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

Diff of /code/trunk/ChangeLog

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

revision 3 by nigel, Sat Feb 24 21:38:01 2007 UTC revision 170 by ph10, Mon Jun 4 11:21:13 2007 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4    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     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
40        PCRE_INFO_JCHANGED.
41    
42     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
43        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
44        for detecting groups that can match an empty string.
45    
46    
47    Version 7.1 24-Apr-07
48    ---------------------
49    
50     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
51        that is more "standard", making use of automake and other Autotools. There
52        is some re-arrangement of the files and adjustment of comments consequent
53        on this.
54    
55     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
56        for recursive directory scanning broke on some systems because the files
57        are not scanned in any specific order and on different systems the order
58        was different. A call to "sort" has been inserted into RunGrepTest for the
59        approprate test as a short-term fix. In the longer term there may be an
60        alternative.
61    
62     3. I had an email from Eric Raymond about problems translating some of PCRE's
63        man pages to HTML (despite the fact that I distribute HTML pages, some
64        people do their own conversions for various reasons). The problems
65        concerned the use of low-level troff macros .br and .in. I have therefore
66        removed all such uses from the man pages (some were redundant, some could
67        be replaced by .nf/.fi pairs). The 132html script that I use to generate
68        HTML has been updated to handle .nf/.fi and to complain if it encounters
69        .br or .in.
70    
71     4. Updated comments in configure.ac that get placed in config.h.in and also
72        arranged for config.h to be included in the distribution, with the name
73        config.h.generic, for the benefit of those who have to compile without
74        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
75    
76     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
77        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
78        makevp.bat for the current PCRE, using the additional files
79        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
80    
81     6. A Windows user reported a minor discrepancy with test 2, which turned out
82        to be caused by a trailing space on an input line that had got lost in his
83        copy. The trailing space was an accident, so I've just removed it.
84    
85     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
86        that is needed.
87    
88     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
89        as "const" (a) because they are and (b) because it helps the PHP
90        maintainers who have recently made a script to detect big data structures
91        in the php code that should be moved to the .rodata section. I remembered
92        to update Builducptable as well, so it won't revert if ucptable.h is ever
93        re-created.
94    
95     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
96        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
97        order to be able to cut out the UTF-8 tables in the latter when UTF-8
98        support is not required. This saves 1.5-2K of code, which is important in
99        some applications.
100    
101        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
102        so as not to refer to the tables, even though these functions will never be
103        called when UTF-8 support is disabled. Otherwise there are problems with a
104        shared library.
105    
106    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
107    
108        (a) It was defining its arguments as char * instead of void *.
109    
110        (b) It was assuming that all moves were upwards in memory; this was true
111            a long time ago when I wrote it, but is no longer the case.
112    
113        The emulated memove() is provided for those environments that have neither
114        memmove() nor bcopy(). I didn't think anyone used it these days, but that
115        is clearly not the case, as these two bugs were recently reported.
116    
117    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
118        and Detrail to create the HTML documentation, the .txt form of the man
119        pages, and it removes trailing spaces from listed files. It also creates
120        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
121        case, it wraps all the #defines with #ifndefs. This script should be run
122        before "make dist".
123    
124    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
125        with Unicode property support.
126    
127        (a) For a maximizing quantifier, if the two different cases of the
128            character were of different lengths in their UTF-8 codings (there are
129            some cases like this - I found 11), and the matching function had to
130            back up over a mixture of the two cases, it incorrectly assumed they
131            were both the same length.
132    
133        (b) When PCRE was configured to use the heap rather than the stack for
134            recursion during matching, it was not correctly preserving the data for
135            the other case of a UTF-8 character when checking ahead for a match
136            while processing a minimizing repeat. If the check also involved
137            matching a wide character, but failed, corruption could cause an
138            erroneous result when trying to check for a repeat of the original
139            character.
140    
141    13. Some tidying changes to the testing mechanism:
142    
143        (a) The RunTest script now detects the internal link size and whether there
144            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
145            values substituted by "configure". (The RunGrepTest script already did
146            this for UTF-8.) The configure.ac script no longer substitutes the
147            relevant variables.
148    
149        (b) The debugging options /B and /D in pcretest show the compiled bytecode
150            with length and offset values. This means that the output is different
151            for different internal link sizes. Test 2 is skipped for link sizes
152            other than 2 because of this, bypassing the problem. Unfortunately,
153            there was also a test in test 3 (the locale tests) that used /B and
154            failed for link sizes other than 2. Rather than cut the whole test out,
155            I have added a new /Z option to pcretest that replaces the length and
156            offset values with spaces. This is now used to make test 3 independent
157            of link size. (Test 2 will be tidied up later.)
158    
159    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
160        segmentation fault instead of returning the appropriate error message.
161    
162    15. In multiline mode when the newline sequence was set to "any", the pattern
163        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
164        This doesn't seem right; it now treats the CRLF combination as the line
165        ending, and so does not match in that case. It's only a pattern such as ^$
166        that would hit this one: something like ^ABC$ would have failed after \r
167        and then tried again after \r\n.
168    
169    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
170        in an attempt to make files that differ only in their line terminators
171        compare equal. This works on Linux.
172    
173    17. Under certain error circumstances pcregrep might try to free random memory
174        as it exited. This is now fixed, thanks to valgrind.
175    
176    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
177        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
178        was because its rules for how to advance for /g after matching an empty
179        string at the end of a line did not allow for this case. They now check for
180        it specially.
181    
182    20. pcretest is supposed to handle patterns and data of any length, by
183        extending its buffers when necessary. It was getting this wrong when the
184        buffer for a data line had to be extended.
185    
186    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
187        CRLF as a newline sequence.
188    
189    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
190        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
191        I have nevertheless tidied it up.
192    
193    23. Added some casts to kill warnings from HP-UX ia64 compiler.
194    
195    24. Added a man page for pcre-config.
196    
197    
198    Version 7.0 19-Dec-06
199    ---------------------
200    
201     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
202        moving to gcc 4.1.1.
203    
204     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
205        sys/time.h, which is documented as needed for this function. It doesn't
206        seem to matter on Linux, but it showed up on some releases of OS X.
207    
208     3. It seems that there are systems where bytes whose values are greater than
209        127 match isprint() in the "C" locale. The "C" locale should be the
210        default when a C program starts up. In most systems, only ASCII printing
211        characters match isprint(). This difference caused the output from pcretest
212        to vary, making some of the tests fail. I have changed pcretest so that:
213    
214        (a) When it is outputting text in the compiled version of a pattern, bytes
215            other than 32-126 are always shown as hex escapes.
216    
217        (b) When it is outputting text that is a matched part of a subject string,
218            it does the same, unless a different locale has been set for the match
219            (using the /L modifier). In this case, it uses isprint() to decide.
220    
221     4. Fixed a major bug that caused incorrect computation of the amount of memory
222        required for a compiled pattern when options that changed within the
223        pattern affected the logic of the preliminary scan that determines the
224        length. The relevant options are -x, and -i in UTF-8 mode. The result was
225        that the computed length was too small. The symptoms of this bug were
226        either the PCRE error "internal error: code overflow" from pcre_compile(),
227        or a glibc crash with a message such as "pcretest: free(): invalid next
228        size (fast)". Examples of patterns that provoked this bug (shown in
229        pcretest format) are:
230    
231          /(?-x: )/x
232          /(?x)(?-x: \s*#\s*)/
233          /((?i)[\x{c0}])/8
234          /(?i:[\x{c0}])/8
235    
236        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
237        is now done differently.
238    
239     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
240        wrapper classes; (b) implement a new function in the C++ scanner that is
241        more efficient than the old way of doing things because it avoids levels of
242        recursion in the regex matching; (c) add a paragraph to the documentation
243        for the FullMatch() function.
244    
245     6. The escape sequence \n was being treated as whatever was defined as
246        "newline". Not only was this contrary to the documentation, which states
247        that \n is character 10 (hex 0A), but it also went horribly wrong when
248        "newline" was defined as CRLF. This has been fixed.
249    
250     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
251        was being set to -1 for the "end of line" case (supposedly a value that no
252        character can have). Though this value is never used (the check for end of
253        line is "zero bytes in current character"), it caused compiler complaints.
254        I've changed it to 0xffffffff.
255    
256     8. In pcre_version.c, the version string was being built by a sequence of
257        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
258        string (as it is for production releases) called a macro with an empty
259        argument. The C standard says the result of this is undefined. The gcc
260        compiler treats it as an empty string (which was what was wanted) but it is
261        reported that Visual C gives an error. The source has been hacked around to
262        avoid this problem.
263    
264     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
265        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
266        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
267        of them did).
268    
269    10. Originally, pcretest opened its input and output without "b"; then I was
270        told that "b" was needed in some environments, so it was added for release
271        5.0 to both the input and output. (It makes no difference on Unix-like
272        systems.) Later I was told that it is wrong for the input on Windows. I've
273        now abstracted the modes into two macros, to make it easier to fiddle with
274        them, and removed "b" from the input mode under Windows.
275    
276    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
277    
278    12. Added -help and --help to pcretest as an official way of being reminded
279        of the options.
280    
281    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
282        and pcrecpp.cc because they annoy compilers at high warning levels.
283    
284    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
285    
286    15. Fixed an occurrence of == in configure.ac that should have been = (shell
287        scripts are not C programs :-) and which was not noticed because it works
288        on Linux.
289    
290    16. pcretest is supposed to handle any length of pattern and data line (as one
291        line or as a continued sequence of lines) by extending its input buffer if
292        necessary. This feature was broken for very long pattern lines, leading to
293        a string of junk being passed to pcre_compile() if the pattern was longer
294        than about 50K.
295    
296    17. I have done a major re-factoring of the way pcre_compile() computes the
297        amount of memory needed for a compiled pattern. Previously, there was code
298        that made a preliminary scan of the pattern in order to do this. That was
299        OK when PCRE was new, but as the facilities have expanded, it has become
300        harder and harder to keep it in step with the real compile phase, and there
301        have been a number of bugs (see for example, 4 above). I have now found a
302        cunning way of running the real compile function in a "fake" mode that
303        enables it to compute how much memory it would need, while actually only
304        ever using a few hundred bytes of working memory and without too many
305        tests of the mode. This should make future maintenance and development
306        easier. A side effect of this work is that the limit of 200 on the nesting
307        depth of parentheses has been removed (though this was never a serious
308        limitation, I suspect). However, there is a downside: pcre_compile() now
309        runs more slowly than before (30% or more, depending on the pattern). I
310        hope this isn't a big issue. There is no effect on runtime performance.
311    
312    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
313        newline (only possible for the last line of a file) and it was a
314        pattern that set a locale (followed by /Lsomething), pcretest crashed.
315    
316    19. Added additional timing features to pcretest. (1) The -tm option now times
317        matching only, not compiling. (2) Both -t and -tm can be followed, as a
318        separate command line item, by a number that specifies the number of
319        repeats to use when timing. The default is 50000; this gives better
320        precision, but takes uncomfortably long for very large patterns.
321    
322    20. Extended pcre_study() to be more clever in cases where a branch of a
323        subpattern has no definite first character. For example, (a*|b*)[cd] would
324        previously give no result from pcre_study(). Now it recognizes that the
325        first character must be a, b, c, or d.
326    
327    21. There was an incorrect error "recursive call could loop indefinitely" if
328        a subpattern (or the entire pattern) that was being tested for matching an
329        empty string contained only one non-empty item after a nested subpattern.
330        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
331        incorrectly, because the \d was being skipped in the check.
332    
333    22. The pcretest program now has a new pattern option /B and a command line
334        option -b, which is equivalent to adding /B to every pattern. This causes
335        it to show the compiled bytecode, without the additional information that
336        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
337        is the same as /B/I).
338    
339    23. A new optimization is now able automatically to treat some sequences such
340        as a*b as a*+b. More specifically, if something simple (such as a character
341        or a simple class like \d) has an unlimited quantifier, and is followed by
342        something that cannot possibly match the quantified thing, the quantifier
343        is automatically "possessified".
344    
345    24. A recursive reference to a subpattern whose number was greater than 39
346        went wrong under certain circumstances in UTF-8 mode. This bug could also
347        have affected the operation of pcre_study().
348    
349    25. Realized that a little bit of performance could be had by replacing
350        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
351    
352    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
353    
354    27. Possessive quantifiers such as a++ were previously implemented by turning
355        them into atomic groups such as ($>a+). Now they have their own opcodes,
356        which improves performance. This includes the automatically created ones
357        from 23 above.
358    
359    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
360        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
361        the first matched character to be a colon. This applied both to named and
362        numbered groups.
363    
364    29. The ucpinternal.h header file was missing its idempotency #ifdef.
365    
366    30. I was sent a "project" file called libpcre.a.dev which I understand makes
367        building PCRE on Windows easier, so I have included it in the distribution.
368    
369    31. There is now a check in pcretest against a ridiculously large number being
370        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
371        loop, the loop is abandoned.
372    
373    32. Forward references to subpatterns in conditions such as (?(2)...) where
374        subpattern 2 is defined later cause pcre_compile() to search forwards in
375        the pattern for the relevant set of parentheses. This search went wrong
376        when there were unescaped parentheses in a character class, parentheses
377        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
378    
379    33. "Subroutine" calls and backreferences were previously restricted to
380        referencing subpatterns earlier in the regex. This restriction has now
381        been removed.
382    
383    34. Added a number of extra features that are going to be in Perl 5.10. On the
384        whole, these are just syntactic alternatives for features that PCRE had
385        previously implemented using the Python syntax or my own invention. The
386        other formats are all retained for compatibility.
387    
388        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
389            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
390            also .NET compatible.
391    
392        (b) A recursion or subroutine call to a named group can now be defined as
393            (?&name) as well as (?P>name).
394    
395        (c) A backreference to a named group can now be defined as \k<name> or
396            \k'name' as well as (?P=name). The new forms, as well as being in Perl
397            5.10, are also .NET compatible.
398    
399        (d) A conditional reference to a named group can now use the syntax
400            (?(<name>) or (?('name') as well as (?(name).
401    
402        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
403            groups (named and numbered) that are never evaluated inline, but can be
404            called as "subroutines" from elsewhere. In effect, the DEFINE condition
405            is always false. There may be only one alternative in such a group.
406    
407        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
408            as the simple (?(R). The condition is true only if the most recent
409            recursion is that of the given number or name. It does not search out
410            through the entire recursion stack.
411    
412        (g) The escape \gN or \g{N} has been added, where N is a positive or
413            negative number, specifying an absolute or relative reference.
414    
415    35. Tidied to get rid of some further signed/unsigned compiler warnings and
416        some "unreachable code" warnings.
417    
418    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
419        things, this adds five new scripts.
420    
421    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
422        There were also incompatibilities regarding the handling of \Q..\E inside
423        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
424        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
425    
426    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
427        matches an empty string, and forcibly breaks the loop. There were bugs in
428        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
429        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
430        separate and independent bugs (that affected different cases) have been
431        fixed.
432    
433    39. Refactored the code to abolish the use of different opcodes for small
434        capturing bracket numbers. This is a tidy that I avoided doing when I
435        removed the limit on the number of capturing brackets for 3.5 back in 2001.
436        The new approach is not only tidier, it makes it possible to reduce the
437        memory needed to fix the previous bug (38).
438    
439    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
440        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
441        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
442        mode.
443    
444    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
445        report.
446    
447    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
448        copy construction and assignment in the C++ wrapper.
449    
450    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
451        couple of bugs that could have given wrong results in the "--newline=crlf"
452        case.
453    
454    44. Added a number of casts and did some reorganization of signed/unsigned int
455        variables following suggestions from Dair Grant. Also renamed the variable
456        "this" as "item" because it is a C++ keyword.
457    
458    45. Arranged for dftables to add
459    
460          #include "pcre_internal.h"
461    
462        to pcre_chartables.c because without it, gcc 4.x may remove the array
463        definition from the final binary if PCRE is built into a static library and
464        dead code stripping is activated.
465    
466    46. For an unanchored pattern, if a match attempt fails at the start of a
467        newline sequence, and the newline setting is CRLF or ANY, and the next two
468        characters are CRLF, advance by two characters instead of one.
469    
470    
471    Version 6.7 04-Jul-06
472    ---------------------
473    
474     1. In order to handle tests when input lines are enormously long, pcretest has
475        been re-factored so that it automatically extends its buffers when
476        necessary. The code is crude, but this _is_ just a test program. The
477        default size has been increased from 32K to 50K.
478    
479     2. The code in pcre_study() was using the value of the re argument before
480        testing it for NULL. (Of course, in any sensible call of the function, it
481        won't be NULL.)
482    
483     3. The memmove() emulation function in pcre_internal.h, which is used on
484        systems that lack both memmove() and bcopy() - that is, hardly ever -
485        was missing a "static" storage class specifier.
486    
487     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
488        containing an extended class (one that cannot be represented by a bitmap
489        because it contains high-valued characters or Unicode property items, e.g.
490        [\pZ]). Almost always one would set UTF-8 mode when processing such a
491        pattern, but PCRE should not loop if you do not (it no longer does).
492        [Detail: two cases were found: (a) a repeated subpattern containing an
493        extended class; (b) a recursive reference to a subpattern that followed a
494        previous extended class. It wasn't skipping over the extended class
495        correctly when UTF-8 mode was not set.]
496    
497     5. A negated single-character class was not being recognized as fixed-length
498        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
499        compile error "lookbehind assertion is not fixed length".
500    
501     6. The RunPerlTest auxiliary script was showing an unexpected difference
502        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
503        write a Perl script that can interpret lines of an input file either as
504        byte characters or as UTF-8, which is what "perltest" was being required to
505        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
506        can't do is switch easily at run time between having the "use utf8;" pragma
507        or not. In the end, I fudged it by using the RunPerlTest script to insert
508        "use utf8;" explicitly for the UTF-8 tests.
509    
510     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
511        the end of the subject string, contrary to the documentation and to what
512        Perl does. This was true of both matching functions. Now it matches only at
513        the start of the subject and immediately after *internal* newlines.
514    
515     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
516        a pointer to an int instead of a pointer to an unsigned long int. This
517        caused problems on 64-bit systems.
518    
519     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
520        instance of the 'standard' template library not being so standard".
521    
522    10. There was no check on the number of named subpatterns nor the maximum
523        length of a subpattern name. The product of these values is used to compute
524        the size of the memory block for a compiled pattern. By supplying a very
525        long subpattern name and a large number of named subpatterns, the size
526        computation could be caused to overflow. This is now prevented by limiting
527        the length of names to 32 characters, and the number of named subpatterns
528        to 10,000.
529    
530    11. Subpatterns that are repeated with specific counts have to be replicated in
531        the compiled pattern. The size of memory for this was computed from the
532        length of the subpattern and the repeat count. The latter is limited to
533        65535, but there was no limit on the former, meaning that integer overflow
534        could in principle occur. The compiled length of a repeated subpattern is
535        now limited to 30,000 bytes in order to prevent this.
536    
537    12. Added the optional facility to have named substrings with the same name.
538    
539    13. Added the ability to use a named substring as a condition, using the
540        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
541        are numbers (not recommended). Forward references are permitted.
542    
543    14. Added forward references in named backreferences (if you see what I mean).
544    
545    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
546        pattern could run off the end of the subject. For example, the pattern
547        "(?s)(.{1,5})"8 did this with the subject "ab".
548    
549    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
550        PCRE_CASELESS was set when matching characters that were quantified with ?
551        or *.
552    
553    17. A character class other than a single negated character that had a minimum
554        but no maximum quantifier - for example [ab]{6,} - was not handled
555        correctly by pce_dfa_exec(). It would match only one character.
556    
557    18. A valid (though odd) pattern that looked like a POSIX character
558        class but used an invalid character after [ (for example [[,abc,]]) caused
559        pcre_compile() to give the error "Failed: internal error: code overflow" or
560        in some cases to crash with a glibc free() error. This could even happen if
561        the pattern terminated after [[ but there just happened to be a sequence of
562        letters, a binary zero, and a closing ] in the memory that followed.
563    
564    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
565        over the years. Originally (before any Unicode support), just the bottom 8
566        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
567        output from "man perlunicode" includes this:
568    
569          The regular expression compiler produces polymorphic opcodes.  That
570          is, the pattern adapts to the data and automatically switches to
571          the Unicode character scheme when presented with Unicode data--or
572          instead uses a traditional byte scheme when presented with byte
573          data.
574    
575        Sadly, a wide octal escape does not cause a switch, and in a string with
576        no other multibyte characters, these octal escapes are treated as before.
577        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
578        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
579        Unicode string.
580    
581        I have not perpetrated such confusion in PCRE. Up till now, it took just
582        the bottom 8 bits, as in old Perl. I have now made octal escapes with
583        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
584        translate to the appropriate multibyte character.
585    
586    29. Applied some refactoring to reduce the number of warnings from Microsoft
587        and Borland compilers. This has included removing the fudge introduced
588        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
589        a warning about an unused variable.
590    
591    21. PCRE has not included VT (character 0x0b) in the set of whitespace
592        characters since release 4.0, because Perl (from release 5.004) does not.
593        [Or at least, is documented not to: some releases seem to be in conflict
594        with the documentation.] However, when a pattern was studied with
595        pcre_study() and all its branches started with \s, PCRE still included VT
596        as a possible starting character. Of course, this did no harm; it just
597        caused an unnecessary match attempt.
598    
599    22. Removed a now-redundant internal flag bit that recorded the fact that case
600        dependency changed within the pattern. This was once needed for "required
601        byte" processing, but is no longer used. This recovers a now-scarce options
602        bit. Also moved the least significant internal flag bit to the most-
603        significant bit of the word, which was not previously used (hangover from
604        the days when it was an int rather than a uint) to free up another bit for
605        the future.
606    
607    23. Added support for CRLF line endings as well as CR and LF. As well as the
608        default being selectable at build time, it can now be changed at runtime
609        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
610        specify that it is scanning data with non-default line endings.
611    
612    24. Changed the definition of CXXLINK to make it agree with the definition of
613        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
614    
615    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
616        recursions. This makes a big different to stack usage for some patterns.
617    
618    26. If a subpattern containing a named recursion or subroutine reference such
619        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
620        the space required for the compiled pattern went wrong and gave too small a
621        value. Depending on the environment, this could lead to "Failed: internal
622        error: code overflow at offset 49" or "glibc detected double free or
623        corruption" errors.
624    
625    27. Applied patches from Google (a) to support the new newline modes and (b) to
626        advance over multibyte UTF-8 characters in GlobalReplace.
627    
628    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
629        difference for some implementation of PCRE in some Windows version.
630    
631    29. Added some extra testing facilities to pcretest:
632    
633        \q<number>   in a data line sets the "match limit" value
634        \Q<number>   in a data line sets the "match recursion limt" value
635        -S <number>  sets the stack size, where <number> is in megabytes
636    
637        The -S option isn't available for Windows.
638    
639    
640    Version 6.6 06-Feb-06
641    ---------------------
642    
643     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
644        in pcreposix.h. I have copied the definition from pcre.h.
645    
646     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
647        because pcre.h is no longer a built file.
648    
649     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
650        not normally included in the compiled code.
651    
652    
653    Version 6.5 01-Feb-06
654    ---------------------
655    
656     1. When using the partial match feature with pcre_dfa_exec(), it was not
657        anchoring the second and subsequent partial matches at the new starting
658        point. This could lead to incorrect results. For example, with the pattern
659        /1234/, partially matching against "123" and then "a4" gave a match.
660    
661     2. Changes to pcregrep:
662    
663        (a) All non-match returns from pcre_exec() were being treated as failures
664            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
665            error message is output. Some extra information is given for the
666            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
667            probably the only errors that are likely to be caused by users (by
668            specifying a regex that has nested indefinite repeats, for instance).
669            If there are more than 20 of these errors, pcregrep is abandoned.
670    
671        (b) A binary zero was treated as data while matching, but terminated the
672            output line if it was written out. This has been fixed: binary zeroes
673            are now no different to any other data bytes.
674    
675        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
676            used to set a locale for matching. The --locale=xxxx long option has
677            been added (no short equivalent) to specify a locale explicitly on the
678            pcregrep command, overriding the environment variables.
679    
680        (d) When -B was used with -n, some line numbers in the output were one less
681            than they should have been.
682    
683        (e) Added the -o (--only-matching) option.
684    
685        (f) If -A or -C was used with -c (count only), some lines of context were
686            accidentally printed for the final match.
687    
688        (g) Added the -H (--with-filename) option.
689    
690        (h) The combination of options -rh failed to suppress file names for files
691            that were found from directory arguments.
692    
693        (i) Added the -D (--devices) and -d (--directories) options.
694    
695        (j) Added the -F (--fixed-strings) option.
696    
697        (k) Allow "-" to be used as a file name for -f as well as for a data file.
698    
699        (l) Added the --colo(u)r option.
700    
701        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
702            is not present by default.
703    
704     3. A nasty bug was discovered in the handling of recursive patterns, that is,
705        items such as (?R) or (?1), when the recursion could match a number of
706        alternatives. If it matched one of the alternatives, but subsequently,
707        outside the recursion, there was a failure, the code tried to back up into
708        the recursion. However, because of the way PCRE is implemented, this is not
709        possible, and the result was an incorrect result from the match.
710    
711        In order to prevent this happening, the specification of recursion has
712        been changed so that all such subpatterns are automatically treated as
713        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
714    
715     4. I had overlooked the fact that, in some locales, there are characters for
716        which isalpha() is true but neither isupper() nor islower() are true. In
717        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
718        and ordfeminine) are like this. This affected the treatment of \w and \W
719        when they appeared in character classes, but not when they appeared outside
720        a character class. The bit map for "word" characters is now created
721        separately from the results of isalnum() instead of just taking it from the
722        upper, lower, and digit maps. (Plus the underscore character, of course.)
723    
724     5. The above bug also affected the handling of POSIX character classes such as
725        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
726        permanent tables. Instead, the bit maps for such a class were previously
727        created as the appropriate unions of the upper, lower, and digit bitmaps.
728        Now they are created by subtraction from the [[:word:]] class, which has
729        its own bitmap.
730    
731     6. The [[:blank:]] character class matches horizontal, but not vertical space.
732        It is created by subtracting the vertical space characters (\x09, \x0a,
733        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
734        subtraction was done in the overall bitmap for a character class, meaning
735        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
736        be recognized. This bug has been fixed.
737    
738     7. Patches from the folks at Google:
739    
740          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
741          real life, but is still worth protecting against".
742    
743          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
744          regular expressions".
745    
746          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
747          have it.
748    
749          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
750          "configure" and the latter not, in order to fix a problem somebody had
751          with compiling the Arg class on HP-UX.
752    
753          (e) Improve the error-handling of the C++ wrapper a little bit.
754    
755          (f) New tests for checking recursion limiting.
756    
757     8. The pcre_memmove() function, which is used only if the environment does not
758        have a standard memmove() function (and is therefore rarely compiled),
759        contained two bugs: (a) use of int instead of size_t, and (b) it was not
760        returning a result (though PCRE never actually uses the result).
761    
762     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
763        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
764        returned instead of calling malloc() with an overflowing number that would
765        most likely cause subsequent chaos.
766    
767    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
768    
769    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
770        with this option is matched, the nmatch and pmatch options of regexec() are
771        ignored.
772    
773    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
774        provided in case anyone wants to the the POSIX interface with UTF-8
775        strings.
776    
777    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
778        C++ linking (needed for some HP-UX environments).
779    
780    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
781        (unused parameter) and in the pcre_printint() function (omitted "default"
782        switch label when the default is to do nothing).
783    
784    15. Added some code to make it possible, when PCRE is compiled as a C++
785        library, to replace subject pointers for pcre_exec() with a smart pointer
786        class, thus making it possible to process discontinuous strings.
787    
788    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
789        much the same function. They were added by different people who were trying
790        to make PCRE easy to compile on non-Unix systems. It has been suggested
791        that PCRE_EXPORT be abolished now that there is more automatic apparatus
792        for compiling on Windows systems. I have therefore replaced it with
793        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
794        defaults to "extern" for C or "extern C" for C++, which works fine on
795        Unix-like systems. It is now possible to override the value of PCRE_DATA_
796        SCOPE with something explicit in config.h. In addition:
797    
798        (a) pcreposix.h still had just "extern" instead of either of these macros;
799            I have replaced it with PCRE_DATA_SCOPE.
800    
801        (b) Functions such as _pcre_xclass(), which are internal to the library,
802            but external in the C sense, all had PCRE_EXPORT in their definitions.
803            This is apparently wrong for the Windows case, so I have removed it.
804            (It makes no difference on Unix-like systems.)
805    
806    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
807        of recursive calls to match(). This is different to MATCH_LIMIT because
808        that limits the total number of calls to match(), not all of which increase
809        the depth of recursion. Limiting the recursion depth limits the amount of
810        stack (or heap if NO_RECURSE is set) that is used. The default can be set
811        when PCRE is compiled, and changed at run time. A patch from Google adds
812        this functionality to the C++ interface.
813    
814    18. Changes to the handling of Unicode character properties:
815    
816        (a) Updated the table to Unicode 4.1.0.
817    
818        (b) Recognize characters that are not in the table as "Cn" (undefined).
819    
820        (c) I revised the way the table is implemented to a much improved format
821            which includes recognition of ranges. It now supports the ranges that
822            are defined in UnicodeData.txt, and it also amalgamates other
823            characters into ranges. This has reduced the number of entries in the
824            table from around 16,000 to around 3,000, thus reducing its size
825            considerably. I realized I did not need to use a tree structure after
826            all - a binary chop search is just as efficient. Having reduced the
827            number of entries, I extended their size from 6 bytes to 8 bytes to
828            allow for more data.
829    
830        (d) Added support for Unicode script names via properties such as \p{Han}.
831    
832    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
833        matching that character.
834    
835    20. When matching a repeated Unicode property with a minimum greater than zero,
836        (for example \pL{2,}), PCRE could look past the end of the subject if it
837        reached it while seeking the minimum number of characters. This could
838        happen only if some of the characters were more than one byte long, because
839        there is a check for at least the minimum number of bytes.
840    
841    21. Refactored the implementation of \p and \P so as to be more general, to
842        allow for more different types of property in future. This has changed the
843        compiled form incompatibly. Anybody with saved compiled patterns that use
844        \p or \P will have to recompile them.
845    
846    22. Added "Any" and "L&" to the supported property types.
847    
848    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
849        but give a compile time error if the value is greater than 0xff.
850    
851    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
852        accidentally not being installed or uninstalled.
853    
854    25. The pcre.h file was built from pcre.h.in, but the only changes that were
855        made were to insert the current release number. This seemed silly, because
856        it made things harder for people building PCRE on systems that don't run
857        "configure". I have turned pcre.h into a distributed file, no longer built
858        by "configure", with the version identification directly included. There is
859        no longer a pcre.h.in file.
860    
861        However, this change necessitated a change to the pcre-config script as
862        well. It is built from pcre-config.in, and one of the substitutions was the
863        release number. I have updated configure.ac so that ./configure now finds
864        the release number by grepping pcre.h.
865    
866    26. Added the ability to run the tests under valgrind.
867    
868    
869    Version 6.4 05-Sep-05
870    ---------------------
871    
872     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
873        "--" to be printed when multiple files were scanned, even when none of the
874        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
875        consider it to be a bug, and have restored the previous behaviour.
876    
877     2. A couple of code tidies to get rid of compiler warnings.
878    
879     3. The pcretest program used to cheat by referring to symbols in the library
880        whose names begin with _pcre_. These are internal symbols that are not
881        really supposed to be visible externally, and in some environments it is
882        possible to suppress them. The cheating is now confined to including
883        certain files from the library's source, which is a bit cleaner.
884    
885     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
886        file's purpose clearer.
887    
888     5. Reorganized pcre_ucp_findchar().
889    
890    
891    Version 6.3 15-Aug-05
892    ---------------------
893    
894     1. The file libpcre.pc.in did not have general read permission in the tarball.
895    
896     2. There were some problems when building without C++ support:
897    
898        (a) If C++ support was not built, "make install" and "make test" still
899            tried to test it.
900    
901        (b) There were problems when the value of CXX was explicitly set. Some
902            changes have been made to try to fix these, and ...
903    
904        (c) --disable-cpp can now be used to explicitly disable C++ support.
905    
906        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
907            backslash in a target when C++ was disabled. This confuses some
908            versions of "make", apparently. Using an intermediate variable solves
909            this. (Same for CPP_LOBJ.)
910    
911     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
912        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
913        necessary on certain architectures.
914    
915     4. Added a setting of -export-symbols-regex to the link command to remove
916        those symbols that are exported in the C sense, but actually are local
917        within the library, and not documented. Their names all begin with
918        "_pcre_". This is not a perfect job, because (a) we have to except some
919        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
920        available (and never for static libraries). I have made a note to try to
921        find a way round (a) in the future.
922    
923    
924    Version 6.2 01-Aug-05
925    ---------------------
926    
927     1. There was no test for integer overflow of quantifier values. A construction
928        such as {1111111111111111} would give undefined results. What is worse, if
929        a minimum quantifier for a parenthesized subpattern overflowed and became
930        negative, the calculation of the memory size went wrong. This could have
931        led to memory overwriting.
932    
933     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
934    
935     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
936        operating environments where this matters.
937    
938     4. Applied Giuseppe Maxia's patch to add additional features for controlling
939        PCRE options from within the C++ wrapper.
940    
941     5. Named capturing subpatterns were not being correctly counted when a pattern
942        was compiled. This caused two problems: (a) If there were more than 100
943        such subpatterns, the calculation of the memory needed for the whole
944        compiled pattern went wrong, leading to an overflow error. (b) Numerical
945        back references of the form \12, where the number was greater than 9, were
946        not recognized as back references, even though there were sufficient
947        previous subpatterns.
948    
949     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
950        versions of gcc, e.g. 2.95.4.
951    
952    
953    Version 6.1 21-Jun-05
954    ---------------------
955    
956     1. There was one reference to the variable "posix" in pcretest.c that was not
957        surrounded by "#if !defined NOPOSIX".
958    
959     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
960        the cross-check on the old pcre_info() function, for the benefit of the
961        cut-down version of PCRE that is currently imported into Exim.
962    
963     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
964        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
965        patterns that start (?i) but I don't think that matters. The use of (?i) is
966        just an example; this all applies to the other options as well.
967    
968     4. Since libtool seems to echo the compile commands it is issuing, the output
969        from "make" can be reduced a bit by putting "@" in front of each libtool
970        compile command.
971    
972     5. Patch from the folks at Google for configure.in to be a bit more thorough
973        in checking for a suitable C++ installation before trying to compile the
974        C++ stuff. This should fix a reported problem when a compiler was present,
975        but no suitable headers.
976    
977     6. The man pages all had just "PCRE" as their title. I have changed them to
978        be the relevant file name. I have also arranged that these names are
979        retained in the file doc/pcre.txt, which is a concatenation in text format
980        of all the man pages except the little individual ones for each function.
981    
982     7. The NON-UNIX-USE file had not been updated for the different set of source
983        files that come with release 6. I also added a few comments about the C++
984        wrapper.
985    
986    
987    Version 6.0 07-Jun-05
988    ---------------------
989    
990     1. Some minor internal re-organization to help with my DFA experiments.
991    
992     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
993        didn't matter for the library itself when fully configured, but did matter
994        when compiling without UCP support, or within Exim, where the ucp files are
995        not imported.
996    
997     3. Refactoring of the library code to split up the various functions into
998        different source modules. The addition of the new DFA matching code (see
999        below) to a single monolithic source would have made it really too
1000        unwieldy, quite apart from causing all the code to be include in a
1001        statically linked application, when only some functions are used. This is
1002        relevant even without the DFA addition now that patterns can be compiled in
1003        one application and matched in another.
1004    
1005        The downside of splitting up is that there have to be some external
1006        functions and data tables that are used internally in different modules of
1007        the library but which are not part of the API. These have all had their
1008        names changed to start with "_pcre_" so that they are unlikely to clash
1009        with other external names.
1010    
1011     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1012        a different (DFA) algorithm. Although it is slower than the original
1013        function, it does have some advantages for certain types of matching
1014        problem.
1015    
1016     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1017        including restarting after a partial match.
1018    
1019     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1020        defined when compiling for Windows was sent to me. I have put it into the
1021        code, though I have no means of testing or verifying it.
1022    
1023     7. Added the pcre_refcount() auxiliary function.
1024    
1025     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1026        match before or at the first newline in the subject string. In pcretest,
1027        the /f option on a pattern can be used to set this.
1028    
1029     9. A repeated \w when used in UTF-8 mode with characters greater than 256
1030        would behave wrongly. This has been present in PCRE since release 4.0.
1031    
1032    10. A number of changes to the pcregrep command:
1033    
1034        (a) Refactored how -x works; insert ^(...)$ instead of setting
1035            PCRE_ANCHORED and checking the length, in preparation for adding
1036            something similar for -w.
1037    
1038        (b) Added the -w (match as a word) option.
1039    
1040        (c) Refactored the way lines are read and buffered so as to have more
1041            than one at a time available.
1042    
1043        (d) Implemented a pcregrep test script.
1044    
1045        (e) Added the -M (multiline match) option. This allows patterns to match
1046            over several lines of the subject. The buffering ensures that at least
1047            8K, or the rest of the document (whichever is the shorter) is available
1048            for matching (and similarly the previous 8K for lookbehind assertions).
1049    
1050        (f) Changed the --help output so that it now says
1051    
1052              -w, --word-regex(p)
1053    
1054            instead of two lines, one with "regex" and the other with "regexp"
1055            because that confused at least one person since the short forms are the
1056            same. (This required a bit of code, as the output is generated
1057            automatically from a table. It wasn't just a text change.)
1058    
1059        (g) -- can be used to terminate pcregrep options if the next thing isn't an
1060            option but starts with a hyphen. Could be a pattern or a path name
1061            starting with a hyphen, for instance.
1062    
1063        (h) "-" can be given as a file name to represent stdin.
1064    
1065        (i) When file names are being printed, "(standard input)" is used for
1066            the standard input, for compatibility with GNU grep. Previously
1067            "<stdin>" was used.
1068    
1069        (j) The option --label=xxx can be used to supply a name to be used for
1070            stdin when file names are being printed. There is no short form.
1071    
1072        (k) Re-factored the options decoding logic because we are going to add
1073            two more options that take data. Such options can now be given in four
1074            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1075    
1076        (l) Added the -A, -B, and -C options for requesting that lines of context
1077            around matches be printed.
1078    
1079        (m) Added the -L option to print the names of files that do not contain
1080            any matching lines, that is, the complement of -l.
1081    
1082        (n) The return code is 2 if any file cannot be opened, but pcregrep does
1083            continue to scan other files.
1084    
1085        (o) The -s option was incorrectly implemented. For compatibility with other
1086            greps, it now suppresses the error message for a non-existent or non-
1087            accessible file (but not the return code). There is a new option called
1088            -q that suppresses the output of matching lines, which was what -s was
1089            previously doing.
1090    
1091        (p) Added --include and --exclude options to specify files for inclusion
1092            and exclusion when recursing.
1093    
1094    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1095        Hopefully, it now does.
1096    
1097    12. Missing cast in pcre_study().
1098    
1099    13. Added an "uninstall" target to the makefile.
1100    
1101    14. Replaced "extern" in the function prototypes in Makefile.in with
1102        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1103        world, but is set differently for Windows.
1104    
1105    15. Added a second compiling function called pcre_compile2(). The only
1106        difference is that it has an extra argument, which is a pointer to an
1107        integer error code. When there is a compile-time failure, this is set
1108        non-zero, in addition to the error test pointer being set to point to an
1109        error message. The new argument may be NULL if no error number is required
1110        (but then you may as well call pcre_compile(), which is now just a
1111        wrapper). This facility is provided because some applications need a
1112        numeric error indication, but it has also enabled me to tidy up the way
1113        compile-time errors are handled in the POSIX wrapper.
1114    
1115    16. Added VPATH=.libs to the makefile; this should help when building with one
1116        prefix path and installing with another. (Or so I'm told by someone who
1117        knows more about this stuff than I do.)
1118    
1119    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1120        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1121        match everything, including newlines. This is not POSIX-compatible, but
1122        somebody wanted the feature. From pcretest it can be activated by using
1123        both the P and the s flags.
1124    
1125    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1126    
1127    19. libpcre.pc was being incorrectly installed as executable.
1128    
1129    20. A couple of places in pcretest check for end-of-line by looking for '\n';
1130        it now also looks for '\r' so that it will work unmodified on Windows.
1131    
1132    21. Added Google's contributed C++ wrapper to the distribution.
1133    
1134    22. Added some untidy missing memory free() calls in pcretest, to keep
1135        Electric Fence happy when testing.
1136    
1137    
1138    
1139    Version 5.0 13-Sep-04
1140    ---------------------
1141    
1142     1. Internal change: literal characters are no longer packed up into items
1143        containing multiple characters in a single byte-string. Each character
1144        is now matched using a separate opcode. However, there may be more than one
1145        byte in the character in UTF-8 mode.
1146    
1147     2. The pcre_callout_block structure has two new fields: pattern_position and
1148        next_item_length. These contain the offset in the pattern to the next match
1149        item, and its length, respectively.
1150    
1151     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1152        insertion of callouts before each pattern item. Added the /C option to
1153        pcretest to make use of this.
1154    
1155     4. On the advice of a Windows user, the lines
1156    
1157          #if defined(_WIN32) || defined(WIN32)
1158          _setmode( _fileno( stdout ), 0x8000 );
1159          #endif  /* defined(_WIN32) || defined(WIN32) */
1160    
1161        have been added to the source of pcretest. This apparently does useful
1162        magic in relation to line terminators.
1163    
1164     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1165        for the benefit of those environments where the "b" makes a difference.
1166    
1167     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1168        to know about it. I have put a hack into configure.in that adds in code
1169        to set GCC=yes if CC=icc. This seems to end up at a point in the
1170        generated configure script that is early enough to affect the setting of
1171        compiler options, which is what is needed, but I have no means of testing
1172        whether it really works. (The user who reported this had patched the
1173        generated configure script, which of course I cannot do.)
1174    
1175        LATER: After change 22 below (new libtool files), the configure script
1176        seems to know about icc (and also ecc). Therefore, I have commented out
1177        this hack in configure.in.
1178    
1179     7. Added support for pkg-config (2 patches were sent in).
1180    
1181     8. Negated POSIX character classes that used a combination of internal tables
1182        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1183        [[:^ascii]]. Typically, they would match almost any characters. The other
1184        POSIX classes were not broken in this way.
1185    
1186     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1187        to find the match, as PCRE was deluded into thinking that the match had to
1188        start at the start point or following a newline. The same bug applied to
1189        patterns with negative forward assertions or any backward assertions
1190        preceding ".*" at the start, unless the pattern required a fixed first
1191        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1192    
1193    10. In UTF-8 mode, when moving forwards in the subject after a failed match
1194        starting at the last subject character, bytes beyond the end of the subject
1195        string were read.
1196    
1197    11. Renamed the variable "class" as "classbits" to make life easier for C++
1198        users. (Previously there was a macro definition, but it apparently wasn't
1199        enough.)
1200    
1201    12. Added the new field "tables" to the extra data so that tables can be passed
1202        in at exec time, or the internal tables can be re-selected. This allows
1203        a compiled regex to be saved and re-used at a later time by a different
1204        program that might have everything at different addresses.
1205    
1206    13. Modified the pcre-config script so that, when run on Solaris, it shows a
1207        -R library as well as a -L library.
1208    
1209    14. The debugging options of pcretest (-d on the command line or D on a
1210        pattern) showed incorrect output for anything following an extended class
1211        that contained multibyte characters and which was followed by a quantifier.
1212    
1213    15. Added optional support for general category Unicode character properties
1214        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1215        support. It adds about 90K to the size of the library. The meanings of the
1216        inbuilt class escapes such as \d and \s have NOT been changed.
1217    
1218    16. Updated pcredemo.c to include calls to free() to release the memory for the
1219        compiled pattern.
1220    
1221    17. The generated file chartables.c was being created in the source directory
1222        instead of in the building directory. This caused the build to fail if the
1223        source directory was different from the building directory, and was
1224        read-only.
1225    
1226    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1227        file. No doubt somebody will tell me if they don't make sense... Also added
1228        Dan Mooney's comments about building on OpenVMS.
1229    
1230    19. Added support for partial matching via the PCRE_PARTIAL option for
1231        pcre_exec() and the \P data escape in pcretest.
1232    
1233    20. Extended pcretest with 3 new pattern features:
1234    
1235        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
1236              write the compiled pattern to the file whose name is "rest-of-line".
1237              This is a straight binary dump of the data, with the saved pointer to
1238              the character tables forced to be NULL. The study data, if any, is
1239              written too. After writing, pcretest reads a new pattern.
1240    
1241        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1242              compiled pattern from the given file. There must not be any
1243              occurrences of "<" in the file name (pretty unlikely); if there are,
1244              pcretest will instead treat the initial "<" as a pattern delimiter.
1245              After reading in the pattern, pcretest goes on to read data lines as
1246              usual.
1247    
1248        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1249              and 16-bit fields in a compiled pattern, to simulate a pattern that
1250              was compiled on a host of opposite endianness.
1251    
1252    21. The pcre-exec() function can now cope with patterns that were compiled on
1253        hosts of opposite endianness, with this restriction:
1254    
1255          As for any compiled expression that is saved and used later, the tables
1256          pointer field cannot be preserved; the extra_data field in the arguments
1257          to pcre_exec() should be used to pass in a tables address if a value
1258          other than the default internal tables were used at compile time.
1259    
1260    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1261        now diagnosed as an error. Previously, most of the time, a negative number
1262        would have been treated as zero, but if in addition "ovector" was passed as
1263        NULL, a crash could occur.
1264    
1265    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1266        new versions from the libtool 1.5 distribution (the last one is a copy of
1267        a file called libtool.m4). This seems to have fixed the need to patch
1268        "configure" to support Darwin 1.3 (which I used to do). However, I still
1269        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1270        workstation).
1271    
1272    24. Changed the PCRE licence to be the more standard "BSD" licence.
1273    
1274    
1275    Version 4.5 01-Dec-03
1276    ---------------------
1277    
1278     1. There has been some re-arrangement of the code for the match() function so
1279        that it can be compiled in a version that does not call itself recursively.
1280        Instead, it keeps those local variables that need separate instances for
1281        each "recursion" in a frame on the heap, and gets/frees frames whenever it
1282        needs to "recurse". Keeping track of where control must go is done by means
1283        of setjmp/longjmp. The whole thing is implemented by a set of macros that
1284        hide most of the details from the main code, and operates only if
1285        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1286        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1287        operating.
1288    
1289        To make it easier for callers to provide specially tailored get/free
1290        functions for this usage, two new functions, pcre_stack_malloc, and
1291        pcre_stack_free, are used. They are always called in strict stacking order,
1292        and the size of block requested is always the same.
1293    
1294        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1295        PCRE has been compiled to use the stack or the heap for recursion. The
1296        -C option of pcretest uses this to show which version is compiled.
1297    
1298        A new data escape \S, is added to pcretest; it causes the amounts of store
1299        obtained and freed by both kinds of malloc/free at match time to be added
1300        to the output.
1301    
1302     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1303        what's available on my current Linux desktop machine.
1304    
1305     3. When matching a UTF-8 string, the test for a valid string at the start has
1306        been extended. If start_offset is not zero, PCRE now checks that it points
1307        to a byte that is the start of a UTF-8 character. If not, it returns
1308        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1309        this is necessary because there may be backward assertions in the pattern.
1310        When matching the same subject several times, it may save resources to use
1311        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1312    
1313     4. The code for checking the validity of UTF-8 strings has been tightened so
1314        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1315        containing "overlong sequences".
1316    
1317     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1318        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1319        should have been "&", but it just so happened that all the cases this let
1320        through by mistake were picked up later in the function.
1321    
1322     6. I had used a variable called "isblank" - this is a C99 function, causing
1323        some compilers to warn. To avoid this, I renamed it (as "blankclass").
1324    
1325     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1326        prompting; (b) run "./pcretest /dev/null" at the start of the test script
1327        so the version is shown; (c) stop "make test" echoing "./RunTest".
1328    
1329     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1330    
1331     9. The prototype for memmove() for systems that don't have it was using
1332        size_t, but the inclusion of the header that defines size_t was later. I've
1333        moved the #includes for the C headers earlier to avoid this.
1334    
1335    10. Added some adjustments to the code to make it easier to compiler on certain
1336        special systems:
1337    
1338          (a) Some "const" qualifiers were missing.
1339          (b) Added the macro EXPORT before all exported functions; by default this
1340              is defined to be empty.
1341          (c) Changed the dftables auxiliary program (that builds chartables.c) so
1342              that it reads its output file name as an argument instead of writing
1343              to the standard output and assuming this can be redirected.
1344    
1345    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
1346        class containing characters with values greater than 255, PCRE compilation
1347        went into a loop.
1348    
1349    12. A recursive reference to a subpattern that was within another subpattern
1350        that had a minimum quantifier of zero caused PCRE to crash. For example,
1351        (x(y(?2))z)? provoked this bug with a subject that got as far as the
1352        recursion. If the recursively-called subpattern itself had a zero repeat,
1353        that was OK.
1354    
1355    13. In pcretest, the buffer for reading a data line was set at 30K, but the
1356        buffer into which it was copied (for escape processing) was still set at
1357        1024, so long lines caused crashes.
1358    
1359    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
1360        "internal error: code overflow...". This applied to any character class
1361        that was followed by a possessive quantifier.
1362    
1363    15. Modified the Makefile to add libpcre.la as a prerequisite for
1364        libpcreposix.la because I was told this is needed for a parallel build to
1365        work.
1366    
1367    16. If a pattern that contained .* following optional items at the start was
1368        studied, the wrong optimizing data was generated, leading to matching
1369        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
1370        matching string must start with a or b or c. The correct conclusion for
1371        this pattern is that a match can start with any character.
1372    
1373    
1374    Version 4.4 13-Aug-03
1375    ---------------------
1376    
1377     1. In UTF-8 mode, a character class containing characters with values between
1378        127 and 255 was not handled correctly if the compiled pattern was studied.
1379        In fixing this, I have also improved the studying algorithm for such
1380        classes (slightly).
1381    
1382     2. Three internal functions had redundant arguments passed to them. Removal
1383        might give a very teeny performance improvement.
1384    
1385     3. Documentation bug: the value of the capture_top field in a callout is *one
1386        more than* the number of the hightest numbered captured substring.
1387    
1388     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
1389        in incorrectly linking with a previously installed version. They now link
1390        explicitly with libpcre.la.
1391    
1392     5. configure.in no longer needs to recognize Cygwin specially.
1393    
1394     6. A problem in pcre.in for Windows platforms is fixed.
1395    
1396     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
1397        pcretest, it used to include the size of the study block as part of its
1398        output. Unfortunately, the structure contains a field that has a different
1399        size on different hardware architectures. This meant that the tests that
1400        showed this size failed. As the block is currently always of a fixed size,
1401        this information isn't actually particularly useful in pcretest output, so
1402        I have just removed it.
1403    
1404     8. Three pre-processor statements accidentally did not start in column 1.
1405        Sadly, there are *still* compilers around that complain, even though
1406        standard C has not required this for well over a decade. Sigh.
1407    
1408     9. In pcretest, the code for checking callouts passed small integers in the
1409        callout_data field, which is a void * field. However, some picky compilers
1410        complained about the casts involved for this on 64-bit systems. Now
1411        pcretest passes the address of the small integer instead, which should get
1412        rid of the warnings.
1413    
1414    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
1415        both compile and run time, and gives an error if an invalid UTF-8 sequence
1416        is found. There is a option for disabling this check in cases where the
1417        string is known to be correct and/or the maximum performance is wanted.
1418    
1419    11. In response to a bug report, I changed one line in Makefile.in from
1420    
1421            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
1422        to
1423            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
1424    
1425        to look similar to other lines, but I have no way of telling whether this
1426        is the right thing to do, as I do not use Windows. No doubt I'll get told
1427        if it's wrong...
1428    
1429    
1430    Version 4.3 21-May-03
1431    ---------------------
1432    
1433    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
1434       Makefile.
1435    
1436    2. Some refactoring to improve the quality of the code:
1437    
1438       (i)   The utf8_table... variables are now declared "const".
1439    
1440       (ii)  The code for \cx, which used the "case flipping" table to upper case
1441             lower case letters, now just substracts 32. This is ASCII-specific,
1442             but the whole concept of \cx is ASCII-specific, so it seems
1443             reasonable.
1444    
1445       (iii) PCRE was using its character types table to recognize decimal and
1446             hexadecimal digits in the pattern. This is silly, because it handles
1447             only 0-9, a-f, and A-F, but the character types table is locale-
1448             specific, which means strange things might happen. A private
1449             table is now used for this - though it costs 256 bytes, a table is
1450             much faster than multiple explicit tests. Of course, the standard
1451             character types table is still used for matching digits in subject
1452             strings against \d.
1453    
1454       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
1455             ending in _t are). So I've renamed it as ESC_tee.
1456    
1457    3. The first argument for regexec() in the POSIX wrapper should have been
1458       defined as "const".
1459    
1460    4. Changed pcretest to use malloc() for its buffers so that they can be
1461       Electric Fenced for debugging.
1462    
1463    5. There were several places in the code where, in UTF-8 mode, PCRE would try
1464       to read one or more bytes before the start of the subject string. Often this
1465       had no effect on PCRE's behaviour, but in some circumstances it could
1466       provoke a segmentation fault.
1467    
1468    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
1469       to try to read one or more bytes before the start of the subject string.
1470    
1471    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
1472       UTF-8 support could misbehave in various ways if the subject string
1473       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
1474       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
1475       back over UTF-8 characters.)
1476    
1477    
1478    Version 4.2 14-Apr-03
1479    ---------------------
1480    
1481    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
1482    
1483    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
1484         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
1485         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
1486         [WIN_PREFIX]: new variable, "cyg" for Cygwin
1487         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
1488           and BUILD_EXEEXT
1489         Note: automatic setting of the BUILD variables is not yet working
1490         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
1491           compile-time but not at link-time
1492         [LINK]: use for linking executables only
1493         make different versions for Windows and non-Windows
1494         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
1495           libraries
1496         [LINK_FOR_BUILD]: new variable
1497         [OBJEXT]: use throughout
1498         [EXEEXT]: use throughout
1499         <winshared>: new target
1500         <wininstall>: new target
1501         <dftables.o>: use native compiler
1502         <dftables>: use native linker
1503         <install>: handle Windows platform correctly
1504         <clean>: ditto
1505         <check>: ditto
1506         copy DLL to top builddir before testing
1507    
1508       As part of these changes, -no-undefined was removed again. This was reported
1509       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
1510       in any case.
1511    
1512    3. Some tidies to get rid of compiler warnings:
1513    
1514       . In the match_data structure, match_limit was an unsigned long int, whereas
1515         match_call_count was an int. I've made them both unsigned long ints.
1516    
1517       . In pcretest the fact that a const uschar * doesn't automatically cast to
1518         a void * provoked a warning.
1519    
1520       . Turning on some more compiler warnings threw up some "shadow" variables
1521         and a few more missing casts.
1522    
1523    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
1524       option, a class that contained a single character with a value between 128
1525       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
1526    
1527    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
1528       option, a class that contained several characters, but with at least one
1529       whose value was between 128 and 255 caused PCRE to crash.
1530    
1531    
1532    Version 4.1 12-Mar-03
1533    ---------------------
1534    
1535    1. Compiling with gcc -pedantic found a couple of places where casts were
1536    needed, and a string in dftables.c that was longer than standard compilers are
1537    required to support.
1538    
1539    2. Compiling with Sun's compiler found a few more places where the code could
1540    be tidied up in order to avoid warnings.
1541    
1542    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
1543    first of these names is deprecated in the latest Autoconf in favour of the name
1544    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
1545    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
1546    analogy I have changed it to CFLAGS_FOR_BUILD.
1547    
1548    4. Added -no-undefined to the linking command in the Makefile, because this is
1549    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
1550    linking step for the pcreposix library.
1551    
1552    5. PCRE was failing to diagnose the case of two named groups with the same
1553    name.
1554    
1555    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
1556    literal character that is needed in the subject for a match, and scans along to
1557    ensure that it is present before embarking on the full matching process. This
1558    saves time in cases of nested unlimited repeats that are never going to match.
1559    Problem: the scan can take a lot of time if the subject is very long (e.g.
1560    megabytes), thus penalizing straightforward matches. It is now done only if the
1561    amount of subject to be scanned is less than 1000 bytes.
1562    
1563    7. A lesser problem with the same optimization is that it was recording the
1564    first character of an anchored pattern as "needed", thus provoking a search
1565    right along the subject, even when the first match of the pattern was going to
1566    fail. The "needed" character is now not set for anchored patterns, unless it
1567    follows something in the pattern that is of non-fixed length. Thus, it still
1568    fulfils its original purpose of finding quick non-matches in cases of nested
1569    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
1570    
1571    
1572    Version 4.0 17-Feb-03
1573    ---------------------
1574    
1575    1. If a comment in an extended regex that started immediately after a meta-item
1576    extended to the end of string, PCRE compiled incorrect data. This could lead to
1577    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
1578    
1579    2. Moved to autoconf 2.53 and libtool 1.4.2.
1580    
1581    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
1582    the special perltest8 script is no longer needed - all the tests can be run
1583    from a single perltest script.
1584    
1585    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
1586    by \s. It has now been removed in PCRE. This means it isn't recognized as
1587    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
1588    class [:space:] *does* include VT, thereby creating a mess.
1589    
1590    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
1591    space and tab.
1592    
1593    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
1594    its new features into the main test script, reducing the number of scripts.
1595    
1596    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
1597    were backward compatible, and made the (?i) apply to the whole pattern, as if
1598    /i were given. Now it behaves more logically, and applies the option setting
1599    only to what follows. PCRE has been changed to follow suit. However, if it
1600    finds options settings right at the start of the pattern, it extracts them into
1601    the global options, as before. Thus, they show up in the info data.
1602    
1603    8. Added support for the \Q...\E escape sequence. Characters in between are
1604    treated as literals. This is slightly different from Perl in that $ and @ are
1605    also handled as literals inside the quotes. In Perl, they will cause variable
1606    interpolation. Note the following examples:
1607    
1608        Pattern            PCRE matches      Perl matches
1609    
1610        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
1611        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
1612        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
1613    
1614    For compatibility with Perl, \Q...\E sequences are recognized inside character
1615    classes as well as outside them.
1616    
1617    9. Re-organized 3 code statements in pcretest to avoid "overflow in
1618    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
1619    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
1620    signed/unsigned warnings.
1621    
1622    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
1623    option for pcretest, so I've replaced it by a simple function that does just
1624    that job.
1625    
1626    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
1627    "pcregrep -".
1628    
1629    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
1630    Java package. This provides some syntactic sugar for simple cases of what my
1631    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
1632    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
1633    item, you can use this simplified notation. Note that only makes sense with
1634    greedy quantifiers. Consequently, the use of the possessive quantifier forces
1635    greediness, whatever the setting of the PCRE_UNGREEDY option.
1636    
1637    13. A change of greediness default within a pattern was not taking effect at
1638    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
1639    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
1640    was abstracted outside.
1641    
1642    14. PCRE now supports the \G assertion. It is true when the current matching
1643    position is at the start point of the match. This differs from \A when the
1644    starting offset is non-zero. Used with the /g option of pcretest (or similar
1645    code), it works in the same way as it does for Perl's /g option. If all
1646    alternatives of a regex begin with \G, the expression is anchored to the start
1647    match position, and the "anchored" flag is set in the compiled expression.
1648    
1649    15. Some bugs concerning the handling of certain option changes within patterns
1650    have been fixed. These applied to options other than (?ims). For example,
1651    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
1652    been the other way round. Some of this was related to change 7 above.
1653    
1654    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
1655    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
1656    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
1657    POSIX classes only within a class (e.g. /[[:alpha:]]/).
1658    
1659    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
1660    mode. Unlike ".", it always matches newline, whatever the setting of
1661    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
1662    assertions. Perl allows it, but it doesn't (in general) work because it can't
1663    calculate the length of the lookbehind. At least, that's the case for Perl
1664    5.8.0 - I've been told they are going to document that it doesn't work in
1665    future.
1666    
1667    18. Added an error diagnosis for escapes that PCRE does not support: these are
1668    \L, \l, \N, \P, \p, \U, \u, and \X.
1669    
1670    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
1671    reading past the end of the pattern in cases such as /[abcd/.
1672    
1673    20. PCRE was getting more memory than necessary for patterns with classes that
1674    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
1675    
1676    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
1677    compiling PCRE for use with Virtual Pascal.
1678    
1679    22. Small fix to the Makefile to make it work properly if the build is done
1680    outside the source tree.
1681    
1682    23. Added a new extension: a condition to go with recursion. If a conditional
1683    subpattern starts with (?(R) the "true" branch is used if recursion has
1684    happened, whereas the "false" branch is used only at the top level.
1685    
1686    24. When there was a very long string of literal characters (over 255 bytes
1687    without UTF support, over 250 bytes with UTF support), the computation of how
1688    much memory was required could be incorrect, leading to segfaults or other
1689    strange effects.
1690    
1691    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
1692    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
1693    there was a subsequent back reference to those brackets. This meant that, for
1694    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
1695    possible to check for precisely this case. All we can do is abandon the
1696    optimization if .* occurs inside capturing brackets when there are any back
1697    references whatsoever. (See below for a better fix that came later.)
1698    
1699    26. The handling of the optimization for finding the first character of a
1700    non-anchored pattern, and for finding a character that is required later in the
1701    match were failing in some cases. This didn't break the matching; it just
1702    failed to optimize when it could. The way this is done has been re-implemented.
1703    
1704    27. Fixed typo in error message for invalid (?R item (it said "(?p").
1705    
1706    28. Added a new feature that provides some of the functionality that Perl
1707    provides with (?{...}). The facility is termed a "callout". The way it is done
1708    in PCRE is for the caller to provide an optional function, by setting
1709    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
1710    global variable. By default it is unset, which disables all calling out. To get
1711    the function called, the regex must include (?C) at appropriate points. This
1712    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
1713    This provides a means of identifying different callout points. When PCRE
1714    reaches such a point in the regex, if pcre_callout has been set, the external
1715    function is called. It is provided with data in a structure called
1716    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
1717    matching continues; if it returns a non-zero value, the match at the current
1718    point fails. However, backtracking will occur if possible. [This was changed
1719    later and other features added - see item 49 below.]
1720    
1721    29. pcretest is upgraded to test the callout functionality. It provides a
1722    callout function that displays information. By default, it shows the start of
1723    the match and the current position in the text. There are some new data escapes
1724    to vary what happens:
1725    
1726        \C+         in addition, show current contents of captured substrings
1727        \C-         do not supply a callout function
1728        \C!n        return 1 when callout number n is reached
1729        \C!n!m      return 1 when callout number n is reached for the mth time
1730    
1731    30. If pcregrep was called with the -l option and just a single file name, it
1732    output "<stdin>" if a match was found, instead of the file name.
1733    
1734    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
1735    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
1736    pcre_exec(). This saves a malloc/free per call. The default value of
1737    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
1738    when configuring.
1739    
1740    32. The default maximum size of a compiled pattern is 64K. There have been a
1741    few cases of people hitting this limit. The code now uses macros to handle the
1742    storing of links as offsets within the compiled pattern. It defaults to 2-byte
1743    links, but this can be changed to 3 or 4 bytes by --with-link-size when
1744    configuring. Tests 2 and 5 work only with 2-byte links because they output
1745    debugging information about compiled patterns.
1746    
1747    33. Internal code re-arrangements:
1748    
1749    (a) Moved the debugging function for printing out a compiled regex into
1750        its own source file (printint.c) and used #include to pull it into
1751        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
1752        separate copies.
1753    
1754    (b) Defined the list of op-code names for debugging as a macro in
1755        internal.h so that it is next to the definition of the opcodes.
1756    
1757    (c) Defined a table of op-code lengths for simpler skipping along compiled
1758        code. This is again a macro in internal.h so that it is next to the
1759        definition of the opcodes.
1760    
1761    34. Added support for recursive calls to individual subpatterns, along the
1762    lines of Robin Houston's patch (but implemented somewhat differently).
1763    
1764    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
1765    allow it to read and process whole directories in Win32. This code was
1766    contributed by Lionel Fourquaux; it has not been tested by me.
1767    
1768    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
1769    used to name a group. Names consist of alphanumerics and underscores, and must
1770    be unique. Back references use the syntax (?P=name) and recursive calls use
1771    (?P>name) which is a PCRE extension to the Python extension. Groups still have
1772    numbers. The function pcre_fullinfo() can be used after compilation to extract
1773    a name/number map. There are three relevant calls:
1774    
1775      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
1776      PCRE_INFO_NAMECOUNT            yields the number of entries
1777      PCRE_INFO_NAMETABLE            yields a pointer to the map.
1778    
1779    The map is a vector of fixed-size entries. The size of each entry depends on
1780    the length of the longest name used. The first two bytes of each entry are the
1781    group number, most significant byte first. There follows the corresponding
1782    name, zero terminated. The names are in alphabetical order.
1783    
1784    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
1785    case instead of 255. Making it the same both with and without UTF-8 support
1786    means that the same test output works with both.
1787    
1788    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
1789    calling malloc() with a zero argument.
1790    
1791    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
1792    optimization. I've improved things by keeping a bitmap of backreferences with
1793    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
1794    fact referenced, the optimization can be applied. It is unlikely that a
1795    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
1796    the match to follow \n) will appear inside brackets with a number greater than
1797    31, but if it does, any back reference > 31 suppresses the optimization.
1798    
1799    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
1800    of disabling numbered capturing parentheses. Any opening parenthesis that is
1801    not followed by ? behaves as if it were followed by ?: but named parentheses
1802    can still be used for capturing (and they will acquire numbers in the usual
1803    way).
1804    
1805    41. Redesigned the return codes from the match() function into yes/no/error so
1806    that errors can be passed back from deep inside the nested calls. A malloc
1807    failure while inside a recursive subpattern call now causes the
1808    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
1809    
1810    42. It is now possible to set a limit on the number of times the match()
1811    function is called in a call to pcre_exec(). This facility makes it possible to
1812    limit the amount of recursion and backtracking, though not in a directly
1813    obvious way, because the match() function is used in a number of different
1814    circumstances. The count starts from zero for each position in the subject
1815    string (for non-anchored patterns). The default limit is, for compatibility, a
1816    large number, namely 10 000 000. You can change this in two ways:
1817    
1818    (a) When configuring PCRE before making, you can use --with-match-limit=n
1819        to set a default value for the compiled library.
1820    
1821    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
1822        a different value is set. See 45 below.
1823    
1824    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
1825    
1826    43. Added a new function pcre_config(int, void *) to enable run-time extraction
1827    of things that can be changed at compile time. The first argument specifies
1828    what is wanted and the second points to where the information is to be placed.
1829    The current list of available information is:
1830    
1831      PCRE_CONFIG_UTF8
1832    
1833    The output is an integer that is set to one if UTF-8 support is available;
1834    otherwise it is set to zero.
1835    
1836      PCRE_CONFIG_NEWLINE
1837    
1838    The output is an integer that it set to the value of the code that is used for
1839    newline. It is either LF (10) or CR (13).
1840    
1841      PCRE_CONFIG_LINK_SIZE
1842    
1843    The output is an integer that contains the number of bytes used for internal
1844    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
1845    
1846      PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
1847    
1848    The output is an integer that contains the threshold above which the POSIX
1849    interface uses malloc() for output vectors. See item 31 above.
1850    
1851      PCRE_CONFIG_MATCH_LIMIT
1852    
1853    The output is an unsigned integer that contains the default limit of the number
1854    of match() calls in a pcre_exec() execution. See 42 above.
1855    
1856    44. pcretest has been upgraded by the addition of the -C option. This causes it
1857    to extract all the available output from the new pcre_config() function, and to
1858    output it. The program then exits immediately.
1859    
1860    45. A need has arisen to pass over additional data with calls to pcre_exec() in
1861    order to support additional features. One way would have been to define
1862    pcre_exec2() (for example) with extra arguments, but this would not have been
1863    extensible, and would also have required all calls to the original function to
1864    be mapped to the new one. Instead, I have chosen to extend the mechanism that
1865    is used for passing in "extra" data from pcre_study().
1866    
1867    The pcre_extra structure is now exposed and defined in pcre.h. It currently
1868    contains the following fields:
1869    
1870      flags         a bitmap indicating which of the following fields are set
1871      study_data    opaque data from pcre_study()
1872      match_limit   a way of specifying a limit on match() calls for a specific
1873                      call to pcre_exec()
1874      callout_data  data for callouts (see 49 below)
1875    
1876    The flag bits are also defined in pcre.h, and are
1877    
1878      PCRE_EXTRA_STUDY_DATA
1879      PCRE_EXTRA_MATCH_LIMIT
1880      PCRE_EXTRA_CALLOUT_DATA
1881    
1882    The pcre_study() function now returns one of these new pcre_extra blocks, with
1883    the actual study data pointed to by the study_data field, and the
1884    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
1885    before. That is, this change is entirely upwards-compatible and requires no
1886    change to existing code.
1887    
1888    If you want to pass in additional data to pcre_exec(), you can either place it
1889    in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
1890    block.
1891    
1892    46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
1893    data string contains the escape sequence \M, pcretest calls pcre_exec() several
1894    times with different match limits, until it finds the minimum value needed for
1895    pcre_exec() to complete. The value is then output. This can be instructive; for
1896    most simple matches the number is quite small, but for pathological cases it
1897    gets very large very quickly.
1898    
1899    47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
1900    returns the size of the data block pointed to by the study_data field in a
1901    pcre_extra block, that is, the value that was passed as the argument to
1902    pcre_malloc() when PCRE was getting memory in which to place the information
1903    created by pcre_study(). The fourth argument should point to a size_t variable.
1904    pcretest has been extended so that this information is shown after a successful
1905    pcre_study() call when information about the compiled regex is being displayed.
1906    
1907    48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
1908    because what follows is always an absolute path. (Later: it turns out that this
1909    is more than cosmetic for MinGW, because it doesn't like empty path
1910    components.)
1911    
1912    49. Some changes have been made to the callout feature (see 28 above):
1913    
1914    (i)  A callout function now has three choices for what it returns:
1915    
1916           0  =>  success, carry on matching
1917         > 0  =>  failure at this point, but backtrack if possible
1918         < 0  =>  serious error, return this value from pcre_exec()
1919    
1920         Negative values should normally be chosen from the set of PCRE_ERROR_xxx
1921         values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
1922         "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
1923         use by callout functions. It will never be used by PCRE itself.
1924    
1925    (ii) The pcre_extra structure (see 45 above) has a void * field called
1926         callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
1927         pcre_callout_block structure has a field of the same name. The contents of
1928         the field passed in the pcre_extra structure are passed to the callout
1929         function in the corresponding field in the callout block. This makes it
1930         easier to use the same callout-containing regex from multiple threads. For
1931         testing, the pcretest program has a new data escape
1932    
1933           \C*n        pass the number n (may be negative) as callout_data
1934    
1935         If the callout function in pcretest receives a non-zero value as
1936         callout_data, it returns that value.
1937    
1938    50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
1939    there were some redundant $(CFLAGS) in commands that are now specified as
1940    $(LINK), which already includes $(CFLAGS).
1941    
1942    51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
1943    has been compiled with UTF-8 support *and* pcre_compile() has been compiled
1944    with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
1945    one-byte characters throughout. Note that case-insensitive matching applies
1946    only to characters whose values are less than 256. PCRE doesn't support the
1947    notion of cases for higher-valued characters.
1948    
1949    (i)   A character class whose characters are all within 0-255 is handled as
1950          a bit map, and the map is inverted for negative classes. Previously, a
1951          character > 255 always failed to match such a class; however it should
1952          match if the class was a negative one (e.g. [^ab]). This has been fixed.
1953    
1954    (ii)  A negated character class with a single character < 255 is coded as
1955          "not this character" (OP_NOT). This wasn't working properly when the test
1956          character was multibyte, either singly or repeated.
1957    
1958    (iii) Repeats of multibyte characters are now handled correctly in UTF-8
1959          mode, for example: \x{100}{2,3}.
1960    
1961    (iv)  The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
1962          singly or repeated) now correctly test multibyte characters. However,
1963          PCRE doesn't recognize any characters with values greater than 255 as
1964          digits, spaces, or word characters. Such characters always match \D, \S,
1965          and \W, and never match \d, \s, or \w.
1966    
1967    (v)   Classes may now contain characters and character ranges with values
1968          greater than 255. For example: [ab\x{100}-\x{400}].
1969    
1970    (vi)  pcregrep now has a --utf-8 option (synonym -u) which makes it call
1971          PCRE in UTF-8 mode.
1972    
1973    52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
1974    PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
1975    retained for backwards compatibility. (Note that LASTLITERAL is also a byte
1976    value.)
1977    
1978    53. The single man page has become too large. I have therefore split it up into
1979    a number of separate man pages. These also give rise to individual HTML pages;
1980    these are now put in a separate directory, and there is an index.html page that
1981    lists them all. Some hyperlinking between the pages has been installed.
1982    
1983    54. Added convenience functions for handling named capturing parentheses.
1984    
1985    55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
1986    aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
1987    true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
1988    are faulted.
1989    
1990    56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
1991    calling configure. These values are used when compiling the dftables.c program
1992    which is run to generate the source of the default character tables. They
1993    default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
1994    you will need to set these values.
1995    
1996    57. Updated the building process for Windows DLL, as provided by Fred Cox.
1997    
1998    
1999    Version 3.9 02-Jan-02
2000    ---------------------
2001    
2002    1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2003    
2004    2. If --disable-static was given, the building process failed when trying to
2005    build pcretest and pcregrep. (For some reason it was using libtool to compile
2006    them, which is not right, as they aren't part of the library.)
2007    
2008    
2009    Version 3.8 18-Dec-01
2010    ---------------------
2011    
2012    1. The experimental UTF-8 code was completely screwed up. It was packing the
2013    bytes in the wrong order. How dumb can you get?
2014    
2015    
2016    Version 3.7 29-Oct-01
2017    ---------------------
2018    
2019    1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2020    This caused pcretest, when used on the test data, to segfault. Unfortunately,
2021    this didn't happen under Solaris 8, where I normally test things.
2022    
2023    2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2024    doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2025    isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2026    this fix an hour or so after the initial 3.7 release.)
2027    
2028    
2029    Version 3.6 23-Oct-01
2030    ---------------------
2031    
2032    1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2033    offsets passed as NULL with zero offset count.
2034    
2035    2. The config.guess and config.sub files had not been updated when I moved to
2036    the latest autoconf.
2037    
2038    
2039    Version 3.5 15-Aug-01
2040    ---------------------
2041    
2042    1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2043    had been forgotten.
2044    
2045    2. By using declared but undefined structures, we can avoid using "void"
2046    definitions in pcre.h while keeping the internal definitions of the structures
2047    private.
2048    
2049    3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2050    user point of view, this means that both static and shared libraries are built
2051    by default, but this can be individually controlled. More of the work of
2052    handling this static/shared cases is now inside libtool instead of PCRE's make
2053    file.
2054    
2055    4. The pcretest utility is now installed along with pcregrep because it is
2056    useful for users (to test regexs) and by doing this, it automatically gets
2057    relinked by libtool. The documentation has been turned into a man page, so
2058    there are now .1, .txt, and .html versions in /doc.
2059    
2060    5. Upgrades to pcregrep:
2061       (i)   Added long-form option names like gnu grep.
2062       (ii)  Added --help to list all options with an explanatory phrase.
2063       (iii) Added -r, --recursive to recurse into sub-directories.
2064       (iv)  Added -f, --file to read patterns from a file.
2065    
2066    6. pcre_exec() was referring to its "code" argument before testing that
2067    argument for NULL (and giving an error if it was NULL).
2068    
2069    7. Upgraded Makefile.in to allow for compiling in a different directory from
2070    the source directory.
2071    
2072    8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2073    options bits, the pointer it was passed was to an int instead of to an unsigned
2074    long int. This mattered only on 64-bit systems.
2075    
2076    9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2077    generated) instead of pcre.in, which it its source. Also made the same change
2078    in several of the .c files.
2079    
2080    10. A new release of gcc defines printf() as a macro, which broke pcretest
2081    because it had an ifdef in the middle of a string argument for printf(). Fixed
2082    by using separate calls to printf().
2083    
2084    11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2085    script, to force use of CR or LF instead of \n in the source. On non-Unix
2086    systems, the value can be set in config.h.
2087    
2088    12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2089    absolute limit. Changed the text of the error message to make this clear, and
2090    likewise updated the man page.
2091    
2092    13. The limit of 99 on the number of capturing subpatterns has been removed.
2093    The new limit is 65535, which I hope will not be a "real" limit.
2094    
2095    
2096    Version 3.4 22-Aug-00
2097    ---------------------
2098    
2099    1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2100    
2101    2. Diagnose condition (?(0) as an error instead of crashing on matching.
2102    
2103    
2104    Version 3.3 01-Aug-00
2105    ---------------------
2106    
2107    1. If an octal character was given, but the value was greater than \377, it
2108    was not getting masked to the least significant bits, as documented. This could
2109    lead to crashes in some systems.
2110    
2111    2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
2112    the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
2113    
2114    3. Added the functions pcre_free_substring() and pcre_free_substring_list().
2115    These just pass their arguments on to (pcre_free)(), but they are provided
2116    because some uses of PCRE bind it to non-C systems that can call its functions,
2117    but cannot call free() or pcre_free() directly.
2118    
2119    4. Add "make test" as a synonym for "make check". Corrected some comments in
2120    the Makefile.
2121    
2122    5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
2123    Makefile.
2124    
2125    6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
2126    command called pgrep for grepping around the active processes.
2127    
2128    7. Added the beginnings of support for UTF-8 character strings.
2129    
2130    8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
2131    RANLIB to ./ltconfig so that they are used by libtool. I think these are all
2132    the relevant ones. (AR is not passed because ./ltconfig does its own figuring
2133    out for the ar command.)
2134    
2135    
2136    Version 3.2 12-May-00
2137    ---------------------
2138    
2139    This is purely a bug fixing release.
2140    
2141    1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
2142    of ZA. This was just one example of several cases that could provoke this bug,
2143    which was introduced by change 9 of version 2.00. The code for breaking
2144    infinite loops after an iteration that matches an empty string was't working
2145    correctly.
2146    
2147    2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
2148    when matched against abbab (for example). After matching an empty string, it
2149    wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
2150    caused it to match further down the string than it should.
2151    
2152    3. The code contained an inclusion of sys/types.h. It isn't clear why this
2153    was there because it doesn't seem to be needed, and it causes trouble on some
2154    systems, as it is not a Standard C header. It has been removed.
2155    
2156    4. Made 4 silly changes to the source to avoid stupid compiler warnings that
2157    were reported on the Macintosh. The changes were from
2158    
2159      while ((c = *(++ptr)) != 0 && c != '\n');
2160    to
2161      while ((c = *(++ptr)) != 0 && c != '\n') ;
2162    
2163    Totally extraordinary, but if that's what it takes...
2164    
2165    5. PCRE is being used in one environment where neither memmove() nor bcopy() is
2166    available. Added HAVE_BCOPY and an autoconf test for it; if neither
2167    HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
2168    assumes the way PCRE uses memmove() (always moving upwards).
2169    
2170    6. PCRE is being used in one environment where strchr() is not available. There
2171    was only one use in pcre.c, and writing it out to avoid strchr() probably gives
2172    faster code anyway.
2173    
2174    
2175    Version 3.1 09-Feb-00
2176    ---------------------
2177    
2178    The only change in this release is the fixing of some bugs in Makefile.in for
2179    the "install" target:
2180    
2181    (1) It was failing to install pcreposix.h.
2182    
2183    (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
2184    
2185    
2186    Version 3.0 01-Feb-00
2187    ---------------------
2188    
2189    1. Add support for the /+ modifier to perltest (to output $` like it does in
2190    pcretest).
2191    
2192    2. Add support for the /g modifier to perltest.
2193    
2194    3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
2195    matches null strings.
2196    
2197    4. Fix perltest so that it doesn't do unwanted things when fed an empty
2198    pattern. Perl treats empty patterns specially - it reuses the most recent
2199    pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
2200    effect.
2201    
2202    5. The POSIX interface was broken in that it was just handing over the POSIX
2203    captured string vector to pcre_exec(), but (since release 2.00) PCRE has
2204    required a bigger vector, with some working space on the end. This means that
2205    the POSIX wrapper now has to get and free some memory, and copy the results.
2206    
2207    6. Added some simple autoconf support, placing the test data and the
2208    documentation in separate directories, re-organizing some of the
2209    information files, and making it build pcre-config (a GNU standard). Also added
2210    libtool support for building PCRE as a shared library, which is now the
2211    default.
2212    
2213    7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
2214    09 are not valid octal constants. Single digits will be used for minor values
2215    less than 10.
2216    
2217    8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
2218    existing programs that set these in the POSIX interface can use PCRE without
2219    modification.
2220    
2221    9. Added a new function, pcre_fullinfo() with an extensible interface. It can
2222    return all that pcre_info() returns, plus additional data. The pcre_info()
2223    function is retained for compatibility, but is considered to be obsolete.
2224    
2225    10. Added experimental recursion feature (?R) to handle one common case that
2226    Perl 5.6 will be able to do with (?p{...}).
2227    
2228    11. Added support for POSIX character classes like [:alpha:], which Perl is
2229    adopting.
2230    
2231    
2232    Version 2.08 31-Aug-99
2233    ----------------------
2234    
2235    1. When startoffset was not zero and the pattern began with ".*", PCRE was not
2236    trying to match at the startoffset position, but instead was moving forward to
2237    the next newline as if a previous match had failed.
2238    
2239    2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
2240    and could get into a loop if a null string was matched other than at the start
2241    of the subject.
2242    
2243    3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
2244    be distinguished at compile time, and for completeness also added PCRE_DATE.
2245    
2246    5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
2247    in GnuWin32 environments.
2248    
2249    
2250    Version 2.07 29-Jul-99
2251    ----------------------
2252    
2253    1. The documentation is now supplied in plain text form and HTML as well as in
2254    the form of man page sources.
2255    
2256    2. C++ compilers don't like assigning (void *) values to other pointer types.
2257    In particular this affects malloc(). Although there is no problem in Standard
2258    C, I've put in casts to keep C++ compilers happy.
2259    
2260    3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
2261    should be (const char *).
2262    
2263    4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
2264    be useful for non-Unix systems who don't want to bother with the POSIX stuff.
2265    However, I haven't made this a standard facility. The documentation doesn't
2266    mention it, and the Makefile doesn't support it.
2267    
2268    5. The Makefile now contains an "install" target, with editable destinations at
2269    the top of the file. The pcretest program is not installed.
2270    
2271    6. pgrep -V now gives the PCRE version number and date.
2272    
2273    7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
2274    causing the entire string to be ignored, instead of just the last character.
2275    
2276    8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
2277    non-matching string, it can take a very, very long time, even for strings of
2278    quite modest length, because of the nested recursion. PCRE now does better in
2279    some of these cases. It does this by remembering the last required literal
2280    character in the pattern, and pre-searching the subject to ensure it is present
2281    before running the real match. In other words, it applies a heuristic to detect
2282    some types of certain failure quickly, and in the above example, if presented
2283    with a string that has no trailing " it gives "no match" very quickly.
2284    
2285    9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
2286    other alternatives are tried instead.
2287    
2288    
2289    Version 2.06 09-Jun-99
2290    ----------------------
2291    
2292    1. Change pcretest's output for amount of store used to show just the code
2293    space, because the remainder (the data block) varies in size between 32-bit and
2294    64-bit systems.
2295    
2296    2. Added an extra argument to pcre_exec() to supply an offset in the subject to
2297    start matching at. This allows lookbehinds to work when searching for multiple
2298    occurrences in a string.
2299    
2300    3. Added additional options to pcretest for testing multiple occurrences:
2301    
2302       /+   outputs the rest of the string that follows a match
2303       /g   loops for multiple occurrences, using the new startoffset argument
2304       /G   loops for multiple occurrences by passing an incremented pointer
2305    
2306    4. PCRE wasn't doing the "first character" optimization for patterns starting
2307    with \b or \B, though it was doing it for other lookbehind assertions. That is,
2308    it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
2309    the letter 'x'. On long subject strings, this gives a significant speed-up.
2310    
2311    
2312    Version 2.05 21-Apr-99
2313    ----------------------
2314    
2315    1. Changed the type of magic_number from int to long int so that it works
2316    properly on 16-bit systems.
2317    
2318    2. Fixed a bug which caused patterns starting with .* not to work correctly
2319    when the subject string contained newline characters. PCRE was assuming
2320    anchoring for such patterns in all cases, which is not correct because .* will
2321    not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
2322    DOTALL is set at top level; otherwise it knows that patterns starting with .*
2323    must be retried after every newline in the subject.
2324    
2325    
2326    Version 2.04 18-Feb-99
2327    ----------------------
2328    
2329    1. For parenthesized subpatterns with repeats whose minimum was zero, the
2330    computation of the store needed to hold the pattern was incorrect (too large).
2331    If such patterns were nested a few deep, this could multiply and become a real
2332    problem.
2333    
2334    2. Added /M option to pcretest to show the memory requirement of a specific
2335    pattern. Made -m a synonym of -s (which does this globally) for compatibility.
2336    
2337    3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
2338    compiled in such a way that the backtracking after subsequent failure was
2339    pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
2340    ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
2341    
2342    
2343    Version 2.03 02-Feb-99
2344    ----------------------
2345    
2346    1. Fixed typo and small mistake in man page.
2347    
2348    2. Added 4th condition (GPL supersedes if conflict) and created separate
2349    LICENCE file containing the conditions.
2350    
2351    3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
2352    Perl, that is the internal \ allows the delimiter to be included in the
2353    pattern. Locked out the use of \ as a delimiter. If \ immediately follows
2354    the final delimiter, add \ to the end of the pattern (to test the error).
2355    
2356    4. Added the convenience functions for extracting substrings after a successful
2357    match. Updated pcretest to make it able to test these functions.
2358    
2359    
2360    Version 2.02 14-Jan-99
2361    ----------------------
2362    
2363    1. Initialized the working variables associated with each extraction so that
2364    their saving and restoring doesn't refer to uninitialized store.
2365    
2366    2. Put dummy code into study.c in order to trick the optimizer of the IBM C
2367    compiler for OS/2 into generating correct code. Apparently IBM isn't going to
2368    fix the problem.
2369    
2370    3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
2371    calls, and wasn't printing the correct value for compiling calls. Increased the
2372    default value of LOOPREPEAT, and the number of significant figures in the
2373    times.
2374    
2375    4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
2376    
2377    5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
2378    a building problem on Windows NT with a FAT file system.
2379    
2380    
2381    Version 2.01 21-Oct-98
2382    ----------------------
2383    
2384    1. Changed the API for pcre_compile() to allow for the provision of a pointer
2385    to character tables built by pcre_maketables() in the current locale. If NULL
2386    is passed, the default tables are used.
2387    
2388    
2389    Version 2.00 24-Sep-98
2390    ----------------------
2391    
2392    1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
2393    it any more.
2394    
2395    2. Allow quantification of (?>) groups, and make it work correctly.
2396    
2397    3. The first character computation wasn't working for (?>) groups.
2398    
2399    4. Correct the implementation of \Z (it is permitted to match on the \n at the
2400    end of the subject) and add 5.005's \z, which really does match only at the
2401    very end of the subject.
2402    
2403    5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
2404    
2405    6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
2406    DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
2407    localized options. All options to pcre_study() were also removed.
2408    
2409    7. Add other new features from 5.005:
2410    
2411       $(?<=           positive lookbehind
2412       $(?<!           negative lookbehind
2413       (?imsx-imsx)    added the unsetting capability
2414                       such a setting is global if at outer level; local otherwise
2415       (?imsx-imsx:)   non-capturing groups with option setting
2416       (?(cond)re|re)  conditional pattern matching
2417    
2418       A backreference to itself in a repeated group matches the previous
2419       captured string.
2420    
2421    8. General tidying up of studying (both automatic and via "study")
2422    consequential on the addition of new assertions.
2423    
2424    9. As in 5.005, unlimited repeated groups that could match an empty substring
2425    are no longer faulted at compile time. Instead, the loop is forcibly broken at
2426    runtime if any iteration does actually match an empty substring.
2427    
2428    10. Include the RunTest script in the distribution.
2429    
2430    11. Added tests from the Perl 5.005_02 distribution. This showed up a few
2431    discrepancies, some of which were old and were also with respect to 5.004. They
2432    have now been fixed.
2433    
2434    
2435    Version 1.09 28-Apr-98
2436    ----------------------
2437    
2438    1. A negated single character class followed by a quantifier with a minimum
2439    value of one (e.g.  [^x]{1,6}  ) was not compiled correctly. This could lead to
2440    program crashes, or just wrong answers. This did not apply to negated classes
2441    containing more than one character, or to minima other than one.
2442    
2443    
2444    Version 1.08 27-Mar-98
2445    ----------------------
2446    
2447    1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
2448    
2449    2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
2450    latter must appear before anything that relies on it in the pattern.
2451    
2452    
2453    Version 1.07 16-Feb-98
2454    ----------------------
2455    
2456    1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
2457    repeat of a potentially empty string).
2458    
2459    
2460    Version 1.06 23-Jan-98
2461    ----------------------
2462    
2463    1. Added Markus Oberhumer's little patches for C++.
2464    
2465    2. Literal strings longer than 255 characters were broken.
2466    
2467    
2468    Version 1.05 23-Dec-97
2469    ----------------------
2470    
2471    1. Negated character classes containing more than one character were failing if
2472    PCRE_CASELESS was set at run time.
2473    
2474    
2475    Version 1.04 19-Dec-97
2476    ----------------------
2477    
2478    1. Corrected the man page, where some "const" qualifiers had been omitted.
2479    
2480    2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
2481    input syntax.
2482    
2483    3. Fixed memory leak which occurred when a regex with back references was
2484    matched with an offsets vector that wasn't big enough. The temporary memory
2485    that is used in this case wasn't being freed if the match failed.
2486    
2487    4. Tidied pcretest to ensure it frees memory that it gets.
2488    
2489    5. Temporary memory was being obtained in the case where the passed offsets
2490    vector was exactly big enough.
2491    
2492    6. Corrected definition of offsetof() from change 5 below.
2493    
2494    7. I had screwed up change 6 below and broken the rules for the use of
2495    setjmp(). Now fixed.
2496    
2497    
2498    Version 1.03 18-Dec-97
2499    ----------------------
2500    
2501    1. A erroneous regex with a missing opening parenthesis was correctly
2502    diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
2503    on some systems.
2504    
2505    2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
2506    it was reported that one broken compiler failed on the former because "code" is
2507    also an independent variable.
2508    
2509    3. The erroneous regex a[]b caused an array overrun reference.
2510    
2511    4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
2512    fail on data ending with that character. (It was going on too far, and checking
2513    the next character, typically a binary zero.) This was specific to the
2514    optimized code for single-character negative classes.
2515    
2516    5. Added a contributed patch from the TIN world which does the following:
2517    
2518      + Add an undef for memmove, in case the the system defines a macro for it.
2519    
2520      + Add a definition of offsetof(), in case there isn't one. (I don't know
2521        the reason behind this - offsetof() is part of the ANSI standard - but
2522        it does no harm).
2523    
2524      + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
2525        most of the places where whitespace preceded '#'. I have given up and
2526        allowed the remaining 2 cases to be at the margin.
2527    
2528      + Rename some variables in pcre to eliminate shadowing. This seems very
2529        pedantic, but does no harm, of course.
2530    
2531    6. Moved the call to setjmp() into its own function, to get rid of warnings
2532    from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
2533    
2534    7. Constructs such as \d{8,} were compiling into the equivalent of
2535    \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
2536    outcome, but in this particular case used more store than had been allocated,
2537    which caused the bug to be discovered because it threw up an internal error.
2538    
2539    8. The debugging code in both pcre and pcretest for outputting the compiled
2540    form of a regex was going wrong in the case of back references followed by
2541    curly-bracketed repeats.
2542    
2543    
2544    Version 1.02 12-Dec-97
2545    ----------------------
2546    
2547    1. Typos in pcre.3 and comments in the source fixed.
2548    
2549    2. Applied a contributed patch to get rid of places where it used to remove
2550    'const' from variables, and fixed some signed/unsigned and uninitialized
2551    variable warnings.
2552    
2553    3. Added the "runtest" target to Makefile.
2554    
2555    4. Set default compiler flag to -O2 rather than just -O.
2556    
2557    
2558    Version 1.01 19-Nov-97
2559    ----------------------
2560    
2561    1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
2562    like /([ab]*)*/, that is, for classes with more than one character in them.
2563    
2564    2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
2565    as /((?>a*))*/ (a PCRE_EXTRA facility).
2566    
2567    
2568    Version 1.00 18-Nov-97
2569    ----------------------
2570    
2571    1. Added compile-time macros to support systems such as SunOS4 which don't have
2572    memmove() or strerror() but have other things that can be used instead.
2573    
2574    2. Arranged that "make clean" removes the executables.
2575    
2576    
2577  Version 0.99 27-Oct-97  Version 0.99 27-Oct-97
2578  ----------------------  ----------------------
2579    

Legend:
Removed from v.3  
changed lines
  Added in v.170

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12