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

Diff of /code/trunk/ChangeLog

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

revision 3 by nigel, Sat Feb 24 21:38:01 2007 UTC revision 294 by ph10, Sat Dec 29 16:55:59 2007 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4    Version 7.5 29-Dec-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    14. Added --file-offsets and --line-offsets to pcregrep.
80    
81    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
82        infinitely looping recursion. The bug was that positive lookaheads were not
83        being skipped when checking for a possible empty match (negative lookaheads
84        and both kinds of lookbehind were skipped).
85    
86    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
87        inclusion of <windows.h> to before rather than after the definition of
88        INVALID_FILE_ATTRIBUTES (patch from David Byron).
89    
90    17. Specifying a possessive quantifier with a specific limit for a Unicode
91        character property caused pcre_compile() to compile bad code, which led at
92        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
93        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
94        caused the error; without that there was no problem.
95    
96    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
97    
98    19. Added --enable-pcretest-libreadline.
99    
100    20. In pcrecpp.cc, the variable 'count' was incremented twice in
101        RE::GlobalReplace(). As a result, the number of replacements returned was
102        double what it should be. I have removed one of the increments.
103    
104    21. Several CMake things:
105    
106        (1) Arranged that, when cmake is used on Unix, the libraries end up with
107            the names libpcre and libpcreposix, not just pcre and pcreposix.
108    
109        (2) The above change means that pcretest and pcregrep are now correctly
110            linked with the newly-built libraries, not previously installed ones.
111    
112        (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
113    
114    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
115        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
116        UTF-8 newline character). The key issue is that the pattern starts .*;
117        this means that the match must be either at the beginning, or after a
118        newline. The bug was in the code for advancing after a failed match and
119        checking that the new position followed a newline. It was not taking
120        account of UTF-8 characters correctly.
121    
122    
123    Version 7.4 21-Sep-07
124    ---------------------
125    
126    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
127        means that a class such as [\s] counted as "explicit reference to CR or
128        LF". That isn't really right - the whole point of the change was to try to
129        help when there was an actual mention of one of the two characters. So now
130        the change happens only if \r or \n (or a literal CR or LF) character is
131        encountered.
132    
133    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
134        of both had grown to the point where there were only 3 bits left.
135        Fortunately, there was spare space in the data structure, and so I have
136        moved the internal flags into a new 16-bit field to free up more option
137        bits.
138    
139    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
140        but did not set the internal JCHANGED flag - either of these is enough to
141        control the way the "get" function works - but the PCRE_INFO_JCHANGED
142        facility is supposed to tell if (?J) was ever used, so now (?J) at the
143        start sets both bits.
144    
145    4.  Added options (at build time, compile time, exec time) to change \R from
146        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
147    
148    5.  doc/pcresyntax.html was missing from the distribution.
149    
150    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
151        compatibility, even though it is no longer used.
152    
153    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
154        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
155        windows.h file is present (where different names are used). [This was
156        reversed later after testing - see 16 below.]
157    
158    8.  Changed all #include <config.h> to #include "config.h". There were also
159        some further <pcre.h> cases that I changed to "pcre.h".
160    
161    9.  When pcregrep was used with the --colour option, it missed the line ending
162        sequence off the lines that it output.
163    
164    10. It was pointed out to me that arrays of string pointers cause lots of
165        relocations when a shared library is dynamically loaded. A technique of
166        using a single long string with a table of offsets can drastically reduce
167        these. I have refactored PCRE in four places to do this. The result is
168        dramatic:
169    
170          Originally:                          290
171          After changing UCP table:            187
172          After changing error message table:   43
173          After changing table of "verbs"       36
174          After changing table of Posix names   22
175    
176        Thanks to the folks working on Gregex for glib for this insight.
177    
178    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
179        unicode-properties was also set.
180    
181    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
182    
183    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
184        checked only for CRLF.
185    
186    14. Added casts to pcretest.c to avoid compiler warnings.
187    
188    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
189    
190    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
191        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
192        entirely. This removes changes made in 7 above.
193    
194    17. The CMake files have been updated, and there is now more information about
195        building with CMake in the NON-UNIX-USE document.
196    
197    
198    Version 7.3 28-Aug-07
199    ---------------------
200    
201     1. In the rejigging of the build system that eventually resulted in 7.1, the
202        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
203        brackets there is not right, since it causes compilers to look for an
204        installed pcre.h, not the version that is in the source that is being
205        compiled (which of course may be different). I have changed it back to:
206    
207          #include "pcre.h"
208    
209        I have a vague recollection that the change was concerned with compiling in
210        different directories, but in the new build system, that is taken care of
211        by the VPATH setting the Makefile.
212    
213     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
214        when the subject happened to end in the byte 0x85 (e.g. if the last
215        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
216        characters but of course it shouldn't be taken as a newline when it is part
217        of another character. The bug was that, for an unlimited repeat of . in
218        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
219        characters when looking for a newline.
220    
221     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
222    
223     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
224        in debug output.
225    
226     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
227        long printing in the pcrecpp unittest when running under MinGW.
228    
229     6. ESC_K was left out of the EBCDIC table.
230    
231     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
232        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
233        limit also applies to "virtual nesting" when a pattern is recursive, and in
234        this case 1000 isn't so big. I have been able to remove this limit at the
235        expense of backing off one optimization in certain circumstances. Normally,
236        when pcre_exec() would call its internal match() function recursively and
237        immediately return the result unconditionally, it uses a "tail recursion"
238        feature to save stack. However, when a subpattern that can match an empty
239        string has an unlimited repetition quantifier, it no longer makes this
240        optimization. That gives it a stack frame in which to save the data for
241        checking that an empty string has been matched. Previously this was taken
242        from the 1000-entry workspace that had been reserved. So now there is no
243        explicit limit, but more stack is used.
244    
245     8. Applied Daniel's patches to solve problems with the import/export magic
246        syntax that is required for Windows, and which was going wrong for the
247        pcreposix and pcrecpp parts of the library. These were overlooked when this
248        problem was solved for the main library.
249    
250     9. There were some crude static tests to avoid integer overflow when computing
251        the size of patterns that contain repeated groups with explicit upper
252        limits. As the maximum quantifier is 65535, the maximum group length was
253        set at 30,000 so that the product of these two numbers did not overflow a
254        32-bit integer. However, it turns out that people want to use groups that
255        are longer than 30,000 bytes (though not repeat them that many times).
256        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
257        made it possible to implement the integer overflow checks in a much more
258        dynamic way, which I have now done. The artificial limitation on group
259        length has been removed - we now have only the limit on the total length of
260        the compiled pattern, which depends on the LINK_SIZE setting.
261    
262    10. Fixed a bug in the documentation for get/copy named substring when
263        duplicate names are permitted. If none of the named substrings are set, the
264        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
265        empty string.
266    
267    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
268        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
269        because the ] is interpreted as the first data character and the
270        terminating ] is not found. PCRE has been made compatible with Perl in this
271        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
272        cause memory overwriting.
273    
274    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
275        string has been matched (to stop an infinite loop). It was not recognizing
276        a conditional subpattern that could match an empty string if that
277        subpattern was within another subpattern. For example, it looped when
278        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
279        condition was not nested. This bug has been fixed.
280    
281    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
282        past the start of the subject in the presence of bytes with the top bit
283        set, for example "\x8aBCD".
284    
285    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
286        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
287    
288    14. Optimized (?!) to (*FAIL).
289    
290    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
291        This restricts code points to be within the range 0 to 0x10FFFF, excluding
292        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
293        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
294        does: it's just the validity check that is more restrictive.
295    
296    16. Inserted checks for integer overflows during escape sequence (backslash)
297        processing, and also fixed erroneous offset values for syntax errors during
298        backslash processing.
299    
300    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
301        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
302    
303    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
304        caused an overrun.
305    
306    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
307        something other than just ASCII characters) inside a group that had an
308        unlimited repeat caused a loop at compile time (while checking to see
309        whether the group could match an empty string).
310    
311    20. Debugging a pattern containing \p or \P could cause a crash. For example,
312        [\P{Any}] did so. (Error in the code for printing property names.)
313    
314    21. An orphan \E inside a character class could cause a crash.
315    
316    22. A repeated capturing bracket such as (A)? could cause a wild memory
317        reference during compilation.
318    
319    23. There are several functions in pcre_compile() that scan along a compiled
320        expression for various reasons (e.g. to see if it's fixed length for look
321        behind). There were bugs in these functions when a repeated \p or \P was
322        present in the pattern. These operators have additional parameters compared
323        with \d, etc, and these were not being taken into account when moving along
324        the compiled data. Specifically:
325    
326        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
327            length.
328    
329        (b) An item such as \pL+ within a repeated group could cause crashes or
330            loops.
331    
332        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
333            "reference to non-existent subpattern" error.
334    
335        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
336    
337    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
338        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
339    
340    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
341    
342    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
343        character were causing crashes (broken optimization).
344    
345    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
346        \p or \P) caused a compile-time loop.
347    
348    28. More problems have arisen in unanchored patterns when CRLF is a valid line
349        break. For example, the unstudied pattern [\r\n]A does not match the string
350        "\r\nA" because change 7.0/46 below moves the current point on by two
351        characters after failing to match at the start. However, the pattern \nA
352        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
353        the same is true. There doesn't seem any very clean way out of this, but
354        what I have chosen to do makes the common cases work: PCRE now takes note
355        of whether there can be an explicit match for \r or \n anywhere in the
356        pattern, and if so, 7.0/46 no longer applies. As part of this change,
357        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
358        pattern has explicit CR or LF references.
359    
360    29. Added (*CR) etc for changing newline setting at start of pattern.
361    
362    
363    Version 7.2 19-Jun-07
364    ---------------------
365    
366     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
367        which is apparently normally available under Windows.
368    
369     2. Re-jig the pcregrep tests with different newline settings in an attempt
370        to make them independent of the local environment's newline setting.
371    
372     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
373    
374     4. Some of the "internals" tests were previously cut out when the link size
375        was not 2, because the output contained actual offsets. The recent new
376        "Z" feature of pcretest means that these can be cut out, making the tests
377        usable with all link sizes.
378    
379     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
380        stack recursion. This gives a massive performance boost under BSD, but just
381        a small improvement under Linux. However, it saves one field in the frame
382        in all cases.
383    
384     6. Added more features from the forthcoming Perl 5.10:
385    
386        (a) (?-n) (where n is a string of digits) is a relative subroutine or
387            recursion call. It refers to the nth most recently opened parentheses.
388    
389        (b) (?+n) is also a relative subroutine call; it refers to the nth next
390            to be opened parentheses.
391    
392        (c) Conditions that refer to capturing parentheses can be specified
393            relatively, for example, (?(-2)... or (?(+3)...
394    
395        (d) \K resets the start of the current match so that everything before
396            is not part of it.
397    
398        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
399    
400        (f) \g{name} is another synonym - part of Perl 5.10's unification of
401            reference syntax.
402    
403        (g) (?| introduces a group in which the numbering of parentheses in each
404            alternative starts with the same number.
405    
406        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
407    
408     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
409        PCRE_INFO_JCHANGED.
410    
411     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
412        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
413        for detecting groups that can match an empty string.
414    
415     9. A pattern with a very large number of alternatives (more than several
416        hundred) was running out of internal workspace during the pre-compile
417        phase, where pcre_compile() figures out how much memory will be needed. A
418        bit of new cunning has reduced the workspace needed for groups with
419        alternatives. The 1000-alternative test pattern now uses 12 bytes of
420        workspace instead of running out of the 4096 that are available.
421    
422    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
423    
424    11. Applied patch from Google to remove an optimization that didn't quite work.
425        The report of the bug said:
426    
427          pcrecpp::RE("a*").FullMatch("aaa") matches, while
428          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
429          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
430    
431    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
432        it matched the wrong number of bytes.
433    
434    
435    Version 7.1 24-Apr-07
436    ---------------------
437    
438     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
439        that is more "standard", making use of automake and other Autotools. There
440        is some re-arrangement of the files and adjustment of comments consequent
441        on this.
442    
443     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
444        for recursive directory scanning broke on some systems because the files
445        are not scanned in any specific order and on different systems the order
446        was different. A call to "sort" has been inserted into RunGrepTest for the
447        approprate test as a short-term fix. In the longer term there may be an
448        alternative.
449    
450     3. I had an email from Eric Raymond about problems translating some of PCRE's
451        man pages to HTML (despite the fact that I distribute HTML pages, some
452        people do their own conversions for various reasons). The problems
453        concerned the use of low-level troff macros .br and .in. I have therefore
454        removed all such uses from the man pages (some were redundant, some could
455        be replaced by .nf/.fi pairs). The 132html script that I use to generate
456        HTML has been updated to handle .nf/.fi and to complain if it encounters
457        .br or .in.
458    
459     4. Updated comments in configure.ac that get placed in config.h.in and also
460        arranged for config.h to be included in the distribution, with the name
461        config.h.generic, for the benefit of those who have to compile without
462        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
463    
464     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
465        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
466        makevp.bat for the current PCRE, using the additional files
467        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
468    
469     6. A Windows user reported a minor discrepancy with test 2, which turned out
470        to be caused by a trailing space on an input line that had got lost in his
471        copy. The trailing space was an accident, so I've just removed it.
472    
473     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
474        that is needed.
475    
476     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
477        as "const" (a) because they are and (b) because it helps the PHP
478        maintainers who have recently made a script to detect big data structures
479        in the php code that should be moved to the .rodata section. I remembered
480        to update Builducptable as well, so it won't revert if ucptable.h is ever
481        re-created.
482    
483     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
484        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
485        order to be able to cut out the UTF-8 tables in the latter when UTF-8
486        support is not required. This saves 1.5-2K of code, which is important in
487        some applications.
488    
489        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
490        so as not to refer to the tables, even though these functions will never be
491        called when UTF-8 support is disabled. Otherwise there are problems with a
492        shared library.
493    
494    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
495    
496        (a) It was defining its arguments as char * instead of void *.
497    
498        (b) It was assuming that all moves were upwards in memory; this was true
499            a long time ago when I wrote it, but is no longer the case.
500    
501        The emulated memove() is provided for those environments that have neither
502        memmove() nor bcopy(). I didn't think anyone used it these days, but that
503        is clearly not the case, as these two bugs were recently reported.
504    
505    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
506        and Detrail to create the HTML documentation, the .txt form of the man
507        pages, and it removes trailing spaces from listed files. It also creates
508        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
509        case, it wraps all the #defines with #ifndefs. This script should be run
510        before "make dist".
511    
512    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
513        with Unicode property support.
514    
515        (a) For a maximizing quantifier, if the two different cases of the
516            character were of different lengths in their UTF-8 codings (there are
517            some cases like this - I found 11), and the matching function had to
518            back up over a mixture of the two cases, it incorrectly assumed they
519            were both the same length.
520    
521        (b) When PCRE was configured to use the heap rather than the stack for
522            recursion during matching, it was not correctly preserving the data for
523            the other case of a UTF-8 character when checking ahead for a match
524            while processing a minimizing repeat. If the check also involved
525            matching a wide character, but failed, corruption could cause an
526            erroneous result when trying to check for a repeat of the original
527            character.
528    
529    13. Some tidying changes to the testing mechanism:
530    
531        (a) The RunTest script now detects the internal link size and whether there
532            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
533            values substituted by "configure". (The RunGrepTest script already did
534            this for UTF-8.) The configure.ac script no longer substitutes the
535            relevant variables.
536    
537        (b) The debugging options /B and /D in pcretest show the compiled bytecode
538            with length and offset values. This means that the output is different
539            for different internal link sizes. Test 2 is skipped for link sizes
540            other than 2 because of this, bypassing the problem. Unfortunately,
541            there was also a test in test 3 (the locale tests) that used /B and
542            failed for link sizes other than 2. Rather than cut the whole test out,
543            I have added a new /Z option to pcretest that replaces the length and
544            offset values with spaces. This is now used to make test 3 independent
545            of link size. (Test 2 will be tidied up later.)
546    
547    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
548        segmentation fault instead of returning the appropriate error message.
549    
550    15. In multiline mode when the newline sequence was set to "any", the pattern
551        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
552        This doesn't seem right; it now treats the CRLF combination as the line
553        ending, and so does not match in that case. It's only a pattern such as ^$
554        that would hit this one: something like ^ABC$ would have failed after \r
555        and then tried again after \r\n.
556    
557    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
558        in an attempt to make files that differ only in their line terminators
559        compare equal. This works on Linux.
560    
561    17. Under certain error circumstances pcregrep might try to free random memory
562        as it exited. This is now fixed, thanks to valgrind.
563    
564    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
565        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
566        was because its rules for how to advance for /g after matching an empty
567        string at the end of a line did not allow for this case. They now check for
568        it specially.
569    
570    20. pcretest is supposed to handle patterns and data of any length, by
571        extending its buffers when necessary. It was getting this wrong when the
572        buffer for a data line had to be extended.
573    
574    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
575        CRLF as a newline sequence.
576    
577    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
578        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
579        I have nevertheless tidied it up.
580    
581    23. Added some casts to kill warnings from HP-UX ia64 compiler.
582    
583    24. Added a man page for pcre-config.
584    
585    
586    Version 7.0 19-Dec-06
587    ---------------------
588    
589     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
590        moving to gcc 4.1.1.
591    
592     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
593        sys/time.h, which is documented as needed for this function. It doesn't
594        seem to matter on Linux, but it showed up on some releases of OS X.
595    
596     3. It seems that there are systems where bytes whose values are greater than
597        127 match isprint() in the "C" locale. The "C" locale should be the
598        default when a C program starts up. In most systems, only ASCII printing
599        characters match isprint(). This difference caused the output from pcretest
600        to vary, making some of the tests fail. I have changed pcretest so that:
601    
602        (a) When it is outputting text in the compiled version of a pattern, bytes
603            other than 32-126 are always shown as hex escapes.
604    
605        (b) When it is outputting text that is a matched part of a subject string,
606            it does the same, unless a different locale has been set for the match
607            (using the /L modifier). In this case, it uses isprint() to decide.
608    
609     4. Fixed a major bug that caused incorrect computation of the amount of memory
610        required for a compiled pattern when options that changed within the
611        pattern affected the logic of the preliminary scan that determines the
612        length. The relevant options are -x, and -i in UTF-8 mode. The result was
613        that the computed length was too small. The symptoms of this bug were
614        either the PCRE error "internal error: code overflow" from pcre_compile(),
615        or a glibc crash with a message such as "pcretest: free(): invalid next
616        size (fast)". Examples of patterns that provoked this bug (shown in
617        pcretest format) are:
618    
619          /(?-x: )/x
620          /(?x)(?-x: \s*#\s*)/
621          /((?i)[\x{c0}])/8
622          /(?i:[\x{c0}])/8
623    
624        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
625        is now done differently.
626    
627     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
628        wrapper classes; (b) implement a new function in the C++ scanner that is
629        more efficient than the old way of doing things because it avoids levels of
630        recursion in the regex matching; (c) add a paragraph to the documentation
631        for the FullMatch() function.
632    
633     6. The escape sequence \n was being treated as whatever was defined as
634        "newline". Not only was this contrary to the documentation, which states
635        that \n is character 10 (hex 0A), but it also went horribly wrong when
636        "newline" was defined as CRLF. This has been fixed.
637    
638     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
639        was being set to -1 for the "end of line" case (supposedly a value that no
640        character can have). Though this value is never used (the check for end of
641        line is "zero bytes in current character"), it caused compiler complaints.
642        I've changed it to 0xffffffff.
643    
644     8. In pcre_version.c, the version string was being built by a sequence of
645        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
646        string (as it is for production releases) called a macro with an empty
647        argument. The C standard says the result of this is undefined. The gcc
648        compiler treats it as an empty string (which was what was wanted) but it is
649        reported that Visual C gives an error. The source has been hacked around to
650        avoid this problem.
651    
652     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
653        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
654        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
655        of them did).
656    
657    10. Originally, pcretest opened its input and output without "b"; then I was
658        told that "b" was needed in some environments, so it was added for release
659        5.0 to both the input and output. (It makes no difference on Unix-like
660        systems.) Later I was told that it is wrong for the input on Windows. I've
661        now abstracted the modes into two macros, to make it easier to fiddle with
662        them, and removed "b" from the input mode under Windows.
663    
664    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
665    
666    12. Added -help and --help to pcretest as an official way of being reminded
667        of the options.
668    
669    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
670        and pcrecpp.cc because they annoy compilers at high warning levels.
671    
672    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
673    
674    15. Fixed an occurrence of == in configure.ac that should have been = (shell
675        scripts are not C programs :-) and which was not noticed because it works
676        on Linux.
677    
678    16. pcretest is supposed to handle any length of pattern and data line (as one
679        line or as a continued sequence of lines) by extending its input buffer if
680        necessary. This feature was broken for very long pattern lines, leading to
681        a string of junk being passed to pcre_compile() if the pattern was longer
682        than about 50K.
683    
684    17. I have done a major re-factoring of the way pcre_compile() computes the
685        amount of memory needed for a compiled pattern. Previously, there was code
686        that made a preliminary scan of the pattern in order to do this. That was
687        OK when PCRE was new, but as the facilities have expanded, it has become
688        harder and harder to keep it in step with the real compile phase, and there
689        have been a number of bugs (see for example, 4 above). I have now found a
690        cunning way of running the real compile function in a "fake" mode that
691        enables it to compute how much memory it would need, while actually only
692        ever using a few hundred bytes of working memory and without too many
693        tests of the mode. This should make future maintenance and development
694        easier. A side effect of this work is that the limit of 200 on the nesting
695        depth of parentheses has been removed (though this was never a serious
696        limitation, I suspect). However, there is a downside: pcre_compile() now
697        runs more slowly than before (30% or more, depending on the pattern). I
698        hope this isn't a big issue. There is no effect on runtime performance.
699    
700    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
701        newline (only possible for the last line of a file) and it was a
702        pattern that set a locale (followed by /Lsomething), pcretest crashed.
703    
704    19. Added additional timing features to pcretest. (1) The -tm option now times
705        matching only, not compiling. (2) Both -t and -tm can be followed, as a
706        separate command line item, by a number that specifies the number of
707        repeats to use when timing. The default is 50000; this gives better
708        precision, but takes uncomfortably long for very large patterns.
709    
710    20. Extended pcre_study() to be more clever in cases where a branch of a
711        subpattern has no definite first character. For example, (a*|b*)[cd] would
712        previously give no result from pcre_study(). Now it recognizes that the
713        first character must be a, b, c, or d.
714    
715    21. There was an incorrect error "recursive call could loop indefinitely" if
716        a subpattern (or the entire pattern) that was being tested for matching an
717        empty string contained only one non-empty item after a nested subpattern.
718        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
719        incorrectly, because the \d was being skipped in the check.
720    
721    22. The pcretest program now has a new pattern option /B and a command line
722        option -b, which is equivalent to adding /B to every pattern. This causes
723        it to show the compiled bytecode, without the additional information that
724        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
725        is the same as /B/I).
726    
727    23. A new optimization is now able automatically to treat some sequences such
728        as a*b as a*+b. More specifically, if something simple (such as a character
729        or a simple class like \d) has an unlimited quantifier, and is followed by
730        something that cannot possibly match the quantified thing, the quantifier
731        is automatically "possessified".
732    
733    24. A recursive reference to a subpattern whose number was greater than 39
734        went wrong under certain circumstances in UTF-8 mode. This bug could also
735        have affected the operation of pcre_study().
736    
737    25. Realized that a little bit of performance could be had by replacing
738        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
739    
740    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
741    
742    27. Possessive quantifiers such as a++ were previously implemented by turning
743        them into atomic groups such as ($>a+). Now they have their own opcodes,
744        which improves performance. This includes the automatically created ones
745        from 23 above.
746    
747    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
748        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
749        the first matched character to be a colon. This applied both to named and
750        numbered groups.
751    
752    29. The ucpinternal.h header file was missing its idempotency #ifdef.
753    
754    30. I was sent a "project" file called libpcre.a.dev which I understand makes
755        building PCRE on Windows easier, so I have included it in the distribution.
756    
757    31. There is now a check in pcretest against a ridiculously large number being
758        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
759        loop, the loop is abandoned.
760    
761    32. Forward references to subpatterns in conditions such as (?(2)...) where
762        subpattern 2 is defined later cause pcre_compile() to search forwards in
763        the pattern for the relevant set of parentheses. This search went wrong
764        when there were unescaped parentheses in a character class, parentheses
765        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
766    
767    33. "Subroutine" calls and backreferences were previously restricted to
768        referencing subpatterns earlier in the regex. This restriction has now
769        been removed.
770    
771    34. Added a number of extra features that are going to be in Perl 5.10. On the
772        whole, these are just syntactic alternatives for features that PCRE had
773        previously implemented using the Python syntax or my own invention. The
774        other formats are all retained for compatibility.
775    
776        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
777            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
778            also .NET compatible.
779    
780        (b) A recursion or subroutine call to a named group can now be defined as
781            (?&name) as well as (?P>name).
782    
783        (c) A backreference to a named group can now be defined as \k<name> or
784            \k'name' as well as (?P=name). The new forms, as well as being in Perl
785            5.10, are also .NET compatible.
786    
787        (d) A conditional reference to a named group can now use the syntax
788            (?(<name>) or (?('name') as well as (?(name).
789    
790        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
791            groups (named and numbered) that are never evaluated inline, but can be
792            called as "subroutines" from elsewhere. In effect, the DEFINE condition
793            is always false. There may be only one alternative in such a group.
794    
795        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
796            as the simple (?(R). The condition is true only if the most recent
797            recursion is that of the given number or name. It does not search out
798            through the entire recursion stack.
799    
800        (g) The escape \gN or \g{N} has been added, where N is a positive or
801            negative number, specifying an absolute or relative reference.
802    
803    35. Tidied to get rid of some further signed/unsigned compiler warnings and
804        some "unreachable code" warnings.
805    
806    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
807        things, this adds five new scripts.
808    
809    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
810        There were also incompatibilities regarding the handling of \Q..\E inside
811        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
812        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
813    
814    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
815        matches an empty string, and forcibly breaks the loop. There were bugs in
816        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
817        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
818        separate and independent bugs (that affected different cases) have been
819        fixed.
820    
821    39. Refactored the code to abolish the use of different opcodes for small
822        capturing bracket numbers. This is a tidy that I avoided doing when I
823        removed the limit on the number of capturing brackets for 3.5 back in 2001.
824        The new approach is not only tidier, it makes it possible to reduce the
825        memory needed to fix the previous bug (38).
826    
827    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
828        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
829        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
830        mode.
831    
832    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
833        report.
834    
835    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
836        copy construction and assignment in the C++ wrapper.
837    
838    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
839        couple of bugs that could have given wrong results in the "--newline=crlf"
840        case.
841    
842    44. Added a number of casts and did some reorganization of signed/unsigned int
843        variables following suggestions from Dair Grant. Also renamed the variable
844        "this" as "item" because it is a C++ keyword.
845    
846    45. Arranged for dftables to add
847    
848          #include "pcre_internal.h"
849    
850        to pcre_chartables.c because without it, gcc 4.x may remove the array
851        definition from the final binary if PCRE is built into a static library and
852        dead code stripping is activated.
853    
854    46. For an unanchored pattern, if a match attempt fails at the start of a
855        newline sequence, and the newline setting is CRLF or ANY, and the next two
856        characters are CRLF, advance by two characters instead of one.
857    
858    
859    Version 6.7 04-Jul-06
860    ---------------------
861    
862     1. In order to handle tests when input lines are enormously long, pcretest has
863        been re-factored so that it automatically extends its buffers when
864        necessary. The code is crude, but this _is_ just a test program. The
865        default size has been increased from 32K to 50K.
866    
867     2. The code in pcre_study() was using the value of the re argument before
868        testing it for NULL. (Of course, in any sensible call of the function, it
869        won't be NULL.)
870    
871     3. The memmove() emulation function in pcre_internal.h, which is used on
872        systems that lack both memmove() and bcopy() - that is, hardly ever -
873        was missing a "static" storage class specifier.
874    
875     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
876        containing an extended class (one that cannot be represented by a bitmap
877        because it contains high-valued characters or Unicode property items, e.g.
878        [\pZ]). Almost always one would set UTF-8 mode when processing such a
879        pattern, but PCRE should not loop if you do not (it no longer does).
880        [Detail: two cases were found: (a) a repeated subpattern containing an
881        extended class; (b) a recursive reference to a subpattern that followed a
882        previous extended class. It wasn't skipping over the extended class
883        correctly when UTF-8 mode was not set.]
884    
885     5. A negated single-character class was not being recognized as fixed-length
886        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
887        compile error "lookbehind assertion is not fixed length".
888    
889     6. The RunPerlTest auxiliary script was showing an unexpected difference
890        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
891        write a Perl script that can interpret lines of an input file either as
892        byte characters or as UTF-8, which is what "perltest" was being required to
893        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
894        can't do is switch easily at run time between having the "use utf8;" pragma
895        or not. In the end, I fudged it by using the RunPerlTest script to insert
896        "use utf8;" explicitly for the UTF-8 tests.
897    
898     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
899        the end of the subject string, contrary to the documentation and to what
900        Perl does. This was true of both matching functions. Now it matches only at
901        the start of the subject and immediately after *internal* newlines.
902    
903     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
904        a pointer to an int instead of a pointer to an unsigned long int. This
905        caused problems on 64-bit systems.
906    
907     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
908        instance of the 'standard' template library not being so standard".
909    
910    10. There was no check on the number of named subpatterns nor the maximum
911        length of a subpattern name. The product of these values is used to compute
912        the size of the memory block for a compiled pattern. By supplying a very
913        long subpattern name and a large number of named subpatterns, the size
914        computation could be caused to overflow. This is now prevented by limiting
915        the length of names to 32 characters, and the number of named subpatterns
916        to 10,000.
917    
918    11. Subpatterns that are repeated with specific counts have to be replicated in
919        the compiled pattern. The size of memory for this was computed from the
920        length of the subpattern and the repeat count. The latter is limited to
921        65535, but there was no limit on the former, meaning that integer overflow
922        could in principle occur. The compiled length of a repeated subpattern is
923        now limited to 30,000 bytes in order to prevent this.
924    
925    12. Added the optional facility to have named substrings with the same name.
926    
927    13. Added the ability to use a named substring as a condition, using the
928        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
929        are numbers (not recommended). Forward references are permitted.
930    
931    14. Added forward references in named backreferences (if you see what I mean).
932    
933    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
934        pattern could run off the end of the subject. For example, the pattern
935        "(?s)(.{1,5})"8 did this with the subject "ab".
936    
937    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
938        PCRE_CASELESS was set when matching characters that were quantified with ?
939        or *.
940    
941    17. A character class other than a single negated character that had a minimum
942        but no maximum quantifier - for example [ab]{6,} - was not handled
943        correctly by pce_dfa_exec(). It would match only one character.
944    
945    18. A valid (though odd) pattern that looked like a POSIX character
946        class but used an invalid character after [ (for example [[,abc,]]) caused
947        pcre_compile() to give the error "Failed: internal error: code overflow" or
948        in some cases to crash with a glibc free() error. This could even happen if
949        the pattern terminated after [[ but there just happened to be a sequence of
950        letters, a binary zero, and a closing ] in the memory that followed.
951    
952    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
953        over the years. Originally (before any Unicode support), just the bottom 8
954        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
955        output from "man perlunicode" includes this:
956    
957          The regular expression compiler produces polymorphic opcodes.  That
958          is, the pattern adapts to the data and automatically switches to
959          the Unicode character scheme when presented with Unicode data--or
960          instead uses a traditional byte scheme when presented with byte
961          data.
962    
963        Sadly, a wide octal escape does not cause a switch, and in a string with
964        no other multibyte characters, these octal escapes are treated as before.
965        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
966        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
967        Unicode string.
968    
969        I have not perpetrated such confusion in PCRE. Up till now, it took just
970        the bottom 8 bits, as in old Perl. I have now made octal escapes with
971        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
972        translate to the appropriate multibyte character.
973    
974    29. Applied some refactoring to reduce the number of warnings from Microsoft
975        and Borland compilers. This has included removing the fudge introduced
976        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
977        a warning about an unused variable.
978    
979    21. PCRE has not included VT (character 0x0b) in the set of whitespace
980        characters since release 4.0, because Perl (from release 5.004) does not.
981        [Or at least, is documented not to: some releases seem to be in conflict
982        with the documentation.] However, when a pattern was studied with
983        pcre_study() and all its branches started with \s, PCRE still included VT
984        as a possible starting character. Of course, this did no harm; it just
985        caused an unnecessary match attempt.
986    
987    22. Removed a now-redundant internal flag bit that recorded the fact that case
988        dependency changed within the pattern. This was once needed for "required
989        byte" processing, but is no longer used. This recovers a now-scarce options
990        bit. Also moved the least significant internal flag bit to the most-
991        significant bit of the word, which was not previously used (hangover from
992        the days when it was an int rather than a uint) to free up another bit for
993        the future.
994    
995    23. Added support for CRLF line endings as well as CR and LF. As well as the
996        default being selectable at build time, it can now be changed at runtime
997        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
998        specify that it is scanning data with non-default line endings.
999    
1000    24. Changed the definition of CXXLINK to make it agree with the definition of
1001        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1002    
1003    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1004        recursions. This makes a big different to stack usage for some patterns.
1005    
1006    26. If a subpattern containing a named recursion or subroutine reference such
1007        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1008        the space required for the compiled pattern went wrong and gave too small a
1009        value. Depending on the environment, this could lead to "Failed: internal
1010        error: code overflow at offset 49" or "glibc detected double free or
1011        corruption" errors.
1012    
1013    27. Applied patches from Google (a) to support the new newline modes and (b) to
1014        advance over multibyte UTF-8 characters in GlobalReplace.
1015    
1016    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1017        difference for some implementation of PCRE in some Windows version.
1018    
1019    29. Added some extra testing facilities to pcretest:
1020    
1021        \q<number>   in a data line sets the "match limit" value
1022        \Q<number>   in a data line sets the "match recursion limt" value
1023        -S <number>  sets the stack size, where <number> is in megabytes
1024    
1025        The -S option isn't available for Windows.
1026    
1027    
1028    Version 6.6 06-Feb-06
1029    ---------------------
1030    
1031     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1032        in pcreposix.h. I have copied the definition from pcre.h.
1033    
1034     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1035        because pcre.h is no longer a built file.
1036    
1037     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1038        not normally included in the compiled code.
1039    
1040    
1041    Version 6.5 01-Feb-06
1042    ---------------------
1043    
1044     1. When using the partial match feature with pcre_dfa_exec(), it was not
1045        anchoring the second and subsequent partial matches at the new starting
1046        point. This could lead to incorrect results. For example, with the pattern
1047        /1234/, partially matching against "123" and then "a4" gave a match.
1048    
1049     2. Changes to pcregrep:
1050    
1051        (a) All non-match returns from pcre_exec() were being treated as failures
1052            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1053            error message is output. Some extra information is given for the
1054            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
1055            probably the only errors that are likely to be caused by users (by
1056            specifying a regex that has nested indefinite repeats, for instance).
1057            If there are more than 20 of these errors, pcregrep is abandoned.
1058    
1059        (b) A binary zero was treated as data while matching, but terminated the
1060            output line if it was written out. This has been fixed: binary zeroes
1061            are now no different to any other data bytes.
1062    
1063        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1064            used to set a locale for matching. The --locale=xxxx long option has
1065            been added (no short equivalent) to specify a locale explicitly on the
1066            pcregrep command, overriding the environment variables.
1067    
1068        (d) When -B was used with -n, some line numbers in the output were one less
1069            than they should have been.
1070    
1071        (e) Added the -o (--only-matching) option.
1072    
1073        (f) If -A or -C was used with -c (count only), some lines of context were
1074            accidentally printed for the final match.
1075    
1076        (g) Added the -H (--with-filename) option.
1077    
1078        (h) The combination of options -rh failed to suppress file names for files
1079            that were found from directory arguments.
1080    
1081        (i) Added the -D (--devices) and -d (--directories) options.
1082    
1083        (j) Added the -F (--fixed-strings) option.
1084    
1085        (k) Allow "-" to be used as a file name for -f as well as for a data file.
1086    
1087        (l) Added the --colo(u)r option.
1088    
1089        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1090            is not present by default.
1091    
1092     3. A nasty bug was discovered in the handling of recursive patterns, that is,
1093        items such as (?R) or (?1), when the recursion could match a number of
1094        alternatives. If it matched one of the alternatives, but subsequently,
1095        outside the recursion, there was a failure, the code tried to back up into
1096        the recursion. However, because of the way PCRE is implemented, this is not
1097        possible, and the result was an incorrect result from the match.
1098    
1099        In order to prevent this happening, the specification of recursion has
1100        been changed so that all such subpatterns are automatically treated as
1101        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1102    
1103     4. I had overlooked the fact that, in some locales, there are characters for
1104        which isalpha() is true but neither isupper() nor islower() are true. In
1105        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1106        and ordfeminine) are like this. This affected the treatment of \w and \W
1107        when they appeared in character classes, but not when they appeared outside
1108        a character class. The bit map for "word" characters is now created
1109        separately from the results of isalnum() instead of just taking it from the
1110        upper, lower, and digit maps. (Plus the underscore character, of course.)
1111    
1112     5. The above bug also affected the handling of POSIX character classes such as
1113        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1114        permanent tables. Instead, the bit maps for such a class were previously
1115        created as the appropriate unions of the upper, lower, and digit bitmaps.
1116        Now they are created by subtraction from the [[:word:]] class, which has
1117        its own bitmap.
1118    
1119     6. The [[:blank:]] character class matches horizontal, but not vertical space.
1120        It is created by subtracting the vertical space characters (\x09, \x0a,
1121        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1122        subtraction was done in the overall bitmap for a character class, meaning
1123        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1124        be recognized. This bug has been fixed.
1125    
1126     7. Patches from the folks at Google:
1127    
1128          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1129          real life, but is still worth protecting against".
1130    
1131          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1132          regular expressions".
1133    
1134          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1135          have it.
1136    
1137          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1138          "configure" and the latter not, in order to fix a problem somebody had
1139          with compiling the Arg class on HP-UX.
1140    
1141          (e) Improve the error-handling of the C++ wrapper a little bit.
1142    
1143          (f) New tests for checking recursion limiting.
1144    
1145     8. The pcre_memmove() function, which is used only if the environment does not
1146        have a standard memmove() function (and is therefore rarely compiled),
1147        contained two bugs: (a) use of int instead of size_t, and (b) it was not
1148        returning a result (though PCRE never actually uses the result).
1149    
1150     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1151        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1152        returned instead of calling malloc() with an overflowing number that would
1153        most likely cause subsequent chaos.
1154    
1155    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1156    
1157    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1158        with this option is matched, the nmatch and pmatch options of regexec() are
1159        ignored.
1160    
1161    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1162        provided in case anyone wants to the the POSIX interface with UTF-8
1163        strings.
1164    
1165    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1166        C++ linking (needed for some HP-UX environments).
1167    
1168    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1169        (unused parameter) and in the pcre_printint() function (omitted "default"
1170        switch label when the default is to do nothing).
1171    
1172    15. Added some code to make it possible, when PCRE is compiled as a C++
1173        library, to replace subject pointers for pcre_exec() with a smart pointer
1174        class, thus making it possible to process discontinuous strings.
1175    
1176    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1177        much the same function. They were added by different people who were trying
1178        to make PCRE easy to compile on non-Unix systems. It has been suggested
1179        that PCRE_EXPORT be abolished now that there is more automatic apparatus
1180        for compiling on Windows systems. I have therefore replaced it with
1181        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1182        defaults to "extern" for C or "extern C" for C++, which works fine on
1183        Unix-like systems. It is now possible to override the value of PCRE_DATA_
1184        SCOPE with something explicit in config.h. In addition:
1185    
1186        (a) pcreposix.h still had just "extern" instead of either of these macros;
1187            I have replaced it with PCRE_DATA_SCOPE.
1188    
1189        (b) Functions such as _pcre_xclass(), which are internal to the library,
1190            but external in the C sense, all had PCRE_EXPORT in their definitions.
1191            This is apparently wrong for the Windows case, so I have removed it.
1192            (It makes no difference on Unix-like systems.)
1193    
1194    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1195        of recursive calls to match(). This is different to MATCH_LIMIT because
1196        that limits the total number of calls to match(), not all of which increase
1197        the depth of recursion. Limiting the recursion depth limits the amount of
1198        stack (or heap if NO_RECURSE is set) that is used. The default can be set
1199        when PCRE is compiled, and changed at run time. A patch from Google adds
1200        this functionality to the C++ interface.
1201    
1202    18. Changes to the handling of Unicode character properties:
1203    
1204        (a) Updated the table to Unicode 4.1.0.
1205    
1206        (b) Recognize characters that are not in the table as "Cn" (undefined).
1207    
1208        (c) I revised the way the table is implemented to a much improved format
1209            which includes recognition of ranges. It now supports the ranges that
1210            are defined in UnicodeData.txt, and it also amalgamates other
1211            characters into ranges. This has reduced the number of entries in the
1212            table from around 16,000 to around 3,000, thus reducing its size
1213            considerably. I realized I did not need to use a tree structure after
1214            all - a binary chop search is just as efficient. Having reduced the
1215            number of entries, I extended their size from 6 bytes to 8 bytes to
1216            allow for more data.
1217    
1218        (d) Added support for Unicode script names via properties such as \p{Han}.
1219    
1220    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1221        matching that character.
1222    
1223    20. When matching a repeated Unicode property with a minimum greater than zero,
1224        (for example \pL{2,}), PCRE could look past the end of the subject if it
1225        reached it while seeking the minimum number of characters. This could
1226        happen only if some of the characters were more than one byte long, because
1227        there is a check for at least the minimum number of bytes.
1228    
1229    21. Refactored the implementation of \p and \P so as to be more general, to
1230        allow for more different types of property in future. This has changed the
1231        compiled form incompatibly. Anybody with saved compiled patterns that use
1232        \p or \P will have to recompile them.
1233    
1234    22. Added "Any" and "L&" to the supported property types.
1235    
1236    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1237        but give a compile time error if the value is greater than 0xff.
1238    
1239    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1240        accidentally not being installed or uninstalled.
1241    
1242    25. The pcre.h file was built from pcre.h.in, but the only changes that were
1243        made were to insert the current release number. This seemed silly, because
1244        it made things harder for people building PCRE on systems that don't run
1245        "configure". I have turned pcre.h into a distributed file, no longer built
1246        by "configure", with the version identification directly included. There is
1247        no longer a pcre.h.in file.
1248    
1249        However, this change necessitated a change to the pcre-config script as
1250        well. It is built from pcre-config.in, and one of the substitutions was the
1251        release number. I have updated configure.ac so that ./configure now finds
1252        the release number by grepping pcre.h.
1253    
1254    26. Added the ability to run the tests under valgrind.
1255    
1256    
1257    Version 6.4 05-Sep-05
1258    ---------------------
1259    
1260     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1261        "--" to be printed when multiple files were scanned, even when none of the
1262        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1263        consider it to be a bug, and have restored the previous behaviour.
1264    
1265     2. A couple of code tidies to get rid of compiler warnings.
1266    
1267     3. The pcretest program used to cheat by referring to symbols in the library
1268        whose names begin with _pcre_. These are internal symbols that are not
1269        really supposed to be visible externally, and in some environments it is
1270        possible to suppress them. The cheating is now confined to including
1271        certain files from the library's source, which is a bit cleaner.
1272    
1273     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1274        file's purpose clearer.
1275    
1276     5. Reorganized pcre_ucp_findchar().
1277    
1278    
1279    Version 6.3 15-Aug-05
1280    ---------------------
1281    
1282     1. The file libpcre.pc.in did not have general read permission in the tarball.
1283    
1284     2. There were some problems when building without C++ support:
1285    
1286        (a) If C++ support was not built, "make install" and "make test" still
1287            tried to test it.
1288    
1289        (b) There were problems when the value of CXX was explicitly set. Some
1290            changes have been made to try to fix these, and ...
1291    
1292        (c) --disable-cpp can now be used to explicitly disable C++ support.
1293    
1294        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1295            backslash in a target when C++ was disabled. This confuses some
1296            versions of "make", apparently. Using an intermediate variable solves
1297            this. (Same for CPP_LOBJ.)
1298    
1299     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1300        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1301        necessary on certain architectures.
1302    
1303     4. Added a setting of -export-symbols-regex to the link command to remove
1304        those symbols that are exported in the C sense, but actually are local
1305        within the library, and not documented. Their names all begin with
1306        "_pcre_". This is not a perfect job, because (a) we have to except some
1307        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1308        available (and never for static libraries). I have made a note to try to
1309        find a way round (a) in the future.
1310    
1311    
1312    Version 6.2 01-Aug-05
1313    ---------------------
1314    
1315     1. There was no test for integer overflow of quantifier values. A construction
1316        such as {1111111111111111} would give undefined results. What is worse, if
1317        a minimum quantifier for a parenthesized subpattern overflowed and became
1318        negative, the calculation of the memory size went wrong. This could have
1319        led to memory overwriting.
1320    
1321     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1322    
1323     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1324        operating environments where this matters.
1325    
1326     4. Applied Giuseppe Maxia's patch to add additional features for controlling
1327        PCRE options from within the C++ wrapper.
1328    
1329     5. Named capturing subpatterns were not being correctly counted when a pattern
1330        was compiled. This caused two problems: (a) If there were more than 100
1331        such subpatterns, the calculation of the memory needed for the whole
1332        compiled pattern went wrong, leading to an overflow error. (b) Numerical
1333        back references of the form \12, where the number was greater than 9, were
1334        not recognized as back references, even though there were sufficient
1335        previous subpatterns.
1336    
1337     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1338        versions of gcc, e.g. 2.95.4.
1339    
1340    
1341    Version 6.1 21-Jun-05
1342    ---------------------
1343    
1344     1. There was one reference to the variable "posix" in pcretest.c that was not
1345        surrounded by "#if !defined NOPOSIX".
1346    
1347     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1348        the cross-check on the old pcre_info() function, for the benefit of the
1349        cut-down version of PCRE that is currently imported into Exim.
1350    
1351     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1352        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1353        patterns that start (?i) but I don't think that matters. The use of (?i) is
1354        just an example; this all applies to the other options as well.
1355    
1356     4. Since libtool seems to echo the compile commands it is issuing, the output
1357        from "make" can be reduced a bit by putting "@" in front of each libtool
1358        compile command.
1359    
1360     5. Patch from the folks at Google for configure.in to be a bit more thorough
1361        in checking for a suitable C++ installation before trying to compile the
1362        C++ stuff. This should fix a reported problem when a compiler was present,
1363        but no suitable headers.
1364    
1365     6. The man pages all had just "PCRE" as their title. I have changed them to
1366        be the relevant file name. I have also arranged that these names are
1367        retained in the file doc/pcre.txt, which is a concatenation in text format
1368        of all the man pages except the little individual ones for each function.
1369    
1370     7. The NON-UNIX-USE file had not been updated for the different set of source
1371        files that come with release 6. I also added a few comments about the C++
1372        wrapper.
1373    
1374    
1375    Version 6.0 07-Jun-05
1376    ---------------------
1377    
1378     1. Some minor internal re-organization to help with my DFA experiments.
1379    
1380     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1381        didn't matter for the library itself when fully configured, but did matter
1382        when compiling without UCP support, or within Exim, where the ucp files are
1383        not imported.
1384    
1385     3. Refactoring of the library code to split up the various functions into
1386        different source modules. The addition of the new DFA matching code (see
1387        below) to a single monolithic source would have made it really too
1388        unwieldy, quite apart from causing all the code to be include in a
1389        statically linked application, when only some functions are used. This is
1390        relevant even without the DFA addition now that patterns can be compiled in
1391        one application and matched in another.
1392    
1393        The downside of splitting up is that there have to be some external
1394        functions and data tables that are used internally in different modules of
1395        the library but which are not part of the API. These have all had their
1396        names changed to start with "_pcre_" so that they are unlikely to clash
1397        with other external names.
1398    
1399     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1400        a different (DFA) algorithm. Although it is slower than the original
1401        function, it does have some advantages for certain types of matching
1402        problem.
1403    
1404     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1405        including restarting after a partial match.
1406    
1407     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1408        defined when compiling for Windows was sent to me. I have put it into the
1409        code, though I have no means of testing or verifying it.
1410    
1411     7. Added the pcre_refcount() auxiliary function.
1412    
1413     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1414        match before or at the first newline in the subject string. In pcretest,
1415        the /f option on a pattern can be used to set this.
1416    
1417     9. A repeated \w when used in UTF-8 mode with characters greater than 256
1418        would behave wrongly. This has been present in PCRE since release 4.0.
1419    
1420    10. A number of changes to the pcregrep command:
1421    
1422        (a) Refactored how -x works; insert ^(...)$ instead of setting
1423            PCRE_ANCHORED and checking the length, in preparation for adding
1424            something similar for -w.
1425    
1426        (b) Added the -w (match as a word) option.
1427    
1428        (c) Refactored the way lines are read and buffered so as to have more
1429            than one at a time available.
1430    
1431        (d) Implemented a pcregrep test script.
1432    
1433        (e) Added the -M (multiline match) option. This allows patterns to match
1434            over several lines of the subject. The buffering ensures that at least
1435            8K, or the rest of the document (whichever is the shorter) is available
1436            for matching (and similarly the previous 8K for lookbehind assertions).
1437    
1438        (f) Changed the --help output so that it now says
1439    
1440              -w, --word-regex(p)
1441    
1442            instead of two lines, one with "regex" and the other with "regexp"
1443            because that confused at least one person since the short forms are the
1444            same. (This required a bit of code, as the output is generated
1445            automatically from a table. It wasn't just a text change.)
1446    
1447        (g) -- can be used to terminate pcregrep options if the next thing isn't an
1448            option but starts with a hyphen. Could be a pattern or a path name
1449            starting with a hyphen, for instance.
1450    
1451        (h) "-" can be given as a file name to represent stdin.
1452    
1453        (i) When file names are being printed, "(standard input)" is used for
1454            the standard input, for compatibility with GNU grep. Previously
1455            "<stdin>" was used.
1456    
1457        (j) The option --label=xxx can be used to supply a name to be used for
1458            stdin when file names are being printed. There is no short form.
1459    
1460        (k) Re-factored the options decoding logic because we are going to add
1461            two more options that take data. Such options can now be given in four
1462            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1463    
1464        (l) Added the -A, -B, and -C options for requesting that lines of context
1465            around matches be printed.
1466    
1467        (m) Added the -L option to print the names of files that do not contain
1468            any matching lines, that is, the complement of -l.
1469    
1470        (n) The return code is 2 if any file cannot be opened, but pcregrep does
1471            continue to scan other files.
1472    
1473        (o) The -s option was incorrectly implemented. For compatibility with other
1474            greps, it now suppresses the error message for a non-existent or non-
1475            accessible file (but not the return code). There is a new option called
1476            -q that suppresses the output of matching lines, which was what -s was
1477            previously doing.
1478    
1479        (p) Added --include and --exclude options to specify files for inclusion
1480            and exclusion when recursing.
1481    
1482    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1483        Hopefully, it now does.
1484    
1485    12. Missing cast in pcre_study().
1486    
1487    13. Added an "uninstall" target to the makefile.
1488    
1489    14. Replaced "extern" in the function prototypes in Makefile.in with
1490        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1491        world, but is set differently for Windows.
1492    
1493    15. Added a second compiling function called pcre_compile2(). The only
1494        difference is that it has an extra argument, which is a pointer to an
1495        integer error code. When there is a compile-time failure, this is set
1496        non-zero, in addition to the error test pointer being set to point to an
1497        error message. The new argument may be NULL if no error number is required
1498        (but then you may as well call pcre_compile(), which is now just a
1499        wrapper). This facility is provided because some applications need a
1500        numeric error indication, but it has also enabled me to tidy up the way
1501        compile-time errors are handled in the POSIX wrapper.
1502    
1503    16. Added VPATH=.libs to the makefile; this should help when building with one
1504        prefix path and installing with another. (Or so I'm told by someone who
1505        knows more about this stuff than I do.)
1506    
1507    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1508        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1509        match everything, including newlines. This is not POSIX-compatible, but
1510        somebody wanted the feature. From pcretest it can be activated by using
1511        both the P and the s flags.
1512    
1513    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1514    
1515    19. libpcre.pc was being incorrectly installed as executable.
1516    
1517    20. A couple of places in pcretest check for end-of-line by looking for '\n';
1518        it now also looks for '\r' so that it will work unmodified on Windows.
1519    
1520    21. Added Google's contributed C++ wrapper to the distribution.
1521    
1522    22. Added some untidy missing memory free() calls in pcretest, to keep
1523        Electric Fence happy when testing.
1524    
1525    
1526    
1527    Version 5.0 13-Sep-04
1528    ---------------------
1529    
1530     1. Internal change: literal characters are no longer packed up into items
1531        containing multiple characters in a single byte-string. Each character
1532        is now matched using a separate opcode. However, there may be more than one
1533        byte in the character in UTF-8 mode.
1534    
1535     2. The pcre_callout_block structure has two new fields: pattern_position and
1536        next_item_length. These contain the offset in the pattern to the next match
1537        item, and its length, respectively.
1538    
1539     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1540        insertion of callouts before each pattern item. Added the /C option to
1541        pcretest to make use of this.
1542    
1543     4. On the advice of a Windows user, the lines
1544    
1545          #if defined(_WIN32) || defined(WIN32)
1546          _setmode( _fileno( stdout ), 0x8000 );
1547          #endif  /* defined(_WIN32) || defined(WIN32) */
1548    
1549        have been added to the source of pcretest. This apparently does useful
1550        magic in relation to line terminators.
1551    
1552     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1553        for the benefit of those environments where the "b" makes a difference.
1554    
1555     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1556        to know about it. I have put a hack into configure.in that adds in code
1557        to set GCC=yes if CC=icc. This seems to end up at a point in the
1558        generated configure script that is early enough to affect the setting of
1559        compiler options, which is what is needed, but I have no means of testing
1560        whether it really works. (The user who reported this had patched the
1561        generated configure script, which of course I cannot do.)
1562    
1563        LATER: After change 22 below (new libtool files), the configure script
1564        seems to know about icc (and also ecc). Therefore, I have commented out
1565        this hack in configure.in.
1566    
1567     7. Added support for pkg-config (2 patches were sent in).
1568    
1569     8. Negated POSIX character classes that used a combination of internal tables
1570        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1571        [[:^ascii]]. Typically, they would match almost any characters. The other
1572        POSIX classes were not broken in this way.
1573    
1574     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1575        to find the match, as PCRE was deluded into thinking that the match had to
1576        start at the start point or following a newline. The same bug applied to
1577        patterns with negative forward assertions or any backward assertions
1578        preceding ".*" at the start, unless the pattern required a fixed first
1579        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1580    
1581    10. In UTF-8 mode, when moving forwards in the subject after a failed match
1582        starting at the last subject character, bytes beyond the end of the subject
1583        string were read.
1584    
1585    11. Renamed the variable "class" as "classbits" to make life easier for C++
1586        users. (Previously there was a macro definition, but it apparently wasn't
1587        enough.)
1588    
1589    12. Added the new field "tables" to the extra data so that tables can be passed
1590        in at exec time, or the internal tables can be re-selected. This allows
1591        a compiled regex to be saved and re-used at a later time by a different
1592        program that might have everything at different addresses.
1593    
1594    13. Modified the pcre-config script so that, when run on Solaris, it shows a
1595        -R library as well as a -L library.
1596    
1597    14. The debugging options of pcretest (-d on the command line or D on a
1598        pattern) showed incorrect output for anything following an extended class
1599        that contained multibyte characters and which was followed by a quantifier.
1600    
1601    15. Added optional support for general category Unicode character properties
1602        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1603        support. It adds about 90K to the size of the library. The meanings of the
1604        inbuilt class escapes such as \d and \s have NOT been changed.
1605    
1606    16. Updated pcredemo.c to include calls to free() to release the memory for the
1607        compiled pattern.
1608    
1609    17. The generated file chartables.c was being created in the source directory
1610        instead of in the building directory. This caused the build to fail if the
1611        source directory was different from the building directory, and was
1612        read-only.
1613    
1614    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1615        file. No doubt somebody will tell me if they don't make sense... Also added
1616        Dan Mooney's comments about building on OpenVMS.
1617    
1618    19. Added support for partial matching via the PCRE_PARTIAL option for
1619        pcre_exec() and the \P data escape in pcretest.
1620    
1621    20. Extended pcretest with 3 new pattern features:
1622    
1623        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
1624              write the compiled pattern to the file whose name is "rest-of-line".
1625              This is a straight binary dump of the data, with the saved pointer to
1626              the character tables forced to be NULL. The study data, if any, is
1627              written too. After writing, pcretest reads a new pattern.
1628    
1629        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1630              compiled pattern from the given file. There must not be any
1631              occurrences of "<" in the file name (pretty unlikely); if there are,
1632              pcretest will instead treat the initial "<" as a pattern delimiter.
1633              After reading in the pattern, pcretest goes on to read data lines as
1634              usual.
1635    
1636        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1637              and 16-bit fields in a compiled pattern, to simulate a pattern that
1638              was compiled on a host of opposite endianness.
1639    
1640    21. The pcre-exec() function can now cope with patterns that were compiled on
1641        hosts of opposite endianness, with this restriction:
1642    
1643          As for any compiled expression that is saved and used later, the tables
1644          pointer field cannot be preserved; the extra_data field in the arguments
1645          to pcre_exec() should be used to pass in a tables address if a value
1646          other than the default internal tables were used at compile time.
1647    
1648    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1649        now diagnosed as an error. Previously, most of the time, a negative number
1650        would have been treated as zero, but if in addition "ovector" was passed as
1651        NULL, a crash could occur.
1652    
1653    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1654        new versions from the libtool 1.5 distribution (the last one is a copy of
1655        a file called libtool.m4). This seems to have fixed the need to patch
1656        "configure" to support Darwin 1.3 (which I used to do). However, I still
1657        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1658        workstation).
1659    
1660    24. Changed the PCRE licence to be the more standard "BSD" licence.
1661    
1662    
1663    Version 4.5 01-Dec-03
1664    ---------------------
1665    
1666     1. There has been some re-arrangement of the code for the match() function so
1667        that it can be compiled in a version that does not call itself recursively.
1668        Instead, it keeps those local variables that need separate instances for
1669        each "recursion" in a frame on the heap, and gets/frees frames whenever it
1670        needs to "recurse". Keeping track of where control must go is done by means
1671        of setjmp/longjmp. The whole thing is implemented by a set of macros that
1672        hide most of the details from the main code, and operates only if
1673        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1674        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1675        operating.
1676    
1677        To make it easier for callers to provide specially tailored get/free
1678        functions for this usage, two new functions, pcre_stack_malloc, and
1679        pcre_stack_free, are used. They are always called in strict stacking order,
1680        and the size of block requested is always the same.
1681    
1682        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1683        PCRE has been compiled to use the stack or the heap for recursion. The
1684        -C option of pcretest uses this to show which version is compiled.
1685    
1686        A new data escape \S, is added to pcretest; it causes the amounts of store
1687        obtained and freed by both kinds of malloc/free at match time to be added
1688        to the output.
1689    
1690     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1691        what's available on my current Linux desktop machine.
1692    
1693     3. When matching a UTF-8 string, the test for a valid string at the start has
1694        been extended. If start_offset is not zero, PCRE now checks that it points
1695        to a byte that is the start of a UTF-8 character. If not, it returns
1696        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1697        this is necessary because there may be backward assertions in the pattern.
1698        When matching the same subject several times, it may save resources to use
1699        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1700    
1701     4. The code for checking the validity of UTF-8 strings has been tightened so
1702        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1703        containing "overlong sequences".
1704    
1705     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1706        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1707        should have been "&", but it just so happened that all the cases this let
1708        through by mistake were picked up later in the function.
1709    
1710     6. I had used a variable called "isblank" - this is a C99 function, causing
1711        some compilers to warn. To avoid this, I renamed it (as "blankclass").
1712    
1713     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1714        prompting; (b) run "./pcretest /dev/null" at the start of the test script
1715        so the version is shown; (c) stop "make test" echoing "./RunTest".
1716    
1717     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1718    
1719     9. The prototype for memmove() for systems that don't have it was using
1720        size_t, but the inclusion of the header that defines size_t was later. I've
1721        moved the #includes for the C headers earlier to avoid this.
1722    
1723    10. Added some adjustments to the code to make it easier to compiler on certain
1724        special systems:
1725    
1726          (a) Some "const" qualifiers were missing.
1727          (b) Added the macro EXPORT before all exported functions; by default this
1728              is defined to be empty.
1729          (c) Changed the dftables auxiliary program (that builds chartables.c) so
1730              that it reads its output file name as an argument instead of writing
1731              to the standard output and assuming this can be redirected.
1732    
1733    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
1734        class containing characters with values greater than 255, PCRE compilation
1735        went into a loop.
1736    
1737    12. A recursive reference to a subpattern that was within another subpattern
1738        that had a minimum quantifier of zero caused PCRE to crash. For example,
1739        (x(y(?2))z)? provoked this bug with a subject that got as far as the
1740        recursion. If the recursively-called subpattern itself had a zero repeat,
1741        that was OK.
1742    
1743    13. In pcretest, the buffer for reading a data line was set at 30K, but the
1744        buffer into which it was copied (for escape processing) was still set at
1745        1024, so long lines caused crashes.
1746    
1747    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
1748        "internal error: code overflow...". This applied to any character class
1749        that was followed by a possessive quantifier.
1750    
1751    15. Modified the Makefile to add libpcre.la as a prerequisite for
1752        libpcreposix.la because I was told this is needed for a parallel build to
1753        work.
1754    
1755    16. If a pattern that contained .* following optional items at the start was
1756        studied, the wrong optimizing data was generated, leading to matching
1757        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
1758        matching string must start with a or b or c. The correct conclusion for
1759        this pattern is that a match can start with any character.
1760    
1761    
1762    Version 4.4 13-Aug-03
1763    ---------------------
1764    
1765     1. In UTF-8 mode, a character class containing characters with values between
1766        127 and 255 was not handled correctly if the compiled pattern was studied.
1767        In fixing this, I have also improved the studying algorithm for such
1768        classes (slightly).
1769    
1770     2. Three internal functions had redundant arguments passed to them. Removal
1771        might give a very teeny performance improvement.
1772    
1773     3. Documentation bug: the value of the capture_top field in a callout is *one
1774        more than* the number of the hightest numbered captured substring.
1775    
1776     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
1777        in incorrectly linking with a previously installed version. They now link
1778        explicitly with libpcre.la.
1779    
1780     5. configure.in no longer needs to recognize Cygwin specially.
1781    
1782     6. A problem in pcre.in for Windows platforms is fixed.
1783    
1784     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
1785        pcretest, it used to include the size of the study block as part of its
1786        output. Unfortunately, the structure contains a field that has a different
1787        size on different hardware architectures. This meant that the tests that
1788        showed this size failed. As the block is currently always of a fixed size,
1789        this information isn't actually particularly useful in pcretest output, so
1790        I have just removed it.
1791    
1792     8. Three pre-processor statements accidentally did not start in column 1.
1793        Sadly, there are *still* compilers around that complain, even though
1794        standard C has not required this for well over a decade. Sigh.
1795    
1796     9. In pcretest, the code for checking callouts passed small integers in the
1797        callout_data field, which is a void * field. However, some picky compilers
1798        complained about the casts involved for this on 64-bit systems. Now
1799        pcretest passes the address of the small integer instead, which should get
1800        rid of the warnings.
1801    
1802    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
1803        both compile and run time, and gives an error if an invalid UTF-8 sequence
1804        is found. There is a option for disabling this check in cases where the
1805        string is known to be correct and/or the maximum performance is wanted.
1806    
1807    11. In response to a bug report, I changed one line in Makefile.in from
1808    
1809            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
1810        to
1811            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
1812    
1813        to look similar to other lines, but I have no way of telling whether this
1814        is the right thing to do, as I do not use Windows. No doubt I'll get told
1815        if it's wrong...
1816    
1817    
1818    Version 4.3 21-May-03
1819    ---------------------
1820    
1821    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
1822       Makefile.
1823    
1824    2. Some refactoring to improve the quality of the code:
1825    
1826       (i)   The utf8_table... variables are now declared "const".
1827    
1828       (ii)  The code for \cx, which used the "case flipping" table to upper case
1829             lower case letters, now just substracts 32. This is ASCII-specific,
1830             but the whole concept of \cx is ASCII-specific, so it seems
1831             reasonable.
1832    
1833       (iii) PCRE was using its character types table to recognize decimal and
1834             hexadecimal digits in the pattern. This is silly, because it handles
1835             only 0-9, a-f, and A-F, but the character types table is locale-
1836             specific, which means strange things might happen. A private
1837             table is now used for this - though it costs 256 bytes, a table is
1838             much faster than multiple explicit tests. Of course, the standard
1839             character types table is still used for matching digits in subject
1840             strings against \d.
1841    
1842       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
1843             ending in _t are). So I've renamed it as ESC_tee.
1844    
1845    3. The first argument for regexec() in the POSIX wrapper should have been
1846       defined as "const".
1847    
1848    4. Changed pcretest to use malloc() for its buffers so that they can be
1849       Electric Fenced for debugging.
1850    
1851    5. There were several places in the code where, in UTF-8 mode, PCRE would try
1852       to read one or more bytes before the start of the subject string. Often this
1853       had no effect on PCRE's behaviour, but in some circumstances it could
1854       provoke a segmentation fault.
1855    
1856    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
1857       to try to read one or more bytes before the start of the subject string.
1858    
1859    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
1860       UTF-8 support could misbehave in various ways if the subject string
1861       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
1862       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
1863       back over UTF-8 characters.)
1864    
1865    
1866    Version 4.2 14-Apr-03
1867    ---------------------
1868    
1869    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
1870    
1871    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
1872         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
1873         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
1874         [WIN_PREFIX]: new variable, "cyg" for Cygwin
1875         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
1876           and BUILD_EXEEXT
1877         Note: automatic setting of the BUILD variables is not yet working
1878         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
1879           compile-time but not at link-time
1880         [LINK]: use for linking executables only
1881         make different versions for Windows and non-Windows
1882         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
1883           libraries
1884         [LINK_FOR_BUILD]: new variable
1885         [OBJEXT]: use throughout
1886         [EXEEXT]: use throughout
1887         <winshared>: new target
1888         <wininstall>: new target
1889         <dftables.o>: use native compiler
1890         <dftables>: use native linker
1891         <install>: handle Windows platform correctly
1892         <clean>: ditto
1893         <check>: ditto
1894         copy DLL to top builddir before testing
1895    
1896       As part of these changes, -no-undefined was removed again. This was reported
1897       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
1898       in any case.
1899    
1900    3. Some tidies to get rid of compiler warnings:
1901    
1902       . In the match_data structure, match_limit was an unsigned long int, whereas
1903         match_call_count was an int. I've made them both unsigned long ints.
1904    
1905       . In pcretest the fact that a const uschar * doesn't automatically cast to
1906         a void * provoked a warning.
1907    
1908       . Turning on some more compiler warnings threw up some "shadow" variables
1909         and a few more missing casts.
1910    
1911    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
1912       option, a class that contained a single character with a value between 128
1913       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
1914    
1915    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
1916       option, a class that contained several characters, but with at least one
1917       whose value was between 128 and 255 caused PCRE to crash.
1918    
1919    
1920    Version 4.1 12-Mar-03
1921    ---------------------
1922    
1923    1. Compiling with gcc -pedantic found a couple of places where casts were
1924    needed, and a string in dftables.c that was longer than standard compilers are
1925    required to support.
1926    
1927    2. Compiling with Sun's compiler found a few more places where the code could
1928    be tidied up in order to avoid warnings.
1929    
1930    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
1931    first of these names is deprecated in the latest Autoconf in favour of the name
1932    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
1933    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
1934    analogy I have changed it to CFLAGS_FOR_BUILD.
1935    
1936    4. Added -no-undefined to the linking command in the Makefile, because this is
1937    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
1938    linking step for the pcreposix library.
1939    
1940    5. PCRE was failing to diagnose the case of two named groups with the same
1941    name.
1942    
1943    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
1944    literal character that is needed in the subject for a match, and scans along to
1945    ensure that it is present before embarking on the full matching process. This
1946    saves time in cases of nested unlimited repeats that are never going to match.
1947    Problem: the scan can take a lot of time if the subject is very long (e.g.
1948    megabytes), thus penalizing straightforward matches. It is now done only if the
1949    amount of subject to be scanned is less than 1000 bytes.
1950    
1951    7. A lesser problem with the same optimization is that it was recording the
1952    first character of an anchored pattern as "needed", thus provoking a search
1953    right along the subject, even when the first match of the pattern was going to
1954    fail. The "needed" character is now not set for anchored patterns, unless it
1955    follows something in the pattern that is of non-fixed length. Thus, it still
1956    fulfils its original purpose of finding quick non-matches in cases of nested
1957    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
1958    
1959    
1960    Version 4.0 17-Feb-03
1961    ---------------------
1962    
1963    1. If a comment in an extended regex that started immediately after a meta-item
1964    extended to the end of string, PCRE compiled incorrect data. This could lead to
1965    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
1966    
1967    2. Moved to autoconf 2.53 and libtool 1.4.2.
1968    
1969    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
1970    the special perltest8 script is no longer needed - all the tests can be run
1971    from a single perltest script.
1972    
1973    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
1974    by \s. It has now been removed in PCRE. This means it isn't recognized as
1975    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
1976    class [:space:] *does* include VT, thereby creating a mess.
1977    
1978    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
1979    space and tab.
1980    
1981    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
1982    its new features into the main test script, reducing the number of scripts.
1983    
1984    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
1985    were backward compatible, and made the (?i) apply to the whole pattern, as if
1986    /i were given. Now it behaves more logically, and applies the option setting
1987    only to what follows. PCRE has been changed to follow suit. However, if it
1988    finds options settings right at the start of the pattern, it extracts them into
1989    the global options, as before. Thus, they show up in the info data.
1990    
1991    8. Added support for the \Q...\E escape sequence. Characters in between are
1992    treated as literals. This is slightly different from Perl in that $ and @ are
1993    also handled as literals inside the quotes. In Perl, they will cause variable
1994    interpolation. Note the following examples:
1995    
1996        Pattern            PCRE matches      Perl matches
1997    
1998        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
1999        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
2000        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
2001    
2002    For compatibility with Perl, \Q...\E sequences are recognized inside character
2003    classes as well as outside them.
2004    
2005    9. Re-organized 3 code statements in pcretest to avoid "overflow in
2006    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
2007    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
2008    signed/unsigned warnings.
2009    
2010    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
2011    option for pcretest, so I've replaced it by a simple function that does just
2012    that job.
2013    
2014    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
2015    "pcregrep -".
2016    
2017    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
2018    Java package. This provides some syntactic sugar for simple cases of what my
2019    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
2020    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
2021    item, you can use this simplified notation. Note that only makes sense with
2022    greedy quantifiers. Consequently, the use of the possessive quantifier forces
2023    greediness, whatever the setting of the PCRE_UNGREEDY option.
2024    
2025    13. A change of greediness default within a pattern was not taking effect at
2026    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
2027    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
2028    was abstracted outside.
2029    
2030    14. PCRE now supports the \G assertion. It is true when the current matching
2031    position is at the start point of the match. This differs from \A when the
2032    starting offset is non-zero. Used with the /g option of pcretest (or similar
2033    code), it works in the same way as it does for Perl's /g option. If all
2034    alternatives of a regex begin with \G, the expression is anchored to the start
2035    match position, and the "anchored" flag is set in the compiled expression.
2036    
2037    15. Some bugs concerning the handling of certain option changes within patterns
2038    have been fixed. These applied to options other than (?ims). For example,
2039    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
2040    been the other way round. Some of this was related to change 7 above.
2041    
2042    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
2043    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
2044    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
2045    POSIX classes only within a class (e.g. /[[:alpha:]]/).
2046    
2047    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
2048    mode. Unlike ".", it always matches newline, whatever the setting of
2049    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
2050    assertions. Perl allows it, but it doesn't (in general) work because it can't
2051    calculate the length of the lookbehind. At least, that's the case for Perl
2052    5.8.0 - I've been told they are going to document that it doesn't work in
2053    future.
2054    
2055    18. Added an error diagnosis for escapes that PCRE does not support: these are
2056    \L, \l, \N, \P, \p, \U, \u, and \X.
2057    
2058    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
2059    reading past the end of the pattern in cases such as /[abcd/.
2060    
2061    20. PCRE was getting more memory than necessary for patterns with classes that
2062    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
2063    
2064    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
2065    compiling PCRE for use with Virtual Pascal.
2066    
2067    22. Small fix to the Makefile to make it work properly if the build is done
2068    outside the source tree.
2069    
2070    23. Added a new extension: a condition to go with recursion. If a conditional
2071    subpattern starts with (?(R) the "true" branch is used if recursion has
2072    happened, whereas the "false" branch is used only at the top level.
2073    
2074    24. When there was a very long string of literal characters (over 255 bytes
2075    without UTF support, over 250 bytes with UTF support), the computation of how
2076    much memory was required could be incorrect, leading to segfaults or other
2077    strange effects.
2078    
2079    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
2080    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
2081    there was a subsequent back reference to those brackets. This meant that, for
2082    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
2083    possible to check for precisely this case. All we can do is abandon the
2084    optimization if .* occurs inside capturing brackets when there are any back
2085    references whatsoever. (See below for a better fix that came later.)
2086    
2087    26. The handling of the optimization for finding the first character of a
2088    non-anchored pattern, and for finding a character that is required later in the
2089    match were failing in some cases. This didn't break the matching; it just
2090    failed to optimize when it could. The way this is done has been re-implemented.
2091    
2092    27. Fixed typo in error message for invalid (?R item (it said "(?p").
2093    
2094    28. Added a new feature that provides some of the functionality that Perl
2095    provides with (?{...}). The facility is termed a "callout". The way it is done
2096    in PCRE is for the caller to provide an optional function, by setting
2097    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
2098    global variable. By default it is unset, which disables all calling out. To get
2099    the function called, the regex must include (?C) at appropriate points. This
2100    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
2101    This provides a means of identifying different callout points. When PCRE
2102    reaches such a point in the regex, if pcre_callout has been set, the external
2103    function is called. It is provided with data in a structure called
2104    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
2105    matching continues; if it returns a non-zero value, the match at the current
2106    point fails. However, backtracking will occur if possible. [This was changed
2107    later and other features added - see item 49 below.]
2108    
2109    29. pcretest is upgraded to test the callout functionality. It provides a
2110    callout function that displays information. By default, it shows the start of
2111    the match and the current position in the text. There are some new data escapes
2112    to vary what happens:
2113    
2114        \C+         in addition, show current contents of captured substrings
2115        \C-         do not supply a callout function
2116        \C!n        return 1 when callout number n is reached
2117        \C!n!m      return 1 when callout number n is reached for the mth time
2118    
2119    30. If pcregrep was called with the -l option and just a single file name, it
2120    output "<stdin>" if a match was found, instead of the file name.
2121    
2122    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
2123    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
2124    pcre_exec(). This saves a malloc/free per call. The default value of
2125    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
2126    when configuring.
2127    
2128    32. The default maximum size of a compiled pattern is 64K. There have been a
2129    few cases of people hitting this limit. The code now uses macros to handle the
2130    storing of links as offsets within the compiled pattern. It defaults to 2-byte
2131    links, but this can be changed to 3 or 4 bytes by --with-link-size when
2132    configuring. Tests 2 and 5 work only with 2-byte links because they output
2133    debugging information about compiled patterns.
2134    
2135    33. Internal code re-arrangements:
2136    
2137    (a) Moved the debugging function for printing out a compiled regex into
2138        its own source file (printint.c) and used #include to pull it into
2139        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
2140        separate copies.
2141    
2142    (b) Defined the list of op-code names for debugging as a macro in
2143        internal.h so that it is next to the definition of the opcodes.
2144    
2145    (c) Defined a table of op-code lengths for simpler skipping along compiled
2146        code. This is again a macro in internal.h so that it is next to the
2147        definition of the opcodes.
2148    
2149    34. Added support for recursive calls to individual subpatterns, along the
2150    lines of Robin Houston's patch (but implemented somewhat differently).
2151    
2152    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
2153    allow it to read and process whole directories in Win32. This code was
2154    contributed by Lionel Fourquaux; it has not been tested by me.
2155    
2156    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
2157    used to name a group. Names consist of alphanumerics and underscores, and must
2158    be unique. Back references use the syntax (?P=name) and recursive calls use
2159    (?P>name) which is a PCRE extension to the Python extension. Groups still have
2160    numbers. The function pcre_fullinfo() can be used after compilation to extract
2161    a name/number map. There are three relevant calls:
2162    
2163      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
2164      PCRE_INFO_NAMECOUNT            yields the number of entries
2165      PCRE_INFO_NAMETABLE            yields a pointer to the map.
2166    
2167    The map is a vector of fixed-size entries. The size of each entry depends on
2168    the length of the longest name used. The first two bytes of each entry are the
2169    group number, most significant byte first. There follows the corresponding
2170    name, zero terminated. The names are in alphabetical order.
2171    
2172    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
2173    case instead of 255. Making it the same both with and without UTF-8 support
2174    means that the same test output works with both.
2175    
2176    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
2177    calling malloc() with a zero argument.
2178    
2179    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
2180    optimization. I've improved things by keeping a bitmap of backreferences with
2181    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
2182    fact referenced, the optimization can be applied. It is unlikely that a
2183    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
2184    the match to follow \n) will appear inside brackets with a number greater than
2185    31, but if it does, any back reference > 31 suppresses the optimization.
2186    
2187    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
2188    of disabling numbered capturing parentheses. Any opening parenthesis that is
2189    not followed by ? behaves as if it were followed by ?: but named parentheses
2190    can still be used for capturing (and they will acquire numbers in the usual
2191    way).
2192    
2193    41. Redesigned the return codes from the match() function into yes/no/error so
2194    that errors can be passed back from deep inside the nested calls. A malloc
2195    failure while inside a recursive subpattern call now causes the
2196    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
2197    
2198    42. It is now possible to set a limit on the number of times the match()
2199    function is called in a call to pcre_exec(). This facility makes it possible to
2200    limit the amount of recursion and backtracking, though not in a directly
2201    obvious way, because the match() function is used in a number of different
2202    circumstances. The count starts from zero for each position in the subject
2203    string (for non-anchored patterns). The default limit is, for compatibility, a
2204    large number, namely 10 000 000. You can change this in two ways:
2205    
2206    (a) When configuring PCRE before making, you can use --with-match-limit=n
2207        to set a default value for the compiled library.
2208    
2209    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
2210        a different value is set. See 45 below.
2211    
2212    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
2213    
2214    43. Added a new function pcre_config(int, void *) to enable run-time extraction
2215    of things that can be changed at compile time. The first argument specifies
2216    what is wanted and the second points to where the information is to be placed.
2217    The current list of available information is:
2218    
2219      PCRE_CONFIG_UTF8
2220    
2221    The output is an integer that is set to one if UTF-8 support is available;
2222    otherwise it is set to zero.
2223    
2224      PCRE_CONFIG_NEWLINE
2225    
2226    The output is an integer that it set to the value of the code that is used for
2227    newline. It is either LF (10) or CR (13).
2228    
2229      PCRE_CONFIG_LINK_SIZE
2230    
2231    The output is an integer that contains the number of bytes used for internal
2232    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
2233    
2234      PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
2235    
2236    The output is an integer that contains the threshold above which the POSIX
2237    interface uses malloc() for output vectors. See item 31 above.
2238    
2239      PCRE_CONFIG_MATCH_LIMIT
2240    
2241    The output is an unsigned integer that contains the default limit of the number
2242    of match() calls in a pcre_exec() execution. See 42 above.
2243    
2244    44. pcretest has been upgraded by the addition of the -C option. This causes it
2245    to extract all the available output from the new pcre_config() function, and to
2246    output it. The program then exits immediately.
2247    
2248    45. A need has arisen to pass over additional data with calls to pcre_exec() in
2249    order to support additional features. One way would have been to define
2250    pcre_exec2() (for example) with extra arguments, but this would not have been
2251    extensible, and would also have required all calls to the original function to
2252    be mapped to the new one. Instead, I have chosen to extend the mechanism that
2253    is used for passing in "extra" data from pcre_study().
2254    
2255    The pcre_extra structure is now exposed and defined in pcre.h. It currently
2256    contains the following fields:
2257    
2258      flags         a bitmap indicating which of the following fields are set
2259      study_data    opaque data from pcre_study()
2260      match_limit   a way of specifying a limit on match() calls for a specific
2261                      call to pcre_exec()
2262      callout_data  data for callouts (see 49 below)
2263    
2264    The flag bits are also defined in pcre.h, and are
2265    
2266      PCRE_EXTRA_STUDY_DATA
2267      PCRE_EXTRA_MATCH_LIMIT
2268      PCRE_EXTRA_CALLOUT_DATA
2269    
2270    The pcre_study() function now returns one of these new pcre_extra blocks, with
2271    the actual study data pointed to by the study_data field, and the
2272    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
2273    before. That is, this change is entirely upwards-compatible and requires no
2274    change to existing code.
2275    
2276    If you want to pass in additional data to pcre_exec(), you can either place it
2277    in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
2278    block.
2279    
2280    46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
2281    data string contains the escape sequence \M, pcretest calls pcre_exec() several
2282    times with different match limits, until it finds the minimum value needed for
2283    pcre_exec() to complete. The value is then output. This can be instructive; for
2284    most simple matches the number is quite small, but for pathological cases it
2285    gets very large very quickly.
2286    
2287    47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
2288    returns the size of the data block pointed to by the study_data field in a
2289    pcre_extra block, that is, the value that was passed as the argument to
2290    pcre_malloc() when PCRE was getting memory in which to place the information
2291    created by pcre_study(). The fourth argument should point to a size_t variable.
2292    pcretest has been extended so that this information is shown after a successful
2293    pcre_study() call when information about the compiled regex is being displayed.
2294    
2295    48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
2296    because what follows is always an absolute path. (Later: it turns out that this
2297    is more than cosmetic for MinGW, because it doesn't like empty path
2298    components.)
2299    
2300    49. Some changes have been made to the callout feature (see 28 above):
2301    
2302    (i)  A callout function now has three choices for what it returns:
2303    
2304           0  =>  success, carry on matching
2305         > 0  =>  failure at this point, but backtrack if possible
2306         < 0  =>  serious error, return this value from pcre_exec()
2307    
2308         Negative values should normally be chosen from the set of PCRE_ERROR_xxx
2309         values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
2310         "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
2311         use by callout functions. It will never be used by PCRE itself.
2312    
2313    (ii) The pcre_extra structure (see 45 above) has a void * field called
2314         callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
2315         pcre_callout_block structure has a field of the same name. The contents of
2316         the field passed in the pcre_extra structure are passed to the callout
2317         function in the corresponding field in the callout block. This makes it
2318         easier to use the same callout-containing regex from multiple threads. For
2319         testing, the pcretest program has a new data escape
2320    
2321           \C*n        pass the number n (may be negative) as callout_data
2322    
2323         If the callout function in pcretest receives a non-zero value as
2324         callout_data, it returns that value.
2325    
2326    50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
2327    there were some redundant $(CFLAGS) in commands that are now specified as
2328    $(LINK), which already includes $(CFLAGS).
2329    
2330    51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
2331    has been compiled with UTF-8 support *and* pcre_compile() has been compiled
2332    with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
2333    one-byte characters throughout. Note that case-insensitive matching applies
2334    only to characters whose values are less than 256. PCRE doesn't support the
2335    notion of cases for higher-valued characters.
2336    
2337    (i)   A character class whose characters are all within 0-255 is handled as
2338          a bit map, and the map is inverted for negative classes. Previously, a
2339          character > 255 always failed to match such a class; however it should
2340          match if the class was a negative one (e.g. [^ab]). This has been fixed.
2341    
2342    (ii)  A negated character class with a single character < 255 is coded as
2343          "not this character" (OP_NOT). This wasn't working properly when the test
2344          character was multibyte, either singly or repeated.
2345    
2346    (iii) Repeats of multibyte characters are now handled correctly in UTF-8
2347          mode, for example: \x{100}{2,3}.
2348    
2349    (iv)  The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
2350          singly or repeated) now correctly test multibyte characters. However,
2351          PCRE doesn't recognize any characters with values greater than 255 as
2352          digits, spaces, or word characters. Such characters always match \D, \S,
2353          and \W, and never match \d, \s, or \w.
2354    
2355    (v)   Classes may now contain characters and character ranges with values
2356          greater than 255. For example: [ab\x{100}-\x{400}].
2357    
2358    (vi)  pcregrep now has a --utf-8 option (synonym -u) which makes it call
2359          PCRE in UTF-8 mode.
2360    
2361    52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
2362    PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
2363    retained for backwards compatibility. (Note that LASTLITERAL is also a byte
2364    value.)
2365    
2366    53. The single man page has become too large. I have therefore split it up into
2367    a number of separate man pages. These also give rise to individual HTML pages;
2368    these are now put in a separate directory, and there is an index.html page that
2369    lists them all. Some hyperlinking between the pages has been installed.
2370    
2371    54. Added convenience functions for handling named capturing parentheses.
2372    
2373    55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2374    aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2375    true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2376    are faulted.
2377    
2378    56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2379    calling configure. These values are used when compiling the dftables.c program
2380    which is run to generate the source of the default character tables. They
2381    default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2382    you will need to set these values.
2383    
2384    57. Updated the building process for Windows DLL, as provided by Fred Cox.
2385    
2386    
2387    Version 3.9 02-Jan-02
2388    ---------------------
2389    
2390    1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2391    
2392    2. If --disable-static was given, the building process failed when trying to
2393    build pcretest and pcregrep. (For some reason it was using libtool to compile
2394    them, which is not right, as they aren't part of the library.)
2395    
2396    
2397    Version 3.8 18-Dec-01
2398    ---------------------
2399    
2400    1. The experimental UTF-8 code was completely screwed up. It was packing the
2401    bytes in the wrong order. How dumb can you get?
2402    
2403    
2404    Version 3.7 29-Oct-01
2405    ---------------------
2406    
2407    1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2408    This caused pcretest, when used on the test data, to segfault. Unfortunately,
2409    this didn't happen under Solaris 8, where I normally test things.
2410    
2411    2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2412    doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2413    isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2414    this fix an hour or so after the initial 3.7 release.)
2415    
2416    
2417    Version 3.6 23-Oct-01
2418    ---------------------
2419    
2420    1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2421    offsets passed as NULL with zero offset count.
2422    
2423    2. The config.guess and config.sub files had not been updated when I moved to
2424    the latest autoconf.
2425    
2426    
2427    Version 3.5 15-Aug-01
2428    ---------------------
2429    
2430    1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2431    had been forgotten.
2432    
2433    2. By using declared but undefined structures, we can avoid using "void"
2434    definitions in pcre.h while keeping the internal definitions of the structures
2435    private.
2436    
2437    3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2438    user point of view, this means that both static and shared libraries are built
2439    by default, but this can be individually controlled. More of the work of
2440    handling this static/shared cases is now inside libtool instead of PCRE's make
2441    file.
2442    
2443    4. The pcretest utility is now installed along with pcregrep because it is
2444    useful for users (to test regexs) and by doing this, it automatically gets
2445    relinked by libtool. The documentation has been turned into a man page, so
2446    there are now .1, .txt, and .html versions in /doc.
2447    
2448    5. Upgrades to pcregrep:
2449       (i)   Added long-form option names like gnu grep.
2450       (ii)  Added --help to list all options with an explanatory phrase.
2451       (iii) Added -r, --recursive to recurse into sub-directories.
2452       (iv)  Added -f, --file to read patterns from a file.
2453    
2454    6. pcre_exec() was referring to its "code" argument before testing that
2455    argument for NULL (and giving an error if it was NULL).
2456    
2457    7. Upgraded Makefile.in to allow for compiling in a different directory from
2458    the source directory.
2459    
2460    8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2461    options bits, the pointer it was passed was to an int instead of to an unsigned
2462    long int. This mattered only on 64-bit systems.
2463    
2464    9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2465    generated) instead of pcre.in, which it its source. Also made the same change
2466    in several of the .c files.
2467    
2468    10. A new release of gcc defines printf() as a macro, which broke pcretest
2469    because it had an ifdef in the middle of a string argument for printf(). Fixed
2470    by using separate calls to printf().
2471    
2472    11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2473    script, to force use of CR or LF instead of \n in the source. On non-Unix
2474    systems, the value can be set in config.h.
2475    
2476    12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2477    absolute limit. Changed the text of the error message to make this clear, and
2478    likewise updated the man page.
2479    
2480    13. The limit of 99 on the number of capturing subpatterns has been removed.
2481    The new limit is 65535, which I hope will not be a "real" limit.
2482    
2483    
2484    Version 3.4 22-Aug-00
2485    ---------------------
2486    
2487    1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2488    
2489    2. Diagnose condition (?(0) as an error instead of crashing on matching.
2490    
2491    
2492    Version 3.3 01-Aug-00
2493    ---------------------
2494    
2495    1. If an octal character was given, but the value was greater than \377, it
2496    was not getting masked to the least significant bits, as documented. This could
2497    lead to crashes in some systems.
2498    
2499    2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
2500    the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
2501    
2502    3. Added the functions pcre_free_substring() and pcre_free_substring_list().
2503    These just pass their arguments on to (pcre_free)(), but they are provided
2504    because some uses of PCRE bind it to non-C systems that can call its functions,
2505    but cannot call free() or pcre_free() directly.
2506    
2507    4. Add "make test" as a synonym for "make check". Corrected some comments in
2508    the Makefile.
2509    
2510    5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
2511    Makefile.
2512    
2513    6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
2514    command called pgrep for grepping around the active processes.
2515    
2516    7. Added the beginnings of support for UTF-8 character strings.
2517    
2518    8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
2519    RANLIB to ./ltconfig so that they are used by libtool. I think these are all
2520    the relevant ones. (AR is not passed because ./ltconfig does its own figuring
2521    out for the ar command.)
2522    
2523    
2524    Version 3.2 12-May-00
2525    ---------------------
2526    
2527    This is purely a bug fixing release.
2528    
2529    1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
2530    of ZA. This was just one example of several cases that could provoke this bug,
2531    which was introduced by change 9 of version 2.00. The code for breaking
2532    infinite loops after an iteration that matches an empty string was't working
2533    correctly.
2534    
2535    2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
2536    when matched against abbab (for example). After matching an empty string, it
2537    wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
2538    caused it to match further down the string than it should.
2539    
2540    3. The code contained an inclusion of sys/types.h. It isn't clear why this
2541    was there because it doesn't seem to be needed, and it causes trouble on some
2542    systems, as it is not a Standard C header. It has been removed.
2543    
2544    4. Made 4 silly changes to the source to avoid stupid compiler warnings that
2545    were reported on the Macintosh. The changes were from
2546    
2547      while ((c = *(++ptr)) != 0 && c != '\n');
2548    to
2549      while ((c = *(++ptr)) != 0 && c != '\n') ;
2550    
2551    Totally extraordinary, but if that's what it takes...
2552    
2553    5. PCRE is being used in one environment where neither memmove() nor bcopy() is
2554    available. Added HAVE_BCOPY and an autoconf test for it; if neither
2555    HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
2556    assumes the way PCRE uses memmove() (always moving upwards).
2557    
2558    6. PCRE is being used in one environment where strchr() is not available. There
2559    was only one use in pcre.c, and writing it out to avoid strchr() probably gives
2560    faster code anyway.
2561    
2562    
2563    Version 3.1 09-Feb-00
2564    ---------------------
2565    
2566    The only change in this release is the fixing of some bugs in Makefile.in for
2567    the "install" target:
2568    
2569    (1) It was failing to install pcreposix.h.
2570    
2571    (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
2572    
2573    
2574    Version 3.0 01-Feb-00
2575    ---------------------
2576    
2577    1. Add support for the /+ modifier to perltest (to output $` like it does in
2578    pcretest).
2579    
2580    2. Add support for the /g modifier to perltest.
2581    
2582    3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
2583    matches null strings.
2584    
2585    4. Fix perltest so that it doesn't do unwanted things when fed an empty
2586    pattern. Perl treats empty patterns specially - it reuses the most recent
2587    pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
2588    effect.
2589    
2590    5. The POSIX interface was broken in that it was just handing over the POSIX
2591    captured string vector to pcre_exec(), but (since release 2.00) PCRE has
2592    required a bigger vector, with some working space on the end. This means that
2593    the POSIX wrapper now has to get and free some memory, and copy the results.
2594    
2595    6. Added some simple autoconf support, placing the test data and the
2596    documentation in separate directories, re-organizing some of the
2597    information files, and making it build pcre-config (a GNU standard). Also added
2598    libtool support for building PCRE as a shared library, which is now the
2599    default.
2600    
2601    7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
2602    09 are not valid octal constants. Single digits will be used for minor values
2603    less than 10.
2604    
2605    8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
2606    existing programs that set these in the POSIX interface can use PCRE without
2607    modification.
2608    
2609    9. Added a new function, pcre_fullinfo() with an extensible interface. It can
2610    return all that pcre_info() returns, plus additional data. The pcre_info()
2611    function is retained for compatibility, but is considered to be obsolete.
2612    
2613    10. Added experimental recursion feature (?R) to handle one common case that
2614    Perl 5.6 will be able to do with (?p{...}).
2615    
2616    11. Added support for POSIX character classes like [:alpha:], which Perl is
2617    adopting.
2618    
2619    
2620    Version 2.08 31-Aug-99
2621    ----------------------
2622    
2623    1. When startoffset was not zero and the pattern began with ".*", PCRE was not
2624    trying to match at the startoffset position, but instead was moving forward to
2625    the next newline as if a previous match had failed.
2626    
2627    2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
2628    and could get into a loop if a null string was matched other than at the start
2629    of the subject.
2630    
2631    3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
2632    be distinguished at compile time, and for completeness also added PCRE_DATE.
2633    
2634    5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
2635    in GnuWin32 environments.
2636    
2637    
2638    Version 2.07 29-Jul-99
2639    ----------------------
2640    
2641    1. The documentation is now supplied in plain text form and HTML as well as in
2642    the form of man page sources.
2643    
2644    2. C++ compilers don't like assigning (void *) values to other pointer types.
2645    In particular this affects malloc(). Although there is no problem in Standard
2646    C, I've put in casts to keep C++ compilers happy.
2647    
2648    3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
2649    should be (const char *).
2650    
2651    4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
2652    be useful for non-Unix systems who don't want to bother with the POSIX stuff.
2653    However, I haven't made this a standard facility. The documentation doesn't
2654    mention it, and the Makefile doesn't support it.
2655    
2656    5. The Makefile now contains an "install" target, with editable destinations at
2657    the top of the file. The pcretest program is not installed.
2658    
2659    6. pgrep -V now gives the PCRE version number and date.
2660    
2661    7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
2662    causing the entire string to be ignored, instead of just the last character.
2663    
2664    8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
2665    non-matching string, it can take a very, very long time, even for strings of
2666    quite modest length, because of the nested recursion. PCRE now does better in
2667    some of these cases. It does this by remembering the last required literal
2668    character in the pattern, and pre-searching the subject to ensure it is present
2669    before running the real match. In other words, it applies a heuristic to detect
2670    some types of certain failure quickly, and in the above example, if presented
2671    with a string that has no trailing " it gives "no match" very quickly.
2672    
2673    9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
2674    other alternatives are tried instead.
2675    
2676    
2677    Version 2.06 09-Jun-99
2678    ----------------------
2679    
2680    1. Change pcretest's output for amount of store used to show just the code
2681    space, because the remainder (the data block) varies in size between 32-bit and
2682    64-bit systems.
2683    
2684    2. Added an extra argument to pcre_exec() to supply an offset in the subject to
2685    start matching at. This allows lookbehinds to work when searching for multiple
2686    occurrences in a string.
2687    
2688    3. Added additional options to pcretest for testing multiple occurrences:
2689    
2690       /+   outputs the rest of the string that follows a match
2691       /g   loops for multiple occurrences, using the new startoffset argument
2692       /G   loops for multiple occurrences by passing an incremented pointer
2693    
2694    4. PCRE wasn't doing the "first character" optimization for patterns starting
2695    with \b or \B, though it was doing it for other lookbehind assertions. That is,
2696    it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
2697    the letter 'x'. On long subject strings, this gives a significant speed-up.
2698    
2699    
2700    Version 2.05 21-Apr-99
2701    ----------------------
2702    
2703    1. Changed the type of magic_number from int to long int so that it works
2704    properly on 16-bit systems.
2705    
2706    2. Fixed a bug which caused patterns starting with .* not to work correctly
2707    when the subject string contained newline characters. PCRE was assuming
2708    anchoring for such patterns in all cases, which is not correct because .* will
2709    not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
2710    DOTALL is set at top level; otherwise it knows that patterns starting with .*
2711    must be retried after every newline in the subject.
2712    
2713    
2714    Version 2.04 18-Feb-99
2715    ----------------------
2716    
2717    1. For parenthesized subpatterns with repeats whose minimum was zero, the
2718    computation of the store needed to hold the pattern was incorrect (too large).
2719    If such patterns were nested a few deep, this could multiply and become a real
2720    problem.
2721    
2722    2. Added /M option to pcretest to show the memory requirement of a specific
2723    pattern. Made -m a synonym of -s (which does this globally) for compatibility.
2724    
2725    3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
2726    compiled in such a way that the backtracking after subsequent failure was
2727    pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
2728    ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
2729    
2730    
2731    Version 2.03 02-Feb-99
2732    ----------------------
2733    
2734    1. Fixed typo and small mistake in man page.
2735    
2736    2. Added 4th condition (GPL supersedes if conflict) and created separate
2737    LICENCE file containing the conditions.
2738    
2739    3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
2740    Perl, that is the internal \ allows the delimiter to be included in the
2741    pattern. Locked out the use of \ as a delimiter. If \ immediately follows
2742    the final delimiter, add \ to the end of the pattern (to test the error).
2743    
2744    4. Added the convenience functions for extracting substrings after a successful
2745    match. Updated pcretest to make it able to test these functions.
2746    
2747    
2748    Version 2.02 14-Jan-99
2749    ----------------------
2750    
2751    1. Initialized the working variables associated with each extraction so that
2752    their saving and restoring doesn't refer to uninitialized store.
2753    
2754    2. Put dummy code into study.c in order to trick the optimizer of the IBM C
2755    compiler for OS/2 into generating correct code. Apparently IBM isn't going to
2756    fix the problem.
2757    
2758    3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
2759    calls, and wasn't printing the correct value for compiling calls. Increased the
2760    default value of LOOPREPEAT, and the number of significant figures in the
2761    times.
2762    
2763    4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
2764    
2765    5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
2766    a building problem on Windows NT with a FAT file system.
2767    
2768    
2769    Version 2.01 21-Oct-98
2770    ----------------------
2771    
2772    1. Changed the API for pcre_compile() to allow for the provision of a pointer
2773    to character tables built by pcre_maketables() in the current locale. If NULL
2774    is passed, the default tables are used.
2775    
2776    
2777    Version 2.00 24-Sep-98
2778    ----------------------
2779    
2780    1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
2781    it any more.
2782    
2783    2. Allow quantification of (?>) groups, and make it work correctly.
2784    
2785    3. The first character computation wasn't working for (?>) groups.
2786    
2787    4. Correct the implementation of \Z (it is permitted to match on the \n at the
2788    end of the subject) and add 5.005's \z, which really does match only at the
2789    very end of the subject.
2790    
2791    5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
2792    
2793    6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
2794    DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
2795    localized options. All options to pcre_study() were also removed.
2796    
2797    7. Add other new features from 5.005:
2798    
2799       $(?<=           positive lookbehind
2800       $(?<!           negative lookbehind
2801       (?imsx-imsx)    added the unsetting capability
2802                       such a setting is global if at outer level; local otherwise
2803       (?imsx-imsx:)   non-capturing groups with option setting
2804       (?(cond)re|re)  conditional pattern matching
2805    
2806       A backreference to itself in a repeated group matches the previous
2807       captured string.
2808    
2809    8. General tidying up of studying (both automatic and via "study")
2810    consequential on the addition of new assertions.
2811    
2812    9. As in 5.005, unlimited repeated groups that could match an empty substring
2813    are no longer faulted at compile time. Instead, the loop is forcibly broken at
2814    runtime if any iteration does actually match an empty substring.
2815    
2816    10. Include the RunTest script in the distribution.
2817    
2818    11. Added tests from the Perl 5.005_02 distribution. This showed up a few
2819    discrepancies, some of which were old and were also with respect to 5.004. They
2820    have now been fixed.
2821    
2822    
2823    Version 1.09 28-Apr-98
2824    ----------------------
2825    
2826    1. A negated single character class followed by a quantifier with a minimum
2827    value of one (e.g.  [^x]{1,6}  ) was not compiled correctly. This could lead to
2828    program crashes, or just wrong answers. This did not apply to negated classes
2829    containing more than one character, or to minima other than one.
2830    
2831    
2832    Version 1.08 27-Mar-98
2833    ----------------------
2834    
2835    1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
2836    
2837    2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
2838    latter must appear before anything that relies on it in the pattern.
2839    
2840    
2841    Version 1.07 16-Feb-98
2842    ----------------------
2843    
2844    1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
2845    repeat of a potentially empty string).
2846    
2847    
2848    Version 1.06 23-Jan-98
2849    ----------------------
2850    
2851    1. Added Markus Oberhumer's little patches for C++.
2852    
2853    2. Literal strings longer than 255 characters were broken.
2854    
2855    
2856    Version 1.05 23-Dec-97
2857    ----------------------
2858    
2859    1. Negated character classes containing more than one character were failing if
2860    PCRE_CASELESS was set at run time.
2861    
2862    
2863    Version 1.04 19-Dec-97
2864    ----------------------
2865    
2866    1. Corrected the man page, where some "const" qualifiers had been omitted.
2867    
2868    2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
2869    input syntax.
2870    
2871    3. Fixed memory leak which occurred when a regex with back references was
2872    matched with an offsets vector that wasn't big enough. The temporary memory
2873    that is used in this case wasn't being freed if the match failed.
2874    
2875    4. Tidied pcretest to ensure it frees memory that it gets.
2876    
2877    5. Temporary memory was being obtained in the case where the passed offsets
2878    vector was exactly big enough.
2879    
2880    6. Corrected definition of offsetof() from change 5 below.
2881    
2882    7. I had screwed up change 6 below and broken the rules for the use of
2883    setjmp(). Now fixed.
2884    
2885    
2886    Version 1.03 18-Dec-97
2887    ----------------------
2888    
2889    1. A erroneous regex with a missing opening parenthesis was correctly
2890    diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
2891    on some systems.
2892    
2893    2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
2894    it was reported that one broken compiler failed on the former because "code" is
2895    also an independent variable.
2896    
2897    3. The erroneous regex a[]b caused an array overrun reference.
2898    
2899    4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
2900    fail on data ending with that character. (It was going on too far, and checking
2901    the next character, typically a binary zero.) This was specific to the
2902    optimized code for single-character negative classes.
2903    
2904    5. Added a contributed patch from the TIN world which does the following:
2905    
2906      + Add an undef for memmove, in case the the system defines a macro for it.
2907    
2908      + Add a definition of offsetof(), in case there isn't one. (I don't know
2909        the reason behind this - offsetof() is part of the ANSI standard - but
2910        it does no harm).
2911    
2912      + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
2913        most of the places where whitespace preceded '#'. I have given up and
2914        allowed the remaining 2 cases to be at the margin.
2915    
2916      + Rename some variables in pcre to eliminate shadowing. This seems very
2917        pedantic, but does no harm, of course.
2918    
2919    6. Moved the call to setjmp() into its own function, to get rid of warnings
2920    from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
2921    
2922    7. Constructs such as \d{8,} were compiling into the equivalent of
2923    \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
2924    outcome, but in this particular case used more store than had been allocated,
2925    which caused the bug to be discovered because it threw up an internal error.
2926    
2927    8. The debugging code in both pcre and pcretest for outputting the compiled
2928    form of a regex was going wrong in the case of back references followed by
2929    curly-bracketed repeats.
2930    
2931    
2932    Version 1.02 12-Dec-97
2933    ----------------------
2934    
2935    1. Typos in pcre.3 and comments in the source fixed.
2936    
2937    2. Applied a contributed patch to get rid of places where it used to remove
2938    'const' from variables, and fixed some signed/unsigned and uninitialized
2939    variable warnings.
2940    
2941    3. Added the "runtest" target to Makefile.
2942    
2943    4. Set default compiler flag to -O2 rather than just -O.
2944    
2945    
2946    Version 1.01 19-Nov-97
2947    ----------------------
2948    
2949    1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
2950    like /([ab]*)*/, that is, for classes with more than one character in them.
2951    
2952    2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
2953    as /((?>a*))*/ (a PCRE_EXTRA facility).
2954    
2955    
2956    Version 1.00 18-Nov-97
2957    ----------------------
2958    
2959    1. Added compile-time macros to support systems such as SunOS4 which don't have
2960    memmove() or strerror() but have other things that can be used instead.
2961    
2962    2. Arranged that "make clean" removes the executables.
2963    
2964    
2965  Version 0.99 27-Oct-97  Version 0.99 27-Oct-97
2966  ----------------------  ----------------------
2967    

Legend:
Removed from v.3  
changed lines
  Added in v.294

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12