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

Diff of /code/trunk/ChangeLog

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

revision 51 by nigel, Sat Feb 24 21:39:37 2007 UTC revision 356 by ph10, Tue Jul 8 14:14:34 2008 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4    Version 8.0 02 Jul-08
5    ---------------------
6    
7    1.  Replaced UCP searching code with optimized version as implemented for Ad
8        Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
9        stage table and inline lookup instead of a function, giving speed ups of 2
10        to 5 times on some simple patterns that I tested. Permission was given to
11        distribute the MultiStage2.py script that generates the tables (it's not in
12        the tarball, but is in the Subversion repository).
13    
14    2.  Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
15        scripts.
16    
17    3.  Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
18        a group with a zero qualifier. The result of the study could be incorrect,
19        or the function might crash, depending on the pattern.
20    
21    4.  Caseless matching was not working for non-ASCII characters in back
22        references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
23        It now works when Unicode Property Support is available.
24    
25    5.  In pcretest, an escape such as \x{de} in the data was always generating
26        a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
27        non-UTF-8 mode. If the value is greater than 255, it gives a warning about
28        truncation.
29    
30    6.  Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
31    
32    
33    Version 7.7 07-May-08
34    ---------------------
35    
36    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
37        a string to a long long, pretend we don't even have a long long." This is
38        done by checking for the strtoq, strtoll, and _strtoi64 functions.
39    
40    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
41        pre-7.6 versions, which defined a global no_arg variable instead of putting
42        it in the RE class. (See also #8 below.)
43    
44    3.  Remove a line of dead code, identified by coverity and reported by Nuno
45        Lopes.
46    
47    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
48    
49        (1) The include/exclude patterns were being applied to the whole pathnames
50            of files, instead of just to the final components.
51    
52        (2) If there was more than one level of directory, the subdirectories were
53            skipped unless they satisfied the include/exclude conditions. This is
54            inconsistent with GNU grep (and could even be seen as contrary to the
55            pcregrep specification - which I improved to make it absolutely clear).
56            The action now is always to scan all levels of directory, and just
57            apply the include/exclude patterns to regular files.
58    
59    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
60        --exclude_dir in the tests to avoid scanning .svn directories.
61    
62    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
63        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
64        doesn't support NULs in patterns.
65    
66    7.  Added some missing "const"s to declarations of static tables in
67        pcre_compile.c and pcre_dfa_exec.c.
68    
69    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
70        caused by fix #2  above. (Subsequently also a second patch to fix the
71        first patch. And a third patch - this was a messy problem.)
72    
73    9.  Applied Craig's patch to remove the use of push_back().
74    
75    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
76        matching function regexec().
77    
78    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
79        which, however, unlike Perl's \g{...}, are subroutine calls, not back
80        references. PCRE supports relative numbers with this syntax (I don't think
81        Oniguruma does).
82    
83    12. Previously, a group with a zero repeat such as (...){0} was completely
84        omitted from the compiled regex. However, this means that if the group
85        was called as a subroutine from elsewhere in the pattern, things went wrong
86        (an internal error was given). Such groups are now left in the compiled
87        pattern, with a new opcode that causes them to be skipped at execution
88        time.
89    
90    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
91        to the way PCRE behaves:
92    
93        (a) A lone ] character is dis-allowed (Perl treats it as data).
94    
95        (b) A back reference to an unmatched subpattern matches an empty string
96            (Perl fails the current match path).
97    
98        (c) A data ] in a character class must be notated as \] because if the
99            first data character in a class is ], it defines an empty class. (In
100            Perl it is not possible to have an empty class.) The empty class []
101            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
102            The negative empty class [^] matches any one character, independently
103            of the DOTALL setting.
104    
105    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
106        non-existent subpattern following a character class starting with ']' and
107        containing () gave an internal compiling error instead of "reference to
108        non-existent subpattern". Fortunately, when the pattern did exist, the
109        compiled code was correct. (When scanning forwards to check for the
110        existencd of the subpattern, it was treating the data ']' as terminating
111        the class, so got the count wrong. When actually compiling, the reference
112        was subsequently set up correctly.)
113    
114    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
115        it was being rejected as not supported by pcre_dfa_exec(), even though
116        other assertions are supported. I have made pcre_dfa_exec() support
117        (*FAIL).
118    
119    16. The implementation of 13c above involved the invention of a new opcode,
120        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
121        cannot be changed at match time, I realized I could make a small
122        improvement to matching performance by compiling OP_ALLANY instead of
123        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
124        on the OP_ANY path.
125    
126    17. Compiling pcretest on Windows with readline support failed without the
127        following two fixes: (1) Make the unistd.h include conditional on
128        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
129    
130    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
131        ncurses library to be included for pcretest when ReadLine support is
132        requested, but also to allow for it to be overridden. This patch came from
133        Daniel Bergström.
134    
135    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
136        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
137        any errors with the current Unicode tables. Thanks to Peter Kankowski for
138        spotting this.
139    
140    
141    Version 7.6 28-Jan-08
142    ---------------------
143    
144    1.  A character class containing a very large number of characters with
145        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
146        overflow.
147    
148    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
149        HAVE_LONG_LONG is not defined.
150    
151    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
152        bring them up to date and include new features. This patch includes:
153    
154        - Fixed PH's badly added libz and libbz2 support.
155        - Fixed a problem with static linking.
156        - Added pcredemo. [But later removed - see 7 below.]
157        - Fixed dftables problem and added an option.
158        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
159            HAVE_LONG_LONG.
160        - Added readline support for pcretest.
161        - Added an listing of the option settings after cmake has run.
162    
163    4.  A user submitted a patch to Makefile that makes it easy to create
164        "pcre.dll" under mingw when using Configure/Make. I added stuff to
165        Makefile.am that cause it to include this special target, without
166        affecting anything else. Note that the same mingw target plus all
167        the other distribution libraries and programs are now supported
168        when configuring with CMake (see 6 below) instead of with
169        Configure/Make.
170    
171    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
172        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
173        exported in the Windows port". It has not yet been confirmed that the patch
174        solves the problem, but it does no harm.
175    
176    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
177        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
178        with CMake, and also correct the comment about stack recursion.
179    
180    7.  Remove the automatic building of pcredemo from the ./configure system and
181        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
182        of a program that users should build themselves after PCRE is installed, so
183        building it automatically is not really right. What is more, it gave
184        trouble in some build environments.
185    
186    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
187    
188    
189    Version 7.5 10-Jan-08
190    ---------------------
191    
192    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
193        values in parens when parsing an RE using the C++ wrapper."
194    
195    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
196        Characters greater than 255 were excluded from the class instead of being
197        included.
198    
199    3.  The same bug as (2) above applied to negated POSIX classes such as
200        [:^space:].
201    
202    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
203        defined or documented. It seems to have been a typo for PCRE_STATIC, so
204        I have changed it.
205    
206    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
207        first named subpattern) and a construct such as (?&a) would reference the
208        first named subpattern whose name started with "a" (in other words, the
209        length check was missing). Both these problems are fixed. "Subpattern name
210        expected" is now given for (?&) (a zero-length name), and this patch also
211        makes it give the same error for \k'' (previously it complained that that
212        was a reference to a non-existent subpattern).
213    
214    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
215        this is right because (?- can be followed by option settings as well as by
216        digits. I have, however, made the messages clearer.
217    
218    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
219        than the number used in the conditional) now cause a compile-time error.
220        This is actually not compatible with Perl, which accepts such patterns, but
221        treats the conditional as always being FALSE (as PCRE used to), but it
222        seems to me that giving a diagnostic is better.
223    
224    8.  Change "alphameric" to the more common word "alphanumeric" in comments
225        and messages.
226    
227    9.  Fix two occurrences of "backslash" in comments that should have been
228        "backspace".
229    
230    10. Remove two redundant lines of code that can never be obeyed (their function
231        was moved elsewhere).
232    
233    11. The program that makes PCRE's Unicode character property table had a bug
234        which caused it to generate incorrect table entries for sequences of
235        characters that have the same character type, but are in different scripts.
236        It amalgamated them into a single range, with the script of the first of
237        them. In other words, some characters were in the wrong script. There were
238        thirteen such cases, affecting characters in the following ranges:
239    
240          U+002b0 - U+002c1
241          U+0060c - U+0060d
242          U+0061e - U+00612
243          U+0064b - U+0065e
244          U+0074d - U+0076d
245          U+01800 - U+01805
246          U+01d00 - U+01d77
247          U+01d9b - U+01dbf
248          U+0200b - U+0200f
249          U+030fc - U+030fe
250          U+03260 - U+0327f
251          U+0fb46 - U+0fbb1
252          U+10450 - U+1049d
253    
254    12. The -o option (show only the matching part of a line) for pcregrep was not
255        compatible with GNU grep in that, if there was more than one match in a
256        line, it showed only the first of them. It now behaves in the same way as
257        GNU grep.
258    
259    13. If the -o and -v options were combined for pcregrep, it printed a blank
260        line for every non-matching line. GNU grep prints nothing, and pcregrep now
261        does the same. The return code can be used to tell if there were any
262        non-matching lines.
263    
264    14. Added --file-offsets and --line-offsets to pcregrep.
265    
266    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
267        infinitely looping recursion. The bug was that positive lookaheads were not
268        being skipped when checking for a possible empty match (negative lookaheads
269        and both kinds of lookbehind were skipped).
270    
271    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
272        inclusion of <windows.h> to before rather than after the definition of
273        INVALID_FILE_ATTRIBUTES (patch from David Byron).
274    
275    17. Specifying a possessive quantifier with a specific limit for a Unicode
276        character property caused pcre_compile() to compile bad code, which led at
277        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
278        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
279        caused the error; without that there was no problem.
280    
281    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
282    
283    19. Added --enable-pcretest-libreadline.
284    
285    20. In pcrecpp.cc, the variable 'count' was incremented twice in
286        RE::GlobalReplace(). As a result, the number of replacements returned was
287        double what it should be. I removed one of the increments, but Craig sent a
288        later patch that removed the other one (the right fix) and added unit tests
289        that check the return values (which was not done before).
290    
291    21. Several CMake things:
292    
293        (1) Arranged that, when cmake is used on Unix, the libraries end up with
294            the names libpcre and libpcreposix, not just pcre and pcreposix.
295    
296        (2) The above change means that pcretest and pcregrep are now correctly
297            linked with the newly-built libraries, not previously installed ones.
298    
299        (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
300    
301    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
302        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
303        UTF-8 newline character). The key issue is that the pattern starts .*;
304        this means that the match must be either at the beginning, or after a
305        newline. The bug was in the code for advancing after a failed match and
306        checking that the new position followed a newline. It was not taking
307        account of UTF-8 characters correctly.
308    
309    23. PCRE was behaving differently from Perl in the way it recognized POSIX
310        character classes. PCRE was not treating the sequence [:...:] as a
311        character class unless the ... were all letters. Perl, however, seems to
312        allow any characters between [: and :], though of course it rejects as
313        unknown any "names" that contain non-letters, because all the known class
314        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
315        for example, whereas PCRE did not - it did not recognize a POSIX character
316        class. This seemed a bit dangerous, so the code has been changed to be
317        closer to Perl. The behaviour is not identical to Perl, because PCRE will
318        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
319        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
320        Perl does, and where it didn't before.
321    
322    24. Rewrite so as to remove the single use of %n from pcregrep because in some
323        Windows environments %n is disabled by default.
324    
325    
326    Version 7.4 21-Sep-07
327    ---------------------
328    
329    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
330        means that a class such as [\s] counted as "explicit reference to CR or
331        LF". That isn't really right - the whole point of the change was to try to
332        help when there was an actual mention of one of the two characters. So now
333        the change happens only if \r or \n (or a literal CR or LF) character is
334        encountered.
335    
336    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
337        of both had grown to the point where there were only 3 bits left.
338        Fortunately, there was spare space in the data structure, and so I have
339        moved the internal flags into a new 16-bit field to free up more option
340        bits.
341    
342    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
343        but did not set the internal JCHANGED flag - either of these is enough to
344        control the way the "get" function works - but the PCRE_INFO_JCHANGED
345        facility is supposed to tell if (?J) was ever used, so now (?J) at the
346        start sets both bits.
347    
348    4.  Added options (at build time, compile time, exec time) to change \R from
349        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
350    
351    5.  doc/pcresyntax.html was missing from the distribution.
352    
353    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
354        compatibility, even though it is no longer used.
355    
356    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
357        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
358        windows.h file is present (where different names are used). [This was
359        reversed later after testing - see 16 below.]
360    
361    8.  Changed all #include <config.h> to #include "config.h". There were also
362        some further <pcre.h> cases that I changed to "pcre.h".
363    
364    9.  When pcregrep was used with the --colour option, it missed the line ending
365        sequence off the lines that it output.
366    
367    10. It was pointed out to me that arrays of string pointers cause lots of
368        relocations when a shared library is dynamically loaded. A technique of
369        using a single long string with a table of offsets can drastically reduce
370        these. I have refactored PCRE in four places to do this. The result is
371        dramatic:
372    
373          Originally:                          290
374          After changing UCP table:            187
375          After changing error message table:   43
376          After changing table of "verbs"       36
377          After changing table of Posix names   22
378    
379        Thanks to the folks working on Gregex for glib for this insight.
380    
381    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
382        unicode-properties was also set.
383    
384    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
385    
386    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
387        checked only for CRLF.
388    
389    14. Added casts to pcretest.c to avoid compiler warnings.
390    
391    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
392    
393    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
394        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
395        entirely. This removes changes made in 7 above.
396    
397    17. The CMake files have been updated, and there is now more information about
398        building with CMake in the NON-UNIX-USE document.
399    
400    
401    Version 7.3 28-Aug-07
402    ---------------------
403    
404     1. In the rejigging of the build system that eventually resulted in 7.1, the
405        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
406        brackets there is not right, since it causes compilers to look for an
407        installed pcre.h, not the version that is in the source that is being
408        compiled (which of course may be different). I have changed it back to:
409    
410          #include "pcre.h"
411    
412        I have a vague recollection that the change was concerned with compiling in
413        different directories, but in the new build system, that is taken care of
414        by the VPATH setting the Makefile.
415    
416     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
417        when the subject happened to end in the byte 0x85 (e.g. if the last
418        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
419        characters but of course it shouldn't be taken as a newline when it is part
420        of another character. The bug was that, for an unlimited repeat of . in
421        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
422        characters when looking for a newline.
423    
424     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
425    
426     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
427        in debug output.
428    
429     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
430        long printing in the pcrecpp unittest when running under MinGW.
431    
432     6. ESC_K was left out of the EBCDIC table.
433    
434     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
435        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
436        limit also applies to "virtual nesting" when a pattern is recursive, and in
437        this case 1000 isn't so big. I have been able to remove this limit at the
438        expense of backing off one optimization in certain circumstances. Normally,
439        when pcre_exec() would call its internal match() function recursively and
440        immediately return the result unconditionally, it uses a "tail recursion"
441        feature to save stack. However, when a subpattern that can match an empty
442        string has an unlimited repetition quantifier, it no longer makes this
443        optimization. That gives it a stack frame in which to save the data for
444        checking that an empty string has been matched. Previously this was taken
445        from the 1000-entry workspace that had been reserved. So now there is no
446        explicit limit, but more stack is used.
447    
448     8. Applied Daniel's patches to solve problems with the import/export magic
449        syntax that is required for Windows, and which was going wrong for the
450        pcreposix and pcrecpp parts of the library. These were overlooked when this
451        problem was solved for the main library.
452    
453     9. There were some crude static tests to avoid integer overflow when computing
454        the size of patterns that contain repeated groups with explicit upper
455        limits. As the maximum quantifier is 65535, the maximum group length was
456        set at 30,000 so that the product of these two numbers did not overflow a
457        32-bit integer. However, it turns out that people want to use groups that
458        are longer than 30,000 bytes (though not repeat them that many times).
459        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
460        made it possible to implement the integer overflow checks in a much more
461        dynamic way, which I have now done. The artificial limitation on group
462        length has been removed - we now have only the limit on the total length of
463        the compiled pattern, which depends on the LINK_SIZE setting.
464    
465    10. Fixed a bug in the documentation for get/copy named substring when
466        duplicate names are permitted. If none of the named substrings are set, the
467        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
468        empty string.
469    
470    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
471        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
472        because the ] is interpreted as the first data character and the
473        terminating ] is not found. PCRE has been made compatible with Perl in this
474        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
475        cause memory overwriting.
476    
477    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
478        string has been matched (to stop an infinite loop). It was not recognizing
479        a conditional subpattern that could match an empty string if that
480        subpattern was within another subpattern. For example, it looped when
481        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
482        condition was not nested. This bug has been fixed.
483    
484    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
485        past the start of the subject in the presence of bytes with the top bit
486        set, for example "\x8aBCD".
487    
488    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
489        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
490    
491    14. Optimized (?!) to (*FAIL).
492    
493    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
494        This restricts code points to be within the range 0 to 0x10FFFF, excluding
495        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
496        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
497        does: it's just the validity check that is more restrictive.
498    
499    16. Inserted checks for integer overflows during escape sequence (backslash)
500        processing, and also fixed erroneous offset values for syntax errors during
501        backslash processing.
502    
503    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
504        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
505    
506    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
507        caused an overrun.
508    
509    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
510        something other than just ASCII characters) inside a group that had an
511        unlimited repeat caused a loop at compile time (while checking to see
512        whether the group could match an empty string).
513    
514    20. Debugging a pattern containing \p or \P could cause a crash. For example,
515        [\P{Any}] did so. (Error in the code for printing property names.)
516    
517    21. An orphan \E inside a character class could cause a crash.
518    
519    22. A repeated capturing bracket such as (A)? could cause a wild memory
520        reference during compilation.
521    
522    23. There are several functions in pcre_compile() that scan along a compiled
523        expression for various reasons (e.g. to see if it's fixed length for look
524        behind). There were bugs in these functions when a repeated \p or \P was
525        present in the pattern. These operators have additional parameters compared
526        with \d, etc, and these were not being taken into account when moving along
527        the compiled data. Specifically:
528    
529        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
530            length.
531    
532        (b) An item such as \pL+ within a repeated group could cause crashes or
533            loops.
534    
535        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
536            "reference to non-existent subpattern" error.
537    
538        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
539    
540    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
541        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
542    
543    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
544    
545    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
546        character were causing crashes (broken optimization).
547    
548    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
549        \p or \P) caused a compile-time loop.
550    
551    28. More problems have arisen in unanchored patterns when CRLF is a valid line
552        break. For example, the unstudied pattern [\r\n]A does not match the string
553        "\r\nA" because change 7.0/46 below moves the current point on by two
554        characters after failing to match at the start. However, the pattern \nA
555        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
556        the same is true. There doesn't seem any very clean way out of this, but
557        what I have chosen to do makes the common cases work: PCRE now takes note
558        of whether there can be an explicit match for \r or \n anywhere in the
559        pattern, and if so, 7.0/46 no longer applies. As part of this change,
560        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
561        pattern has explicit CR or LF references.
562    
563    29. Added (*CR) etc for changing newline setting at start of pattern.
564    
565    
566    Version 7.2 19-Jun-07
567    ---------------------
568    
569     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
570        which is apparently normally available under Windows.
571    
572     2. Re-jig the pcregrep tests with different newline settings in an attempt
573        to make them independent of the local environment's newline setting.
574    
575     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
576    
577     4. Some of the "internals" tests were previously cut out when the link size
578        was not 2, because the output contained actual offsets. The recent new
579        "Z" feature of pcretest means that these can be cut out, making the tests
580        usable with all link sizes.
581    
582     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
583        stack recursion. This gives a massive performance boost under BSD, but just
584        a small improvement under Linux. However, it saves one field in the frame
585        in all cases.
586    
587     6. Added more features from the forthcoming Perl 5.10:
588    
589        (a) (?-n) (where n is a string of digits) is a relative subroutine or
590            recursion call. It refers to the nth most recently opened parentheses.
591    
592        (b) (?+n) is also a relative subroutine call; it refers to the nth next
593            to be opened parentheses.
594    
595        (c) Conditions that refer to capturing parentheses can be specified
596            relatively, for example, (?(-2)... or (?(+3)...
597    
598        (d) \K resets the start of the current match so that everything before
599            is not part of it.
600    
601        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
602    
603        (f) \g{name} is another synonym - part of Perl 5.10's unification of
604            reference syntax.
605    
606        (g) (?| introduces a group in which the numbering of parentheses in each
607            alternative starts with the same number.
608    
609        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
610    
611     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
612        PCRE_INFO_JCHANGED.
613    
614     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
615        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
616        for detecting groups that can match an empty string.
617    
618     9. A pattern with a very large number of alternatives (more than several
619        hundred) was running out of internal workspace during the pre-compile
620        phase, where pcre_compile() figures out how much memory will be needed. A
621        bit of new cunning has reduced the workspace needed for groups with
622        alternatives. The 1000-alternative test pattern now uses 12 bytes of
623        workspace instead of running out of the 4096 that are available.
624    
625    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
626    
627    11. Applied patch from Google to remove an optimization that didn't quite work.
628        The report of the bug said:
629    
630          pcrecpp::RE("a*").FullMatch("aaa") matches, while
631          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
632          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
633    
634    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
635        it matched the wrong number of bytes.
636    
637    
638    Version 7.1 24-Apr-07
639    ---------------------
640    
641     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
642        that is more "standard", making use of automake and other Autotools. There
643        is some re-arrangement of the files and adjustment of comments consequent
644        on this.
645    
646     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
647        for recursive directory scanning broke on some systems because the files
648        are not scanned in any specific order and on different systems the order
649        was different. A call to "sort" has been inserted into RunGrepTest for the
650        approprate test as a short-term fix. In the longer term there may be an
651        alternative.
652    
653     3. I had an email from Eric Raymond about problems translating some of PCRE's
654        man pages to HTML (despite the fact that I distribute HTML pages, some
655        people do their own conversions for various reasons). The problems
656        concerned the use of low-level troff macros .br and .in. I have therefore
657        removed all such uses from the man pages (some were redundant, some could
658        be replaced by .nf/.fi pairs). The 132html script that I use to generate
659        HTML has been updated to handle .nf/.fi and to complain if it encounters
660        .br or .in.
661    
662     4. Updated comments in configure.ac that get placed in config.h.in and also
663        arranged for config.h to be included in the distribution, with the name
664        config.h.generic, for the benefit of those who have to compile without
665        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
666    
667     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
668        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
669        makevp.bat for the current PCRE, using the additional files
670        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
671    
672     6. A Windows user reported a minor discrepancy with test 2, which turned out
673        to be caused by a trailing space on an input line that had got lost in his
674        copy. The trailing space was an accident, so I've just removed it.
675    
676     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
677        that is needed.
678    
679     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
680        as "const" (a) because they are and (b) because it helps the PHP
681        maintainers who have recently made a script to detect big data structures
682        in the php code that should be moved to the .rodata section. I remembered
683        to update Builducptable as well, so it won't revert if ucptable.h is ever
684        re-created.
685    
686     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
687        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
688        order to be able to cut out the UTF-8 tables in the latter when UTF-8
689        support is not required. This saves 1.5-2K of code, which is important in
690        some applications.
691    
692        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
693        so as not to refer to the tables, even though these functions will never be
694        called when UTF-8 support is disabled. Otherwise there are problems with a
695        shared library.
696    
697    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
698    
699        (a) It was defining its arguments as char * instead of void *.
700    
701        (b) It was assuming that all moves were upwards in memory; this was true
702            a long time ago when I wrote it, but is no longer the case.
703    
704        The emulated memove() is provided for those environments that have neither
705        memmove() nor bcopy(). I didn't think anyone used it these days, but that
706        is clearly not the case, as these two bugs were recently reported.
707    
708    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
709        and Detrail to create the HTML documentation, the .txt form of the man
710        pages, and it removes trailing spaces from listed files. It also creates
711        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
712        case, it wraps all the #defines with #ifndefs. This script should be run
713        before "make dist".
714    
715    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
716        with Unicode property support.
717    
718        (a) For a maximizing quantifier, if the two different cases of the
719            character were of different lengths in their UTF-8 codings (there are
720            some cases like this - I found 11), and the matching function had to
721            back up over a mixture of the two cases, it incorrectly assumed they
722            were both the same length.
723    
724        (b) When PCRE was configured to use the heap rather than the stack for
725            recursion during matching, it was not correctly preserving the data for
726            the other case of a UTF-8 character when checking ahead for a match
727            while processing a minimizing repeat. If the check also involved
728            matching a wide character, but failed, corruption could cause an
729            erroneous result when trying to check for a repeat of the original
730            character.
731    
732    13. Some tidying changes to the testing mechanism:
733    
734        (a) The RunTest script now detects the internal link size and whether there
735            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
736            values substituted by "configure". (The RunGrepTest script already did
737            this for UTF-8.) The configure.ac script no longer substitutes the
738            relevant variables.
739    
740        (b) The debugging options /B and /D in pcretest show the compiled bytecode
741            with length and offset values. This means that the output is different
742            for different internal link sizes. Test 2 is skipped for link sizes
743            other than 2 because of this, bypassing the problem. Unfortunately,
744            there was also a test in test 3 (the locale tests) that used /B and
745            failed for link sizes other than 2. Rather than cut the whole test out,
746            I have added a new /Z option to pcretest that replaces the length and
747            offset values with spaces. This is now used to make test 3 independent
748            of link size. (Test 2 will be tidied up later.)
749    
750    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
751        segmentation fault instead of returning the appropriate error message.
752    
753    15. In multiline mode when the newline sequence was set to "any", the pattern
754        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
755        This doesn't seem right; it now treats the CRLF combination as the line
756        ending, and so does not match in that case. It's only a pattern such as ^$
757        that would hit this one: something like ^ABC$ would have failed after \r
758        and then tried again after \r\n.
759    
760    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
761        in an attempt to make files that differ only in their line terminators
762        compare equal. This works on Linux.
763    
764    17. Under certain error circumstances pcregrep might try to free random memory
765        as it exited. This is now fixed, thanks to valgrind.
766    
767    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
768        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
769        was because its rules for how to advance for /g after matching an empty
770        string at the end of a line did not allow for this case. They now check for
771        it specially.
772    
773    20. pcretest is supposed to handle patterns and data of any length, by
774        extending its buffers when necessary. It was getting this wrong when the
775        buffer for a data line had to be extended.
776    
777    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
778        CRLF as a newline sequence.
779    
780    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
781        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
782        I have nevertheless tidied it up.
783    
784    23. Added some casts to kill warnings from HP-UX ia64 compiler.
785    
786    24. Added a man page for pcre-config.
787    
788    
789    Version 7.0 19-Dec-06
790    ---------------------
791    
792     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
793        moving to gcc 4.1.1.
794    
795     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
796        sys/time.h, which is documented as needed for this function. It doesn't
797        seem to matter on Linux, but it showed up on some releases of OS X.
798    
799     3. It seems that there are systems where bytes whose values are greater than
800        127 match isprint() in the "C" locale. The "C" locale should be the
801        default when a C program starts up. In most systems, only ASCII printing
802        characters match isprint(). This difference caused the output from pcretest
803        to vary, making some of the tests fail. I have changed pcretest so that:
804    
805        (a) When it is outputting text in the compiled version of a pattern, bytes
806            other than 32-126 are always shown as hex escapes.
807    
808        (b) When it is outputting text that is a matched part of a subject string,
809            it does the same, unless a different locale has been set for the match
810            (using the /L modifier). In this case, it uses isprint() to decide.
811    
812     4. Fixed a major bug that caused incorrect computation of the amount of memory
813        required for a compiled pattern when options that changed within the
814        pattern affected the logic of the preliminary scan that determines the
815        length. The relevant options are -x, and -i in UTF-8 mode. The result was
816        that the computed length was too small. The symptoms of this bug were
817        either the PCRE error "internal error: code overflow" from pcre_compile(),
818        or a glibc crash with a message such as "pcretest: free(): invalid next
819        size (fast)". Examples of patterns that provoked this bug (shown in
820        pcretest format) are:
821    
822          /(?-x: )/x
823          /(?x)(?-x: \s*#\s*)/
824          /((?i)[\x{c0}])/8
825          /(?i:[\x{c0}])/8
826    
827        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
828        is now done differently.
829    
830     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
831        wrapper classes; (b) implement a new function in the C++ scanner that is
832        more efficient than the old way of doing things because it avoids levels of
833        recursion in the regex matching; (c) add a paragraph to the documentation
834        for the FullMatch() function.
835    
836     6. The escape sequence \n was being treated as whatever was defined as
837        "newline". Not only was this contrary to the documentation, which states
838        that \n is character 10 (hex 0A), but it also went horribly wrong when
839        "newline" was defined as CRLF. This has been fixed.
840    
841     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
842        was being set to -1 for the "end of line" case (supposedly a value that no
843        character can have). Though this value is never used (the check for end of
844        line is "zero bytes in current character"), it caused compiler complaints.
845        I've changed it to 0xffffffff.
846    
847     8. In pcre_version.c, the version string was being built by a sequence of
848        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
849        string (as it is for production releases) called a macro with an empty
850        argument. The C standard says the result of this is undefined. The gcc
851        compiler treats it as an empty string (which was what was wanted) but it is
852        reported that Visual C gives an error. The source has been hacked around to
853        avoid this problem.
854    
855     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
856        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
857        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
858        of them did).
859    
860    10. Originally, pcretest opened its input and output without "b"; then I was
861        told that "b" was needed in some environments, so it was added for release
862        5.0 to both the input and output. (It makes no difference on Unix-like
863        systems.) Later I was told that it is wrong for the input on Windows. I've
864        now abstracted the modes into two macros, to make it easier to fiddle with
865        them, and removed "b" from the input mode under Windows.
866    
867    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
868    
869    12. Added -help and --help to pcretest as an official way of being reminded
870        of the options.
871    
872    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
873        and pcrecpp.cc because they annoy compilers at high warning levels.
874    
875    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
876    
877    15. Fixed an occurrence of == in configure.ac that should have been = (shell
878        scripts are not C programs :-) and which was not noticed because it works
879        on Linux.
880    
881    16. pcretest is supposed to handle any length of pattern and data line (as one
882        line or as a continued sequence of lines) by extending its input buffer if
883        necessary. This feature was broken for very long pattern lines, leading to
884        a string of junk being passed to pcre_compile() if the pattern was longer
885        than about 50K.
886    
887    17. I have done a major re-factoring of the way pcre_compile() computes the
888        amount of memory needed for a compiled pattern. Previously, there was code
889        that made a preliminary scan of the pattern in order to do this. That was
890        OK when PCRE was new, but as the facilities have expanded, it has become
891        harder and harder to keep it in step with the real compile phase, and there
892        have been a number of bugs (see for example, 4 above). I have now found a
893        cunning way of running the real compile function in a "fake" mode that
894        enables it to compute how much memory it would need, while actually only
895        ever using a few hundred bytes of working memory and without too many
896        tests of the mode. This should make future maintenance and development
897        easier. A side effect of this work is that the limit of 200 on the nesting
898        depth of parentheses has been removed (though this was never a serious
899        limitation, I suspect). However, there is a downside: pcre_compile() now
900        runs more slowly than before (30% or more, depending on the pattern). I
901        hope this isn't a big issue. There is no effect on runtime performance.
902    
903    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
904        newline (only possible for the last line of a file) and it was a
905        pattern that set a locale (followed by /Lsomething), pcretest crashed.
906    
907    19. Added additional timing features to pcretest. (1) The -tm option now times
908        matching only, not compiling. (2) Both -t and -tm can be followed, as a
909        separate command line item, by a number that specifies the number of
910        repeats to use when timing. The default is 50000; this gives better
911        precision, but takes uncomfortably long for very large patterns.
912    
913    20. Extended pcre_study() to be more clever in cases where a branch of a
914        subpattern has no definite first character. For example, (a*|b*)[cd] would
915        previously give no result from pcre_study(). Now it recognizes that the
916        first character must be a, b, c, or d.
917    
918    21. There was an incorrect error "recursive call could loop indefinitely" if
919        a subpattern (or the entire pattern) that was being tested for matching an
920        empty string contained only one non-empty item after a nested subpattern.
921        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
922        incorrectly, because the \d was being skipped in the check.
923    
924    22. The pcretest program now has a new pattern option /B and a command line
925        option -b, which is equivalent to adding /B to every pattern. This causes
926        it to show the compiled bytecode, without the additional information that
927        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
928        is the same as /B/I).
929    
930    23. A new optimization is now able automatically to treat some sequences such
931        as a*b as a*+b. More specifically, if something simple (such as a character
932        or a simple class like \d) has an unlimited quantifier, and is followed by
933        something that cannot possibly match the quantified thing, the quantifier
934        is automatically "possessified".
935    
936    24. A recursive reference to a subpattern whose number was greater than 39
937        went wrong under certain circumstances in UTF-8 mode. This bug could also
938        have affected the operation of pcre_study().
939    
940    25. Realized that a little bit of performance could be had by replacing
941        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
942    
943    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
944    
945    27. Possessive quantifiers such as a++ were previously implemented by turning
946        them into atomic groups such as ($>a+). Now they have their own opcodes,
947        which improves performance. This includes the automatically created ones
948        from 23 above.
949    
950    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
951        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
952        the first matched character to be a colon. This applied both to named and
953        numbered groups.
954    
955    29. The ucpinternal.h header file was missing its idempotency #ifdef.
956    
957    30. I was sent a "project" file called libpcre.a.dev which I understand makes
958        building PCRE on Windows easier, so I have included it in the distribution.
959    
960    31. There is now a check in pcretest against a ridiculously large number being
961        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
962        loop, the loop is abandoned.
963    
964    32. Forward references to subpatterns in conditions such as (?(2)...) where
965        subpattern 2 is defined later cause pcre_compile() to search forwards in
966        the pattern for the relevant set of parentheses. This search went wrong
967        when there were unescaped parentheses in a character class, parentheses
968        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
969    
970    33. "Subroutine" calls and backreferences were previously restricted to
971        referencing subpatterns earlier in the regex. This restriction has now
972        been removed.
973    
974    34. Added a number of extra features that are going to be in Perl 5.10. On the
975        whole, these are just syntactic alternatives for features that PCRE had
976        previously implemented using the Python syntax or my own invention. The
977        other formats are all retained for compatibility.
978    
979        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
980            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
981            also .NET compatible.
982    
983        (b) A recursion or subroutine call to a named group can now be defined as
984            (?&name) as well as (?P>name).
985    
986        (c) A backreference to a named group can now be defined as \k<name> or
987            \k'name' as well as (?P=name). The new forms, as well as being in Perl
988            5.10, are also .NET compatible.
989    
990        (d) A conditional reference to a named group can now use the syntax
991            (?(<name>) or (?('name') as well as (?(name).
992    
993        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
994            groups (named and numbered) that are never evaluated inline, but can be
995            called as "subroutines" from elsewhere. In effect, the DEFINE condition
996            is always false. There may be only one alternative in such a group.
997    
998        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
999            as the simple (?(R). The condition is true only if the most recent
1000            recursion is that of the given number or name. It does not search out
1001            through the entire recursion stack.
1002    
1003        (g) The escape \gN or \g{N} has been added, where N is a positive or
1004            negative number, specifying an absolute or relative reference.
1005    
1006    35. Tidied to get rid of some further signed/unsigned compiler warnings and
1007        some "unreachable code" warnings.
1008    
1009    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1010        things, this adds five new scripts.
1011    
1012    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1013        There were also incompatibilities regarding the handling of \Q..\E inside
1014        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1015        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1016    
1017    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1018        matches an empty string, and forcibly breaks the loop. There were bugs in
1019        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
1020        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
1021        separate and independent bugs (that affected different cases) have been
1022        fixed.
1023    
1024    39. Refactored the code to abolish the use of different opcodes for small
1025        capturing bracket numbers. This is a tidy that I avoided doing when I
1026        removed the limit on the number of capturing brackets for 3.5 back in 2001.
1027        The new approach is not only tidier, it makes it possible to reduce the
1028        memory needed to fix the previous bug (38).
1029    
1030    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1031        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1032        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1033        mode.
1034    
1035    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1036        report.
1037    
1038    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1039        copy construction and assignment in the C++ wrapper.
1040    
1041    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1042        couple of bugs that could have given wrong results in the "--newline=crlf"
1043        case.
1044    
1045    44. Added a number of casts and did some reorganization of signed/unsigned int
1046        variables following suggestions from Dair Grant. Also renamed the variable
1047        "this" as "item" because it is a C++ keyword.
1048    
1049    45. Arranged for dftables to add
1050    
1051          #include "pcre_internal.h"
1052    
1053        to pcre_chartables.c because without it, gcc 4.x may remove the array
1054        definition from the final binary if PCRE is built into a static library and
1055        dead code stripping is activated.
1056    
1057    46. For an unanchored pattern, if a match attempt fails at the start of a
1058        newline sequence, and the newline setting is CRLF or ANY, and the next two
1059        characters are CRLF, advance by two characters instead of one.
1060    
1061    
1062    Version 6.7 04-Jul-06
1063    ---------------------
1064    
1065     1. In order to handle tests when input lines are enormously long, pcretest has
1066        been re-factored so that it automatically extends its buffers when
1067        necessary. The code is crude, but this _is_ just a test program. The
1068        default size has been increased from 32K to 50K.
1069    
1070     2. The code in pcre_study() was using the value of the re argument before
1071        testing it for NULL. (Of course, in any sensible call of the function, it
1072        won't be NULL.)
1073    
1074     3. The memmove() emulation function in pcre_internal.h, which is used on
1075        systems that lack both memmove() and bcopy() - that is, hardly ever -
1076        was missing a "static" storage class specifier.
1077    
1078     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1079        containing an extended class (one that cannot be represented by a bitmap
1080        because it contains high-valued characters or Unicode property items, e.g.
1081        [\pZ]). Almost always one would set UTF-8 mode when processing such a
1082        pattern, but PCRE should not loop if you do not (it no longer does).
1083        [Detail: two cases were found: (a) a repeated subpattern containing an
1084        extended class; (b) a recursive reference to a subpattern that followed a
1085        previous extended class. It wasn't skipping over the extended class
1086        correctly when UTF-8 mode was not set.]
1087    
1088     5. A negated single-character class was not being recognized as fixed-length
1089        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1090        compile error "lookbehind assertion is not fixed length".
1091    
1092     6. The RunPerlTest auxiliary script was showing an unexpected difference
1093        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1094        write a Perl script that can interpret lines of an input file either as
1095        byte characters or as UTF-8, which is what "perltest" was being required to
1096        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1097        can't do is switch easily at run time between having the "use utf8;" pragma
1098        or not. In the end, I fudged it by using the RunPerlTest script to insert
1099        "use utf8;" explicitly for the UTF-8 tests.
1100    
1101     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1102        the end of the subject string, contrary to the documentation and to what
1103        Perl does. This was true of both matching functions. Now it matches only at
1104        the start of the subject and immediately after *internal* newlines.
1105    
1106     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1107        a pointer to an int instead of a pointer to an unsigned long int. This
1108        caused problems on 64-bit systems.
1109    
1110     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1111        instance of the 'standard' template library not being so standard".
1112    
1113    10. There was no check on the number of named subpatterns nor the maximum
1114        length of a subpattern name. The product of these values is used to compute
1115        the size of the memory block for a compiled pattern. By supplying a very
1116        long subpattern name and a large number of named subpatterns, the size
1117        computation could be caused to overflow. This is now prevented by limiting
1118        the length of names to 32 characters, and the number of named subpatterns
1119        to 10,000.
1120    
1121    11. Subpatterns that are repeated with specific counts have to be replicated in
1122        the compiled pattern. The size of memory for this was computed from the
1123        length of the subpattern and the repeat count. The latter is limited to
1124        65535, but there was no limit on the former, meaning that integer overflow
1125        could in principle occur. The compiled length of a repeated subpattern is
1126        now limited to 30,000 bytes in order to prevent this.
1127    
1128    12. Added the optional facility to have named substrings with the same name.
1129    
1130    13. Added the ability to use a named substring as a condition, using the
1131        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1132        are numbers (not recommended). Forward references are permitted.
1133    
1134    14. Added forward references in named backreferences (if you see what I mean).
1135    
1136    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1137        pattern could run off the end of the subject. For example, the pattern
1138        "(?s)(.{1,5})"8 did this with the subject "ab".
1139    
1140    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1141        PCRE_CASELESS was set when matching characters that were quantified with ?
1142        or *.
1143    
1144    17. A character class other than a single negated character that had a minimum
1145        but no maximum quantifier - for example [ab]{6,} - was not handled
1146        correctly by pce_dfa_exec(). It would match only one character.
1147    
1148    18. A valid (though odd) pattern that looked like a POSIX character
1149        class but used an invalid character after [ (for example [[,abc,]]) caused
1150        pcre_compile() to give the error "Failed: internal error: code overflow" or
1151        in some cases to crash with a glibc free() error. This could even happen if
1152        the pattern terminated after [[ but there just happened to be a sequence of
1153        letters, a binary zero, and a closing ] in the memory that followed.
1154    
1155    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1156        over the years. Originally (before any Unicode support), just the bottom 8
1157        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1158        output from "man perlunicode" includes this:
1159    
1160          The regular expression compiler produces polymorphic opcodes.  That
1161          is, the pattern adapts to the data and automatically switches to
1162          the Unicode character scheme when presented with Unicode data--or
1163          instead uses a traditional byte scheme when presented with byte
1164          data.
1165    
1166        Sadly, a wide octal escape does not cause a switch, and in a string with
1167        no other multibyte characters, these octal escapes are treated as before.
1168        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
1169        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1170        Unicode string.
1171    
1172        I have not perpetrated such confusion in PCRE. Up till now, it took just
1173        the bottom 8 bits, as in old Perl. I have now made octal escapes with
1174        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1175        translate to the appropriate multibyte character.
1176    
1177    29. Applied some refactoring to reduce the number of warnings from Microsoft
1178        and Borland compilers. This has included removing the fudge introduced
1179        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1180        a warning about an unused variable.
1181    
1182    21. PCRE has not included VT (character 0x0b) in the set of whitespace
1183        characters since release 4.0, because Perl (from release 5.004) does not.
1184        [Or at least, is documented not to: some releases seem to be in conflict
1185        with the documentation.] However, when a pattern was studied with
1186        pcre_study() and all its branches started with \s, PCRE still included VT
1187        as a possible starting character. Of course, this did no harm; it just
1188        caused an unnecessary match attempt.
1189    
1190    22. Removed a now-redundant internal flag bit that recorded the fact that case
1191        dependency changed within the pattern. This was once needed for "required
1192        byte" processing, but is no longer used. This recovers a now-scarce options
1193        bit. Also moved the least significant internal flag bit to the most-
1194        significant bit of the word, which was not previously used (hangover from
1195        the days when it was an int rather than a uint) to free up another bit for
1196        the future.
1197    
1198    23. Added support for CRLF line endings as well as CR and LF. As well as the
1199        default being selectable at build time, it can now be changed at runtime
1200        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1201        specify that it is scanning data with non-default line endings.
1202    
1203    24. Changed the definition of CXXLINK to make it agree with the definition of
1204        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1205    
1206    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1207        recursions. This makes a big different to stack usage for some patterns.
1208    
1209    26. If a subpattern containing a named recursion or subroutine reference such
1210        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1211        the space required for the compiled pattern went wrong and gave too small a
1212        value. Depending on the environment, this could lead to "Failed: internal
1213        error: code overflow at offset 49" or "glibc detected double free or
1214        corruption" errors.
1215    
1216    27. Applied patches from Google (a) to support the new newline modes and (b) to
1217        advance over multibyte UTF-8 characters in GlobalReplace.
1218    
1219    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1220        difference for some implementation of PCRE in some Windows version.
1221    
1222    29. Added some extra testing facilities to pcretest:
1223    
1224        \q<number>   in a data line sets the "match limit" value
1225        \Q<number>   in a data line sets the "match recursion limt" value
1226        -S <number>  sets the stack size, where <number> is in megabytes
1227    
1228        The -S option isn't available for Windows.
1229    
1230    
1231    Version 6.6 06-Feb-06
1232    ---------------------
1233    
1234     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1235        in pcreposix.h. I have copied the definition from pcre.h.
1236    
1237     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1238        because pcre.h is no longer a built file.
1239    
1240     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1241        not normally included in the compiled code.
1242    
1243    
1244    Version 6.5 01-Feb-06
1245    ---------------------
1246    
1247     1. When using the partial match feature with pcre_dfa_exec(), it was not
1248        anchoring the second and subsequent partial matches at the new starting
1249        point. This could lead to incorrect results. For example, with the pattern
1250        /1234/, partially matching against "123" and then "a4" gave a match.
1251    
1252     2. Changes to pcregrep:
1253    
1254        (a) All non-match returns from pcre_exec() were being treated as failures
1255            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1256            error message is output. Some extra information is given for the
1257            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
1258            probably the only errors that are likely to be caused by users (by
1259            specifying a regex that has nested indefinite repeats, for instance).
1260            If there are more than 20 of these errors, pcregrep is abandoned.
1261    
1262        (b) A binary zero was treated as data while matching, but terminated the
1263            output line if it was written out. This has been fixed: binary zeroes
1264            are now no different to any other data bytes.
1265    
1266        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1267            used to set a locale for matching. The --locale=xxxx long option has
1268            been added (no short equivalent) to specify a locale explicitly on the
1269            pcregrep command, overriding the environment variables.
1270    
1271        (d) When -B was used with -n, some line numbers in the output were one less
1272            than they should have been.
1273    
1274        (e) Added the -o (--only-matching) option.
1275    
1276        (f) If -A or -C was used with -c (count only), some lines of context were
1277            accidentally printed for the final match.
1278    
1279        (g) Added the -H (--with-filename) option.
1280    
1281        (h) The combination of options -rh failed to suppress file names for files
1282            that were found from directory arguments.
1283    
1284        (i) Added the -D (--devices) and -d (--directories) options.
1285    
1286        (j) Added the -F (--fixed-strings) option.
1287    
1288        (k) Allow "-" to be used as a file name for -f as well as for a data file.
1289    
1290        (l) Added the --colo(u)r option.
1291    
1292        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1293            is not present by default.
1294    
1295     3. A nasty bug was discovered in the handling of recursive patterns, that is,
1296        items such as (?R) or (?1), when the recursion could match a number of
1297        alternatives. If it matched one of the alternatives, but subsequently,
1298        outside the recursion, there was a failure, the code tried to back up into
1299        the recursion. However, because of the way PCRE is implemented, this is not
1300        possible, and the result was an incorrect result from the match.
1301    
1302        In order to prevent this happening, the specification of recursion has
1303        been changed so that all such subpatterns are automatically treated as
1304        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1305    
1306     4. I had overlooked the fact that, in some locales, there are characters for
1307        which isalpha() is true but neither isupper() nor islower() are true. In
1308        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1309        and ordfeminine) are like this. This affected the treatment of \w and \W
1310        when they appeared in character classes, but not when they appeared outside
1311        a character class. The bit map for "word" characters is now created
1312        separately from the results of isalnum() instead of just taking it from the
1313        upper, lower, and digit maps. (Plus the underscore character, of course.)
1314    
1315     5. The above bug also affected the handling of POSIX character classes such as
1316        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1317        permanent tables. Instead, the bit maps for such a class were previously
1318        created as the appropriate unions of the upper, lower, and digit bitmaps.
1319        Now they are created by subtraction from the [[:word:]] class, which has
1320        its own bitmap.
1321    
1322     6. The [[:blank:]] character class matches horizontal, but not vertical space.
1323        It is created by subtracting the vertical space characters (\x09, \x0a,
1324        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1325        subtraction was done in the overall bitmap for a character class, meaning
1326        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1327        be recognized. This bug has been fixed.
1328    
1329     7. Patches from the folks at Google:
1330    
1331          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1332          real life, but is still worth protecting against".
1333    
1334          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1335          regular expressions".
1336    
1337          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1338          have it.
1339    
1340          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1341          "configure" and the latter not, in order to fix a problem somebody had
1342          with compiling the Arg class on HP-UX.
1343    
1344          (e) Improve the error-handling of the C++ wrapper a little bit.
1345    
1346          (f) New tests for checking recursion limiting.
1347    
1348     8. The pcre_memmove() function, which is used only if the environment does not
1349        have a standard memmove() function (and is therefore rarely compiled),
1350        contained two bugs: (a) use of int instead of size_t, and (b) it was not
1351        returning a result (though PCRE never actually uses the result).
1352    
1353     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1354        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1355        returned instead of calling malloc() with an overflowing number that would
1356        most likely cause subsequent chaos.
1357    
1358    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1359    
1360    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1361        with this option is matched, the nmatch and pmatch options of regexec() are
1362        ignored.
1363    
1364    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1365        provided in case anyone wants to the the POSIX interface with UTF-8
1366        strings.
1367    
1368    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1369        C++ linking (needed for some HP-UX environments).
1370    
1371    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1372        (unused parameter) and in the pcre_printint() function (omitted "default"
1373        switch label when the default is to do nothing).
1374    
1375    15. Added some code to make it possible, when PCRE is compiled as a C++
1376        library, to replace subject pointers for pcre_exec() with a smart pointer
1377        class, thus making it possible to process discontinuous strings.
1378    
1379    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1380        much the same function. They were added by different people who were trying
1381        to make PCRE easy to compile on non-Unix systems. It has been suggested
1382        that PCRE_EXPORT be abolished now that there is more automatic apparatus
1383        for compiling on Windows systems. I have therefore replaced it with
1384        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1385        defaults to "extern" for C or "extern C" for C++, which works fine on
1386        Unix-like systems. It is now possible to override the value of PCRE_DATA_
1387        SCOPE with something explicit in config.h. In addition:
1388    
1389        (a) pcreposix.h still had just "extern" instead of either of these macros;
1390            I have replaced it with PCRE_DATA_SCOPE.
1391    
1392        (b) Functions such as _pcre_xclass(), which are internal to the library,
1393            but external in the C sense, all had PCRE_EXPORT in their definitions.
1394            This is apparently wrong for the Windows case, so I have removed it.
1395            (It makes no difference on Unix-like systems.)
1396    
1397    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1398        of recursive calls to match(). This is different to MATCH_LIMIT because
1399        that limits the total number of calls to match(), not all of which increase
1400        the depth of recursion. Limiting the recursion depth limits the amount of
1401        stack (or heap if NO_RECURSE is set) that is used. The default can be set
1402        when PCRE is compiled, and changed at run time. A patch from Google adds
1403        this functionality to the C++ interface.
1404    
1405    18. Changes to the handling of Unicode character properties:
1406    
1407        (a) Updated the table to Unicode 4.1.0.
1408    
1409        (b) Recognize characters that are not in the table as "Cn" (undefined).
1410    
1411        (c) I revised the way the table is implemented to a much improved format
1412            which includes recognition of ranges. It now supports the ranges that
1413            are defined in UnicodeData.txt, and it also amalgamates other
1414            characters into ranges. This has reduced the number of entries in the
1415            table from around 16,000 to around 3,000, thus reducing its size
1416            considerably. I realized I did not need to use a tree structure after
1417            all - a binary chop search is just as efficient. Having reduced the
1418            number of entries, I extended their size from 6 bytes to 8 bytes to
1419            allow for more data.
1420    
1421        (d) Added support for Unicode script names via properties such as \p{Han}.
1422    
1423    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1424        matching that character.
1425    
1426    20. When matching a repeated Unicode property with a minimum greater than zero,
1427        (for example \pL{2,}), PCRE could look past the end of the subject if it
1428        reached it while seeking the minimum number of characters. This could
1429        happen only if some of the characters were more than one byte long, because
1430        there is a check for at least the minimum number of bytes.
1431    
1432    21. Refactored the implementation of \p and \P so as to be more general, to
1433        allow for more different types of property in future. This has changed the
1434        compiled form incompatibly. Anybody with saved compiled patterns that use
1435        \p or \P will have to recompile them.
1436    
1437    22. Added "Any" and "L&" to the supported property types.
1438    
1439    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1440        but give a compile time error if the value is greater than 0xff.
1441    
1442    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1443        accidentally not being installed or uninstalled.
1444    
1445    25. The pcre.h file was built from pcre.h.in, but the only changes that were
1446        made were to insert the current release number. This seemed silly, because
1447        it made things harder for people building PCRE on systems that don't run
1448        "configure". I have turned pcre.h into a distributed file, no longer built
1449        by "configure", with the version identification directly included. There is
1450        no longer a pcre.h.in file.
1451    
1452        However, this change necessitated a change to the pcre-config script as
1453        well. It is built from pcre-config.in, and one of the substitutions was the
1454        release number. I have updated configure.ac so that ./configure now finds
1455        the release number by grepping pcre.h.
1456    
1457    26. Added the ability to run the tests under valgrind.
1458    
1459    
1460    Version 6.4 05-Sep-05
1461    ---------------------
1462    
1463     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1464        "--" to be printed when multiple files were scanned, even when none of the
1465        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1466        consider it to be a bug, and have restored the previous behaviour.
1467    
1468     2. A couple of code tidies to get rid of compiler warnings.
1469    
1470     3. The pcretest program used to cheat by referring to symbols in the library
1471        whose names begin with _pcre_. These are internal symbols that are not
1472        really supposed to be visible externally, and in some environments it is
1473        possible to suppress them. The cheating is now confined to including
1474        certain files from the library's source, which is a bit cleaner.
1475    
1476     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1477        file's purpose clearer.
1478    
1479     5. Reorganized pcre_ucp_findchar().
1480    
1481    
1482    Version 6.3 15-Aug-05
1483    ---------------------
1484    
1485     1. The file libpcre.pc.in did not have general read permission in the tarball.
1486    
1487     2. There were some problems when building without C++ support:
1488    
1489        (a) If C++ support was not built, "make install" and "make test" still
1490            tried to test it.
1491    
1492        (b) There were problems when the value of CXX was explicitly set. Some
1493            changes have been made to try to fix these, and ...
1494    
1495        (c) --disable-cpp can now be used to explicitly disable C++ support.
1496    
1497        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1498            backslash in a target when C++ was disabled. This confuses some
1499            versions of "make", apparently. Using an intermediate variable solves
1500            this. (Same for CPP_LOBJ.)
1501    
1502     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1503        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1504        necessary on certain architectures.
1505    
1506     4. Added a setting of -export-symbols-regex to the link command to remove
1507        those symbols that are exported in the C sense, but actually are local
1508        within the library, and not documented. Their names all begin with
1509        "_pcre_". This is not a perfect job, because (a) we have to except some
1510        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1511        available (and never for static libraries). I have made a note to try to
1512        find a way round (a) in the future.
1513    
1514    
1515    Version 6.2 01-Aug-05
1516    ---------------------
1517    
1518     1. There was no test for integer overflow of quantifier values. A construction
1519        such as {1111111111111111} would give undefined results. What is worse, if
1520        a minimum quantifier for a parenthesized subpattern overflowed and became
1521        negative, the calculation of the memory size went wrong. This could have
1522        led to memory overwriting.
1523    
1524     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1525    
1526     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1527        operating environments where this matters.
1528    
1529     4. Applied Giuseppe Maxia's patch to add additional features for controlling
1530        PCRE options from within the C++ wrapper.
1531    
1532     5. Named capturing subpatterns were not being correctly counted when a pattern
1533        was compiled. This caused two problems: (a) If there were more than 100
1534        such subpatterns, the calculation of the memory needed for the whole
1535        compiled pattern went wrong, leading to an overflow error. (b) Numerical
1536        back references of the form \12, where the number was greater than 9, were
1537        not recognized as back references, even though there were sufficient
1538        previous subpatterns.
1539    
1540     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1541        versions of gcc, e.g. 2.95.4.
1542    
1543    
1544    Version 6.1 21-Jun-05
1545    ---------------------
1546    
1547     1. There was one reference to the variable "posix" in pcretest.c that was not
1548        surrounded by "#if !defined NOPOSIX".
1549    
1550     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1551        the cross-check on the old pcre_info() function, for the benefit of the
1552        cut-down version of PCRE that is currently imported into Exim.
1553    
1554     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1555        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1556        patterns that start (?i) but I don't think that matters. The use of (?i) is
1557        just an example; this all applies to the other options as well.
1558    
1559     4. Since libtool seems to echo the compile commands it is issuing, the output
1560        from "make" can be reduced a bit by putting "@" in front of each libtool
1561        compile command.
1562    
1563     5. Patch from the folks at Google for configure.in to be a bit more thorough
1564        in checking for a suitable C++ installation before trying to compile the
1565        C++ stuff. This should fix a reported problem when a compiler was present,
1566        but no suitable headers.
1567    
1568     6. The man pages all had just "PCRE" as their title. I have changed them to
1569        be the relevant file name. I have also arranged that these names are
1570        retained in the file doc/pcre.txt, which is a concatenation in text format
1571        of all the man pages except the little individual ones for each function.
1572    
1573     7. The NON-UNIX-USE file had not been updated for the different set of source
1574        files that come with release 6. I also added a few comments about the C++
1575        wrapper.
1576    
1577    
1578    Version 6.0 07-Jun-05
1579    ---------------------
1580    
1581     1. Some minor internal re-organization to help with my DFA experiments.
1582    
1583     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1584        didn't matter for the library itself when fully configured, but did matter
1585        when compiling without UCP support, or within Exim, where the ucp files are
1586        not imported.
1587    
1588     3. Refactoring of the library code to split up the various functions into
1589        different source modules. The addition of the new DFA matching code (see
1590        below) to a single monolithic source would have made it really too
1591        unwieldy, quite apart from causing all the code to be include in a
1592        statically linked application, when only some functions are used. This is
1593        relevant even without the DFA addition now that patterns can be compiled in
1594        one application and matched in another.
1595    
1596        The downside of splitting up is that there have to be some external
1597        functions and data tables that are used internally in different modules of
1598        the library but which are not part of the API. These have all had their
1599        names changed to start with "_pcre_" so that they are unlikely to clash
1600        with other external names.
1601    
1602     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1603        a different (DFA) algorithm. Although it is slower than the original
1604        function, it does have some advantages for certain types of matching
1605        problem.
1606    
1607     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1608        including restarting after a partial match.
1609    
1610     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1611        defined when compiling for Windows was sent to me. I have put it into the
1612        code, though I have no means of testing or verifying it.
1613    
1614     7. Added the pcre_refcount() auxiliary function.
1615    
1616     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1617        match before or at the first newline in the subject string. In pcretest,
1618        the /f option on a pattern can be used to set this.
1619    
1620     9. A repeated \w when used in UTF-8 mode with characters greater than 256
1621        would behave wrongly. This has been present in PCRE since release 4.0.
1622    
1623    10. A number of changes to the pcregrep command:
1624    
1625        (a) Refactored how -x works; insert ^(...)$ instead of setting
1626            PCRE_ANCHORED and checking the length, in preparation for adding
1627            something similar for -w.
1628    
1629        (b) Added the -w (match as a word) option.
1630    
1631        (c) Refactored the way lines are read and buffered so as to have more
1632            than one at a time available.
1633    
1634        (d) Implemented a pcregrep test script.
1635    
1636        (e) Added the -M (multiline match) option. This allows patterns to match
1637            over several lines of the subject. The buffering ensures that at least
1638            8K, or the rest of the document (whichever is the shorter) is available
1639            for matching (and similarly the previous 8K for lookbehind assertions).
1640    
1641        (f) Changed the --help output so that it now says
1642    
1643              -w, --word-regex(p)
1644    
1645            instead of two lines, one with "regex" and the other with "regexp"
1646            because that confused at least one person since the short forms are the
1647            same. (This required a bit of code, as the output is generated
1648            automatically from a table. It wasn't just a text change.)
1649    
1650        (g) -- can be used to terminate pcregrep options if the next thing isn't an
1651            option but starts with a hyphen. Could be a pattern or a path name
1652            starting with a hyphen, for instance.
1653    
1654        (h) "-" can be given as a file name to represent stdin.
1655    
1656        (i) When file names are being printed, "(standard input)" is used for
1657            the standard input, for compatibility with GNU grep. Previously
1658            "<stdin>" was used.
1659    
1660        (j) The option --label=xxx can be used to supply a name to be used for
1661            stdin when file names are being printed. There is no short form.
1662    
1663        (k) Re-factored the options decoding logic because we are going to add
1664            two more options that take data. Such options can now be given in four
1665            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1666    
1667        (l) Added the -A, -B, and -C options for requesting that lines of context
1668            around matches be printed.
1669    
1670        (m) Added the -L option to print the names of files that do not contain
1671            any matching lines, that is, the complement of -l.
1672    
1673        (n) The return code is 2 if any file cannot be opened, but pcregrep does
1674            continue to scan other files.
1675    
1676        (o) The -s option was incorrectly implemented. For compatibility with other
1677            greps, it now suppresses the error message for a non-existent or non-
1678            accessible file (but not the return code). There is a new option called
1679            -q that suppresses the output of matching lines, which was what -s was
1680            previously doing.
1681    
1682        (p) Added --include and --exclude options to specify files for inclusion
1683            and exclusion when recursing.
1684    
1685    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1686        Hopefully, it now does.
1687    
1688    12. Missing cast in pcre_study().
1689    
1690    13. Added an "uninstall" target to the makefile.
1691    
1692    14. Replaced "extern" in the function prototypes in Makefile.in with
1693        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1694        world, but is set differently for Windows.
1695    
1696    15. Added a second compiling function called pcre_compile2(). The only
1697        difference is that it has an extra argument, which is a pointer to an
1698        integer error code. When there is a compile-time failure, this is set
1699        non-zero, in addition to the error test pointer being set to point to an
1700        error message. The new argument may be NULL if no error number is required
1701        (but then you may as well call pcre_compile(), which is now just a
1702        wrapper). This facility is provided because some applications need a
1703        numeric error indication, but it has also enabled me to tidy up the way
1704        compile-time errors are handled in the POSIX wrapper.
1705    
1706    16. Added VPATH=.libs to the makefile; this should help when building with one
1707        prefix path and installing with another. (Or so I'm told by someone who
1708        knows more about this stuff than I do.)
1709    
1710    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1711        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1712        match everything, including newlines. This is not POSIX-compatible, but
1713        somebody wanted the feature. From pcretest it can be activated by using
1714        both the P and the s flags.
1715    
1716    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1717    
1718    19. libpcre.pc was being incorrectly installed as executable.
1719    
1720    20. A couple of places in pcretest check for end-of-line by looking for '\n';
1721        it now also looks for '\r' so that it will work unmodified on Windows.
1722    
1723    21. Added Google's contributed C++ wrapper to the distribution.
1724    
1725    22. Added some untidy missing memory free() calls in pcretest, to keep
1726        Electric Fence happy when testing.
1727    
1728    
1729    
1730    Version 5.0 13-Sep-04
1731    ---------------------
1732    
1733     1. Internal change: literal characters are no longer packed up into items
1734        containing multiple characters in a single byte-string. Each character
1735        is now matched using a separate opcode. However, there may be more than one
1736        byte in the character in UTF-8 mode.
1737    
1738     2. The pcre_callout_block structure has two new fields: pattern_position and
1739        next_item_length. These contain the offset in the pattern to the next match
1740        item, and its length, respectively.
1741    
1742     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1743        insertion of callouts before each pattern item. Added the /C option to
1744        pcretest to make use of this.
1745    
1746     4. On the advice of a Windows user, the lines
1747    
1748          #if defined(_WIN32) || defined(WIN32)
1749          _setmode( _fileno( stdout ), 0x8000 );
1750          #endif  /* defined(_WIN32) || defined(WIN32) */
1751    
1752        have been added to the source of pcretest. This apparently does useful
1753        magic in relation to line terminators.
1754    
1755     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1756        for the benefit of those environments where the "b" makes a difference.
1757    
1758     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1759        to know about it. I have put a hack into configure.in that adds in code
1760        to set GCC=yes if CC=icc. This seems to end up at a point in the
1761        generated configure script that is early enough to affect the setting of
1762        compiler options, which is what is needed, but I have no means of testing
1763        whether it really works. (The user who reported this had patched the
1764        generated configure script, which of course I cannot do.)
1765    
1766        LATER: After change 22 below (new libtool files), the configure script
1767        seems to know about icc (and also ecc). Therefore, I have commented out
1768        this hack in configure.in.
1769    
1770     7. Added support for pkg-config (2 patches were sent in).
1771    
1772     8. Negated POSIX character classes that used a combination of internal tables
1773        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1774        [[:^ascii]]. Typically, they would match almost any characters. The other
1775        POSIX classes were not broken in this way.
1776    
1777     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1778        to find the match, as PCRE was deluded into thinking that the match had to
1779        start at the start point or following a newline. The same bug applied to
1780        patterns with negative forward assertions or any backward assertions
1781        preceding ".*" at the start, unless the pattern required a fixed first
1782        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1783    
1784    10. In UTF-8 mode, when moving forwards in the subject after a failed match
1785        starting at the last subject character, bytes beyond the end of the subject
1786        string were read.
1787    
1788    11. Renamed the variable "class" as "classbits" to make life easier for C++
1789        users. (Previously there was a macro definition, but it apparently wasn't
1790        enough.)
1791    
1792    12. Added the new field "tables" to the extra data so that tables can be passed
1793        in at exec time, or the internal tables can be re-selected. This allows
1794        a compiled regex to be saved and re-used at a later time by a different
1795        program that might have everything at different addresses.
1796    
1797    13. Modified the pcre-config script so that, when run on Solaris, it shows a
1798        -R library as well as a -L library.
1799    
1800    14. The debugging options of pcretest (-d on the command line or D on a
1801        pattern) showed incorrect output for anything following an extended class
1802        that contained multibyte characters and which was followed by a quantifier.
1803    
1804    15. Added optional support for general category Unicode character properties
1805        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1806        support. It adds about 90K to the size of the library. The meanings of the
1807        inbuilt class escapes such as \d and \s have NOT been changed.
1808    
1809    16. Updated pcredemo.c to include calls to free() to release the memory for the
1810        compiled pattern.
1811    
1812    17. The generated file chartables.c was being created in the source directory
1813        instead of in the building directory. This caused the build to fail if the
1814        source directory was different from the building directory, and was
1815        read-only.
1816    
1817    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1818        file. No doubt somebody will tell me if they don't make sense... Also added
1819        Dan Mooney's comments about building on OpenVMS.
1820    
1821    19. Added support for partial matching via the PCRE_PARTIAL option for
1822        pcre_exec() and the \P data escape in pcretest.
1823    
1824    20. Extended pcretest with 3 new pattern features:
1825    
1826        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
1827              write the compiled pattern to the file whose name is "rest-of-line".
1828              This is a straight binary dump of the data, with the saved pointer to
1829              the character tables forced to be NULL. The study data, if any, is
1830              written too. After writing, pcretest reads a new pattern.
1831    
1832        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1833              compiled pattern from the given file. There must not be any
1834              occurrences of "<" in the file name (pretty unlikely); if there are,
1835              pcretest will instead treat the initial "<" as a pattern delimiter.
1836              After reading in the pattern, pcretest goes on to read data lines as
1837              usual.
1838    
1839        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1840              and 16-bit fields in a compiled pattern, to simulate a pattern that
1841              was compiled on a host of opposite endianness.
1842    
1843    21. The pcre-exec() function can now cope with patterns that were compiled on
1844        hosts of opposite endianness, with this restriction:
1845    
1846          As for any compiled expression that is saved and used later, the tables
1847          pointer field cannot be preserved; the extra_data field in the arguments
1848          to pcre_exec() should be used to pass in a tables address if a value
1849          other than the default internal tables were used at compile time.
1850    
1851    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1852        now diagnosed as an error. Previously, most of the time, a negative number
1853        would have been treated as zero, but if in addition "ovector" was passed as
1854        NULL, a crash could occur.
1855    
1856    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1857        new versions from the libtool 1.5 distribution (the last one is a copy of
1858        a file called libtool.m4). This seems to have fixed the need to patch
1859        "configure" to support Darwin 1.3 (which I used to do). However, I still
1860        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1861        workstation).
1862    
1863    24. Changed the PCRE licence to be the more standard "BSD" licence.
1864    
1865    
1866    Version 4.5 01-Dec-03
1867    ---------------------
1868    
1869     1. There has been some re-arrangement of the code for the match() function so
1870        that it can be compiled in a version that does not call itself recursively.
1871        Instead, it keeps those local variables that need separate instances for
1872        each "recursion" in a frame on the heap, and gets/frees frames whenever it
1873        needs to "recurse". Keeping track of where control must go is done by means
1874        of setjmp/longjmp. The whole thing is implemented by a set of macros that
1875        hide most of the details from the main code, and operates only if
1876        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1877        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1878        operating.
1879    
1880        To make it easier for callers to provide specially tailored get/free
1881        functions for this usage, two new functions, pcre_stack_malloc, and
1882        pcre_stack_free, are used. They are always called in strict stacking order,
1883        and the size of block requested is always the same.
1884    
1885        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1886        PCRE has been compiled to use the stack or the heap for recursion. The
1887        -C option of pcretest uses this to show which version is compiled.
1888    
1889        A new data escape \S, is added to pcretest; it causes the amounts of store
1890        obtained and freed by both kinds of malloc/free at match time to be added
1891        to the output.
1892    
1893     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1894        what's available on my current Linux desktop machine.
1895    
1896     3. When matching a UTF-8 string, the test for a valid string at the start has
1897        been extended. If start_offset is not zero, PCRE now checks that it points
1898        to a byte that is the start of a UTF-8 character. If not, it returns
1899        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1900        this is necessary because there may be backward assertions in the pattern.
1901        When matching the same subject several times, it may save resources to use
1902        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1903    
1904     4. The code for checking the validity of UTF-8 strings has been tightened so
1905        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1906        containing "overlong sequences".
1907    
1908     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1909        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1910        should have been "&", but it just so happened that all the cases this let
1911        through by mistake were picked up later in the function.
1912    
1913     6. I had used a variable called "isblank" - this is a C99 function, causing
1914        some compilers to warn. To avoid this, I renamed it (as "blankclass").
1915    
1916     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1917        prompting; (b) run "./pcretest /dev/null" at the start of the test script
1918        so the version is shown; (c) stop "make test" echoing "./RunTest".
1919    
1920     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1921    
1922     9. The prototype for memmove() for systems that don't have it was using
1923        size_t, but the inclusion of the header that defines size_t was later. I've
1924        moved the #includes for the C headers earlier to avoid this.
1925    
1926    10. Added some adjustments to the code to make it easier to compiler on certain
1927        special systems:
1928    
1929          (a) Some "const" qualifiers were missing.
1930          (b) Added the macro EXPORT before all exported functions; by default this
1931              is defined to be empty.
1932          (c) Changed the dftables auxiliary program (that builds chartables.c) so
1933              that it reads its output file name as an argument instead of writing
1934              to the standard output and assuming this can be redirected.
1935    
1936    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
1937        class containing characters with values greater than 255, PCRE compilation
1938        went into a loop.
1939    
1940    12. A recursive reference to a subpattern that was within another subpattern
1941        that had a minimum quantifier of zero caused PCRE to crash. For example,
1942        (x(y(?2))z)? provoked this bug with a subject that got as far as the
1943        recursion. If the recursively-called subpattern itself had a zero repeat,
1944        that was OK.
1945    
1946    13. In pcretest, the buffer for reading a data line was set at 30K, but the
1947        buffer into which it was copied (for escape processing) was still set at
1948        1024, so long lines caused crashes.
1949    
1950    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
1951        "internal error: code overflow...". This applied to any character class
1952        that was followed by a possessive quantifier.
1953    
1954    15. Modified the Makefile to add libpcre.la as a prerequisite for
1955        libpcreposix.la because I was told this is needed for a parallel build to
1956        work.
1957    
1958    16. If a pattern that contained .* following optional items at the start was
1959        studied, the wrong optimizing data was generated, leading to matching
1960        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
1961        matching string must start with a or b or c. The correct conclusion for
1962        this pattern is that a match can start with any character.
1963    
1964    
1965    Version 4.4 13-Aug-03
1966    ---------------------
1967    
1968     1. In UTF-8 mode, a character class containing characters with values between
1969        127 and 255 was not handled correctly if the compiled pattern was studied.
1970        In fixing this, I have also improved the studying algorithm for such
1971        classes (slightly).
1972    
1973     2. Three internal functions had redundant arguments passed to them. Removal
1974        might give a very teeny performance improvement.
1975    
1976     3. Documentation bug: the value of the capture_top field in a callout is *one
1977        more than* the number of the hightest numbered captured substring.
1978    
1979     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
1980        in incorrectly linking with a previously installed version. They now link
1981        explicitly with libpcre.la.
1982    
1983     5. configure.in no longer needs to recognize Cygwin specially.
1984    
1985     6. A problem in pcre.in for Windows platforms is fixed.
1986    
1987     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
1988        pcretest, it used to include the size of the study block as part of its
1989        output. Unfortunately, the structure contains a field that has a different
1990        size on different hardware architectures. This meant that the tests that
1991        showed this size failed. As the block is currently always of a fixed size,
1992        this information isn't actually particularly useful in pcretest output, so
1993        I have just removed it.
1994    
1995     8. Three pre-processor statements accidentally did not start in column 1.
1996        Sadly, there are *still* compilers around that complain, even though
1997        standard C has not required this for well over a decade. Sigh.
1998    
1999     9. In pcretest, the code for checking callouts passed small integers in the
2000        callout_data field, which is a void * field. However, some picky compilers
2001        complained about the casts involved for this on 64-bit systems. Now
2002        pcretest passes the address of the small integer instead, which should get
2003        rid of the warnings.
2004    
2005    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
2006        both compile and run time, and gives an error if an invalid UTF-8 sequence
2007        is found. There is a option for disabling this check in cases where the
2008        string is known to be correct and/or the maximum performance is wanted.
2009    
2010    11. In response to a bug report, I changed one line in Makefile.in from
2011    
2012            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
2013        to
2014            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
2015    
2016        to look similar to other lines, but I have no way of telling whether this
2017        is the right thing to do, as I do not use Windows. No doubt I'll get told
2018        if it's wrong...
2019    
2020    
2021    Version 4.3 21-May-03
2022    ---------------------
2023    
2024    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
2025       Makefile.
2026    
2027    2. Some refactoring to improve the quality of the code:
2028    
2029       (i)   The utf8_table... variables are now declared "const".
2030    
2031       (ii)  The code for \cx, which used the "case flipping" table to upper case
2032             lower case letters, now just substracts 32. This is ASCII-specific,
2033             but the whole concept of \cx is ASCII-specific, so it seems
2034             reasonable.
2035    
2036       (iii) PCRE was using its character types table to recognize decimal and
2037             hexadecimal digits in the pattern. This is silly, because it handles
2038             only 0-9, a-f, and A-F, but the character types table is locale-
2039             specific, which means strange things might happen. A private
2040             table is now used for this - though it costs 256 bytes, a table is
2041             much faster than multiple explicit tests. Of course, the standard
2042             character types table is still used for matching digits in subject
2043             strings against \d.
2044    
2045       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
2046             ending in _t are). So I've renamed it as ESC_tee.
2047    
2048    3. The first argument for regexec() in the POSIX wrapper should have been
2049       defined as "const".
2050    
2051    4. Changed pcretest to use malloc() for its buffers so that they can be
2052       Electric Fenced for debugging.
2053    
2054    5. There were several places in the code where, in UTF-8 mode, PCRE would try
2055       to read one or more bytes before the start of the subject string. Often this
2056       had no effect on PCRE's behaviour, but in some circumstances it could
2057       provoke a segmentation fault.
2058    
2059    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
2060       to try to read one or more bytes before the start of the subject string.
2061    
2062    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
2063       UTF-8 support could misbehave in various ways if the subject string
2064       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
2065       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
2066       back over UTF-8 characters.)
2067    
2068    
2069    Version 4.2 14-Apr-03
2070    ---------------------
2071    
2072    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
2073    
2074    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
2075         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
2076         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
2077         [WIN_PREFIX]: new variable, "cyg" for Cygwin
2078         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
2079           and BUILD_EXEEXT
2080         Note: automatic setting of the BUILD variables is not yet working
2081         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
2082           compile-time but not at link-time
2083         [LINK]: use for linking executables only
2084         make different versions for Windows and non-Windows
2085         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
2086           libraries
2087         [LINK_FOR_BUILD]: new variable
2088         [OBJEXT]: use throughout
2089         [EXEEXT]: use throughout
2090         <winshared>: new target
2091         <wininstall>: new target
2092         <dftables.o>: use native compiler
2093         <dftables>: use native linker
2094         <install>: handle Windows platform correctly
2095         <clean>: ditto
2096         <check>: ditto
2097         copy DLL to top builddir before testing
2098    
2099       As part of these changes, -no-undefined was removed again. This was reported
2100       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
2101       in any case.
2102    
2103    3. Some tidies to get rid of compiler warnings:
2104    
2105       . In the match_data structure, match_limit was an unsigned long int, whereas
2106         match_call_count was an int. I've made them both unsigned long ints.
2107    
2108       . In pcretest the fact that a const uschar * doesn't automatically cast to
2109         a void * provoked a warning.
2110    
2111       . Turning on some more compiler warnings threw up some "shadow" variables
2112         and a few more missing casts.
2113    
2114    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
2115       option, a class that contained a single character with a value between 128
2116       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
2117    
2118    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
2119       option, a class that contained several characters, but with at least one
2120       whose value was between 128 and 255 caused PCRE to crash.
2121    
2122    
2123    Version 4.1 12-Mar-03
2124    ---------------------
2125    
2126    1. Compiling with gcc -pedantic found a couple of places where casts were
2127    needed, and a string in dftables.c that was longer than standard compilers are
2128    required to support.
2129    
2130    2. Compiling with Sun's compiler found a few more places where the code could
2131    be tidied up in order to avoid warnings.
2132    
2133    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
2134    first of these names is deprecated in the latest Autoconf in favour of the name
2135    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
2136    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
2137    analogy I have changed it to CFLAGS_FOR_BUILD.
2138    
2139    4. Added -no-undefined to the linking command in the Makefile, because this is
2140    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
2141    linking step for the pcreposix library.
2142    
2143    5. PCRE was failing to diagnose the case of two named groups with the same
2144    name.
2145    
2146    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
2147    literal character that is needed in the subject for a match, and scans along to
2148    ensure that it is present before embarking on the full matching process. This
2149    saves time in cases of nested unlimited repeats that are never going to match.
2150    Problem: the scan can take a lot of time if the subject is very long (e.g.
2151    megabytes), thus penalizing straightforward matches. It is now done only if the
2152    amount of subject to be scanned is less than 1000 bytes.
2153    
2154    7. A lesser problem with the same optimization is that it was recording the
2155    first character of an anchored pattern as "needed", thus provoking a search
2156    right along the subject, even when the first match of the pattern was going to
2157    fail. The "needed" character is now not set for anchored patterns, unless it
2158    follows something in the pattern that is of non-fixed length. Thus, it still
2159    fulfils its original purpose of finding quick non-matches in cases of nested
2160    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
2161    
2162    
2163    Version 4.0 17-Feb-03
2164    ---------------------
2165    
2166    1. If a comment in an extended regex that started immediately after a meta-item
2167    extended to the end of string, PCRE compiled incorrect data. This could lead to
2168    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
2169    
2170    2. Moved to autoconf 2.53 and libtool 1.4.2.
2171    
2172    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
2173    the special perltest8 script is no longer needed - all the tests can be run
2174    from a single perltest script.
2175    
2176    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
2177    by \s. It has now been removed in PCRE. This means it isn't recognized as
2178    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
2179    class [:space:] *does* include VT, thereby creating a mess.
2180    
2181    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
2182    space and tab.
2183    
2184    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
2185    its new features into the main test script, reducing the number of scripts.
2186    
2187    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
2188    were backward compatible, and made the (?i) apply to the whole pattern, as if
2189    /i were given. Now it behaves more logically, and applies the option setting
2190    only to what follows. PCRE has been changed to follow suit. However, if it
2191    finds options settings right at the start of the pattern, it extracts them into
2192    the global options, as before. Thus, they show up in the info data.
2193    
2194    8. Added support for the \Q...\E escape sequence. Characters in between are
2195    treated as literals. This is slightly different from Perl in that $ and @ are
2196    also handled as literals inside the quotes. In Perl, they will cause variable
2197    interpolation. Note the following examples:
2198    
2199        Pattern            PCRE matches      Perl matches
2200    
2201        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
2202        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
2203        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
2204    
2205    For compatibility with Perl, \Q...\E sequences are recognized inside character
2206    classes as well as outside them.
2207    
2208    9. Re-organized 3 code statements in pcretest to avoid "overflow in
2209    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
2210    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
2211    signed/unsigned warnings.
2212    
2213    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
2214    option for pcretest, so I've replaced it by a simple function that does just
2215    that job.
2216    
2217    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
2218    "pcregrep -".
2219    
2220    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
2221    Java package. This provides some syntactic sugar for simple cases of what my
2222    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
2223    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
2224    item, you can use this simplified notation. Note that only makes sense with
2225    greedy quantifiers. Consequently, the use of the possessive quantifier forces
2226    greediness, whatever the setting of the PCRE_UNGREEDY option.
2227    
2228    13. A change of greediness default within a pattern was not taking effect at
2229    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
2230    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
2231    was abstracted outside.
2232    
2233    14. PCRE now supports the \G assertion. It is true when the current matching
2234    position is at the start point of the match. This differs from \A when the
2235    starting offset is non-zero. Used with the /g option of pcretest (or similar
2236    code), it works in the same way as it does for Perl's /g option. If all
2237    alternatives of a regex begin with \G, the expression is anchored to the start
2238    match position, and the "anchored" flag is set in the compiled expression.
2239    
2240    15. Some bugs concerning the handling of certain option changes within patterns
2241    have been fixed. These applied to options other than (?ims). For example,
2242    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
2243    been the other way round. Some of this was related to change 7 above.
2244    
2245    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
2246    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
2247    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
2248    POSIX classes only within a class (e.g. /[[:alpha:]]/).
2249    
2250    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
2251    mode. Unlike ".", it always matches newline, whatever the setting of
2252    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
2253    assertions. Perl allows it, but it doesn't (in general) work because it can't
2254    calculate the length of the lookbehind. At least, that's the case for Perl
2255    5.8.0 - I've been told they are going to document that it doesn't work in
2256    future.
2257    
2258    18. Added an error diagnosis for escapes that PCRE does not support: these are
2259    \L, \l, \N, \P, \p, \U, \u, and \X.
2260    
2261    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
2262    reading past the end of the pattern in cases such as /[abcd/.
2263    
2264    20. PCRE was getting more memory than necessary for patterns with classes that
2265    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
2266    
2267    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
2268    compiling PCRE for use with Virtual Pascal.
2269    
2270    22. Small fix to the Makefile to make it work properly if the build is done
2271    outside the source tree.
2272    
2273    23. Added a new extension: a condition to go with recursion. If a conditional
2274    subpattern starts with (?(R) the "true" branch is used if recursion has
2275    happened, whereas the "false" branch is used only at the top level.
2276    
2277    24. When there was a very long string of literal characters (over 255 bytes
2278    without UTF support, over 250 bytes with UTF support), the computation of how
2279    much memory was required could be incorrect, leading to segfaults or other
2280    strange effects.
2281    
2282    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
2283    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
2284    there was a subsequent back reference to those brackets. This meant that, for
2285    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
2286    possible to check for precisely this case. All we can do is abandon the
2287    optimization if .* occurs inside capturing brackets when there are any back
2288    references whatsoever. (See below for a better fix that came later.)
2289    
2290    26. The handling of the optimization for finding the first character of a
2291    non-anchored pattern, and for finding a character that is required later in the
2292    match were failing in some cases. This didn't break the matching; it just
2293    failed to optimize when it could. The way this is done has been re-implemented.
2294    
2295    27. Fixed typo in error message for invalid (?R item (it said "(?p").
2296    
2297    28. Added a new feature that provides some of the functionality that Perl
2298    provides with (?{...}). The facility is termed a "callout". The way it is done
2299    in PCRE is for the caller to provide an optional function, by setting
2300    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
2301    global variable. By default it is unset, which disables all calling out. To get
2302    the function called, the regex must include (?C) at appropriate points. This
2303    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
2304    This provides a means of identifying different callout points. When PCRE
2305    reaches such a point in the regex, if pcre_callout has been set, the external
2306    function is called. It is provided with data in a structure called
2307    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
2308    matching continues; if it returns a non-zero value, the match at the current
2309    point fails. However, backtracking will occur if possible. [This was changed
2310    later and other features added - see item 49 below.]
2311    
2312    29. pcretest is upgraded to test the callout functionality. It provides a
2313    callout function that displays information. By default, it shows the start of
2314    the match and the current position in the text. There are some new data escapes
2315    to vary what happens:
2316    
2317        \C+         in addition, show current contents of captured substrings
2318        \C-         do not supply a callout function
2319        \C!n        return 1 when callout number n is reached
2320        \C!n!m      return 1 when callout number n is reached for the mth time
2321    
2322    30. If pcregrep was called with the -l option and just a single file name, it
2323    output "<stdin>" if a match was found, instead of the file name.
2324    
2325    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
2326    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
2327    pcre_exec(). This saves a malloc/free per call. The default value of
2328    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
2329    when configuring.
2330    
2331    32. The default maximum size of a compiled pattern is 64K. There have been a
2332    few cases of people hitting this limit. The code now uses macros to handle the
2333    storing of links as offsets within the compiled pattern. It defaults to 2-byte
2334    links, but this can be changed to 3 or 4 bytes by --with-link-size when
2335    configuring. Tests 2 and 5 work only with 2-byte links because they output
2336    debugging information about compiled patterns.
2337    
2338    33. Internal code re-arrangements:
2339    
2340    (a) Moved the debugging function for printing out a compiled regex into
2341        its own source file (printint.c) and used #include to pull it into
2342        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
2343        separate copies.
2344    
2345    (b) Defined the list of op-code names for debugging as a macro in
2346        internal.h so that it is next to the definition of the opcodes.
2347    
2348    (c) Defined a table of op-code lengths for simpler skipping along compiled
2349        code. This is again a macro in internal.h so that it is next to the
2350        definition of the opcodes.
2351    
2352    34. Added support for recursive calls to individual subpatterns, along the
2353    lines of Robin Houston's patch (but implemented somewhat differently).
2354    
2355    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
2356    allow it to read and process whole directories in Win32. This code was
2357    contributed by Lionel Fourquaux; it has not been tested by me.
2358    
2359    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
2360    used to name a group. Names consist of alphanumerics and underscores, and must
2361    be unique. Back references use the syntax (?P=name) and recursive calls use
2362    (?P>name) which is a PCRE extension to the Python extension. Groups still have
2363    numbers. The function pcre_fullinfo() can be used after compilation to extract
2364    a name/number map. There are three relevant calls:
2365    
2366      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
2367      PCRE_INFO_NAMECOUNT            yields the number of entries
2368      PCRE_INFO_NAMETABLE            yields a pointer to the map.
2369    
2370    The map is a vector of fixed-size entries. The size of each entry depends on
2371    the length of the longest name used. The first two bytes of each entry are the
2372    group number, most significant byte first. There follows the corresponding
2373    name, zero terminated. The names are in alphabetical order.
2374    
2375    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
2376    case instead of 255. Making it the same both with and without UTF-8 support
2377    means that the same test output works with both.
2378    
2379    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
2380    calling malloc() with a zero argument.
2381    
2382    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
2383    optimization. I've improved things by keeping a bitmap of backreferences with
2384    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
2385    fact referenced, the optimization can be applied. It is unlikely that a
2386    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
2387    the match to follow \n) will appear inside brackets with a number greater than
2388    31, but if it does, any back reference > 31 suppresses the optimization.
2389    
2390    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
2391    of disabling numbered capturing parentheses. Any opening parenthesis that is
2392    not followed by ? behaves as if it were followed by ?: but named parentheses
2393    can still be used for capturing (and they will acquire numbers in the usual
2394    way).
2395    
2396    41. Redesigned the return codes from the match() function into yes/no/error so
2397    that errors can be passed back from deep inside the nested calls. A malloc
2398    failure while inside a recursive subpattern call now causes the
2399    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
2400    
2401    42. It is now possible to set a limit on the number of times the match()
2402    function is called in a call to pcre_exec(). This facility makes it possible to
2403    limit the amount of recursion and backtracking, though not in a directly
2404    obvious way, because the match() function is used in a number of different
2405    circumstances. The count starts from zero for each position in the subject
2406    string (for non-anchored patterns). The default limit is, for compatibility, a
2407    large number, namely 10 000 000. You can change this in two ways:
2408    
2409    (a) When configuring PCRE before making, you can use --with-match-limit=n
2410        to set a default value for the compiled library.
2411    
2412    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
2413        a different value is set. See 45 below.
2414    
2415    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
2416    
2417    43. Added a new function pcre_config(int, void *) to enable run-time extraction
2418    of things that can be changed at compile time. The first argument specifies
2419    what is wanted and the second points to where the information is to be placed.
2420    The current list of available information is:
2421    
2422      PCRE_CONFIG_UTF8
2423    
2424    The output is an integer that is set to one if UTF-8 support is available;
2425    otherwise it is set to zero.
2426    
2427      PCRE_CONFIG_NEWLINE
2428    
2429    The output is an integer that it set to the value of the code that is used for
2430    newline. It is either LF (10) or CR (13).
2431    
2432      PCRE_CONFIG_LINK_SIZE
2433    
2434    The output is an integer that contains the number of bytes used for internal
2435    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
2436    
2437      PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
2438    
2439    The output is an integer that contains the threshold above which the POSIX
2440    interface uses malloc() for output vectors. See item 31 above.
2441    
2442      PCRE_CONFIG_MATCH_LIMIT
2443    
2444    The output is an unsigned integer that contains the default limit of the number
2445    of match() calls in a pcre_exec() execution. See 42 above.
2446    
2447    44. pcretest has been upgraded by the addition of the -C option. This causes it
2448    to extract all the available output from the new pcre_config() function, and to
2449    output it. The program then exits immediately.
2450    
2451    45. A need has arisen to pass over additional data with calls to pcre_exec() in
2452    order to support additional features. One way would have been to define
2453    pcre_exec2() (for example) with extra arguments, but this would not have been
2454    extensible, and would also have required all calls to the original function to
2455    be mapped to the new one. Instead, I have chosen to extend the mechanism that
2456    is used for passing in "extra" data from pcre_study().
2457    
2458    The pcre_extra structure is now exposed and defined in pcre.h. It currently
2459    contains the following fields:
2460    
2461      flags         a bitmap indicating which of the following fields are set
2462      study_data    opaque data from pcre_study()
2463      match_limit   a way of specifying a limit on match() calls for a specific
2464                      call to pcre_exec()
2465      callout_data  data for callouts (see 49 below)
2466    
2467    The flag bits are also defined in pcre.h, and are
2468    
2469      PCRE_EXTRA_STUDY_DATA
2470      PCRE_EXTRA_MATCH_LIMIT
2471      PCRE_EXTRA_CALLOUT_DATA
2472    
2473    The pcre_study() function now returns one of these new pcre_extra blocks, with
2474    the actual study data pointed to by the study_data field, and the
2475    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
2476    before. That is, this change is entirely upwards-compatible and requires no
2477    change to existing code.
2478    
2479    If you want to pass in additional data to pcre_exec(), you can either place it
2480    in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
2481    block.
2482    
2483    46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
2484    data string contains the escape sequence \M, pcretest calls pcre_exec() several
2485    times with different match limits, until it finds the minimum value needed for
2486    pcre_exec() to complete. The value is then output. This can be instructive; for
2487    most simple matches the number is quite small, but for pathological cases it
2488    gets very large very quickly.
2489    
2490    47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
2491    returns the size of the data block pointed to by the study_data field in a
2492    pcre_extra block, that is, the value that was passed as the argument to
2493    pcre_malloc() when PCRE was getting memory in which to place the information
2494    created by pcre_study(). The fourth argument should point to a size_t variable.
2495    pcretest has been extended so that this information is shown after a successful
2496    pcre_study() call when information about the compiled regex is being displayed.
2497    
2498    48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
2499    because what follows is always an absolute path. (Later: it turns out that this
2500    is more than cosmetic for MinGW, because it doesn't like empty path
2501    components.)
2502    
2503    49. Some changes have been made to the callout feature (see 28 above):
2504    
2505    (i)  A callout function now has three choices for what it returns:
2506    
2507           0  =>  success, carry on matching
2508         > 0  =>  failure at this point, but backtrack if possible
2509         < 0  =>  serious error, return this value from pcre_exec()
2510    
2511         Negative values should normally be chosen from the set of PCRE_ERROR_xxx
2512         values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
2513         "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
2514         use by callout functions. It will never be used by PCRE itself.
2515    
2516    (ii) The pcre_extra structure (see 45 above) has a void * field called
2517         callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
2518         pcre_callout_block structure has a field of the same name. The contents of
2519         the field passed in the pcre_extra structure are passed to the callout
2520         function in the corresponding field in the callout block. This makes it
2521         easier to use the same callout-containing regex from multiple threads. For
2522         testing, the pcretest program has a new data escape
2523    
2524           \C*n        pass the number n (may be negative) as callout_data
2525    
2526         If the callout function in pcretest receives a non-zero value as
2527         callout_data, it returns that value.
2528    
2529    50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
2530    there were some redundant $(CFLAGS) in commands that are now specified as
2531    $(LINK), which already includes $(CFLAGS).
2532    
2533    51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
2534    has been compiled with UTF-8 support *and* pcre_compile() has been compiled
2535    with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
2536    one-byte characters throughout. Note that case-insensitive matching applies
2537    only to characters whose values are less than 256. PCRE doesn't support the
2538    notion of cases for higher-valued characters.
2539    
2540    (i)   A character class whose characters are all within 0-255 is handled as
2541          a bit map, and the map is inverted for negative classes. Previously, a
2542          character > 255 always failed to match such a class; however it should
2543          match if the class was a negative one (e.g. [^ab]). This has been fixed.
2544    
2545    (ii)  A negated character class with a single character < 255 is coded as
2546          "not this character" (OP_NOT). This wasn't working properly when the test
2547          character was multibyte, either singly or repeated.
2548    
2549    (iii) Repeats of multibyte characters are now handled correctly in UTF-8
2550          mode, for example: \x{100}{2,3}.
2551    
2552    (iv)  The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
2553          singly or repeated) now correctly test multibyte characters. However,
2554          PCRE doesn't recognize any characters with values greater than 255 as
2555          digits, spaces, or word characters. Such characters always match \D, \S,
2556          and \W, and never match \d, \s, or \w.
2557    
2558    (v)   Classes may now contain characters and character ranges with values
2559          greater than 255. For example: [ab\x{100}-\x{400}].
2560    
2561    (vi)  pcregrep now has a --utf-8 option (synonym -u) which makes it call
2562          PCRE in UTF-8 mode.
2563    
2564    52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
2565    PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
2566    retained for backwards compatibility. (Note that LASTLITERAL is also a byte
2567    value.)
2568    
2569    53. The single man page has become too large. I have therefore split it up into
2570    a number of separate man pages. These also give rise to individual HTML pages;
2571    these are now put in a separate directory, and there is an index.html page that
2572    lists them all. Some hyperlinking between the pages has been installed.
2573    
2574    54. Added convenience functions for handling named capturing parentheses.
2575    
2576    55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2577    aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2578    true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2579    are faulted.
2580    
2581    56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2582    calling configure. These values are used when compiling the dftables.c program
2583    which is run to generate the source of the default character tables. They
2584    default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2585    you will need to set these values.
2586    
2587    57. Updated the building process for Windows DLL, as provided by Fred Cox.
2588    
2589    
2590    Version 3.9 02-Jan-02
2591    ---------------------
2592    
2593    1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2594    
2595    2. If --disable-static was given, the building process failed when trying to
2596    build pcretest and pcregrep. (For some reason it was using libtool to compile
2597    them, which is not right, as they aren't part of the library.)
2598    
2599    
2600    Version 3.8 18-Dec-01
2601    ---------------------
2602    
2603    1. The experimental UTF-8 code was completely screwed up. It was packing the
2604    bytes in the wrong order. How dumb can you get?
2605    
2606    
2607    Version 3.7 29-Oct-01
2608    ---------------------
2609    
2610    1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2611    This caused pcretest, when used on the test data, to segfault. Unfortunately,
2612    this didn't happen under Solaris 8, where I normally test things.
2613    
2614    2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2615    doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2616    isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2617    this fix an hour or so after the initial 3.7 release.)
2618    
2619    
2620    Version 3.6 23-Oct-01
2621    ---------------------
2622    
2623    1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2624    offsets passed as NULL with zero offset count.
2625    
2626    2. The config.guess and config.sub files had not been updated when I moved to
2627    the latest autoconf.
2628    
2629    
2630    Version 3.5 15-Aug-01
2631    ---------------------
2632    
2633    1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2634    had been forgotten.
2635    
2636    2. By using declared but undefined structures, we can avoid using "void"
2637    definitions in pcre.h while keeping the internal definitions of the structures
2638    private.
2639    
2640    3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2641    user point of view, this means that both static and shared libraries are built
2642    by default, but this can be individually controlled. More of the work of
2643    handling this static/shared cases is now inside libtool instead of PCRE's make
2644    file.
2645    
2646    4. The pcretest utility is now installed along with pcregrep because it is
2647    useful for users (to test regexs) and by doing this, it automatically gets
2648    relinked by libtool. The documentation has been turned into a man page, so
2649    there are now .1, .txt, and .html versions in /doc.
2650    
2651    5. Upgrades to pcregrep:
2652       (i)   Added long-form option names like gnu grep.
2653       (ii)  Added --help to list all options with an explanatory phrase.
2654       (iii) Added -r, --recursive to recurse into sub-directories.
2655       (iv)  Added -f, --file to read patterns from a file.
2656    
2657    6. pcre_exec() was referring to its "code" argument before testing that
2658    argument for NULL (and giving an error if it was NULL).
2659    
2660    7. Upgraded Makefile.in to allow for compiling in a different directory from
2661    the source directory.
2662    
2663    8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2664    options bits, the pointer it was passed was to an int instead of to an unsigned
2665    long int. This mattered only on 64-bit systems.
2666    
2667    9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2668    generated) instead of pcre.in, which it its source. Also made the same change
2669    in several of the .c files.
2670    
2671    10. A new release of gcc defines printf() as a macro, which broke pcretest
2672    because it had an ifdef in the middle of a string argument for printf(). Fixed
2673    by using separate calls to printf().
2674    
2675    11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2676    script, to force use of CR or LF instead of \n in the source. On non-Unix
2677    systems, the value can be set in config.h.
2678    
2679    12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2680    absolute limit. Changed the text of the error message to make this clear, and
2681    likewise updated the man page.
2682    
2683    13. The limit of 99 on the number of capturing subpatterns has been removed.
2684    The new limit is 65535, which I hope will not be a "real" limit.
2685    
2686    
2687  Version 3.4 22-Aug-00  Version 3.4 22-Aug-00
2688  ---------------------  ---------------------

Legend:
Removed from v.51  
changed lines
  Added in v.356

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12