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

Diff of /code/trunk/ChangeLog

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

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

Legend:
Removed from v.49  
changed lines
  Added in v.284

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12