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

Diff of /code/trunk/ChangeLog

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

revision 61 by nigel, Sat Feb 24 21:39:58 2007 UTC revision 171 by ph10, Mon Jun 4 14:28:58 2007 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4  Version 3.0 02-Jan-02  Version 7.2 01-May-07
5    ---------------------
6    
7     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
8        which is apparently normally available under Windows.
9    
10     2. Re-jig the pcregrep tests with different newline settings in an attempt
11        to make them independent of the local environment's newline setting.
12    
13     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
14    
15     4. Some of the "internals" tests were previously cut out when the link size
16        was not 2, because the output contained actual offsets. The recent new
17        "Z" feature of pcretest means that these can be cut out, making the tests
18        usable with all link sizes.
19    
20     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
21        stack recursion. This gives a massive performance boost under BSD, but just
22        a small improvement under Linux. However, it saves one field in the frame
23        in all cases.
24    
25     6. Added more features from the forthcoming Perl 5.10:
26    
27        (a) (?-n) (where n is a string of digits) is a relative subroutine or
28            recursion call. It refers to the nth most recently opened parentheses.
29    
30        (b) (?+n) is also a relative subroutine call; it refers to the nth next
31            to be opened parentheses.
32    
33        (c) Conditions that refer to capturing parentheses can be specified
34            relatively, for example, (?(-2)... or (?(+3)...
35    
36        (d) \K resets the start of the current match so that everything before
37            is not part of it.
38    
39        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
40    
41        (f) \g{name} is another synonym - part of Perl 5.10's unification of
42            reference syntax.
43    
44     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
45        PCRE_INFO_JCHANGED.
46    
47     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
48        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
49        for detecting groups that can match an empty string.
50    
51    
52    Version 7.1 24-Apr-07
53    ---------------------
54    
55     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
56        that is more "standard", making use of automake and other Autotools. There
57        is some re-arrangement of the files and adjustment of comments consequent
58        on this.
59    
60     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
61        for recursive directory scanning broke on some systems because the files
62        are not scanned in any specific order and on different systems the order
63        was different. A call to "sort" has been inserted into RunGrepTest for the
64        approprate test as a short-term fix. In the longer term there may be an
65        alternative.
66    
67     3. I had an email from Eric Raymond about problems translating some of PCRE's
68        man pages to HTML (despite the fact that I distribute HTML pages, some
69        people do their own conversions for various reasons). The problems
70        concerned the use of low-level troff macros .br and .in. I have therefore
71        removed all such uses from the man pages (some were redundant, some could
72        be replaced by .nf/.fi pairs). The 132html script that I use to generate
73        HTML has been updated to handle .nf/.fi and to complain if it encounters
74        .br or .in.
75    
76     4. Updated comments in configure.ac that get placed in config.h.in and also
77        arranged for config.h to be included in the distribution, with the name
78        config.h.generic, for the benefit of those who have to compile without
79        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
80    
81     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
82        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
83        makevp.bat for the current PCRE, using the additional files
84        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
85    
86     6. A Windows user reported a minor discrepancy with test 2, which turned out
87        to be caused by a trailing space on an input line that had got lost in his
88        copy. The trailing space was an accident, so I've just removed it.
89    
90     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
91        that is needed.
92    
93     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
94        as "const" (a) because they are and (b) because it helps the PHP
95        maintainers who have recently made a script to detect big data structures
96        in the php code that should be moved to the .rodata section. I remembered
97        to update Builducptable as well, so it won't revert if ucptable.h is ever
98        re-created.
99    
100     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
101        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
102        order to be able to cut out the UTF-8 tables in the latter when UTF-8
103        support is not required. This saves 1.5-2K of code, which is important in
104        some applications.
105    
106        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
107        so as not to refer to the tables, even though these functions will never be
108        called when UTF-8 support is disabled. Otherwise there are problems with a
109        shared library.
110    
111    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
112    
113        (a) It was defining its arguments as char * instead of void *.
114    
115        (b) It was assuming that all moves were upwards in memory; this was true
116            a long time ago when I wrote it, but is no longer the case.
117    
118        The emulated memove() is provided for those environments that have neither
119        memmove() nor bcopy(). I didn't think anyone used it these days, but that
120        is clearly not the case, as these two bugs were recently reported.
121    
122    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
123        and Detrail to create the HTML documentation, the .txt form of the man
124        pages, and it removes trailing spaces from listed files. It also creates
125        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
126        case, it wraps all the #defines with #ifndefs. This script should be run
127        before "make dist".
128    
129    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
130        with Unicode property support.
131    
132        (a) For a maximizing quantifier, if the two different cases of the
133            character were of different lengths in their UTF-8 codings (there are
134            some cases like this - I found 11), and the matching function had to
135            back up over a mixture of the two cases, it incorrectly assumed they
136            were both the same length.
137    
138        (b) When PCRE was configured to use the heap rather than the stack for
139            recursion during matching, it was not correctly preserving the data for
140            the other case of a UTF-8 character when checking ahead for a match
141            while processing a minimizing repeat. If the check also involved
142            matching a wide character, but failed, corruption could cause an
143            erroneous result when trying to check for a repeat of the original
144            character.
145    
146    13. Some tidying changes to the testing mechanism:
147    
148        (a) The RunTest script now detects the internal link size and whether there
149            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
150            values substituted by "configure". (The RunGrepTest script already did
151            this for UTF-8.) The configure.ac script no longer substitutes the
152            relevant variables.
153    
154        (b) The debugging options /B and /D in pcretest show the compiled bytecode
155            with length and offset values. This means that the output is different
156            for different internal link sizes. Test 2 is skipped for link sizes
157            other than 2 because of this, bypassing the problem. Unfortunately,
158            there was also a test in test 3 (the locale tests) that used /B and
159            failed for link sizes other than 2. Rather than cut the whole test out,
160            I have added a new /Z option to pcretest that replaces the length and
161            offset values with spaces. This is now used to make test 3 independent
162            of link size. (Test 2 will be tidied up later.)
163    
164    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
165        segmentation fault instead of returning the appropriate error message.
166    
167    15. In multiline mode when the newline sequence was set to "any", the pattern
168        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
169        This doesn't seem right; it now treats the CRLF combination as the line
170        ending, and so does not match in that case. It's only a pattern such as ^$
171        that would hit this one: something like ^ABC$ would have failed after \r
172        and then tried again after \r\n.
173    
174    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
175        in an attempt to make files that differ only in their line terminators
176        compare equal. This works on Linux.
177    
178    17. Under certain error circumstances pcregrep might try to free random memory
179        as it exited. This is now fixed, thanks to valgrind.
180    
181    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
182        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
183        was because its rules for how to advance for /g after matching an empty
184        string at the end of a line did not allow for this case. They now check for
185        it specially.
186    
187    20. pcretest is supposed to handle patterns and data of any length, by
188        extending its buffers when necessary. It was getting this wrong when the
189        buffer for a data line had to be extended.
190    
191    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
192        CRLF as a newline sequence.
193    
194    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
195        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
196        I have nevertheless tidied it up.
197    
198    23. Added some casts to kill warnings from HP-UX ia64 compiler.
199    
200    24. Added a man page for pcre-config.
201    
202    
203    Version 7.0 19-Dec-06
204    ---------------------
205    
206     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
207        moving to gcc 4.1.1.
208    
209     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
210        sys/time.h, which is documented as needed for this function. It doesn't
211        seem to matter on Linux, but it showed up on some releases of OS X.
212    
213     3. It seems that there are systems where bytes whose values are greater than
214        127 match isprint() in the "C" locale. The "C" locale should be the
215        default when a C program starts up. In most systems, only ASCII printing
216        characters match isprint(). This difference caused the output from pcretest
217        to vary, making some of the tests fail. I have changed pcretest so that:
218    
219        (a) When it is outputting text in the compiled version of a pattern, bytes
220            other than 32-126 are always shown as hex escapes.
221    
222        (b) When it is outputting text that is a matched part of a subject string,
223            it does the same, unless a different locale has been set for the match
224            (using the /L modifier). In this case, it uses isprint() to decide.
225    
226     4. Fixed a major bug that caused incorrect computation of the amount of memory
227        required for a compiled pattern when options that changed within the
228        pattern affected the logic of the preliminary scan that determines the
229        length. The relevant options are -x, and -i in UTF-8 mode. The result was
230        that the computed length was too small. The symptoms of this bug were
231        either the PCRE error "internal error: code overflow" from pcre_compile(),
232        or a glibc crash with a message such as "pcretest: free(): invalid next
233        size (fast)". Examples of patterns that provoked this bug (shown in
234        pcretest format) are:
235    
236          /(?-x: )/x
237          /(?x)(?-x: \s*#\s*)/
238          /((?i)[\x{c0}])/8
239          /(?i:[\x{c0}])/8
240    
241        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
242        is now done differently.
243    
244     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
245        wrapper classes; (b) implement a new function in the C++ scanner that is
246        more efficient than the old way of doing things because it avoids levels of
247        recursion in the regex matching; (c) add a paragraph to the documentation
248        for the FullMatch() function.
249    
250     6. The escape sequence \n was being treated as whatever was defined as
251        "newline". Not only was this contrary to the documentation, which states
252        that \n is character 10 (hex 0A), but it also went horribly wrong when
253        "newline" was defined as CRLF. This has been fixed.
254    
255     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
256        was being set to -1 for the "end of line" case (supposedly a value that no
257        character can have). Though this value is never used (the check for end of
258        line is "zero bytes in current character"), it caused compiler complaints.
259        I've changed it to 0xffffffff.
260    
261     8. In pcre_version.c, the version string was being built by a sequence of
262        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
263        string (as it is for production releases) called a macro with an empty
264        argument. The C standard says the result of this is undefined. The gcc
265        compiler treats it as an empty string (which was what was wanted) but it is
266        reported that Visual C gives an error. The source has been hacked around to
267        avoid this problem.
268    
269     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
270        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
271        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
272        of them did).
273    
274    10. Originally, pcretest opened its input and output without "b"; then I was
275        told that "b" was needed in some environments, so it was added for release
276        5.0 to both the input and output. (It makes no difference on Unix-like
277        systems.) Later I was told that it is wrong for the input on Windows. I've
278        now abstracted the modes into two macros, to make it easier to fiddle with
279        them, and removed "b" from the input mode under Windows.
280    
281    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
282    
283    12. Added -help and --help to pcretest as an official way of being reminded
284        of the options.
285    
286    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
287        and pcrecpp.cc because they annoy compilers at high warning levels.
288    
289    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
290    
291    15. Fixed an occurrence of == in configure.ac that should have been = (shell
292        scripts are not C programs :-) and which was not noticed because it works
293        on Linux.
294    
295    16. pcretest is supposed to handle any length of pattern and data line (as one
296        line or as a continued sequence of lines) by extending its input buffer if
297        necessary. This feature was broken for very long pattern lines, leading to
298        a string of junk being passed to pcre_compile() if the pattern was longer
299        than about 50K.
300    
301    17. I have done a major re-factoring of the way pcre_compile() computes the
302        amount of memory needed for a compiled pattern. Previously, there was code
303        that made a preliminary scan of the pattern in order to do this. That was
304        OK when PCRE was new, but as the facilities have expanded, it has become
305        harder and harder to keep it in step with the real compile phase, and there
306        have been a number of bugs (see for example, 4 above). I have now found a
307        cunning way of running the real compile function in a "fake" mode that
308        enables it to compute how much memory it would need, while actually only
309        ever using a few hundred bytes of working memory and without too many
310        tests of the mode. This should make future maintenance and development
311        easier. A side effect of this work is that the limit of 200 on the nesting
312        depth of parentheses has been removed (though this was never a serious
313        limitation, I suspect). However, there is a downside: pcre_compile() now
314        runs more slowly than before (30% or more, depending on the pattern). I
315        hope this isn't a big issue. There is no effect on runtime performance.
316    
317    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
318        newline (only possible for the last line of a file) and it was a
319        pattern that set a locale (followed by /Lsomething), pcretest crashed.
320    
321    19. Added additional timing features to pcretest. (1) The -tm option now times
322        matching only, not compiling. (2) Both -t and -tm can be followed, as a
323        separate command line item, by a number that specifies the number of
324        repeats to use when timing. The default is 50000; this gives better
325        precision, but takes uncomfortably long for very large patterns.
326    
327    20. Extended pcre_study() to be more clever in cases where a branch of a
328        subpattern has no definite first character. For example, (a*|b*)[cd] would
329        previously give no result from pcre_study(). Now it recognizes that the
330        first character must be a, b, c, or d.
331    
332    21. There was an incorrect error "recursive call could loop indefinitely" if
333        a subpattern (or the entire pattern) that was being tested for matching an
334        empty string contained only one non-empty item after a nested subpattern.
335        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
336        incorrectly, because the \d was being skipped in the check.
337    
338    22. The pcretest program now has a new pattern option /B and a command line
339        option -b, which is equivalent to adding /B to every pattern. This causes
340        it to show the compiled bytecode, without the additional information that
341        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
342        is the same as /B/I).
343    
344    23. A new optimization is now able automatically to treat some sequences such
345        as a*b as a*+b. More specifically, if something simple (such as a character
346        or a simple class like \d) has an unlimited quantifier, and is followed by
347        something that cannot possibly match the quantified thing, the quantifier
348        is automatically "possessified".
349    
350    24. A recursive reference to a subpattern whose number was greater than 39
351        went wrong under certain circumstances in UTF-8 mode. This bug could also
352        have affected the operation of pcre_study().
353    
354    25. Realized that a little bit of performance could be had by replacing
355        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
356    
357    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
358    
359    27. Possessive quantifiers such as a++ were previously implemented by turning
360        them into atomic groups such as ($>a+). Now they have their own opcodes,
361        which improves performance. This includes the automatically created ones
362        from 23 above.
363    
364    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
365        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
366        the first matched character to be a colon. This applied both to named and
367        numbered groups.
368    
369    29. The ucpinternal.h header file was missing its idempotency #ifdef.
370    
371    30. I was sent a "project" file called libpcre.a.dev which I understand makes
372        building PCRE on Windows easier, so I have included it in the distribution.
373    
374    31. There is now a check in pcretest against a ridiculously large number being
375        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
376        loop, the loop is abandoned.
377    
378    32. Forward references to subpatterns in conditions such as (?(2)...) where
379        subpattern 2 is defined later cause pcre_compile() to search forwards in
380        the pattern for the relevant set of parentheses. This search went wrong
381        when there were unescaped parentheses in a character class, parentheses
382        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
383    
384    33. "Subroutine" calls and backreferences were previously restricted to
385        referencing subpatterns earlier in the regex. This restriction has now
386        been removed.
387    
388    34. Added a number of extra features that are going to be in Perl 5.10. On the
389        whole, these are just syntactic alternatives for features that PCRE had
390        previously implemented using the Python syntax or my own invention. The
391        other formats are all retained for compatibility.
392    
393        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
394            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
395            also .NET compatible.
396    
397        (b) A recursion or subroutine call to a named group can now be defined as
398            (?&name) as well as (?P>name).
399    
400        (c) A backreference to a named group can now be defined as \k<name> or
401            \k'name' as well as (?P=name). The new forms, as well as being in Perl
402            5.10, are also .NET compatible.
403    
404        (d) A conditional reference to a named group can now use the syntax
405            (?(<name>) or (?('name') as well as (?(name).
406    
407        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
408            groups (named and numbered) that are never evaluated inline, but can be
409            called as "subroutines" from elsewhere. In effect, the DEFINE condition
410            is always false. There may be only one alternative in such a group.
411    
412        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
413            as the simple (?(R). The condition is true only if the most recent
414            recursion is that of the given number or name. It does not search out
415            through the entire recursion stack.
416    
417        (g) The escape \gN or \g{N} has been added, where N is a positive or
418            negative number, specifying an absolute or relative reference.
419    
420    35. Tidied to get rid of some further signed/unsigned compiler warnings and
421        some "unreachable code" warnings.
422    
423    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
424        things, this adds five new scripts.
425    
426    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
427        There were also incompatibilities regarding the handling of \Q..\E inside
428        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
429        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
430    
431    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
432        matches an empty string, and forcibly breaks the loop. There were bugs in
433        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
434        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
435        separate and independent bugs (that affected different cases) have been
436        fixed.
437    
438    39. Refactored the code to abolish the use of different opcodes for small
439        capturing bracket numbers. This is a tidy that I avoided doing when I
440        removed the limit on the number of capturing brackets for 3.5 back in 2001.
441        The new approach is not only tidier, it makes it possible to reduce the
442        memory needed to fix the previous bug (38).
443    
444    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
445        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
446        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
447        mode.
448    
449    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
450        report.
451    
452    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
453        copy construction and assignment in the C++ wrapper.
454    
455    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
456        couple of bugs that could have given wrong results in the "--newline=crlf"
457        case.
458    
459    44. Added a number of casts and did some reorganization of signed/unsigned int
460        variables following suggestions from Dair Grant. Also renamed the variable
461        "this" as "item" because it is a C++ keyword.
462    
463    45. Arranged for dftables to add
464    
465          #include "pcre_internal.h"
466    
467        to pcre_chartables.c because without it, gcc 4.x may remove the array
468        definition from the final binary if PCRE is built into a static library and
469        dead code stripping is activated.
470    
471    46. For an unanchored pattern, if a match attempt fails at the start of a
472        newline sequence, and the newline setting is CRLF or ANY, and the next two
473        characters are CRLF, advance by two characters instead of one.
474    
475    
476    Version 6.7 04-Jul-06
477    ---------------------
478    
479     1. In order to handle tests when input lines are enormously long, pcretest has
480        been re-factored so that it automatically extends its buffers when
481        necessary. The code is crude, but this _is_ just a test program. The
482        default size has been increased from 32K to 50K.
483    
484     2. The code in pcre_study() was using the value of the re argument before
485        testing it for NULL. (Of course, in any sensible call of the function, it
486        won't be NULL.)
487    
488     3. The memmove() emulation function in pcre_internal.h, which is used on
489        systems that lack both memmove() and bcopy() - that is, hardly ever -
490        was missing a "static" storage class specifier.
491    
492     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
493        containing an extended class (one that cannot be represented by a bitmap
494        because it contains high-valued characters or Unicode property items, e.g.
495        [\pZ]). Almost always one would set UTF-8 mode when processing such a
496        pattern, but PCRE should not loop if you do not (it no longer does).
497        [Detail: two cases were found: (a) a repeated subpattern containing an
498        extended class; (b) a recursive reference to a subpattern that followed a
499        previous extended class. It wasn't skipping over the extended class
500        correctly when UTF-8 mode was not set.]
501    
502     5. A negated single-character class was not being recognized as fixed-length
503        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
504        compile error "lookbehind assertion is not fixed length".
505    
506     6. The RunPerlTest auxiliary script was showing an unexpected difference
507        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
508        write a Perl script that can interpret lines of an input file either as
509        byte characters or as UTF-8, which is what "perltest" was being required to
510        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
511        can't do is switch easily at run time between having the "use utf8;" pragma
512        or not. In the end, I fudged it by using the RunPerlTest script to insert
513        "use utf8;" explicitly for the UTF-8 tests.
514    
515     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
516        the end of the subject string, contrary to the documentation and to what
517        Perl does. This was true of both matching functions. Now it matches only at
518        the start of the subject and immediately after *internal* newlines.
519    
520     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
521        a pointer to an int instead of a pointer to an unsigned long int. This
522        caused problems on 64-bit systems.
523    
524     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
525        instance of the 'standard' template library not being so standard".
526    
527    10. There was no check on the number of named subpatterns nor the maximum
528        length of a subpattern name. The product of these values is used to compute
529        the size of the memory block for a compiled pattern. By supplying a very
530        long subpattern name and a large number of named subpatterns, the size
531        computation could be caused to overflow. This is now prevented by limiting
532        the length of names to 32 characters, and the number of named subpatterns
533        to 10,000.
534    
535    11. Subpatterns that are repeated with specific counts have to be replicated in
536        the compiled pattern. The size of memory for this was computed from the
537        length of the subpattern and the repeat count. The latter is limited to
538        65535, but there was no limit on the former, meaning that integer overflow
539        could in principle occur. The compiled length of a repeated subpattern is
540        now limited to 30,000 bytes in order to prevent this.
541    
542    12. Added the optional facility to have named substrings with the same name.
543    
544    13. Added the ability to use a named substring as a condition, using the
545        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
546        are numbers (not recommended). Forward references are permitted.
547    
548    14. Added forward references in named backreferences (if you see what I mean).
549    
550    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
551        pattern could run off the end of the subject. For example, the pattern
552        "(?s)(.{1,5})"8 did this with the subject "ab".
553    
554    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
555        PCRE_CASELESS was set when matching characters that were quantified with ?
556        or *.
557    
558    17. A character class other than a single negated character that had a minimum
559        but no maximum quantifier - for example [ab]{6,} - was not handled
560        correctly by pce_dfa_exec(). It would match only one character.
561    
562    18. A valid (though odd) pattern that looked like a POSIX character
563        class but used an invalid character after [ (for example [[,abc,]]) caused
564        pcre_compile() to give the error "Failed: internal error: code overflow" or
565        in some cases to crash with a glibc free() error. This could even happen if
566        the pattern terminated after [[ but there just happened to be a sequence of
567        letters, a binary zero, and a closing ] in the memory that followed.
568    
569    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
570        over the years. Originally (before any Unicode support), just the bottom 8
571        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
572        output from "man perlunicode" includes this:
573    
574          The regular expression compiler produces polymorphic opcodes.  That
575          is, the pattern adapts to the data and automatically switches to
576          the Unicode character scheme when presented with Unicode data--or
577          instead uses a traditional byte scheme when presented with byte
578          data.
579    
580        Sadly, a wide octal escape does not cause a switch, and in a string with
581        no other multibyte characters, these octal escapes are treated as before.
582        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
583        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
584        Unicode string.
585    
586        I have not perpetrated such confusion in PCRE. Up till now, it took just
587        the bottom 8 bits, as in old Perl. I have now made octal escapes with
588        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
589        translate to the appropriate multibyte character.
590    
591    29. Applied some refactoring to reduce the number of warnings from Microsoft
592        and Borland compilers. This has included removing the fudge introduced
593        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
594        a warning about an unused variable.
595    
596    21. PCRE has not included VT (character 0x0b) in the set of whitespace
597        characters since release 4.0, because Perl (from release 5.004) does not.
598        [Or at least, is documented not to: some releases seem to be in conflict
599        with the documentation.] However, when a pattern was studied with
600        pcre_study() and all its branches started with \s, PCRE still included VT
601        as a possible starting character. Of course, this did no harm; it just
602        caused an unnecessary match attempt.
603    
604    22. Removed a now-redundant internal flag bit that recorded the fact that case
605        dependency changed within the pattern. This was once needed for "required
606        byte" processing, but is no longer used. This recovers a now-scarce options
607        bit. Also moved the least significant internal flag bit to the most-
608        significant bit of the word, which was not previously used (hangover from
609        the days when it was an int rather than a uint) to free up another bit for
610        the future.
611    
612    23. Added support for CRLF line endings as well as CR and LF. As well as the
613        default being selectable at build time, it can now be changed at runtime
614        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
615        specify that it is scanning data with non-default line endings.
616    
617    24. Changed the definition of CXXLINK to make it agree with the definition of
618        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
619    
620    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
621        recursions. This makes a big different to stack usage for some patterns.
622    
623    26. If a subpattern containing a named recursion or subroutine reference such
624        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
625        the space required for the compiled pattern went wrong and gave too small a
626        value. Depending on the environment, this could lead to "Failed: internal
627        error: code overflow at offset 49" or "glibc detected double free or
628        corruption" errors.
629    
630    27. Applied patches from Google (a) to support the new newline modes and (b) to
631        advance over multibyte UTF-8 characters in GlobalReplace.
632    
633    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
634        difference for some implementation of PCRE in some Windows version.
635    
636    29. Added some extra testing facilities to pcretest:
637    
638        \q<number>   in a data line sets the "match limit" value
639        \Q<number>   in a data line sets the "match recursion limt" value
640        -S <number>  sets the stack size, where <number> is in megabytes
641    
642        The -S option isn't available for Windows.
643    
644    
645    Version 6.6 06-Feb-06
646    ---------------------
647    
648     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
649        in pcreposix.h. I have copied the definition from pcre.h.
650    
651     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
652        because pcre.h is no longer a built file.
653    
654     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
655        not normally included in the compiled code.
656    
657    
658    Version 6.5 01-Feb-06
659    ---------------------
660    
661     1. When using the partial match feature with pcre_dfa_exec(), it was not
662        anchoring the second and subsequent partial matches at the new starting
663        point. This could lead to incorrect results. For example, with the pattern
664        /1234/, partially matching against "123" and then "a4" gave a match.
665    
666     2. Changes to pcregrep:
667    
668        (a) All non-match returns from pcre_exec() were being treated as failures
669            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
670            error message is output. Some extra information is given for the
671            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
672            probably the only errors that are likely to be caused by users (by
673            specifying a regex that has nested indefinite repeats, for instance).
674            If there are more than 20 of these errors, pcregrep is abandoned.
675    
676        (b) A binary zero was treated as data while matching, but terminated the
677            output line if it was written out. This has been fixed: binary zeroes
678            are now no different to any other data bytes.
679    
680        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
681            used to set a locale for matching. The --locale=xxxx long option has
682            been added (no short equivalent) to specify a locale explicitly on the
683            pcregrep command, overriding the environment variables.
684    
685        (d) When -B was used with -n, some line numbers in the output were one less
686            than they should have been.
687    
688        (e) Added the -o (--only-matching) option.
689    
690        (f) If -A or -C was used with -c (count only), some lines of context were
691            accidentally printed for the final match.
692    
693        (g) Added the -H (--with-filename) option.
694    
695        (h) The combination of options -rh failed to suppress file names for files
696            that were found from directory arguments.
697    
698        (i) Added the -D (--devices) and -d (--directories) options.
699    
700        (j) Added the -F (--fixed-strings) option.
701    
702        (k) Allow "-" to be used as a file name for -f as well as for a data file.
703    
704        (l) Added the --colo(u)r option.
705    
706        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
707            is not present by default.
708    
709     3. A nasty bug was discovered in the handling of recursive patterns, that is,
710        items such as (?R) or (?1), when the recursion could match a number of
711        alternatives. If it matched one of the alternatives, but subsequently,
712        outside the recursion, there was a failure, the code tried to back up into
713        the recursion. However, because of the way PCRE is implemented, this is not
714        possible, and the result was an incorrect result from the match.
715    
716        In order to prevent this happening, the specification of recursion has
717        been changed so that all such subpatterns are automatically treated as
718        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
719    
720     4. I had overlooked the fact that, in some locales, there are characters for
721        which isalpha() is true but neither isupper() nor islower() are true. In
722        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
723        and ordfeminine) are like this. This affected the treatment of \w and \W
724        when they appeared in character classes, but not when they appeared outside
725        a character class. The bit map for "word" characters is now created
726        separately from the results of isalnum() instead of just taking it from the
727        upper, lower, and digit maps. (Plus the underscore character, of course.)
728    
729     5. The above bug also affected the handling of POSIX character classes such as
730        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
731        permanent tables. Instead, the bit maps for such a class were previously
732        created as the appropriate unions of the upper, lower, and digit bitmaps.
733        Now they are created by subtraction from the [[:word:]] class, which has
734        its own bitmap.
735    
736     6. The [[:blank:]] character class matches horizontal, but not vertical space.
737        It is created by subtracting the vertical space characters (\x09, \x0a,
738        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
739        subtraction was done in the overall bitmap for a character class, meaning
740        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
741        be recognized. This bug has been fixed.
742    
743     7. Patches from the folks at Google:
744    
745          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
746          real life, but is still worth protecting against".
747    
748          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
749          regular expressions".
750    
751          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
752          have it.
753    
754          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
755          "configure" and the latter not, in order to fix a problem somebody had
756          with compiling the Arg class on HP-UX.
757    
758          (e) Improve the error-handling of the C++ wrapper a little bit.
759    
760          (f) New tests for checking recursion limiting.
761    
762     8. The pcre_memmove() function, which is used only if the environment does not
763        have a standard memmove() function (and is therefore rarely compiled),
764        contained two bugs: (a) use of int instead of size_t, and (b) it was not
765        returning a result (though PCRE never actually uses the result).
766    
767     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
768        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
769        returned instead of calling malloc() with an overflowing number that would
770        most likely cause subsequent chaos.
771    
772    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
773    
774    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
775        with this option is matched, the nmatch and pmatch options of regexec() are
776        ignored.
777    
778    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
779        provided in case anyone wants to the the POSIX interface with UTF-8
780        strings.
781    
782    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
783        C++ linking (needed for some HP-UX environments).
784    
785    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
786        (unused parameter) and in the pcre_printint() function (omitted "default"
787        switch label when the default is to do nothing).
788    
789    15. Added some code to make it possible, when PCRE is compiled as a C++
790        library, to replace subject pointers for pcre_exec() with a smart pointer
791        class, thus making it possible to process discontinuous strings.
792    
793    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
794        much the same function. They were added by different people who were trying
795        to make PCRE easy to compile on non-Unix systems. It has been suggested
796        that PCRE_EXPORT be abolished now that there is more automatic apparatus
797        for compiling on Windows systems. I have therefore replaced it with
798        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
799        defaults to "extern" for C or "extern C" for C++, which works fine on
800        Unix-like systems. It is now possible to override the value of PCRE_DATA_
801        SCOPE with something explicit in config.h. In addition:
802    
803        (a) pcreposix.h still had just "extern" instead of either of these macros;
804            I have replaced it with PCRE_DATA_SCOPE.
805    
806        (b) Functions such as _pcre_xclass(), which are internal to the library,
807            but external in the C sense, all had PCRE_EXPORT in their definitions.
808            This is apparently wrong for the Windows case, so I have removed it.
809            (It makes no difference on Unix-like systems.)
810    
811    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
812        of recursive calls to match(). This is different to MATCH_LIMIT because
813        that limits the total number of calls to match(), not all of which increase
814        the depth of recursion. Limiting the recursion depth limits the amount of
815        stack (or heap if NO_RECURSE is set) that is used. The default can be set
816        when PCRE is compiled, and changed at run time. A patch from Google adds
817        this functionality to the C++ interface.
818    
819    18. Changes to the handling of Unicode character properties:
820    
821        (a) Updated the table to Unicode 4.1.0.
822    
823        (b) Recognize characters that are not in the table as "Cn" (undefined).
824    
825        (c) I revised the way the table is implemented to a much improved format
826            which includes recognition of ranges. It now supports the ranges that
827            are defined in UnicodeData.txt, and it also amalgamates other
828            characters into ranges. This has reduced the number of entries in the
829            table from around 16,000 to around 3,000, thus reducing its size
830            considerably. I realized I did not need to use a tree structure after
831            all - a binary chop search is just as efficient. Having reduced the
832            number of entries, I extended their size from 6 bytes to 8 bytes to
833            allow for more data.
834    
835        (d) Added support for Unicode script names via properties such as \p{Han}.
836    
837    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
838        matching that character.
839    
840    20. When matching a repeated Unicode property with a minimum greater than zero,
841        (for example \pL{2,}), PCRE could look past the end of the subject if it
842        reached it while seeking the minimum number of characters. This could
843        happen only if some of the characters were more than one byte long, because
844        there is a check for at least the minimum number of bytes.
845    
846    21. Refactored the implementation of \p and \P so as to be more general, to
847        allow for more different types of property in future. This has changed the
848        compiled form incompatibly. Anybody with saved compiled patterns that use
849        \p or \P will have to recompile them.
850    
851    22. Added "Any" and "L&" to the supported property types.
852    
853    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
854        but give a compile time error if the value is greater than 0xff.
855    
856    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
857        accidentally not being installed or uninstalled.
858    
859    25. The pcre.h file was built from pcre.h.in, but the only changes that were
860        made were to insert the current release number. This seemed silly, because
861        it made things harder for people building PCRE on systems that don't run
862        "configure". I have turned pcre.h into a distributed file, no longer built
863        by "configure", with the version identification directly included. There is
864        no longer a pcre.h.in file.
865    
866        However, this change necessitated a change to the pcre-config script as
867        well. It is built from pcre-config.in, and one of the substitutions was the
868        release number. I have updated configure.ac so that ./configure now finds
869        the release number by grepping pcre.h.
870    
871    26. Added the ability to run the tests under valgrind.
872    
873    
874    Version 6.4 05-Sep-05
875    ---------------------
876    
877     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
878        "--" to be printed when multiple files were scanned, even when none of the
879        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
880        consider it to be a bug, and have restored the previous behaviour.
881    
882     2. A couple of code tidies to get rid of compiler warnings.
883    
884     3. The pcretest program used to cheat by referring to symbols in the library
885        whose names begin with _pcre_. These are internal symbols that are not
886        really supposed to be visible externally, and in some environments it is
887        possible to suppress them. The cheating is now confined to including
888        certain files from the library's source, which is a bit cleaner.
889    
890     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
891        file's purpose clearer.
892    
893     5. Reorganized pcre_ucp_findchar().
894    
895    
896    Version 6.3 15-Aug-05
897    ---------------------
898    
899     1. The file libpcre.pc.in did not have general read permission in the tarball.
900    
901     2. There were some problems when building without C++ support:
902    
903        (a) If C++ support was not built, "make install" and "make test" still
904            tried to test it.
905    
906        (b) There were problems when the value of CXX was explicitly set. Some
907            changes have been made to try to fix these, and ...
908    
909        (c) --disable-cpp can now be used to explicitly disable C++ support.
910    
911        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
912            backslash in a target when C++ was disabled. This confuses some
913            versions of "make", apparently. Using an intermediate variable solves
914            this. (Same for CPP_LOBJ.)
915    
916     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
917        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
918        necessary on certain architectures.
919    
920     4. Added a setting of -export-symbols-regex to the link command to remove
921        those symbols that are exported in the C sense, but actually are local
922        within the library, and not documented. Their names all begin with
923        "_pcre_". This is not a perfect job, because (a) we have to except some
924        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
925        available (and never for static libraries). I have made a note to try to
926        find a way round (a) in the future.
927    
928    
929    Version 6.2 01-Aug-05
930    ---------------------
931    
932     1. There was no test for integer overflow of quantifier values. A construction
933        such as {1111111111111111} would give undefined results. What is worse, if
934        a minimum quantifier for a parenthesized subpattern overflowed and became
935        negative, the calculation of the memory size went wrong. This could have
936        led to memory overwriting.
937    
938     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
939    
940     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
941        operating environments where this matters.
942    
943     4. Applied Giuseppe Maxia's patch to add additional features for controlling
944        PCRE options from within the C++ wrapper.
945    
946     5. Named capturing subpatterns were not being correctly counted when a pattern
947        was compiled. This caused two problems: (a) If there were more than 100
948        such subpatterns, the calculation of the memory needed for the whole
949        compiled pattern went wrong, leading to an overflow error. (b) Numerical
950        back references of the form \12, where the number was greater than 9, were
951        not recognized as back references, even though there were sufficient
952        previous subpatterns.
953    
954     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
955        versions of gcc, e.g. 2.95.4.
956    
957    
958    Version 6.1 21-Jun-05
959    ---------------------
960    
961     1. There was one reference to the variable "posix" in pcretest.c that was not
962        surrounded by "#if !defined NOPOSIX".
963    
964     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
965        the cross-check on the old pcre_info() function, for the benefit of the
966        cut-down version of PCRE that is currently imported into Exim.
967    
968     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
969        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
970        patterns that start (?i) but I don't think that matters. The use of (?i) is
971        just an example; this all applies to the other options as well.
972    
973     4. Since libtool seems to echo the compile commands it is issuing, the output
974        from "make" can be reduced a bit by putting "@" in front of each libtool
975        compile command.
976    
977     5. Patch from the folks at Google for configure.in to be a bit more thorough
978        in checking for a suitable C++ installation before trying to compile the
979        C++ stuff. This should fix a reported problem when a compiler was present,
980        but no suitable headers.
981    
982     6. The man pages all had just "PCRE" as their title. I have changed them to
983        be the relevant file name. I have also arranged that these names are
984        retained in the file doc/pcre.txt, which is a concatenation in text format
985        of all the man pages except the little individual ones for each function.
986    
987     7. The NON-UNIX-USE file had not been updated for the different set of source
988        files that come with release 6. I also added a few comments about the C++
989        wrapper.
990    
991    
992    Version 6.0 07-Jun-05
993    ---------------------
994    
995     1. Some minor internal re-organization to help with my DFA experiments.
996    
997     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
998        didn't matter for the library itself when fully configured, but did matter
999        when compiling without UCP support, or within Exim, where the ucp files are
1000        not imported.
1001    
1002     3. Refactoring of the library code to split up the various functions into
1003        different source modules. The addition of the new DFA matching code (see
1004        below) to a single monolithic source would have made it really too
1005        unwieldy, quite apart from causing all the code to be include in a
1006        statically linked application, when only some functions are used. This is
1007        relevant even without the DFA addition now that patterns can be compiled in
1008        one application and matched in another.
1009    
1010        The downside of splitting up is that there have to be some external
1011        functions and data tables that are used internally in different modules of
1012        the library but which are not part of the API. These have all had their
1013        names changed to start with "_pcre_" so that they are unlikely to clash
1014        with other external names.
1015    
1016     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1017        a different (DFA) algorithm. Although it is slower than the original
1018        function, it does have some advantages for certain types of matching
1019        problem.
1020    
1021     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1022        including restarting after a partial match.
1023    
1024     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1025        defined when compiling for Windows was sent to me. I have put it into the
1026        code, though I have no means of testing or verifying it.
1027    
1028     7. Added the pcre_refcount() auxiliary function.
1029    
1030     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1031        match before or at the first newline in the subject string. In pcretest,
1032        the /f option on a pattern can be used to set this.
1033    
1034     9. A repeated \w when used in UTF-8 mode with characters greater than 256
1035        would behave wrongly. This has been present in PCRE since release 4.0.
1036    
1037    10. A number of changes to the pcregrep command:
1038    
1039        (a) Refactored how -x works; insert ^(...)$ instead of setting
1040            PCRE_ANCHORED and checking the length, in preparation for adding
1041            something similar for -w.
1042    
1043        (b) Added the -w (match as a word) option.
1044    
1045        (c) Refactored the way lines are read and buffered so as to have more
1046            than one at a time available.
1047    
1048        (d) Implemented a pcregrep test script.
1049    
1050        (e) Added the -M (multiline match) option. This allows patterns to match
1051            over several lines of the subject. The buffering ensures that at least
1052            8K, or the rest of the document (whichever is the shorter) is available
1053            for matching (and similarly the previous 8K for lookbehind assertions).
1054    
1055        (f) Changed the --help output so that it now says
1056    
1057              -w, --word-regex(p)
1058    
1059            instead of two lines, one with "regex" and the other with "regexp"
1060            because that confused at least one person since the short forms are the
1061            same. (This required a bit of code, as the output is generated
1062            automatically from a table. It wasn't just a text change.)
1063    
1064        (g) -- can be used to terminate pcregrep options if the next thing isn't an
1065            option but starts with a hyphen. Could be a pattern or a path name
1066            starting with a hyphen, for instance.
1067    
1068        (h) "-" can be given as a file name to represent stdin.
1069    
1070        (i) When file names are being printed, "(standard input)" is used for
1071            the standard input, for compatibility with GNU grep. Previously
1072            "<stdin>" was used.
1073    
1074        (j) The option --label=xxx can be used to supply a name to be used for
1075            stdin when file names are being printed. There is no short form.
1076    
1077        (k) Re-factored the options decoding logic because we are going to add
1078            two more options that take data. Such options can now be given in four
1079            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1080    
1081        (l) Added the -A, -B, and -C options for requesting that lines of context
1082            around matches be printed.
1083    
1084        (m) Added the -L option to print the names of files that do not contain
1085            any matching lines, that is, the complement of -l.
1086    
1087        (n) The return code is 2 if any file cannot be opened, but pcregrep does
1088            continue to scan other files.
1089    
1090        (o) The -s option was incorrectly implemented. For compatibility with other
1091            greps, it now suppresses the error message for a non-existent or non-
1092            accessible file (but not the return code). There is a new option called
1093            -q that suppresses the output of matching lines, which was what -s was
1094            previously doing.
1095    
1096        (p) Added --include and --exclude options to specify files for inclusion
1097            and exclusion when recursing.
1098    
1099    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1100        Hopefully, it now does.
1101    
1102    12. Missing cast in pcre_study().
1103    
1104    13. Added an "uninstall" target to the makefile.
1105    
1106    14. Replaced "extern" in the function prototypes in Makefile.in with
1107        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1108        world, but is set differently for Windows.
1109    
1110    15. Added a second compiling function called pcre_compile2(). The only
1111        difference is that it has an extra argument, which is a pointer to an
1112        integer error code. When there is a compile-time failure, this is set
1113        non-zero, in addition to the error test pointer being set to point to an
1114        error message. The new argument may be NULL if no error number is required
1115        (but then you may as well call pcre_compile(), which is now just a
1116        wrapper). This facility is provided because some applications need a
1117        numeric error indication, but it has also enabled me to tidy up the way
1118        compile-time errors are handled in the POSIX wrapper.
1119    
1120    16. Added VPATH=.libs to the makefile; this should help when building with one
1121        prefix path and installing with another. (Or so I'm told by someone who
1122        knows more about this stuff than I do.)
1123    
1124    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1125        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1126        match everything, including newlines. This is not POSIX-compatible, but
1127        somebody wanted the feature. From pcretest it can be activated by using
1128        both the P and the s flags.
1129    
1130    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1131    
1132    19. libpcre.pc was being incorrectly installed as executable.
1133    
1134    20. A couple of places in pcretest check for end-of-line by looking for '\n';
1135        it now also looks for '\r' so that it will work unmodified on Windows.
1136    
1137    21. Added Google's contributed C++ wrapper to the distribution.
1138    
1139    22. Added some untidy missing memory free() calls in pcretest, to keep
1140        Electric Fence happy when testing.
1141    
1142    
1143    
1144    Version 5.0 13-Sep-04
1145    ---------------------
1146    
1147     1. Internal change: literal characters are no longer packed up into items
1148        containing multiple characters in a single byte-string. Each character
1149        is now matched using a separate opcode. However, there may be more than one
1150        byte in the character in UTF-8 mode.
1151    
1152     2. The pcre_callout_block structure has two new fields: pattern_position and
1153        next_item_length. These contain the offset in the pattern to the next match
1154        item, and its length, respectively.
1155    
1156     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1157        insertion of callouts before each pattern item. Added the /C option to
1158        pcretest to make use of this.
1159    
1160     4. On the advice of a Windows user, the lines
1161    
1162          #if defined(_WIN32) || defined(WIN32)
1163          _setmode( _fileno( stdout ), 0x8000 );
1164          #endif  /* defined(_WIN32) || defined(WIN32) */
1165    
1166        have been added to the source of pcretest. This apparently does useful
1167        magic in relation to line terminators.
1168    
1169     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1170        for the benefit of those environments where the "b" makes a difference.
1171    
1172     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1173        to know about it. I have put a hack into configure.in that adds in code
1174        to set GCC=yes if CC=icc. This seems to end up at a point in the
1175        generated configure script that is early enough to affect the setting of
1176        compiler options, which is what is needed, but I have no means of testing
1177        whether it really works. (The user who reported this had patched the
1178        generated configure script, which of course I cannot do.)
1179    
1180        LATER: After change 22 below (new libtool files), the configure script
1181        seems to know about icc (and also ecc). Therefore, I have commented out
1182        this hack in configure.in.
1183    
1184     7. Added support for pkg-config (2 patches were sent in).
1185    
1186     8. Negated POSIX character classes that used a combination of internal tables
1187        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1188        [[:^ascii]]. Typically, they would match almost any characters. The other
1189        POSIX classes were not broken in this way.
1190    
1191     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1192        to find the match, as PCRE was deluded into thinking that the match had to
1193        start at the start point or following a newline. The same bug applied to
1194        patterns with negative forward assertions or any backward assertions
1195        preceding ".*" at the start, unless the pattern required a fixed first
1196        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1197    
1198    10. In UTF-8 mode, when moving forwards in the subject after a failed match
1199        starting at the last subject character, bytes beyond the end of the subject
1200        string were read.
1201    
1202    11. Renamed the variable "class" as "classbits" to make life easier for C++
1203        users. (Previously there was a macro definition, but it apparently wasn't
1204        enough.)
1205    
1206    12. Added the new field "tables" to the extra data so that tables can be passed
1207        in at exec time, or the internal tables can be re-selected. This allows
1208        a compiled regex to be saved and re-used at a later time by a different
1209        program that might have everything at different addresses.
1210    
1211    13. Modified the pcre-config script so that, when run on Solaris, it shows a
1212        -R library as well as a -L library.
1213    
1214    14. The debugging options of pcretest (-d on the command line or D on a
1215        pattern) showed incorrect output for anything following an extended class
1216        that contained multibyte characters and which was followed by a quantifier.
1217    
1218    15. Added optional support for general category Unicode character properties
1219        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1220        support. It adds about 90K to the size of the library. The meanings of the
1221        inbuilt class escapes such as \d and \s have NOT been changed.
1222    
1223    16. Updated pcredemo.c to include calls to free() to release the memory for the
1224        compiled pattern.
1225    
1226    17. The generated file chartables.c was being created in the source directory
1227        instead of in the building directory. This caused the build to fail if the
1228        source directory was different from the building directory, and was
1229        read-only.
1230    
1231    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1232        file. No doubt somebody will tell me if they don't make sense... Also added
1233        Dan Mooney's comments about building on OpenVMS.
1234    
1235    19. Added support for partial matching via the PCRE_PARTIAL option for
1236        pcre_exec() and the \P data escape in pcretest.
1237    
1238    20. Extended pcretest with 3 new pattern features:
1239    
1240        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
1241              write the compiled pattern to the file whose name is "rest-of-line".
1242              This is a straight binary dump of the data, with the saved pointer to
1243              the character tables forced to be NULL. The study data, if any, is
1244              written too. After writing, pcretest reads a new pattern.
1245    
1246        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1247              compiled pattern from the given file. There must not be any
1248              occurrences of "<" in the file name (pretty unlikely); if there are,
1249              pcretest will instead treat the initial "<" as a pattern delimiter.
1250              After reading in the pattern, pcretest goes on to read data lines as
1251              usual.
1252    
1253        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1254              and 16-bit fields in a compiled pattern, to simulate a pattern that
1255              was compiled on a host of opposite endianness.
1256    
1257    21. The pcre-exec() function can now cope with patterns that were compiled on
1258        hosts of opposite endianness, with this restriction:
1259    
1260          As for any compiled expression that is saved and used later, the tables
1261          pointer field cannot be preserved; the extra_data field in the arguments
1262          to pcre_exec() should be used to pass in a tables address if a value
1263          other than the default internal tables were used at compile time.
1264    
1265    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1266        now diagnosed as an error. Previously, most of the time, a negative number
1267        would have been treated as zero, but if in addition "ovector" was passed as
1268        NULL, a crash could occur.
1269    
1270    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1271        new versions from the libtool 1.5 distribution (the last one is a copy of
1272        a file called libtool.m4). This seems to have fixed the need to patch
1273        "configure" to support Darwin 1.3 (which I used to do). However, I still
1274        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1275        workstation).
1276    
1277    24. Changed the PCRE licence to be the more standard "BSD" licence.
1278    
1279    
1280    Version 4.5 01-Dec-03
1281    ---------------------
1282    
1283     1. There has been some re-arrangement of the code for the match() function so
1284        that it can be compiled in a version that does not call itself recursively.
1285        Instead, it keeps those local variables that need separate instances for
1286        each "recursion" in a frame on the heap, and gets/frees frames whenever it
1287        needs to "recurse". Keeping track of where control must go is done by means
1288        of setjmp/longjmp. The whole thing is implemented by a set of macros that
1289        hide most of the details from the main code, and operates only if
1290        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1291        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1292        operating.
1293    
1294        To make it easier for callers to provide specially tailored get/free
1295        functions for this usage, two new functions, pcre_stack_malloc, and
1296        pcre_stack_free, are used. They are always called in strict stacking order,
1297        and the size of block requested is always the same.
1298    
1299        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1300        PCRE has been compiled to use the stack or the heap for recursion. The
1301        -C option of pcretest uses this to show which version is compiled.
1302    
1303        A new data escape \S, is added to pcretest; it causes the amounts of store
1304        obtained and freed by both kinds of malloc/free at match time to be added
1305        to the output.
1306    
1307     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1308        what's available on my current Linux desktop machine.
1309    
1310     3. When matching a UTF-8 string, the test for a valid string at the start has
1311        been extended. If start_offset is not zero, PCRE now checks that it points
1312        to a byte that is the start of a UTF-8 character. If not, it returns
1313        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1314        this is necessary because there may be backward assertions in the pattern.
1315        When matching the same subject several times, it may save resources to use
1316        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1317    
1318     4. The code for checking the validity of UTF-8 strings has been tightened so
1319        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1320        containing "overlong sequences".
1321    
1322     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1323        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1324        should have been "&", but it just so happened that all the cases this let
1325        through by mistake were picked up later in the function.
1326    
1327     6. I had used a variable called "isblank" - this is a C99 function, causing
1328        some compilers to warn. To avoid this, I renamed it (as "blankclass").
1329    
1330     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1331        prompting; (b) run "./pcretest /dev/null" at the start of the test script
1332        so the version is shown; (c) stop "make test" echoing "./RunTest".
1333    
1334     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1335    
1336     9. The prototype for memmove() for systems that don't have it was using
1337        size_t, but the inclusion of the header that defines size_t was later. I've
1338        moved the #includes for the C headers earlier to avoid this.
1339    
1340    10. Added some adjustments to the code to make it easier to compiler on certain
1341        special systems:
1342    
1343          (a) Some "const" qualifiers were missing.
1344          (b) Added the macro EXPORT before all exported functions; by default this
1345              is defined to be empty.
1346          (c) Changed the dftables auxiliary program (that builds chartables.c) so
1347              that it reads its output file name as an argument instead of writing
1348              to the standard output and assuming this can be redirected.
1349    
1350    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
1351        class containing characters with values greater than 255, PCRE compilation
1352        went into a loop.
1353    
1354    12. A recursive reference to a subpattern that was within another subpattern
1355        that had a minimum quantifier of zero caused PCRE to crash. For example,
1356        (x(y(?2))z)? provoked this bug with a subject that got as far as the
1357        recursion. If the recursively-called subpattern itself had a zero repeat,
1358        that was OK.
1359    
1360    13. In pcretest, the buffer for reading a data line was set at 30K, but the
1361        buffer into which it was copied (for escape processing) was still set at
1362        1024, so long lines caused crashes.
1363    
1364    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
1365        "internal error: code overflow...". This applied to any character class
1366        that was followed by a possessive quantifier.
1367    
1368    15. Modified the Makefile to add libpcre.la as a prerequisite for
1369        libpcreposix.la because I was told this is needed for a parallel build to
1370        work.
1371    
1372    16. If a pattern that contained .* following optional items at the start was
1373        studied, the wrong optimizing data was generated, leading to matching
1374        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
1375        matching string must start with a or b or c. The correct conclusion for
1376        this pattern is that a match can start with any character.
1377    
1378    
1379    Version 4.4 13-Aug-03
1380    ---------------------
1381    
1382     1. In UTF-8 mode, a character class containing characters with values between
1383        127 and 255 was not handled correctly if the compiled pattern was studied.
1384        In fixing this, I have also improved the studying algorithm for such
1385        classes (slightly).
1386    
1387     2. Three internal functions had redundant arguments passed to them. Removal
1388        might give a very teeny performance improvement.
1389    
1390     3. Documentation bug: the value of the capture_top field in a callout is *one
1391        more than* the number of the hightest numbered captured substring.
1392    
1393     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
1394        in incorrectly linking with a previously installed version. They now link
1395        explicitly with libpcre.la.
1396    
1397     5. configure.in no longer needs to recognize Cygwin specially.
1398    
1399     6. A problem in pcre.in for Windows platforms is fixed.
1400    
1401     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
1402        pcretest, it used to include the size of the study block as part of its
1403        output. Unfortunately, the structure contains a field that has a different
1404        size on different hardware architectures. This meant that the tests that
1405        showed this size failed. As the block is currently always of a fixed size,
1406        this information isn't actually particularly useful in pcretest output, so
1407        I have just removed it.
1408    
1409     8. Three pre-processor statements accidentally did not start in column 1.
1410        Sadly, there are *still* compilers around that complain, even though
1411        standard C has not required this for well over a decade. Sigh.
1412    
1413     9. In pcretest, the code for checking callouts passed small integers in the
1414        callout_data field, which is a void * field. However, some picky compilers
1415        complained about the casts involved for this on 64-bit systems. Now
1416        pcretest passes the address of the small integer instead, which should get
1417        rid of the warnings.
1418    
1419    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
1420        both compile and run time, and gives an error if an invalid UTF-8 sequence
1421        is found. There is a option for disabling this check in cases where the
1422        string is known to be correct and/or the maximum performance is wanted.
1423    
1424    11. In response to a bug report, I changed one line in Makefile.in from
1425    
1426            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
1427        to
1428            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
1429    
1430        to look similar to other lines, but I have no way of telling whether this
1431        is the right thing to do, as I do not use Windows. No doubt I'll get told
1432        if it's wrong...
1433    
1434    
1435    Version 4.3 21-May-03
1436    ---------------------
1437    
1438    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
1439       Makefile.
1440    
1441    2. Some refactoring to improve the quality of the code:
1442    
1443       (i)   The utf8_table... variables are now declared "const".
1444    
1445       (ii)  The code for \cx, which used the "case flipping" table to upper case
1446             lower case letters, now just substracts 32. This is ASCII-specific,
1447             but the whole concept of \cx is ASCII-specific, so it seems
1448             reasonable.
1449    
1450       (iii) PCRE was using its character types table to recognize decimal and
1451             hexadecimal digits in the pattern. This is silly, because it handles
1452             only 0-9, a-f, and A-F, but the character types table is locale-
1453             specific, which means strange things might happen. A private
1454             table is now used for this - though it costs 256 bytes, a table is
1455             much faster than multiple explicit tests. Of course, the standard
1456             character types table is still used for matching digits in subject
1457             strings against \d.
1458    
1459       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
1460             ending in _t are). So I've renamed it as ESC_tee.
1461    
1462    3. The first argument for regexec() in the POSIX wrapper should have been
1463       defined as "const".
1464    
1465    4. Changed pcretest to use malloc() for its buffers so that they can be
1466       Electric Fenced for debugging.
1467    
1468    5. There were several places in the code where, in UTF-8 mode, PCRE would try
1469       to read one or more bytes before the start of the subject string. Often this
1470       had no effect on PCRE's behaviour, but in some circumstances it could
1471       provoke a segmentation fault.
1472    
1473    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
1474       to try to read one or more bytes before the start of the subject string.
1475    
1476    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
1477       UTF-8 support could misbehave in various ways if the subject string
1478       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
1479       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
1480       back over UTF-8 characters.)
1481    
1482    
1483    Version 4.2 14-Apr-03
1484    ---------------------
1485    
1486    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
1487    
1488    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
1489         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
1490         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
1491         [WIN_PREFIX]: new variable, "cyg" for Cygwin
1492         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
1493           and BUILD_EXEEXT
1494         Note: automatic setting of the BUILD variables is not yet working
1495         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
1496           compile-time but not at link-time
1497         [LINK]: use for linking executables only
1498         make different versions for Windows and non-Windows
1499         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
1500           libraries
1501         [LINK_FOR_BUILD]: new variable
1502         [OBJEXT]: use throughout
1503         [EXEEXT]: use throughout
1504         <winshared>: new target
1505         <wininstall>: new target
1506         <dftables.o>: use native compiler
1507         <dftables>: use native linker
1508         <install>: handle Windows platform correctly
1509         <clean>: ditto
1510         <check>: ditto
1511         copy DLL to top builddir before testing
1512    
1513       As part of these changes, -no-undefined was removed again. This was reported
1514       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
1515       in any case.
1516    
1517    3. Some tidies to get rid of compiler warnings:
1518    
1519       . In the match_data structure, match_limit was an unsigned long int, whereas
1520         match_call_count was an int. I've made them both unsigned long ints.
1521    
1522       . In pcretest the fact that a const uschar * doesn't automatically cast to
1523         a void * provoked a warning.
1524    
1525       . Turning on some more compiler warnings threw up some "shadow" variables
1526         and a few more missing casts.
1527    
1528    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
1529       option, a class that contained a single character with a value between 128
1530       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
1531    
1532    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
1533       option, a class that contained several characters, but with at least one
1534       whose value was between 128 and 255 caused PCRE to crash.
1535    
1536    
1537    Version 4.1 12-Mar-03
1538    ---------------------
1539    
1540    1. Compiling with gcc -pedantic found a couple of places where casts were
1541    needed, and a string in dftables.c that was longer than standard compilers are
1542    required to support.
1543    
1544    2. Compiling with Sun's compiler found a few more places where the code could
1545    be tidied up in order to avoid warnings.
1546    
1547    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
1548    first of these names is deprecated in the latest Autoconf in favour of the name
1549    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
1550    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
1551    analogy I have changed it to CFLAGS_FOR_BUILD.
1552    
1553    4. Added -no-undefined to the linking command in the Makefile, because this is
1554    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
1555    linking step for the pcreposix library.
1556    
1557    5. PCRE was failing to diagnose the case of two named groups with the same
1558    name.
1559    
1560    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
1561    literal character that is needed in the subject for a match, and scans along to
1562    ensure that it is present before embarking on the full matching process. This
1563    saves time in cases of nested unlimited repeats that are never going to match.
1564    Problem: the scan can take a lot of time if the subject is very long (e.g.
1565    megabytes), thus penalizing straightforward matches. It is now done only if the
1566    amount of subject to be scanned is less than 1000 bytes.
1567    
1568    7. A lesser problem with the same optimization is that it was recording the
1569    first character of an anchored pattern as "needed", thus provoking a search
1570    right along the subject, even when the first match of the pattern was going to
1571    fail. The "needed" character is now not set for anchored patterns, unless it
1572    follows something in the pattern that is of non-fixed length. Thus, it still
1573    fulfils its original purpose of finding quick non-matches in cases of nested
1574    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
1575    
1576    
1577    Version 4.0 17-Feb-03
1578    ---------------------
1579    
1580    1. If a comment in an extended regex that started immediately after a meta-item
1581    extended to the end of string, PCRE compiled incorrect data. This could lead to
1582    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
1583    
1584    2. Moved to autoconf 2.53 and libtool 1.4.2.
1585    
1586    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
1587    the special perltest8 script is no longer needed - all the tests can be run
1588    from a single perltest script.
1589    
1590    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
1591    by \s. It has now been removed in PCRE. This means it isn't recognized as
1592    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
1593    class [:space:] *does* include VT, thereby creating a mess.
1594    
1595    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
1596    space and tab.
1597    
1598    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
1599    its new features into the main test script, reducing the number of scripts.
1600    
1601    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
1602    were backward compatible, and made the (?i) apply to the whole pattern, as if
1603    /i were given. Now it behaves more logically, and applies the option setting
1604    only to what follows. PCRE has been changed to follow suit. However, if it
1605    finds options settings right at the start of the pattern, it extracts them into
1606    the global options, as before. Thus, they show up in the info data.
1607    
1608    8. Added support for the \Q...\E escape sequence. Characters in between are
1609    treated as literals. This is slightly different from Perl in that $ and @ are
1610    also handled as literals inside the quotes. In Perl, they will cause variable
1611    interpolation. Note the following examples:
1612    
1613        Pattern            PCRE matches      Perl matches
1614    
1615        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
1616        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
1617        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
1618    
1619    For compatibility with Perl, \Q...\E sequences are recognized inside character
1620    classes as well as outside them.
1621    
1622    9. Re-organized 3 code statements in pcretest to avoid "overflow in
1623    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
1624    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
1625    signed/unsigned warnings.
1626    
1627    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
1628    option for pcretest, so I've replaced it by a simple function that does just
1629    that job.
1630    
1631    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
1632    "pcregrep -".
1633    
1634    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
1635    Java package. This provides some syntactic sugar for simple cases of what my
1636    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
1637    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
1638    item, you can use this simplified notation. Note that only makes sense with
1639    greedy quantifiers. Consequently, the use of the possessive quantifier forces
1640    greediness, whatever the setting of the PCRE_UNGREEDY option.
1641    
1642    13. A change of greediness default within a pattern was not taking effect at
1643    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
1644    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
1645    was abstracted outside.
1646    
1647    14. PCRE now supports the \G assertion. It is true when the current matching
1648    position is at the start point of the match. This differs from \A when the
1649    starting offset is non-zero. Used with the /g option of pcretest (or similar
1650    code), it works in the same way as it does for Perl's /g option. If all
1651    alternatives of a regex begin with \G, the expression is anchored to the start
1652    match position, and the "anchored" flag is set in the compiled expression.
1653    
1654    15. Some bugs concerning the handling of certain option changes within patterns
1655    have been fixed. These applied to options other than (?ims). For example,
1656    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
1657    been the other way round. Some of this was related to change 7 above.
1658    
1659    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
1660    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
1661    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
1662    POSIX classes only within a class (e.g. /[[:alpha:]]/).
1663    
1664    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
1665    mode. Unlike ".", it always matches newline, whatever the setting of
1666    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
1667    assertions. Perl allows it, but it doesn't (in general) work because it can't
1668    calculate the length of the lookbehind. At least, that's the case for Perl
1669    5.8.0 - I've been told they are going to document that it doesn't work in
1670    future.
1671    
1672    18. Added an error diagnosis for escapes that PCRE does not support: these are
1673    \L, \l, \N, \P, \p, \U, \u, and \X.
1674    
1675    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
1676    reading past the end of the pattern in cases such as /[abcd/.
1677    
1678    20. PCRE was getting more memory than necessary for patterns with classes that
1679    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
1680    
1681    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
1682    compiling PCRE for use with Virtual Pascal.
1683    
1684    22. Small fix to the Makefile to make it work properly if the build is done
1685    outside the source tree.
1686    
1687    23. Added a new extension: a condition to go with recursion. If a conditional
1688    subpattern starts with (?(R) the "true" branch is used if recursion has
1689    happened, whereas the "false" branch is used only at the top level.
1690    
1691    24. When there was a very long string of literal characters (over 255 bytes
1692    without UTF support, over 250 bytes with UTF support), the computation of how
1693    much memory was required could be incorrect, leading to segfaults or other
1694    strange effects.
1695    
1696    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
1697    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
1698    there was a subsequent back reference to those brackets. This meant that, for
1699    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
1700    possible to check for precisely this case. All we can do is abandon the
1701    optimization if .* occurs inside capturing brackets when there are any back
1702    references whatsoever. (See below for a better fix that came later.)
1703    
1704    26. The handling of the optimization for finding the first character of a
1705    non-anchored pattern, and for finding a character that is required later in the
1706    match were failing in some cases. This didn't break the matching; it just
1707    failed to optimize when it could. The way this is done has been re-implemented.
1708    
1709    27. Fixed typo in error message for invalid (?R item (it said "(?p").
1710    
1711    28. Added a new feature that provides some of the functionality that Perl
1712    provides with (?{...}). The facility is termed a "callout". The way it is done
1713    in PCRE is for the caller to provide an optional function, by setting
1714    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
1715    global variable. By default it is unset, which disables all calling out. To get
1716    the function called, the regex must include (?C) at appropriate points. This
1717    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
1718    This provides a means of identifying different callout points. When PCRE
1719    reaches such a point in the regex, if pcre_callout has been set, the external
1720    function is called. It is provided with data in a structure called
1721    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
1722    matching continues; if it returns a non-zero value, the match at the current
1723    point fails. However, backtracking will occur if possible. [This was changed
1724    later and other features added - see item 49 below.]
1725    
1726    29. pcretest is upgraded to test the callout functionality. It provides a
1727    callout function that displays information. By default, it shows the start of
1728    the match and the current position in the text. There are some new data escapes
1729    to vary what happens:
1730    
1731        \C+         in addition, show current contents of captured substrings
1732        \C-         do not supply a callout function
1733        \C!n        return 1 when callout number n is reached
1734        \C!n!m      return 1 when callout number n is reached for the mth time
1735    
1736    30. If pcregrep was called with the -l option and just a single file name, it
1737    output "<stdin>" if a match was found, instead of the file name.
1738    
1739    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
1740    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
1741    pcre_exec(). This saves a malloc/free per call. The default value of
1742    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
1743    when configuring.
1744    
1745    32. The default maximum size of a compiled pattern is 64K. There have been a
1746    few cases of people hitting this limit. The code now uses macros to handle the
1747    storing of links as offsets within the compiled pattern. It defaults to 2-byte
1748    links, but this can be changed to 3 or 4 bytes by --with-link-size when
1749    configuring. Tests 2 and 5 work only with 2-byte links because they output
1750    debugging information about compiled patterns.
1751    
1752    33. Internal code re-arrangements:
1753    
1754    (a) Moved the debugging function for printing out a compiled regex into
1755        its own source file (printint.c) and used #include to pull it into
1756        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
1757        separate copies.
1758    
1759    (b) Defined the list of op-code names for debugging as a macro in
1760        internal.h so that it is next to the definition of the opcodes.
1761    
1762    (c) Defined a table of op-code lengths for simpler skipping along compiled
1763        code. This is again a macro in internal.h so that it is next to the
1764        definition of the opcodes.
1765    
1766    34. Added support for recursive calls to individual subpatterns, along the
1767    lines of Robin Houston's patch (but implemented somewhat differently).
1768    
1769    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
1770    allow it to read and process whole directories in Win32. This code was
1771    contributed by Lionel Fourquaux; it has not been tested by me.
1772    
1773    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
1774    used to name a group. Names consist of alphanumerics and underscores, and must
1775    be unique. Back references use the syntax (?P=name) and recursive calls use
1776    (?P>name) which is a PCRE extension to the Python extension. Groups still have
1777    numbers. The function pcre_fullinfo() can be used after compilation to extract
1778    a name/number map. There are three relevant calls:
1779    
1780      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
1781      PCRE_INFO_NAMECOUNT            yields the number of entries
1782      PCRE_INFO_NAMETABLE            yields a pointer to the map.
1783    
1784    The map is a vector of fixed-size entries. The size of each entry depends on
1785    the length of the longest name used. The first two bytes of each entry are the
1786    group number, most significant byte first. There follows the corresponding
1787    name, zero terminated. The names are in alphabetical order.
1788    
1789    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
1790    case instead of 255. Making it the same both with and without UTF-8 support
1791    means that the same test output works with both.
1792    
1793    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
1794    calling malloc() with a zero argument.
1795    
1796    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
1797    optimization. I've improved things by keeping a bitmap of backreferences with
1798    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
1799    fact referenced, the optimization can be applied. It is unlikely that a
1800    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
1801    the match to follow \n) will appear inside brackets with a number greater than
1802    31, but if it does, any back reference > 31 suppresses the optimization.
1803    
1804    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
1805    of disabling numbered capturing parentheses. Any opening parenthesis that is
1806    not followed by ? behaves as if it were followed by ?: but named parentheses
1807    can still be used for capturing (and they will acquire numbers in the usual
1808    way).
1809    
1810    41. Redesigned the return codes from the match() function into yes/no/error so
1811    that errors can be passed back from deep inside the nested calls. A malloc
1812    failure while inside a recursive subpattern call now causes the
1813    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
1814    
1815    42. It is now possible to set a limit on the number of times the match()
1816    function is called in a call to pcre_exec(). This facility makes it possible to
1817    limit the amount of recursion and backtracking, though not in a directly
1818    obvious way, because the match() function is used in a number of different
1819    circumstances. The count starts from zero for each position in the subject
1820    string (for non-anchored patterns). The default limit is, for compatibility, a
1821    large number, namely 10 000 000. You can change this in two ways:
1822    
1823    (a) When configuring PCRE before making, you can use --with-match-limit=n
1824        to set a default value for the compiled library.
1825    
1826    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
1827        a different value is set. See 45 below.
1828    
1829    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
1830    
1831    43. Added a new function pcre_config(int, void *) to enable run-time extraction
1832    of things that can be changed at compile time. The first argument specifies
1833    what is wanted and the second points to where the information is to be placed.
1834    The current list of available information is:
1835    
1836      PCRE_CONFIG_UTF8
1837    
1838    The output is an integer that is set to one if UTF-8 support is available;
1839    otherwise it is set to zero.
1840    
1841      PCRE_CONFIG_NEWLINE
1842    
1843    The output is an integer that it set to the value of the code that is used for
1844    newline. It is either LF (10) or CR (13).
1845    
1846      PCRE_CONFIG_LINK_SIZE
1847    
1848    The output is an integer that contains the number of bytes used for internal
1849    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
1850    
1851      PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
1852    
1853    The output is an integer that contains the threshold above which the POSIX
1854    interface uses malloc() for output vectors. See item 31 above.
1855    
1856      PCRE_CONFIG_MATCH_LIMIT
1857    
1858    The output is an unsigned integer that contains the default limit of the number
1859    of match() calls in a pcre_exec() execution. See 42 above.
1860    
1861    44. pcretest has been upgraded by the addition of the -C option. This causes it
1862    to extract all the available output from the new pcre_config() function, and to
1863    output it. The program then exits immediately.
1864    
1865    45. A need has arisen to pass over additional data with calls to pcre_exec() in
1866    order to support additional features. One way would have been to define
1867    pcre_exec2() (for example) with extra arguments, but this would not have been
1868    extensible, and would also have required all calls to the original function to
1869    be mapped to the new one. Instead, I have chosen to extend the mechanism that
1870    is used for passing in "extra" data from pcre_study().
1871    
1872    The pcre_extra structure is now exposed and defined in pcre.h. It currently
1873    contains the following fields:
1874    
1875      flags         a bitmap indicating which of the following fields are set
1876      study_data    opaque data from pcre_study()
1877      match_limit   a way of specifying a limit on match() calls for a specific
1878                      call to pcre_exec()
1879      callout_data  data for callouts (see 49 below)
1880    
1881    The flag bits are also defined in pcre.h, and are
1882    
1883      PCRE_EXTRA_STUDY_DATA
1884      PCRE_EXTRA_MATCH_LIMIT
1885      PCRE_EXTRA_CALLOUT_DATA
1886    
1887    The pcre_study() function now returns one of these new pcre_extra blocks, with
1888    the actual study data pointed to by the study_data field, and the
1889    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
1890    before. That is, this change is entirely upwards-compatible and requires no
1891    change to existing code.
1892    
1893    If you want to pass in additional data to pcre_exec(), you can either place it
1894    in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
1895    block.
1896    
1897    46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
1898    data string contains the escape sequence \M, pcretest calls pcre_exec() several
1899    times with different match limits, until it finds the minimum value needed for
1900    pcre_exec() to complete. The value is then output. This can be instructive; for
1901    most simple matches the number is quite small, but for pathological cases it
1902    gets very large very quickly.
1903    
1904    47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
1905    returns the size of the data block pointed to by the study_data field in a
1906    pcre_extra block, that is, the value that was passed as the argument to
1907    pcre_malloc() when PCRE was getting memory in which to place the information
1908    created by pcre_study(). The fourth argument should point to a size_t variable.
1909    pcretest has been extended so that this information is shown after a successful
1910    pcre_study() call when information about the compiled regex is being displayed.
1911    
1912    48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
1913    because what follows is always an absolute path. (Later: it turns out that this
1914    is more than cosmetic for MinGW, because it doesn't like empty path
1915    components.)
1916    
1917    49. Some changes have been made to the callout feature (see 28 above):
1918    
1919    (i)  A callout function now has three choices for what it returns:
1920    
1921           0  =>  success, carry on matching
1922         > 0  =>  failure at this point, but backtrack if possible
1923         < 0  =>  serious error, return this value from pcre_exec()
1924    
1925         Negative values should normally be chosen from the set of PCRE_ERROR_xxx
1926         values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
1927         "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
1928         use by callout functions. It will never be used by PCRE itself.
1929    
1930    (ii) The pcre_extra structure (see 45 above) has a void * field called
1931         callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
1932         pcre_callout_block structure has a field of the same name. The contents of
1933         the field passed in the pcre_extra structure are passed to the callout
1934         function in the corresponding field in the callout block. This makes it
1935         easier to use the same callout-containing regex from multiple threads. For
1936         testing, the pcretest program has a new data escape
1937    
1938           \C*n        pass the number n (may be negative) as callout_data
1939    
1940         If the callout function in pcretest receives a non-zero value as
1941         callout_data, it returns that value.
1942    
1943    50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
1944    there were some redundant $(CFLAGS) in commands that are now specified as
1945    $(LINK), which already includes $(CFLAGS).
1946    
1947    51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
1948    has been compiled with UTF-8 support *and* pcre_compile() has been compiled
1949    with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
1950    one-byte characters throughout. Note that case-insensitive matching applies
1951    only to characters whose values are less than 256. PCRE doesn't support the
1952    notion of cases for higher-valued characters.
1953    
1954    (i)   A character class whose characters are all within 0-255 is handled as
1955          a bit map, and the map is inverted for negative classes. Previously, a
1956          character > 255 always failed to match such a class; however it should
1957          match if the class was a negative one (e.g. [^ab]). This has been fixed.
1958    
1959    (ii)  A negated character class with a single character < 255 is coded as
1960          "not this character" (OP_NOT). This wasn't working properly when the test
1961          character was multibyte, either singly or repeated.
1962    
1963    (iii) Repeats of multibyte characters are now handled correctly in UTF-8
1964          mode, for example: \x{100}{2,3}.
1965    
1966    (iv)  The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
1967          singly or repeated) now correctly test multibyte characters. However,
1968          PCRE doesn't recognize any characters with values greater than 255 as
1969          digits, spaces, or word characters. Such characters always match \D, \S,
1970          and \W, and never match \d, \s, or \w.
1971    
1972    (v)   Classes may now contain characters and character ranges with values
1973          greater than 255. For example: [ab\x{100}-\x{400}].
1974    
1975    (vi)  pcregrep now has a --utf-8 option (synonym -u) which makes it call
1976          PCRE in UTF-8 mode.
1977    
1978    52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
1979    PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
1980    retained for backwards compatibility. (Note that LASTLITERAL is also a byte
1981    value.)
1982    
1983    53. The single man page has become too large. I have therefore split it up into
1984    a number of separate man pages. These also give rise to individual HTML pages;
1985    these are now put in a separate directory, and there is an index.html page that
1986    lists them all. Some hyperlinking between the pages has been installed.
1987    
1988    54. Added convenience functions for handling named capturing parentheses.
1989    
1990    55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
1991    aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
1992    true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
1993    are faulted.
1994    
1995    56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
1996    calling configure. These values are used when compiling the dftables.c program
1997    which is run to generate the source of the default character tables. They
1998    default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
1999    you will need to set these values.
2000    
2001    57. Updated the building process for Windows DLL, as provided by Fred Cox.
2002    
2003    
2004    Version 3.9 02-Jan-02
2005  ---------------------  ---------------------
2006    
2007  1. A bit of extraneous text had somehow crept into the pcregrep documentation.  1. A bit of extraneous text had somehow crept into the pcregrep documentation.

Legend:
Removed from v.61  
changed lines
  Added in v.171

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12