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

Diff of /code/trunk/ChangeLog

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

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

Legend:
Removed from v.29  
changed lines
  Added in v.292

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12