() auxiliary function.
1259    
1260     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1261        match before or at the first newline in the subject string. In pcretest,
1262        the /f option on a pattern can be used to set this.
1263    
1264     9. A repeated \w when used in UTF-8 mode with characters greater than 256
1265        would behave wrongly. This has been present in PCRE since release 4.0.
1266    
1267    10. A number of changes to the pcregrep command:
1268    
1269        (a) Refactored how -x works; insert ^(...)$ instead of setting
1270            PCRE_ANCHORED and checking the length, in preparation for adding
1271            something similar for -w.
1272    
1273        (b) Added the -w (match as a word) option.
1274    
1275        (c) Refactored the way lines are read and buffered so as to have more
1276            than one at a time available.
1277    
1278        (d) Implemented a pcregrep test script.
1279    
1280        (e) Added the -M (multiline match) option. This allows patterns to match
1281            over several lines of the subject. The buffering ensures that at least
1282            8K, or the rest of the document (whichever is the shorter) is available
1283            for matching (and similarly the previous 8K for lookbehind assertions).
1284    
1285        (f) Changed the --help output so that it now says
1286    
1287              -w, --word-regex(p)
1288    
1289            instead of two lines, one with "regex" and the other with "regexp"
1290            because that confused at least one person since the short forms are the
1291            same. (This required a bit of code, as the output is generated
1292            automatically from a table. It wasn't just a text change.)
1293    
1294        (g) -- can be used to terminate pcregrep options if the next thing isn't an
1295            option but starts with a hyphen. Could be a pattern or a path name
1296            starting with a hyphen, for instance.
1297    
1298        (h) "-" can be given as a file name to represent stdin.
1299    
1300        (i) When file names are being printed, "(standard input)" is used for
1301            the standard input, for compatibility with GNU grep. Previously
1302            "<stdin>" was used.
1303    
1304        (j) The option --label=xxx can be used to supply a name to be used for
1305            stdin when file names are being printed. There is no short form.
1306    
1307        (k) Re-factored the options decoding logic because we are going to add
1308            two more options that take data. Such options can now be given in four
1309            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1310    
1311        (l) Added the -A, -B, and -C options for requesting that lines of context
1312            around matches be printed.
1313    
1314        (m) Added the -L option to print the names of files that do not contain
1315            any matching lines, that is, the complement of -l.
1316    
1317        (n) The return code is 2 if any file cannot be opened, but pcregrep does
1318            continue to scan other files.
1319    
1320        (o) The -s option was incorrectly implemented. For compatibility with other
1321            greps, it now suppresses the error message for a non-existent or non-
1322            accessible file (but not the return code). There is a new option called
1323            -q that suppresses the output of matching lines, which was what -s was
1324            previously doing.
1325    
1326        (p) Added --include and --exclude options to specify files for inclusion
1327            and exclusion when recursing.
1328    
1329    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1330        Hopefully, it now does.
1331    
1332    12. Missing cast in pcre_study().
1333    
1334    13. Added an "uninstall" target to the makefile.
1335    
1336    14. Replaced "extern" in the function prototypes in Makefile.in with
1337        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1338        world, but is set differently for Windows.
1339    
1340    15. Added a second compiling function called pcre_compile2(). The only
1341        difference is that it has an extra argument, which is a pointer to an
1342        integer error code. When there is a compile-time failure, this is set
1343        non-zero, in addition to the error test pointer being set to point to an
1344        error message. The new argument may be NULL if no error number is required
1345        (but then you may as well call pcre_compile(), which is now just a
1346        wrapper). This facility is provided because some applications need a
1347        numeric error indication, but it has also enabled me to tidy up the way
1348        compile-time errors are handled in the POSIX wrapper.
1349    
1350    16. Added VPATH=.libs to the makefile; this should help when building with one
1351        prefix path and installing with another. (Or so I'm told by someone who
1352        knows more about this stuff than I do.)
1353    
1354    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1355        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1356        match everything, including newlines. This is not POSIX-compatible, but
1357        somebody wanted the feature. From pcretest it can be activated by using
1358        both the P and the s flags.
1359    
1360    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1361    
1362    19. libpcre.pc was being incorrectly installed as executable.
1363    
1364    20. A couple of places in pcretest check for end-of-line by looking for '\n';
1365        it now also looks for '\r' so that it will work unmodified on Windows.
1366    
1367    21. Added Google's contributed C++ wrapper to the distribution.
1368    
1369    22. Added some untidy missing memory free() calls in pcretest, to keep
1370        Electric Fence happy when testing.
1371    
1372    
1373    
1374    Version 5.0 13-Sep-04
1375    ---------------------
1376    
1377     1. Internal change: literal characters are no longer packed up into items
1378        containing multiple characters in a single byte-string. Each character
1379        is now matched using a separate opcode. However, there may be more than one
1380        byte in the character in UTF-8 mode.
1381    
1382     2. The pcre_callout_block structure has two new fields: pattern_position and
1383        next_item_length. These contain the offset in the pattern to the next match
1384        item, and its length, respectively.
1385    
1386     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1387        insertion of callouts before each pattern item. Added the /C option to
1388        pcretest to make use of this.
1389    
1390     4. On the advice of a Windows user, the lines
1391    
1392          #if defined(_WIN32) || defined(WIN32)
1393          _setmode( _fileno( stdout ), 0x8000 );
1394          #endif  /* defined(_WIN32) || defined(WIN32) */
1395    
1396        have been added to the source of pcretest. This apparently does useful
1397        magic in relation to line terminators.
1398    
1399     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1400        for the benefit of those environments where the "b" makes a difference.
1401    
1402     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1403        to know about it. I have put a hack into configure.in that adds in code
1404        to set GCC=yes if CC=icc. This seems to end up at a point in the
1405        generated configure script that is early enough to affect the setting of
1406        compiler options, which is what is needed, but I have no means of testing
1407        whether it really works. (The user who reported this had patched the
1408        generated configure script, which of course I cannot do.)
1409    
1410        LATER: After change 22 below (new libtool files), the configure script
1411        seems to know about icc (and also ecc). Therefore, I have commented out
1412        this hack in configure.in.
1413    
1414     7. Added support for pkg-config (2 patches were sent in).
1415    
1416     8. Negated POSIX character classes that used a combination of internal tables
1417        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1418        [[:^ascii]]. Typically, they would match almost any characters. The other
1419        POSIX classes were not broken in this way.
1420    
1421     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1422        to find the match, as PCRE was deluded into thinking that the match had to
1423        start at the start point or following a newline. The same bug applied to
1424        patterns with negative forward assertions or any backward assertions
1425        preceding ".*" at the start, unless the pattern required a fixed first
1426        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1427    
1428    10. In UTF-8 mode, when moving forwards in the subject after a failed match
1429        starting at the last subject character, bytes beyond the end of the subject
1430        string were read.
1431    
1432    11. Renamed the variable "class" as "classbits" to make life easier for C++
1433        users. (Previously there was a macro definition, but it apparently wasn't
1434        enough.)
1435    
1436    12. Added the new field "tables" to the extra data so that tables can be passed
1437        in at exec time, or the internal tables can be re-selected. This allows
1438        a compiled regex to be saved and re-used at a later time by a different
1439        program that might have everything at different addresses.
1440    
1441    13. Modified the pcre-config script so that, when run on Solaris, it shows a
1442        -R library as well as a -L library.
1443    
1444    14. The debugging options of pcretest (-d on the command line or D on a
1445        pattern) showed incorrect output for anything following an extended class
1446        that contained multibyte characters and which was followed by a quantifier.
1447    
1448    15. Added optional support for general category Unicode character properties
1449        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1450        support. It adds about 90K to the size of the library. The meanings of the
1451        inbuilt class escapes such as \d and \s have NOT been changed.
1452    
1453    16. Updated pcredemo.c to include calls to free() to release the memory for the
1454        compiled pattern.
1455    
1456    17. The generated file chartables.c was being created in the source directory
1457        instead of in the building directory. This caused the build to fail if the
1458        source directory was different from the building directory, and was
1459        read-only.
1460    
1461    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1462        file. No doubt somebody will tell me if they don't make sense... Also added
1463        Dan Mooney's comments about building on OpenVMS.
1464    
1465    19. Added support for partial matching via the PCRE_PARTIAL option for
1466        pcre_exec() and the \P data escape in pcretest.
1467    
1468    20. Extended pcretest with 3 new pattern features:
1469    
1470        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
1471              write the compiled pattern to the file whose name is "rest-of-line".
1472              This is a straight binary dump of the data, with the saved pointer to
1473              the character tables forced to be NULL. The study data, if any, is
1474              written too. After writing, pcretest reads a new pattern.
1475    
1476        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1477              compiled pattern from the given file. There must not be any
1478              occurrences of "<" in the file name (pretty unlikely); if there are,
1479              pcretest will instead treat the initial "<" as a pattern delimiter.
1480              After reading in the pattern, pcretest goes on to read data lines as
1481              usual.
1482    
1483        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1484              and 16-bit fields in a compiled pattern, to simulate a pattern that
1485              was compiled on a host of opposite endianness.
1486    
1487    21. The pcre-exec() function can now cope with patterns that were compiled on
1488        hosts of opposite endianness, with this restriction:
1489    
1490          As for any compiled expression that is saved and used later, the tables
1491          pointer field cannot be preserved; the extra_data field in the arguments
1492          to pcre_exec() should be used to pass in a tables address if a value
1493          other than the default internal tables were used at compile time.
1494    
1495    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1496        now diagnosed as an error. Previously, most of the time, a negative number
1497        would have been treated as zero, but if in addition "ovector" was passed as
1498        NULL, a crash could occur.
1499    
1500    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1501        new versions from the libtool 1.5 distribution (the last one is a copy of
1502        a file called libtool.m4). This seems to have fixed the need to patch
1503        "configure" to support Darwin 1.3 (which I used to do). However, I still
1504        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1505        workstation).
1506    
1507    24. Changed the PCRE licence to be the more standard "BSD" licence.
1508    
1509    
1510    Version 4.5 01-Dec-03
1511    ---------------------
1512    
1513     1. There has been some re-arrangement of the code for the match() function so
1514        that it can be compiled in a version that does not call itself recursively.
1515        Instead, it keeps those local variables that need separate instances for
1516        each "recursion" in a frame on the heap, and gets/frees frames whenever it
1517        needs to "recurse". Keeping track of where control must go is done by means
1518        of setjmp/longjmp. The whole thing is implemented by a set of macros that
1519        hide most of the details from the main code, and operates only if
1520        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1521        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1522        operating.
1523    
1524        To make it easier for callers to provide specially tailored get/free
1525        functions for this usage, two new functions, pcre_stack_malloc, and
1526        pcre_stack_free, are used. They are always called in strict stacking order,
1527        and the size of block requested is always the same.
1528    
1529        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1530        PCRE has been compiled to use the stack or the heap for recursion. The
1531        -C option of pcretest uses this to show which version is compiled.
1532    
1533        A new data escape \S, is added to pcretest; it causes the amounts of store
1534        obtained and freed by both kinds of malloc/free at match time to be added
1535        to the output.
1536    
1537     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1538        what's available on my current Linux desktop machine.
1539    
1540     3. When matching a UTF-8 string, the test for a valid string at the start has
1541        been extended. If start_offset is not zero, PCRE now checks that it points
1542        to a byte that is the start of a UTF-8 character. If not, it returns
1543        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1544        this is necessary because there may be backward assertions in the pattern.
1545        When matching the same subject several times, it may save resources to use
1546        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1547    
1548     4. The code for checking the validity of UTF-8 strings has been tightened so
1549        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1550        containing "overlong sequences".
1551    
1552     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1553        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1554        should have been "&", but it just so happened that all the cases this let
1555        through by mistake were picked up later in the function.
1556    
1557     6. I had used a variable called "isblank" - this is a C99 function, causing
1558        some compilers to warn. To avoid this, I renamed it (as "blankclass").
1559    
1560     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1561        prompting; (b) run "./pcretest /dev/null" at the start of the test script
1562        so the version is shown; (c) stop "make test" echoing "./RunTest".
1563    
1564     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1565    
1566     9. The prototype for memmove() for systems that don't have it was using
1567        size_t, but the inclusion of the header that defines size_t was later. I've
1568        moved the #includes for the C headers earlier to avoid this.
1569    
1570    10. Added some adjustments to the code to make it easier to compiler on certain
1571        special systems:
1572    
1573          (a) Some "const" qualifiers were missing.
1574          (b) Added the macro EXPORT before all exported functions; by default this
1575              is defined to be empty.
1576          (c) Changed the dftables auxiliary program (that builds chartables.c) so
1577              that it reads its output file name as an argument instead of writing
1578              to the standard output and assuming this can be redirected.
1579    
1580    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
1581        class containing characters with values greater than 255, PCRE compilation
1582        went into a loop.
1583    
1584    12. A recursive reference to a subpattern that was within another subpattern
1585        that had a minimum quantifier of zero caused PCRE to crash. For example,
1586        (x(y(?2))z)? provoked this bug with a subject that got as far as the
1587        recursion. If the recursively-called subpattern itself had a zero repeat,
1588        that was OK.
1589    
1590    13. In pcretest, the buffer for reading a data line was set at 30K, but the
1591        buffer into which it was copied (for escape processing) was still set at
1592        1024, so long lines caused crashes.
1593    
1594    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
1595        "internal error: code overflow...". This applied to any character class
1596        that was followed by a possessive quantifier.
1597    
1598    15. Modified the Makefile to add libpcre.la as a prerequisite for
1599        libpcreposix.la because I was told this is needed for a parallel build to
1600        work.
1601    
1602    16. If a pattern that contained .* following optional items at the start was
1603        studied, the wrong optimizing data was generated, leading to matching
1604        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
1605        matching string must start with a or b or c. The correct conclusion for
1606        this pattern is that a match can start with any character.
1607    
1608    
1609    Version 4.4 13-Aug-03
1610    ---------------------
1611    
1612     1. In UTF-8 mode, a character class containing characters with values between
1613        127 and 255 was not handled correctly if the compiled pattern was studied.
1614        In fixing this, I have also improved the studying algorithm for such
1615        classes (slightly).
1616    
1617     2. Three internal functions had redundant arguments passed to them. Removal
1618        might give a very teeny performance improvement.
1619    
1620     3. Documentation bug: the value of the capture_top field in a callout is *one
1621        more than* the number of the hightest numbered captured substring.
1622    
1623     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
1624        in incorrectly linking with a previously installed version. They now link
1625        explicitly with libpcre.la.
1626    
1627     5. configure.in no longer needs to recognize Cygwin specially.
1628    
1629     6. A problem in pcre.in for Windows platforms is fixed.
1630    
1631     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
1632        pcretest, it used to include the size of the study block as part of its
1633        output. Unfortunately, the structure contains a field that has a different
1634        size on different hardware architectures. This meant that the tests that
1635        showed this size failed. As the block is currently always of a fixed size,
1636        this information isn't actually particularly useful in pcretest output, so
1637        I have just removed it.
1638    
1639     8. Three pre-processor statements accidentally did not start in column 1.
1640        Sadly, there are *still* compilers around that complain, even though
1641        standard C has not required this for well over a decade. Sigh.
1642    
1643     9. In pcretest, the code for checking callouts passed small integers in the
1644        callout_data field, which is a void * field. However, some picky compilers
1645        complained about the casts involved for this on 64-bit systems. Now
1646        pcretest passes the address of the small integer instead, which should get
1647        rid of the warnings.
1648    
1649    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
1650        both compile and run time, and gives an error if an invalid UTF-8 sequence
1651        is found. There is a option for disabling this check in cases where the
1652        string is known to be correct and/or the maximum performance is wanted.
1653    
1654    11. In response to a bug report, I changed one line in Makefile.in from
1655    
1656            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
1657        to
1658            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
1659    
1660        to look similar to other lines, but I have no way of telling whether this
1661        is the right thing to do, as I do not use Windows. No doubt I'll get told
1662        if it's wrong...
1663    
1664    
1665    Version 4.3 21-May-03
1666    ---------------------
1667    
1668    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
1669       Makefile.
1670    
1671    2. Some refactoring to improve the quality of the code:
1672    
1673       (i)   The utf8_table... variables are now declared "const".
1674    
1675       (ii)  The code for \cx, which used the "case flipping" table to upper case
1676             lower case letters, now just substracts 32. This is ASCII-specific,
1677             but the whole concept of \cx is ASCII-specific, so it seems
1678             reasonable.
1679    
1680       (iii) PCRE was using its character types table to recognize decimal and
1681             hexadecimal digits in the pattern. This is silly, because it handles
1682             only 0-9, a-f, and A-F, but the character types table is locale-
1683             specific, which means strange things might happen. A private
1684             table is now used for this - though it costs 256 bytes, a table is
1685             much faster than multiple explicit tests. Of course, the standard
1686             character types table is still used for matching digits in subject
1687             strings against \d.
1688    
1689       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
1690             ending in _t are). So I've renamed it as ESC_tee.
1691    
1692    3. The first argument for regexec() in the POSIX wrapper should have been
1693       defined as "const".
1694    
1695    4. Changed pcretest to use malloc() for its buffers so that they can be
1696       Electric Fenced for debugging.
1697    
1698    5. There were several places in the code where, in UTF-8 mode, PCRE would try
1699       to read one or more bytes before the start of the subject string. Often this
1700       had no effect on PCRE's behaviour, but in some circumstances it could
1701       provoke a segmentation fault.
1702    
1703    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
1704       to try to read one or more bytes before the start of the subject string.
1705    
1706    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
1707       UTF-8 support could misbehave in various ways if the subject string
1708       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
1709       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
1710       back over UTF-8 characters.)
1711    
1712    
1713    Version 4.2 14-Apr-03
1714    ---------------------
1715    
1716    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
1717    
1718    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
1719         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
1720         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
1721         [WIN_PREFIX]: new variable, "cyg" for Cygwin
1722         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
1723           and BUILD_EXEEXT
1724         Note: automatic setting of the BUILD variables is not yet working
1725         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
1726           compile-time but not at link-time
1727         [LINK]: use for linking executables only
1728         make different versions for Windows and non-Windows
1729         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
1730           libraries
1731         [LINK_FOR_BUILD]: new variable
1732         [OBJEXT]: use throughout
1733         [EXEEXT]: use throughout
1734         <winshared>: new target
1735         <wininstall>: new target
1736         <dftables.o>: use native compiler
1737         <dftables>: use native linker
1738         <install>: handle Windows platform correctly
1739         <clean>: ditto
1740         <check>: ditto
1741         copy DLL to top builddir before testing
1742    
1743       As part of these changes, -no-undefined was removed again. This was reported
1744       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
1745       in any case.
1746    
1747    3. Some tidies to get rid of compiler warnings:
1748    
1749       . In the match_data structure, match_limit was an unsigned long int, whereas
1750         match_call_count was an int. I've made them both unsigned long ints.
1751    
1752       . In pcretest the fact that a const uschar * doesn't automatically cast to
1753         a void * provoked a warning.
1754    
1755       . Turning on some more compiler warnings threw up some "shadow" variables
1756         and a few more missing casts.
1757    
1758    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
1759       option, a class that contained a single character with a value between 128
1760       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
1761    
1762    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
1763       option, a class that contained several characters, but with at least one
1764       whose value was between 128 and 255 caused PCRE to crash.
1765    
1766    
1767    Version 4.1 12-Mar-03
1768    ---------------------
1769    
1770    1. Compiling with gcc -pedantic found a couple of places where casts were
1771    needed, and a string in dftables.c that was longer than standard compilers are
1772    required to support.
1773    
1774    2. Compiling with Sun's compiler found a few more places where the code could
1775    be tidied up in order to avoid warnings.
1776    
1777    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
1778    first of these names is deprecated in the latest Autoconf in favour of the name
1779    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
1780    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
1781    analogy I have changed it to CFLAGS_FOR_BUILD.
1782    
1783    4. Added -no-undefined to the linking command in the Makefile, because this is
1784    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
1785    linking step for the pcreposix library.
1786    
1787    5. PCRE was failing to diagnose the case of two named groups with the same
1788    name.
1789    
1790    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
1791    literal character that is needed in the subject for a match, and scans along to
1792    ensure that it is present before embarking on the full matching process. This
1793    saves time in cases of nested unlimited repeats that are never going to match.
1794    Problem: the scan can take a lot of time if the subject is very long (e.g.
1795    megabytes), thus penalizing straightforward matches. It is now done only if the
1796    amount of subject to be scanned is less than 1000 bytes.
1797    
1798    7. A lesser problem with the same optimization is that it was recording the
1799    first character of an anchored pattern as "needed", thus provoking a search
1800    right along the subject, even when the first match of the pattern was going to
1801    fail. The "needed" character is now not set for anchored patterns, unless it
1802    follows something in the pattern that is of non-fixed length. Thus, it still
1803    fulfils its original purpose of finding quick non-matches in cases of nested
1804    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
1805    
1806    
1807    Version 4.0 17-Feb-03
1808    ---------------------
1809    
1810    1. If a comment in an extended regex that started immediately after a meta-item
1811    extended to the end of string, PCRE compiled incorrect data. This could lead to
1812    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
1813    
1814    2. Moved to autoconf 2.53 and libtool 1.4.2.
1815    
1816    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
1817    the special perltest8 script is no longer needed - all the tests can be run
1818    from a single perltest script.
1819    
1820    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
1821    by \s. It has now been removed in PCRE. This means it isn't recognized as
1822    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
1823    class [:space:] *does* include VT, thereby creating a mess.
1824    
1825    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
1826    space and tab.
1827    
1828    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
1829    its new features into the main test script, reducing the number of scripts.
1830    
1831    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
1832    were backward compatible, and made the (?i) apply to the whole pattern, as if
1833    /i were given. Now it behaves more logically, and applies the option setting
1834    only to what follows. PCRE has been changed to follow suit. However, if it
1835    finds options settings right at the start of the pattern, it extracts them into
1836    the global options, as before. Thus, they show up in the info data.
1837    
1838    8. Added support for the \Q...\E escape sequence. Characters in between are
1839    treated as literals. This is slightly different from Perl in that $ and @ are
1840    also handled as literals inside the quotes. In Perl, they will cause variable
1841    interpolation. Note the following examples:
1842    
1843        Pattern            PCRE matches      Perl matches
1844    
1845        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
1846        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
1847        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
1848    
1849    For compatibility with Perl, \Q...\E sequences are recognized inside character
1850    classes as well as outside them.
1851    
1852    9. Re-organized 3 code statements in pcretest to avoid "overflow in
1853    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
1854    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
1855    signed/unsigned warnings.
1856    
1857    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
1858    option for pcretest, so I've replaced it by a simple function that does just
1859    that job.
1860    
1861    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
1862    "pcregrep -".
1863    
1864    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
1865    Java package. This provides some syntactic sugar for simple cases of what my
1866    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
1867    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
1868    item, you can use this simplified notation. Note that only makes sense with
1869    greedy quantifiers. Consequently, the use of the possessive quantifier forces
1870    greediness, whatever the setting of the PCRE_UNGREEDY option.
1871    
1872    13. A change of greediness default within a pattern was not taking effect at
1873    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
1874    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
1875    was abstracted outside.
1876    
1877    14. PCRE now supports the \G assertion. It is true when the current matching
1878    position is at the start point of the match. This differs from \A when the
1879    starting offset is non-zero. Used with the /g option of pcretest (or similar
1880    code), it works in the same way as it does for Perl's /g option. If all
1881    alternatives of a regex begin with \G, the expression is anchored to the start
1882    match position, and the "anchored" flag is set in the compiled expression.
1883    
1884    15. Some bugs concerning the handling of certain option changes within patterns
1885    have been fixed. These applied to options other than (?ims). For example,
1886    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
1887    been the other way round. Some of this was related to change 7 above.
1888    
1889    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
1890    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
1891    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
1892    POSIX classes only within a class (e.g. /[[:alpha:]]/).
1893    
1894    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
1895    mode. Unlike ".", it always matches newline, whatever the setting of
1896    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
1897    assertions. Perl allows it, but it doesn't (in general) work because it can't
1898    calculate the length of the lookbehind. At least, that's the case for Perl
1899    5.8.0 - I've been told they are going to document that it doesn't work in
1900    future.
1901    
1902    18. Added an error diagnosis for escapes that PCRE does not support: these are
1903    \L, \l, \N, \P, \p, \U, \u, and \X.
1904    
1905    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
1906    reading past the end of the pattern in cases such as /[abcd/.
1907    
1908    20. PCRE was getting more memory than necessary for patterns with classes that
1909    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
1910    
1911    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
1912    compiling PCRE for use with Virtual Pascal.
1913    
1914    22. Small fix to the Makefile to make it work properly if the build is done
1915    outside the source tree.
1916    
1917    23. Added a new extension: a condition to go with recursion. If a conditional
1918    subpattern starts with (?(R) the "true" branch is used if recursion has
1919    happened, whereas the "false" branch is used only at the top level.
1920    
1921    24. When there was a very long string of literal characters (over 255 bytes
1922    without UTF support, over 250 bytes with UTF support), the computation of how
1923    much memory was required could be incorrect, leading to segfaults or other
1924    strange effects.
1925    
1926    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
1927    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
1928    there was a subsequent back reference to those brackets. This meant that, for
1929    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
1930    possible to check for precisely this case. All we can do is abandon the
1931    optimization if .* occurs inside capturing brackets when there are any back
1932    references whatsoever. (See below for a better fix that came later.)
1933    
1934    26. The handling of the optimization for finding the first character of a
1935    non-anchored pattern, and for finding a character that is required later in the
1936    match were failing in some cases. This didn't break the matching; it just
1937    failed to optimize when it could. The way this is done has been re-implemented.
1938    
1939    27. Fixed typo in error message for invalid (?R item (it said "(?p").
1940    
1941    28. Added a new feature that provides some of the functionality that Perl
1942    provides with (?{...}). The facility is termed a "callout". The way it is done
1943    in PCRE is for the caller to provide an optional function, by setting
1944    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
1945    global variable. By default it is unset, which disables all calling out. To get
1946    the function called, the regex must include (?C) at appropriate points. This
1947    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
1948    This provides a means of identifying different callout points. When PCRE
1949    reaches such a point in the regex, if pcre_callout has been set, the external
1950    function is called. It is provided with data in a structure called
1951    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
1952    matching continues; if it returns a non-zero value, the match at the current
1953    point fails. However, backtracking will occur if possible. [This was changed
1954    later and other features added - see item 49 below.]
1955    
1956    29. pcretest is upgraded to test the callout functionality. It provides a
1957    callout function that displays information. By default, it shows the start of
1958    the match and the current position in the text. There are some new data escapes
1959    to vary what happens:
1960    
1961        \C+         in addition, show current contents of captured substrings
1962        \C-         do not supply a callout function
1963        \C!n        return 1 when callout number n is reached
1964        \C!n!m      return 1 when callout number n is reached for the mth time
1965    
1966    30. If pcregrep was called with the -l option and just a single file name, it
1967    output "<stdin>" if a match was found, instead of the file name.
1968    
1969    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
1970    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
1971    pcre_exec(). This saves a malloc/free per call. The default value of
1972    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
1973    when configuring.
1974    
1975    32. The default maximum size of a compiled pattern is 64K. There have been a
1976    few cases of people hitting this limit. The code now uses macros to handle the
1977    storing of links as offsets within the compiled pattern. It defaults to 2-byte
1978    links, but this can be changed to 3 or 4 bytes by --with-link-size when
1979    configuring. Tests 2 and 5 work only with 2-byte links because they output
1980    debugging information about compiled patterns.
1981    
1982    33. Internal code re-arrangements:
1983    
1984    (a) Moved the debugging function for printing out a compiled regex into
1985        its own source file (printint.c) and used #include to pull it into
1986        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
1987        separate copies.
1988    
1989    (b) Defined the list of op-code names for debugging as a macro in
1990        internal.h so that it is next to the definition of the opcodes.
1991    
1992    (c) Defined a table of op-code lengths for simpler skipping along compiled
1993        code. This is again a macro in internal.h so that it is next to the
1994        definition of the opcodes.
1995    
1996    34. Added support for recursive calls to individual subpatterns, along the
1997    lines of Robin Houston's patch (but implemented somewhat differently).
1998    
1999    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
2000    allow it to read and process whole directories in Win32. This code was
2001    contributed by Lionel Fourquaux; it has not been tested by me.
2002    
2003    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
2004    used to name a group. Names consist of alphanumerics and underscores, and must
2005    be unique. Back references use the syntax (?P=name) and recursive calls use
2006    (?P>name) which is a PCRE extension to the Python extension. Groups still have
2007    numbers. The function pcre_fullinfo() can be used after compilation to extract
2008    a name/number map. There are three relevant calls:
2009    
2010      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
2011      PCRE_INFO_NAMECOUNT            yields the number of entries
2012      PCRE_INFO_NAMETABLE            yields a pointer to the map.
2013    
2014    The map is a vector of fixed-size entries. The size of each entry depends on
2015    the length of the longest name used. The first two bytes of each entry are the
2016    group number, most significant byte first. There follows the corresponding
2017    name, zero terminated. The names are in alphabetical order.
2018    
2019    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
2020    case instead of 255. Making it the same both with and without UTF-8 support
2021    means that the same test output works with both.
2022    
2023    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
2024    calling malloc() with a zero argument.
2025    
2026    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
2027    optimization. I've improved things by keeping a bitmap of backreferences with
2028    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
2029    fact referenced, the optimization can be applied. It is unlikely that a
2030    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
2031    the match to follow \n) will appear inside brackets with a number greater than
2032    31, but if it does, any back reference > 31 suppresses the optimization.
2033    
2034    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
2035    of disabling numbered capturing parentheses. Any opening parenthesis that is
2036    not followed by ? behaves as if it were followed by ?: but named parentheses
2037    can still be used for capturing (and they will acquire numbers in the usual
2038    way).
2039    
2040    41. Redesigned the return codes from the match() function into yes/no/error so
2041    that errors can be passed back from deep inside the nested calls. A malloc
2042    failure while inside a recursive subpattern call now causes the
2043    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
2044    
2045    42. It is now possible to set a limit on the number of times the match()
2046    function is called in a call to pcre_exec(). This facility makes it possible to
2047    limit the amount of recursion and backtracking, though not in a directly
2048    obvious way, because the match() function is used in a number of different
2049    circumstances. The count starts from zero for each position in the subject
2050    string (for non-anchored patterns). The default limit is, for compatibility, a
2051    large number, namely 10 000 000. You can change this in two ways:
2052    
2053    (a) When configuring PCRE before making, you can use --with-match-limit=n
2054        to set a default value for the compiled library.
2055    
2056    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
2057        a different value is set. See 45 below.
2058    
2059    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
2060    
2061    43. Added a new function pcre_config(int, void *) to enable run-time extraction
2062    of things that can be changed at compile time. The first argument specifies
2063    what is wanted and the second points to where the information is to be placed.
2064    The current list of available information is:
2065    
2066      PCRE_CONFIG_UTF8
2067    
2068    The output is an integer that is set to one if UTF-8 support is available;
2069    otherwise it is set to zero.
2070    
2071      PCRE_CONFIG_NEWLINE
2072    
2073    The output is an integer that it set to the value of the code that is used for
2074    newline. It is either LF (10) or CR (13).
2075    
2076      PCRE_CONFIG_LINK_SIZE
2077    
2078    The output is an integer that contains the number of bytes used for internal
2079    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
2080    
2081      PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
2082    
2083    The output is an integer that contains the threshold above which the POSIX
2084    interface uses malloc() for output vectors. See item 31 above.
2085    
2086      PCRE_CONFIG_MATCH_LIMIT
2087    
2088    The output is an unsigned integer that contains the default limit of the number
2089    of match() calls in a pcre_exec() execution. See 42 above.
2090    
2091    44. pcretest has been upgraded by the addition of the -C option. This causes it
2092    to extract all the available output from the new pcre_config() function, and to
2093    output it. The program then exits immediately.
2094    
2095    45. A need has arisen to pass over additional data with calls to pcre_exec() in
2096    order to support additional features. One way would have been to define
2097    pcre_exec2() (for example) with extra arguments, but this would not have been
2098    extensible, and would also have required all calls to the original function to
2099    be mapped to the new one. Instead, I have chosen to extend the mechanism that
2100    is used for passing in "extra" data from pcre_study().
2101    
2102    The pcre_extra structure is now exposed and defined in pcre.h. It currently
2103    contains the following fields:
2104    
2105      flags         a bitmap indicating which of the following fields are set
2106      study_data    opaque data from pcre_study()
2107      match_limit   a way of specifying a limit on match() calls for a specific
2108                      call to pcre_exec()
2109      callout_data  data for callouts (see 49 below)
2110    
2111    The flag bits are also defined in pcre.h, and are
2112    
2113      PCRE_EXTRA_STUDY_DATA
2114      PCRE_EXTRA_MATCH_LIMIT
2115      PCRE_EXTRA_CALLOUT_DATA
2116    
2117    The pcre_study() function now returns one of these new pcre_extra blocks, with
2118    the actual study data pointed to by the study_data field, and the
2119    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
2120    before. That is, this change is entirely upwards-compatible and requires no
2121    change to existing code.
2122    
2123    If you want to pass in additional data to pcre_exec(), you can either place it
2124    in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
2125    block.
2126    
2127    46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
2128    data string contains the escape sequence \M, pcretest calls pcre_exec() several
2129    times with different match limits, until it finds the minimum value needed for
2130    pcre_exec() to complete. The value is then output. This can be instructive; for
2131    most simple matches the number is quite small, but for pathological cases it
2132    gets very large very quickly.
2133    
2134    47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
2135    returns the size of the data block pointed to by the study_data field in a
2136    pcre_extra block, that is, the value that was passed as the argument to
2137    pcre_malloc() when PCRE was getting memory in which to place the information
2138    created by pcre_study(). The fourth argument should point to a size_t variable.
2139    pcretest has been extended so that this information is shown after a successful
2140    pcre_study() call when information about the compiled regex is being displayed.
2141    
2142    48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
2143    because what follows is always an absolute path. (Later: it turns out that this
2144    is more than cosmetic for MinGW, because it doesn't like empty path
2145    components.)
2146    
2147    49. Some changes have been made to the callout feature (see 28 above):
2148    
2149    (i)  A callout function now has three choices for what it returns:
2150    
2151           0  =>  success, carry on matching
2152         > 0  =>  failure at this point, but backtrack if possible
2153         < 0  =>  serious error, return this value from pcre_exec()
2154    
2155         Negative values should normally be chosen from the set of PCRE_ERROR_xxx
2156         values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
2157         "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
2158         use by callout functions. It will never be used by PCRE itself.
2159    
2160    (ii) The pcre_extra structure (see 45 above) has a void * field called
2161         callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
2162         pcre_callout_block structure has a field of the same name. The contents of
2163         the field passed in the pcre_extra structure are passed to the callout
2164         function in the corresponding field in the callout block. This makes it
2165         easier to use the same callout-containing regex from multiple threads. For
2166         testing, the pcretest program has a new data escape
2167    
2168           \C*n        pass the number n (may be negative) as callout_data
2169    
2170         If the callout function in pcretest receives a non-zero value as
2171         callout_data, it returns that value.
2172    
2173    50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
2174    there were some redundant $(CFLAGS) in commands that are now specified as
2175    $(LINK), which already includes $(CFLAGS).
2176    
2177    51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
2178    has been compiled with UTF-8 support *and* pcre_compile() has been compiled
2179    with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
2180    one-byte characters throughout. Note that case-insensitive matching applies
2181    only to characters whose values are less than 256. PCRE doesn't support the
2182    notion of cases for higher-valued characters.
2183    
2184    (i)   A character class whose characters are all within 0-255 is handled as
2185          a bit map, and the map is inverted for negative classes. Previously, a
2186          character > 255 always failed to match such a class; however it should
2187          match if the class was a negative one (e.g. [^ab]). This has been fixed.
2188    
2189    (ii)  A negated character class with a single character < 255 is coded as
2190          "not this character" (OP_NOT). This wasn't working properly when the test
2191          character was multibyte, either singly or repeated.
2192    
2193    (iii) Repeats of multibyte characters are now handled correctly in UTF-8
2194          mode, for example: \x{100}{2,3}.
2195    
2196    (iv)  The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
2197          singly or repeated) now correctly test multibyte characters. However,
2198          PCRE doesn't recognize any characters with values greater than 255 as
2199          digits, spaces, or word characters. Such characters always match \D, \S,
2200          and \W, and never match \d, \s, or \w.
2201    
2202    (v)   Classes may now contain characters and character ranges with values
2203          greater than 255. For example: [ab\x{100}-\x{400}].
2204    
2205    (vi)  pcregrep now has a --utf-8 option (synonym -u) which makes it call
2206          PCRE in UTF-8 mode.
2207    
2208    52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
2209    PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
2210    retained for backwards compatibility. (Note that LASTLITERAL is also a byte
2211    value.)
2212    
2213    53. The single man page has become too large. I have therefore split it up into
2214    a number of separate man pages. These also give rise to individual HTML pages;
2215    these are now put in a separate directory, and there is an index.html page that
2216    lists them all. Some hyperlinking between the pages has been installed.
2217    
2218    54. Added convenience functions for handling named capturing parentheses.
2219    
2220    55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2221    aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2222    true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2223    are faulted.
2224    
2225    56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2226    calling configure. These values are used when compiling the dftables.c program
2227    which is run to generate the source of the default character tables. They
2228    default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2229    you will need to set these values.
2230    
2231    57. Updated the building process for Windows DLL, as provided by Fred Cox.
2232    
2233    
2234    Version 3.9 02-Jan-02
2235    ---------------------
2236    
2237    1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2238    
2239    2. If --disable-static was given, the building process failed when trying to
2240    build pcretest and pcregrep. (For some reason it was using libtool to compile
2241    them, which is not right, as they aren't part of the library.)
2242    
2243    
2244    Version 3.8 18-Dec-01
2245    ---------------------
2246    
2247    1. The experimental UTF-8 code was completely screwed up. It was packing the
2248    bytes in the wrong order. How dumb can you get?
2249    
2250    
2251    Version 3.7 29-Oct-01
2252    ---------------------
2253    
2254    1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2255    This caused pcretest, when used on the test data, to segfault. Unfortunately,
2256    this didn't happen under Solaris 8, where I normally test things.
2257    
2258    2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2259    doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2260    isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2261    this fix an hour or so after the initial 3.7 release.)
2262    
2263    
2264    Version 3.6 23-Oct-01
2265    ---------------------
2266    
2267    1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2268    offsets passed as NULL with zero offset count.
2269    
2270    2. The config.guess and config.sub files had not been updated when I moved to
2271    the latest autoconf.
2272    
2273    
2274    Version 3.5 15-Aug-01
2275    ---------------------
2276    
2277    1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2278    had been forgotten.
2279    
2280    2. By using declared but undefined structures, we can avoid using "void"
2281    definitions in pcre.h while keeping the internal definitions of the structures
2282    private.
2283    
2284    3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2285    user point of view, this means that both static and shared libraries are built
2286    by default, but this can be individually controlled. More of the work of
2287    handling this static/shared cases is now inside libtool instead of PCRE's make
2288    file.
2289    
2290    4. The pcretest utility is now installed along with pcregrep because it is
2291    useful for users (to test regexs) and by doing this, it automatically gets
2292    relinked by libtool. The documentation has been turned into a man page, so
2293    there are now .1, .txt, and .html versions in /doc.
2294    
2295    5. Upgrades to pcregrep:
2296       (i)   Added long-form option names like gnu grep.
2297       (ii)  Added --help to list all options with an explanatory phrase.
2298       (iii) Added -r, --recursive to recurse into sub-directories.
2299       (iv)  Added -f, --file to read patterns from a file.
2300    
2301    6. pcre_exec() was referring to its "code" argument before testing that
2302    argument for NULL (and giving an error if it was NULL).
2303    
2304    7. Upgraded Makefile.in to allow for compiling in a different directory from
2305    the source directory.
2306    
2307    8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2308    options bits, the pointer it was passed was to an int instead of to an unsigned
2309    long int. This mattered only on 64-bit systems.
2310    
2311    9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2312    generated) instead of pcre.in, which it its source. Also made the same change
2313    in several of the .c files.
2314    
2315    10. A new release of gcc defines printf() as a macro, which broke pcretest
2316    because it had an ifdef in the middle of a string argument for printf(). Fixed
2317    by using separate calls to printf().
2318    
2319    11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2320    script, to force use of CR or LF instead of \n in the source. On non-Unix
2321    systems, the value can be set in config.h.
2322    
2323    12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2324    absolute limit. Changed the text of the error message to make this clear, and
2325    likewise updated the man page.
2326    
2327    13. The limit of 99 on the number of capturing subpatterns has been removed.
2328    The new limit is 65535, which I hope will not be a "real" limit.
2329    
2330    
2331    Version 3.4 22-Aug-00
2332    ---------------------
2333    
2334    1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2335    
2336    2. Diagnose condition (?(0) as an error instead of crashing on matching.
2337    
2338    
2339    Version 3.3 01-Aug-00
2340    ---------------------
2341    
2342    1. If an octal character was given, but the value was greater than \377, it
2343    was not getting masked to the least significant bits, as documented. This could
2344    lead to crashes in some systems.
2345    
2346    2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
2347    the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
2348    
2349    3. Added the functions pcre_free_substring() and pcre_free_substring_list().
2350    These just pass their arguments on to (pcre_free)(), but they are provided
2351    because some uses of PCRE bind it to non-C systems that can call its functions,
2352    but cannot call free() or pcre_free() directly.
2353    
2354    4. Add "make test" as a synonym for "make check". Corrected some comments in
2355    the Makefile.
2356    
2357    5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
2358    Makefile.
2359    
2360    6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
2361    command called pgrep for grepping around the active processes.
2362    
2363    7. Added the beginnings of support for UTF-8 character strings.
2364    
2365    8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
2366    RANLIB to ./ltconfig so that they are used by libtool. I think these are all
2367    the relevant ones. (AR is not passed because ./ltconfig does its own figuring
2368    out for the ar command.)
2369    
2370    
2371    Version 3.2 12-May-00
2372    ---------------------
2373    
2374    This is purely a bug fixing release.
2375    
2376    1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
2377    of ZA. This was just one example of several cases that could provoke this bug,
2378    which was introduced by change 9 of version 2.00. The code for breaking
2379    infinite loops after an iteration that matches an empty string was't working
2380    correctly.
2381    
2382    2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
2383    when matched against abbab (for example). After matching an empty string, it
2384    wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
2385    caused it to match further down the string than it should.
2386    
2387    3. The code contained an inclusion of sys/types.h. It isn't clear why this
2388    was there because it doesn't seem to be needed, and it causes trouble on some
2389    systems, as it is not a Standard C header. It has been removed.
2390    
2391    4. Made 4 silly changes to the source to avoid stupid compiler warnings that
2392    were reported on the Macintosh. The changes were from
2393    
2394      while ((c = *(++ptr)) != 0 && c != '\n');
2395    to
2396      while ((c = *(++ptr)) != 0 && c != '\n') ;
2397    
2398    Totally extraordinary, but if that's what it takes...
2399    
2400    5. PCRE is being used in one environment where neither memmove() nor bcopy() is
2401    available. Added HAVE_BCOPY and an autoconf test for it; if neither
2402    HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
2403    assumes the way PCRE uses memmove() (always moving upwards).
2404    
2405    6. PCRE is being used in one environment where strchr() is not available. There
2406    was only one use in pcre.c, and writing it out to avoid strchr() probably gives
2407    faster code anyway.
2408    
2409    
2410  Version 3.1 09-Feb-00  Version 3.1 09-Feb-00
2411  ---------------------  ---------------------

Legend:
Removed from v.45  
changed lines
  Added in v.239

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12