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

Diff of /code/trunk/ChangeLog

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

revision 67 by nigel, Sat Feb 24 21:40:13 2007 UTC revision 311 by ph10, Wed Jan 23 17:34:58 2008 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4    Version 7.6 21-Jan-08
5    ---------------------
6    
7    1.  A character class containing a very large number of characters with
8        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
9        overflow.
10    
11    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
12        HAVE_LONG_LONG is not defined.
13    
14    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
15        bring them up to date and include new features. This patch includes:
16    
17        - Fixed PH's badly added libz and libbz2 support.
18        - Fixed a problem with static linking.
19        - Added pcredemo.
20        - Fixed dftables problem and added an option.
21        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
22            HAVE_LONG_LONG.
23        - Added readline support for pcretest.
24        - Added an listing of the option settings after cmake has run.
25    
26    4.  A user submitted a patch to Makefile that makes it easy to created a dll
27        under mingw. I added stuff to Makefile.am that cause it to include this
28        special target, without affecting anything else.
29    
30    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
31        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
32        exported in the Windows port". It has not yet been confirmed that the patch
33        solves the problem, but it does no harm.
34    
35    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
36        NON_STANDARD_LIB_SUFFIX and correct the comment about stack recursion.
37    
38    
39    Version 7.5 10-Jan-08
40    ---------------------
41    
42    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
43        values in parens when parsing an RE using the C++ wrapper."
44    
45    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
46        Characters greater than 255 were excluded from the class instead of being
47        included.
48    
49    3.  The same bug as (2) above applied to negated POSIX classes such as
50        [:^space:].
51    
52    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
53        defined or documented. It seems to have been a typo for PCRE_STATIC, so
54        I have changed it.
55    
56    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
57        first named subpattern) and a construct such as (?&a) would reference the
58        first named subpattern whose name started with "a" (in other words, the
59        length check was missing). Both these problems are fixed. "Subpattern name
60        expected" is now given for (?&) (a zero-length name), and this patch also
61        makes it give the same error for \k'' (previously it complained that that
62        was a reference to a non-existent subpattern).
63    
64    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
65        this is right because (?- can be followed by option settings as well as by
66        digits. I have, however, made the messages clearer.
67    
68    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
69        than the number used in the conditional) now cause a compile-time error.
70        This is actually not compatible with Perl, which accepts such patterns, but
71        treats the conditional as always being FALSE (as PCRE used to), but it
72        seems to me that giving a diagnostic is better.
73    
74    8.  Change "alphameric" to the more common word "alphanumeric" in comments
75        and messages.
76    
77    9.  Fix two occurrences of "backslash" in comments that should have been
78        "backspace".
79    
80    10. Remove two redundant lines of code that can never be obeyed (their function
81        was moved elsewhere).
82    
83    11. The program that makes PCRE's Unicode character property table had a bug
84        which caused it to generate incorrect table entries for sequences of
85        characters that have the same character type, but are in different scripts.
86        It amalgamated them into a single range, with the script of the first of
87        them. In other words, some characters were in the wrong script. There were
88        thirteen such cases, affecting characters in the following ranges:
89    
90          U+002b0 - U+002c1
91          U+0060c - U+0060d
92          U+0061e - U+00612
93          U+0064b - U+0065e
94          U+0074d - U+0076d
95          U+01800 - U+01805
96          U+01d00 - U+01d77
97          U+01d9b - U+01dbf
98          U+0200b - U+0200f
99          U+030fc - U+030fe
100          U+03260 - U+0327f
101          U+0fb46 - U+0fbb1
102          U+10450 - U+1049d
103    
104    12. The -o option (show only the matching part of a line) for pcregrep was not
105        compatible with GNU grep in that, if there was more than one match in a
106        line, it showed only the first of them. It now behaves in the same way as
107        GNU grep.
108    
109    13. If the -o and -v options were combined for pcregrep, it printed a blank
110        line for every non-matching line. GNU grep prints nothing, and pcregrep now
111        does the same. The return code can be used to tell if there were any
112        non-matching lines.
113    
114    14. Added --file-offsets and --line-offsets to pcregrep.
115    
116    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
117        infinitely looping recursion. The bug was that positive lookaheads were not
118        being skipped when checking for a possible empty match (negative lookaheads
119        and both kinds of lookbehind were skipped).
120    
121    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
122        inclusion of <windows.h> to before rather than after the definition of
123        INVALID_FILE_ATTRIBUTES (patch from David Byron).
124    
125    17. Specifying a possessive quantifier with a specific limit for a Unicode
126        character property caused pcre_compile() to compile bad code, which led at
127        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
128        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
129        caused the error; without that there was no problem.
130    
131    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
132    
133    19. Added --enable-pcretest-libreadline.
134    
135    20. In pcrecpp.cc, the variable 'count' was incremented twice in
136        RE::GlobalReplace(). As a result, the number of replacements returned was
137        double what it should be. I removed one of the increments, but Craig sent a
138        later patch that removed the other one (the right fix) and added unit tests
139        that check the return values (which was not done before).
140    
141    21. Several CMake things:
142    
143        (1) Arranged that, when cmake is used on Unix, the libraries end up with
144            the names libpcre and libpcreposix, not just pcre and pcreposix.
145    
146        (2) The above change means that pcretest and pcregrep are now correctly
147            linked with the newly-built libraries, not previously installed ones.
148    
149        (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
150    
151    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
152        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
153        UTF-8 newline character). The key issue is that the pattern starts .*;
154        this means that the match must be either at the beginning, or after a
155        newline. The bug was in the code for advancing after a failed match and
156        checking that the new position followed a newline. It was not taking
157        account of UTF-8 characters correctly.
158    
159    23. PCRE was behaving differently from Perl in the way it recognized POSIX
160        character classes. PCRE was not treating the sequence [:...:] as a
161        character class unless the ... were all letters. Perl, however, seems to
162        allow any characters between [: and :], though of course it rejects as
163        unknown any "names" that contain non-letters, because all the known class
164        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
165        for example, whereas PCRE did not - it did not recognize a POSIX character
166        class. This seemed a bit dangerous, so the code has been changed to be
167        closer to Perl. The behaviour is not identical to Perl, because PCRE will
168        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
169        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
170        Perl does, and where it didn't before.
171    
172    24. Rewrite so as to remove the single use of %n from pcregrep because in some
173        Windows environments %n is disabled by default.
174    
175    
176    Version 7.4 21-Sep-07
177    ---------------------
178    
179    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
180        means that a class such as [\s] counted as "explicit reference to CR or
181        LF". That isn't really right - the whole point of the change was to try to
182        help when there was an actual mention of one of the two characters. So now
183        the change happens only if \r or \n (or a literal CR or LF) character is
184        encountered.
185    
186    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
187        of both had grown to the point where there were only 3 bits left.
188        Fortunately, there was spare space in the data structure, and so I have
189        moved the internal flags into a new 16-bit field to free up more option
190        bits.
191    
192    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
193        but did not set the internal JCHANGED flag - either of these is enough to
194        control the way the "get" function works - but the PCRE_INFO_JCHANGED
195        facility is supposed to tell if (?J) was ever used, so now (?J) at the
196        start sets both bits.
197    
198    4.  Added options (at build time, compile time, exec time) to change \R from
199        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
200    
201    5.  doc/pcresyntax.html was missing from the distribution.
202    
203    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
204        compatibility, even though it is no longer used.
205    
206    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
207        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
208        windows.h file is present (where different names are used). [This was
209        reversed later after testing - see 16 below.]
210    
211    8.  Changed all #include <config.h> to #include "config.h". There were also
212        some further <pcre.h> cases that I changed to "pcre.h".
213    
214    9.  When pcregrep was used with the --colour option, it missed the line ending
215        sequence off the lines that it output.
216    
217    10. It was pointed out to me that arrays of string pointers cause lots of
218        relocations when a shared library is dynamically loaded. A technique of
219        using a single long string with a table of offsets can drastically reduce
220        these. I have refactored PCRE in four places to do this. The result is
221        dramatic:
222    
223          Originally:                          290
224          After changing UCP table:            187
225          After changing error message table:   43
226          After changing table of "verbs"       36
227          After changing table of Posix names   22
228    
229        Thanks to the folks working on Gregex for glib for this insight.
230    
231    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
232        unicode-properties was also set.
233    
234    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
235    
236    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
237        checked only for CRLF.
238    
239    14. Added casts to pcretest.c to avoid compiler warnings.
240    
241    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
242    
243    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
244        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
245        entirely. This removes changes made in 7 above.
246    
247    17. The CMake files have been updated, and there is now more information about
248        building with CMake in the NON-UNIX-USE document.
249    
250    
251    Version 7.3 28-Aug-07
252    ---------------------
253    
254     1. In the rejigging of the build system that eventually resulted in 7.1, the
255        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
256        brackets there is not right, since it causes compilers to look for an
257        installed pcre.h, not the version that is in the source that is being
258        compiled (which of course may be different). I have changed it back to:
259    
260          #include "pcre.h"
261    
262        I have a vague recollection that the change was concerned with compiling in
263        different directories, but in the new build system, that is taken care of
264        by the VPATH setting the Makefile.
265    
266     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
267        when the subject happened to end in the byte 0x85 (e.g. if the last
268        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
269        characters but of course it shouldn't be taken as a newline when it is part
270        of another character. The bug was that, for an unlimited repeat of . in
271        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
272        characters when looking for a newline.
273    
274     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
275    
276     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
277        in debug output.
278    
279     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
280        long printing in the pcrecpp unittest when running under MinGW.
281    
282     6. ESC_K was left out of the EBCDIC table.
283    
284     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
285        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
286        limit also applies to "virtual nesting" when a pattern is recursive, and in
287        this case 1000 isn't so big. I have been able to remove this limit at the
288        expense of backing off one optimization in certain circumstances. Normally,
289        when pcre_exec() would call its internal match() function recursively and
290        immediately return the result unconditionally, it uses a "tail recursion"
291        feature to save stack. However, when a subpattern that can match an empty
292        string has an unlimited repetition quantifier, it no longer makes this
293        optimization. That gives it a stack frame in which to save the data for
294        checking that an empty string has been matched. Previously this was taken
295        from the 1000-entry workspace that had been reserved. So now there is no
296        explicit limit, but more stack is used.
297    
298     8. Applied Daniel's patches to solve problems with the import/export magic
299        syntax that is required for Windows, and which was going wrong for the
300        pcreposix and pcrecpp parts of the library. These were overlooked when this
301        problem was solved for the main library.
302    
303     9. There were some crude static tests to avoid integer overflow when computing
304        the size of patterns that contain repeated groups with explicit upper
305        limits. As the maximum quantifier is 65535, the maximum group length was
306        set at 30,000 so that the product of these two numbers did not overflow a
307        32-bit integer. However, it turns out that people want to use groups that
308        are longer than 30,000 bytes (though not repeat them that many times).
309        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
310        made it possible to implement the integer overflow checks in a much more
311        dynamic way, which I have now done. The artificial limitation on group
312        length has been removed - we now have only the limit on the total length of
313        the compiled pattern, which depends on the LINK_SIZE setting.
314    
315    10. Fixed a bug in the documentation for get/copy named substring when
316        duplicate names are permitted. If none of the named substrings are set, the
317        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
318        empty string.
319    
320    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
321        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
322        because the ] is interpreted as the first data character and the
323        terminating ] is not found. PCRE has been made compatible with Perl in this
324        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
325        cause memory overwriting.
326    
327    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
328        string has been matched (to stop an infinite loop). It was not recognizing
329        a conditional subpattern that could match an empty string if that
330        subpattern was within another subpattern. For example, it looped when
331        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
332        condition was not nested. This bug has been fixed.
333    
334    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
335        past the start of the subject in the presence of bytes with the top bit
336        set, for example "\x8aBCD".
337    
338    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
339        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
340    
341    14. Optimized (?!) to (*FAIL).
342    
343    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
344        This restricts code points to be within the range 0 to 0x10FFFF, excluding
345        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
346        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
347        does: it's just the validity check that is more restrictive.
348    
349    16. Inserted checks for integer overflows during escape sequence (backslash)
350        processing, and also fixed erroneous offset values for syntax errors during
351        backslash processing.
352    
353    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
354        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
355    
356    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
357        caused an overrun.
358    
359    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
360        something other than just ASCII characters) inside a group that had an
361        unlimited repeat caused a loop at compile time (while checking to see
362        whether the group could match an empty string).
363    
364    20. Debugging a pattern containing \p or \P could cause a crash. For example,
365        [\P{Any}] did so. (Error in the code for printing property names.)
366    
367    21. An orphan \E inside a character class could cause a crash.
368    
369    22. A repeated capturing bracket such as (A)? could cause a wild memory
370        reference during compilation.
371    
372    23. There are several functions in pcre_compile() that scan along a compiled
373        expression for various reasons (e.g. to see if it's fixed length for look
374        behind). There were bugs in these functions when a repeated \p or \P was
375        present in the pattern. These operators have additional parameters compared
376        with \d, etc, and these were not being taken into account when moving along
377        the compiled data. Specifically:
378    
379        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
380            length.
381    
382        (b) An item such as \pL+ within a repeated group could cause crashes or
383            loops.
384    
385        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
386            "reference to non-existent subpattern" error.
387    
388        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
389    
390    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
391        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
392    
393    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
394    
395    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
396        character were causing crashes (broken optimization).
397    
398    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
399        \p or \P) caused a compile-time loop.
400    
401    28. More problems have arisen in unanchored patterns when CRLF is a valid line
402        break. For example, the unstudied pattern [\r\n]A does not match the string
403        "\r\nA" because change 7.0/46 below moves the current point on by two
404        characters after failing to match at the start. However, the pattern \nA
405        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
406        the same is true. There doesn't seem any very clean way out of this, but
407        what I have chosen to do makes the common cases work: PCRE now takes note
408        of whether there can be an explicit match for \r or \n anywhere in the
409        pattern, and if so, 7.0/46 no longer applies. As part of this change,
410        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
411        pattern has explicit CR or LF references.
412    
413    29. Added (*CR) etc for changing newline setting at start of pattern.
414    
415    
416    Version 7.2 19-Jun-07
417    ---------------------
418    
419     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
420        which is apparently normally available under Windows.
421    
422     2. Re-jig the pcregrep tests with different newline settings in an attempt
423        to make them independent of the local environment's newline setting.
424    
425     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
426    
427     4. Some of the "internals" tests were previously cut out when the link size
428        was not 2, because the output contained actual offsets. The recent new
429        "Z" feature of pcretest means that these can be cut out, making the tests
430        usable with all link sizes.
431    
432     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
433        stack recursion. This gives a massive performance boost under BSD, but just
434        a small improvement under Linux. However, it saves one field in the frame
435        in all cases.
436    
437     6. Added more features from the forthcoming Perl 5.10:
438    
439        (a) (?-n) (where n is a string of digits) is a relative subroutine or
440            recursion call. It refers to the nth most recently opened parentheses.
441    
442        (b) (?+n) is also a relative subroutine call; it refers to the nth next
443            to be opened parentheses.
444    
445        (c) Conditions that refer to capturing parentheses can be specified
446            relatively, for example, (?(-2)... or (?(+3)...
447    
448        (d) \K resets the start of the current match so that everything before
449            is not part of it.
450    
451        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
452    
453        (f) \g{name} is another synonym - part of Perl 5.10's unification of
454            reference syntax.
455    
456        (g) (?| introduces a group in which the numbering of parentheses in each
457            alternative starts with the same number.
458    
459        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
460    
461     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
462        PCRE_INFO_JCHANGED.
463    
464     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
465        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
466        for detecting groups that can match an empty string.
467    
468     9. A pattern with a very large number of alternatives (more than several
469        hundred) was running out of internal workspace during the pre-compile
470        phase, where pcre_compile() figures out how much memory will be needed. A
471        bit of new cunning has reduced the workspace needed for groups with
472        alternatives. The 1000-alternative test pattern now uses 12 bytes of
473        workspace instead of running out of the 4096 that are available.
474    
475    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
476    
477    11. Applied patch from Google to remove an optimization that didn't quite work.
478        The report of the bug said:
479    
480          pcrecpp::RE("a*").FullMatch("aaa") matches, while
481          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
482          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
483    
484    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
485        it matched the wrong number of bytes.
486    
487    
488    Version 7.1 24-Apr-07
489    ---------------------
490    
491     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
492        that is more "standard", making use of automake and other Autotools. There
493        is some re-arrangement of the files and adjustment of comments consequent
494        on this.
495    
496     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
497        for recursive directory scanning broke on some systems because the files
498        are not scanned in any specific order and on different systems the order
499        was different. A call to "sort" has been inserted into RunGrepTest for the
500        approprate test as a short-term fix. In the longer term there may be an
501        alternative.
502    
503     3. I had an email from Eric Raymond about problems translating some of PCRE's
504        man pages to HTML (despite the fact that I distribute HTML pages, some
505        people do their own conversions for various reasons). The problems
506        concerned the use of low-level troff macros .br and .in. I have therefore
507        removed all such uses from the man pages (some were redundant, some could
508        be replaced by .nf/.fi pairs). The 132html script that I use to generate
509        HTML has been updated to handle .nf/.fi and to complain if it encounters
510        .br or .in.
511    
512     4. Updated comments in configure.ac that get placed in config.h.in and also
513        arranged for config.h to be included in the distribution, with the name
514        config.h.generic, for the benefit of those who have to compile without
515        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
516    
517     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
518        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
519        makevp.bat for the current PCRE, using the additional files
520        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
521    
522     6. A Windows user reported a minor discrepancy with test 2, which turned out
523        to be caused by a trailing space on an input line that had got lost in his
524        copy. The trailing space was an accident, so I've just removed it.
525    
526     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
527        that is needed.
528    
529     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
530        as "const" (a) because they are and (b) because it helps the PHP
531        maintainers who have recently made a script to detect big data structures
532        in the php code that should be moved to the .rodata section. I remembered
533        to update Builducptable as well, so it won't revert if ucptable.h is ever
534        re-created.
535    
536     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
537        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
538        order to be able to cut out the UTF-8 tables in the latter when UTF-8
539        support is not required. This saves 1.5-2K of code, which is important in
540        some applications.
541    
542        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
543        so as not to refer to the tables, even though these functions will never be
544        called when UTF-8 support is disabled. Otherwise there are problems with a
545        shared library.
546    
547    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
548    
549        (a) It was defining its arguments as char * instead of void *.
550    
551        (b) It was assuming that all moves were upwards in memory; this was true
552            a long time ago when I wrote it, but is no longer the case.
553    
554        The emulated memove() is provided for those environments that have neither
555        memmove() nor bcopy(). I didn't think anyone used it these days, but that
556        is clearly not the case, as these two bugs were recently reported.
557    
558    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
559        and Detrail to create the HTML documentation, the .txt form of the man
560        pages, and it removes trailing spaces from listed files. It also creates
561        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
562        case, it wraps all the #defines with #ifndefs. This script should be run
563        before "make dist".
564    
565    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
566        with Unicode property support.
567    
568        (a) For a maximizing quantifier, if the two different cases of the
569            character were of different lengths in their UTF-8 codings (there are
570            some cases like this - I found 11), and the matching function had to
571            back up over a mixture of the two cases, it incorrectly assumed they
572            were both the same length.
573    
574        (b) When PCRE was configured to use the heap rather than the stack for
575            recursion during matching, it was not correctly preserving the data for
576            the other case of a UTF-8 character when checking ahead for a match
577            while processing a minimizing repeat. If the check also involved
578            matching a wide character, but failed, corruption could cause an
579            erroneous result when trying to check for a repeat of the original
580            character.
581    
582    13. Some tidying changes to the testing mechanism:
583    
584        (a) The RunTest script now detects the internal link size and whether there
585            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
586            values substituted by "configure". (The RunGrepTest script already did
587            this for UTF-8.) The configure.ac script no longer substitutes the
588            relevant variables.
589    
590        (b) The debugging options /B and /D in pcretest show the compiled bytecode
591            with length and offset values. This means that the output is different
592            for different internal link sizes. Test 2 is skipped for link sizes
593            other than 2 because of this, bypassing the problem. Unfortunately,
594            there was also a test in test 3 (the locale tests) that used /B and
595            failed for link sizes other than 2. Rather than cut the whole test out,
596            I have added a new /Z option to pcretest that replaces the length and
597            offset values with spaces. This is now used to make test 3 independent
598            of link size. (Test 2 will be tidied up later.)
599    
600    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
601        segmentation fault instead of returning the appropriate error message.
602    
603    15. In multiline mode when the newline sequence was set to "any", the pattern
604        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
605        This doesn't seem right; it now treats the CRLF combination as the line
606        ending, and so does not match in that case. It's only a pattern such as ^$
607        that would hit this one: something like ^ABC$ would have failed after \r
608        and then tried again after \r\n.
609    
610    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
611        in an attempt to make files that differ only in their line terminators
612        compare equal. This works on Linux.
613    
614    17. Under certain error circumstances pcregrep might try to free random memory
615        as it exited. This is now fixed, thanks to valgrind.
616    
617    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
618        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
619        was because its rules for how to advance for /g after matching an empty
620        string at the end of a line did not allow for this case. They now check for
621        it specially.
622    
623    20. pcretest is supposed to handle patterns and data of any length, by
624        extending its buffers when necessary. It was getting this wrong when the
625        buffer for a data line had to be extended.
626    
627    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
628        CRLF as a newline sequence.
629    
630    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
631        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
632        I have nevertheless tidied it up.
633    
634    23. Added some casts to kill warnings from HP-UX ia64 compiler.
635    
636    24. Added a man page for pcre-config.
637    
638    
639    Version 7.0 19-Dec-06
640    ---------------------
641    
642     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
643        moving to gcc 4.1.1.
644    
645     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
646        sys/time.h, which is documented as needed for this function. It doesn't
647        seem to matter on Linux, but it showed up on some releases of OS X.
648    
649     3. It seems that there are systems where bytes whose values are greater than
650        127 match isprint() in the "C" locale. The "C" locale should be the
651        default when a C program starts up. In most systems, only ASCII printing
652        characters match isprint(). This difference caused the output from pcretest
653        to vary, making some of the tests fail. I have changed pcretest so that:
654    
655        (a) When it is outputting text in the compiled version of a pattern, bytes
656            other than 32-126 are always shown as hex escapes.
657    
658        (b) When it is outputting text that is a matched part of a subject string,
659            it does the same, unless a different locale has been set for the match
660            (using the /L modifier). In this case, it uses isprint() to decide.
661    
662     4. Fixed a major bug that caused incorrect computation of the amount of memory
663        required for a compiled pattern when options that changed within the
664        pattern affected the logic of the preliminary scan that determines the
665        length. The relevant options are -x, and -i in UTF-8 mode. The result was
666        that the computed length was too small. The symptoms of this bug were
667        either the PCRE error "internal error: code overflow" from pcre_compile(),
668        or a glibc crash with a message such as "pcretest: free(): invalid next
669        size (fast)". Examples of patterns that provoked this bug (shown in
670        pcretest format) are:
671    
672          /(?-x: )/x
673          /(?x)(?-x: \s*#\s*)/
674          /((?i)[\x{c0}])/8
675          /(?i:[\x{c0}])/8
676    
677        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
678        is now done differently.
679    
680     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
681        wrapper classes; (b) implement a new function in the C++ scanner that is
682        more efficient than the old way of doing things because it avoids levels of
683        recursion in the regex matching; (c) add a paragraph to the documentation
684        for the FullMatch() function.
685    
686     6. The escape sequence \n was being treated as whatever was defined as
687        "newline". Not only was this contrary to the documentation, which states
688        that \n is character 10 (hex 0A), but it also went horribly wrong when
689        "newline" was defined as CRLF. This has been fixed.
690    
691     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
692        was being set to -1 for the "end of line" case (supposedly a value that no
693        character can have). Though this value is never used (the check for end of
694        line is "zero bytes in current character"), it caused compiler complaints.
695        I've changed it to 0xffffffff.
696    
697     8. In pcre_version.c, the version string was being built by a sequence of
698        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
699        string (as it is for production releases) called a macro with an empty
700        argument. The C standard says the result of this is undefined. The gcc
701        compiler treats it as an empty string (which was what was wanted) but it is
702        reported that Visual C gives an error. The source has been hacked around to
703        avoid this problem.
704    
705     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
706        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
707        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
708        of them did).
709    
710    10. Originally, pcretest opened its input and output without "b"; then I was
711        told that "b" was needed in some environments, so it was added for release
712        5.0 to both the input and output. (It makes no difference on Unix-like
713        systems.) Later I was told that it is wrong for the input on Windows. I've
714        now abstracted the modes into two macros, to make it easier to fiddle with
715        them, and removed "b" from the input mode under Windows.
716    
717    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
718    
719    12. Added -help and --help to pcretest as an official way of being reminded
720        of the options.
721    
722    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
723        and pcrecpp.cc because they annoy compilers at high warning levels.
724    
725    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
726    
727    15. Fixed an occurrence of == in configure.ac that should have been = (shell
728        scripts are not C programs :-) and which was not noticed because it works
729        on Linux.
730    
731    16. pcretest is supposed to handle any length of pattern and data line (as one
732        line or as a continued sequence of lines) by extending its input buffer if
733        necessary. This feature was broken for very long pattern lines, leading to
734        a string of junk being passed to pcre_compile() if the pattern was longer
735        than about 50K.
736    
737    17. I have done a major re-factoring of the way pcre_compile() computes the
738        amount of memory needed for a compiled pattern. Previously, there was code
739        that made a preliminary scan of the pattern in order to do this. That was
740        OK when PCRE was new, but as the facilities have expanded, it has become
741        harder and harder to keep it in step with the real compile phase, and there
742        have been a number of bugs (see for example, 4 above). I have now found a
743        cunning way of running the real compile function in a "fake" mode that
744        enables it to compute how much memory it would need, while actually only
745        ever using a few hundred bytes of working memory and without too many
746        tests of the mode. This should make future maintenance and development
747        easier. A side effect of this work is that the limit of 200 on the nesting
748        depth of parentheses has been removed (though this was never a serious
749        limitation, I suspect). However, there is a downside: pcre_compile() now
750        runs more slowly than before (30% or more, depending on the pattern). I
751        hope this isn't a big issue. There is no effect on runtime performance.
752    
753    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
754        newline (only possible for the last line of a file) and it was a
755        pattern that set a locale (followed by /Lsomething), pcretest crashed.
756    
757    19. Added additional timing features to pcretest. (1) The -tm option now times
758        matching only, not compiling. (2) Both -t and -tm can be followed, as a
759        separate command line item, by a number that specifies the number of
760        repeats to use when timing. The default is 50000; this gives better
761        precision, but takes uncomfortably long for very large patterns.
762    
763    20. Extended pcre_study() to be more clever in cases where a branch of a
764        subpattern has no definite first character. For example, (a*|b*)[cd] would
765        previously give no result from pcre_study(). Now it recognizes that the
766        first character must be a, b, c, or d.
767    
768    21. There was an incorrect error "recursive call could loop indefinitely" if
769        a subpattern (or the entire pattern) that was being tested for matching an
770        empty string contained only one non-empty item after a nested subpattern.
771        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
772        incorrectly, because the \d was being skipped in the check.
773    
774    22. The pcretest program now has a new pattern option /B and a command line
775        option -b, which is equivalent to adding /B to every pattern. This causes
776        it to show the compiled bytecode, without the additional information that
777        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
778        is the same as /B/I).
779    
780    23. A new optimization is now able automatically to treat some sequences such
781        as a*b as a*+b. More specifically, if something simple (such as a character
782        or a simple class like \d) has an unlimited quantifier, and is followed by
783        something that cannot possibly match the quantified thing, the quantifier
784        is automatically "possessified".
785    
786    24. A recursive reference to a subpattern whose number was greater than 39
787        went wrong under certain circumstances in UTF-8 mode. This bug could also
788        have affected the operation of pcre_study().
789    
790    25. Realized that a little bit of performance could be had by replacing
791        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
792    
793    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
794    
795    27. Possessive quantifiers such as a++ were previously implemented by turning
796        them into atomic groups such as ($>a+). Now they have their own opcodes,
797        which improves performance. This includes the automatically created ones
798        from 23 above.
799    
800    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
801        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
802        the first matched character to be a colon. This applied both to named and
803        numbered groups.
804    
805    29. The ucpinternal.h header file was missing its idempotency #ifdef.
806    
807    30. I was sent a "project" file called libpcre.a.dev which I understand makes
808        building PCRE on Windows easier, so I have included it in the distribution.
809    
810    31. There is now a check in pcretest against a ridiculously large number being
811        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
812        loop, the loop is abandoned.
813    
814    32. Forward references to subpatterns in conditions such as (?(2)...) where
815        subpattern 2 is defined later cause pcre_compile() to search forwards in
816        the pattern for the relevant set of parentheses. This search went wrong
817        when there were unescaped parentheses in a character class, parentheses
818        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
819    
820    33. "Subroutine" calls and backreferences were previously restricted to
821        referencing subpatterns earlier in the regex. This restriction has now
822        been removed.
823    
824    34. Added a number of extra features that are going to be in Perl 5.10. On the
825        whole, these are just syntactic alternatives for features that PCRE had
826        previously implemented using the Python syntax or my own invention. The
827        other formats are all retained for compatibility.
828    
829        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
830            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
831            also .NET compatible.
832    
833        (b) A recursion or subroutine call to a named group can now be defined as
834            (?&name) as well as (?P>name).
835    
836        (c) A backreference to a named group can now be defined as \k<name> or
837            \k'name' as well as (?P=name). The new forms, as well as being in Perl
838            5.10, are also .NET compatible.
839    
840        (d) A conditional reference to a named group can now use the syntax
841            (?(<name>) or (?('name') as well as (?(name).
842    
843        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
844            groups (named and numbered) that are never evaluated inline, but can be
845            called as "subroutines" from elsewhere. In effect, the DEFINE condition
846            is always false. There may be only one alternative in such a group.
847    
848        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
849            as the simple (?(R). The condition is true only if the most recent
850            recursion is that of the given number or name. It does not search out
851            through the entire recursion stack.
852    
853        (g) The escape \gN or \g{N} has been added, where N is a positive or
854            negative number, specifying an absolute or relative reference.
855    
856    35. Tidied to get rid of some further signed/unsigned compiler warnings and
857        some "unreachable code" warnings.
858    
859    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
860        things, this adds five new scripts.
861    
862    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
863        There were also incompatibilities regarding the handling of \Q..\E inside
864        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
865        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
866    
867    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
868        matches an empty string, and forcibly breaks the loop. There were bugs in
869        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
870        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
871        separate and independent bugs (that affected different cases) have been
872        fixed.
873    
874    39. Refactored the code to abolish the use of different opcodes for small
875        capturing bracket numbers. This is a tidy that I avoided doing when I
876        removed the limit on the number of capturing brackets for 3.5 back in 2001.
877        The new approach is not only tidier, it makes it possible to reduce the
878        memory needed to fix the previous bug (38).
879    
880    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
881        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
882        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
883        mode.
884    
885    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
886        report.
887    
888    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
889        copy construction and assignment in the C++ wrapper.
890    
891    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
892        couple of bugs that could have given wrong results in the "--newline=crlf"
893        case.
894    
895    44. Added a number of casts and did some reorganization of signed/unsigned int
896        variables following suggestions from Dair Grant. Also renamed the variable
897        "this" as "item" because it is a C++ keyword.
898    
899    45. Arranged for dftables to add
900    
901          #include "pcre_internal.h"
902    
903        to pcre_chartables.c because without it, gcc 4.x may remove the array
904        definition from the final binary if PCRE is built into a static library and
905        dead code stripping is activated.
906    
907    46. For an unanchored pattern, if a match attempt fails at the start of a
908        newline sequence, and the newline setting is CRLF or ANY, and the next two
909        characters are CRLF, advance by two characters instead of one.
910    
911    
912    Version 6.7 04-Jul-06
913    ---------------------
914    
915     1. In order to handle tests when input lines are enormously long, pcretest has
916        been re-factored so that it automatically extends its buffers when
917        necessary. The code is crude, but this _is_ just a test program. The
918        default size has been increased from 32K to 50K.
919    
920     2. The code in pcre_study() was using the value of the re argument before
921        testing it for NULL. (Of course, in any sensible call of the function, it
922        won't be NULL.)
923    
924     3. The memmove() emulation function in pcre_internal.h, which is used on
925        systems that lack both memmove() and bcopy() - that is, hardly ever -
926        was missing a "static" storage class specifier.
927    
928     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
929        containing an extended class (one that cannot be represented by a bitmap
930        because it contains high-valued characters or Unicode property items, e.g.
931        [\pZ]). Almost always one would set UTF-8 mode when processing such a
932        pattern, but PCRE should not loop if you do not (it no longer does).
933        [Detail: two cases were found: (a) a repeated subpattern containing an
934        extended class; (b) a recursive reference to a subpattern that followed a
935        previous extended class. It wasn't skipping over the extended class
936        correctly when UTF-8 mode was not set.]
937    
938     5. A negated single-character class was not being recognized as fixed-length
939        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
940        compile error "lookbehind assertion is not fixed length".
941    
942     6. The RunPerlTest auxiliary script was showing an unexpected difference
943        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
944        write a Perl script that can interpret lines of an input file either as
945        byte characters or as UTF-8, which is what "perltest" was being required to
946        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
947        can't do is switch easily at run time between having the "use utf8;" pragma
948        or not. In the end, I fudged it by using the RunPerlTest script to insert
949        "use utf8;" explicitly for the UTF-8 tests.
950    
951     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
952        the end of the subject string, contrary to the documentation and to what
953        Perl does. This was true of both matching functions. Now it matches only at
954        the start of the subject and immediately after *internal* newlines.
955    
956     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
957        a pointer to an int instead of a pointer to an unsigned long int. This
958        caused problems on 64-bit systems.
959    
960     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
961        instance of the 'standard' template library not being so standard".
962    
963    10. There was no check on the number of named subpatterns nor the maximum
964        length of a subpattern name. The product of these values is used to compute
965        the size of the memory block for a compiled pattern. By supplying a very
966        long subpattern name and a large number of named subpatterns, the size
967        computation could be caused to overflow. This is now prevented by limiting
968        the length of names to 32 characters, and the number of named subpatterns
969        to 10,000.
970    
971    11. Subpatterns that are repeated with specific counts have to be replicated in
972        the compiled pattern. The size of memory for this was computed from the
973        length of the subpattern and the repeat count. The latter is limited to
974        65535, but there was no limit on the former, meaning that integer overflow
975        could in principle occur. The compiled length of a repeated subpattern is
976        now limited to 30,000 bytes in order to prevent this.
977    
978    12. Added the optional facility to have named substrings with the same name.
979    
980    13. Added the ability to use a named substring as a condition, using the
981        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
982        are numbers (not recommended). Forward references are permitted.
983    
984    14. Added forward references in named backreferences (if you see what I mean).
985    
986    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
987        pattern could run off the end of the subject. For example, the pattern
988        "(?s)(.{1,5})"8 did this with the subject "ab".
989    
990    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
991        PCRE_CASELESS was set when matching characters that were quantified with ?
992        or *.
993    
994    17. A character class other than a single negated character that had a minimum
995        but no maximum quantifier - for example [ab]{6,} - was not handled
996        correctly by pce_dfa_exec(). It would match only one character.
997    
998    18. A valid (though odd) pattern that looked like a POSIX character
999        class but used an invalid character after [ (for example [[,abc,]]) caused
1000        pcre_compile() to give the error "Failed: internal error: code overflow" or
1001        in some cases to crash with a glibc free() error. This could even happen if
1002        the pattern terminated after [[ but there just happened to be a sequence of
1003        letters, a binary zero, and a closing ] in the memory that followed.
1004    
1005    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1006        over the years. Originally (before any Unicode support), just the bottom 8
1007        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1008        output from "man perlunicode" includes this:
1009    
1010          The regular expression compiler produces polymorphic opcodes.  That
1011          is, the pattern adapts to the data and automatically switches to
1012          the Unicode character scheme when presented with Unicode data--or
1013          instead uses a traditional byte scheme when presented with byte
1014          data.
1015    
1016        Sadly, a wide octal escape does not cause a switch, and in a string with
1017        no other multibyte characters, these octal escapes are treated as before.
1018        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
1019        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1020        Unicode string.
1021    
1022        I have not perpetrated such confusion in PCRE. Up till now, it took just
1023        the bottom 8 bits, as in old Perl. I have now made octal escapes with
1024        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1025        translate to the appropriate multibyte character.
1026    
1027    29. Applied some refactoring to reduce the number of warnings from Microsoft
1028        and Borland compilers. This has included removing the fudge introduced
1029        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1030        a warning about an unused variable.
1031    
1032    21. PCRE has not included VT (character 0x0b) in the set of whitespace
1033        characters since release 4.0, because Perl (from release 5.004) does not.
1034        [Or at least, is documented not to: some releases seem to be in conflict
1035        with the documentation.] However, when a pattern was studied with
1036        pcre_study() and all its branches started with \s, PCRE still included VT
1037        as a possible starting character. Of course, this did no harm; it just
1038        caused an unnecessary match attempt.
1039    
1040    22. Removed a now-redundant internal flag bit that recorded the fact that case
1041        dependency changed within the pattern. This was once needed for "required
1042        byte" processing, but is no longer used. This recovers a now-scarce options
1043        bit. Also moved the least significant internal flag bit to the most-
1044        significant bit of the word, which was not previously used (hangover from
1045        the days when it was an int rather than a uint) to free up another bit for
1046        the future.
1047    
1048    23. Added support for CRLF line endings as well as CR and LF. As well as the
1049        default being selectable at build time, it can now be changed at runtime
1050        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1051        specify that it is scanning data with non-default line endings.
1052    
1053    24. Changed the definition of CXXLINK to make it agree with the definition of
1054        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1055    
1056    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1057        recursions. This makes a big different to stack usage for some patterns.
1058    
1059    26. If a subpattern containing a named recursion or subroutine reference such
1060        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1061        the space required for the compiled pattern went wrong and gave too small a
1062        value. Depending on the environment, this could lead to "Failed: internal
1063        error: code overflow at offset 49" or "glibc detected double free or
1064        corruption" errors.
1065    
1066    27. Applied patches from Google (a) to support the new newline modes and (b) to
1067        advance over multibyte UTF-8 characters in GlobalReplace.
1068    
1069    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1070        difference for some implementation of PCRE in some Windows version.
1071    
1072    29. Added some extra testing facilities to pcretest:
1073    
1074        \q<number>   in a data line sets the "match limit" value
1075        \Q<number>   in a data line sets the "match recursion limt" value
1076        -S <number>  sets the stack size, where <number> is in megabytes
1077    
1078        The -S option isn't available for Windows.
1079    
1080    
1081    Version 6.6 06-Feb-06
1082    ---------------------
1083    
1084     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1085        in pcreposix.h. I have copied the definition from pcre.h.
1086    
1087     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1088        because pcre.h is no longer a built file.
1089    
1090     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1091        not normally included in the compiled code.
1092    
1093    
1094    Version 6.5 01-Feb-06
1095    ---------------------
1096    
1097     1. When using the partial match feature with pcre_dfa_exec(), it was not
1098        anchoring the second and subsequent partial matches at the new starting
1099        point. This could lead to incorrect results. For example, with the pattern
1100        /1234/, partially matching against "123" and then "a4" gave a match.
1101    
1102     2. Changes to pcregrep:
1103    
1104        (a) All non-match returns from pcre_exec() were being treated as failures
1105            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1106            error message is output. Some extra information is given for the
1107            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
1108            probably the only errors that are likely to be caused by users (by
1109            specifying a regex that has nested indefinite repeats, for instance).
1110            If there are more than 20 of these errors, pcregrep is abandoned.
1111    
1112        (b) A binary zero was treated as data while matching, but terminated the
1113            output line if it was written out. This has been fixed: binary zeroes
1114            are now no different to any other data bytes.
1115    
1116        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1117            used to set a locale for matching. The --locale=xxxx long option has
1118            been added (no short equivalent) to specify a locale explicitly on the
1119            pcregrep command, overriding the environment variables.
1120    
1121        (d) When -B was used with -n, some line numbers in the output were one less
1122            than they should have been.
1123    
1124        (e) Added the -o (--only-matching) option.
1125    
1126        (f) If -A or -C was used with -c (count only), some lines of context were
1127            accidentally printed for the final match.
1128    
1129        (g) Added the -H (--with-filename) option.
1130    
1131        (h) The combination of options -rh failed to suppress file names for files
1132            that were found from directory arguments.
1133    
1134        (i) Added the -D (--devices) and -d (--directories) options.
1135    
1136        (j) Added the -F (--fixed-strings) option.
1137    
1138        (k) Allow "-" to be used as a file name for -f as well as for a data file.
1139    
1140        (l) Added the --colo(u)r option.
1141    
1142        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1143            is not present by default.
1144    
1145     3. A nasty bug was discovered in the handling of recursive patterns, that is,
1146        items such as (?R) or (?1), when the recursion could match a number of
1147        alternatives. If it matched one of the alternatives, but subsequently,
1148        outside the recursion, there was a failure, the code tried to back up into
1149        the recursion. However, because of the way PCRE is implemented, this is not
1150        possible, and the result was an incorrect result from the match.
1151    
1152        In order to prevent this happening, the specification of recursion has
1153        been changed so that all such subpatterns are automatically treated as
1154        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1155    
1156     4. I had overlooked the fact that, in some locales, there are characters for
1157        which isalpha() is true but neither isupper() nor islower() are true. In
1158        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1159        and ordfeminine) are like this. This affected the treatment of \w and \W
1160        when they appeared in character classes, but not when they appeared outside
1161        a character class. The bit map for "word" characters is now created
1162        separately from the results of isalnum() instead of just taking it from the
1163        upper, lower, and digit maps. (Plus the underscore character, of course.)
1164    
1165     5. The above bug also affected the handling of POSIX character classes such as
1166        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1167        permanent tables. Instead, the bit maps for such a class were previously
1168        created as the appropriate unions of the upper, lower, and digit bitmaps.
1169        Now they are created by subtraction from the [[:word:]] class, which has
1170        its own bitmap.
1171    
1172     6. The [[:blank:]] character class matches horizontal, but not vertical space.
1173        It is created by subtracting the vertical space characters (\x09, \x0a,
1174        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1175        subtraction was done in the overall bitmap for a character class, meaning
1176        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1177        be recognized. This bug has been fixed.
1178    
1179     7. Patches from the folks at Google:
1180    
1181          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1182          real life, but is still worth protecting against".
1183    
1184          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1185          regular expressions".
1186    
1187          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1188          have it.
1189    
1190          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1191          "configure" and the latter not, in order to fix a problem somebody had
1192          with compiling the Arg class on HP-UX.
1193    
1194          (e) Improve the error-handling of the C++ wrapper a little bit.
1195    
1196          (f) New tests for checking recursion limiting.
1197    
1198     8. The pcre_memmove() function, which is used only if the environment does not
1199        have a standard memmove() function (and is therefore rarely compiled),
1200        contained two bugs: (a) use of int instead of size_t, and (b) it was not
1201        returning a result (though PCRE never actually uses the result).
1202    
1203     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1204        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1205        returned instead of calling malloc() with an overflowing number that would
1206        most likely cause subsequent chaos.
1207    
1208    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1209    
1210    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1211        with this option is matched, the nmatch and pmatch options of regexec() are
1212        ignored.
1213    
1214    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1215        provided in case anyone wants to the the POSIX interface with UTF-8
1216        strings.
1217    
1218    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1219        C++ linking (needed for some HP-UX environments).
1220    
1221    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1222        (unused parameter) and in the pcre_printint() function (omitted "default"
1223        switch label when the default is to do nothing).
1224    
1225    15. Added some code to make it possible, when PCRE is compiled as a C++
1226        library, to replace subject pointers for pcre_exec() with a smart pointer
1227        class, thus making it possible to process discontinuous strings.
1228    
1229    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1230        much the same function. They were added by different people who were trying
1231        to make PCRE easy to compile on non-Unix systems. It has been suggested
1232        that PCRE_EXPORT be abolished now that there is more automatic apparatus
1233        for compiling on Windows systems. I have therefore replaced it with
1234        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1235        defaults to "extern" for C or "extern C" for C++, which works fine on
1236        Unix-like systems. It is now possible to override the value of PCRE_DATA_
1237        SCOPE with something explicit in config.h. In addition:
1238    
1239        (a) pcreposix.h still had just "extern" instead of either of these macros;
1240            I have replaced it with PCRE_DATA_SCOPE.
1241    
1242        (b) Functions such as _pcre_xclass(), which are internal to the library,
1243            but external in the C sense, all had PCRE_EXPORT in their definitions.
1244            This is apparently wrong for the Windows case, so I have removed it.
1245            (It makes no difference on Unix-like systems.)
1246    
1247    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1248        of recursive calls to match(). This is different to MATCH_LIMIT because
1249        that limits the total number of calls to match(), not all of which increase
1250        the depth of recursion. Limiting the recursion depth limits the amount of
1251        stack (or heap if NO_RECURSE is set) that is used. The default can be set
1252        when PCRE is compiled, and changed at run time. A patch from Google adds
1253        this functionality to the C++ interface.
1254    
1255    18. Changes to the handling of Unicode character properties:
1256    
1257        (a) Updated the table to Unicode 4.1.0.
1258    
1259        (b) Recognize characters that are not in the table as "Cn" (undefined).
1260    
1261        (c) I revised the way the table is implemented to a much improved format
1262            which includes recognition of ranges. It now supports the ranges that
1263            are defined in UnicodeData.txt, and it also amalgamates other
1264            characters into ranges. This has reduced the number of entries in the
1265            table from around 16,000 to around 3,000, thus reducing its size
1266            considerably. I realized I did not need to use a tree structure after
1267            all - a binary chop search is just as efficient. Having reduced the
1268            number of entries, I extended their size from 6 bytes to 8 bytes to
1269            allow for more data.
1270    
1271        (d) Added support for Unicode script names via properties such as \p{Han}.
1272    
1273    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1274        matching that character.
1275    
1276    20. When matching a repeated Unicode property with a minimum greater than zero,
1277        (for example \pL{2,}), PCRE could look past the end of the subject if it
1278        reached it while seeking the minimum number of characters. This could
1279        happen only if some of the characters were more than one byte long, because
1280        there is a check for at least the minimum number of bytes.
1281    
1282    21. Refactored the implementation of \p and \P so as to be more general, to
1283        allow for more different types of property in future. This has changed the
1284        compiled form incompatibly. Anybody with saved compiled patterns that use
1285        \p or \P will have to recompile them.
1286    
1287    22. Added "Any" and "L&" to the supported property types.
1288    
1289    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1290        but give a compile time error if the value is greater than 0xff.
1291    
1292    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1293        accidentally not being installed or uninstalled.
1294    
1295    25. The pcre.h file was built from pcre.h.in, but the only changes that were
1296        made were to insert the current release number. This seemed silly, because
1297        it made things harder for people building PCRE on systems that don't run
1298        "configure". I have turned pcre.h into a distributed file, no longer built
1299        by "configure", with the version identification directly included. There is
1300        no longer a pcre.h.in file.
1301    
1302        However, this change necessitated a change to the pcre-config script as
1303        well. It is built from pcre-config.in, and one of the substitutions was the
1304        release number. I have updated configure.ac so that ./configure now finds
1305        the release number by grepping pcre.h.
1306    
1307    26. Added the ability to run the tests under valgrind.
1308    
1309    
1310    Version 6.4 05-Sep-05
1311    ---------------------
1312    
1313     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1314        "--" to be printed when multiple files were scanned, even when none of the
1315        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1316        consider it to be a bug, and have restored the previous behaviour.
1317    
1318     2. A couple of code tidies to get rid of compiler warnings.
1319    
1320     3. The pcretest program used to cheat by referring to symbols in the library
1321        whose names begin with _pcre_. These are internal symbols that are not
1322        really supposed to be visible externally, and in some environments it is
1323        possible to suppress them. The cheating is now confined to including
1324        certain files from the library's source, which is a bit cleaner.
1325    
1326     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1327        file's purpose clearer.
1328    
1329     5. Reorganized pcre_ucp_findchar().
1330    
1331    
1332    Version 6.3 15-Aug-05
1333    ---------------------
1334    
1335     1. The file libpcre.pc.in did not have general read permission in the tarball.
1336    
1337     2. There were some problems when building without C++ support:
1338    
1339        (a) If C++ support was not built, "make install" and "make test" still
1340            tried to test it.
1341    
1342        (b) There were problems when the value of CXX was explicitly set. Some
1343            changes have been made to try to fix these, and ...
1344    
1345        (c) --disable-cpp can now be used to explicitly disable C++ support.
1346    
1347        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1348            backslash in a target when C++ was disabled. This confuses some
1349            versions of "make", apparently. Using an intermediate variable solves
1350            this. (Same for CPP_LOBJ.)
1351    
1352     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1353        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1354        necessary on certain architectures.
1355    
1356     4. Added a setting of -export-symbols-regex to the link command to remove
1357        those symbols that are exported in the C sense, but actually are local
1358        within the library, and not documented. Their names all begin with
1359        "_pcre_". This is not a perfect job, because (a) we have to except some
1360        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1361        available (and never for static libraries). I have made a note to try to
1362        find a way round (a) in the future.
1363    
1364    
1365    Version 6.2 01-Aug-05
1366    ---------------------
1367    
1368     1. There was no test for integer overflow of quantifier values. A construction
1369        such as {1111111111111111} would give undefined results. What is worse, if
1370        a minimum quantifier for a parenthesized subpattern overflowed and became
1371        negative, the calculation of the memory size went wrong. This could have
1372        led to memory overwriting.
1373    
1374     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1375    
1376     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1377        operating environments where this matters.
1378    
1379     4. Applied Giuseppe Maxia's patch to add additional features for controlling
1380        PCRE options from within the C++ wrapper.
1381    
1382     5. Named capturing subpatterns were not being correctly counted when a pattern
1383        was compiled. This caused two problems: (a) If there were more than 100
1384        such subpatterns, the calculation of the memory needed for the whole
1385        compiled pattern went wrong, leading to an overflow error. (b) Numerical
1386        back references of the form \12, where the number was greater than 9, were
1387        not recognized as back references, even though there were sufficient
1388        previous subpatterns.
1389    
1390     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1391        versions of gcc, e.g. 2.95.4.
1392    
1393    
1394    Version 6.1 21-Jun-05
1395    ---------------------
1396    
1397     1. There was one reference to the variable "posix" in pcretest.c that was not
1398        surrounded by "#if !defined NOPOSIX".
1399    
1400     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1401        the cross-check on the old pcre_info() function, for the benefit of the
1402        cut-down version of PCRE that is currently imported into Exim.
1403    
1404     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1405        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1406        patterns that start (?i) but I don't think that matters. The use of (?i) is
1407        just an example; this all applies to the other options as well.
1408    
1409     4. Since libtool seems to echo the compile commands it is issuing, the output
1410        from "make" can be reduced a bit by putting "@" in front of each libtool
1411        compile command.
1412    
1413     5. Patch from the folks at Google for configure.in to be a bit more thorough
1414        in checking for a suitable C++ installation before trying to compile the
1415        C++ stuff. This should fix a reported problem when a compiler was present,
1416        but no suitable headers.
1417    
1418     6. The man pages all had just "PCRE" as their title. I have changed them to
1419        be the relevant file name. I have also arranged that these names are
1420        retained in the file doc/pcre.txt, which is a concatenation in text format
1421        of all the man pages except the little individual ones for each function.
1422    
1423     7. The NON-UNIX-USE file had not been updated for the different set of source
1424        files that come with release 6. I also added a few comments about the C++
1425        wrapper.
1426    
1427    
1428    Version 6.0 07-Jun-05
1429    ---------------------
1430    
1431     1. Some minor internal re-organization to help with my DFA experiments.
1432    
1433     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1434        didn't matter for the library itself when fully configured, but did matter
1435        when compiling without UCP support, or within Exim, where the ucp files are
1436        not imported.
1437    
1438     3. Refactoring of the library code to split up the various functions into
1439        different source modules. The addition of the new DFA matching code (see
1440        below) to a single monolithic source would have made it really too
1441        unwieldy, quite apart from causing all the code to be include in a
1442        statically linked application, when only some functions are used. This is
1443        relevant even without the DFA addition now that patterns can be compiled in
1444        one application and matched in another.
1445    
1446        The downside of splitting up is that there have to be some external
1447        functions and data tables that are used internally in different modules of
1448        the library but which are not part of the API. These have all had their
1449        names changed to start with "_pcre_" so that they are unlikely to clash
1450        with other external names.
1451    
1452     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1453        a different (DFA) algorithm. Although it is slower than the original
1454        function, it does have some advantages for certain types of matching
1455        problem.
1456    
1457     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1458        including restarting after a partial match.
1459    
1460     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1461        defined when compiling for Windows was sent to me. I have put it into the
1462        code, though I have no means of testing or verifying it.
1463    
1464     7. Added the pcre_refcount() auxiliary function.
1465    
1466     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1467        match before or at the first newline in the subject string. In pcretest,
1468        the /f option on a pattern can be used to set this.
1469    
1470     9. A repeated \w when used in UTF-8 mode with characters greater than 256
1471        would behave wrongly. This has been present in PCRE since release 4.0.
1472    
1473    10. A number of changes to the pcregrep command:
1474    
1475        (a) Refactored how -x works; insert ^(...)$ instead of setting
1476            PCRE_ANCHORED and checking the length, in preparation for adding
1477            something similar for -w.
1478    
1479        (b) Added the -w (match as a word) option.
1480    
1481        (c) Refactored the way lines are read and buffered so as to have more
1482            than one at a time available.
1483    
1484        (d) Implemented a pcregrep test script.
1485    
1486        (e) Added the -M (multiline match) option. This allows patterns to match
1487            over several lines of the subject. The buffering ensures that at least
1488            8K, or the rest of the document (whichever is the shorter) is available
1489            for matching (and similarly the previous 8K for lookbehind assertions).
1490    
1491        (f) Changed the --help output so that it now says
1492    
1493              -w, --word-regex(p)
1494    
1495            instead of two lines, one with "regex" and the other with "regexp"
1496            because that confused at least one person since the short forms are the
1497            same. (This required a bit of code, as the output is generated
1498            automatically from a table. It wasn't just a text change.)
1499    
1500        (g) -- can be used to terminate pcregrep options if the next thing isn't an
1501            option but starts with a hyphen. Could be a pattern or a path name
1502            starting with a hyphen, for instance.
1503    
1504        (h) "-" can be given as a file name to represent stdin.
1505    
1506        (i) When file names are being printed, "(standard input)" is used for
1507            the standard input, for compatibility with GNU grep. Previously
1508            "<stdin>" was used.
1509    
1510        (j) The option --label=xxx can be used to supply a name to be used for
1511            stdin when file names are being printed. There is no short form.
1512    
1513        (k) Re-factored the options decoding logic because we are going to add
1514            two more options that take data. Such options can now be given in four
1515            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1516    
1517        (l) Added the -A, -B, and -C options for requesting that lines of context
1518            around matches be printed.
1519    
1520        (m) Added the -L option to print the names of files that do not contain
1521            any matching lines, that is, the complement of -l.
1522    
1523        (n) The return code is 2 if any file cannot be opened, but pcregrep does
1524            continue to scan other files.
1525    
1526        (o) The -s option was incorrectly implemented. For compatibility with other
1527            greps, it now suppresses the error message for a non-existent or non-
1528            accessible file (but not the return code). There is a new option called
1529            -q that suppresses the output of matching lines, which was what -s was
1530            previously doing.
1531    
1532        (p) Added --include and --exclude options to specify files for inclusion
1533            and exclusion when recursing.
1534    
1535    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1536        Hopefully, it now does.
1537    
1538    12. Missing cast in pcre_study().
1539    
1540    13. Added an "uninstall" target to the makefile.
1541    
1542    14. Replaced "extern" in the function prototypes in Makefile.in with
1543        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1544        world, but is set differently for Windows.
1545    
1546    15. Added a second compiling function called pcre_compile2(). The only
1547        difference is that it has an extra argument, which is a pointer to an
1548        integer error code. When there is a compile-time failure, this is set
1549        non-zero, in addition to the error test pointer being set to point to an
1550        error message. The new argument may be NULL if no error number is required
1551        (but then you may as well call pcre_compile(), which is now just a
1552        wrapper). This facility is provided because some applications need a
1553        numeric error indication, but it has also enabled me to tidy up the way
1554        compile-time errors are handled in the POSIX wrapper.
1555    
1556    16. Added VPATH=.libs to the makefile; this should help when building with one
1557        prefix path and installing with another. (Or so I'm told by someone who
1558        knows more about this stuff than I do.)
1559    
1560    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1561        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1562        match everything, including newlines. This is not POSIX-compatible, but
1563        somebody wanted the feature. From pcretest it can be activated by using
1564        both the P and the s flags.
1565    
1566    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1567    
1568    19. libpcre.pc was being incorrectly installed as executable.
1569    
1570    20. A couple of places in pcretest check for end-of-line by looking for '\n';
1571        it now also looks for '\r' so that it will work unmodified on Windows.
1572    
1573    21. Added Google's contributed C++ wrapper to the distribution.
1574    
1575    22. Added some untidy missing memory free() calls in pcretest, to keep
1576        Electric Fence happy when testing.
1577    
1578    
1579    
1580    Version 5.0 13-Sep-04
1581    ---------------------
1582    
1583     1. Internal change: literal characters are no longer packed up into items
1584        containing multiple characters in a single byte-string. Each character
1585        is now matched using a separate opcode. However, there may be more than one
1586        byte in the character in UTF-8 mode.
1587    
1588     2. The pcre_callout_block structure has two new fields: pattern_position and
1589        next_item_length. These contain the offset in the pattern to the next match
1590        item, and its length, respectively.
1591    
1592     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1593        insertion of callouts before each pattern item. Added the /C option to
1594        pcretest to make use of this.
1595    
1596     4. On the advice of a Windows user, the lines
1597    
1598          #if defined(_WIN32) || defined(WIN32)
1599          _setmode( _fileno( stdout ), 0x8000 );
1600          #endif  /* defined(_WIN32) || defined(WIN32) */
1601    
1602        have been added to the source of pcretest. This apparently does useful
1603        magic in relation to line terminators.
1604    
1605     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1606        for the benefit of those environments where the "b" makes a difference.
1607    
1608     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1609        to know about it. I have put a hack into configure.in that adds in code
1610        to set GCC=yes if CC=icc. This seems to end up at a point in the
1611        generated configure script that is early enough to affect the setting of
1612        compiler options, which is what is needed, but I have no means of testing
1613        whether it really works. (The user who reported this had patched the
1614        generated configure script, which of course I cannot do.)
1615    
1616        LATER: After change 22 below (new libtool files), the configure script
1617        seems to know about icc (and also ecc). Therefore, I have commented out
1618        this hack in configure.in.
1619    
1620     7. Added support for pkg-config (2 patches were sent in).
1621    
1622     8. Negated POSIX character classes that used a combination of internal tables
1623        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1624        [[:^ascii]]. Typically, they would match almost any characters. The other
1625        POSIX classes were not broken in this way.
1626    
1627     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1628        to find the match, as PCRE was deluded into thinking that the match had to
1629        start at the start point or following a newline. The same bug applied to
1630        patterns with negative forward assertions or any backward assertions
1631        preceding ".*" at the start, unless the pattern required a fixed first
1632        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1633    
1634    10. In UTF-8 mode, when moving forwards in the subject after a failed match
1635        starting at the last subject character, bytes beyond the end of the subject
1636        string were read.
1637    
1638    11. Renamed the variable "class" as "classbits" to make life easier for C++
1639        users. (Previously there was a macro definition, but it apparently wasn't
1640        enough.)
1641    
1642    12. Added the new field "tables" to the extra data so that tables can be passed
1643        in at exec time, or the internal tables can be re-selected. This allows
1644        a compiled regex to be saved and re-used at a later time by a different
1645        program that might have everything at different addresses.
1646    
1647    13. Modified the pcre-config script so that, when run on Solaris, it shows a
1648        -R library as well as a -L library.
1649    
1650    14. The debugging options of pcretest (-d on the command line or D on a
1651        pattern) showed incorrect output for anything following an extended class
1652        that contained multibyte characters and which was followed by a quantifier.
1653    
1654    15. Added optional support for general category Unicode character properties
1655        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1656        support. It adds about 90K to the size of the library. The meanings of the
1657        inbuilt class escapes such as \d and \s have NOT been changed.
1658    
1659    16. Updated pcredemo.c to include calls to free() to release the memory for the
1660        compiled pattern.
1661    
1662    17. The generated file chartables.c was being created in the source directory
1663        instead of in the building directory. This caused the build to fail if the
1664        source directory was different from the building directory, and was
1665        read-only.
1666    
1667    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1668        file. No doubt somebody will tell me if they don't make sense... Also added
1669        Dan Mooney's comments about building on OpenVMS.
1670    
1671    19. Added support for partial matching via the PCRE_PARTIAL option for
1672        pcre_exec() and the \P data escape in pcretest.
1673    
1674    20. Extended pcretest with 3 new pattern features:
1675    
1676        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
1677              write the compiled pattern to the file whose name is "rest-of-line".
1678              This is a straight binary dump of the data, with the saved pointer to
1679              the character tables forced to be NULL. The study data, if any, is
1680              written too. After writing, pcretest reads a new pattern.
1681    
1682        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1683              compiled pattern from the given file. There must not be any
1684              occurrences of "<" in the file name (pretty unlikely); if there are,
1685              pcretest will instead treat the initial "<" as a pattern delimiter.
1686              After reading in the pattern, pcretest goes on to read data lines as
1687              usual.
1688    
1689        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1690              and 16-bit fields in a compiled pattern, to simulate a pattern that
1691              was compiled on a host of opposite endianness.
1692    
1693    21. The pcre-exec() function can now cope with patterns that were compiled on
1694        hosts of opposite endianness, with this restriction:
1695    
1696          As for any compiled expression that is saved and used later, the tables
1697          pointer field cannot be preserved; the extra_data field in the arguments
1698          to pcre_exec() should be used to pass in a tables address if a value
1699          other than the default internal tables were used at compile time.
1700    
1701    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1702        now diagnosed as an error. Previously, most of the time, a negative number
1703        would have been treated as zero, but if in addition "ovector" was passed as
1704        NULL, a crash could occur.
1705    
1706    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1707        new versions from the libtool 1.5 distribution (the last one is a copy of
1708        a file called libtool.m4). This seems to have fixed the need to patch
1709        "configure" to support Darwin 1.3 (which I used to do). However, I still
1710        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1711        workstation).
1712    
1713    24. Changed the PCRE licence to be the more standard "BSD" licence.
1714    
1715    
1716    Version 4.5 01-Dec-03
1717    ---------------------
1718    
1719     1. There has been some re-arrangement of the code for the match() function so
1720        that it can be compiled in a version that does not call itself recursively.
1721        Instead, it keeps those local variables that need separate instances for
1722        each "recursion" in a frame on the heap, and gets/frees frames whenever it
1723        needs to "recurse". Keeping track of where control must go is done by means
1724        of setjmp/longjmp. The whole thing is implemented by a set of macros that
1725        hide most of the details from the main code, and operates only if
1726        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1727        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1728        operating.
1729    
1730        To make it easier for callers to provide specially tailored get/free
1731        functions for this usage, two new functions, pcre_stack_malloc, and
1732        pcre_stack_free, are used. They are always called in strict stacking order,
1733        and the size of block requested is always the same.
1734    
1735        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1736        PCRE has been compiled to use the stack or the heap for recursion. The
1737        -C option of pcretest uses this to show which version is compiled.
1738    
1739        A new data escape \S, is added to pcretest; it causes the amounts of store
1740        obtained and freed by both kinds of malloc/free at match time to be added
1741        to the output.
1742    
1743     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1744        what's available on my current Linux desktop machine.
1745    
1746     3. When matching a UTF-8 string, the test for a valid string at the start has
1747        been extended. If start_offset is not zero, PCRE now checks that it points
1748        to a byte that is the start of a UTF-8 character. If not, it returns
1749        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1750        this is necessary because there may be backward assertions in the pattern.
1751        When matching the same subject several times, it may save resources to use
1752        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1753    
1754     4. The code for checking the validity of UTF-8 strings has been tightened so
1755        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1756        containing "overlong sequences".
1757    
1758     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1759        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1760        should have been "&", but it just so happened that all the cases this let
1761        through by mistake were picked up later in the function.
1762    
1763     6. I had used a variable called "isblank" - this is a C99 function, causing
1764        some compilers to warn. To avoid this, I renamed it (as "blankclass").
1765    
1766     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1767        prompting; (b) run "./pcretest /dev/null" at the start of the test script
1768        so the version is shown; (c) stop "make test" echoing "./RunTest".
1769    
1770     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1771    
1772     9. The prototype for memmove() for systems that don't have it was using
1773        size_t, but the inclusion of the header that defines size_t was later. I've
1774        moved the #includes for the C headers earlier to avoid this.
1775    
1776    10. Added some adjustments to the code to make it easier to compiler on certain
1777        special systems:
1778    
1779          (a) Some "const" qualifiers were missing.
1780          (b) Added the macro EXPORT before all exported functions; by default this
1781              is defined to be empty.
1782          (c) Changed the dftables auxiliary program (that builds chartables.c) so
1783              that it reads its output file name as an argument instead of writing
1784              to the standard output and assuming this can be redirected.
1785    
1786    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
1787        class containing characters with values greater than 255, PCRE compilation
1788        went into a loop.
1789    
1790    12. A recursive reference to a subpattern that was within another subpattern
1791        that had a minimum quantifier of zero caused PCRE to crash. For example,
1792        (x(y(?2))z)? provoked this bug with a subject that got as far as the
1793        recursion. If the recursively-called subpattern itself had a zero repeat,
1794        that was OK.
1795    
1796    13. In pcretest, the buffer for reading a data line was set at 30K, but the
1797        buffer into which it was copied (for escape processing) was still set at
1798        1024, so long lines caused crashes.
1799    
1800    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
1801        "internal error: code overflow...". This applied to any character class
1802        that was followed by a possessive quantifier.
1803    
1804    15. Modified the Makefile to add libpcre.la as a prerequisite for
1805        libpcreposix.la because I was told this is needed for a parallel build to
1806        work.
1807    
1808    16. If a pattern that contained .* following optional items at the start was
1809        studied, the wrong optimizing data was generated, leading to matching
1810        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
1811        matching string must start with a or b or c. The correct conclusion for
1812        this pattern is that a match can start with any character.
1813    
1814    
1815    Version 4.4 13-Aug-03
1816    ---------------------
1817    
1818     1. In UTF-8 mode, a character class containing characters with values between
1819        127 and 255 was not handled correctly if the compiled pattern was studied.
1820        In fixing this, I have also improved the studying algorithm for such
1821        classes (slightly).
1822    
1823     2. Three internal functions had redundant arguments passed to them. Removal
1824        might give a very teeny performance improvement.
1825    
1826     3. Documentation bug: the value of the capture_top field in a callout is *one
1827        more than* the number of the hightest numbered captured substring.
1828    
1829     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
1830        in incorrectly linking with a previously installed version. They now link
1831        explicitly with libpcre.la.
1832    
1833     5. configure.in no longer needs to recognize Cygwin specially.
1834    
1835     6. A problem in pcre.in for Windows platforms is fixed.
1836    
1837     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
1838        pcretest, it used to include the size of the study block as part of its
1839        output. Unfortunately, the structure contains a field that has a different
1840        size on different hardware architectures. This meant that the tests that
1841        showed this size failed. As the block is currently always of a fixed size,
1842        this information isn't actually particularly useful in pcretest output, so
1843        I have just removed it.
1844    
1845     8. Three pre-processor statements accidentally did not start in column 1.
1846        Sadly, there are *still* compilers around that complain, even though
1847        standard C has not required this for well over a decade. Sigh.
1848    
1849     9. In pcretest, the code for checking callouts passed small integers in the
1850        callout_data field, which is a void * field. However, some picky compilers
1851        complained about the casts involved for this on 64-bit systems. Now
1852        pcretest passes the address of the small integer instead, which should get
1853        rid of the warnings.
1854    
1855    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
1856        both compile and run time, and gives an error if an invalid UTF-8 sequence
1857        is found. There is a option for disabling this check in cases where the
1858        string is known to be correct and/or the maximum performance is wanted.
1859    
1860    11. In response to a bug report, I changed one line in Makefile.in from
1861    
1862            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
1863        to
1864            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
1865    
1866        to look similar to other lines, but I have no way of telling whether this
1867        is the right thing to do, as I do not use Windows. No doubt I'll get told
1868        if it's wrong...
1869    
1870    
1871    Version 4.3 21-May-03
1872    ---------------------
1873    
1874    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
1875       Makefile.
1876    
1877    2. Some refactoring to improve the quality of the code:
1878    
1879       (i)   The utf8_table... variables are now declared "const".
1880    
1881       (ii)  The code for \cx, which used the "case flipping" table to upper case
1882             lower case letters, now just substracts 32. This is ASCII-specific,
1883             but the whole concept of \cx is ASCII-specific, so it seems
1884             reasonable.
1885    
1886       (iii) PCRE was using its character types table to recognize decimal and
1887             hexadecimal digits in the pattern. This is silly, because it handles
1888             only 0-9, a-f, and A-F, but the character types table is locale-
1889             specific, which means strange things might happen. A private
1890             table is now used for this - though it costs 256 bytes, a table is
1891             much faster than multiple explicit tests. Of course, the standard
1892             character types table is still used for matching digits in subject
1893             strings against \d.
1894    
1895       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
1896             ending in _t are). So I've renamed it as ESC_tee.
1897    
1898    3. The first argument for regexec() in the POSIX wrapper should have been
1899       defined as "const".
1900    
1901    4. Changed pcretest to use malloc() for its buffers so that they can be
1902       Electric Fenced for debugging.
1903    
1904    5. There were several places in the code where, in UTF-8 mode, PCRE would try
1905       to read one or more bytes before the start of the subject string. Often this
1906       had no effect on PCRE's behaviour, but in some circumstances it could
1907       provoke a segmentation fault.
1908    
1909    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
1910       to try to read one or more bytes before the start of the subject string.
1911    
1912    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
1913       UTF-8 support could misbehave in various ways if the subject string
1914       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
1915       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
1916       back over UTF-8 characters.)
1917    
1918    
1919  Version 4.2 14-Apr-03  Version 4.2 14-Apr-03
1920  ---------------------  ---------------------
1921    

Legend:
Removed from v.67  
changed lines
  Added in v.311

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12