/[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 279 by ph10, Tue Dec 4 20:01:43 2007 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4    Version 7.5 12-Nov-07
5    ---------------------
6    
7    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
8        values in parens when parsing an RE using the C++ wrapper."
9    
10    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
11        Characters greater than 255 were excluded from the class instead of being
12        included.
13    
14    3.  The same bug as (2) above applied to negated POSIX classes such as
15        [:^space:].
16    
17    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
18        defined or documented. It seems to have been a typo for PCRE_STATIC, so
19        I have changed it.
20    
21    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
22        first named subpattern) and a construct such as (?&a) would reference the
23        first named subpattern whose name started with "a" (in other words, the
24        length check was missing). Both these problems are fixed. "Subpattern name
25        expected" is now given for (?&) (a zero-length name), and this patch also
26        makes it give the same error for \k'' (previously it complained that that
27        was a reference to a non-existent subpattern).
28    
29    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
30        this is right because (?- can be followed by option settings as well as by
31        digits. I have, however, made the messages clearer.
32    
33    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
34        than the number used in the conditional) now cause a compile-time error.
35        This is actually not compatible with Perl, which accepts such patterns, but
36        treats the conditional as always being FALSE (as PCRE used to), but it
37        seems to me that giving a diagnostic is better.
38    
39    8.  Change "alphameric" to the more common word "alphanumeric" in comments
40        and messages.
41    
42    9.  Fix two occurrences of "backslash" in comments that should have been
43        "backspace".
44    
45    10. Remove two redundant lines of code that can never be obeyed (their function
46        was moved elsewhere).
47    
48    11. The program that makes PCRE's Unicode character property table had a bug
49        which caused it to generate incorrect table entries for sequences of
50        characters that have the same character type, but are in different scripts.
51        It amalgamated them into a single range, with the script of the first of
52        them. In other words, some characters were in the wrong script. There were
53        thirteen such cases, affecting characters in the following ranges:
54    
55          U+002b0 - U+002c1
56          U+0060c - U+0060d
57          U+0061e - U+00612
58          U+0064b - U+0065e
59          U+0074d - U+0076d
60          U+01800 - U+01805
61          U+01d00 - U+01d77
62          U+01d9b - U+01dbf
63          U+0200b - U+0200f
64          U+030fc - U+030fe
65          U+03260 - U+0327f
66          U+0fb46 - U+0fbb1
67          U+10450 - U+1049d
68    
69    12. The -o option (show only the matching part of a line) for pcregrep was not
70        compatible with GNU grep in that, if there was more than one match in a
71        line, it showed only the first of them. It now behaves in the same way as
72        GNU grep.
73    
74    13. If the -o and -v options were combined for pcregrep, it printed a blank
75        line for every non-matching line. GNU grep prints nothing, and pcregrep now
76        does the same. The return code can be used to tell if there were any
77        non-matching lines.
78    
79    
80    Version 7.4 21-Sep-07
81    ---------------------
82    
83    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
84        means that a class such as [\s] counted as "explicit reference to CR or
85        LF". That isn't really right - the whole point of the change was to try to
86        help when there was an actual mention of one of the two characters. So now
87        the change happens only if \r or \n (or a literal CR or LF) character is
88        encountered.
89    
90    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
91        of both had grown to the point where there were only 3 bits left.
92        Fortunately, there was spare space in the data structure, and so I have
93        moved the internal flags into a new 16-bit field to free up more option
94        bits.
95    
96    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
97        but did not set the internal JCHANGED flag - either of these is enough to
98        control the way the "get" function works - but the PCRE_INFO_JCHANGED
99        facility is supposed to tell if (?J) was ever used, so now (?J) at the
100        start sets both bits.
101    
102    4.  Added options (at build time, compile time, exec time) to change \R from
103        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
104    
105    5.  doc/pcresyntax.html was missing from the distribution.
106    
107    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
108        compatibility, even though it is no longer used.
109    
110    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
111        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
112        windows.h file is present (where different names are used). [This was
113        reversed later after testing - see 16 below.]
114    
115    8.  Changed all #include <config.h> to #include "config.h". There were also
116        some further <pcre.h> cases that I changed to "pcre.h".
117    
118    9.  When pcregrep was used with the --colour option, it missed the line ending
119        sequence off the lines that it output.
120    
121    10. It was pointed out to me that arrays of string pointers cause lots of
122        relocations when a shared library is dynamically loaded. A technique of
123        using a single long string with a table of offsets can drastically reduce
124        these. I have refactored PCRE in four places to do this. The result is
125        dramatic:
126    
127          Originally:                          290
128          After changing UCP table:            187
129          After changing error message table:   43
130          After changing table of "verbs"       36
131          After changing table of Posix names   22
132    
133        Thanks to the folks working on Gregex for glib for this insight.
134    
135    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
136        unicode-properties was also set.
137    
138    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
139    
140    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
141        checked only for CRLF.
142    
143    14. Added casts to pcretest.c to avoid compiler warnings.
144    
145    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
146    
147    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
148        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
149        entirely. This removes changes made in 7 above.
150    
151    17. The CMake files have been updated, and there is now more information about
152        building with CMake in the NON-UNIX-USE document.
153    
154    
155    Version 7.3 28-Aug-07
156    ---------------------
157    
158     1. In the rejigging of the build system that eventually resulted in 7.1, the
159        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
160        brackets there is not right, since it causes compilers to look for an
161        installed pcre.h, not the version that is in the source that is being
162        compiled (which of course may be different). I have changed it back to:
163    
164          #include "pcre.h"
165    
166        I have a vague recollection that the change was concerned with compiling in
167        different directories, but in the new build system, that is taken care of
168        by the VPATH setting the Makefile.
169    
170     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
171        when the subject happened to end in the byte 0x85 (e.g. if the last
172        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
173        characters but of course it shouldn't be taken as a newline when it is part
174        of another character. The bug was that, for an unlimited repeat of . in
175        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
176        characters when looking for a newline.
177    
178     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
179    
180     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
181        in debug output.
182    
183     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
184        long printing in the pcrecpp unittest when running under MinGW.
185    
186     6. ESC_K was left out of the EBCDIC table.
187    
188     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
189        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
190        limit also applies to "virtual nesting" when a pattern is recursive, and in
191        this case 1000 isn't so big. I have been able to remove this limit at the
192        expense of backing off one optimization in certain circumstances. Normally,
193        when pcre_exec() would call its internal match() function recursively and
194        immediately return the result unconditionally, it uses a "tail recursion"
195        feature to save stack. However, when a subpattern that can match an empty
196        string has an unlimited repetition quantifier, it no longer makes this
197        optimization. That gives it a stack frame in which to save the data for
198        checking that an empty string has been matched. Previously this was taken
199        from the 1000-entry workspace that had been reserved. So now there is no
200        explicit limit, but more stack is used.
201    
202     8. Applied Daniel's patches to solve problems with the import/export magic
203        syntax that is required for Windows, and which was going wrong for the
204        pcreposix and pcrecpp parts of the library. These were overlooked when this
205        problem was solved for the main library.
206    
207     9. There were some crude static tests to avoid integer overflow when computing
208        the size of patterns that contain repeated groups with explicit upper
209        limits. As the maximum quantifier is 65535, the maximum group length was
210        set at 30,000 so that the product of these two numbers did not overflow a
211        32-bit integer. However, it turns out that people want to use groups that
212        are longer than 30,000 bytes (though not repeat them that many times).
213        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
214        made it possible to implement the integer overflow checks in a much more
215        dynamic way, which I have now done. The artificial limitation on group
216        length has been removed - we now have only the limit on the total length of
217        the compiled pattern, which depends on the LINK_SIZE setting.
218    
219    10. Fixed a bug in the documentation for get/copy named substring when
220        duplicate names are permitted. If none of the named substrings are set, the
221        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
222        empty string.
223    
224    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
225        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
226        because the ] is interpreted as the first data character and the
227        terminating ] is not found. PCRE has been made compatible with Perl in this
228        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
229        cause memory overwriting.
230    
231    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
232        string has been matched (to stop an infinite loop). It was not recognizing
233        a conditional subpattern that could match an empty string if that
234        subpattern was within another subpattern. For example, it looped when
235        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
236        condition was not nested. This bug has been fixed.
237    
238    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
239        past the start of the subject in the presence of bytes with the top bit
240        set, for example "\x8aBCD".
241    
242    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
243        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
244    
245    14. Optimized (?!) to (*FAIL).
246    
247    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
248        This restricts code points to be within the range 0 to 0x10FFFF, excluding
249        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
250        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
251        does: it's just the validity check that is more restrictive.
252    
253    16. Inserted checks for integer overflows during escape sequence (backslash)
254        processing, and also fixed erroneous offset values for syntax errors during
255        backslash processing.
256    
257    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
258        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
259    
260    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
261        caused an overrun.
262    
263    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
264        something other than just ASCII characters) inside a group that had an
265        unlimited repeat caused a loop at compile time (while checking to see
266        whether the group could match an empty string).
267    
268    20. Debugging a pattern containing \p or \P could cause a crash. For example,
269        [\P{Any}] did so. (Error in the code for printing property names.)
270    
271    21. An orphan \E inside a character class could cause a crash.
272    
273    22. A repeated capturing bracket such as (A)? could cause a wild memory
274        reference during compilation.
275    
276    23. There are several functions in pcre_compile() that scan along a compiled
277        expression for various reasons (e.g. to see if it's fixed length for look
278        behind). There were bugs in these functions when a repeated \p or \P was
279        present in the pattern. These operators have additional parameters compared
280        with \d, etc, and these were not being taken into account when moving along
281        the compiled data. Specifically:
282    
283        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
284            length.
285    
286        (b) An item such as \pL+ within a repeated group could cause crashes or
287            loops.
288    
289        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
290            "reference to non-existent subpattern" error.
291    
292        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
293    
294    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
295        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
296    
297    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
298    
299    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
300        character were causing crashes (broken optimization).
301    
302    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
303        \p or \P) caused a compile-time loop.
304    
305    28. More problems have arisen in unanchored patterns when CRLF is a valid line
306        break. For example, the unstudied pattern [\r\n]A does not match the string
307        "\r\nA" because change 7.0/46 below moves the current point on by two
308        characters after failing to match at the start. However, the pattern \nA
309        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
310        the same is true. There doesn't seem any very clean way out of this, but
311        what I have chosen to do makes the common cases work: PCRE now takes note
312        of whether there can be an explicit match for \r or \n anywhere in the
313        pattern, and if so, 7.0/46 no longer applies. As part of this change,
314        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
315        pattern has explicit CR or LF references.
316    
317    29. Added (*CR) etc for changing newline setting at start of pattern.
318    
319    
320    Version 7.2 19-Jun-07
321    ---------------------
322    
323     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
324        which is apparently normally available under Windows.
325    
326     2. Re-jig the pcregrep tests with different newline settings in an attempt
327        to make them independent of the local environment's newline setting.
328    
329     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
330    
331     4. Some of the "internals" tests were previously cut out when the link size
332        was not 2, because the output contained actual offsets. The recent new
333        "Z" feature of pcretest means that these can be cut out, making the tests
334        usable with all link sizes.
335    
336     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
337        stack recursion. This gives a massive performance boost under BSD, but just
338        a small improvement under Linux. However, it saves one field in the frame
339        in all cases.
340    
341     6. Added more features from the forthcoming Perl 5.10:
342    
343        (a) (?-n) (where n is a string of digits) is a relative subroutine or
344            recursion call. It refers to the nth most recently opened parentheses.
345    
346        (b) (?+n) is also a relative subroutine call; it refers to the nth next
347            to be opened parentheses.
348    
349        (c) Conditions that refer to capturing parentheses can be specified
350            relatively, for example, (?(-2)... or (?(+3)...
351    
352        (d) \K resets the start of the current match so that everything before
353            is not part of it.
354    
355        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
356    
357        (f) \g{name} is another synonym - part of Perl 5.10's unification of
358            reference syntax.
359    
360        (g) (?| introduces a group in which the numbering of parentheses in each
361            alternative starts with the same number.
362    
363        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
364    
365     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
366        PCRE_INFO_JCHANGED.
367    
368     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
369        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
370        for detecting groups that can match an empty string.
371    
372     9. A pattern with a very large number of alternatives (more than several
373        hundred) was running out of internal workspace during the pre-compile
374        phase, where pcre_compile() figures out how much memory will be needed. A
375        bit of new cunning has reduced the workspace needed for groups with
376        alternatives. The 1000-alternative test pattern now uses 12 bytes of
377        workspace instead of running out of the 4096 that are available.
378    
379    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
380    
381    11. Applied patch from Google to remove an optimization that didn't quite work.
382        The report of the bug said:
383    
384          pcrecpp::RE("a*").FullMatch("aaa") matches, while
385          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
386          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
387    
388    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
389        it matched the wrong number of bytes.
390    
391    
392    Version 7.1 24-Apr-07
393    ---------------------
394    
395     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
396        that is more "standard", making use of automake and other Autotools. There
397        is some re-arrangement of the files and adjustment of comments consequent
398        on this.
399    
400     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
401        for recursive directory scanning broke on some systems because the files
402        are not scanned in any specific order and on different systems the order
403        was different. A call to "sort" has been inserted into RunGrepTest for the
404        approprate test as a short-term fix. In the longer term there may be an
405        alternative.
406    
407     3. I had an email from Eric Raymond about problems translating some of PCRE's
408        man pages to HTML (despite the fact that I distribute HTML pages, some
409        people do their own conversions for various reasons). The problems
410        concerned the use of low-level troff macros .br and .in. I have therefore
411        removed all such uses from the man pages (some were redundant, some could
412        be replaced by .nf/.fi pairs). The 132html script that I use to generate
413        HTML has been updated to handle .nf/.fi and to complain if it encounters
414        .br or .in.
415    
416     4. Updated comments in configure.ac that get placed in config.h.in and also
417        arranged for config.h to be included in the distribution, with the name
418        config.h.generic, for the benefit of those who have to compile without
419        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
420    
421     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
422        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
423        makevp.bat for the current PCRE, using the additional files
424        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
425    
426     6. A Windows user reported a minor discrepancy with test 2, which turned out
427        to be caused by a trailing space on an input line that had got lost in his
428        copy. The trailing space was an accident, so I've just removed it.
429    
430     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
431        that is needed.
432    
433     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
434        as "const" (a) because they are and (b) because it helps the PHP
435        maintainers who have recently made a script to detect big data structures
436        in the php code that should be moved to the .rodata section. I remembered
437        to update Builducptable as well, so it won't revert if ucptable.h is ever
438        re-created.
439    
440     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
441        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
442        order to be able to cut out the UTF-8 tables in the latter when UTF-8
443        support is not required. This saves 1.5-2K of code, which is important in
444        some applications.
445    
446        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
447        so as not to refer to the tables, even though these functions will never be
448        called when UTF-8 support is disabled. Otherwise there are problems with a
449        shared library.
450    
451    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
452    
453        (a) It was defining its arguments as char * instead of void *.
454    
455        (b) It was assuming that all moves were upwards in memory; this was true
456            a long time ago when I wrote it, but is no longer the case.
457    
458        The emulated memove() is provided for those environments that have neither
459        memmove() nor bcopy(). I didn't think anyone used it these days, but that
460        is clearly not the case, as these two bugs were recently reported.
461    
462    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
463        and Detrail to create the HTML documentation, the .txt form of the man
464        pages, and it removes trailing spaces from listed files. It also creates
465        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
466        case, it wraps all the #defines with #ifndefs. This script should be run
467        before "make dist".
468    
469    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
470        with Unicode property support.
471    
472        (a) For a maximizing quantifier, if the two different cases of the
473            character were of different lengths in their UTF-8 codings (there are
474            some cases like this - I found 11), and the matching function had to
475            back up over a mixture of the two cases, it incorrectly assumed they
476            were both the same length.
477    
478        (b) When PCRE was configured to use the heap rather than the stack for
479            recursion during matching, it was not correctly preserving the data for
480            the other case of a UTF-8 character when checking ahead for a match
481            while processing a minimizing repeat. If the check also involved
482            matching a wide character, but failed, corruption could cause an
483            erroneous result when trying to check for a repeat of the original
484            character.
485    
486    13. Some tidying changes to the testing mechanism:
487    
488        (a) The RunTest script now detects the internal link size and whether there
489            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
490            values substituted by "configure". (The RunGrepTest script already did
491            this for UTF-8.) The configure.ac script no longer substitutes the
492            relevant variables.
493    
494        (b) The debugging options /B and /D in pcretest show the compiled bytecode
495            with length and offset values. This means that the output is different
496            for different internal link sizes. Test 2 is skipped for link sizes
497            other than 2 because of this, bypassing the problem. Unfortunately,
498            there was also a test in test 3 (the locale tests) that used /B and
499            failed for link sizes other than 2. Rather than cut the whole test out,
500            I have added a new /Z option to pcretest that replaces the length and
501            offset values with spaces. This is now used to make test 3 independent
502            of link size. (Test 2 will be tidied up later.)
503    
504    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
505        segmentation fault instead of returning the appropriate error message.
506    
507    15. In multiline mode when the newline sequence was set to "any", the pattern
508        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
509        This doesn't seem right; it now treats the CRLF combination as the line
510        ending, and so does not match in that case. It's only a pattern such as ^$
511        that would hit this one: something like ^ABC$ would have failed after \r
512        and then tried again after \r\n.
513    
514    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
515        in an attempt to make files that differ only in their line terminators
516        compare equal. This works on Linux.
517    
518    17. Under certain error circumstances pcregrep might try to free random memory
519        as it exited. This is now fixed, thanks to valgrind.
520    
521    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
522        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
523        was because its rules for how to advance for /g after matching an empty
524        string at the end of a line did not allow for this case. They now check for
525        it specially.
526    
527    20. pcretest is supposed to handle patterns and data of any length, by
528        extending its buffers when necessary. It was getting this wrong when the
529        buffer for a data line had to be extended.
530    
531    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
532        CRLF as a newline sequence.
533    
534    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
535        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
536        I have nevertheless tidied it up.
537    
538    23. Added some casts to kill warnings from HP-UX ia64 compiler.
539    
540    24. Added a man page for pcre-config.
541    
542    
543    Version 7.0 19-Dec-06
544    ---------------------
545    
546     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
547        moving to gcc 4.1.1.
548    
549     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
550        sys/time.h, which is documented as needed for this function. It doesn't
551        seem to matter on Linux, but it showed up on some releases of OS X.
552    
553     3. It seems that there are systems where bytes whose values are greater than
554        127 match isprint() in the "C" locale. The "C" locale should be the
555        default when a C program starts up. In most systems, only ASCII printing
556        characters match isprint(). This difference caused the output from pcretest
557        to vary, making some of the tests fail. I have changed pcretest so that:
558    
559        (a) When it is outputting text in the compiled version of a pattern, bytes
560            other than 32-126 are always shown as hex escapes.
561    
562        (b) When it is outputting text that is a matched part of a subject string,
563            it does the same, unless a different locale has been set for the match
564            (using the /L modifier). In this case, it uses isprint() to decide.
565    
566     4. Fixed a major bug that caused incorrect computation of the amount of memory
567        required for a compiled pattern when options that changed within the
568        pattern affected the logic of the preliminary scan that determines the
569        length. The relevant options are -x, and -i in UTF-8 mode. The result was
570        that the computed length was too small. The symptoms of this bug were
571        either the PCRE error "internal error: code overflow" from pcre_compile(),
572        or a glibc crash with a message such as "pcretest: free(): invalid next
573        size (fast)". Examples of patterns that provoked this bug (shown in
574        pcretest format) are:
575    
576          /(?-x: )/x
577          /(?x)(?-x: \s*#\s*)/
578          /((?i)[\x{c0}])/8
579          /(?i:[\x{c0}])/8
580    
581        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
582        is now done differently.
583    
584     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
585        wrapper classes; (b) implement a new function in the C++ scanner that is
586        more efficient than the old way of doing things because it avoids levels of
587        recursion in the regex matching; (c) add a paragraph to the documentation
588        for the FullMatch() function.
589    
590     6. The escape sequence \n was being treated as whatever was defined as
591        "newline". Not only was this contrary to the documentation, which states
592        that \n is character 10 (hex 0A), but it also went horribly wrong when
593        "newline" was defined as CRLF. This has been fixed.
594    
595     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
596        was being set to -1 for the "end of line" case (supposedly a value that no
597        character can have). Though this value is never used (the check for end of
598        line is "zero bytes in current character"), it caused compiler complaints.
599        I've changed it to 0xffffffff.
600    
601     8. In pcre_version.c, the version string was being built by a sequence of
602        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
603        string (as it is for production releases) called a macro with an empty
604        argument. The C standard says the result of this is undefined. The gcc
605        compiler treats it as an empty string (which was what was wanted) but it is
606        reported that Visual C gives an error. The source has been hacked around to
607        avoid this problem.
608    
609     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
610        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
611        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
612        of them did).
613    
614    10. Originally, pcretest opened its input and output without "b"; then I was
615        told that "b" was needed in some environments, so it was added for release
616        5.0 to both the input and output. (It makes no difference on Unix-like
617        systems.) Later I was told that it is wrong for the input on Windows. I've
618        now abstracted the modes into two macros, to make it easier to fiddle with
619        them, and removed "b" from the input mode under Windows.
620    
621    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
622    
623    12. Added -help and --help to pcretest as an official way of being reminded
624        of the options.
625    
626    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
627        and pcrecpp.cc because they annoy compilers at high warning levels.
628    
629    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
630    
631    15. Fixed an occurrence of == in configure.ac that should have been = (shell
632        scripts are not C programs :-) and which was not noticed because it works
633        on Linux.
634    
635    16. pcretest is supposed to handle any length of pattern and data line (as one
636        line or as a continued sequence of lines) by extending its input buffer if
637        necessary. This feature was broken for very long pattern lines, leading to
638        a string of junk being passed to pcre_compile() if the pattern was longer
639        than about 50K.
640    
641    17. I have done a major re-factoring of the way pcre_compile() computes the
642        amount of memory needed for a compiled pattern. Previously, there was code
643        that made a preliminary scan of the pattern in order to do this. That was
644        OK when PCRE was new, but as the facilities have expanded, it has become
645        harder and harder to keep it in step with the real compile phase, and there
646        have been a number of bugs (see for example, 4 above). I have now found a
647        cunning way of running the real compile function in a "fake" mode that
648        enables it to compute how much memory it would need, while actually only
649        ever using a few hundred bytes of working memory and without too many
650        tests of the mode. This should make future maintenance and development
651        easier. A side effect of this work is that the limit of 200 on the nesting
652        depth of parentheses has been removed (though this was never a serious
653        limitation, I suspect). However, there is a downside: pcre_compile() now
654        runs more slowly than before (30% or more, depending on the pattern). I
655        hope this isn't a big issue. There is no effect on runtime performance.
656    
657    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
658        newline (only possible for the last line of a file) and it was a
659        pattern that set a locale (followed by /Lsomething), pcretest crashed.
660    
661    19. Added additional timing features to pcretest. (1) The -tm option now times
662        matching only, not compiling. (2) Both -t and -tm can be followed, as a
663        separate command line item, by a number that specifies the number of
664        repeats to use when timing. The default is 50000; this gives better
665        precision, but takes uncomfortably long for very large patterns.
666    
667    20. Extended pcre_study() to be more clever in cases where a branch of a
668        subpattern has no definite first character. For example, (a*|b*)[cd] would
669        previously give no result from pcre_study(). Now it recognizes that the
670        first character must be a, b, c, or d.
671    
672    21. There was an incorrect error "recursive call could loop indefinitely" if
673        a subpattern (or the entire pattern) that was being tested for matching an
674        empty string contained only one non-empty item after a nested subpattern.
675        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
676        incorrectly, because the \d was being skipped in the check.
677    
678    22. The pcretest program now has a new pattern option /B and a command line
679        option -b, which is equivalent to adding /B to every pattern. This causes
680        it to show the compiled bytecode, without the additional information that
681        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
682        is the same as /B/I).
683    
684    23. A new optimization is now able automatically to treat some sequences such
685        as a*b as a*+b. More specifically, if something simple (such as a character
686        or a simple class like \d) has an unlimited quantifier, and is followed by
687        something that cannot possibly match the quantified thing, the quantifier
688        is automatically "possessified".
689    
690    24. A recursive reference to a subpattern whose number was greater than 39
691        went wrong under certain circumstances in UTF-8 mode. This bug could also
692        have affected the operation of pcre_study().
693    
694    25. Realized that a little bit of performance could be had by replacing
695        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
696    
697    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
698    
699    27. Possessive quantifiers such as a++ were previously implemented by turning
700        them into atomic groups such as ($>a+). Now they have their own opcodes,
701        which improves performance. This includes the automatically created ones
702        from 23 above.
703    
704    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
705        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
706        the first matched character to be a colon. This applied both to named and
707        numbered groups.
708    
709    29. The ucpinternal.h header file was missing its idempotency #ifdef.
710    
711    30. I was sent a "project" file called libpcre.a.dev which I understand makes
712        building PCRE on Windows easier, so I have included it in the distribution.
713    
714    31. There is now a check in pcretest against a ridiculously large number being
715        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
716        loop, the loop is abandoned.
717    
718    32. Forward references to subpatterns in conditions such as (?(2)...) where
719        subpattern 2 is defined later cause pcre_compile() to search forwards in
720        the pattern for the relevant set of parentheses. This search went wrong
721        when there were unescaped parentheses in a character class, parentheses
722        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
723    
724    33. "Subroutine" calls and backreferences were previously restricted to
725        referencing subpatterns earlier in the regex. This restriction has now
726        been removed.
727    
728    34. Added a number of extra features that are going to be in Perl 5.10. On the
729        whole, these are just syntactic alternatives for features that PCRE had
730        previously implemented using the Python syntax or my own invention. The
731        other formats are all retained for compatibility.
732    
733        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
734            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
735            also .NET compatible.
736    
737        (b) A recursion or subroutine call to a named group can now be defined as
738            (?&name) as well as (?P>name).
739    
740        (c) A backreference to a named group can now be defined as \k<name> or
741            \k'name' as well as (?P=name). The new forms, as well as being in Perl
742            5.10, are also .NET compatible.
743    
744        (d) A conditional reference to a named group can now use the syntax
745            (?(<name>) or (?('name') as well as (?(name).
746    
747        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
748            groups (named and numbered) that are never evaluated inline, but can be
749            called as "subroutines" from elsewhere. In effect, the DEFINE condition
750            is always false. There may be only one alternative in such a group.
751    
752        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
753            as the simple (?(R). The condition is true only if the most recent
754            recursion is that of the given number or name. It does not search out
755            through the entire recursion stack.
756    
757        (g) The escape \gN or \g{N} has been added, where N is a positive or
758            negative number, specifying an absolute or relative reference.
759    
760    35. Tidied to get rid of some further signed/unsigned compiler warnings and
761        some "unreachable code" warnings.
762    
763    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
764        things, this adds five new scripts.
765    
766    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
767        There were also incompatibilities regarding the handling of \Q..\E inside
768        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
769        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
770    
771    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
772        matches an empty string, and forcibly breaks the loop. There were bugs in
773        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
774        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
775        separate and independent bugs (that affected different cases) have been
776        fixed.
777    
778    39. Refactored the code to abolish the use of different opcodes for small
779        capturing bracket numbers. This is a tidy that I avoided doing when I
780        removed the limit on the number of capturing brackets for 3.5 back in 2001.
781        The new approach is not only tidier, it makes it possible to reduce the
782        memory needed to fix the previous bug (38).
783    
784    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
785        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
786        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
787        mode.
788    
789    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
790        report.
791    
792    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
793        copy construction and assignment in the C++ wrapper.
794    
795    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
796        couple of bugs that could have given wrong results in the "--newline=crlf"
797        case.
798    
799    44. Added a number of casts and did some reorganization of signed/unsigned int
800        variables following suggestions from Dair Grant. Also renamed the variable
801        "this" as "item" because it is a C++ keyword.
802    
803    45. Arranged for dftables to add
804    
805          #include "pcre_internal.h"
806    
807        to pcre_chartables.c because without it, gcc 4.x may remove the array
808        definition from the final binary if PCRE is built into a static library and
809        dead code stripping is activated.
810    
811    46. For an unanchored pattern, if a match attempt fails at the start of a
812        newline sequence, and the newline setting is CRLF or ANY, and the next two
813        characters are CRLF, advance by two characters instead of one.
814    
815    
816    Version 6.7 04-Jul-06
817    ---------------------
818    
819     1. In order to handle tests when input lines are enormously long, pcretest has
820        been re-factored so that it automatically extends its buffers when
821        necessary. The code is crude, but this _is_ just a test program. The
822        default size has been increased from 32K to 50K.
823    
824     2. The code in pcre_study() was using the value of the re argument before
825        testing it for NULL. (Of course, in any sensible call of the function, it
826        won't be NULL.)
827    
828     3. The memmove() emulation function in pcre_internal.h, which is used on
829        systems that lack both memmove() and bcopy() - that is, hardly ever -
830        was missing a "static" storage class specifier.
831    
832     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
833        containing an extended class (one that cannot be represented by a bitmap
834        because it contains high-valued characters or Unicode property items, e.g.
835        [\pZ]). Almost always one would set UTF-8 mode when processing such a
836        pattern, but PCRE should not loop if you do not (it no longer does).
837        [Detail: two cases were found: (a) a repeated subpattern containing an
838        extended class; (b) a recursive reference to a subpattern that followed a
839        previous extended class. It wasn't skipping over the extended class
840        correctly when UTF-8 mode was not set.]
841    
842     5. A negated single-character class was not being recognized as fixed-length
843        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
844        compile error "lookbehind assertion is not fixed length".
845    
846     6. The RunPerlTest auxiliary script was showing an unexpected difference
847        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
848        write a Perl script that can interpret lines of an input file either as
849        byte characters or as UTF-8, which is what "perltest" was being required to
850        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
851        can't do is switch easily at run time between having the "use utf8;" pragma
852        or not. In the end, I fudged it by using the RunPerlTest script to insert
853        "use utf8;" explicitly for the UTF-8 tests.
854    
855     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
856        the end of the subject string, contrary to the documentation and to what
857        Perl does. This was true of both matching functions. Now it matches only at
858        the start of the subject and immediately after *internal* newlines.
859    
860     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
861        a pointer to an int instead of a pointer to an unsigned long int. This
862        caused problems on 64-bit systems.
863    
864     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
865        instance of the 'standard' template library not being so standard".
866    
867    10. There was no check on the number of named subpatterns nor the maximum
868        length of a subpattern name. The product of these values is used to compute
869        the size of the memory block for a compiled pattern. By supplying a very
870        long subpattern name and a large number of named subpatterns, the size
871        computation could be caused to overflow. This is now prevented by limiting
872        the length of names to 32 characters, and the number of named subpatterns
873        to 10,000.
874    
875    11. Subpatterns that are repeated with specific counts have to be replicated in
876        the compiled pattern. The size of memory for this was computed from the
877        length of the subpattern and the repeat count. The latter is limited to
878        65535, but there was no limit on the former, meaning that integer overflow
879        could in principle occur. The compiled length of a repeated subpattern is
880        now limited to 30,000 bytes in order to prevent this.
881    
882    12. Added the optional facility to have named substrings with the same name.
883    
884    13. Added the ability to use a named substring as a condition, using the
885        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
886        are numbers (not recommended). Forward references are permitted.
887    
888    14. Added forward references in named backreferences (if you see what I mean).
889    
890    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
891        pattern could run off the end of the subject. For example, the pattern
892        "(?s)(.{1,5})"8 did this with the subject "ab".
893    
894    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
895        PCRE_CASELESS was set when matching characters that were quantified with ?
896        or *.
897    
898    17. A character class other than a single negated character that had a minimum
899        but no maximum quantifier - for example [ab]{6,} - was not handled
900        correctly by pce_dfa_exec(). It would match only one character.
901    
902    18. A valid (though odd) pattern that looked like a POSIX character
903        class but used an invalid character after [ (for example [[,abc,]]) caused
904        pcre_compile() to give the error "Failed: internal error: code overflow" or
905        in some cases to crash with a glibc free() error. This could even happen if
906        the pattern terminated after [[ but there just happened to be a sequence of
907        letters, a binary zero, and a closing ] in the memory that followed.
908    
909    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
910        over the years. Originally (before any Unicode support), just the bottom 8
911        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
912        output from "man perlunicode" includes this:
913    
914          The regular expression compiler produces polymorphic opcodes.  That
915          is, the pattern adapts to the data and automatically switches to
916          the Unicode character scheme when presented with Unicode data--or
917          instead uses a traditional byte scheme when presented with byte
918          data.
919    
920        Sadly, a wide octal escape does not cause a switch, and in a string with
921        no other multibyte characters, these octal escapes are treated as before.
922        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
923        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
924        Unicode string.
925    
926        I have not perpetrated such confusion in PCRE. Up till now, it took just
927        the bottom 8 bits, as in old Perl. I have now made octal escapes with
928        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
929        translate to the appropriate multibyte character.
930    
931    29. Applied some refactoring to reduce the number of warnings from Microsoft
932        and Borland compilers. This has included removing the fudge introduced
933        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
934        a warning about an unused variable.
935    
936    21. PCRE has not included VT (character 0x0b) in the set of whitespace
937        characters since release 4.0, because Perl (from release 5.004) does not.
938        [Or at least, is documented not to: some releases seem to be in conflict
939        with the documentation.] However, when a pattern was studied with
940        pcre_study() and all its branches started with \s, PCRE still included VT
941        as a possible starting character. Of course, this did no harm; it just
942        caused an unnecessary match attempt.
943    
944    22. Removed a now-redundant internal flag bit that recorded the fact that case
945        dependency changed within the pattern. This was once needed for "required
946        byte" processing, but is no longer used. This recovers a now-scarce options
947        bit. Also moved the least significant internal flag bit to the most-
948        significant bit of the word, which was not previously used (hangover from
949        the days when it was an int rather than a uint) to free up another bit for
950        the future.
951    
952    23. Added support for CRLF line endings as well as CR and LF. As well as the
953        default being selectable at build time, it can now be changed at runtime
954        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
955        specify that it is scanning data with non-default line endings.
956    
957    24. Changed the definition of CXXLINK to make it agree with the definition of
958        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
959    
960    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
961        recursions. This makes a big different to stack usage for some patterns.
962    
963    26. If a subpattern containing a named recursion or subroutine reference such
964        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
965        the space required for the compiled pattern went wrong and gave too small a
966        value. Depending on the environment, this could lead to "Failed: internal
967        error: code overflow at offset 49" or "glibc detected double free or
968        corruption" errors.
969    
970    27. Applied patches from Google (a) to support the new newline modes and (b) to
971        advance over multibyte UTF-8 characters in GlobalReplace.
972    
973    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
974        difference for some implementation of PCRE in some Windows version.
975    
976    29. Added some extra testing facilities to pcretest:
977    
978        \q<number>   in a data line sets the "match limit" value
979        \Q<number>   in a data line sets the "match recursion limt" value
980        -S <number>  sets the stack size, where <number> is in megabytes
981    
982        The -S option isn't available for Windows.
983    
984    
985    Version 6.6 06-Feb-06
986    ---------------------
987    
988     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
989        in pcreposix.h. I have copied the definition from pcre.h.
990    
991     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
992        because pcre.h is no longer a built file.
993    
994     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
995        not normally included in the compiled code.
996    
997    
998    Version 6.5 01-Feb-06
999    ---------------------
1000    
1001     1. When using the partial match feature with pcre_dfa_exec(), it was not
1002        anchoring the second and subsequent partial matches at the new starting
1003        point. This could lead to incorrect results. For example, with the pattern
1004        /1234/, partially matching against "123" and then "a4" gave a match.
1005    
1006     2. Changes to pcregrep:
1007    
1008        (a) All non-match returns from pcre_exec() were being treated as failures
1009            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1010            error message is output. Some extra information is given for the
1011            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
1012            probably the only errors that are likely to be caused by users (by
1013            specifying a regex that has nested indefinite repeats, for instance).
1014            If there are more than 20 of these errors, pcregrep is abandoned.
1015    
1016        (b) A binary zero was treated as data while matching, but terminated the
1017            output line if it was written out. This has been fixed: binary zeroes
1018            are now no different to any other data bytes.
1019    
1020        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1021            used to set a locale for matching. The --locale=xxxx long option has
1022            been added (no short equivalent) to specify a locale explicitly on the
1023            pcregrep command, overriding the environment variables.
1024    
1025        (d) When -B was used with -n, some line numbers in the output were one less
1026            than they should have been.
1027    
1028        (e) Added the -o (--only-matching) option.
1029    
1030        (f) If -A or -C was used with -c (count only), some lines of context were
1031            accidentally printed for the final match.
1032    
1033        (g) Added the -H (--with-filename) option.
1034    
1035        (h) The combination of options -rh failed to suppress file names for files
1036            that were found from directory arguments.
1037    
1038        (i) Added the -D (--devices) and -d (--directories) options.
1039    
1040        (j) Added the -F (--fixed-strings) option.
1041    
1042        (k) Allow "-" to be used as a file name for -f as well as for a data file.
1043    
1044        (l) Added the --colo(u)r option.
1045    
1046        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1047            is not present by default.
1048    
1049     3. A nasty bug was discovered in the handling of recursive patterns, that is,
1050        items such as (?R) or (?1), when the recursion could match a number of
1051        alternatives. If it matched one of the alternatives, but subsequently,
1052        outside the recursion, there was a failure, the code tried to back up into
1053        the recursion. However, because of the way PCRE is implemented, this is not
1054        possible, and the result was an incorrect result from the match.
1055    
1056        In order to prevent this happening, the specification of recursion has
1057        been changed so that all such subpatterns are automatically treated as
1058        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1059    
1060     4. I had overlooked the fact that, in some locales, there are characters for
1061        which isalpha() is true but neither isupper() nor islower() are true. In
1062        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1063        and ordfeminine) are like this. This affected the treatment of \w and \W
1064        when they appeared in character classes, but not when they appeared outside
1065        a character class. The bit map for "word" characters is now created
1066        separately from the results of isalnum() instead of just taking it from the
1067        upper, lower, and digit maps. (Plus the underscore character, of course.)
1068    
1069     5. The above bug also affected the handling of POSIX character classes such as
1070        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1071        permanent tables. Instead, the bit maps for such a class were previously
1072        created as the appropriate unions of the upper, lower, and digit bitmaps.
1073        Now they are created by subtraction from the [[:word:]] class, which has
1074        its own bitmap.
1075    
1076     6. The [[:blank:]] character class matches horizontal, but not vertical space.
1077        It is created by subtracting the vertical space characters (\x09, \x0a,
1078        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1079        subtraction was done in the overall bitmap for a character class, meaning
1080        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1081        be recognized. This bug has been fixed.
1082    
1083     7. Patches from the folks at Google:
1084    
1085          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1086          real life, but is still worth protecting against".
1087    
1088          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1089          regular expressions".
1090    
1091          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1092          have it.
1093    
1094          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1095          "configure" and the latter not, in order to fix a problem somebody had
1096          with compiling the Arg class on HP-UX.
1097    
1098          (e) Improve the error-handling of the C++ wrapper a little bit.
1099    
1100          (f) New tests for checking recursion limiting.
1101    
1102     8. The pcre_memmove() function, which is used only if the environment does not
1103        have a standard memmove() function (and is therefore rarely compiled),
1104        contained two bugs: (a) use of int instead of size_t, and (b) it was not
1105        returning a result (though PCRE never actually uses the result).
1106    
1107     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1108        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1109        returned instead of calling malloc() with an overflowing number that would
1110        most likely cause subsequent chaos.
1111    
1112    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1113    
1114    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1115        with this option is matched, the nmatch and pmatch options of regexec() are
1116        ignored.
1117    
1118    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1119        provided in case anyone wants to the the POSIX interface with UTF-8
1120        strings.
1121    
1122    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1123        C++ linking (needed for some HP-UX environments).
1124    
1125    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1126        (unused parameter) and in the pcre_printint() function (omitted "default"
1127        switch label when the default is to do nothing).
1128    
1129    15. Added some code to make it possible, when PCRE is compiled as a C++
1130        library, to replace subject pointers for pcre_exec() with a smart pointer
1131        class, thus making it possible to process discontinuous strings.
1132    
1133    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1134        much the same function. They were added by different people who were trying
1135        to make PCRE easy to compile on non-Unix systems. It has been suggested
1136        that PCRE_EXPORT be abolished now that there is more automatic apparatus
1137        for compiling on Windows systems. I have therefore replaced it with
1138        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1139        defaults to "extern" for C or "extern C" for C++, which works fine on
1140        Unix-like systems. It is now possible to override the value of PCRE_DATA_
1141        SCOPE with something explicit in config.h. In addition:
1142    
1143        (a) pcreposix.h still had just "extern" instead of either of these macros;
1144            I have replaced it with PCRE_DATA_SCOPE.
1145    
1146        (b) Functions such as _pcre_xclass(), which are internal to the library,
1147            but external in the C sense, all had PCRE_EXPORT in their definitions.
1148            This is apparently wrong for the Windows case, so I have removed it.
1149            (It makes no difference on Unix-like systems.)
1150    
1151    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1152        of recursive calls to match(). This is different to MATCH_LIMIT because
1153        that limits the total number of calls to match(), not all of which increase
1154        the depth of recursion. Limiting the recursion depth limits the amount of
1155        stack (or heap if NO_RECURSE is set) that is used. The default can be set
1156        when PCRE is compiled, and changed at run time. A patch from Google adds
1157        this functionality to the C++ interface.
1158    
1159    18. Changes to the handling of Unicode character properties:
1160    
1161        (a) Updated the table to Unicode 4.1.0.
1162    
1163        (b) Recognize characters that are not in the table as "Cn" (undefined).
1164    
1165        (c) I revised the way the table is implemented to a much improved format
1166            which includes recognition of ranges. It now supports the ranges that
1167            are defined in UnicodeData.txt, and it also amalgamates other
1168            characters into ranges. This has reduced the number of entries in the
1169            table from around 16,000 to around 3,000, thus reducing its size
1170            considerably. I realized I did not need to use a tree structure after
1171            all - a binary chop search is just as efficient. Having reduced the
1172            number of entries, I extended their size from 6 bytes to 8 bytes to
1173            allow for more data.
1174    
1175        (d) Added support for Unicode script names via properties such as \p{Han}.
1176    
1177    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1178        matching that character.
1179    
1180    20. When matching a repeated Unicode property with a minimum greater than zero,
1181        (for example \pL{2,}), PCRE could look past the end of the subject if it
1182        reached it while seeking the minimum number of characters. This could
1183        happen only if some of the characters were more than one byte long, because
1184        there is a check for at least the minimum number of bytes.
1185    
1186    21. Refactored the implementation of \p and \P so as to be more general, to
1187        allow for more different types of property in future. This has changed the
1188        compiled form incompatibly. Anybody with saved compiled patterns that use
1189        \p or \P will have to recompile them.
1190    
1191    22. Added "Any" and "L&" to the supported property types.
1192    
1193    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1194        but give a compile time error if the value is greater than 0xff.
1195    
1196    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1197        accidentally not being installed or uninstalled.
1198    
1199    25. The pcre.h file was built from pcre.h.in, but the only changes that were
1200        made were to insert the current release number. This seemed silly, because
1201        it made things harder for people building PCRE on systems that don't run
1202        "configure". I have turned pcre.h into a distributed file, no longer built
1203        by "configure", with the version identification directly included. There is
1204        no longer a pcre.h.in file.
1205    
1206        However, this change necessitated a change to the pcre-config script as
1207        well. It is built from pcre-config.in, and one of the substitutions was the
1208        release number. I have updated configure.ac so that ./configure now finds
1209        the release number by grepping pcre.h.
1210    
1211    26. Added the ability to run the tests under valgrind.
1212    
1213    
1214    Version 6.4 05-Sep-05
1215    ---------------------
1216    
1217     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1218        "--" to be printed when multiple files were scanned, even when none of the
1219        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1220        consider it to be a bug, and have restored the previous behaviour.
1221    
1222     2. A couple of code tidies to get rid of compiler warnings.
1223    
1224     3. The pcretest program used to cheat by referring to symbols in the library
1225        whose names begin with _pcre_. These are internal symbols that are not
1226        really supposed to be visible externally, and in some environments it is
1227        possible to suppress them. The cheating is now confined to including
1228        certain files from the library's source, which is a bit cleaner.
1229    
1230     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1231        file's purpose clearer.
1232    
1233     5. Reorganized pcre_ucp_findchar().
1234    
1235    
1236    Version 6.3 15-Aug-05
1237    ---------------------
1238    
1239     1. The file libpcre.pc.in did not have general read permission in the tarball.
1240    
1241     2. There were some problems when building without C++ support:
1242    
1243        (a) If C++ support was not built, "make install" and "make test" still
1244            tried to test it.
1245    
1246        (b) There were problems when the value of CXX was explicitly set. Some
1247            changes have been made to try to fix these, and ...
1248    
1249        (c) --disable-cpp can now be used to explicitly disable C++ support.
1250    
1251        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1252            backslash in a target when C++ was disabled. This confuses some
1253            versions of "make", apparently. Using an intermediate variable solves
1254            this. (Same for CPP_LOBJ.)
1255    
1256     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1257        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1258        necessary on certain architectures.
1259    
1260     4. Added a setting of -export-symbols-regex to the link command to remove
1261        those symbols that are exported in the C sense, but actually are local
1262        within the library, and not documented. Their names all begin with
1263        "_pcre_". This is not a perfect job, because (a) we have to except some
1264        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1265        available (and never for static libraries). I have made a note to try to
1266        find a way round (a) in the future.
1267    
1268    
1269    Version 6.2 01-Aug-05
1270    ---------------------
1271    
1272     1. There was no test for integer overflow of quantifier values. A construction
1273        such as {1111111111111111} would give undefined results. What is worse, if
1274        a minimum quantifier for a parenthesized subpattern overflowed and became
1275        negative, the calculation of the memory size went wrong. This could have
1276        led to memory overwriting.
1277    
1278     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1279    
1280     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1281        operating environments where this matters.
1282    
1283     4. Applied Giuseppe Maxia's patch to add additional features for controlling
1284        PCRE options from within the C++ wrapper.
1285    
1286     5. Named capturing subpatterns were not being correctly counted when a pattern
1287        was compiled. This caused two problems: (a) If there were more than 100
1288        such subpatterns, the calculation of the memory needed for the whole
1289        compiled pattern went wrong, leading to an overflow error. (b) Numerical
1290        back references of the form \12, where the number was greater than 9, were
1291        not recognized as back references, even though there were sufficient
1292        previous subpatterns.
1293    
1294     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1295        versions of gcc, e.g. 2.95.4.
1296    
1297    
1298    Version 6.1 21-Jun-05
1299    ---------------------
1300    
1301     1. There was one reference to the variable "posix" in pcretest.c that was not
1302        surrounded by "#if !defined NOPOSIX".
1303    
1304     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1305        the cross-check on the old pcre_info() function, for the benefit of the
1306        cut-down version of PCRE that is currently imported into Exim.
1307    
1308     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1309        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1310        patterns that start (?i) but I don't think that matters. The use of (?i) is
1311        just an example; this all applies to the other options as well.
1312    
1313     4. Since libtool seems to echo the compile commands it is issuing, the output
1314        from "make" can be reduced a bit by putting "@" in front of each libtool
1315        compile command.
1316    
1317     5. Patch from the folks at Google for configure.in to be a bit more thorough
1318        in checking for a suitable C++ installation before trying to compile the
1319        C++ stuff. This should fix a reported problem when a compiler was present,
1320        but no suitable headers.
1321    
1322     6. The man pages all had just "PCRE" as their title. I have changed them to
1323        be the relevant file name. I have also arranged that these names are
1324        retained in the file doc/pcre.txt, which is a concatenation in text format
1325        of all the man pages except the little individual ones for each function.
1326    
1327     7. The NON-UNIX-USE file had not been updated for the different set of source
1328        files that come with release 6. I also added a few comments about the C++
1329        wrapper.
1330    
1331    
1332    Version 6.0 07-Jun-05
1333    ---------------------
1334    
1335     1. Some minor internal re-organization to help with my DFA experiments.
1336    
1337     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1338        didn't matter for the library itself when fully configured, but did matter
1339        when compiling without UCP support, or within Exim, where the ucp files are
1340        not imported.
1341    
1342     3. Refactoring of the library code to split up the various functions into
1343        different source modules. The addition of the new DFA matching code (see
1344        below) to a single monolithic source would have made it really too
1345        unwieldy, quite apart from causing all the code to be include in a
1346        statically linked application, when only some functions are used. This is
1347        relevant even without the DFA addition now that patterns can be compiled in
1348        one application and matched in another.
1349    
1350        The downside of splitting up is that there have to be some external
1351        functions and data tables that are used internally in different modules of
1352        the library but which are not part of the API. These have all had their
1353        names changed to start with "_pcre_" so that they are unlikely to clash
1354        with other external names.
1355    
1356     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1357        a different (DFA) algorithm. Although it is slower than the original
1358        function, it does have some advantages for certain types of matching
1359        problem.
1360    
1361     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1362        including restarting after a partial match.
1363    
1364     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1365        defined when compiling for Windows was sent to me. I have put it into the
1366        code, though I have no means of testing or verifying it.
1367    
1368     7. Added the pcre_refcount() auxiliary function.
1369    
1370     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1371        match before or at the first newline in the subject string. In pcretest,
1372        the /f option on a pattern can be used to set this.
1373    
1374     9. A repeated \w when used in UTF-8 mode with characters greater than 256
1375        would behave wrongly. This has been present in PCRE since release 4.0.
1376    
1377    10. A number of changes to the pcregrep command:
1378    
1379        (a) Refactored how -x works; insert ^(...)$ instead of setting
1380            PCRE_ANCHORED and checking the length, in preparation for adding
1381            something similar for -w.
1382    
1383        (b) Added the -w (match as a word) option.
1384    
1385        (c) Refactored the way lines are read and buffered so as to have more
1386            than one at a time available.
1387    
1388        (d) Implemented a pcregrep test script.
1389    
1390        (e) Added the -M (multiline match) option. This allows patterns to match
1391            over several lines of the subject. The buffering ensures that at least
1392            8K, or the rest of the document (whichever is the shorter) is available
1393            for matching (and similarly the previous 8K for lookbehind assertions).
1394    
1395        (f) Changed the --help output so that it now says
1396    
1397              -w, --word-regex(p)
1398    
1399            instead of two lines, one with "regex" and the other with "regexp"
1400            because that confused at least one person since the short forms are the
1401            same. (This required a bit of code, as the output is generated
1402            automatically from a table. It wasn't just a text change.)
1403    
1404        (g) -- can be used to terminate pcregrep options if the next thing isn't an
1405            option but starts with a hyphen. Could be a pattern or a path name
1406            starting with a hyphen, for instance.
1407    
1408        (h) "-" can be given as a file name to represent stdin.
1409    
1410        (i) When file names are being printed, "(standard input)" is used for
1411            the standard input, for compatibility with GNU grep. Previously
1412            "<stdin>" was used.
1413    
1414        (j) The option --label=xxx can be used to supply a name to be used for
1415            stdin when file names are being printed. There is no short form.
1416    
1417        (k) Re-factored the options decoding logic because we are going to add
1418            two more options that take data. Such options can now be given in four
1419            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1420    
1421        (l) Added the -A, -B, and -C options for requesting that lines of context
1422            around matches be printed.
1423    
1424        (m) Added the -L option to print the names of files that do not contain
1425            any matching lines, that is, the complement of -l.
1426    
1427        (n) The return code is 2 if any file cannot be opened, but pcregrep does
1428            continue to scan other files.
1429    
1430        (o) The -s option was incorrectly implemented. For compatibility with other
1431            greps, it now suppresses the error message for a non-existent or non-
1432            accessible file (but not the return code). There is a new option called
1433            -q that suppresses the output of matching lines, which was what -s was
1434            previously doing.
1435    
1436        (p) Added --include and --exclude options to specify files for inclusion
1437            and exclusion when recursing.
1438    
1439    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1440        Hopefully, it now does.
1441    
1442    12. Missing cast in pcre_study().
1443    
1444    13. Added an "uninstall" target to the makefile.
1445    
1446    14. Replaced "extern" in the function prototypes in Makefile.in with
1447        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1448        world, but is set differently for Windows.
1449    
1450    15. Added a second compiling function called pcre_compile2(). The only
1451        difference is that it has an extra argument, which is a pointer to an
1452        integer error code. When there is a compile-time failure, this is set
1453        non-zero, in addition to the error test pointer being set to point to an
1454        error message. The new argument may be NULL if no error number is required
1455        (but then you may as well call pcre_compile(), which is now just a
1456        wrapper). This facility is provided because some applications need a
1457        numeric error indication, but it has also enabled me to tidy up the way
1458        compile-time errors are handled in the POSIX wrapper.
1459    
1460    16. Added VPATH=.libs to the makefile; this should help when building with one
1461        prefix path and installing with another. (Or so I'm told by someone who
1462        knows more about this stuff than I do.)
1463    
1464    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1465        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1466        match everything, including newlines. This is not POSIX-compatible, but
1467        somebody wanted the feature. From pcretest it can be activated by using
1468        both the P and the s flags.
1469    
1470    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1471    
1472    19. libpcre.pc was being incorrectly installed as executable.
1473    
1474    20. A couple of places in pcretest check for end-of-line by looking for '\n';
1475        it now also looks for '\r' so that it will work unmodified on Windows.
1476    
1477    21. Added Google's contributed C++ wrapper to the distribution.
1478    
1479    22. Added some untidy missing memory free() calls in pcretest, to keep
1480        Electric Fence happy when testing.
1481    
1482    
1483    
1484    Version 5.0 13-Sep-04
1485    ---------------------
1486    
1487     1. Internal change: literal characters are no longer packed up into items
1488        containing multiple characters in a single byte-string. Each character
1489        is now matched using a separate opcode. However, there may be more than one
1490        byte in the character in UTF-8 mode.
1491    
1492     2. The pcre_callout_block structure has two new fields: pattern_position and
1493        next_item_length. These contain the offset in the pattern to the next match
1494        item, and its length, respectively.
1495    
1496     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1497        insertion of callouts before each pattern item. Added the /C option to
1498        pcretest to make use of this.
1499    
1500     4. On the advice of a Windows user, the lines
1501    
1502          #if defined(_WIN32) || defined(WIN32)
1503          _setmode( _fileno( stdout ), 0x8000 );
1504          #endif  /* defined(_WIN32) || defined(WIN32) */
1505    
1506        have been added to the source of pcretest. This apparently does useful
1507        magic in relation to line terminators.
1508    
1509     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1510        for the benefit of those environments where the "b" makes a difference.
1511    
1512     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1513        to know about it. I have put a hack into configure.in that adds in code
1514        to set GCC=yes if CC=icc. This seems to end up at a point in the
1515        generated configure script that is early enough to affect the setting of
1516        compiler options, which is what is needed, but I have no means of testing
1517        whether it really works. (The user who reported this had patched the
1518        generated configure script, which of course I cannot do.)
1519    
1520        LATER: After change 22 below (new libtool files), the configure script
1521        seems to know about icc (and also ecc). Therefore, I have commented out
1522        this hack in configure.in.
1523    
1524     7. Added support for pkg-config (2 patches were sent in).
1525    
1526     8. Negated POSIX character classes that used a combination of internal tables
1527        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1528        [[:^ascii]]. Typically, they would match almost any characters. The other
1529        POSIX classes were not broken in this way.
1530    
1531     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1532        to find the match, as PCRE was deluded into thinking that the match had to
1533        start at the start point or following a newline. The same bug applied to
1534        patterns with negative forward assertions or any backward assertions
1535        preceding ".*" at the start, unless the pattern required a fixed first
1536        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1537    
1538    10. In UTF-8 mode, when moving forwards in the subject after a failed match
1539        starting at the last subject character, bytes beyond the end of the subject
1540        string were read.
1541    
1542    11. Renamed the variable "class" as "classbits" to make life easier for C++
1543        users. (Previously there was a macro definition, but it apparently wasn't
1544        enough.)
1545    
1546    12. Added the new field "tables" to the extra data so that tables can be passed
1547        in at exec time, or the internal tables can be re-selected. This allows
1548        a compiled regex to be saved and re-used at a later time by a different
1549        program that might have everything at different addresses.
1550    
1551    13. Modified the pcre-config script so that, when run on Solaris, it shows a
1552        -R library as well as a -L library.
1553    
1554    14. The debugging options of pcretest (-d on the command line or D on a
1555        pattern) showed incorrect output for anything following an extended class
1556        that contained multibyte characters and which was followed by a quantifier.
1557    
1558    15. Added optional support for general category Unicode character properties
1559        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1560        support. It adds about 90K to the size of the library. The meanings of the
1561        inbuilt class escapes such as \d and \s have NOT been changed.
1562    
1563    16. Updated pcredemo.c to include calls to free() to release the memory for the
1564        compiled pattern.
1565    
1566    17. The generated file chartables.c was being created in the source directory
1567        instead of in the building directory. This caused the build to fail if the
1568        source directory was different from the building directory, and was
1569        read-only.
1570    
1571    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1572        file. No doubt somebody will tell me if they don't make sense... Also added
1573        Dan Mooney's comments about building on OpenVMS.
1574    
1575    19. Added support for partial matching via the PCRE_PARTIAL option for
1576        pcre_exec() and the \P data escape in pcretest.
1577    
1578    20. Extended pcretest with 3 new pattern features:
1579    
1580        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
1581              write the compiled pattern to the file whose name is "rest-of-line".
1582              This is a straight binary dump of the data, with the saved pointer to
1583              the character tables forced to be NULL. The study data, if any, is
1584              written too. After writing, pcretest reads a new pattern.
1585    
1586        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1587              compiled pattern from the given file. There must not be any
1588              occurrences of "<" in the file name (pretty unlikely); if there are,
1589              pcretest will instead treat the initial "<" as a pattern delimiter.
1590              After reading in the pattern, pcretest goes on to read data lines as
1591              usual.
1592    
1593        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1594              and 16-bit fields in a compiled pattern, to simulate a pattern that
1595              was compiled on a host of opposite endianness.
1596    
1597    21. The pcre-exec() function can now cope with patterns that were compiled on
1598        hosts of opposite endianness, with this restriction:
1599    
1600          As for any compiled expression that is saved and used later, the tables
1601          pointer field cannot be preserved; the extra_data field in the arguments
1602          to pcre_exec() should be used to pass in a tables address if a value
1603          other than the default internal tables were used at compile time.
1604    
1605    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1606        now diagnosed as an error. Previously, most of the time, a negative number
1607        would have been treated as zero, but if in addition "ovector" was passed as
1608        NULL, a crash could occur.
1609    
1610    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1611        new versions from the libtool 1.5 distribution (the last one is a copy of
1612        a file called libtool.m4). This seems to have fixed the need to patch
1613        "configure" to support Darwin 1.3 (which I used to do). However, I still
1614        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1615        workstation).
1616    
1617    24. Changed the PCRE licence to be the more standard "BSD" licence.
1618    
1619    
1620    Version 4.5 01-Dec-03
1621    ---------------------
1622    
1623     1. There has been some re-arrangement of the code for the match() function so
1624        that it can be compiled in a version that does not call itself recursively.
1625        Instead, it keeps those local variables that need separate instances for
1626        each "recursion" in a frame on the heap, and gets/frees frames whenever it
1627        needs to "recurse". Keeping track of where control must go is done by means
1628        of setjmp/longjmp. The whole thing is implemented by a set of macros that
1629        hide most of the details from the main code, and operates only if
1630        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1631        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1632        operating.
1633    
1634        To make it easier for callers to provide specially tailored get/free
1635        functions for this usage, two new functions, pcre_stack_malloc, and
1636        pcre_stack_free, are used. They are always called in strict stacking order,
1637        and the size of block requested is always the same.
1638    
1639        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1640        PCRE has been compiled to use the stack or the heap for recursion. The
1641        -C option of pcretest uses this to show which version is compiled.
1642    
1643        A new data escape \S, is added to pcretest; it causes the amounts of store
1644        obtained and freed by both kinds of malloc/free at match time to be added
1645        to the output.
1646    
1647     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1648        what's available on my current Linux desktop machine.
1649    
1650     3. When matching a UTF-8 string, the test for a valid string at the start has
1651        been extended. If start_offset is not zero, PCRE now checks that it points
1652        to a byte that is the start of a UTF-8 character. If not, it returns
1653        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1654        this is necessary because there may be backward assertions in the pattern.
1655        When matching the same subject several times, it may save resources to use
1656        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1657    
1658     4. The code for checking the validity of UTF-8 strings has been tightened so
1659        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1660        containing "overlong sequences".
1661    
1662     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1663        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1664        should have been "&", but it just so happened that all the cases this let
1665        through by mistake were picked up later in the function.
1666    
1667     6. I had used a variable called "isblank" - this is a C99 function, causing
1668        some compilers to warn. To avoid this, I renamed it (as "blankclass").
1669    
1670     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1671        prompting; (b) run "./pcretest /dev/null" at the start of the test script
1672        so the version is shown; (c) stop "make test" echoing "./RunTest".
1673    
1674     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1675    
1676     9. The prototype for memmove() for systems that don't have it was using
1677        size_t, but the inclusion of the header that defines size_t was later. I've
1678        moved the #includes for the C headers earlier to avoid this.
1679    
1680    10. Added some adjustments to the code to make it easier to compiler on certain
1681        special systems:
1682    
1683          (a) Some "const" qualifiers were missing.
1684          (b) Added the macro EXPORT before all exported functions; by default this
1685              is defined to be empty.
1686          (c) Changed the dftables auxiliary program (that builds chartables.c) so
1687              that it reads its output file name as an argument instead of writing
1688              to the standard output and assuming this can be redirected.
1689    
1690    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
1691        class containing characters with values greater than 255, PCRE compilation
1692        went into a loop.
1693    
1694    12. A recursive reference to a subpattern that was within another subpattern
1695        that had a minimum quantifier of zero caused PCRE to crash. For example,
1696        (x(y(?2))z)? provoked this bug with a subject that got as far as the
1697        recursion. If the recursively-called subpattern itself had a zero repeat,
1698        that was OK.
1699    
1700    13. In pcretest, the buffer for reading a data line was set at 30K, but the
1701        buffer into which it was copied (for escape processing) was still set at
1702        1024, so long lines caused crashes.
1703    
1704    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
1705        "internal error: code overflow...". This applied to any character class
1706        that was followed by a possessive quantifier.
1707    
1708    15. Modified the Makefile to add libpcre.la as a prerequisite for
1709        libpcreposix.la because I was told this is needed for a parallel build to
1710        work.
1711    
1712    16. If a pattern that contained .* following optional items at the start was
1713        studied, the wrong optimizing data was generated, leading to matching
1714        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
1715        matching string must start with a or b or c. The correct conclusion for
1716        this pattern is that a match can start with any character.
1717    
1718    
1719    Version 4.4 13-Aug-03
1720    ---------------------
1721    
1722     1. In UTF-8 mode, a character class containing characters with values between
1723        127 and 255 was not handled correctly if the compiled pattern was studied.
1724        In fixing this, I have also improved the studying algorithm for such
1725        classes (slightly).
1726    
1727     2. Three internal functions had redundant arguments passed to them. Removal
1728        might give a very teeny performance improvement.
1729    
1730     3. Documentation bug: the value of the capture_top field in a callout is *one
1731        more than* the number of the hightest numbered captured substring.
1732    
1733     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
1734        in incorrectly linking with a previously installed version. They now link
1735        explicitly with libpcre.la.
1736    
1737     5. configure.in no longer needs to recognize Cygwin specially.
1738    
1739     6. A problem in pcre.in for Windows platforms is fixed.
1740    
1741     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
1742        pcretest, it used to include the size of the study block as part of its
1743        output. Unfortunately, the structure contains a field that has a different
1744        size on different hardware architectures. This meant that the tests that
1745        showed this size failed. As the block is currently always of a fixed size,
1746        this information isn't actually particularly useful in pcretest output, so
1747        I have just removed it.
1748    
1749     8. Three pre-processor statements accidentally did not start in column 1.
1750        Sadly, there are *still* compilers around that complain, even though
1751        standard C has not required this for well over a decade. Sigh.
1752    
1753     9. In pcretest, the code for checking callouts passed small integers in the
1754        callout_data field, which is a void * field. However, some picky compilers
1755        complained about the casts involved for this on 64-bit systems. Now
1756        pcretest passes the address of the small integer instead, which should get
1757        rid of the warnings.
1758    
1759    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
1760        both compile and run time, and gives an error if an invalid UTF-8 sequence
1761        is found. There is a option for disabling this check in cases where the
1762        string is known to be correct and/or the maximum performance is wanted.
1763    
1764    11. In response to a bug report, I changed one line in Makefile.in from
1765    
1766            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
1767        to
1768            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
1769    
1770        to look similar to other lines, but I have no way of telling whether this
1771        is the right thing to do, as I do not use Windows. No doubt I'll get told
1772        if it's wrong...
1773    
1774    
1775    Version 4.3 21-May-03
1776    ---------------------
1777    
1778    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
1779       Makefile.
1780    
1781    2. Some refactoring to improve the quality of the code:
1782    
1783       (i)   The utf8_table... variables are now declared "const".
1784    
1785       (ii)  The code for \cx, which used the "case flipping" table to upper case
1786             lower case letters, now just substracts 32. This is ASCII-specific,
1787             but the whole concept of \cx is ASCII-specific, so it seems
1788             reasonable.
1789    
1790       (iii) PCRE was using its character types table to recognize decimal and
1791             hexadecimal digits in the pattern. This is silly, because it handles
1792             only 0-9, a-f, and A-F, but the character types table is locale-
1793             specific, which means strange things might happen. A private
1794             table is now used for this - though it costs 256 bytes, a table is
1795             much faster than multiple explicit tests. Of course, the standard
1796             character types table is still used for matching digits in subject
1797             strings against \d.
1798    
1799       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
1800             ending in _t are). So I've renamed it as ESC_tee.
1801    
1802    3. The first argument for regexec() in the POSIX wrapper should have been
1803       defined as "const".
1804    
1805    4. Changed pcretest to use malloc() for its buffers so that they can be
1806       Electric Fenced for debugging.
1807    
1808    5. There were several places in the code where, in UTF-8 mode, PCRE would try
1809       to read one or more bytes before the start of the subject string. Often this
1810       had no effect on PCRE's behaviour, but in some circumstances it could
1811       provoke a segmentation fault.
1812    
1813    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
1814       to try to read one or more bytes before the start of the subject string.
1815    
1816    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
1817       UTF-8 support could misbehave in various ways if the subject string
1818       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
1819       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
1820       back over UTF-8 characters.)
1821    
1822    
1823    Version 4.2 14-Apr-03
1824    ---------------------
1825    
1826    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
1827    
1828    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
1829         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
1830         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
1831         [WIN_PREFIX]: new variable, "cyg" for Cygwin
1832         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
1833           and BUILD_EXEEXT
1834         Note: automatic setting of the BUILD variables is not yet working
1835         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
1836           compile-time but not at link-time
1837         [LINK]: use for linking executables only
1838         make different versions for Windows and non-Windows
1839         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
1840           libraries
1841         [LINK_FOR_BUILD]: new variable
1842         [OBJEXT]: use throughout
1843         [EXEEXT]: use throughout
1844         <winshared>: new target
1845         <wininstall>: new target
1846         <dftables.o>: use native compiler
1847         <dftables>: use native linker
1848         <install>: handle Windows platform correctly
1849         <clean>: ditto
1850         <check>: ditto
1851         copy DLL to top builddir before testing
1852    
1853       As part of these changes, -no-undefined was removed again. This was reported
1854       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
1855       in any case.
1856    
1857    3. Some tidies to get rid of compiler warnings:
1858    
1859       . In the match_data structure, match_limit was an unsigned long int, whereas
1860         match_call_count was an int. I've made them both unsigned long ints.
1861    
1862       . In pcretest the fact that a const uschar * doesn't automatically cast to
1863         a void * provoked a warning.
1864    
1865       . Turning on some more compiler warnings threw up some "shadow" variables
1866         and a few more missing casts.
1867    
1868    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
1869       option, a class that contained a single character with a value between 128
1870       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
1871    
1872    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
1873       option, a class that contained several characters, but with at least one
1874       whose value was between 128 and 255 caused PCRE to crash.
1875    
1876    
1877    Version 4.1 12-Mar-03
1878    ---------------------
1879    
1880    1. Compiling with gcc -pedantic found a couple of places where casts were
1881    needed, and a string in dftables.c that was longer than standard compilers are
1882    required to support.
1883    
1884    2. Compiling with Sun's compiler found a few more places where the code could
1885    be tidied up in order to avoid warnings.
1886    
1887    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
1888    first of these names is deprecated in the latest Autoconf in favour of the name
1889    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
1890    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
1891    analogy I have changed it to CFLAGS_FOR_BUILD.
1892    
1893    4. Added -no-undefined to the linking command in the Makefile, because this is
1894    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
1895    linking step for the pcreposix library.
1896    
1897    5. PCRE was failing to diagnose the case of two named groups with the same
1898    name.
1899    
1900    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
1901    literal character that is needed in the subject for a match, and scans along to
1902    ensure that it is present before embarking on the full matching process. This
1903    saves time in cases of nested unlimited repeats that are never going to match.
1904    Problem: the scan can take a lot of time if the subject is very long (e.g.
1905    megabytes), thus penalizing straightforward matches. It is now done only if the
1906    amount of subject to be scanned is less than 1000 bytes.
1907    
1908    7. A lesser problem with the same optimization is that it was recording the
1909    first character of an anchored pattern as "needed", thus provoking a search
1910    right along the subject, even when the first match of the pattern was going to
1911    fail. The "needed" character is now not set for anchored patterns, unless it
1912    follows something in the pattern that is of non-fixed length. Thus, it still
1913    fulfils its original purpose of finding quick non-matches in cases of nested
1914    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
1915    
1916    
1917    Version 4.0 17-Feb-03
1918    ---------------------
1919    
1920    1. If a comment in an extended regex that started immediately after a meta-item
1921    extended to the end of string, PCRE compiled incorrect data. This could lead to
1922    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
1923    
1924    2. Moved to autoconf 2.53 and libtool 1.4.2.
1925    
1926    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
1927    the special perltest8 script is no longer needed - all the tests can be run
1928    from a single perltest script.
1929    
1930    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
1931    by \s. It has now been removed in PCRE. This means it isn't recognized as
1932    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
1933    class [:space:] *does* include VT, thereby creating a mess.
1934    
1935    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
1936    space and tab.
1937    
1938    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
1939    its new features into the main test script, reducing the number of scripts.
1940    
1941    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
1942    were backward compatible, and made the (?i) apply to the whole pattern, as if
1943    /i were given. Now it behaves more logically, and applies the option setting
1944    only to what follows. PCRE has been changed to follow suit. However, if it
1945    finds options settings right at the start of the pattern, it extracts them into
1946    the global options, as before. Thus, they show up in the info data.
1947    
1948    8. Added support for the \Q...\E escape sequence. Characters in between are
1949    treated as literals. This is slightly different from Perl in that $ and @ are
1950    also handled as literals inside the quotes. In Perl, they will cause variable
1951    interpolation. Note the following examples:
1952    
1953        Pattern            PCRE matches      Perl matches
1954    
1955        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
1956        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
1957        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
1958    
1959    For compatibility with Perl, \Q...\E sequences are recognized inside character
1960    classes as well as outside them.
1961    
1962    9. Re-organized 3 code statements in pcretest to avoid "overflow in
1963    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
1964    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
1965    signed/unsigned warnings.
1966    
1967    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
1968    option for pcretest, so I've replaced it by a simple function that does just
1969    that job.
1970    
1971    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
1972    "pcregrep -".
1973    
1974    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
1975    Java package. This provides some syntactic sugar for simple cases of what my
1976    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
1977    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
1978    item, you can use this simplified notation. Note that only makes sense with
1979    greedy quantifiers. Consequently, the use of the possessive quantifier forces
1980    greediness, whatever the setting of the PCRE_UNGREEDY option.
1981    
1982    13. A change of greediness default within a pattern was not taking effect at
1983    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
1984    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
1985    was abstracted outside.
1986    
1987    14. PCRE now supports the \G assertion. It is true when the current matching
1988    position is at the start point of the match. This differs from \A when the
1989    starting offset is non-zero. Used with the /g option of pcretest (or similar
1990    code), it works in the same way as it does for Perl's /g option. If all
1991    alternatives of a regex begin with \G, the expression is anchored to the start
1992    match position, and the "anchored" flag is set in the compiled expression.
1993    
1994    15. Some bugs concerning the handling of certain option changes within patterns
1995    have been fixed. These applied to options other than (?ims). For example,
1996    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
1997    been the other way round. Some of this was related to change 7 above.
1998    
1999    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
2000    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
2001    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
2002    POSIX classes only within a class (e.g. /[[:alpha:]]/).
2003    
2004    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
2005    mode. Unlike ".", it always matches newline, whatever the setting of
2006    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
2007    assertions. Perl allows it, but it doesn't (in general) work because it can't
2008    calculate the length of the lookbehind. At least, that's the case for Perl
2009    5.8.0 - I've been told they are going to document that it doesn't work in
2010    future.
2011    
2012    18. Added an error diagnosis for escapes that PCRE does not support: these are
2013    \L, \l, \N, \P, \p, \U, \u, and \X.
2014    
2015    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
2016    reading past the end of the pattern in cases such as /[abcd/.
2017    
2018    20. PCRE was getting more memory than necessary for patterns with classes that
2019    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
2020    
2021    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
2022    compiling PCRE for use with Virtual Pascal.
2023    
2024    22. Small fix to the Makefile to make it work properly if the build is done
2025    outside the source tree.
2026    
2027    23. Added a new extension: a condition to go with recursion. If a conditional
2028    subpattern starts with (?(R) the "true" branch is used if recursion has
2029    happened, whereas the "false" branch is used only at the top level.
2030    
2031    24. When there was a very long string of literal characters (over 255 bytes
2032    without UTF support, over 250 bytes with UTF support), the computation of how
2033    much memory was required could be incorrect, leading to segfaults or other
2034    strange effects.
2035    
2036    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
2037    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
2038    there was a subsequent back reference to those brackets. This meant that, for
2039    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
2040    possible to check for precisely this case. All we can do is abandon the
2041    optimization if .* occurs inside capturing brackets when there are any back
2042    references whatsoever. (See below for a better fix that came later.)
2043    
2044    26. The handling of the optimization for finding the first character of a
2045    non-anchored pattern, and for finding a character that is required later in the
2046    match were failing in some cases. This didn't break the matching; it just
2047    failed to optimize when it could. The way this is done has been re-implemented.
2048    
2049    27. Fixed typo in error message for invalid (?R item (it said "(?p").
2050    
2051    28. Added a new feature that provides some of the functionality that Perl
2052    provides with (?{...}). The facility is termed a "callout". The way it is done
2053    in PCRE is for the caller to provide an optional function, by setting
2054    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
2055    global variable. By default it is unset, which disables all calling out. To get
2056    the function called, the regex must include (?C) at appropriate points. This
2057    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
2058    This provides a means of identifying different callout points. When PCRE
2059    reaches such a point in the regex, if pcre_callout has been set, the external
2060    function is called. It is provided with data in a structure called
2061    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
2062    matching continues; if it returns a non-zero value, the match at the current
2063    point fails. However, backtracking will occur if possible. [This was changed
2064    later and other features added - see item 49 below.]
2065    
2066    29. pcretest is upgraded to test the callout functionality. It provides a
2067    callout function that displays information. By default, it shows the start of
2068    the match and the current position in the text. There are some new data escapes
2069    to vary what happens:
2070    
2071        \C+         in addition, show current contents of captured substrings
2072        \C-         do not supply a callout function
2073        \C!n        return 1 when callout number n is reached
2074        \C!n!m      return 1 when callout number n is reached for the mth time
2075    
2076    30. If pcregrep was called with the -l option and just a single file name, it
2077    output "<stdin>" if a match was found, instead of the file name.
2078    
2079    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
2080    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
2081    pcre_exec(). This saves a malloc/free per call. The default value of
2082    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
2083    when configuring.
2084    
2085    32. The default maximum size of a compiled pattern is 64K. There have been a
2086    few cases of people hitting this limit. The code now uses macros to handle the
2087    storing of links as offsets within the compiled pattern. It defaults to 2-byte
2088    links, but this can be changed to 3 or 4 bytes by --with-link-size when
2089    configuring. Tests 2 and 5 work only with 2-byte links because they output
2090    debugging information about compiled patterns.
2091    
2092    33. Internal code re-arrangements:
2093    
2094    (a) Moved the debugging function for printing out a compiled regex into
2095        its own source file (printint.c) and used #include to pull it into
2096        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
2097        separate copies.
2098    
2099    (b) Defined the list of op-code names for debugging as a macro in
2100        internal.h so that it is next to the definition of the opcodes.
2101    
2102    (c) Defined a table of op-code lengths for simpler skipping along compiled
2103        code. This is again a macro in internal.h so that it is next to the
2104        definition of the opcodes.
2105    
2106    34. Added support for recursive calls to individual subpatterns, along the
2107    lines of Robin Houston's patch (but implemented somewhat differently).
2108    
2109    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
2110    allow it to read and process whole directories in Win32. This code was
2111    contributed by Lionel Fourquaux; it has not been tested by me.
2112    
2113    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
2114    used to name a group. Names consist of alphanumerics and underscores, and must
2115    be unique. Back references use the syntax (?P=name) and recursive calls use
2116    (?P>name) which is a PCRE extension to the Python extension. Groups still have
2117    numbers. The function pcre_fullinfo() can be used after compilation to extract
2118    a name/number map. There are three relevant calls:
2119    
2120      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
2121      PCRE_INFO_NAMECOUNT            yields the number of entries
2122      PCRE_INFO_NAMETABLE            yields a pointer to the map.
2123    
2124    The map is a vector of fixed-size entries. The size of each entry depends on
2125    the length of the longest name used. The first two bytes of each entry are the
2126    group number, most significant byte first. There follows the corresponding
2127    name, zero terminated. The names are in alphabetical order.
2128    
2129    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
2130    case instead of 255. Making it the same both with and without UTF-8 support
2131    means that the same test output works with both.
2132    
2133    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
2134    calling malloc() with a zero argument.
2135    
2136    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
2137    optimization. I've improved things by keeping a bitmap of backreferences with
2138    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
2139    fact referenced, the optimization can be applied. It is unlikely that a
2140    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
2141    the match to follow \n) will appear inside brackets with a number greater than
2142    31, but if it does, any back reference > 31 suppresses the optimization.
2143    
2144    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
2145    of disabling numbered capturing parentheses. Any opening parenthesis that is
2146    not followed by ? behaves as if it were followed by ?: but named parentheses
2147    can still be used for capturing (and they will acquire numbers in the usual
2148    way).
2149    
2150    41. Redesigned the return codes from the match() function into yes/no/error so
2151    that errors can be passed back from deep inside the nested calls. A malloc
2152    failure while inside a recursive subpattern call now causes the
2153    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
2154    
2155    42. It is now possible to set a limit on the number of times the match()
2156    function is called in a call to pcre_exec(). This facility makes it possible to
2157    limit the amount of recursion and backtracking, though not in a directly
2158    obvious way, because the match() function is used in a number of different
2159    circumstances. The count starts from zero for each position in the subject
2160    string (for non-anchored patterns). The default limit is, for compatibility, a
2161    large number, namely 10 000 000. You can change this in two ways:
2162    
2163    (a) When configuring PCRE before making, you can use --with-match-limit=n
2164        to set a default value for the compiled library.
2165    
2166    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
2167        a different value is set. See 45 below.
2168    
2169    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
2170    
2171    43. Added a new function pcre_config(int, void *) to enable run-time extraction
2172    of things that can be changed at compile time. The first argument specifies
2173    what is wanted and the second points to where the information is to be placed.
2174    The current list of available information is:
2175    
2176      PCRE_CONFIG_UTF8
2177    
2178    The output is an integer that is set to one if UTF-8 support is available;
2179    otherwise it is set to zero.
2180    
2181      PCRE_CONFIG_NEWLINE
2182    
2183    The output is an integer that it set to the value of the code that is used for
2184    newline. It is either LF (10) or CR (13).
2185    
2186      PCRE_CONFIG_LINK_SIZE
2187    
2188    The output is an integer that contains the number of bytes used for internal
2189    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
2190    
2191      PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
2192    
2193    The output is an integer that contains the threshold above which the POSIX
2194    interface uses malloc() for output vectors. See item 31 above.
2195    
2196      PCRE_CONFIG_MATCH_LIMIT
2197    
2198    The output is an unsigned integer that contains the default limit of the number
2199    of match() calls in a pcre_exec() execution. See 42 above.
2200    
2201    44. pcretest has been upgraded by the addition of the -C option. This causes it
2202    to extract all the available output from the new pcre_config() function, and to
2203    output it. The program then exits immediately.
2204    
2205    45. A need has arisen to pass over additional data with calls to pcre_exec() in
2206    order to support additional features. One way would have been to define
2207    pcre_exec2() (for example) with extra arguments, but this would not have been
2208    extensible, and would also have required all calls to the original function to
2209    be mapped to the new one. Instead, I have chosen to extend the mechanism that
2210    is used for passing in "extra" data from pcre_study().
2211    
2212    The pcre_extra structure is now exposed and defined in pcre.h. It currently
2213    contains the following fields:
2214    
2215      flags         a bitmap indicating which of the following fields are set
2216      study_data    opaque data from pcre_study()
2217      match_limit   a way of specifying a limit on match() calls for a specific
2218                      call to pcre_exec()
2219      callout_data  data for callouts (see 49 below)
2220    
2221    The flag bits are also defined in pcre.h, and are
2222    
2223      PCRE_EXTRA_STUDY_DATA
2224      PCRE_EXTRA_MATCH_LIMIT
2225      PCRE_EXTRA_CALLOUT_DATA
2226    
2227    The pcre_study() function now returns one of these new pcre_extra blocks, with
2228    the actual study data pointed to by the study_data field, and the
2229    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
2230    before. That is, this change is entirely upwards-compatible and requires no
2231    change to existing code.
2232    
2233    If you want to pass in additional data to pcre_exec(), you can either place it
2234    in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
2235    block.
2236    
2237    46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
2238    data string contains the escape sequence \M, pcretest calls pcre_exec() several
2239    times with different match limits, until it finds the minimum value needed for
2240    pcre_exec() to complete. The value is then output. This can be instructive; for
2241    most simple matches the number is quite small, but for pathological cases it
2242    gets very large very quickly.
2243    
2244    47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
2245    returns the size of the data block pointed to by the study_data field in a
2246    pcre_extra block, that is, the value that was passed as the argument to
2247    pcre_malloc() when PCRE was getting memory in which to place the information
2248    created by pcre_study(). The fourth argument should point to a size_t variable.
2249    pcretest has been extended so that this information is shown after a successful
2250    pcre_study() call when information about the compiled regex is being displayed.
2251    
2252    48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
2253    because what follows is always an absolute path. (Later: it turns out that this
2254    is more than cosmetic for MinGW, because it doesn't like empty path
2255    components.)
2256    
2257    49. Some changes have been made to the callout feature (see 28 above):
2258    
2259    (i)  A callout function now has three choices for what it returns:
2260    
2261           0  =>  success, carry on matching
2262         > 0  =>  failure at this point, but backtrack if possible
2263         < 0  =>  serious error, return this value from pcre_exec()
2264    
2265         Negative values should normally be chosen from the set of PCRE_ERROR_xxx
2266         values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
2267         "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
2268         use by callout functions. It will never be used by PCRE itself.
2269    
2270    (ii) The pcre_extra structure (see 45 above) has a void * field called
2271         callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
2272         pcre_callout_block structure has a field of the same name. The contents of
2273         the field passed in the pcre_extra structure are passed to the callout
2274         function in the corresponding field in the callout block. This makes it
2275         easier to use the same callout-containing regex from multiple threads. For
2276         testing, the pcretest program has a new data escape
2277    
2278           \C*n        pass the number n (may be negative) as callout_data
2279    
2280         If the callout function in pcretest receives a non-zero value as
2281         callout_data, it returns that value.
2282    
2283    50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
2284    there were some redundant $(CFLAGS) in commands that are now specified as
2285    $(LINK), which already includes $(CFLAGS).
2286    
2287    51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
2288    has been compiled with UTF-8 support *and* pcre_compile() has been compiled
2289    with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
2290    one-byte characters throughout. Note that case-insensitive matching applies
2291    only to characters whose values are less than 256. PCRE doesn't support the
2292    notion of cases for higher-valued characters.
2293    
2294    (i)   A character class whose characters are all within 0-255 is handled as
2295          a bit map, and the map is inverted for negative classes. Previously, a
2296          character > 255 always failed to match such a class; however it should
2297          match if the class was a negative one (e.g. [^ab]). This has been fixed.
2298    
2299    (ii)  A negated character class with a single character < 255 is coded as
2300          "not this character" (OP_NOT). This wasn't working properly when the test
2301          character was multibyte, either singly or repeated.
2302    
2303    (iii) Repeats of multibyte characters are now handled correctly in UTF-8
2304          mode, for example: \x{100}{2,3}.
2305    
2306    (iv)  The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
2307          singly or repeated) now correctly test multibyte characters. However,
2308          PCRE doesn't recognize any characters with values greater than 255 as
2309          digits, spaces, or word characters. Such characters always match \D, \S,
2310          and \W, and never match \d, \s, or \w.
2311    
2312    (v)   Classes may now contain characters and character ranges with values
2313          greater than 255. For example: [ab\x{100}-\x{400}].
2314    
2315    (vi)  pcregrep now has a --utf-8 option (synonym -u) which makes it call
2316          PCRE in UTF-8 mode.
2317    
2318    52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
2319    PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
2320    retained for backwards compatibility. (Note that LASTLITERAL is also a byte
2321    value.)
2322    
2323    53. The single man page has become too large. I have therefore split it up into
2324    a number of separate man pages. These also give rise to individual HTML pages;
2325    these are now put in a separate directory, and there is an index.html page that
2326    lists them all. Some hyperlinking between the pages has been installed.
2327    
2328    54. Added convenience functions for handling named capturing parentheses.
2329    
2330    55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2331    aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2332    true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2333    are faulted.
2334    
2335    56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2336    calling configure. These values are used when compiling the dftables.c program
2337    which is run to generate the source of the default character tables. They
2338    default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2339    you will need to set these values.
2340    
2341    57. Updated the building process for Windows DLL, as provided by Fred Cox.
2342    
2343    
2344    Version 3.9 02-Jan-02
2345    ---------------------
2346    
2347    1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2348    
2349    2. If --disable-static was given, the building process failed when trying to
2350    build pcretest and pcregrep. (For some reason it was using libtool to compile
2351    them, which is not right, as they aren't part of the library.)
2352    
2353    
2354    Version 3.8 18-Dec-01
2355    ---------------------
2356    
2357    1. The experimental UTF-8 code was completely screwed up. It was packing the
2358    bytes in the wrong order. How dumb can you get?
2359    
2360    
2361    Version 3.7 29-Oct-01
2362    ---------------------
2363    
2364    1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2365    This caused pcretest, when used on the test data, to segfault. Unfortunately,
2366    this didn't happen under Solaris 8, where I normally test things.
2367    
2368    2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2369    doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2370    isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2371    this fix an hour or so after the initial 3.7 release.)
2372    
2373    
2374    Version 3.6 23-Oct-01
2375    ---------------------
2376    
2377    1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2378    offsets passed as NULL with zero offset count.
2379    
2380    2. The config.guess and config.sub files had not been updated when I moved to
2381    the latest autoconf.
2382    
2383    
2384    Version 3.5 15-Aug-01
2385    ---------------------
2386    
2387    1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2388    had been forgotten.
2389    
2390    2. By using declared but undefined structures, we can avoid using "void"
2391    definitions in pcre.h while keeping the internal definitions of the structures
2392    private.
2393    
2394    3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2395    user point of view, this means that both static and shared libraries are built
2396    by default, but this can be individually controlled. More of the work of
2397    handling this static/shared cases is now inside libtool instead of PCRE's make
2398    file.
2399    
2400    4. The pcretest utility is now installed along with pcregrep because it is
2401    useful for users (to test regexs) and by doing this, it automatically gets
2402    relinked by libtool. The documentation has been turned into a man page, so
2403    there are now .1, .txt, and .html versions in /doc.
2404    
2405    5. Upgrades to pcregrep:
2406       (i)   Added long-form option names like gnu grep.
2407       (ii)  Added --help to list all options with an explanatory phrase.
2408       (iii) Added -r, --recursive to recurse into sub-directories.
2409       (iv)  Added -f, --file to read patterns from a file.
2410    
2411    6. pcre_exec() was referring to its "code" argument before testing that
2412    argument for NULL (and giving an error if it was NULL).
2413    
2414    7. Upgraded Makefile.in to allow for compiling in a different directory from
2415    the source directory.
2416    
2417    8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2418    options bits, the pointer it was passed was to an int instead of to an unsigned
2419    long int. This mattered only on 64-bit systems.
2420    
2421    9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2422    generated) instead of pcre.in, which it its source. Also made the same change
2423    in several of the .c files.
2424    
2425    10. A new release of gcc defines printf() as a macro, which broke pcretest
2426    because it had an ifdef in the middle of a string argument for printf(). Fixed
2427    by using separate calls to printf().
2428    
2429    11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2430    script, to force use of CR or LF instead of \n in the source. On non-Unix
2431    systems, the value can be set in config.h.
2432    
2433    12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2434    absolute limit. Changed the text of the error message to make this clear, and
2435    likewise updated the man page.
2436    
2437    13. The limit of 99 on the number of capturing subpatterns has been removed.
2438    The new limit is 65535, which I hope will not be a "real" limit.
2439    
2440    
2441    Version 3.4 22-Aug-00
2442    ---------------------
2443    
2444    1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2445    
2446    2. Diagnose condition (?(0) as an error instead of crashing on matching.
2447    
2448    
2449    Version 3.3 01-Aug-00
2450    ---------------------
2451    
2452    1. If an octal character was given, but the value was greater than \377, it
2453    was not getting masked to the least significant bits, as documented. This could
2454    lead to crashes in some systems.
2455    
2456    2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
2457    the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
2458    
2459    3. Added the functions pcre_free_substring() and pcre_free_substring_list().
2460    These just pass their arguments on to (pcre_free)(), but they are provided
2461    because some uses of PCRE bind it to non-C systems that can call its functions,
2462    but cannot call free() or pcre_free() directly.
2463    
2464    4. Add "make test" as a synonym for "make check". Corrected some comments in
2465    the Makefile.
2466    
2467    5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
2468    Makefile.
2469    
2470    6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
2471    command called pgrep for grepping around the active processes.
2472    
2473    7. Added the beginnings of support for UTF-8 character strings.
2474    
2475    8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
2476    RANLIB to ./ltconfig so that they are used by libtool. I think these are all
2477    the relevant ones. (AR is not passed because ./ltconfig does its own figuring
2478    out for the ar command.)
2479    
2480    
2481    Version 3.2 12-May-00
2482    ---------------------
2483    
2484    This is purely a bug fixing release.
2485    
2486    1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
2487    of ZA. This was just one example of several cases that could provoke this bug,
2488    which was introduced by change 9 of version 2.00. The code for breaking
2489    infinite loops after an iteration that matches an empty string was't working
2490    correctly.
2491    
2492    2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
2493    when matched against abbab (for example). After matching an empty string, it
2494    wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
2495    caused it to match further down the string than it should.
2496    
2497    3. The code contained an inclusion of sys/types.h. It isn't clear why this
2498    was there because it doesn't seem to be needed, and it causes trouble on some
2499    systems, as it is not a Standard C header. It has been removed.
2500    
2501    4. Made 4 silly changes to the source to avoid stupid compiler warnings that
2502    were reported on the Macintosh. The changes were from
2503    
2504      while ((c = *(++ptr)) != 0 && c != '\n');
2505    to
2506      while ((c = *(++ptr)) != 0 && c != '\n') ;
2507    
2508    Totally extraordinary, but if that's what it takes...
2509    
2510    5. PCRE is being used in one environment where neither memmove() nor bcopy() is
2511    available. Added HAVE_BCOPY and an autoconf test for it; if neither
2512    HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
2513    assumes the way PCRE uses memmove() (always moving upwards).
2514    
2515    6. PCRE is being used in one environment where strchr() is not available. There
2516    was only one use in pcre.c, and writing it out to avoid strchr() probably gives
2517    faster code anyway.
2518    
2519    
2520  Version 3.1 09-Feb-00  Version 3.1 09-Feb-00
2521  ---------------------  ---------------------

Legend:
Removed from v.45  
changed lines
  Added in v.279

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12