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

Diff of /code/trunk/ChangeLog

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

revision 71 by nigel, Sat Feb 24 21:40:24 2007 UTC revision 488 by ph10, Mon Jan 11 15:29:42 2010 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4    Version 8.01 01-Jan-10
5    ----------------------
6    
7    1.  If a pattern contained a conditional subpattern with only one branch (in
8        particular, this includes all (DEFINE) patterns), a call to pcre_study()
9        computed the wrong minimum data length (which is of course zero for such
10        subpatterns).
11    
12    2.  For patterns such as (?i)a(?-i)b|c where an option setting at the start of
13        the pattern is reset in the first branch, pcre_compile() failed with
14        "internal error: code overflow at offset...". This happened only when
15        the reset was to the original external option setting. (An optimization
16        abstracts leading options settings into an external setting, which was the
17        cause of this.)
18    
19    3.  A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
20        of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
21        assertion pattern did not match (meaning that the assertion was true), it
22        was incorrectly treated as false if the SKIP had been reached during the
23        matching. This also applied to assertions used as conditions.
24    
25    4.  If an item that is not supported by pcre_dfa_exec() was encountered in an
26        assertion subpattern, including such a pattern used as a condition,
27        unpredictable results occurred, instead of the error return
28        PCRE_ERROR_DFA_UITEM.
29    
30    5.  The C++ GlobalReplace function was not working like Perl for the special
31        situation when an empty string is matched. It now does the fancy magic
32        stuff that is necessary.
33    
34    6.  In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
35        removed. (These were left over from very, very early versions of PCRE.)
36    
37    7.  Some cosmetic changes to the code to make life easier when compiling it
38        as part of something else:
39    
40        (a) Change DEBUG to PCRE_DEBUG.
41    
42        (b) In pcre_compile(), rename the member of the "branch_chain" structure
43            called "current" as "current_branch", to prevent a collision with the
44            Linux macro when compiled as a kernel module.
45    
46        (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
47            prevent a collision with the Linux macro when compiled as a kernel
48            module.
49    
50    8.  In pcre_compile() there are some checks for integer overflows that used to
51        cast potentially large values to (double). This has been changed to that
52        when building, a check for int64_t is made, and if it is found, it is used
53        instead, thus avoiding the use of floating point arithmetic. (There is no
54        other use of FP in PCRE.) If int64_t is not found, the fallback is to
55        double.
56    
57    9.  Added two casts to avoid signed/unsigned warnings from VS Studio Express
58        2005 (difference between two addresses compared to an unsigned value).
59    
60    10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
61        custom one, because of the following reported problem in Windows:
62    
63          - libbz2 uses the Pascal calling convention (WINAPI) for the functions
64            under Win32.
65          - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
66            therefore missing the function definition.
67          - The compiler thus generates a "C" signature for the test function.
68          - The linker fails to find the "C" function.
69          - PCRE fails to configure if asked to do so against libbz2.
70    
71    11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
72        messages were output:
73    
74          Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
75          rerunning libtoolize, to keep the correct libtool macros in-tree.
76          Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
77    
78        I have done both of these things.
79    
80    12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
81        most of the time, it *can* run out if it is given a pattern that contains a
82        runaway infinite recursion. I updated the discussion in the pcrestack man
83        page.
84    
85    13. Now that we have gone to the x.xx style of version numbers, the minor
86        version may start with zero. Using 08 or 09 is a bad idea because users
87        might check the value of PCRE_MINOR in their code, and 08 or 09 may be
88        interpreted as invalid octal numbers. I've updated the previous comment in
89        configure.ac, and also added a check that gives an error if 08 or 09 are
90        used.
91    
92    14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
93        causing partial matching to fail when the end of the subject matched \W
94        in a UTF-8 pattern where \W was quantified with a minimum of 3.
95    
96    15. There were some discrepancies between the declarations in pcre_internal.h
97        of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
98        their definitions. The declarations used "const uschar *" and the
99        definitions used USPTR. Even though USPTR is normally defined as "const
100        unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
101        reported that: "This difference in casting confuses some C++ compilers, for
102        example, SunCC recognizes above declarations as different functions and
103        generates broken code for hbpcre." I have changed the declarations to use
104        USPTR.
105    
106    16. GNU libtool is named differently on some systems. The autogen.sh script now
107        tries several variants such as glibtoolize (MacOSX) and libtoolize1x
108        (FreeBSD).
109    
110    17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
111        (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
112        comment: "Figure out how to create a longlong from a string: strtoll and
113        equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
114        instance, but it only takes 2 args instead of 3!"
115    
116    18. A subtle bug concerned with back references has been fixed by a change of
117        specification, with a corresponding code fix. A pattern such as
118        ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
119        refers, was giving matches when it shouldn't. For example, xa=xaaa would
120        match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
121        same bug. Such groups have to be quantified to be useful, or contained
122        inside another quantified group. (If there's no repetition, the reference
123        can never match.) The problem arises because, having left the group and
124        moved on to the rest of the pattern, a later failure that backtracks into
125        the group uses the captured value from the final iteration of the group
126        rather than the correct earlier one. I have fixed this in PCRE by forcing
127        any group that contains a reference to itself to be an atomic group; that
128        is, there cannot be any backtracking into it once it has completed. This is
129        similar to recursive and subroutine calls.
130    
131    
132    Version 8.00 19-Oct-09
133    ----------------------
134    
135    1.  The table for translating pcre_compile() error codes into POSIX error codes
136        was out-of-date, and there was no check on the pcre_compile() error code
137        being within the table. This could lead to an OK return being given in
138        error.
139    
140    2.  Changed the call to open a subject file in pcregrep from fopen(pathname,
141        "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
142        in a Windows environment.
143    
144    3.  The pcregrep --count option prints the count for each file even when it is
145        zero, as does GNU grep. However, pcregrep was also printing all files when
146        --files-with-matches was added. Now, when both options are given, it prints
147        counts only for those files that have at least one match. (GNU grep just
148        prints the file name in this circumstance, but including the count seems
149        more useful - otherwise, why use --count?) Also ensured that the
150        combination -clh just lists non-zero counts, with no names.
151    
152    4.  The long form of the pcregrep -F option was incorrectly implemented as
153        --fixed_strings instead of --fixed-strings. This is an incompatible change,
154        but it seems right to fix it, and I didn't think it was worth preserving
155        the old behaviour.
156    
157    5.  The command line items --regex=pattern and --regexp=pattern were not
158        recognized by pcregrep, which required --regex pattern or --regexp pattern
159        (with a space rather than an '='). The man page documented the '=' forms,
160        which are compatible with GNU grep; these now work.
161    
162    6.  No libpcreposix.pc file was created for pkg-config; there was just
163        libpcre.pc and libpcrecpp.pc. The omission has been rectified.
164    
165    7.  Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
166        when UCP support is not needed, by modifying the Python script that
167        generates it from Unicode data files. This should not matter if the module
168        is correctly used as a library, but I received one complaint about 50K of
169        unwanted data. My guess is that the person linked everything into his
170        program rather than using a library. Anyway, it does no harm.
171    
172    8.  A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
173        was a minimum greater than 1 for a wide character in a possessive
174        repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
175        which had an unlimited repeat of a nested, fixed maximum repeat of a wide
176        character. Chaos in the form of incorrect output or a compiling loop could
177        result.
178    
179    9.  The restrictions on what a pattern can contain when partial matching is
180        requested for pcre_exec() have been removed. All patterns can now be
181        partially matched by this function. In addition, if there are at least two
182        slots in the offset vector, the offset of the earliest inspected character
183        for the match and the offset of the end of the subject are set in them when
184        PCRE_ERROR_PARTIAL is returned.
185    
186    10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
187        synonymous with PCRE_PARTIAL, for backwards compatibility, and
188        PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
189        and may be more useful for multi-segment matching.
190    
191    11. Partial matching with pcre_exec() is now more intuitive. A partial match
192        used to be given if ever the end of the subject was reached; now it is
193        given only if matching could not proceed because another character was
194        needed. This makes a difference in some odd cases such as Z(*FAIL) with the
195        string "Z", which now yields "no match" instead of "partial match". In the
196        case of pcre_dfa_exec(), "no match" is given if every matching path for the
197        final character ended with (*FAIL).
198    
199    12. Restarting a match using pcre_dfa_exec() after a partial match did not work
200        if the pattern had a "must contain" character that was already found in the
201        earlier partial match, unless partial matching was again requested. For
202        example, with the pattern /dog.(body)?/, the "must contain" character is
203        "g". If the first part-match was for the string "dog", restarting with
204        "sbody" failed. This bug has been fixed.
205    
206    13. The string returned by pcre_dfa_exec() after a partial match has been
207        changed so that it starts at the first inspected character rather than the
208        first character of the match. This makes a difference only if the pattern
209        starts with a lookbehind assertion or \b or \B (\K is not supported by
210        pcre_dfa_exec()). It's an incompatible change, but it makes the two
211        matching functions compatible, and I think it's the right thing to do.
212    
213    14. Added a pcredemo man page, created automatically from the pcredemo.c file,
214        so that the demonstration program is easily available in environments where
215        PCRE has not been installed from source.
216    
217    15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
218        libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
219        library.
220    
221    16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
222        It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
223        is not the first non-POSIX option to be added. Clearly some people find
224        these options useful.
225    
226    17. If a caller to the POSIX matching function regexec() passes a non-zero
227        value for nmatch with a NULL value for pmatch, the value of
228        nmatch is forced to zero.
229    
230    18. RunGrepTest did not have a test for the availability of the -u option of
231        the diff command, as RunTest does. It now checks in the same way as
232        RunTest, and also checks for the -b option.
233    
234    19. If an odd number of negated classes containing just a single character
235        interposed, within parentheses, between a forward reference to a named
236        subpattern and the definition of the subpattern, compilation crashed with
237        an internal error, complaining that it could not find the referenced
238        subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
239        [The bug was that it was starting one character too far in when skipping
240        over the character class, thus treating the ] as data rather than
241        terminating the class. This meant it could skip too much.]
242    
243    20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
244        /g option in pcretest when the pattern contains \K, which makes it possible
245        to have an empty string match not at the start, even when the pattern is
246        anchored. Updated pcretest and pcredemo to use this option.
247    
248    21. If the maximum number of capturing subpatterns in a recursion was greater
249        than the maximum at the outer level, the higher number was returned, but
250        with unset values at the outer level. The correct (outer level) value is
251        now given.
252    
253    22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
254        PCRE did not set those parentheses (unlike Perl). I have now found a way to
255        make it do so. The string so far is captured, making this feature
256        compatible with Perl.
257    
258    23. The tests have been re-organized, adding tests 11 and 12, to make it
259        possible to check the Perl 5.10 features against Perl 5.10.
260    
261    24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
262        pattern matches a fixed length string. PCRE did not allow this; now it
263        does. Neither allows recursion.
264    
265    25. I finally figured out how to implement a request to provide the minimum
266        length of subject string that was needed in order to match a given pattern.
267        (It was back references and recursion that I had previously got hung up
268        on.) This code has now been added to pcre_study(); it finds a lower bound
269        to the length of subject needed. It is not necessarily the greatest lower
270        bound, but using it to avoid searching strings that are too short does give
271        some useful speed-ups. The value is available to calling programs via
272        pcre_fullinfo().
273    
274    26. While implementing 25, I discovered to my embarrassment that pcretest had
275        not been passing the result of pcre_study() to pcre_dfa_exec(), so the
276        study optimizations had never been tested with that matching function.
277        Oops. What is worse, even when it was passed study data, there was a bug in
278        pcre_dfa_exec() that meant it never actually used it. Double oops. There
279        were also very few tests of studied patterns with pcre_dfa_exec().
280    
281    27. If (?| is used to create subpatterns with duplicate numbers, they are now
282        allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
283        on the other side of the coin, they are no longer allowed to have different
284        names, because these cannot be distinguished in PCRE, and this has caused
285        confusion. (This is a difference from Perl.)
286    
287    28. When duplicate subpattern names are present (necessarily with different
288        numbers, as required by 27 above), and a test is made by name in a
289        conditional pattern, either for a subpattern having been matched, or for
290        recursion in such a pattern, all the associated numbered subpatterns are
291        tested, and the overall condition is true if the condition is true for any
292        one of them. This is the way Perl works, and is also more like the way
293        testing by number works.
294    
295    
296    Version 7.9 11-Apr-09
297    ---------------------
298    
299    1.  When building with support for bzlib/zlib (pcregrep) and/or readline
300        (pcretest), all targets were linked against these libraries. This included
301        libpcre, libpcreposix, and libpcrecpp, even though they do not use these
302        libraries. This caused unwanted dependencies to be created. This problem
303        has been fixed, and now only pcregrep is linked with bzlib/zlib and only
304        pcretest is linked with readline.
305    
306    2.  The "typedef int BOOL" in pcre_internal.h that was included inside the
307        "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
308        moved outside it again, because FALSE and TRUE are already defined in AIX,
309        but BOOL is not.
310    
311    3.  The pcre_config() function was treating the PCRE_MATCH_LIMIT and
312        PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
313    
314    4.  The pcregrep documentation said spaces were inserted as well as colons (or
315        hyphens) following file names and line numbers when outputting matching
316        lines. This is not true; no spaces are inserted. I have also clarified the
317        wording for the --colour (or --color) option.
318    
319    5.  In pcregrep, when --colour was used with -o, the list of matching strings
320        was not coloured; this is different to GNU grep, so I have changed it to be
321        the same.
322    
323    6.  When --colo(u)r was used in pcregrep, only the first matching substring in
324        each matching line was coloured. Now it goes on to look for further matches
325        of any of the test patterns, which is the same behaviour as GNU grep.
326    
327    7.  A pattern that could match an empty string could cause pcregrep to loop; it
328        doesn't make sense to accept an empty string match in pcregrep, so I have
329        locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
330        seems to be how GNU grep behaves.
331    
332    8.  The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
333        start or after a newline", because the conditional assertion was not being
334        correctly handled. The rule now is that both the assertion and what follows
335        in the first alternative must satisfy the test.
336    
337    9.  If auto-callout was enabled in a pattern with a conditional group whose
338        condition was an assertion, PCRE could crash during matching, both with
339        pcre_exec() and pcre_dfa_exec().
340    
341    10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
342        used for matching.
343    
344    11. Unicode property support in character classes was not working for
345        characters (bytes) greater than 127 when not in UTF-8 mode.
346    
347    12. Added the -M command line option to pcretest.
348    
349    14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
350    
351    15. Added the PCRE_NO_START_OPTIMIZE match-time option.
352    
353    16. Added comments and documentation about mis-use of no_arg in the C++
354        wrapper.
355    
356    17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
357        from Martin Jerabek that uses macro names for all relevant character and
358        string constants.
359    
360    18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
361        SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
362        SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
363        these, but not everybody uses configure.
364    
365    19. A conditional group that had only one branch was not being correctly
366        recognized as an item that could match an empty string. This meant that an
367        enclosing group might also not be so recognized, causing infinite looping
368        (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
369        with the subject "ab", where knowledge that the repeated group can match
370        nothing is needed in order to break the loop.
371    
372    20. If a pattern that was compiled with callouts was matched using pcre_dfa_
373        exec(), but without supplying a callout function, matching went wrong.
374    
375    21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
376        leak if the size of the offset vector was greater than 30. When the vector
377        is smaller, the saved offsets during recursion go onto a local stack
378        vector, but for larger vectors malloc() is used. It was failing to free
379        when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
380        error, in fact).
381    
382    22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
383        heapframe that is used only when UTF-8 support is enabled. This caused no
384        problem, but was untidy.
385    
386    23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
387        CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
388        included within another project.
389    
390    24. Steven Van Ingelgem's patches to add more options to the CMake support,
391        slightly modified by me:
392    
393          (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
394              not building pcregrep.
395    
396          (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
397              if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
398    
399    25. Forward references, both numeric and by name, in patterns that made use of
400        duplicate group numbers, could behave incorrectly or give incorrect errors,
401        because when scanning forward to find the reference group, PCRE was not
402        taking into account the duplicate group numbers. A pattern such as
403        ^X(?3)(a)(?|(b)|(q))(Y) is an example.
404    
405    26. Changed a few more instances of "const unsigned char *" to USPTR, making
406        the feature of a custom pointer more persuasive (as requested by a user).
407    
408    27. Wrapped the definitions of fileno and isatty for Windows, which appear in
409        pcretest.c, inside #ifndefs, because it seems they are sometimes already
410        pre-defined.
411    
412    28. Added support for (*UTF8) at the start of a pattern.
413    
414    29. Arrange for flags added by the "release type" setting in CMake to be shown
415        in the configuration summary.
416    
417    
418    Version 7.8 05-Sep-08
419    ---------------------
420    
421    1.  Replaced UCP searching code with optimized version as implemented for Ad
422        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
423        stage table and inline lookup instead of a function, giving speed ups of 2
424        to 5 times on some simple patterns that I tested. Permission was given to
425        distribute the MultiStage2.py script that generates the tables (it's not in
426        the tarball, but is in the Subversion repository).
427    
428    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
429        scripts.
430    
431    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
432        a group with a zero qualifier. The result of the study could be incorrect,
433        or the function might crash, depending on the pattern.
434    
435    4.  Caseless matching was not working for non-ASCII characters in back
436        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
437        It now works when Unicode Property Support is available.
438    
439    5.  In pcretest, an escape such as \x{de} in the data was always generating
440        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
441        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
442        truncation.
443    
444    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
445    
446    7.  Added two (int) casts to pcregrep when printing the difference of two
447        pointers, in case they are 64-bit values.
448    
449    8.  Added comments about Mac OS X stack usage to the pcrestack man page and to
450        test 2 if it fails.
451    
452    9.  Added PCRE_CALL_CONVENTION just before the names of all exported functions,
453        and a #define of that name to empty if it is not externally set. This is to
454        allow users of MSVC to set it if necessary.
455    
456    10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
457        the convenience functions in the pcre_get.c source file.
458    
459    11. An option change at the start of a pattern that had top-level alternatives
460        could cause overwriting and/or a crash. This command provoked a crash in
461        some environments:
462    
463          printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
464    
465        This potential security problem was recorded as CVE-2008-2371.
466    
467    12. For a pattern where the match had to start at the beginning or immediately
468        after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
469        pcre_dfa_exec() could read past the end of the passed subject if there was
470        no match. To help with detecting such bugs (e.g. with valgrind), I modified
471        pcretest so that it places the subject at the end of its malloc-ed buffer.
472    
473    13. The change to pcretest in 12 above threw up a couple more cases when pcre_
474        exec() might read past the end of the data buffer in UTF-8 mode.
475    
476    14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
477        the data contained the byte 0x85 as part of a UTF-8 character within its
478        first line. This applied both to normal and DFA matching.
479    
480    15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
481        /^[^d]*?$/8 failed to match "abc".
482    
483    16. Added a missing copyright notice to pcrecpp_internal.h.
484    
485    17. Make it more clear in the documentation that values returned from
486        pcre_exec() in ovector are byte offsets, not character counts.
487    
488    18. Tidied a few places to stop certain compilers from issuing warnings.
489    
490    19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
491        supplied by Stefan Weber. I made a further small update for 7.8 because
492        there is a change of source arrangements: the pcre_searchfuncs.c module is
493        replaced by pcre_ucd.c.
494    
495    
496    Version 7.7 07-May-08
497    ---------------------
498    
499    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
500        a string to a long long, pretend we don't even have a long long." This is
501        done by checking for the strtoq, strtoll, and _strtoi64 functions.
502    
503    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
504        pre-7.6 versions, which defined a global no_arg variable instead of putting
505        it in the RE class. (See also #8 below.)
506    
507    3.  Remove a line of dead code, identified by coverity and reported by Nuno
508        Lopes.
509    
510    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
511    
512        (1) The include/exclude patterns were being applied to the whole pathnames
513            of files, instead of just to the final components.
514    
515        (2) If there was more than one level of directory, the subdirectories were
516            skipped unless they satisfied the include/exclude conditions. This is
517            inconsistent with GNU grep (and could even be seen as contrary to the
518            pcregrep specification - which I improved to make it absolutely clear).
519            The action now is always to scan all levels of directory, and just
520            apply the include/exclude patterns to regular files.
521    
522    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
523        --exclude_dir in the tests to avoid scanning .svn directories.
524    
525    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
526        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
527        doesn't support NULs in patterns.
528    
529    7.  Added some missing "const"s to declarations of static tables in
530        pcre_compile.c and pcre_dfa_exec.c.
531    
532    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
533        caused by fix #2  above. (Subsequently also a second patch to fix the
534        first patch. And a third patch - this was a messy problem.)
535    
536    9.  Applied Craig's patch to remove the use of push_back().
537    
538    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
539        matching function regexec().
540    
541    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
542        which, however, unlike Perl's \g{...}, are subroutine calls, not back
543        references. PCRE supports relative numbers with this syntax (I don't think
544        Oniguruma does).
545    
546    12. Previously, a group with a zero repeat such as (...){0} was completely
547        omitted from the compiled regex. However, this means that if the group
548        was called as a subroutine from elsewhere in the pattern, things went wrong
549        (an internal error was given). Such groups are now left in the compiled
550        pattern, with a new opcode that causes them to be skipped at execution
551        time.
552    
553    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
554        to the way PCRE behaves:
555    
556        (a) A lone ] character is dis-allowed (Perl treats it as data).
557    
558        (b) A back reference to an unmatched subpattern matches an empty string
559            (Perl fails the current match path).
560    
561        (c) A data ] in a character class must be notated as \] because if the
562            first data character in a class is ], it defines an empty class. (In
563            Perl it is not possible to have an empty class.) The empty class []
564            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
565            The negative empty class [^] matches any one character, independently
566            of the DOTALL setting.
567    
568    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
569        non-existent subpattern following a character class starting with ']' and
570        containing () gave an internal compiling error instead of "reference to
571        non-existent subpattern". Fortunately, when the pattern did exist, the
572        compiled code was correct. (When scanning forwards to check for the
573        existencd of the subpattern, it was treating the data ']' as terminating
574        the class, so got the count wrong. When actually compiling, the reference
575        was subsequently set up correctly.)
576    
577    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
578        it was being rejected as not supported by pcre_dfa_exec(), even though
579        other assertions are supported. I have made pcre_dfa_exec() support
580        (*FAIL).
581    
582    16. The implementation of 13c above involved the invention of a new opcode,
583        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
584        cannot be changed at match time, I realized I could make a small
585        improvement to matching performance by compiling OP_ALLANY instead of
586        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
587        on the OP_ANY path.
588    
589    17. Compiling pcretest on Windows with readline support failed without the
590        following two fixes: (1) Make the unistd.h include conditional on
591        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
592    
593    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
594        ncurses library to be included for pcretest when ReadLine support is
595        requested, but also to allow for it to be overridden. This patch came from
596        Daniel Bergström.
597    
598    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
599        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
600        any errors with the current Unicode tables. Thanks to Peter Kankowski for
601        spotting this.
602    
603    
604    Version 7.6 28-Jan-08
605    ---------------------
606    
607    1.  A character class containing a very large number of characters with
608        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
609        overflow.
610    
611    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
612        HAVE_LONG_LONG is not defined.
613    
614    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
615        bring them up to date and include new features. This patch includes:
616    
617        - Fixed PH's badly added libz and libbz2 support.
618        - Fixed a problem with static linking.
619        - Added pcredemo. [But later removed - see 7 below.]
620        - Fixed dftables problem and added an option.
621        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
622            HAVE_LONG_LONG.
623        - Added readline support for pcretest.
624        - Added an listing of the option settings after cmake has run.
625    
626    4.  A user submitted a patch to Makefile that makes it easy to create
627        "pcre.dll" under mingw when using Configure/Make. I added stuff to
628        Makefile.am that cause it to include this special target, without
629        affecting anything else. Note that the same mingw target plus all
630        the other distribution libraries and programs are now supported
631        when configuring with CMake (see 6 below) instead of with
632        Configure/Make.
633    
634    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
635        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
636        exported in the Windows port". It has not yet been confirmed that the patch
637        solves the problem, but it does no harm.
638    
639    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
640        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
641        with CMake, and also correct the comment about stack recursion.
642    
643    7.  Remove the automatic building of pcredemo from the ./configure system and
644        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
645        of a program that users should build themselves after PCRE is installed, so
646        building it automatically is not really right. What is more, it gave
647        trouble in some build environments.
648    
649    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
650    
651    
652    Version 7.5 10-Jan-08
653    ---------------------
654    
655    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
656        values in parens when parsing an RE using the C++ wrapper."
657    
658    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
659        Characters greater than 255 were excluded from the class instead of being
660        included.
661    
662    3.  The same bug as (2) above applied to negated POSIX classes such as
663        [:^space:].
664    
665    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
666        defined or documented. It seems to have been a typo for PCRE_STATIC, so
667        I have changed it.
668    
669    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
670        first named subpattern) and a construct such as (?&a) would reference the
671        first named subpattern whose name started with "a" (in other words, the
672        length check was missing). Both these problems are fixed. "Subpattern name
673        expected" is now given for (?&) (a zero-length name), and this patch also
674        makes it give the same error for \k'' (previously it complained that that
675        was a reference to a non-existent subpattern).
676    
677    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
678        this is right because (?- can be followed by option settings as well as by
679        digits. I have, however, made the messages clearer.
680    
681    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
682        than the number used in the conditional) now cause a compile-time error.
683        This is actually not compatible with Perl, which accepts such patterns, but
684        treats the conditional as always being FALSE (as PCRE used to), but it
685        seems to me that giving a diagnostic is better.
686    
687    8.  Change "alphameric" to the more common word "alphanumeric" in comments
688        and messages.
689    
690    9.  Fix two occurrences of "backslash" in comments that should have been
691        "backspace".
692    
693    10. Remove two redundant lines of code that can never be obeyed (their function
694        was moved elsewhere).
695    
696    11. The program that makes PCRE's Unicode character property table had a bug
697        which caused it to generate incorrect table entries for sequences of
698        characters that have the same character type, but are in different scripts.
699        It amalgamated them into a single range, with the script of the first of
700        them. In other words, some characters were in the wrong script. There were
701        thirteen such cases, affecting characters in the following ranges:
702    
703          U+002b0 - U+002c1
704          U+0060c - U+0060d
705          U+0061e - U+00612
706          U+0064b - U+0065e
707          U+0074d - U+0076d
708          U+01800 - U+01805
709          U+01d00 - U+01d77
710          U+01d9b - U+01dbf
711          U+0200b - U+0200f
712          U+030fc - U+030fe
713          U+03260 - U+0327f
714          U+0fb46 - U+0fbb1
715          U+10450 - U+1049d
716    
717    12. The -o option (show only the matching part of a line) for pcregrep was not
718        compatible with GNU grep in that, if there was more than one match in a
719        line, it showed only the first of them. It now behaves in the same way as
720        GNU grep.
721    
722    13. If the -o and -v options were combined for pcregrep, it printed a blank
723        line for every non-matching line. GNU grep prints nothing, and pcregrep now
724        does the same. The return code can be used to tell if there were any
725        non-matching lines.
726    
727    14. Added --file-offsets and --line-offsets to pcregrep.
728    
729    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
730        infinitely looping recursion. The bug was that positive lookaheads were not
731        being skipped when checking for a possible empty match (negative lookaheads
732        and both kinds of lookbehind were skipped).
733    
734    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
735        inclusion of <windows.h> to before rather than after the definition of
736        INVALID_FILE_ATTRIBUTES (patch from David Byron).
737    
738    17. Specifying a possessive quantifier with a specific limit for a Unicode
739        character property caused pcre_compile() to compile bad code, which led at
740        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
741        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
742        caused the error; without that there was no problem.
743    
744    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
745    
746    19. Added --enable-pcretest-libreadline.
747    
748    20. In pcrecpp.cc, the variable 'count' was incremented twice in
749        RE::GlobalReplace(). As a result, the number of replacements returned was
750        double what it should be. I removed one of the increments, but Craig sent a
751        later patch that removed the other one (the right fix) and added unit tests
752        that check the return values (which was not done before).
753    
754    21. Several CMake things:
755    
756        (1) Arranged that, when cmake is used on Unix, the libraries end up with
757            the names libpcre and libpcreposix, not just pcre and pcreposix.
758    
759        (2) The above change means that pcretest and pcregrep are now correctly
760            linked with the newly-built libraries, not previously installed ones.
761    
762        (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
763    
764    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
765        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
766        UTF-8 newline character). The key issue is that the pattern starts .*;
767        this means that the match must be either at the beginning, or after a
768        newline. The bug was in the code for advancing after a failed match and
769        checking that the new position followed a newline. It was not taking
770        account of UTF-8 characters correctly.
771    
772    23. PCRE was behaving differently from Perl in the way it recognized POSIX
773        character classes. PCRE was not treating the sequence [:...:] as a
774        character class unless the ... were all letters. Perl, however, seems to
775        allow any characters between [: and :], though of course it rejects as
776        unknown any "names" that contain non-letters, because all the known class
777        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
778        for example, whereas PCRE did not - it did not recognize a POSIX character
779        class. This seemed a bit dangerous, so the code has been changed to be
780        closer to Perl. The behaviour is not identical to Perl, because PCRE will
781        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
782        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
783        Perl does, and where it didn't before.
784    
785    24. Rewrite so as to remove the single use of %n from pcregrep because in some
786        Windows environments %n is disabled by default.
787    
788    
789    Version 7.4 21-Sep-07
790    ---------------------
791    
792    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
793        means that a class such as [\s] counted as "explicit reference to CR or
794        LF". That isn't really right - the whole point of the change was to try to
795        help when there was an actual mention of one of the two characters. So now
796        the change happens only if \r or \n (or a literal CR or LF) character is
797        encountered.
798    
799    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
800        of both had grown to the point where there were only 3 bits left.
801        Fortunately, there was spare space in the data structure, and so I have
802        moved the internal flags into a new 16-bit field to free up more option
803        bits.
804    
805    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
806        but did not set the internal JCHANGED flag - either of these is enough to
807        control the way the "get" function works - but the PCRE_INFO_JCHANGED
808        facility is supposed to tell if (?J) was ever used, so now (?J) at the
809        start sets both bits.
810    
811    4.  Added options (at build time, compile time, exec time) to change \R from
812        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
813    
814    5.  doc/pcresyntax.html was missing from the distribution.
815    
816    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
817        compatibility, even though it is no longer used.
818    
819    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
820        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
821        windows.h file is present (where different names are used). [This was
822        reversed later after testing - see 16 below.]
823    
824    8.  Changed all #include <config.h> to #include "config.h". There were also
825        some further <pcre.h> cases that I changed to "pcre.h".
826    
827    9.  When pcregrep was used with the --colour option, it missed the line ending
828        sequence off the lines that it output.
829    
830    10. It was pointed out to me that arrays of string pointers cause lots of
831        relocations when a shared library is dynamically loaded. A technique of
832        using a single long string with a table of offsets can drastically reduce
833        these. I have refactored PCRE in four places to do this. The result is
834        dramatic:
835    
836          Originally:                          290
837          After changing UCP table:            187
838          After changing error message table:   43
839          After changing table of "verbs"       36
840          After changing table of Posix names   22
841    
842        Thanks to the folks working on Gregex for glib for this insight.
843    
844    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
845        unicode-properties was also set.
846    
847    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
848    
849    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
850        checked only for CRLF.
851    
852    14. Added casts to pcretest.c to avoid compiler warnings.
853    
854    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
855    
856    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
857        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
858        entirely. This removes changes made in 7 above.
859    
860    17. The CMake files have been updated, and there is now more information about
861        building with CMake in the NON-UNIX-USE document.
862    
863    
864    Version 7.3 28-Aug-07
865    ---------------------
866    
867     1. In the rejigging of the build system that eventually resulted in 7.1, the
868        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
869        brackets there is not right, since it causes compilers to look for an
870        installed pcre.h, not the version that is in the source that is being
871        compiled (which of course may be different). I have changed it back to:
872    
873          #include "pcre.h"
874    
875        I have a vague recollection that the change was concerned with compiling in
876        different directories, but in the new build system, that is taken care of
877        by the VPATH setting the Makefile.
878    
879     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
880        when the subject happened to end in the byte 0x85 (e.g. if the last
881        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
882        characters but of course it shouldn't be taken as a newline when it is part
883        of another character. The bug was that, for an unlimited repeat of . in
884        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
885        characters when looking for a newline.
886    
887     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
888    
889     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
890        in debug output.
891    
892     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
893        long printing in the pcrecpp unittest when running under MinGW.
894    
895     6. ESC_K was left out of the EBCDIC table.
896    
897     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
898        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
899        limit also applies to "virtual nesting" when a pattern is recursive, and in
900        this case 1000 isn't so big. I have been able to remove this limit at the
901        expense of backing off one optimization in certain circumstances. Normally,
902        when pcre_exec() would call its internal match() function recursively and
903        immediately return the result unconditionally, it uses a "tail recursion"
904        feature to save stack. However, when a subpattern that can match an empty
905        string has an unlimited repetition quantifier, it no longer makes this
906        optimization. That gives it a stack frame in which to save the data for
907        checking that an empty string has been matched. Previously this was taken
908        from the 1000-entry workspace that had been reserved. So now there is no
909        explicit limit, but more stack is used.
910    
911     8. Applied Daniel's patches to solve problems with the import/export magic
912        syntax that is required for Windows, and which was going wrong for the
913        pcreposix and pcrecpp parts of the library. These were overlooked when this
914        problem was solved for the main library.
915    
916     9. There were some crude static tests to avoid integer overflow when computing
917        the size of patterns that contain repeated groups with explicit upper
918        limits. As the maximum quantifier is 65535, the maximum group length was
919        set at 30,000 so that the product of these two numbers did not overflow a
920        32-bit integer. However, it turns out that people want to use groups that
921        are longer than 30,000 bytes (though not repeat them that many times).
922        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
923        made it possible to implement the integer overflow checks in a much more
924        dynamic way, which I have now done. The artificial limitation on group
925        length has been removed - we now have only the limit on the total length of
926        the compiled pattern, which depends on the LINK_SIZE setting.
927    
928    10. Fixed a bug in the documentation for get/copy named substring when
929        duplicate names are permitted. If none of the named substrings are set, the
930        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
931        empty string.
932    
933    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
934        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
935        because the ] is interpreted as the first data character and the
936        terminating ] is not found. PCRE has been made compatible with Perl in this
937        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
938        cause memory overwriting.
939    
940    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
941        string has been matched (to stop an infinite loop). It was not recognizing
942        a conditional subpattern that could match an empty string if that
943        subpattern was within another subpattern. For example, it looped when
944        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
945        condition was not nested. This bug has been fixed.
946    
947    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
948        past the start of the subject in the presence of bytes with the top bit
949        set, for example "\x8aBCD".
950    
951    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
952        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
953    
954    14. Optimized (?!) to (*FAIL).
955    
956    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
957        This restricts code points to be within the range 0 to 0x10FFFF, excluding
958        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
959        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
960        does: it's just the validity check that is more restrictive.
961    
962    16. Inserted checks for integer overflows during escape sequence (backslash)
963        processing, and also fixed erroneous offset values for syntax errors during
964        backslash processing.
965    
966    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
967        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
968    
969    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
970        caused an overrun.
971    
972    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
973        something other than just ASCII characters) inside a group that had an
974        unlimited repeat caused a loop at compile time (while checking to see
975        whether the group could match an empty string).
976    
977    20. Debugging a pattern containing \p or \P could cause a crash. For example,
978        [\P{Any}] did so. (Error in the code for printing property names.)
979    
980    21. An orphan \E inside a character class could cause a crash.
981    
982    22. A repeated capturing bracket such as (A)? could cause a wild memory
983        reference during compilation.
984    
985    23. There are several functions in pcre_compile() that scan along a compiled
986        expression for various reasons (e.g. to see if it's fixed length for look
987        behind). There were bugs in these functions when a repeated \p or \P was
988        present in the pattern. These operators have additional parameters compared
989        with \d, etc, and these were not being taken into account when moving along
990        the compiled data. Specifically:
991    
992        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
993            length.
994    
995        (b) An item such as \pL+ within a repeated group could cause crashes or
996            loops.
997    
998        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
999            "reference to non-existent subpattern" error.
1000    
1001        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
1002    
1003    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
1004        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
1005    
1006    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
1007    
1008    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
1009        character were causing crashes (broken optimization).
1010    
1011    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
1012        \p or \P) caused a compile-time loop.
1013    
1014    28. More problems have arisen in unanchored patterns when CRLF is a valid line
1015        break. For example, the unstudied pattern [\r\n]A does not match the string
1016        "\r\nA" because change 7.0/46 below moves the current point on by two
1017        characters after failing to match at the start. However, the pattern \nA
1018        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
1019        the same is true. There doesn't seem any very clean way out of this, but
1020        what I have chosen to do makes the common cases work: PCRE now takes note
1021        of whether there can be an explicit match for \r or \n anywhere in the
1022        pattern, and if so, 7.0/46 no longer applies. As part of this change,
1023        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
1024        pattern has explicit CR or LF references.
1025    
1026    29. Added (*CR) etc for changing newline setting at start of pattern.
1027    
1028    
1029    Version 7.2 19-Jun-07
1030    ---------------------
1031    
1032     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
1033        which is apparently normally available under Windows.
1034    
1035     2. Re-jig the pcregrep tests with different newline settings in an attempt
1036        to make them independent of the local environment's newline setting.
1037    
1038     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
1039    
1040     4. Some of the "internals" tests were previously cut out when the link size
1041        was not 2, because the output contained actual offsets. The recent new
1042        "Z" feature of pcretest means that these can be cut out, making the tests
1043        usable with all link sizes.
1044    
1045     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
1046        stack recursion. This gives a massive performance boost under BSD, but just
1047        a small improvement under Linux. However, it saves one field in the frame
1048        in all cases.
1049    
1050     6. Added more features from the forthcoming Perl 5.10:
1051    
1052        (a) (?-n) (where n is a string of digits) is a relative subroutine or
1053            recursion call. It refers to the nth most recently opened parentheses.
1054    
1055        (b) (?+n) is also a relative subroutine call; it refers to the nth next
1056            to be opened parentheses.
1057    
1058        (c) Conditions that refer to capturing parentheses can be specified
1059            relatively, for example, (?(-2)... or (?(+3)...
1060    
1061        (d) \K resets the start of the current match so that everything before
1062            is not part of it.
1063    
1064        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
1065    
1066        (f) \g{name} is another synonym - part of Perl 5.10's unification of
1067            reference syntax.
1068    
1069        (g) (?| introduces a group in which the numbering of parentheses in each
1070            alternative starts with the same number.
1071    
1072        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
1073    
1074     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
1075        PCRE_INFO_JCHANGED.
1076    
1077     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
1078        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
1079        for detecting groups that can match an empty string.
1080    
1081     9. A pattern with a very large number of alternatives (more than several
1082        hundred) was running out of internal workspace during the pre-compile
1083        phase, where pcre_compile() figures out how much memory will be needed. A
1084        bit of new cunning has reduced the workspace needed for groups with
1085        alternatives. The 1000-alternative test pattern now uses 12 bytes of
1086        workspace instead of running out of the 4096 that are available.
1087    
1088    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
1089    
1090    11. Applied patch from Google to remove an optimization that didn't quite work.
1091        The report of the bug said:
1092    
1093          pcrecpp::RE("a*").FullMatch("aaa") matches, while
1094          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
1095          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
1096    
1097    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
1098        it matched the wrong number of bytes.
1099    
1100    
1101    Version 7.1 24-Apr-07
1102    ---------------------
1103    
1104     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
1105        that is more "standard", making use of automake and other Autotools. There
1106        is some re-arrangement of the files and adjustment of comments consequent
1107        on this.
1108    
1109     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
1110        for recursive directory scanning broke on some systems because the files
1111        are not scanned in any specific order and on different systems the order
1112        was different. A call to "sort" has been inserted into RunGrepTest for the
1113        approprate test as a short-term fix. In the longer term there may be an
1114        alternative.
1115    
1116     3. I had an email from Eric Raymond about problems translating some of PCRE's
1117        man pages to HTML (despite the fact that I distribute HTML pages, some
1118        people do their own conversions for various reasons). The problems
1119        concerned the use of low-level troff macros .br and .in. I have therefore
1120        removed all such uses from the man pages (some were redundant, some could
1121        be replaced by .nf/.fi pairs). The 132html script that I use to generate
1122        HTML has been updated to handle .nf/.fi and to complain if it encounters
1123        .br or .in.
1124    
1125     4. Updated comments in configure.ac that get placed in config.h.in and also
1126        arranged for config.h to be included in the distribution, with the name
1127        config.h.generic, for the benefit of those who have to compile without
1128        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
1129    
1130     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
1131        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
1132        makevp.bat for the current PCRE, using the additional files
1133        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
1134    
1135     6. A Windows user reported a minor discrepancy with test 2, which turned out
1136        to be caused by a trailing space on an input line that had got lost in his
1137        copy. The trailing space was an accident, so I've just removed it.
1138    
1139     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
1140        that is needed.
1141    
1142     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
1143        as "const" (a) because they are and (b) because it helps the PHP
1144        maintainers who have recently made a script to detect big data structures
1145        in the php code that should be moved to the .rodata section. I remembered
1146        to update Builducptable as well, so it won't revert if ucptable.h is ever
1147        re-created.
1148    
1149     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
1150        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
1151        order to be able to cut out the UTF-8 tables in the latter when UTF-8
1152        support is not required. This saves 1.5-2K of code, which is important in
1153        some applications.
1154    
1155        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
1156        so as not to refer to the tables, even though these functions will never be
1157        called when UTF-8 support is disabled. Otherwise there are problems with a
1158        shared library.
1159    
1160    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
1161    
1162        (a) It was defining its arguments as char * instead of void *.
1163    
1164        (b) It was assuming that all moves were upwards in memory; this was true
1165            a long time ago when I wrote it, but is no longer the case.
1166    
1167        The emulated memove() is provided for those environments that have neither
1168        memmove() nor bcopy(). I didn't think anyone used it these days, but that
1169        is clearly not the case, as these two bugs were recently reported.
1170    
1171    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
1172        and Detrail to create the HTML documentation, the .txt form of the man
1173        pages, and it removes trailing spaces from listed files. It also creates
1174        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
1175        case, it wraps all the #defines with #ifndefs. This script should be run
1176        before "make dist".
1177    
1178    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
1179        with Unicode property support.
1180    
1181        (a) For a maximizing quantifier, if the two different cases of the
1182            character were of different lengths in their UTF-8 codings (there are
1183            some cases like this - I found 11), and the matching function had to
1184            back up over a mixture of the two cases, it incorrectly assumed they
1185            were both the same length.
1186    
1187        (b) When PCRE was configured to use the heap rather than the stack for
1188            recursion during matching, it was not correctly preserving the data for
1189            the other case of a UTF-8 character when checking ahead for a match
1190            while processing a minimizing repeat. If the check also involved
1191            matching a wide character, but failed, corruption could cause an
1192            erroneous result when trying to check for a repeat of the original
1193            character.
1194    
1195    13. Some tidying changes to the testing mechanism:
1196    
1197        (a) The RunTest script now detects the internal link size and whether there
1198            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
1199            values substituted by "configure". (The RunGrepTest script already did
1200            this for UTF-8.) The configure.ac script no longer substitutes the
1201            relevant variables.
1202    
1203        (b) The debugging options /B and /D in pcretest show the compiled bytecode
1204            with length and offset values. This means that the output is different
1205            for different internal link sizes. Test 2 is skipped for link sizes
1206            other than 2 because of this, bypassing the problem. Unfortunately,
1207            there was also a test in test 3 (the locale tests) that used /B and
1208            failed for link sizes other than 2. Rather than cut the whole test out,
1209            I have added a new /Z option to pcretest that replaces the length and
1210            offset values with spaces. This is now used to make test 3 independent
1211            of link size. (Test 2 will be tidied up later.)
1212    
1213    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
1214        segmentation fault instead of returning the appropriate error message.
1215    
1216    15. In multiline mode when the newline sequence was set to "any", the pattern
1217        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
1218        This doesn't seem right; it now treats the CRLF combination as the line
1219        ending, and so does not match in that case. It's only a pattern such as ^$
1220        that would hit this one: something like ^ABC$ would have failed after \r
1221        and then tried again after \r\n.
1222    
1223    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
1224        in an attempt to make files that differ only in their line terminators
1225        compare equal. This works on Linux.
1226    
1227    17. Under certain error circumstances pcregrep might try to free random memory
1228        as it exited. This is now fixed, thanks to valgrind.
1229    
1230    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
1231        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
1232        was because its rules for how to advance for /g after matching an empty
1233        string at the end of a line did not allow for this case. They now check for
1234        it specially.
1235    
1236    20. pcretest is supposed to handle patterns and data of any length, by
1237        extending its buffers when necessary. It was getting this wrong when the
1238        buffer for a data line had to be extended.
1239    
1240    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
1241        CRLF as a newline sequence.
1242    
1243    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
1244        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
1245        I have nevertheless tidied it up.
1246    
1247    23. Added some casts to kill warnings from HP-UX ia64 compiler.
1248    
1249    24. Added a man page for pcre-config.
1250    
1251    
1252    Version 7.0 19-Dec-06
1253    ---------------------
1254    
1255     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
1256        moving to gcc 4.1.1.
1257    
1258     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
1259        sys/time.h, which is documented as needed for this function. It doesn't
1260        seem to matter on Linux, but it showed up on some releases of OS X.
1261    
1262     3. It seems that there are systems where bytes whose values are greater than
1263        127 match isprint() in the "C" locale. The "C" locale should be the
1264        default when a C program starts up. In most systems, only ASCII printing
1265        characters match isprint(). This difference caused the output from pcretest
1266        to vary, making some of the tests fail. I have changed pcretest so that:
1267    
1268        (a) When it is outputting text in the compiled version of a pattern, bytes
1269            other than 32-126 are always shown as hex escapes.
1270    
1271        (b) When it is outputting text that is a matched part of a subject string,
1272            it does the same, unless a different locale has been set for the match
1273            (using the /L modifier). In this case, it uses isprint() to decide.
1274    
1275     4. Fixed a major bug that caused incorrect computation of the amount of memory
1276        required for a compiled pattern when options that changed within the
1277        pattern affected the logic of the preliminary scan that determines the
1278        length. The relevant options are -x, and -i in UTF-8 mode. The result was
1279        that the computed length was too small. The symptoms of this bug were
1280        either the PCRE error "internal error: code overflow" from pcre_compile(),
1281        or a glibc crash with a message such as "pcretest: free(): invalid next
1282        size (fast)". Examples of patterns that provoked this bug (shown in
1283        pcretest format) are:
1284    
1285          /(?-x: )/x
1286          /(?x)(?-x: \s*#\s*)/
1287          /((?i)[\x{c0}])/8
1288          /(?i:[\x{c0}])/8
1289    
1290        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
1291        is now done differently.
1292    
1293     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
1294        wrapper classes; (b) implement a new function in the C++ scanner that is
1295        more efficient than the old way of doing things because it avoids levels of
1296        recursion in the regex matching; (c) add a paragraph to the documentation
1297        for the FullMatch() function.
1298    
1299     6. The escape sequence \n was being treated as whatever was defined as
1300        "newline". Not only was this contrary to the documentation, which states
1301        that \n is character 10 (hex 0A), but it also went horribly wrong when
1302        "newline" was defined as CRLF. This has been fixed.
1303    
1304     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
1305        was being set to -1 for the "end of line" case (supposedly a value that no
1306        character can have). Though this value is never used (the check for end of
1307        line is "zero bytes in current character"), it caused compiler complaints.
1308        I've changed it to 0xffffffff.
1309    
1310     8. In pcre_version.c, the version string was being built by a sequence of
1311        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
1312        string (as it is for production releases) called a macro with an empty
1313        argument. The C standard says the result of this is undefined. The gcc
1314        compiler treats it as an empty string (which was what was wanted) but it is
1315        reported that Visual C gives an error. The source has been hacked around to
1316        avoid this problem.
1317    
1318     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
1319        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
1320        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
1321        of them did).
1322    
1323    10. Originally, pcretest opened its input and output without "b"; then I was
1324        told that "b" was needed in some environments, so it was added for release
1325        5.0 to both the input and output. (It makes no difference on Unix-like
1326        systems.) Later I was told that it is wrong for the input on Windows. I've
1327        now abstracted the modes into two macros, to make it easier to fiddle with
1328        them, and removed "b" from the input mode under Windows.
1329    
1330    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
1331    
1332    12. Added -help and --help to pcretest as an official way of being reminded
1333        of the options.
1334    
1335    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
1336        and pcrecpp.cc because they annoy compilers at high warning levels.
1337    
1338    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1339    
1340    15. Fixed an occurrence of == in configure.ac that should have been = (shell
1341        scripts are not C programs :-) and which was not noticed because it works
1342        on Linux.
1343    
1344    16. pcretest is supposed to handle any length of pattern and data line (as one
1345        line or as a continued sequence of lines) by extending its input buffer if
1346        necessary. This feature was broken for very long pattern lines, leading to
1347        a string of junk being passed to pcre_compile() if the pattern was longer
1348        than about 50K.
1349    
1350    17. I have done a major re-factoring of the way pcre_compile() computes the
1351        amount of memory needed for a compiled pattern. Previously, there was code
1352        that made a preliminary scan of the pattern in order to do this. That was
1353        OK when PCRE was new, but as the facilities have expanded, it has become
1354        harder and harder to keep it in step with the real compile phase, and there
1355        have been a number of bugs (see for example, 4 above). I have now found a
1356        cunning way of running the real compile function in a "fake" mode that
1357        enables it to compute how much memory it would need, while actually only
1358        ever using a few hundred bytes of working memory and without too many
1359        tests of the mode. This should make future maintenance and development
1360        easier. A side effect of this work is that the limit of 200 on the nesting
1361        depth of parentheses has been removed (though this was never a serious
1362        limitation, I suspect). However, there is a downside: pcre_compile() now
1363        runs more slowly than before (30% or more, depending on the pattern). I
1364        hope this isn't a big issue. There is no effect on runtime performance.
1365    
1366    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1367        newline (only possible for the last line of a file) and it was a
1368        pattern that set a locale (followed by /Lsomething), pcretest crashed.
1369    
1370    19. Added additional timing features to pcretest. (1) The -tm option now times
1371        matching only, not compiling. (2) Both -t and -tm can be followed, as a
1372        separate command line item, by a number that specifies the number of
1373        repeats to use when timing. The default is 50000; this gives better
1374        precision, but takes uncomfortably long for very large patterns.
1375    
1376    20. Extended pcre_study() to be more clever in cases where a branch of a
1377        subpattern has no definite first character. For example, (a*|b*)[cd] would
1378        previously give no result from pcre_study(). Now it recognizes that the
1379        first character must be a, b, c, or d.
1380    
1381    21. There was an incorrect error "recursive call could loop indefinitely" if
1382        a subpattern (or the entire pattern) that was being tested for matching an
1383        empty string contained only one non-empty item after a nested subpattern.
1384        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1385        incorrectly, because the \d was being skipped in the check.
1386    
1387    22. The pcretest program now has a new pattern option /B and a command line
1388        option -b, which is equivalent to adding /B to every pattern. This causes
1389        it to show the compiled bytecode, without the additional information that
1390        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1391        is the same as /B/I).
1392    
1393    23. A new optimization is now able automatically to treat some sequences such
1394        as a*b as a*+b. More specifically, if something simple (such as a character
1395        or a simple class like \d) has an unlimited quantifier, and is followed by
1396        something that cannot possibly match the quantified thing, the quantifier
1397        is automatically "possessified".
1398    
1399    24. A recursive reference to a subpattern whose number was greater than 39
1400        went wrong under certain circumstances in UTF-8 mode. This bug could also
1401        have affected the operation of pcre_study().
1402    
1403    25. Realized that a little bit of performance could be had by replacing
1404        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1405    
1406    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1407    
1408    27. Possessive quantifiers such as a++ were previously implemented by turning
1409        them into atomic groups such as ($>a+). Now they have their own opcodes,
1410        which improves performance. This includes the automatically created ones
1411        from 23 above.
1412    
1413    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1414        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1415        the first matched character to be a colon. This applied both to named and
1416        numbered groups.
1417    
1418    29. The ucpinternal.h header file was missing its idempotency #ifdef.
1419    
1420    30. I was sent a "project" file called libpcre.a.dev which I understand makes
1421        building PCRE on Windows easier, so I have included it in the distribution.
1422    
1423    31. There is now a check in pcretest against a ridiculously large number being
1424        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1425        loop, the loop is abandoned.
1426    
1427    32. Forward references to subpatterns in conditions such as (?(2)...) where
1428        subpattern 2 is defined later cause pcre_compile() to search forwards in
1429        the pattern for the relevant set of parentheses. This search went wrong
1430        when there were unescaped parentheses in a character class, parentheses
1431        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1432    
1433    33. "Subroutine" calls and backreferences were previously restricted to
1434        referencing subpatterns earlier in the regex. This restriction has now
1435        been removed.
1436    
1437    34. Added a number of extra features that are going to be in Perl 5.10. On the
1438        whole, these are just syntactic alternatives for features that PCRE had
1439        previously implemented using the Python syntax or my own invention. The
1440        other formats are all retained for compatibility.
1441    
1442        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1443            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1444            also .NET compatible.
1445    
1446        (b) A recursion or subroutine call to a named group can now be defined as
1447            (?&name) as well as (?P>name).
1448    
1449        (c) A backreference to a named group can now be defined as \k<name> or
1450            \k'name' as well as (?P=name). The new forms, as well as being in Perl
1451            5.10, are also .NET compatible.
1452    
1453        (d) A conditional reference to a named group can now use the syntax
1454            (?(<name>) or (?('name') as well as (?(name).
1455    
1456        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1457            groups (named and numbered) that are never evaluated inline, but can be
1458            called as "subroutines" from elsewhere. In effect, the DEFINE condition
1459            is always false. There may be only one alternative in such a group.
1460    
1461        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1462            as the simple (?(R). The condition is true only if the most recent
1463            recursion is that of the given number or name. It does not search out
1464            through the entire recursion stack.
1465    
1466        (g) The escape \gN or \g{N} has been added, where N is a positive or
1467            negative number, specifying an absolute or relative reference.
1468    
1469    35. Tidied to get rid of some further signed/unsigned compiler warnings and
1470        some "unreachable code" warnings.
1471    
1472    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1473        things, this adds five new scripts.
1474    
1475    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1476        There were also incompatibilities regarding the handling of \Q..\E inside
1477        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1478        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1479    
1480    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1481        matches an empty string, and forcibly breaks the loop. There were bugs in
1482        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
1483        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
1484        separate and independent bugs (that affected different cases) have been
1485        fixed.
1486    
1487    39. Refactored the code to abolish the use of different opcodes for small
1488        capturing bracket numbers. This is a tidy that I avoided doing when I
1489        removed the limit on the number of capturing brackets for 3.5 back in 2001.
1490        The new approach is not only tidier, it makes it possible to reduce the
1491        memory needed to fix the previous bug (38).
1492    
1493    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1494        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1495        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1496        mode.
1497    
1498    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1499        report.
1500    
1501    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1502        copy construction and assignment in the C++ wrapper.
1503    
1504    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1505        couple of bugs that could have given wrong results in the "--newline=crlf"
1506        case.
1507    
1508    44. Added a number of casts and did some reorganization of signed/unsigned int
1509        variables following suggestions from Dair Grant. Also renamed the variable
1510        "this" as "item" because it is a C++ keyword.
1511    
1512    45. Arranged for dftables to add
1513    
1514          #include "pcre_internal.h"
1515    
1516        to pcre_chartables.c because without it, gcc 4.x may remove the array
1517        definition from the final binary if PCRE is built into a static library and
1518        dead code stripping is activated.
1519    
1520    46. For an unanchored pattern, if a match attempt fails at the start of a
1521        newline sequence, and the newline setting is CRLF or ANY, and the next two
1522        characters are CRLF, advance by two characters instead of one.
1523    
1524    
1525    Version 6.7 04-Jul-06
1526    ---------------------
1527    
1528     1. In order to handle tests when input lines are enormously long, pcretest has
1529        been re-factored so that it automatically extends its buffers when
1530        necessary. The code is crude, but this _is_ just a test program. The
1531        default size has been increased from 32K to 50K.
1532    
1533     2. The code in pcre_study() was using the value of the re argument before
1534        testing it for NULL. (Of course, in any sensible call of the function, it
1535        won't be NULL.)
1536    
1537     3. The memmove() emulation function in pcre_internal.h, which is used on
1538        systems that lack both memmove() and bcopy() - that is, hardly ever -
1539        was missing a "static" storage class specifier.
1540    
1541     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1542        containing an extended class (one that cannot be represented by a bitmap
1543        because it contains high-valued characters or Unicode property items, e.g.
1544        [\pZ]). Almost always one would set UTF-8 mode when processing such a
1545        pattern, but PCRE should not loop if you do not (it no longer does).
1546        [Detail: two cases were found: (a) a repeated subpattern containing an
1547        extended class; (b) a recursive reference to a subpattern that followed a
1548        previous extended class. It wasn't skipping over the extended class
1549        correctly when UTF-8 mode was not set.]
1550    
1551     5. A negated single-character class was not being recognized as fixed-length
1552        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1553        compile error "lookbehind assertion is not fixed length".
1554    
1555     6. The RunPerlTest auxiliary script was showing an unexpected difference
1556        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1557        write a Perl script that can interpret lines of an input file either as
1558        byte characters or as UTF-8, which is what "perltest" was being required to
1559        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1560        can't do is switch easily at run time between having the "use utf8;" pragma
1561        or not. In the end, I fudged it by using the RunPerlTest script to insert
1562        "use utf8;" explicitly for the UTF-8 tests.
1563    
1564     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1565        the end of the subject string, contrary to the documentation and to what
1566        Perl does. This was true of both matching functions. Now it matches only at
1567        the start of the subject and immediately after *internal* newlines.
1568    
1569     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1570        a pointer to an int instead of a pointer to an unsigned long int. This
1571        caused problems on 64-bit systems.
1572    
1573     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1574        instance of the 'standard' template library not being so standard".
1575    
1576    10. There was no check on the number of named subpatterns nor the maximum
1577        length of a subpattern name. The product of these values is used to compute
1578        the size of the memory block for a compiled pattern. By supplying a very
1579        long subpattern name and a large number of named subpatterns, the size
1580        computation could be caused to overflow. This is now prevented by limiting
1581        the length of names to 32 characters, and the number of named subpatterns
1582        to 10,000.
1583    
1584    11. Subpatterns that are repeated with specific counts have to be replicated in
1585        the compiled pattern. The size of memory for this was computed from the
1586        length of the subpattern and the repeat count. The latter is limited to
1587        65535, but there was no limit on the former, meaning that integer overflow
1588        could in principle occur. The compiled length of a repeated subpattern is
1589        now limited to 30,000 bytes in order to prevent this.
1590    
1591    12. Added the optional facility to have named substrings with the same name.
1592    
1593    13. Added the ability to use a named substring as a condition, using the
1594        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1595        are numbers (not recommended). Forward references are permitted.
1596    
1597    14. Added forward references in named backreferences (if you see what I mean).
1598    
1599    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1600        pattern could run off the end of the subject. For example, the pattern
1601        "(?s)(.{1,5})"8 did this with the subject "ab".
1602    
1603    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1604        PCRE_CASELESS was set when matching characters that were quantified with ?
1605        or *.
1606    
1607    17. A character class other than a single negated character that had a minimum
1608        but no maximum quantifier - for example [ab]{6,} - was not handled
1609        correctly by pce_dfa_exec(). It would match only one character.
1610    
1611    18. A valid (though odd) pattern that looked like a POSIX character
1612        class but used an invalid character after [ (for example [[,abc,]]) caused
1613        pcre_compile() to give the error "Failed: internal error: code overflow" or
1614        in some cases to crash with a glibc free() error. This could even happen if
1615        the pattern terminated after [[ but there just happened to be a sequence of
1616        letters, a binary zero, and a closing ] in the memory that followed.
1617    
1618    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1619        over the years. Originally (before any Unicode support), just the bottom 8
1620        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1621        output from "man perlunicode" includes this:
1622    
1623          The regular expression compiler produces polymorphic opcodes.  That
1624          is, the pattern adapts to the data and automatically switches to
1625          the Unicode character scheme when presented with Unicode data--or
1626          instead uses a traditional byte scheme when presented with byte
1627          data.
1628    
1629        Sadly, a wide octal escape does not cause a switch, and in a string with
1630        no other multibyte characters, these octal escapes are treated as before.
1631        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
1632        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1633        Unicode string.
1634    
1635        I have not perpetrated such confusion in PCRE. Up till now, it took just
1636        the bottom 8 bits, as in old Perl. I have now made octal escapes with
1637        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1638        translate to the appropriate multibyte character.
1639    
1640    29. Applied some refactoring to reduce the number of warnings from Microsoft
1641        and Borland compilers. This has included removing the fudge introduced
1642        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1643        a warning about an unused variable.
1644    
1645    21. PCRE has not included VT (character 0x0b) in the set of whitespace
1646        characters since release 4.0, because Perl (from release 5.004) does not.
1647        [Or at least, is documented not to: some releases seem to be in conflict
1648        with the documentation.] However, when a pattern was studied with
1649        pcre_study() and all its branches started with \s, PCRE still included VT
1650        as a possible starting character. Of course, this did no harm; it just
1651        caused an unnecessary match attempt.
1652    
1653    22. Removed a now-redundant internal flag bit that recorded the fact that case
1654        dependency changed within the pattern. This was once needed for "required
1655        byte" processing, but is no longer used. This recovers a now-scarce options
1656        bit. Also moved the least significant internal flag bit to the most-
1657        significant bit of the word, which was not previously used (hangover from
1658        the days when it was an int rather than a uint) to free up another bit for
1659        the future.
1660    
1661    23. Added support for CRLF line endings as well as CR and LF. As well as the
1662        default being selectable at build time, it can now be changed at runtime
1663        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1664        specify that it is scanning data with non-default line endings.
1665    
1666    24. Changed the definition of CXXLINK to make it agree with the definition of
1667        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1668    
1669    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1670        recursions. This makes a big different to stack usage for some patterns.
1671    
1672    26. If a subpattern containing a named recursion or subroutine reference such
1673        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1674        the space required for the compiled pattern went wrong and gave too small a
1675        value. Depending on the environment, this could lead to "Failed: internal
1676        error: code overflow at offset 49" or "glibc detected double free or
1677        corruption" errors.
1678    
1679    27. Applied patches from Google (a) to support the new newline modes and (b) to
1680        advance over multibyte UTF-8 characters in GlobalReplace.
1681    
1682    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1683        difference for some implementation of PCRE in some Windows version.
1684    
1685    29. Added some extra testing facilities to pcretest:
1686    
1687        \q<number>   in a data line sets the "match limit" value
1688        \Q<number>   in a data line sets the "match recursion limt" value
1689        -S <number>  sets the stack size, where <number> is in megabytes
1690    
1691        The -S option isn't available for Windows.
1692    
1693    
1694    Version 6.6 06-Feb-06
1695    ---------------------
1696    
1697     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1698        in pcreposix.h. I have copied the definition from pcre.h.
1699    
1700     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1701        because pcre.h is no longer a built file.
1702    
1703     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1704        not normally included in the compiled code.
1705    
1706    
1707    Version 6.5 01-Feb-06
1708    ---------------------
1709    
1710     1. When using the partial match feature with pcre_dfa_exec(), it was not
1711        anchoring the second and subsequent partial matches at the new starting
1712        point. This could lead to incorrect results. For example, with the pattern
1713        /1234/, partially matching against "123" and then "a4" gave a match.
1714    
1715     2. Changes to pcregrep:
1716    
1717        (a) All non-match returns from pcre_exec() were being treated as failures
1718            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1719            error message is output. Some extra information is given for the
1720            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
1721            probably the only errors that are likely to be caused by users (by
1722            specifying a regex that has nested indefinite repeats, for instance).
1723            If there are more than 20 of these errors, pcregrep is abandoned.
1724    
1725        (b) A binary zero was treated as data while matching, but terminated the
1726            output line if it was written out. This has been fixed: binary zeroes
1727            are now no different to any other data bytes.
1728    
1729        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1730            used to set a locale for matching. The --locale=xxxx long option has
1731            been added (no short equivalent) to specify a locale explicitly on the
1732            pcregrep command, overriding the environment variables.
1733    
1734        (d) When -B was used with -n, some line numbers in the output were one less
1735            than they should have been.
1736    
1737        (e) Added the -o (--only-matching) option.
1738    
1739        (f) If -A or -C was used with -c (count only), some lines of context were
1740            accidentally printed for the final match.
1741    
1742        (g) Added the -H (--with-filename) option.
1743    
1744        (h) The combination of options -rh failed to suppress file names for files
1745            that were found from directory arguments.
1746    
1747        (i) Added the -D (--devices) and -d (--directories) options.
1748    
1749        (j) Added the -F (--fixed-strings) option.
1750    
1751        (k) Allow "-" to be used as a file name for -f as well as for a data file.
1752    
1753        (l) Added the --colo(u)r option.
1754    
1755        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1756            is not present by default.
1757    
1758     3. A nasty bug was discovered in the handling of recursive patterns, that is,
1759        items such as (?R) or (?1), when the recursion could match a number of
1760        alternatives. If it matched one of the alternatives, but subsequently,
1761        outside the recursion, there was a failure, the code tried to back up into
1762        the recursion. However, because of the way PCRE is implemented, this is not
1763        possible, and the result was an incorrect result from the match.
1764    
1765        In order to prevent this happening, the specification of recursion has
1766        been changed so that all such subpatterns are automatically treated as
1767        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1768    
1769     4. I had overlooked the fact that, in some locales, there are characters for
1770        which isalpha() is true but neither isupper() nor islower() are true. In
1771        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1772        and ordfeminine) are like this. This affected the treatment of \w and \W
1773        when they appeared in character classes, but not when they appeared outside
1774        a character class. The bit map for "word" characters is now created
1775        separately from the results of isalnum() instead of just taking it from the
1776        upper, lower, and digit maps. (Plus the underscore character, of course.)
1777    
1778     5. The above bug also affected the handling of POSIX character classes such as
1779        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1780        permanent tables. Instead, the bit maps for such a class were previously
1781        created as the appropriate unions of the upper, lower, and digit bitmaps.
1782        Now they are created by subtraction from the [[:word:]] class, which has
1783        its own bitmap.
1784    
1785     6. The [[:blank:]] character class matches horizontal, but not vertical space.
1786        It is created by subtracting the vertical space characters (\x09, \x0a,
1787        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1788        subtraction was done in the overall bitmap for a character class, meaning
1789        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1790        be recognized. This bug has been fixed.
1791    
1792     7. Patches from the folks at Google:
1793    
1794          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1795          real life, but is still worth protecting against".
1796    
1797          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1798          regular expressions".
1799    
1800          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1801          have it.
1802    
1803          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1804          "configure" and the latter not, in order to fix a problem somebody had
1805          with compiling the Arg class on HP-UX.
1806    
1807          (e) Improve the error-handling of the C++ wrapper a little bit.
1808    
1809          (f) New tests for checking recursion limiting.
1810    
1811     8. The pcre_memmove() function, which is used only if the environment does not
1812        have a standard memmove() function (and is therefore rarely compiled),
1813        contained two bugs: (a) use of int instead of size_t, and (b) it was not
1814        returning a result (though PCRE never actually uses the result).
1815    
1816     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1817        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1818        returned instead of calling malloc() with an overflowing number that would
1819        most likely cause subsequent chaos.
1820    
1821    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1822    
1823    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1824        with this option is matched, the nmatch and pmatch options of regexec() are
1825        ignored.
1826    
1827    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1828        provided in case anyone wants to the the POSIX interface with UTF-8
1829        strings.
1830    
1831    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1832        C++ linking (needed for some HP-UX environments).
1833    
1834    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1835        (unused parameter) and in the pcre_printint() function (omitted "default"
1836        switch label when the default is to do nothing).
1837    
1838    15. Added some code to make it possible, when PCRE is compiled as a C++
1839        library, to replace subject pointers for pcre_exec() with a smart pointer
1840        class, thus making it possible to process discontinuous strings.
1841    
1842    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1843        much the same function. They were added by different people who were trying
1844        to make PCRE easy to compile on non-Unix systems. It has been suggested
1845        that PCRE_EXPORT be abolished now that there is more automatic apparatus
1846        for compiling on Windows systems. I have therefore replaced it with
1847        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1848        defaults to "extern" for C or "extern C" for C++, which works fine on
1849        Unix-like systems. It is now possible to override the value of PCRE_DATA_
1850        SCOPE with something explicit in config.h. In addition:
1851    
1852        (a) pcreposix.h still had just "extern" instead of either of these macros;
1853            I have replaced it with PCRE_DATA_SCOPE.
1854    
1855        (b) Functions such as _pcre_xclass(), which are internal to the library,
1856            but external in the C sense, all had PCRE_EXPORT in their definitions.
1857            This is apparently wrong for the Windows case, so I have removed it.
1858            (It makes no difference on Unix-like systems.)
1859    
1860    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1861        of recursive calls to match(). This is different to MATCH_LIMIT because
1862        that limits the total number of calls to match(), not all of which increase
1863        the depth of recursion. Limiting the recursion depth limits the amount of
1864        stack (or heap if NO_RECURSE is set) that is used. The default can be set
1865        when PCRE is compiled, and changed at run time. A patch from Google adds
1866        this functionality to the C++ interface.
1867    
1868    18. Changes to the handling of Unicode character properties:
1869    
1870        (a) Updated the table to Unicode 4.1.0.
1871    
1872        (b) Recognize characters that are not in the table as "Cn" (undefined).
1873    
1874        (c) I revised the way the table is implemented to a much improved format
1875            which includes recognition of ranges. It now supports the ranges that
1876            are defined in UnicodeData.txt, and it also amalgamates other
1877            characters into ranges. This has reduced the number of entries in the
1878            table from around 16,000 to around 3,000, thus reducing its size
1879            considerably. I realized I did not need to use a tree structure after
1880            all - a binary chop search is just as efficient. Having reduced the
1881            number of entries, I extended their size from 6 bytes to 8 bytes to
1882            allow for more data.
1883    
1884        (d) Added support for Unicode script names via properties such as \p{Han}.
1885    
1886    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1887        matching that character.
1888    
1889    20. When matching a repeated Unicode property with a minimum greater than zero,
1890        (for example \pL{2,}), PCRE could look past the end of the subject if it
1891        reached it while seeking the minimum number of characters. This could
1892        happen only if some of the characters were more than one byte long, because
1893        there is a check for at least the minimum number of bytes.
1894    
1895    21. Refactored the implementation of \p and \P so as to be more general, to
1896        allow for more different types of property in future. This has changed the
1897        compiled form incompatibly. Anybody with saved compiled patterns that use
1898        \p or \P will have to recompile them.
1899    
1900    22. Added "Any" and "L&" to the supported property types.
1901    
1902    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1903        but give a compile time error if the value is greater than 0xff.
1904    
1905    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1906        accidentally not being installed or uninstalled.
1907    
1908    25. The pcre.h file was built from pcre.h.in, but the only changes that were
1909        made were to insert the current release number. This seemed silly, because
1910        it made things harder for people building PCRE on systems that don't run
1911        "configure". I have turned pcre.h into a distributed file, no longer built
1912        by "configure", with the version identification directly included. There is
1913        no longer a pcre.h.in file.
1914    
1915        However, this change necessitated a change to the pcre-config script as
1916        well. It is built from pcre-config.in, and one of the substitutions was the
1917        release number. I have updated configure.ac so that ./configure now finds
1918        the release number by grepping pcre.h.
1919    
1920    26. Added the ability to run the tests under valgrind.
1921    
1922    
1923    Version 6.4 05-Sep-05
1924    ---------------------
1925    
1926     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1927        "--" to be printed when multiple files were scanned, even when none of the
1928        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1929        consider it to be a bug, and have restored the previous behaviour.
1930    
1931     2. A couple of code tidies to get rid of compiler warnings.
1932    
1933     3. The pcretest program used to cheat by referring to symbols in the library
1934        whose names begin with _pcre_. These are internal symbols that are not
1935        really supposed to be visible externally, and in some environments it is
1936        possible to suppress them. The cheating is now confined to including
1937        certain files from the library's source, which is a bit cleaner.
1938    
1939     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1940        file's purpose clearer.
1941    
1942     5. Reorganized pcre_ucp_findchar().
1943    
1944    
1945    Version 6.3 15-Aug-05
1946    ---------------------
1947    
1948     1. The file libpcre.pc.in did not have general read permission in the tarball.
1949    
1950     2. There were some problems when building without C++ support:
1951    
1952        (a) If C++ support was not built, "make install" and "make test" still
1953            tried to test it.
1954    
1955        (b) There were problems when the value of CXX was explicitly set. Some
1956            changes have been made to try to fix these, and ...
1957    
1958        (c) --disable-cpp can now be used to explicitly disable C++ support.
1959    
1960        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1961            backslash in a target when C++ was disabled. This confuses some
1962            versions of "make", apparently. Using an intermediate variable solves
1963            this. (Same for CPP_LOBJ.)
1964    
1965     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1966        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1967        necessary on certain architectures.
1968    
1969     4. Added a setting of -export-symbols-regex to the link command to remove
1970        those symbols that are exported in the C sense, but actually are local
1971        within the library, and not documented. Their names all begin with
1972        "_pcre_". This is not a perfect job, because (a) we have to except some
1973        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1974        available (and never for static libraries). I have made a note to try to
1975        find a way round (a) in the future.
1976    
1977    
1978    Version 6.2 01-Aug-05
1979    ---------------------
1980    
1981     1. There was no test for integer overflow of quantifier values. A construction
1982        such as {1111111111111111} would give undefined results. What is worse, if
1983        a minimum quantifier for a parenthesized subpattern overflowed and became
1984        negative, the calculation of the memory size went wrong. This could have
1985        led to memory overwriting.
1986    
1987     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1988    
1989     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1990        operating environments where this matters.
1991    
1992     4. Applied Giuseppe Maxia's patch to add additional features for controlling
1993        PCRE options from within the C++ wrapper.
1994    
1995     5. Named capturing subpatterns were not being correctly counted when a pattern
1996        was compiled. This caused two problems: (a) If there were more than 100
1997        such subpatterns, the calculation of the memory needed for the whole
1998        compiled pattern went wrong, leading to an overflow error. (b) Numerical
1999        back references of the form \12, where the number was greater than 9, were
2000        not recognized as back references, even though there were sufficient
2001        previous subpatterns.
2002    
2003     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
2004        versions of gcc, e.g. 2.95.4.
2005    
2006    
2007    Version 6.1 21-Jun-05
2008    ---------------------
2009    
2010     1. There was one reference to the variable "posix" in pcretest.c that was not
2011        surrounded by "#if !defined NOPOSIX".
2012    
2013     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
2014        the cross-check on the old pcre_info() function, for the benefit of the
2015        cut-down version of PCRE that is currently imported into Exim.
2016    
2017     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
2018        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
2019        patterns that start (?i) but I don't think that matters. The use of (?i) is
2020        just an example; this all applies to the other options as well.
2021    
2022     4. Since libtool seems to echo the compile commands it is issuing, the output
2023        from "make" can be reduced a bit by putting "@" in front of each libtool
2024        compile command.
2025    
2026     5. Patch from the folks at Google for configure.in to be a bit more thorough
2027        in checking for a suitable C++ installation before trying to compile the
2028        C++ stuff. This should fix a reported problem when a compiler was present,
2029        but no suitable headers.
2030    
2031     6. The man pages all had just "PCRE" as their title. I have changed them to
2032        be the relevant file name. I have also arranged that these names are
2033        retained in the file doc/pcre.txt, which is a concatenation in text format
2034        of all the man pages except the little individual ones for each function.
2035    
2036     7. The NON-UNIX-USE file had not been updated for the different set of source
2037        files that come with release 6. I also added a few comments about the C++
2038        wrapper.
2039    
2040    
2041    Version 6.0 07-Jun-05
2042    ---------------------
2043    
2044     1. Some minor internal re-organization to help with my DFA experiments.
2045    
2046     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
2047        didn't matter for the library itself when fully configured, but did matter
2048        when compiling without UCP support, or within Exim, where the ucp files are
2049        not imported.
2050    
2051     3. Refactoring of the library code to split up the various functions into
2052        different source modules. The addition of the new DFA matching code (see
2053        below) to a single monolithic source would have made it really too
2054        unwieldy, quite apart from causing all the code to be include in a
2055        statically linked application, when only some functions are used. This is
2056        relevant even without the DFA addition now that patterns can be compiled in
2057        one application and matched in another.
2058    
2059        The downside of splitting up is that there have to be some external
2060        functions and data tables that are used internally in different modules of
2061        the library but which are not part of the API. These have all had their
2062        names changed to start with "_pcre_" so that they are unlikely to clash
2063        with other external names.
2064    
2065     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
2066        a different (DFA) algorithm. Although it is slower than the original
2067        function, it does have some advantages for certain types of matching
2068        problem.
2069    
2070     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
2071        including restarting after a partial match.
2072    
2073     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
2074        defined when compiling for Windows was sent to me. I have put it into the
2075        code, though I have no means of testing or verifying it.
2076    
2077     7. Added the pcre_refcount() auxiliary function.
2078    
2079     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
2080        match before or at the first newline in the subject string. In pcretest,
2081        the /f option on a pattern can be used to set this.
2082    
2083     9. A repeated \w when used in UTF-8 mode with characters greater than 256
2084        would behave wrongly. This has been present in PCRE since release 4.0.
2085    
2086    10. A number of changes to the pcregrep command:
2087    
2088        (a) Refactored how -x works; insert ^(...)$ instead of setting
2089            PCRE_ANCHORED and checking the length, in preparation for adding
2090            something similar for -w.
2091    
2092        (b) Added the -w (match as a word) option.
2093    
2094        (c) Refactored the way lines are read and buffered so as to have more
2095            than one at a time available.
2096    
2097        (d) Implemented a pcregrep test script.
2098    
2099        (e) Added the -M (multiline match) option. This allows patterns to match
2100            over several lines of the subject. The buffering ensures that at least
2101            8K, or the rest of the document (whichever is the shorter) is available
2102            for matching (and similarly the previous 8K for lookbehind assertions).
2103    
2104        (f) Changed the --help output so that it now says
2105    
2106              -w, --word-regex(p)
2107    
2108            instead of two lines, one with "regex" and the other with "regexp"
2109            because that confused at least one person since the short forms are the
2110            same. (This required a bit of code, as the output is generated
2111            automatically from a table. It wasn't just a text change.)
2112    
2113        (g) -- can be used to terminate pcregrep options if the next thing isn't an
2114            option but starts with a hyphen. Could be a pattern or a path name
2115            starting with a hyphen, for instance.
2116    
2117        (h) "-" can be given as a file name to represent stdin.
2118    
2119        (i) When file names are being printed, "(standard input)" is used for
2120            the standard input, for compatibility with GNU grep. Previously
2121            "<stdin>" was used.
2122    
2123        (j) The option --label=xxx can be used to supply a name to be used for
2124            stdin when file names are being printed. There is no short form.
2125    
2126        (k) Re-factored the options decoding logic because we are going to add
2127            two more options that take data. Such options can now be given in four
2128            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
2129    
2130        (l) Added the -A, -B, and -C options for requesting that lines of context
2131            around matches be printed.
2132    
2133        (m) Added the -L option to print the names of files that do not contain
2134            any matching lines, that is, the complement of -l.
2135    
2136        (n) The return code is 2 if any file cannot be opened, but pcregrep does
2137            continue to scan other files.
2138    
2139        (o) The -s option was incorrectly implemented. For compatibility with other
2140            greps, it now suppresses the error message for a non-existent or non-
2141            accessible file (but not the return code). There is a new option called
2142            -q that suppresses the output of matching lines, which was what -s was
2143            previously doing.
2144    
2145        (p) Added --include and --exclude options to specify files for inclusion
2146            and exclusion when recursing.
2147    
2148    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
2149        Hopefully, it now does.
2150    
2151    12. Missing cast in pcre_study().
2152    
2153    13. Added an "uninstall" target to the makefile.
2154    
2155    14. Replaced "extern" in the function prototypes in Makefile.in with
2156        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
2157        world, but is set differently for Windows.
2158    
2159    15. Added a second compiling function called pcre_compile2(). The only
2160        difference is that it has an extra argument, which is a pointer to an
2161        integer error code. When there is a compile-time failure, this is set
2162        non-zero, in addition to the error test pointer being set to point to an
2163        error message. The new argument may be NULL if no error number is required
2164        (but then you may as well call pcre_compile(), which is now just a
2165        wrapper). This facility is provided because some applications need a
2166        numeric error indication, but it has also enabled me to tidy up the way
2167        compile-time errors are handled in the POSIX wrapper.
2168    
2169    16. Added VPATH=.libs to the makefile; this should help when building with one
2170        prefix path and installing with another. (Or so I'm told by someone who
2171        knows more about this stuff than I do.)
2172    
2173    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
2174        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
2175        match everything, including newlines. This is not POSIX-compatible, but
2176        somebody wanted the feature. From pcretest it can be activated by using
2177        both the P and the s flags.
2178    
2179    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
2180    
2181    19. libpcre.pc was being incorrectly installed as executable.
2182    
2183    20. A couple of places in pcretest check for end-of-line by looking for '\n';
2184        it now also looks for '\r' so that it will work unmodified on Windows.
2185    
2186    21. Added Google's contributed C++ wrapper to the distribution.
2187    
2188    22. Added some untidy missing memory free() calls in pcretest, to keep
2189        Electric Fence happy when testing.
2190    
2191    
2192    
2193    Version 5.0 13-Sep-04
2194    ---------------------
2195    
2196     1. Internal change: literal characters are no longer packed up into items
2197        containing multiple characters in a single byte-string. Each character
2198        is now matched using a separate opcode. However, there may be more than one
2199        byte in the character in UTF-8 mode.
2200    
2201     2. The pcre_callout_block structure has two new fields: pattern_position and
2202        next_item_length. These contain the offset in the pattern to the next match
2203        item, and its length, respectively.
2204    
2205     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
2206        insertion of callouts before each pattern item. Added the /C option to
2207        pcretest to make use of this.
2208    
2209     4. On the advice of a Windows user, the lines
2210    
2211          #if defined(_WIN32) || defined(WIN32)
2212          _setmode( _fileno( stdout ), 0x8000 );
2213          #endif  /* defined(_WIN32) || defined(WIN32) */
2214    
2215        have been added to the source of pcretest. This apparently does useful
2216        magic in relation to line terminators.
2217    
2218     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
2219        for the benefit of those environments where the "b" makes a difference.
2220    
2221     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
2222        to know about it. I have put a hack into configure.in that adds in code
2223        to set GCC=yes if CC=icc. This seems to end up at a point in the
2224        generated configure script that is early enough to affect the setting of
2225        compiler options, which is what is needed, but I have no means of testing
2226        whether it really works. (The user who reported this had patched the
2227        generated configure script, which of course I cannot do.)
2228    
2229        LATER: After change 22 below (new libtool files), the configure script
2230        seems to know about icc (and also ecc). Therefore, I have commented out
2231        this hack in configure.in.
2232    
2233     7. Added support for pkg-config (2 patches were sent in).
2234    
2235     8. Negated POSIX character classes that used a combination of internal tables
2236        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
2237        [[:^ascii]]. Typically, they would match almost any characters. The other
2238        POSIX classes were not broken in this way.
2239    
2240     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
2241        to find the match, as PCRE was deluded into thinking that the match had to
2242        start at the start point or following a newline. The same bug applied to
2243        patterns with negative forward assertions or any backward assertions
2244        preceding ".*" at the start, unless the pattern required a fixed first
2245        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
2246    
2247    10. In UTF-8 mode, when moving forwards in the subject after a failed match
2248        starting at the last subject character, bytes beyond the end of the subject
2249        string were read.
2250    
2251    11. Renamed the variable "class" as "classbits" to make life easier for C++
2252        users. (Previously there was a macro definition, but it apparently wasn't
2253        enough.)
2254    
2255    12. Added the new field "tables" to the extra data so that tables can be passed
2256        in at exec time, or the internal tables can be re-selected. This allows
2257        a compiled regex to be saved and re-used at a later time by a different
2258        program that might have everything at different addresses.
2259    
2260    13. Modified the pcre-config script so that, when run on Solaris, it shows a
2261        -R library as well as a -L library.
2262    
2263    14. The debugging options of pcretest (-d on the command line or D on a
2264        pattern) showed incorrect output for anything following an extended class
2265        that contained multibyte characters and which was followed by a quantifier.
2266    
2267    15. Added optional support for general category Unicode character properties
2268        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
2269        support. It adds about 90K to the size of the library. The meanings of the
2270        inbuilt class escapes such as \d and \s have NOT been changed.
2271    
2272    16. Updated pcredemo.c to include calls to free() to release the memory for the
2273        compiled pattern.
2274    
2275    17. The generated file chartables.c was being created in the source directory
2276        instead of in the building directory. This caused the build to fail if the
2277        source directory was different from the building directory, and was
2278        read-only.
2279    
2280    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
2281        file. No doubt somebody will tell me if they don't make sense... Also added
2282        Dan Mooney's comments about building on OpenVMS.
2283    
2284    19. Added support for partial matching via the PCRE_PARTIAL option for
2285        pcre_exec() and the \P data escape in pcretest.
2286    
2287    20. Extended pcretest with 3 new pattern features:
2288    
2289        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
2290              write the compiled pattern to the file whose name is "rest-of-line".
2291              This is a straight binary dump of the data, with the saved pointer to
2292              the character tables forced to be NULL. The study data, if any, is
2293              written too. After writing, pcretest reads a new pattern.
2294    
2295        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
2296              compiled pattern from the given file. There must not be any
2297              occurrences of "<" in the file name (pretty unlikely); if there are,
2298              pcretest will instead treat the initial "<" as a pattern delimiter.
2299              After reading in the pattern, pcretest goes on to read data lines as
2300              usual.
2301    
2302        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
2303              and 16-bit fields in a compiled pattern, to simulate a pattern that
2304              was compiled on a host of opposite endianness.
2305    
2306    21. The pcre-exec() function can now cope with patterns that were compiled on
2307        hosts of opposite endianness, with this restriction:
2308    
2309          As for any compiled expression that is saved and used later, the tables
2310          pointer field cannot be preserved; the extra_data field in the arguments
2311          to pcre_exec() should be used to pass in a tables address if a value
2312          other than the default internal tables were used at compile time.
2313    
2314    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
2315        now diagnosed as an error. Previously, most of the time, a negative number
2316        would have been treated as zero, but if in addition "ovector" was passed as
2317        NULL, a crash could occur.
2318    
2319    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
2320        new versions from the libtool 1.5 distribution (the last one is a copy of
2321        a file called libtool.m4). This seems to have fixed the need to patch
2322        "configure" to support Darwin 1.3 (which I used to do). However, I still
2323        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
2324        workstation).
2325    
2326    24. Changed the PCRE licence to be the more standard "BSD" licence.
2327    
2328    
2329    Version 4.5 01-Dec-03
2330    ---------------------
2331    
2332     1. There has been some re-arrangement of the code for the match() function so
2333        that it can be compiled in a version that does not call itself recursively.
2334        Instead, it keeps those local variables that need separate instances for
2335        each "recursion" in a frame on the heap, and gets/frees frames whenever it
2336        needs to "recurse". Keeping track of where control must go is done by means
2337        of setjmp/longjmp. The whole thing is implemented by a set of macros that
2338        hide most of the details from the main code, and operates only if
2339        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
2340        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
2341        operating.
2342    
2343        To make it easier for callers to provide specially tailored get/free
2344        functions for this usage, two new functions, pcre_stack_malloc, and
2345        pcre_stack_free, are used. They are always called in strict stacking order,
2346        and the size of block requested is always the same.
2347    
2348        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
2349        PCRE has been compiled to use the stack or the heap for recursion. The
2350        -C option of pcretest uses this to show which version is compiled.
2351    
2352        A new data escape \S, is added to pcretest; it causes the amounts of store
2353        obtained and freed by both kinds of malloc/free at match time to be added
2354        to the output.
2355    
2356     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
2357        what's available on my current Linux desktop machine.
2358    
2359     3. When matching a UTF-8 string, the test for a valid string at the start has
2360        been extended. If start_offset is not zero, PCRE now checks that it points
2361        to a byte that is the start of a UTF-8 character. If not, it returns
2362        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
2363        this is necessary because there may be backward assertions in the pattern.
2364        When matching the same subject several times, it may save resources to use
2365        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
2366    
2367     4. The code for checking the validity of UTF-8 strings has been tightened so
2368        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
2369        containing "overlong sequences".
2370    
2371     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
2372        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
2373        should have been "&", but it just so happened that all the cases this let
2374        through by mistake were picked up later in the function.
2375    
2376     6. I had used a variable called "isblank" - this is a C99 function, causing
2377        some compilers to warn. To avoid this, I renamed it (as "blankclass").
2378    
2379     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
2380        prompting; (b) run "./pcretest /dev/null" at the start of the test script
2381        so the version is shown; (c) stop "make test" echoing "./RunTest".
2382    
2383     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
2384    
2385     9. The prototype for memmove() for systems that don't have it was using
2386        size_t, but the inclusion of the header that defines size_t was later. I've
2387        moved the #includes for the C headers earlier to avoid this.
2388    
2389    10. Added some adjustments to the code to make it easier to compiler on certain
2390        special systems:
2391    
2392          (a) Some "const" qualifiers were missing.
2393          (b) Added the macro EXPORT before all exported functions; by default this
2394              is defined to be empty.
2395          (c) Changed the dftables auxiliary program (that builds chartables.c) so
2396              that it reads its output file name as an argument instead of writing
2397              to the standard output and assuming this can be redirected.
2398    
2399    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
2400        class containing characters with values greater than 255, PCRE compilation
2401        went into a loop.
2402    
2403    12. A recursive reference to a subpattern that was within another subpattern
2404        that had a minimum quantifier of zero caused PCRE to crash. For example,
2405        (x(y(?2))z)? provoked this bug with a subject that got as far as the
2406        recursion. If the recursively-called subpattern itself had a zero repeat,
2407        that was OK.
2408    
2409    13. In pcretest, the buffer for reading a data line was set at 30K, but the
2410        buffer into which it was copied (for escape processing) was still set at
2411        1024, so long lines caused crashes.
2412    
2413    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
2414        "internal error: code overflow...". This applied to any character class
2415        that was followed by a possessive quantifier.
2416    
2417    15. Modified the Makefile to add libpcre.la as a prerequisite for
2418        libpcreposix.la because I was told this is needed for a parallel build to
2419        work.
2420    
2421    16. If a pattern that contained .* following optional items at the start was
2422        studied, the wrong optimizing data was generated, leading to matching
2423        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
2424        matching string must start with a or b or c. The correct conclusion for
2425        this pattern is that a match can start with any character.
2426    
2427    
2428  Version 4.4 13-Aug-03  Version 4.4 13-Aug-03
2429  ---------------------  ---------------------
2430    

Legend:
Removed from v.71  
changed lines
  Added in v.488

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12