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

Diff of /code/trunk/ChangeLog

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

revision 15 by nigel, Sat Feb 24 21:38:25 2007 UTC revision 344 by ph10, Tue Apr 22 09:16:15 2008 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4    Version 7.7 05-Mar-08
5    ---------------------
6    
7    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
8        a string to a long long, pretend we don't even have a long long." This is
9        done by checking for the strtoq, strtoll, and _strtoi64 functions.
10    
11    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
12        pre-7.6 versions, which defined a global no_arg variable instead of putting
13        it in the RE class. (See also #8 below.)
14    
15    3.  Remove a line of dead code, identified by coverity and reported by Nuno
16        Lopes.
17    
18    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
19    
20        (1) The include/exclude patterns were being applied to the whole pathnames
21            of files, instead of just to the final components.
22    
23        (2) If there was more than one level of directory, the subdirectories were
24            skipped unless they satisfied the include/exclude conditions. This is
25            inconsistent with GNU grep (and could even be seen as contrary to the
26            pcregrep specification - which I improved to make it absolutely clear).
27            The action now is always to scan all levels of directory, and just
28            apply the include/exclude patterns to regular files.
29    
30    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
31        --exclude_dir in the tests to avoid scanning .svn directories.
32    
33    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
34        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
35        doesn't support NULs in patterns.
36    
37    7.  Added some missing "const"s to declarations of static tables in
38        pcre_compile.c and pcre_dfa_exec.c.
39    
40    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
41        caused by fix #2  above. (Subsequently also a second patch to fix the
42        first patch. And a third patch - this was a messy problem.)
43    
44    9.  Applied Craig's patch to remove the use of push_back().
45    
46    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
47        matching function regexec().
48    
49    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
50        which, however, unlike Perl's \g{...}, are subroutine calls, not back
51        references. PCRE supports relative numbers with this syntax (I don't think
52        Oniguruma does).
53    
54    12. Previously, a group with a zero repeat such as (...){0} was completely
55        omitted from the compiled regex. However, this means that if the group
56        was called as a subroutine from elsewhere in the pattern, things went wrong
57        (an internal error was given). Such groups are now left in the compiled
58        pattern, with a new opcode that causes them to be skipped at execution
59        time.
60    
61    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
62        to the way PCRE behaves:
63    
64        (a) A lone ] character is dis-allowed (Perl treats it as data).
65    
66        (b) A back reference to an unmatched subpattern matches an empty string
67            (Perl fails the current match path).
68    
69        (c) A data ] in a character class must be notated as \] because if the
70            first data character in a class is ], it defines an empty class. (In
71            Perl it is not possible to have an empty class.) The empty class []
72            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
73            The negative empty class [^] matches any one character, independently
74            of the DOTALL setting.
75    
76    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
77        non-existent subpattern following a character class starting with ']' and
78        containing () gave an internal compiling error instead of "reference to
79        non-existent subpattern". Fortunately, when the pattern did exist, the
80        compiled code was correct. (When scanning forwards to check for the
81        existencd of the subpattern, it was treating the data ']' as terminating
82        the class, so got the count wrong. When actually compiling, the reference
83        was subsequently set up correctly.)
84    
85    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
86        it was being rejected as not supported by pcre_dfa_exec(), even though
87        other assertions are supported. I have made pcre_dfa_exec() support
88        (*FAIL).
89    
90    16. The implementation of 13c above involved the invention of a new opcode,
91        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
92        cannot be changed at match time, I realized I could make a small
93        improvement to matching performance by compiling OP_ALLANY instead of
94        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
95        on the OP_ANY path.
96    
97    17. Compiling pcretest on Windows with readline support failed without the
98        following two fixes: (1) Make the unistd.h include conditional on
99        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
100    
101    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
102        ncurses library to be included for pcretest when ReadLine support is
103        requested, but also to allow for it to be overridden. This patch came from
104        Daniel Bergström.
105    
106    
107    Version 7.6 28-Jan-08
108    ---------------------
109    
110    1.  A character class containing a very large number of characters with
111        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
112        overflow.
113    
114    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
115        HAVE_LONG_LONG is not defined.
116    
117    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
118        bring them up to date and include new features. This patch includes:
119    
120        - Fixed PH's badly added libz and libbz2 support.
121        - Fixed a problem with static linking.
122        - Added pcredemo. [But later removed - see 7 below.]
123        - Fixed dftables problem and added an option.
124        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
125            HAVE_LONG_LONG.
126        - Added readline support for pcretest.
127        - Added an listing of the option settings after cmake has run.
128    
129    4.  A user submitted a patch to Makefile that makes it easy to create
130        "pcre.dll" under mingw when using Configure/Make. I added stuff to
131        Makefile.am that cause it to include this special target, without
132        affecting anything else. Note that the same mingw target plus all
133        the other distribution libraries and programs are now supported
134        when configuring with CMake (see 6 below) instead of with
135        Configure/Make.
136    
137    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
138        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
139        exported in the Windows port". It has not yet been confirmed that the patch
140        solves the problem, but it does no harm.
141    
142    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
143        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
144        with CMake, and also correct the comment about stack recursion.
145    
146    7.  Remove the automatic building of pcredemo from the ./configure system and
147        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
148        of a program that users should build themselves after PCRE is installed, so
149        building it automatically is not really right. What is more, it gave
150        trouble in some build environments.
151    
152    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
153    
154    
155    Version 7.5 10-Jan-08
156    ---------------------
157    
158    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
159        values in parens when parsing an RE using the C++ wrapper."
160    
161    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
162        Characters greater than 255 were excluded from the class instead of being
163        included.
164    
165    3.  The same bug as (2) above applied to negated POSIX classes such as
166        [:^space:].
167    
168    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
169        defined or documented. It seems to have been a typo for PCRE_STATIC, so
170        I have changed it.
171    
172    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
173        first named subpattern) and a construct such as (?&a) would reference the
174        first named subpattern whose name started with "a" (in other words, the
175        length check was missing). Both these problems are fixed. "Subpattern name
176        expected" is now given for (?&) (a zero-length name), and this patch also
177        makes it give the same error for \k'' (previously it complained that that
178        was a reference to a non-existent subpattern).
179    
180    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
181        this is right because (?- can be followed by option settings as well as by
182        digits. I have, however, made the messages clearer.
183    
184    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
185        than the number used in the conditional) now cause a compile-time error.
186        This is actually not compatible with Perl, which accepts such patterns, but
187        treats the conditional as always being FALSE (as PCRE used to), but it
188        seems to me that giving a diagnostic is better.
189    
190    8.  Change "alphameric" to the more common word "alphanumeric" in comments
191        and messages.
192    
193    9.  Fix two occurrences of "backslash" in comments that should have been
194        "backspace".
195    
196    10. Remove two redundant lines of code that can never be obeyed (their function
197        was moved elsewhere).
198    
199    11. The program that makes PCRE's Unicode character property table had a bug
200        which caused it to generate incorrect table entries for sequences of
201        characters that have the same character type, but are in different scripts.
202        It amalgamated them into a single range, with the script of the first of
203        them. In other words, some characters were in the wrong script. There were
204        thirteen such cases, affecting characters in the following ranges:
205    
206          U+002b0 - U+002c1
207          U+0060c - U+0060d
208          U+0061e - U+00612
209          U+0064b - U+0065e
210          U+0074d - U+0076d
211          U+01800 - U+01805
212          U+01d00 - U+01d77
213          U+01d9b - U+01dbf
214          U+0200b - U+0200f
215          U+030fc - U+030fe
216          U+03260 - U+0327f
217          U+0fb46 - U+0fbb1
218          U+10450 - U+1049d
219    
220    12. The -o option (show only the matching part of a line) for pcregrep was not
221        compatible with GNU grep in that, if there was more than one match in a
222        line, it showed only the first of them. It now behaves in the same way as
223        GNU grep.
224    
225    13. If the -o and -v options were combined for pcregrep, it printed a blank
226        line for every non-matching line. GNU grep prints nothing, and pcregrep now
227        does the same. The return code can be used to tell if there were any
228        non-matching lines.
229    
230    14. Added --file-offsets and --line-offsets to pcregrep.
231    
232    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
233        infinitely looping recursion. The bug was that positive lookaheads were not
234        being skipped when checking for a possible empty match (negative lookaheads
235        and both kinds of lookbehind were skipped).
236    
237    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
238        inclusion of <windows.h> to before rather than after the definition of
239        INVALID_FILE_ATTRIBUTES (patch from David Byron).
240    
241    17. Specifying a possessive quantifier with a specific limit for a Unicode
242        character property caused pcre_compile() to compile bad code, which led at
243        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
244        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
245        caused the error; without that there was no problem.
246    
247    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
248    
249    19. Added --enable-pcretest-libreadline.
250    
251    20. In pcrecpp.cc, the variable 'count' was incremented twice in
252        RE::GlobalReplace(). As a result, the number of replacements returned was
253        double what it should be. I removed one of the increments, but Craig sent a
254        later patch that removed the other one (the right fix) and added unit tests
255        that check the return values (which was not done before).
256    
257    21. Several CMake things:
258    
259        (1) Arranged that, when cmake is used on Unix, the libraries end up with
260            the names libpcre and libpcreposix, not just pcre and pcreposix.
261    
262        (2) The above change means that pcretest and pcregrep are now correctly
263            linked with the newly-built libraries, not previously installed ones.
264    
265        (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
266    
267    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
268        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
269        UTF-8 newline character). The key issue is that the pattern starts .*;
270        this means that the match must be either at the beginning, or after a
271        newline. The bug was in the code for advancing after a failed match and
272        checking that the new position followed a newline. It was not taking
273        account of UTF-8 characters correctly.
274    
275    23. PCRE was behaving differently from Perl in the way it recognized POSIX
276        character classes. PCRE was not treating the sequence [:...:] as a
277        character class unless the ... were all letters. Perl, however, seems to
278        allow any characters between [: and :], though of course it rejects as
279        unknown any "names" that contain non-letters, because all the known class
280        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
281        for example, whereas PCRE did not - it did not recognize a POSIX character
282        class. This seemed a bit dangerous, so the code has been changed to be
283        closer to Perl. The behaviour is not identical to Perl, because PCRE will
284        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
285        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
286        Perl does, and where it didn't before.
287    
288    24. Rewrite so as to remove the single use of %n from pcregrep because in some
289        Windows environments %n is disabled by default.
290    
291    
292    Version 7.4 21-Sep-07
293    ---------------------
294    
295    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
296        means that a class such as [\s] counted as "explicit reference to CR or
297        LF". That isn't really right - the whole point of the change was to try to
298        help when there was an actual mention of one of the two characters. So now
299        the change happens only if \r or \n (or a literal CR or LF) character is
300        encountered.
301    
302    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
303        of both had grown to the point where there were only 3 bits left.
304        Fortunately, there was spare space in the data structure, and so I have
305        moved the internal flags into a new 16-bit field to free up more option
306        bits.
307    
308    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
309        but did not set the internal JCHANGED flag - either of these is enough to
310        control the way the "get" function works - but the PCRE_INFO_JCHANGED
311        facility is supposed to tell if (?J) was ever used, so now (?J) at the
312        start sets both bits.
313    
314    4.  Added options (at build time, compile time, exec time) to change \R from
315        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
316    
317    5.  doc/pcresyntax.html was missing from the distribution.
318    
319    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
320        compatibility, even though it is no longer used.
321    
322    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
323        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
324        windows.h file is present (where different names are used). [This was
325        reversed later after testing - see 16 below.]
326    
327    8.  Changed all #include <config.h> to #include "config.h". There were also
328        some further <pcre.h> cases that I changed to "pcre.h".
329    
330    9.  When pcregrep was used with the --colour option, it missed the line ending
331        sequence off the lines that it output.
332    
333    10. It was pointed out to me that arrays of string pointers cause lots of
334        relocations when a shared library is dynamically loaded. A technique of
335        using a single long string with a table of offsets can drastically reduce
336        these. I have refactored PCRE in four places to do this. The result is
337        dramatic:
338    
339          Originally:                          290
340          After changing UCP table:            187
341          After changing error message table:   43
342          After changing table of "verbs"       36
343          After changing table of Posix names   22
344    
345        Thanks to the folks working on Gregex for glib for this insight.
346    
347    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
348        unicode-properties was also set.
349    
350    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
351    
352    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
353        checked only for CRLF.
354    
355    14. Added casts to pcretest.c to avoid compiler warnings.
356    
357    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
358    
359    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
360        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
361        entirely. This removes changes made in 7 above.
362    
363    17. The CMake files have been updated, and there is now more information about
364        building with CMake in the NON-UNIX-USE document.
365    
366    
367    Version 7.3 28-Aug-07
368    ---------------------
369    
370     1. In the rejigging of the build system that eventually resulted in 7.1, the
371        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
372        brackets there is not right, since it causes compilers to look for an
373        installed pcre.h, not the version that is in the source that is being
374        compiled (which of course may be different). I have changed it back to:
375    
376          #include "pcre.h"
377    
378        I have a vague recollection that the change was concerned with compiling in
379        different directories, but in the new build system, that is taken care of
380        by the VPATH setting the Makefile.
381    
382     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
383        when the subject happened to end in the byte 0x85 (e.g. if the last
384        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
385        characters but of course it shouldn't be taken as a newline when it is part
386        of another character. The bug was that, for an unlimited repeat of . in
387        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
388        characters when looking for a newline.
389    
390     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
391    
392     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
393        in debug output.
394    
395     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
396        long printing in the pcrecpp unittest when running under MinGW.
397    
398     6. ESC_K was left out of the EBCDIC table.
399    
400     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
401        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
402        limit also applies to "virtual nesting" when a pattern is recursive, and in
403        this case 1000 isn't so big. I have been able to remove this limit at the
404        expense of backing off one optimization in certain circumstances. Normally,
405        when pcre_exec() would call its internal match() function recursively and
406        immediately return the result unconditionally, it uses a "tail recursion"
407        feature to save stack. However, when a subpattern that can match an empty
408        string has an unlimited repetition quantifier, it no longer makes this
409        optimization. That gives it a stack frame in which to save the data for
410        checking that an empty string has been matched. Previously this was taken
411        from the 1000-entry workspace that had been reserved. So now there is no
412        explicit limit, but more stack is used.
413    
414     8. Applied Daniel's patches to solve problems with the import/export magic
415        syntax that is required for Windows, and which was going wrong for the
416        pcreposix and pcrecpp parts of the library. These were overlooked when this
417        problem was solved for the main library.
418    
419     9. There were some crude static tests to avoid integer overflow when computing
420        the size of patterns that contain repeated groups with explicit upper
421        limits. As the maximum quantifier is 65535, the maximum group length was
422        set at 30,000 so that the product of these two numbers did not overflow a
423        32-bit integer. However, it turns out that people want to use groups that
424        are longer than 30,000 bytes (though not repeat them that many times).
425        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
426        made it possible to implement the integer overflow checks in a much more
427        dynamic way, which I have now done. The artificial limitation on group
428        length has been removed - we now have only the limit on the total length of
429        the compiled pattern, which depends on the LINK_SIZE setting.
430    
431    10. Fixed a bug in the documentation for get/copy named substring when
432        duplicate names are permitted. If none of the named substrings are set, the
433        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
434        empty string.
435    
436    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
437        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
438        because the ] is interpreted as the first data character and the
439        terminating ] is not found. PCRE has been made compatible with Perl in this
440        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
441        cause memory overwriting.
442    
443    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
444        string has been matched (to stop an infinite loop). It was not recognizing
445        a conditional subpattern that could match an empty string if that
446        subpattern was within another subpattern. For example, it looped when
447        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
448        condition was not nested. This bug has been fixed.
449    
450    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
451        past the start of the subject in the presence of bytes with the top bit
452        set, for example "\x8aBCD".
453    
454    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
455        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
456    
457    14. Optimized (?!) to (*FAIL).
458    
459    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
460        This restricts code points to be within the range 0 to 0x10FFFF, excluding
461        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
462        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
463        does: it's just the validity check that is more restrictive.
464    
465    16. Inserted checks for integer overflows during escape sequence (backslash)
466        processing, and also fixed erroneous offset values for syntax errors during
467        backslash processing.
468    
469    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
470        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
471    
472    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
473        caused an overrun.
474    
475    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
476        something other than just ASCII characters) inside a group that had an
477        unlimited repeat caused a loop at compile time (while checking to see
478        whether the group could match an empty string).
479    
480    20. Debugging a pattern containing \p or \P could cause a crash. For example,
481        [\P{Any}] did so. (Error in the code for printing property names.)
482    
483    21. An orphan \E inside a character class could cause a crash.
484    
485    22. A repeated capturing bracket such as (A)? could cause a wild memory
486        reference during compilation.
487    
488    23. There are several functions in pcre_compile() that scan along a compiled
489        expression for various reasons (e.g. to see if it's fixed length for look
490        behind). There were bugs in these functions when a repeated \p or \P was
491        present in the pattern. These operators have additional parameters compared
492        with \d, etc, and these were not being taken into account when moving along
493        the compiled data. Specifically:
494    
495        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
496            length.
497    
498        (b) An item such as \pL+ within a repeated group could cause crashes or
499            loops.
500    
501        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
502            "reference to non-existent subpattern" error.
503    
504        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
505    
506    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
507        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
508    
509    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
510    
511    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
512        character were causing crashes (broken optimization).
513    
514    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
515        \p or \P) caused a compile-time loop.
516    
517    28. More problems have arisen in unanchored patterns when CRLF is a valid line
518        break. For example, the unstudied pattern [\r\n]A does not match the string
519        "\r\nA" because change 7.0/46 below moves the current point on by two
520        characters after failing to match at the start. However, the pattern \nA
521        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
522        the same is true. There doesn't seem any very clean way out of this, but
523        what I have chosen to do makes the common cases work: PCRE now takes note
524        of whether there can be an explicit match for \r or \n anywhere in the
525        pattern, and if so, 7.0/46 no longer applies. As part of this change,
526        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
527        pattern has explicit CR or LF references.
528    
529    29. Added (*CR) etc for changing newline setting at start of pattern.
530    
531    
532    Version 7.2 19-Jun-07
533    ---------------------
534    
535     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
536        which is apparently normally available under Windows.
537    
538     2. Re-jig the pcregrep tests with different newline settings in an attempt
539        to make them independent of the local environment's newline setting.
540    
541     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
542    
543     4. Some of the "internals" tests were previously cut out when the link size
544        was not 2, because the output contained actual offsets. The recent new
545        "Z" feature of pcretest means that these can be cut out, making the tests
546        usable with all link sizes.
547    
548     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
549        stack recursion. This gives a massive performance boost under BSD, but just
550        a small improvement under Linux. However, it saves one field in the frame
551        in all cases.
552    
553     6. Added more features from the forthcoming Perl 5.10:
554    
555        (a) (?-n) (where n is a string of digits) is a relative subroutine or
556            recursion call. It refers to the nth most recently opened parentheses.
557    
558        (b) (?+n) is also a relative subroutine call; it refers to the nth next
559            to be opened parentheses.
560    
561        (c) Conditions that refer to capturing parentheses can be specified
562            relatively, for example, (?(-2)... or (?(+3)...
563    
564        (d) \K resets the start of the current match so that everything before
565            is not part of it.
566    
567        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
568    
569        (f) \g{name} is another synonym - part of Perl 5.10's unification of
570            reference syntax.
571    
572        (g) (?| introduces a group in which the numbering of parentheses in each
573            alternative starts with the same number.
574    
575        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
576    
577     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
578        PCRE_INFO_JCHANGED.
579    
580     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
581        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
582        for detecting groups that can match an empty string.
583    
584     9. A pattern with a very large number of alternatives (more than several
585        hundred) was running out of internal workspace during the pre-compile
586        phase, where pcre_compile() figures out how much memory will be needed. A
587        bit of new cunning has reduced the workspace needed for groups with
588        alternatives. The 1000-alternative test pattern now uses 12 bytes of
589        workspace instead of running out of the 4096 that are available.
590    
591    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
592    
593    11. Applied patch from Google to remove an optimization that didn't quite work.
594        The report of the bug said:
595    
596          pcrecpp::RE("a*").FullMatch("aaa") matches, while
597          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
598          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
599    
600    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
601        it matched the wrong number of bytes.
602    
603    
604    Version 7.1 24-Apr-07
605    ---------------------
606    
607     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
608        that is more "standard", making use of automake and other Autotools. There
609        is some re-arrangement of the files and adjustment of comments consequent
610        on this.
611    
612     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
613        for recursive directory scanning broke on some systems because the files
614        are not scanned in any specific order and on different systems the order
615        was different. A call to "sort" has been inserted into RunGrepTest for the
616        approprate test as a short-term fix. In the longer term there may be an
617        alternative.
618    
619     3. I had an email from Eric Raymond about problems translating some of PCRE's
620        man pages to HTML (despite the fact that I distribute HTML pages, some
621        people do their own conversions for various reasons). The problems
622        concerned the use of low-level troff macros .br and .in. I have therefore
623        removed all such uses from the man pages (some were redundant, some could
624        be replaced by .nf/.fi pairs). The 132html script that I use to generate
625        HTML has been updated to handle .nf/.fi and to complain if it encounters
626        .br or .in.
627    
628     4. Updated comments in configure.ac that get placed in config.h.in and also
629        arranged for config.h to be included in the distribution, with the name
630        config.h.generic, for the benefit of those who have to compile without
631        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
632    
633     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
634        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
635        makevp.bat for the current PCRE, using the additional files
636        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
637    
638     6. A Windows user reported a minor discrepancy with test 2, which turned out
639        to be caused by a trailing space on an input line that had got lost in his
640        copy. The trailing space was an accident, so I've just removed it.
641    
642     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
643        that is needed.
644    
645     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
646        as "const" (a) because they are and (b) because it helps the PHP
647        maintainers who have recently made a script to detect big data structures
648        in the php code that should be moved to the .rodata section. I remembered
649        to update Builducptable as well, so it won't revert if ucptable.h is ever
650        re-created.
651    
652     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
653        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
654        order to be able to cut out the UTF-8 tables in the latter when UTF-8
655        support is not required. This saves 1.5-2K of code, which is important in
656        some applications.
657    
658        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
659        so as not to refer to the tables, even though these functions will never be
660        called when UTF-8 support is disabled. Otherwise there are problems with a
661        shared library.
662    
663    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
664    
665        (a) It was defining its arguments as char * instead of void *.
666    
667        (b) It was assuming that all moves were upwards in memory; this was true
668            a long time ago when I wrote it, but is no longer the case.
669    
670        The emulated memove() is provided for those environments that have neither
671        memmove() nor bcopy(). I didn't think anyone used it these days, but that
672        is clearly not the case, as these two bugs were recently reported.
673    
674    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
675        and Detrail to create the HTML documentation, the .txt form of the man
676        pages, and it removes trailing spaces from listed files. It also creates
677        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
678        case, it wraps all the #defines with #ifndefs. This script should be run
679        before "make dist".
680    
681    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
682        with Unicode property support.
683    
684        (a) For a maximizing quantifier, if the two different cases of the
685            character were of different lengths in their UTF-8 codings (there are
686            some cases like this - I found 11), and the matching function had to
687            back up over a mixture of the two cases, it incorrectly assumed they
688            were both the same length.
689    
690        (b) When PCRE was configured to use the heap rather than the stack for
691            recursion during matching, it was not correctly preserving the data for
692            the other case of a UTF-8 character when checking ahead for a match
693            while processing a minimizing repeat. If the check also involved
694            matching a wide character, but failed, corruption could cause an
695            erroneous result when trying to check for a repeat of the original
696            character.
697    
698    13. Some tidying changes to the testing mechanism:
699    
700        (a) The RunTest script now detects the internal link size and whether there
701            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
702            values substituted by "configure". (The RunGrepTest script already did
703            this for UTF-8.) The configure.ac script no longer substitutes the
704            relevant variables.
705    
706        (b) The debugging options /B and /D in pcretest show the compiled bytecode
707            with length and offset values. This means that the output is different
708            for different internal link sizes. Test 2 is skipped for link sizes
709            other than 2 because of this, bypassing the problem. Unfortunately,
710            there was also a test in test 3 (the locale tests) that used /B and
711            failed for link sizes other than 2. Rather than cut the whole test out,
712            I have added a new /Z option to pcretest that replaces the length and
713            offset values with spaces. This is now used to make test 3 independent
714            of link size. (Test 2 will be tidied up later.)
715    
716    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
717        segmentation fault instead of returning the appropriate error message.
718    
719    15. In multiline mode when the newline sequence was set to "any", the pattern
720        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
721        This doesn't seem right; it now treats the CRLF combination as the line
722        ending, and so does not match in that case. It's only a pattern such as ^$
723        that would hit this one: something like ^ABC$ would have failed after \r
724        and then tried again after \r\n.
725    
726    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
727        in an attempt to make files that differ only in their line terminators
728        compare equal. This works on Linux.
729    
730    17. Under certain error circumstances pcregrep might try to free random memory
731        as it exited. This is now fixed, thanks to valgrind.
732    
733    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
734        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
735        was because its rules for how to advance for /g after matching an empty
736        string at the end of a line did not allow for this case. They now check for
737        it specially.
738    
739    20. pcretest is supposed to handle patterns and data of any length, by
740        extending its buffers when necessary. It was getting this wrong when the
741        buffer for a data line had to be extended.
742    
743    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
744        CRLF as a newline sequence.
745    
746    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
747        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
748        I have nevertheless tidied it up.
749    
750    23. Added some casts to kill warnings from HP-UX ia64 compiler.
751    
752    24. Added a man page for pcre-config.
753    
754    
755    Version 7.0 19-Dec-06
756    ---------------------
757    
758     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
759        moving to gcc 4.1.1.
760    
761     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
762        sys/time.h, which is documented as needed for this function. It doesn't
763        seem to matter on Linux, but it showed up on some releases of OS X.
764    
765     3. It seems that there are systems where bytes whose values are greater than
766        127 match isprint() in the "C" locale. The "C" locale should be the
767        default when a C program starts up. In most systems, only ASCII printing
768        characters match isprint(). This difference caused the output from pcretest
769        to vary, making some of the tests fail. I have changed pcretest so that:
770    
771        (a) When it is outputting text in the compiled version of a pattern, bytes
772            other than 32-126 are always shown as hex escapes.
773    
774        (b) When it is outputting text that is a matched part of a subject string,
775            it does the same, unless a different locale has been set for the match
776            (using the /L modifier). In this case, it uses isprint() to decide.
777    
778     4. Fixed a major bug that caused incorrect computation of the amount of memory
779        required for a compiled pattern when options that changed within the
780        pattern affected the logic of the preliminary scan that determines the
781        length. The relevant options are -x, and -i in UTF-8 mode. The result was
782        that the computed length was too small. The symptoms of this bug were
783        either the PCRE error "internal error: code overflow" from pcre_compile(),
784        or a glibc crash with a message such as "pcretest: free(): invalid next
785        size (fast)". Examples of patterns that provoked this bug (shown in
786        pcretest format) are:
787    
788          /(?-x: )/x
789          /(?x)(?-x: \s*#\s*)/
790          /((?i)[\x{c0}])/8
791          /(?i:[\x{c0}])/8
792    
793        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
794        is now done differently.
795    
796     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
797        wrapper classes; (b) implement a new function in the C++ scanner that is
798        more efficient than the old way of doing things because it avoids levels of
799        recursion in the regex matching; (c) add a paragraph to the documentation
800        for the FullMatch() function.
801    
802     6. The escape sequence \n was being treated as whatever was defined as
803        "newline". Not only was this contrary to the documentation, which states
804        that \n is character 10 (hex 0A), but it also went horribly wrong when
805        "newline" was defined as CRLF. This has been fixed.
806    
807     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
808        was being set to -1 for the "end of line" case (supposedly a value that no
809        character can have). Though this value is never used (the check for end of
810        line is "zero bytes in current character"), it caused compiler complaints.
811        I've changed it to 0xffffffff.
812    
813     8. In pcre_version.c, the version string was being built by a sequence of
814        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
815        string (as it is for production releases) called a macro with an empty
816        argument. The C standard says the result of this is undefined. The gcc
817        compiler treats it as an empty string (which was what was wanted) but it is
818        reported that Visual C gives an error. The source has been hacked around to
819        avoid this problem.
820    
821     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
822        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
823        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
824        of them did).
825    
826    10. Originally, pcretest opened its input and output without "b"; then I was
827        told that "b" was needed in some environments, so it was added for release
828        5.0 to both the input and output. (It makes no difference on Unix-like
829        systems.) Later I was told that it is wrong for the input on Windows. I've
830        now abstracted the modes into two macros, to make it easier to fiddle with
831        them, and removed "b" from the input mode under Windows.
832    
833    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
834    
835    12. Added -help and --help to pcretest as an official way of being reminded
836        of the options.
837    
838    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
839        and pcrecpp.cc because they annoy compilers at high warning levels.
840    
841    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
842    
843    15. Fixed an occurrence of == in configure.ac that should have been = (shell
844        scripts are not C programs :-) and which was not noticed because it works
845        on Linux.
846    
847    16. pcretest is supposed to handle any length of pattern and data line (as one
848        line or as a continued sequence of lines) by extending its input buffer if
849        necessary. This feature was broken for very long pattern lines, leading to
850        a string of junk being passed to pcre_compile() if the pattern was longer
851        than about 50K.
852    
853    17. I have done a major re-factoring of the way pcre_compile() computes the
854        amount of memory needed for a compiled pattern. Previously, there was code
855        that made a preliminary scan of the pattern in order to do this. That was
856        OK when PCRE was new, but as the facilities have expanded, it has become
857        harder and harder to keep it in step with the real compile phase, and there
858        have been a number of bugs (see for example, 4 above). I have now found a
859        cunning way of running the real compile function in a "fake" mode that
860        enables it to compute how much memory it would need, while actually only
861        ever using a few hundred bytes of working memory and without too many
862        tests of the mode. This should make future maintenance and development
863        easier. A side effect of this work is that the limit of 200 on the nesting
864        depth of parentheses has been removed (though this was never a serious
865        limitation, I suspect). However, there is a downside: pcre_compile() now
866        runs more slowly than before (30% or more, depending on the pattern). I
867        hope this isn't a big issue. There is no effect on runtime performance.
868    
869    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
870        newline (only possible for the last line of a file) and it was a
871        pattern that set a locale (followed by /Lsomething), pcretest crashed.
872    
873    19. Added additional timing features to pcretest. (1) The -tm option now times
874        matching only, not compiling. (2) Both -t and -tm can be followed, as a
875        separate command line item, by a number that specifies the number of
876        repeats to use when timing. The default is 50000; this gives better
877        precision, but takes uncomfortably long for very large patterns.
878    
879    20. Extended pcre_study() to be more clever in cases where a branch of a
880        subpattern has no definite first character. For example, (a*|b*)[cd] would
881        previously give no result from pcre_study(). Now it recognizes that the
882        first character must be a, b, c, or d.
883    
884    21. There was an incorrect error "recursive call could loop indefinitely" if
885        a subpattern (or the entire pattern) that was being tested for matching an
886        empty string contained only one non-empty item after a nested subpattern.
887        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
888        incorrectly, because the \d was being skipped in the check.
889    
890    22. The pcretest program now has a new pattern option /B and a command line
891        option -b, which is equivalent to adding /B to every pattern. This causes
892        it to show the compiled bytecode, without the additional information that
893        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
894        is the same as /B/I).
895    
896    23. A new optimization is now able automatically to treat some sequences such
897        as a*b as a*+b. More specifically, if something simple (such as a character
898        or a simple class like \d) has an unlimited quantifier, and is followed by
899        something that cannot possibly match the quantified thing, the quantifier
900        is automatically "possessified".
901    
902    24. A recursive reference to a subpattern whose number was greater than 39
903        went wrong under certain circumstances in UTF-8 mode. This bug could also
904        have affected the operation of pcre_study().
905    
906    25. Realized that a little bit of performance could be had by replacing
907        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
908    
909    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
910    
911    27. Possessive quantifiers such as a++ were previously implemented by turning
912        them into atomic groups such as ($>a+). Now they have their own opcodes,
913        which improves performance. This includes the automatically created ones
914        from 23 above.
915    
916    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
917        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
918        the first matched character to be a colon. This applied both to named and
919        numbered groups.
920    
921    29. The ucpinternal.h header file was missing its idempotency #ifdef.
922    
923    30. I was sent a "project" file called libpcre.a.dev which I understand makes
924        building PCRE on Windows easier, so I have included it in the distribution.
925    
926    31. There is now a check in pcretest against a ridiculously large number being
927        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
928        loop, the loop is abandoned.
929    
930    32. Forward references to subpatterns in conditions such as (?(2)...) where
931        subpattern 2 is defined later cause pcre_compile() to search forwards in
932        the pattern for the relevant set of parentheses. This search went wrong
933        when there were unescaped parentheses in a character class, parentheses
934        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
935    
936    33. "Subroutine" calls and backreferences were previously restricted to
937        referencing subpatterns earlier in the regex. This restriction has now
938        been removed.
939    
940    34. Added a number of extra features that are going to be in Perl 5.10. On the
941        whole, these are just syntactic alternatives for features that PCRE had
942        previously implemented using the Python syntax or my own invention. The
943        other formats are all retained for compatibility.
944    
945        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
946            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
947            also .NET compatible.
948    
949        (b) A recursion or subroutine call to a named group can now be defined as
950            (?&name) as well as (?P>name).
951    
952        (c) A backreference to a named group can now be defined as \k<name> or
953            \k'name' as well as (?P=name). The new forms, as well as being in Perl
954            5.10, are also .NET compatible.
955    
956        (d) A conditional reference to a named group can now use the syntax
957            (?(<name>) or (?('name') as well as (?(name).
958    
959        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
960            groups (named and numbered) that are never evaluated inline, but can be
961            called as "subroutines" from elsewhere. In effect, the DEFINE condition
962            is always false. There may be only one alternative in such a group.
963    
964        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
965            as the simple (?(R). The condition is true only if the most recent
966            recursion is that of the given number or name. It does not search out
967            through the entire recursion stack.
968    
969        (g) The escape \gN or \g{N} has been added, where N is a positive or
970            negative number, specifying an absolute or relative reference.
971    
972    35. Tidied to get rid of some further signed/unsigned compiler warnings and
973        some "unreachable code" warnings.
974    
975    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
976        things, this adds five new scripts.
977    
978    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
979        There were also incompatibilities regarding the handling of \Q..\E inside
980        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
981        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
982    
983    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
984        matches an empty string, and forcibly breaks the loop. There were bugs in
985        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
986        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
987        separate and independent bugs (that affected different cases) have been
988        fixed.
989    
990    39. Refactored the code to abolish the use of different opcodes for small
991        capturing bracket numbers. This is a tidy that I avoided doing when I
992        removed the limit on the number of capturing brackets for 3.5 back in 2001.
993        The new approach is not only tidier, it makes it possible to reduce the
994        memory needed to fix the previous bug (38).
995    
996    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
997        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
998        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
999        mode.
1000    
1001    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1002        report.
1003    
1004    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1005        copy construction and assignment in the C++ wrapper.
1006    
1007    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1008        couple of bugs that could have given wrong results in the "--newline=crlf"
1009        case.
1010    
1011    44. Added a number of casts and did some reorganization of signed/unsigned int
1012        variables following suggestions from Dair Grant. Also renamed the variable
1013        "this" as "item" because it is a C++ keyword.
1014    
1015    45. Arranged for dftables to add
1016    
1017          #include "pcre_internal.h"
1018    
1019        to pcre_chartables.c because without it, gcc 4.x may remove the array
1020        definition from the final binary if PCRE is built into a static library and
1021        dead code stripping is activated.
1022    
1023    46. For an unanchored pattern, if a match attempt fails at the start of a
1024        newline sequence, and the newline setting is CRLF or ANY, and the next two
1025        characters are CRLF, advance by two characters instead of one.
1026    
1027    
1028    Version 6.7 04-Jul-06
1029    ---------------------
1030    
1031     1. In order to handle tests when input lines are enormously long, pcretest has
1032        been re-factored so that it automatically extends its buffers when
1033        necessary. The code is crude, but this _is_ just a test program. The
1034        default size has been increased from 32K to 50K.
1035    
1036     2. The code in pcre_study() was using the value of the re argument before
1037        testing it for NULL. (Of course, in any sensible call of the function, it
1038        won't be NULL.)
1039    
1040     3. The memmove() emulation function in pcre_internal.h, which is used on
1041        systems that lack both memmove() and bcopy() - that is, hardly ever -
1042        was missing a "static" storage class specifier.
1043    
1044     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1045        containing an extended class (one that cannot be represented by a bitmap
1046        because it contains high-valued characters or Unicode property items, e.g.
1047        [\pZ]). Almost always one would set UTF-8 mode when processing such a
1048        pattern, but PCRE should not loop if you do not (it no longer does).
1049        [Detail: two cases were found: (a) a repeated subpattern containing an
1050        extended class; (b) a recursive reference to a subpattern that followed a
1051        previous extended class. It wasn't skipping over the extended class
1052        correctly when UTF-8 mode was not set.]
1053    
1054     5. A negated single-character class was not being recognized as fixed-length
1055        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1056        compile error "lookbehind assertion is not fixed length".
1057    
1058     6. The RunPerlTest auxiliary script was showing an unexpected difference
1059        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1060        write a Perl script that can interpret lines of an input file either as
1061        byte characters or as UTF-8, which is what "perltest" was being required to
1062        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1063        can't do is switch easily at run time between having the "use utf8;" pragma
1064        or not. In the end, I fudged it by using the RunPerlTest script to insert
1065        "use utf8;" explicitly for the UTF-8 tests.
1066    
1067     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1068        the end of the subject string, contrary to the documentation and to what
1069        Perl does. This was true of both matching functions. Now it matches only at
1070        the start of the subject and immediately after *internal* newlines.
1071    
1072     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1073        a pointer to an int instead of a pointer to an unsigned long int. This
1074        caused problems on 64-bit systems.
1075    
1076     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1077        instance of the 'standard' template library not being so standard".
1078    
1079    10. There was no check on the number of named subpatterns nor the maximum
1080        length of a subpattern name. The product of these values is used to compute
1081        the size of the memory block for a compiled pattern. By supplying a very
1082        long subpattern name and a large number of named subpatterns, the size
1083        computation could be caused to overflow. This is now prevented by limiting
1084        the length of names to 32 characters, and the number of named subpatterns
1085        to 10,000.
1086    
1087    11. Subpatterns that are repeated with specific counts have to be replicated in
1088        the compiled pattern. The size of memory for this was computed from the
1089        length of the subpattern and the repeat count. The latter is limited to
1090        65535, but there was no limit on the former, meaning that integer overflow
1091        could in principle occur. The compiled length of a repeated subpattern is
1092        now limited to 30,000 bytes in order to prevent this.
1093    
1094    12. Added the optional facility to have named substrings with the same name.
1095    
1096    13. Added the ability to use a named substring as a condition, using the
1097        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1098        are numbers (not recommended). Forward references are permitted.
1099    
1100    14. Added forward references in named backreferences (if you see what I mean).
1101    
1102    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1103        pattern could run off the end of the subject. For example, the pattern
1104        "(?s)(.{1,5})"8 did this with the subject "ab".
1105    
1106    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1107        PCRE_CASELESS was set when matching characters that were quantified with ?
1108        or *.
1109    
1110    17. A character class other than a single negated character that had a minimum
1111        but no maximum quantifier - for example [ab]{6,} - was not handled
1112        correctly by pce_dfa_exec(). It would match only one character.
1113    
1114    18. A valid (though odd) pattern that looked like a POSIX character
1115        class but used an invalid character after [ (for example [[,abc,]]) caused
1116        pcre_compile() to give the error "Failed: internal error: code overflow" or
1117        in some cases to crash with a glibc free() error. This could even happen if
1118        the pattern terminated after [[ but there just happened to be a sequence of
1119        letters, a binary zero, and a closing ] in the memory that followed.
1120    
1121    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1122        over the years. Originally (before any Unicode support), just the bottom 8
1123        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1124        output from "man perlunicode" includes this:
1125    
1126          The regular expression compiler produces polymorphic opcodes.  That
1127          is, the pattern adapts to the data and automatically switches to
1128          the Unicode character scheme when presented with Unicode data--or
1129          instead uses a traditional byte scheme when presented with byte
1130          data.
1131    
1132        Sadly, a wide octal escape does not cause a switch, and in a string with
1133        no other multibyte characters, these octal escapes are treated as before.
1134        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
1135        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1136        Unicode string.
1137    
1138        I have not perpetrated such confusion in PCRE. Up till now, it took just
1139        the bottom 8 bits, as in old Perl. I have now made octal escapes with
1140        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1141        translate to the appropriate multibyte character.
1142    
1143    29. Applied some refactoring to reduce the number of warnings from Microsoft
1144        and Borland compilers. This has included removing the fudge introduced
1145        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1146        a warning about an unused variable.
1147    
1148    21. PCRE has not included VT (character 0x0b) in the set of whitespace
1149        characters since release 4.0, because Perl (from release 5.004) does not.
1150        [Or at least, is documented not to: some releases seem to be in conflict
1151        with the documentation.] However, when a pattern was studied with
1152        pcre_study() and all its branches started with \s, PCRE still included VT
1153        as a possible starting character. Of course, this did no harm; it just
1154        caused an unnecessary match attempt.
1155    
1156    22. Removed a now-redundant internal flag bit that recorded the fact that case
1157        dependency changed within the pattern. This was once needed for "required
1158        byte" processing, but is no longer used. This recovers a now-scarce options
1159        bit. Also moved the least significant internal flag bit to the most-
1160        significant bit of the word, which was not previously used (hangover from
1161        the days when it was an int rather than a uint) to free up another bit for
1162        the future.
1163    
1164    23. Added support for CRLF line endings as well as CR and LF. As well as the
1165        default being selectable at build time, it can now be changed at runtime
1166        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1167        specify that it is scanning data with non-default line endings.
1168    
1169    24. Changed the definition of CXXLINK to make it agree with the definition of
1170        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1171    
1172    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1173        recursions. This makes a big different to stack usage for some patterns.
1174    
1175    26. If a subpattern containing a named recursion or subroutine reference such
1176        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1177        the space required for the compiled pattern went wrong and gave too small a
1178        value. Depending on the environment, this could lead to "Failed: internal
1179        error: code overflow at offset 49" or "glibc detected double free or
1180        corruption" errors.
1181    
1182    27. Applied patches from Google (a) to support the new newline modes and (b) to
1183        advance over multibyte UTF-8 characters in GlobalReplace.
1184    
1185    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1186        difference for some implementation of PCRE in some Windows version.
1187    
1188    29. Added some extra testing facilities to pcretest:
1189    
1190        \q<number>   in a data line sets the "match limit" value
1191        \Q<number>   in a data line sets the "match recursion limt" value
1192        -S <number>  sets the stack size, where <number> is in megabytes
1193    
1194        The -S option isn't available for Windows.
1195    
1196    
1197    Version 6.6 06-Feb-06
1198    ---------------------
1199    
1200     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1201        in pcreposix.h. I have copied the definition from pcre.h.
1202    
1203     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1204        because pcre.h is no longer a built file.
1205    
1206     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1207        not normally included in the compiled code.
1208    
1209    
1210    Version 6.5 01-Feb-06
1211    ---------------------
1212    
1213     1. When using the partial match feature with pcre_dfa_exec(), it was not
1214        anchoring the second and subsequent partial matches at the new starting
1215        point. This could lead to incorrect results. For example, with the pattern
1216        /1234/, partially matching against "123" and then "a4" gave a match.
1217    
1218     2. Changes to pcregrep:
1219    
1220        (a) All non-match returns from pcre_exec() were being treated as failures
1221            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1222            error message is output. Some extra information is given for the
1223            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
1224            probably the only errors that are likely to be caused by users (by
1225            specifying a regex that has nested indefinite repeats, for instance).
1226            If there are more than 20 of these errors, pcregrep is abandoned.
1227    
1228        (b) A binary zero was treated as data while matching, but terminated the
1229            output line if it was written out. This has been fixed: binary zeroes
1230            are now no different to any other data bytes.
1231    
1232        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1233            used to set a locale for matching. The --locale=xxxx long option has
1234            been added (no short equivalent) to specify a locale explicitly on the
1235            pcregrep command, overriding the environment variables.
1236    
1237        (d) When -B was used with -n, some line numbers in the output were one less
1238            than they should have been.
1239    
1240        (e) Added the -o (--only-matching) option.
1241    
1242        (f) If -A or -C was used with -c (count only), some lines of context were
1243            accidentally printed for the final match.
1244    
1245        (g) Added the -H (--with-filename) option.
1246    
1247        (h) The combination of options -rh failed to suppress file names for files
1248            that were found from directory arguments.
1249    
1250        (i) Added the -D (--devices) and -d (--directories) options.
1251    
1252        (j) Added the -F (--fixed-strings) option.
1253    
1254        (k) Allow "-" to be used as a file name for -f as well as for a data file.
1255    
1256        (l) Added the --colo(u)r option.
1257    
1258        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1259            is not present by default.
1260    
1261     3. A nasty bug was discovered in the handling of recursive patterns, that is,
1262        items such as (?R) or (?1), when the recursion could match a number of
1263        alternatives. If it matched one of the alternatives, but subsequently,
1264        outside the recursion, there was a failure, the code tried to back up into
1265        the recursion. However, because of the way PCRE is implemented, this is not
1266        possible, and the result was an incorrect result from the match.
1267    
1268        In order to prevent this happening, the specification of recursion has
1269        been changed so that all such subpatterns are automatically treated as
1270        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1271    
1272     4. I had overlooked the fact that, in some locales, there are characters for
1273        which isalpha() is true but neither isupper() nor islower() are true. In
1274        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1275        and ordfeminine) are like this. This affected the treatment of \w and \W
1276        when they appeared in character classes, but not when they appeared outside
1277        a character class. The bit map for "word" characters is now created
1278        separately from the results of isalnum() instead of just taking it from the
1279        upper, lower, and digit maps. (Plus the underscore character, of course.)
1280    
1281     5. The above bug also affected the handling of POSIX character classes such as
1282        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1283        permanent tables. Instead, the bit maps for such a class were previously
1284        created as the appropriate unions of the upper, lower, and digit bitmaps.
1285        Now they are created by subtraction from the [[:word:]] class, which has
1286        its own bitmap.
1287    
1288     6. The [[:blank:]] character class matches horizontal, but not vertical space.
1289        It is created by subtracting the vertical space characters (\x09, \x0a,
1290        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1291        subtraction was done in the overall bitmap for a character class, meaning
1292        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1293        be recognized. This bug has been fixed.
1294    
1295     7. Patches from the folks at Google:
1296    
1297          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1298          real life, but is still worth protecting against".
1299    
1300          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1301          regular expressions".
1302    
1303          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1304          have it.
1305    
1306          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1307          "configure" and the latter not, in order to fix a problem somebody had
1308          with compiling the Arg class on HP-UX.
1309    
1310          (e) Improve the error-handling of the C++ wrapper a little bit.
1311    
1312          (f) New tests for checking recursion limiting.
1313    
1314     8. The pcre_memmove() function, which is used only if the environment does not
1315        have a standard memmove() function (and is therefore rarely compiled),
1316        contained two bugs: (a) use of int instead of size_t, and (b) it was not
1317        returning a result (though PCRE never actually uses the result).
1318    
1319     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1320        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1321        returned instead of calling malloc() with an overflowing number that would
1322        most likely cause subsequent chaos.
1323    
1324    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1325    
1326    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1327        with this option is matched, the nmatch and pmatch options of regexec() are
1328        ignored.
1329    
1330    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1331        provided in case anyone wants to the the POSIX interface with UTF-8
1332        strings.
1333    
1334    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1335        C++ linking (needed for some HP-UX environments).
1336    
1337    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1338        (unused parameter) and in the pcre_printint() function (omitted "default"
1339        switch label when the default is to do nothing).
1340    
1341    15. Added some code to make it possible, when PCRE is compiled as a C++
1342        library, to replace subject pointers for pcre_exec() with a smart pointer
1343        class, thus making it possible to process discontinuous strings.
1344    
1345    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1346        much the same function. They were added by different people who were trying
1347        to make PCRE easy to compile on non-Unix systems. It has been suggested
1348        that PCRE_EXPORT be abolished now that there is more automatic apparatus
1349        for compiling on Windows systems. I have therefore replaced it with
1350        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1351        defaults to "extern" for C or "extern C" for C++, which works fine on
1352        Unix-like systems. It is now possible to override the value of PCRE_DATA_
1353        SCOPE with something explicit in config.h. In addition:
1354    
1355        (a) pcreposix.h still had just "extern" instead of either of these macros;
1356            I have replaced it with PCRE_DATA_SCOPE.
1357    
1358        (b) Functions such as _pcre_xclass(), which are internal to the library,
1359            but external in the C sense, all had PCRE_EXPORT in their definitions.
1360            This is apparently wrong for the Windows case, so I have removed it.
1361            (It makes no difference on Unix-like systems.)
1362    
1363    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1364        of recursive calls to match(). This is different to MATCH_LIMIT because
1365        that limits the total number of calls to match(), not all of which increase
1366        the depth of recursion. Limiting the recursion depth limits the amount of
1367        stack (or heap if NO_RECURSE is set) that is used. The default can be set
1368        when PCRE is compiled, and changed at run time. A patch from Google adds
1369        this functionality to the C++ interface.
1370    
1371    18. Changes to the handling of Unicode character properties:
1372    
1373        (a) Updated the table to Unicode 4.1.0.
1374    
1375        (b) Recognize characters that are not in the table as "Cn" (undefined).
1376    
1377        (c) I revised the way the table is implemented to a much improved format
1378            which includes recognition of ranges. It now supports the ranges that
1379            are defined in UnicodeData.txt, and it also amalgamates other
1380            characters into ranges. This has reduced the number of entries in the
1381            table from around 16,000 to around 3,000, thus reducing its size
1382            considerably. I realized I did not need to use a tree structure after
1383            all - a binary chop search is just as efficient. Having reduced the
1384            number of entries, I extended their size from 6 bytes to 8 bytes to
1385            allow for more data.
1386    
1387        (d) Added support for Unicode script names via properties such as \p{Han}.
1388    
1389    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1390        matching that character.
1391    
1392    20. When matching a repeated Unicode property with a minimum greater than zero,
1393        (for example \pL{2,}), PCRE could look past the end of the subject if it
1394        reached it while seeking the minimum number of characters. This could
1395        happen only if some of the characters were more than one byte long, because
1396        there is a check for at least the minimum number of bytes.
1397    
1398    21. Refactored the implementation of \p and \P so as to be more general, to
1399        allow for more different types of property in future. This has changed the
1400        compiled form incompatibly. Anybody with saved compiled patterns that use
1401        \p or \P will have to recompile them.
1402    
1403    22. Added "Any" and "L&" to the supported property types.
1404    
1405    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1406        but give a compile time error if the value is greater than 0xff.
1407    
1408    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1409        accidentally not being installed or uninstalled.
1410    
1411    25. The pcre.h file was built from pcre.h.in, but the only changes that were
1412        made were to insert the current release number. This seemed silly, because
1413        it made things harder for people building PCRE on systems that don't run
1414        "configure". I have turned pcre.h into a distributed file, no longer built
1415        by "configure", with the version identification directly included. There is
1416        no longer a pcre.h.in file.
1417    
1418        However, this change necessitated a change to the pcre-config script as
1419        well. It is built from pcre-config.in, and one of the substitutions was the
1420        release number. I have updated configure.ac so that ./configure now finds
1421        the release number by grepping pcre.h.
1422    
1423    26. Added the ability to run the tests under valgrind.
1424    
1425    
1426    Version 6.4 05-Sep-05
1427    ---------------------
1428    
1429     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1430        "--" to be printed when multiple files were scanned, even when none of the
1431        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1432        consider it to be a bug, and have restored the previous behaviour.
1433    
1434     2. A couple of code tidies to get rid of compiler warnings.
1435    
1436     3. The pcretest program used to cheat by referring to symbols in the library
1437        whose names begin with _pcre_. These are internal symbols that are not
1438        really supposed to be visible externally, and in some environments it is
1439        possible to suppress them. The cheating is now confined to including
1440        certain files from the library's source, which is a bit cleaner.
1441    
1442     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1443        file's purpose clearer.
1444    
1445     5. Reorganized pcre_ucp_findchar().
1446    
1447    
1448    Version 6.3 15-Aug-05
1449    ---------------------
1450    
1451     1. The file libpcre.pc.in did not have general read permission in the tarball.
1452    
1453     2. There were some problems when building without C++ support:
1454    
1455        (a) If C++ support was not built, "make install" and "make test" still
1456            tried to test it.
1457    
1458        (b) There were problems when the value of CXX was explicitly set. Some
1459            changes have been made to try to fix these, and ...
1460    
1461        (c) --disable-cpp can now be used to explicitly disable C++ support.
1462    
1463        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1464            backslash in a target when C++ was disabled. This confuses some
1465            versions of "make", apparently. Using an intermediate variable solves
1466            this. (Same for CPP_LOBJ.)
1467    
1468     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1469        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1470        necessary on certain architectures.
1471    
1472     4. Added a setting of -export-symbols-regex to the link command to remove
1473        those symbols that are exported in the C sense, but actually are local
1474        within the library, and not documented. Their names all begin with
1475        "_pcre_". This is not a perfect job, because (a) we have to except some
1476        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1477        available (and never for static libraries). I have made a note to try to
1478        find a way round (a) in the future.
1479    
1480    
1481    Version 6.2 01-Aug-05
1482    ---------------------
1483    
1484     1. There was no test for integer overflow of quantifier values. A construction
1485        such as {1111111111111111} would give undefined results. What is worse, if
1486        a minimum quantifier for a parenthesized subpattern overflowed and became
1487        negative, the calculation of the memory size went wrong. This could have
1488        led to memory overwriting.
1489    
1490     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1491    
1492     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1493        operating environments where this matters.
1494    
1495     4. Applied Giuseppe Maxia's patch to add additional features for controlling
1496        PCRE options from within the C++ wrapper.
1497    
1498     5. Named capturing subpatterns were not being correctly counted when a pattern
1499        was compiled. This caused two problems: (a) If there were more than 100
1500        such subpatterns, the calculation of the memory needed for the whole
1501        compiled pattern went wrong, leading to an overflow error. (b) Numerical
1502        back references of the form \12, where the number was greater than 9, were
1503        not recognized as back references, even though there were sufficient
1504        previous subpatterns.
1505    
1506     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1507        versions of gcc, e.g. 2.95.4.
1508    
1509    
1510    Version 6.1 21-Jun-05
1511    ---------------------
1512    
1513     1. There was one reference to the variable "posix" in pcretest.c that was not
1514        surrounded by "#if !defined NOPOSIX".
1515    
1516     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1517        the cross-check on the old pcre_info() function, for the benefit of the
1518        cut-down version of PCRE that is currently imported into Exim.
1519    
1520     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1521        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1522        patterns that start (?i) but I don't think that matters. The use of (?i) is
1523        just an example; this all applies to the other options as well.
1524    
1525     4. Since libtool seems to echo the compile commands it is issuing, the output
1526        from "make" can be reduced a bit by putting "@" in front of each libtool
1527        compile command.
1528    
1529     5. Patch from the folks at Google for configure.in to be a bit more thorough
1530        in checking for a suitable C++ installation before trying to compile the
1531        C++ stuff. This should fix a reported problem when a compiler was present,
1532        but no suitable headers.
1533    
1534     6. The man pages all had just "PCRE" as their title. I have changed them to
1535        be the relevant file name. I have also arranged that these names are
1536        retained in the file doc/pcre.txt, which is a concatenation in text format
1537        of all the man pages except the little individual ones for each function.
1538    
1539     7. The NON-UNIX-USE file had not been updated for the different set of source
1540        files that come with release 6. I also added a few comments about the C++
1541        wrapper.
1542    
1543    
1544    Version 6.0 07-Jun-05
1545    ---------------------
1546    
1547     1. Some minor internal re-organization to help with my DFA experiments.
1548    
1549     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1550        didn't matter for the library itself when fully configured, but did matter
1551        when compiling without UCP support, or within Exim, where the ucp files are
1552        not imported.
1553    
1554     3. Refactoring of the library code to split up the various functions into
1555        different source modules. The addition of the new DFA matching code (see
1556        below) to a single monolithic source would have made it really too
1557        unwieldy, quite apart from causing all the code to be include in a
1558        statically linked application, when only some functions are used. This is
1559        relevant even without the DFA addition now that patterns can be compiled in
1560        one application and matched in another.
1561    
1562        The downside of splitting up is that there have to be some external
1563        functions and data tables that are used internally in different modules of
1564        the library but which are not part of the API. These have all had their
1565        names changed to start with "_pcre_" so that they are unlikely to clash
1566        with other external names.
1567    
1568     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1569        a different (DFA) algorithm. Although it is slower than the original
1570        function, it does have some advantages for certain types of matching
1571        problem.
1572    
1573     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1574        including restarting after a partial match.
1575    
1576     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1577        defined when compiling for Windows was sent to me. I have put it into the
1578        code, though I have no means of testing or verifying it.
1579    
1580     7. Added the pcre_refcount() auxiliary function.
1581    
1582     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1583        match before or at the first newline in the subject string. In pcretest,
1584        the /f option on a pattern can be used to set this.
1585    
1586     9. A repeated \w when used in UTF-8 mode with characters greater than 256
1587        would behave wrongly. This has been present in PCRE since release 4.0.
1588    
1589    10. A number of changes to the pcregrep command:
1590    
1591        (a) Refactored how -x works; insert ^(...)$ instead of setting
1592            PCRE_ANCHORED and checking the length, in preparation for adding
1593            something similar for -w.
1594    
1595        (b) Added the -w (match as a word) option.
1596    
1597        (c) Refactored the way lines are read and buffered so as to have more
1598            than one at a time available.
1599    
1600        (d) Implemented a pcregrep test script.
1601    
1602        (e) Added the -M (multiline match) option. This allows patterns to match
1603            over several lines of the subject. The buffering ensures that at least
1604            8K, or the rest of the document (whichever is the shorter) is available
1605            for matching (and similarly the previous 8K for lookbehind assertions).
1606    
1607        (f) Changed the --help output so that it now says
1608    
1609              -w, --word-regex(p)
1610    
1611            instead of two lines, one with "regex" and the other with "regexp"
1612            because that confused at least one person since the short forms are the
1613            same. (This required a bit of code, as the output is generated
1614            automatically from a table. It wasn't just a text change.)
1615    
1616        (g) -- can be used to terminate pcregrep options if the next thing isn't an
1617            option but starts with a hyphen. Could be a pattern or a path name
1618            starting with a hyphen, for instance.
1619    
1620        (h) "-" can be given as a file name to represent stdin.
1621    
1622        (i) When file names are being printed, "(standard input)" is used for
1623            the standard input, for compatibility with GNU grep. Previously
1624            "<stdin>" was used.
1625    
1626        (j) The option --label=xxx can be used to supply a name to be used for
1627            stdin when file names are being printed. There is no short form.
1628    
1629        (k) Re-factored the options decoding logic because we are going to add
1630            two more options that take data. Such options can now be given in four
1631            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1632    
1633        (l) Added the -A, -B, and -C options for requesting that lines of context
1634            around matches be printed.
1635    
1636        (m) Added the -L option to print the names of files that do not contain
1637            any matching lines, that is, the complement of -l.
1638    
1639        (n) The return code is 2 if any file cannot be opened, but pcregrep does
1640            continue to scan other files.
1641    
1642        (o) The -s option was incorrectly implemented. For compatibility with other
1643            greps, it now suppresses the error message for a non-existent or non-
1644            accessible file (but not the return code). There is a new option called
1645            -q that suppresses the output of matching lines, which was what -s was
1646            previously doing.
1647    
1648        (p) Added --include and --exclude options to specify files for inclusion
1649            and exclusion when recursing.
1650    
1651    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1652        Hopefully, it now does.
1653    
1654    12. Missing cast in pcre_study().
1655    
1656    13. Added an "uninstall" target to the makefile.
1657    
1658    14. Replaced "extern" in the function prototypes in Makefile.in with
1659        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1660        world, but is set differently for Windows.
1661    
1662    15. Added a second compiling function called pcre_compile2(). The only
1663        difference is that it has an extra argument, which is a pointer to an
1664        integer error code. When there is a compile-time failure, this is set
1665        non-zero, in addition to the error test pointer being set to point to an
1666        error message. The new argument may be NULL if no error number is required
1667        (but then you may as well call pcre_compile(), which is now just a
1668        wrapper). This facility is provided because some applications need a
1669        numeric error indication, but it has also enabled me to tidy up the way
1670        compile-time errors are handled in the POSIX wrapper.
1671    
1672    16. Added VPATH=.libs to the makefile; this should help when building with one
1673        prefix path and installing with another. (Or so I'm told by someone who
1674        knows more about this stuff than I do.)
1675    
1676    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1677        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1678        match everything, including newlines. This is not POSIX-compatible, but
1679        somebody wanted the feature. From pcretest it can be activated by using
1680        both the P and the s flags.
1681    
1682    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1683    
1684    19. libpcre.pc was being incorrectly installed as executable.
1685    
1686    20. A couple of places in pcretest check for end-of-line by looking for '\n';
1687        it now also looks for '\r' so that it will work unmodified on Windows.
1688    
1689    21. Added Google's contributed C++ wrapper to the distribution.
1690    
1691    22. Added some untidy missing memory free() calls in pcretest, to keep
1692        Electric Fence happy when testing.
1693    
1694    
1695    
1696    Version 5.0 13-Sep-04
1697    ---------------------
1698    
1699     1. Internal change: literal characters are no longer packed up into items
1700        containing multiple characters in a single byte-string. Each character
1701        is now matched using a separate opcode. However, there may be more than one
1702        byte in the character in UTF-8 mode.
1703    
1704     2. The pcre_callout_block structure has two new fields: pattern_position and
1705        next_item_length. These contain the offset in the pattern to the next match
1706        item, and its length, respectively.
1707    
1708     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1709        insertion of callouts before each pattern item. Added the /C option to
1710        pcretest to make use of this.
1711    
1712     4. On the advice of a Windows user, the lines
1713    
1714          #if defined(_WIN32) || defined(WIN32)
1715          _setmode( _fileno( stdout ), 0x8000 );
1716          #endif  /* defined(_WIN32) || defined(WIN32) */
1717    
1718        have been added to the source of pcretest. This apparently does useful
1719        magic in relation to line terminators.
1720    
1721     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1722        for the benefit of those environments where the "b" makes a difference.
1723    
1724     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1725        to know about it. I have put a hack into configure.in that adds in code
1726        to set GCC=yes if CC=icc. This seems to end up at a point in the
1727        generated configure script that is early enough to affect the setting of
1728        compiler options, which is what is needed, but I have no means of testing
1729        whether it really works. (The user who reported this had patched the
1730        generated configure script, which of course I cannot do.)
1731    
1732        LATER: After change 22 below (new libtool files), the configure script
1733        seems to know about icc (and also ecc). Therefore, I have commented out
1734        this hack in configure.in.
1735    
1736     7. Added support for pkg-config (2 patches were sent in).
1737    
1738     8. Negated POSIX character classes that used a combination of internal tables
1739        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1740        [[:^ascii]]. Typically, they would match almost any characters. The other
1741        POSIX classes were not broken in this way.
1742    
1743     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1744        to find the match, as PCRE was deluded into thinking that the match had to
1745        start at the start point or following a newline. The same bug applied to
1746        patterns with negative forward assertions or any backward assertions
1747        preceding ".*" at the start, unless the pattern required a fixed first
1748        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1749    
1750    10. In UTF-8 mode, when moving forwards in the subject after a failed match
1751        starting at the last subject character, bytes beyond the end of the subject
1752        string were read.
1753    
1754    11. Renamed the variable "class" as "classbits" to make life easier for C++
1755        users. (Previously there was a macro definition, but it apparently wasn't
1756        enough.)
1757    
1758    12. Added the new field "tables" to the extra data so that tables can be passed
1759        in at exec time, or the internal tables can be re-selected. This allows
1760        a compiled regex to be saved and re-used at a later time by a different
1761        program that might have everything at different addresses.
1762    
1763    13. Modified the pcre-config script so that, when run on Solaris, it shows a
1764        -R library as well as a -L library.
1765    
1766    14. The debugging options of pcretest (-d on the command line or D on a
1767        pattern) showed incorrect output for anything following an extended class
1768        that contained multibyte characters and which was followed by a quantifier.
1769    
1770    15. Added optional support for general category Unicode character properties
1771        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1772        support. It adds about 90K to the size of the library. The meanings of the
1773        inbuilt class escapes such as \d and \s have NOT been changed.
1774    
1775    16. Updated pcredemo.c to include calls to free() to release the memory for the
1776        compiled pattern.
1777    
1778    17. The generated file chartables.c was being created in the source directory
1779        instead of in the building directory. This caused the build to fail if the
1780        source directory was different from the building directory, and was
1781        read-only.
1782    
1783    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1784        file. No doubt somebody will tell me if they don't make sense... Also added
1785        Dan Mooney's comments about building on OpenVMS.
1786    
1787    19. Added support for partial matching via the PCRE_PARTIAL option for
1788        pcre_exec() and the \P data escape in pcretest.
1789    
1790    20. Extended pcretest with 3 new pattern features:
1791    
1792        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
1793              write the compiled pattern to the file whose name is "rest-of-line".
1794              This is a straight binary dump of the data, with the saved pointer to
1795              the character tables forced to be NULL. The study data, if any, is
1796              written too. After writing, pcretest reads a new pattern.
1797    
1798        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1799              compiled pattern from the given file. There must not be any
1800              occurrences of "<" in the file name (pretty unlikely); if there are,
1801              pcretest will instead treat the initial "<" as a pattern delimiter.
1802              After reading in the pattern, pcretest goes on to read data lines as
1803              usual.
1804    
1805        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1806              and 16-bit fields in a compiled pattern, to simulate a pattern that
1807              was compiled on a host of opposite endianness.
1808    
1809    21. The pcre-exec() function can now cope with patterns that were compiled on
1810        hosts of opposite endianness, with this restriction:
1811    
1812          As for any compiled expression that is saved and used later, the tables
1813          pointer field cannot be preserved; the extra_data field in the arguments
1814          to pcre_exec() should be used to pass in a tables address if a value
1815          other than the default internal tables were used at compile time.
1816    
1817    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1818        now diagnosed as an error. Previously, most of the time, a negative number
1819        would have been treated as zero, but if in addition "ovector" was passed as
1820        NULL, a crash could occur.
1821    
1822    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1823        new versions from the libtool 1.5 distribution (the last one is a copy of
1824        a file called libtool.m4). This seems to have fixed the need to patch
1825        "configure" to support Darwin 1.3 (which I used to do). However, I still
1826        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1827        workstation).
1828    
1829    24. Changed the PCRE licence to be the more standard "BSD" licence.
1830    
1831    
1832    Version 4.5 01-Dec-03
1833    ---------------------
1834    
1835     1. There has been some re-arrangement of the code for the match() function so
1836        that it can be compiled in a version that does not call itself recursively.
1837        Instead, it keeps those local variables that need separate instances for
1838        each "recursion" in a frame on the heap, and gets/frees frames whenever it
1839        needs to "recurse". Keeping track of where control must go is done by means
1840        of setjmp/longjmp. The whole thing is implemented by a set of macros that
1841        hide most of the details from the main code, and operates only if
1842        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1843        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1844        operating.
1845    
1846        To make it easier for callers to provide specially tailored get/free
1847        functions for this usage, two new functions, pcre_stack_malloc, and
1848        pcre_stack_free, are used. They are always called in strict stacking order,
1849        and the size of block requested is always the same.
1850    
1851        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1852        PCRE has been compiled to use the stack or the heap for recursion. The
1853        -C option of pcretest uses this to show which version is compiled.
1854    
1855        A new data escape \S, is added to pcretest; it causes the amounts of store
1856        obtained and freed by both kinds of malloc/free at match time to be added
1857        to the output.
1858    
1859     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1860        what's available on my current Linux desktop machine.
1861    
1862     3. When matching a UTF-8 string, the test for a valid string at the start has
1863        been extended. If start_offset is not zero, PCRE now checks that it points
1864        to a byte that is the start of a UTF-8 character. If not, it returns
1865        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1866        this is necessary because there may be backward assertions in the pattern.
1867        When matching the same subject several times, it may save resources to use
1868        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1869    
1870     4. The code for checking the validity of UTF-8 strings has been tightened so
1871        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1872        containing "overlong sequences".
1873    
1874     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1875        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1876        should have been "&", but it just so happened that all the cases this let
1877        through by mistake were picked up later in the function.
1878    
1879     6. I had used a variable called "isblank" - this is a C99 function, causing
1880        some compilers to warn. To avoid this, I renamed it (as "blankclass").
1881    
1882     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1883        prompting; (b) run "./pcretest /dev/null" at the start of the test script
1884        so the version is shown; (c) stop "make test" echoing "./RunTest".
1885    
1886     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1887    
1888     9. The prototype for memmove() for systems that don't have it was using
1889        size_t, but the inclusion of the header that defines size_t was later. I've
1890        moved the #includes for the C headers earlier to avoid this.
1891    
1892    10. Added some adjustments to the code to make it easier to compiler on certain
1893        special systems:
1894    
1895          (a) Some "const" qualifiers were missing.
1896          (b) Added the macro EXPORT before all exported functions; by default this
1897              is defined to be empty.
1898          (c) Changed the dftables auxiliary program (that builds chartables.c) so
1899              that it reads its output file name as an argument instead of writing
1900              to the standard output and assuming this can be redirected.
1901    
1902    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
1903        class containing characters with values greater than 255, PCRE compilation
1904        went into a loop.
1905    
1906    12. A recursive reference to a subpattern that was within another subpattern
1907        that had a minimum quantifier of zero caused PCRE to crash. For example,
1908        (x(y(?2))z)? provoked this bug with a subject that got as far as the
1909        recursion. If the recursively-called subpattern itself had a zero repeat,
1910        that was OK.
1911    
1912    13. In pcretest, the buffer for reading a data line was set at 30K, but the
1913        buffer into which it was copied (for escape processing) was still set at
1914        1024, so long lines caused crashes.
1915    
1916    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
1917        "internal error: code overflow...". This applied to any character class
1918        that was followed by a possessive quantifier.
1919    
1920    15. Modified the Makefile to add libpcre.la as a prerequisite for
1921        libpcreposix.la because I was told this is needed for a parallel build to
1922        work.
1923    
1924    16. If a pattern that contained .* following optional items at the start was
1925        studied, the wrong optimizing data was generated, leading to matching
1926        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
1927        matching string must start with a or b or c. The correct conclusion for
1928        this pattern is that a match can start with any character.
1929    
1930    
1931    Version 4.4 13-Aug-03
1932    ---------------------
1933    
1934     1. In UTF-8 mode, a character class containing characters with values between
1935        127 and 255 was not handled correctly if the compiled pattern was studied.
1936        In fixing this, I have also improved the studying algorithm for such
1937        classes (slightly).
1938    
1939     2. Three internal functions had redundant arguments passed to them. Removal
1940        might give a very teeny performance improvement.
1941    
1942     3. Documentation bug: the value of the capture_top field in a callout is *one
1943        more than* the number of the hightest numbered captured substring.
1944    
1945     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
1946        in incorrectly linking with a previously installed version. They now link
1947        explicitly with libpcre.la.
1948    
1949     5. configure.in no longer needs to recognize Cygwin specially.
1950    
1951     6. A problem in pcre.in for Windows platforms is fixed.
1952    
1953     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
1954        pcretest, it used to include the size of the study block as part of its
1955        output. Unfortunately, the structure contains a field that has a different
1956        size on different hardware architectures. This meant that the tests that
1957        showed this size failed. As the block is currently always of a fixed size,
1958        this information isn't actually particularly useful in pcretest output, so
1959        I have just removed it.
1960    
1961     8. Three pre-processor statements accidentally did not start in column 1.
1962        Sadly, there are *still* compilers around that complain, even though
1963        standard C has not required this for well over a decade. Sigh.
1964    
1965     9. In pcretest, the code for checking callouts passed small integers in the
1966        callout_data field, which is a void * field. However, some picky compilers
1967        complained about the casts involved for this on 64-bit systems. Now
1968        pcretest passes the address of the small integer instead, which should get
1969        rid of the warnings.
1970    
1971    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
1972        both compile and run time, and gives an error if an invalid UTF-8 sequence
1973        is found. There is a option for disabling this check in cases where the
1974        string is known to be correct and/or the maximum performance is wanted.
1975    
1976    11. In response to a bug report, I changed one line in Makefile.in from
1977    
1978            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
1979        to
1980            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
1981    
1982        to look similar to other lines, but I have no way of telling whether this
1983        is the right thing to do, as I do not use Windows. No doubt I'll get told
1984        if it's wrong...
1985    
1986    
1987    Version 4.3 21-May-03
1988    ---------------------
1989    
1990    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
1991       Makefile.
1992    
1993    2. Some refactoring to improve the quality of the code:
1994    
1995       (i)   The utf8_table... variables are now declared "const".
1996    
1997       (ii)  The code for \cx, which used the "case flipping" table to upper case
1998             lower case letters, now just substracts 32. This is ASCII-specific,
1999             but the whole concept of \cx is ASCII-specific, so it seems
2000             reasonable.
2001    
2002       (iii) PCRE was using its character types table to recognize decimal and
2003             hexadecimal digits in the pattern. This is silly, because it handles
2004             only 0-9, a-f, and A-F, but the character types table is locale-
2005             specific, which means strange things might happen. A private
2006             table is now used for this - though it costs 256 bytes, a table is
2007             much faster than multiple explicit tests. Of course, the standard
2008             character types table is still used for matching digits in subject
2009             strings against \d.
2010    
2011       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
2012             ending in _t are). So I've renamed it as ESC_tee.
2013    
2014    3. The first argument for regexec() in the POSIX wrapper should have been
2015       defined as "const".
2016    
2017    4. Changed pcretest to use malloc() for its buffers so that they can be
2018       Electric Fenced for debugging.
2019    
2020    5. There were several places in the code where, in UTF-8 mode, PCRE would try
2021       to read one or more bytes before the start of the subject string. Often this
2022       had no effect on PCRE's behaviour, but in some circumstances it could
2023       provoke a segmentation fault.
2024    
2025    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
2026       to try to read one or more bytes before the start of the subject string.
2027    
2028    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
2029       UTF-8 support could misbehave in various ways if the subject string
2030       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
2031       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
2032       back over UTF-8 characters.)
2033    
2034    
2035    Version 4.2 14-Apr-03
2036    ---------------------
2037    
2038    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
2039    
2040    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
2041         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
2042         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
2043         [WIN_PREFIX]: new variable, "cyg" for Cygwin
2044         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
2045           and BUILD_EXEEXT
2046         Note: automatic setting of the BUILD variables is not yet working
2047         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
2048           compile-time but not at link-time
2049         [LINK]: use for linking executables only
2050         make different versions for Windows and non-Windows
2051         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
2052           libraries
2053         [LINK_FOR_BUILD]: new variable
2054         [OBJEXT]: use throughout
2055         [EXEEXT]: use throughout
2056         <winshared>: new target
2057         <wininstall>: new target
2058         <dftables.o>: use native compiler
2059         <dftables>: use native linker
2060         <install>: handle Windows platform correctly
2061         <clean>: ditto
2062         <check>: ditto
2063         copy DLL to top builddir before testing
2064    
2065       As part of these changes, -no-undefined was removed again. This was reported
2066       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
2067       in any case.
2068    
2069    3. Some tidies to get rid of compiler warnings:
2070    
2071       . In the match_data structure, match_limit was an unsigned long int, whereas
2072         match_call_count was an int. I've made them both unsigned long ints.
2073    
2074       . In pcretest the fact that a const uschar * doesn't automatically cast to
2075         a void * provoked a warning.
2076    
2077       . Turning on some more compiler warnings threw up some "shadow" variables
2078         and a few more missing casts.
2079    
2080    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
2081       option, a class that contained a single character with a value between 128
2082       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
2083    
2084    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
2085       option, a class that contained several characters, but with at least one
2086       whose value was between 128 and 255 caused PCRE to crash.
2087    
2088    
2089    Version 4.1 12-Mar-03
2090    ---------------------
2091    
2092    1. Compiling with gcc -pedantic found a couple of places where casts were
2093    needed, and a string in dftables.c that was longer than standard compilers are
2094    required to support.
2095    
2096    2. Compiling with Sun's compiler found a few more places where the code could
2097    be tidied up in order to avoid warnings.
2098    
2099    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
2100    first of these names is deprecated in the latest Autoconf in favour of the name
2101    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
2102    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
2103    analogy I have changed it to CFLAGS_FOR_BUILD.
2104    
2105    4. Added -no-undefined to the linking command in the Makefile, because this is
2106    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
2107    linking step for the pcreposix library.
2108    
2109    5. PCRE was failing to diagnose the case of two named groups with the same
2110    name.
2111    
2112    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
2113    literal character that is needed in the subject for a match, and scans along to
2114    ensure that it is present before embarking on the full matching process. This
2115    saves time in cases of nested unlimited repeats that are never going to match.
2116    Problem: the scan can take a lot of time if the subject is very long (e.g.
2117    megabytes), thus penalizing straightforward matches. It is now done only if the
2118    amount of subject to be scanned is less than 1000 bytes.
2119    
2120    7. A lesser problem with the same optimization is that it was recording the
2121    first character of an anchored pattern as "needed", thus provoking a search
2122    right along the subject, even when the first match of the pattern was going to
2123    fail. The "needed" character is now not set for anchored patterns, unless it
2124    follows something in the pattern that is of non-fixed length. Thus, it still
2125    fulfils its original purpose of finding quick non-matches in cases of nested
2126    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
2127    
2128    
2129    Version 4.0 17-Feb-03
2130    ---------------------
2131    
2132    1. If a comment in an extended regex that started immediately after a meta-item
2133    extended to the end of string, PCRE compiled incorrect data. This could lead to
2134    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
2135    
2136    2. Moved to autoconf 2.53 and libtool 1.4.2.
2137    
2138    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
2139    the special perltest8 script is no longer needed - all the tests can be run
2140    from a single perltest script.
2141    
2142    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
2143    by \s. It has now been removed in PCRE. This means it isn't recognized as
2144    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
2145    class [:space:] *does* include VT, thereby creating a mess.
2146    
2147    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
2148    space and tab.
2149    
2150    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
2151    its new features into the main test script, reducing the number of scripts.
2152    
2153    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
2154    were backward compatible, and made the (?i) apply to the whole pattern, as if
2155    /i were given. Now it behaves more logically, and applies the option setting
2156    only to what follows. PCRE has been changed to follow suit. However, if it
2157    finds options settings right at the start of the pattern, it extracts them into
2158    the global options, as before. Thus, they show up in the info data.
2159    
2160    8. Added support for the \Q...\E escape sequence. Characters in between are
2161    treated as literals. This is slightly different from Perl in that $ and @ are
2162    also handled as literals inside the quotes. In Perl, they will cause variable
2163    interpolation. Note the following examples:
2164    
2165        Pattern            PCRE matches      Perl matches
2166    
2167        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
2168        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
2169        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
2170    
2171    For compatibility with Perl, \Q...\E sequences are recognized inside character
2172    classes as well as outside them.
2173    
2174    9. Re-organized 3 code statements in pcretest to avoid "overflow in
2175    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
2176    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
2177    signed/unsigned warnings.
2178    
2179    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
2180    option for pcretest, so I've replaced it by a simple function that does just
2181    that job.
2182    
2183    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
2184    "pcregrep -".
2185    
2186    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
2187    Java package. This provides some syntactic sugar for simple cases of what my
2188    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
2189    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
2190    item, you can use this simplified notation. Note that only makes sense with
2191    greedy quantifiers. Consequently, the use of the possessive quantifier forces
2192    greediness, whatever the setting of the PCRE_UNGREEDY option.
2193    
2194    13. A change of greediness default within a pattern was not taking effect at
2195    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
2196    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
2197    was abstracted outside.
2198    
2199    14. PCRE now supports the \G assertion. It is true when the current matching
2200    position is at the start point of the match. This differs from \A when the
2201    starting offset is non-zero. Used with the /g option of pcretest (or similar
2202    code), it works in the same way as it does for Perl's /g option. If all
2203    alternatives of a regex begin with \G, the expression is anchored to the start
2204    match position, and the "anchored" flag is set in the compiled expression.
2205    
2206    15. Some bugs concerning the handling of certain option changes within patterns
2207    have been fixed. These applied to options other than (?ims). For example,
2208    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
2209    been the other way round. Some of this was related to change 7 above.
2210    
2211    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
2212    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
2213    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
2214    POSIX classes only within a class (e.g. /[[:alpha:]]/).
2215    
2216    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
2217    mode. Unlike ".", it always matches newline, whatever the setting of
2218    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
2219    assertions. Perl allows it, but it doesn't (in general) work because it can't
2220    calculate the length of the lookbehind. At least, that's the case for Perl
2221    5.8.0 - I've been told they are going to document that it doesn't work in
2222    future.
2223    
2224    18. Added an error diagnosis for escapes that PCRE does not support: these are
2225    \L, \l, \N, \P, \p, \U, \u, and \X.
2226    
2227    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
2228    reading past the end of the pattern in cases such as /[abcd/.
2229    
2230    20. PCRE was getting more memory than necessary for patterns with classes that
2231    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
2232    
2233    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
2234    compiling PCRE for use with Virtual Pascal.
2235    
2236    22. Small fix to the Makefile to make it work properly if the build is done
2237    outside the source tree.
2238    
2239    23. Added a new extension: a condition to go with recursion. If a conditional
2240    subpattern starts with (?(R) the "true" branch is used if recursion has
2241    happened, whereas the "false" branch is used only at the top level.
2242    
2243    24. When there was a very long string of literal characters (over 255 bytes
2244    without UTF support, over 250 bytes with UTF support), the computation of how
2245    much memory was required could be incorrect, leading to segfaults or other
2246    strange effects.
2247    
2248    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
2249    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
2250    there was a subsequent back reference to those brackets. This meant that, for
2251    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
2252    possible to check for precisely this case. All we can do is abandon the
2253    optimization if .* occurs inside capturing brackets when there are any back
2254    references whatsoever. (See below for a better fix that came later.)
2255    
2256    26. The handling of the optimization for finding the first character of a
2257    non-anchored pattern, and for finding a character that is required later in the
2258    match were failing in some cases. This didn't break the matching; it just
2259    failed to optimize when it could. The way this is done has been re-implemented.
2260    
2261    27. Fixed typo in error message for invalid (?R item (it said "(?p").
2262    
2263    28. Added a new feature that provides some of the functionality that Perl
2264    provides with (?{...}). The facility is termed a "callout". The way it is done
2265    in PCRE is for the caller to provide an optional function, by setting
2266    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
2267    global variable. By default it is unset, which disables all calling out. To get
2268    the function called, the regex must include (?C) at appropriate points. This
2269    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
2270    This provides a means of identifying different callout points. When PCRE
2271    reaches such a point in the regex, if pcre_callout has been set, the external
2272    function is called. It is provided with data in a structure called
2273    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
2274    matching continues; if it returns a non-zero value, the match at the current
2275    point fails. However, backtracking will occur if possible. [This was changed
2276    later and other features added - see item 49 below.]
2277    
2278    29. pcretest is upgraded to test the callout functionality. It provides a
2279    callout function that displays information. By default, it shows the start of
2280    the match and the current position in the text. There are some new data escapes
2281    to vary what happens:
2282    
2283        \C+         in addition, show current contents of captured substrings
2284        \C-         do not supply a callout function
2285        \C!n        return 1 when callout number n is reached
2286        \C!n!m      return 1 when callout number n is reached for the mth time
2287    
2288    30. If pcregrep was called with the -l option and just a single file name, it
2289    output "<stdin>" if a match was found, instead of the file name.
2290    
2291    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
2292    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
2293    pcre_exec(). This saves a malloc/free per call. The default value of
2294    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
2295    when configuring.
2296    
2297    32. The default maximum size of a compiled pattern is 64K. There have been a
2298    few cases of people hitting this limit. The code now uses macros to handle the
2299    storing of links as offsets within the compiled pattern. It defaults to 2-byte
2300    links, but this can be changed to 3 or 4 bytes by --with-link-size when
2301    configuring. Tests 2 and 5 work only with 2-byte links because they output
2302    debugging information about compiled patterns.
2303    
2304    33. Internal code re-arrangements:
2305    
2306    (a) Moved the debugging function for printing out a compiled regex into
2307        its own source file (printint.c) and used #include to pull it into
2308        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
2309        separate copies.
2310    
2311    (b) Defined the list of op-code names for debugging as a macro in
2312        internal.h so that it is next to the definition of the opcodes.
2313    
2314    (c) Defined a table of op-code lengths for simpler skipping along compiled
2315        code. This is again a macro in internal.h so that it is next to the
2316        definition of the opcodes.
2317    
2318    34. Added support for recursive calls to individual subpatterns, along the
2319    lines of Robin Houston's patch (but implemented somewhat differently).
2320    
2321    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
2322    allow it to read and process whole directories in Win32. This code was
2323    contributed by Lionel Fourquaux; it has not been tested by me.
2324    
2325    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
2326    used to name a group. Names consist of alphanumerics and underscores, and must
2327    be unique. Back references use the syntax (?P=name) and recursive calls use
2328    (?P>name) which is a PCRE extension to the Python extension. Groups still have
2329    numbers. The function pcre_fullinfo() can be used after compilation to extract
2330    a name/number map. There are three relevant calls:
2331    
2332      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
2333      PCRE_INFO_NAMECOUNT            yields the number of entries
2334      PCRE_INFO_NAMETABLE            yields a pointer to the map.
2335    
2336    The map is a vector of fixed-size entries. The size of each entry depends on
2337    the length of the longest name used. The first two bytes of each entry are the
2338    group number, most significant byte first. There follows the corresponding
2339    name, zero terminated. The names are in alphabetical order.
2340    
2341    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
2342    case instead of 255. Making it the same both with and without UTF-8 support
2343    means that the same test output works with both.
2344    
2345    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
2346    calling malloc() with a zero argument.
2347    
2348    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
2349    optimization. I've improved things by keeping a bitmap of backreferences with
2350    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
2351    fact referenced, the optimization can be applied. It is unlikely that a
2352    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
2353    the match to follow \n) will appear inside brackets with a number greater than
2354    31, but if it does, any back reference > 31 suppresses the optimization.
2355    
2356    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
2357    of disabling numbered capturing parentheses. Any opening parenthesis that is
2358    not followed by ? behaves as if it were followed by ?: but named parentheses
2359    can still be used for capturing (and they will acquire numbers in the usual
2360    way).
2361    
2362    41. Redesigned the return codes from the match() function into yes/no/error so
2363    that errors can be passed back from deep inside the nested calls. A malloc
2364    failure while inside a recursive subpattern call now causes the
2365    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
2366    
2367    42. It is now possible to set a limit on the number of times the match()
2368    function is called in a call to pcre_exec(). This facility makes it possible to
2369    limit the amount of recursion and backtracking, though not in a directly
2370    obvious way, because the match() function is used in a number of different
2371    circumstances. The count starts from zero for each position in the subject
2372    string (for non-anchored patterns). The default limit is, for compatibility, a
2373    large number, namely 10 000 000. You can change this in two ways:
2374    
2375    (a) When configuring PCRE before making, you can use --with-match-limit=n
2376        to set a default value for the compiled library.
2377    
2378    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
2379        a different value is set. See 45 below.
2380    
2381    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
2382    
2383    43. Added a new function pcre_config(int, void *) to enable run-time extraction
2384    of things that can be changed at compile time. The first argument specifies
2385    what is wanted and the second points to where the information is to be placed.
2386    The current list of available information is:
2387    
2388      PCRE_CONFIG_UTF8
2389    
2390    The output is an integer that is set to one if UTF-8 support is available;
2391    otherwise it is set to zero.
2392    
2393      PCRE_CONFIG_NEWLINE
2394    
2395    The output is an integer that it set to the value of the code that is used for
2396    newline. It is either LF (10) or CR (13).
2397    
2398      PCRE_CONFIG_LINK_SIZE
2399    
2400    The output is an integer that contains the number of bytes used for internal
2401    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
2402    
2403      PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
2404    
2405    The output is an integer that contains the threshold above which the POSIX
2406    interface uses malloc() for output vectors. See item 31 above.
2407    
2408      PCRE_CONFIG_MATCH_LIMIT
2409    
2410    The output is an unsigned integer that contains the default limit of the number
2411    of match() calls in a pcre_exec() execution. See 42 above.
2412    
2413    44. pcretest has been upgraded by the addition of the -C option. This causes it
2414    to extract all the available output from the new pcre_config() function, and to
2415    output it. The program then exits immediately.
2416    
2417    45. A need has arisen to pass over additional data with calls to pcre_exec() in
2418    order to support additional features. One way would have been to define
2419    pcre_exec2() (for example) with extra arguments, but this would not have been
2420    extensible, and would also have required all calls to the original function to
2421    be mapped to the new one. Instead, I have chosen to extend the mechanism that
2422    is used for passing in "extra" data from pcre_study().
2423    
2424    The pcre_extra structure is now exposed and defined in pcre.h. It currently
2425    contains the following fields:
2426    
2427      flags         a bitmap indicating which of the following fields are set
2428      study_data    opaque data from pcre_study()
2429      match_limit   a way of specifying a limit on match() calls for a specific
2430                      call to pcre_exec()
2431      callout_data  data for callouts (see 49 below)
2432    
2433    The flag bits are also defined in pcre.h, and are
2434    
2435      PCRE_EXTRA_STUDY_DATA
2436      PCRE_EXTRA_MATCH_LIMIT
2437      PCRE_EXTRA_CALLOUT_DATA
2438    
2439    The pcre_study() function now returns one of these new pcre_extra blocks, with
2440    the actual study data pointed to by the study_data field, and the
2441    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
2442    before. That is, this change is entirely upwards-compatible and requires no
2443    change to existing code.
2444    
2445    If you want to pass in additional data to pcre_exec(), you can either place it
2446    in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
2447    block.
2448    
2449    46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
2450    data string contains the escape sequence \M, pcretest calls pcre_exec() several
2451    times with different match limits, until it finds the minimum value needed for
2452    pcre_exec() to complete. The value is then output. This can be instructive; for
2453    most simple matches the number is quite small, but for pathological cases it
2454    gets very large very quickly.
2455    
2456    47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
2457    returns the size of the data block pointed to by the study_data field in a
2458    pcre_extra block, that is, the value that was passed as the argument to
2459    pcre_malloc() when PCRE was getting memory in which to place the information
2460    created by pcre_study(). The fourth argument should point to a size_t variable.
2461    pcretest has been extended so that this information is shown after a successful
2462    pcre_study() call when information about the compiled regex is being displayed.
2463    
2464    48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
2465    because what follows is always an absolute path. (Later: it turns out that this
2466    is more than cosmetic for MinGW, because it doesn't like empty path
2467    components.)
2468    
2469    49. Some changes have been made to the callout feature (see 28 above):
2470    
2471    (i)  A callout function now has three choices for what it returns:
2472    
2473           0  =>  success, carry on matching
2474         > 0  =>  failure at this point, but backtrack if possible
2475         < 0  =>  serious error, return this value from pcre_exec()
2476    
2477         Negative values should normally be chosen from the set of PCRE_ERROR_xxx
2478         values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
2479         "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
2480         use by callout functions. It will never be used by PCRE itself.
2481    
2482    (ii) The pcre_extra structure (see 45 above) has a void * field called
2483         callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
2484         pcre_callout_block structure has a field of the same name. The contents of
2485         the field passed in the pcre_extra structure are passed to the callout
2486         function in the corresponding field in the callout block. This makes it
2487         easier to use the same callout-containing regex from multiple threads. For
2488         testing, the pcretest program has a new data escape
2489    
2490           \C*n        pass the number n (may be negative) as callout_data
2491    
2492         If the callout function in pcretest receives a non-zero value as
2493         callout_data, it returns that value.
2494    
2495    50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
2496    there were some redundant $(CFLAGS) in commands that are now specified as
2497    $(LINK), which already includes $(CFLAGS).
2498    
2499    51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
2500    has been compiled with UTF-8 support *and* pcre_compile() has been compiled
2501    with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
2502    one-byte characters throughout. Note that case-insensitive matching applies
2503    only to characters whose values are less than 256. PCRE doesn't support the
2504    notion of cases for higher-valued characters.
2505    
2506    (i)   A character class whose characters are all within 0-255 is handled as
2507          a bit map, and the map is inverted for negative classes. Previously, a
2508          character > 255 always failed to match such a class; however it should
2509          match if the class was a negative one (e.g. [^ab]). This has been fixed.
2510    
2511    (ii)  A negated character class with a single character < 255 is coded as
2512          "not this character" (OP_NOT). This wasn't working properly when the test
2513          character was multibyte, either singly or repeated.
2514    
2515    (iii) Repeats of multibyte characters are now handled correctly in UTF-8
2516          mode, for example: \x{100}{2,3}.
2517    
2518    (iv)  The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
2519          singly or repeated) now correctly test multibyte characters. However,
2520          PCRE doesn't recognize any characters with values greater than 255 as
2521          digits, spaces, or word characters. Such characters always match \D, \S,
2522          and \W, and never match \d, \s, or \w.
2523    
2524    (v)   Classes may now contain characters and character ranges with values
2525          greater than 255. For example: [ab\x{100}-\x{400}].
2526    
2527    (vi)  pcregrep now has a --utf-8 option (synonym -u) which makes it call
2528          PCRE in UTF-8 mode.
2529    
2530    52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
2531    PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
2532    retained for backwards compatibility. (Note that LASTLITERAL is also a byte
2533    value.)
2534    
2535    53. The single man page has become too large. I have therefore split it up into
2536    a number of separate man pages. These also give rise to individual HTML pages;
2537    these are now put in a separate directory, and there is an index.html page that
2538    lists them all. Some hyperlinking between the pages has been installed.
2539    
2540    54. Added convenience functions for handling named capturing parentheses.
2541    
2542    55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2543    aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2544    true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2545    are faulted.
2546    
2547    56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2548    calling configure. These values are used when compiling the dftables.c program
2549    which is run to generate the source of the default character tables. They
2550    default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2551    you will need to set these values.
2552    
2553    57. Updated the building process for Windows DLL, as provided by Fred Cox.
2554    
2555    
2556    Version 3.9 02-Jan-02
2557    ---------------------
2558    
2559    1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2560    
2561    2. If --disable-static was given, the building process failed when trying to
2562    build pcretest and pcregrep. (For some reason it was using libtool to compile
2563    them, which is not right, as they aren't part of the library.)
2564    
2565    
2566    Version 3.8 18-Dec-01
2567    ---------------------
2568    
2569    1. The experimental UTF-8 code was completely screwed up. It was packing the
2570    bytes in the wrong order. How dumb can you get?
2571    
2572    
2573    Version 3.7 29-Oct-01
2574    ---------------------
2575    
2576    1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2577    This caused pcretest, when used on the test data, to segfault. Unfortunately,
2578    this didn't happen under Solaris 8, where I normally test things.
2579    
2580    2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2581    doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2582    isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2583    this fix an hour or so after the initial 3.7 release.)
2584    
2585    
2586    Version 3.6 23-Oct-01
2587    ---------------------
2588    
2589    1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2590    offsets passed as NULL with zero offset count.
2591    
2592    2. The config.guess and config.sub files had not been updated when I moved to
2593    the latest autoconf.
2594    
2595    
2596    Version 3.5 15-Aug-01
2597    ---------------------
2598    
2599    1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2600    had been forgotten.
2601    
2602    2. By using declared but undefined structures, we can avoid using "void"
2603    definitions in pcre.h while keeping the internal definitions of the structures
2604    private.
2605    
2606    3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2607    user point of view, this means that both static and shared libraries are built
2608    by default, but this can be individually controlled. More of the work of
2609    handling this static/shared cases is now inside libtool instead of PCRE's make
2610    file.
2611    
2612    4. The pcretest utility is now installed along with pcregrep because it is
2613    useful for users (to test regexs) and by doing this, it automatically gets
2614    relinked by libtool. The documentation has been turned into a man page, so
2615    there are now .1, .txt, and .html versions in /doc.
2616    
2617    5. Upgrades to pcregrep:
2618       (i)   Added long-form option names like gnu grep.
2619       (ii)  Added --help to list all options with an explanatory phrase.
2620       (iii) Added -r, --recursive to recurse into sub-directories.
2621       (iv)  Added -f, --file to read patterns from a file.
2622    
2623    6. pcre_exec() was referring to its "code" argument before testing that
2624    argument for NULL (and giving an error if it was NULL).
2625    
2626    7. Upgraded Makefile.in to allow for compiling in a different directory from
2627    the source directory.
2628    
2629    8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2630    options bits, the pointer it was passed was to an int instead of to an unsigned
2631    long int. This mattered only on 64-bit systems.
2632    
2633    9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2634    generated) instead of pcre.in, which it its source. Also made the same change
2635    in several of the .c files.
2636    
2637    10. A new release of gcc defines printf() as a macro, which broke pcretest
2638    because it had an ifdef in the middle of a string argument for printf(). Fixed
2639    by using separate calls to printf().
2640    
2641    11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2642    script, to force use of CR or LF instead of \n in the source. On non-Unix
2643    systems, the value can be set in config.h.
2644    
2645    12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2646    absolute limit. Changed the text of the error message to make this clear, and
2647    likewise updated the man page.
2648    
2649    13. The limit of 99 on the number of capturing subpatterns has been removed.
2650    The new limit is 65535, which I hope will not be a "real" limit.
2651    
2652    
2653    Version 3.4 22-Aug-00
2654    ---------------------
2655    
2656    1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2657    
2658    2. Diagnose condition (?(0) as an error instead of crashing on matching.
2659    
2660    
2661    Version 3.3 01-Aug-00
2662    ---------------------
2663    
2664    1. If an octal character was given, but the value was greater than \377, it
2665    was not getting masked to the least significant bits, as documented. This could
2666    lead to crashes in some systems.
2667    
2668    2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
2669    the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
2670    
2671    3. Added the functions pcre_free_substring() and pcre_free_substring_list().
2672    These just pass their arguments on to (pcre_free)(), but they are provided
2673    because some uses of PCRE bind it to non-C systems that can call its functions,
2674    but cannot call free() or pcre_free() directly.
2675    
2676    4. Add "make test" as a synonym for "make check". Corrected some comments in
2677    the Makefile.
2678    
2679    5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
2680    Makefile.
2681    
2682    6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
2683    command called pgrep for grepping around the active processes.
2684    
2685    7. Added the beginnings of support for UTF-8 character strings.
2686    
2687    8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
2688    RANLIB to ./ltconfig so that they are used by libtool. I think these are all
2689    the relevant ones. (AR is not passed because ./ltconfig does its own figuring
2690    out for the ar command.)
2691    
2692    
2693    Version 3.2 12-May-00
2694    ---------------------
2695    
2696    This is purely a bug fixing release.
2697    
2698    1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
2699    of ZA. This was just one example of several cases that could provoke this bug,
2700    which was introduced by change 9 of version 2.00. The code for breaking
2701    infinite loops after an iteration that matches an empty string was't working
2702    correctly.
2703    
2704    2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
2705    when matched against abbab (for example). After matching an empty string, it
2706    wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
2707    caused it to match further down the string than it should.
2708    
2709    3. The code contained an inclusion of sys/types.h. It isn't clear why this
2710    was there because it doesn't seem to be needed, and it causes trouble on some
2711    systems, as it is not a Standard C header. It has been removed.
2712    
2713    4. Made 4 silly changes to the source to avoid stupid compiler warnings that
2714    were reported on the Macintosh. The changes were from
2715    
2716      while ((c = *(++ptr)) != 0 && c != '\n');
2717    to
2718      while ((c = *(++ptr)) != 0 && c != '\n') ;
2719    
2720    Totally extraordinary, but if that's what it takes...
2721    
2722    5. PCRE is being used in one environment where neither memmove() nor bcopy() is
2723    available. Added HAVE_BCOPY and an autoconf test for it; if neither
2724    HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
2725    assumes the way PCRE uses memmove() (always moving upwards).
2726    
2727    6. PCRE is being used in one environment where strchr() is not available. There
2728    was only one use in pcre.c, and writing it out to avoid strchr() probably gives
2729    faster code anyway.
2730    
2731    
2732    Version 3.1 09-Feb-00
2733    ---------------------
2734    
2735    The only change in this release is the fixing of some bugs in Makefile.in for
2736    the "install" target:
2737    
2738    (1) It was failing to install pcreposix.h.
2739    
2740    (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
2741    
2742    
2743    Version 3.0 01-Feb-00
2744    ---------------------
2745    
2746    1. Add support for the /+ modifier to perltest (to output $` like it does in
2747    pcretest).
2748    
2749    2. Add support for the /g modifier to perltest.
2750    
2751    3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
2752    matches null strings.
2753    
2754    4. Fix perltest so that it doesn't do unwanted things when fed an empty
2755    pattern. Perl treats empty patterns specially - it reuses the most recent
2756    pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
2757    effect.
2758    
2759    5. The POSIX interface was broken in that it was just handing over the POSIX
2760    captured string vector to pcre_exec(), but (since release 2.00) PCRE has
2761    required a bigger vector, with some working space on the end. This means that
2762    the POSIX wrapper now has to get and free some memory, and copy the results.
2763    
2764    6. Added some simple autoconf support, placing the test data and the
2765    documentation in separate directories, re-organizing some of the
2766    information files, and making it build pcre-config (a GNU standard). Also added
2767    libtool support for building PCRE as a shared library, which is now the
2768    default.
2769    
2770    7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
2771    09 are not valid octal constants. Single digits will be used for minor values
2772    less than 10.
2773    
2774    8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
2775    existing programs that set these in the POSIX interface can use PCRE without
2776    modification.
2777    
2778    9. Added a new function, pcre_fullinfo() with an extensible interface. It can
2779    return all that pcre_info() returns, plus additional data. The pcre_info()
2780    function is retained for compatibility, but is considered to be obsolete.
2781    
2782    10. Added experimental recursion feature (?R) to handle one common case that
2783    Perl 5.6 will be able to do with (?p{...}).
2784    
2785    11. Added support for POSIX character classes like [:alpha:], which Perl is
2786    adopting.
2787    
2788    
2789    Version 2.08 31-Aug-99
2790    ----------------------
2791    
2792    1. When startoffset was not zero and the pattern began with ".*", PCRE was not
2793    trying to match at the startoffset position, but instead was moving forward to
2794    the next newline as if a previous match had failed.
2795    
2796    2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
2797    and could get into a loop if a null string was matched other than at the start
2798    of the subject.
2799    
2800    3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
2801    be distinguished at compile time, and for completeness also added PCRE_DATE.
2802    
2803    5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
2804    in GnuWin32 environments.
2805    
2806    
2807    Version 2.07 29-Jul-99
2808    ----------------------
2809    
2810    1. The documentation is now supplied in plain text form and HTML as well as in
2811    the form of man page sources.
2812    
2813    2. C++ compilers don't like assigning (void *) values to other pointer types.
2814    In particular this affects malloc(). Although there is no problem in Standard
2815    C, I've put in casts to keep C++ compilers happy.
2816    
2817    3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
2818    should be (const char *).
2819    
2820    4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
2821    be useful for non-Unix systems who don't want to bother with the POSIX stuff.
2822    However, I haven't made this a standard facility. The documentation doesn't
2823    mention it, and the Makefile doesn't support it.
2824    
2825    5. The Makefile now contains an "install" target, with editable destinations at
2826    the top of the file. The pcretest program is not installed.
2827    
2828    6. pgrep -V now gives the PCRE version number and date.
2829    
2830    7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
2831    causing the entire string to be ignored, instead of just the last character.
2832    
2833    8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
2834    non-matching string, it can take a very, very long time, even for strings of
2835    quite modest length, because of the nested recursion. PCRE now does better in
2836    some of these cases. It does this by remembering the last required literal
2837    character in the pattern, and pre-searching the subject to ensure it is present
2838    before running the real match. In other words, it applies a heuristic to detect
2839    some types of certain failure quickly, and in the above example, if presented
2840    with a string that has no trailing " it gives "no match" very quickly.
2841    
2842    9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
2843    other alternatives are tried instead.
2844    
2845    
2846    Version 2.06 09-Jun-99
2847    ----------------------
2848    
2849    1. Change pcretest's output for amount of store used to show just the code
2850    space, because the remainder (the data block) varies in size between 32-bit and
2851    64-bit systems.
2852    
2853    2. Added an extra argument to pcre_exec() to supply an offset in the subject to
2854    start matching at. This allows lookbehinds to work when searching for multiple
2855    occurrences in a string.
2856    
2857    3. Added additional options to pcretest for testing multiple occurrences:
2858    
2859       /+   outputs the rest of the string that follows a match
2860       /g   loops for multiple occurrences, using the new startoffset argument
2861       /G   loops for multiple occurrences by passing an incremented pointer
2862    
2863    4. PCRE wasn't doing the "first character" optimization for patterns starting
2864    with \b or \B, though it was doing it for other lookbehind assertions. That is,
2865    it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
2866    the letter 'x'. On long subject strings, this gives a significant speed-up.
2867    
2868    
2869    Version 2.05 21-Apr-99
2870    ----------------------
2871    
2872    1. Changed the type of magic_number from int to long int so that it works
2873    properly on 16-bit systems.
2874    
2875    2. Fixed a bug which caused patterns starting with .* not to work correctly
2876    when the subject string contained newline characters. PCRE was assuming
2877    anchoring for such patterns in all cases, which is not correct because .* will
2878    not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
2879    DOTALL is set at top level; otherwise it knows that patterns starting with .*
2880    must be retried after every newline in the subject.
2881    
2882    
2883    Version 2.04 18-Feb-99
2884    ----------------------
2885    
2886    1. For parenthesized subpatterns with repeats whose minimum was zero, the
2887    computation of the store needed to hold the pattern was incorrect (too large).
2888    If such patterns were nested a few deep, this could multiply and become a real
2889    problem.
2890    
2891    2. Added /M option to pcretest to show the memory requirement of a specific
2892    pattern. Made -m a synonym of -s (which does this globally) for compatibility.
2893    
2894    3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
2895    compiled in such a way that the backtracking after subsequent failure was
2896    pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
2897    ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
2898    
2899    
2900    Version 2.03 02-Feb-99
2901    ----------------------
2902    
2903    1. Fixed typo and small mistake in man page.
2904    
2905    2. Added 4th condition (GPL supersedes if conflict) and created separate
2906    LICENCE file containing the conditions.
2907    
2908    3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
2909    Perl, that is the internal \ allows the delimiter to be included in the
2910    pattern. Locked out the use of \ as a delimiter. If \ immediately follows
2911    the final delimiter, add \ to the end of the pattern (to test the error).
2912    
2913    4. Added the convenience functions for extracting substrings after a successful
2914    match. Updated pcretest to make it able to test these functions.
2915    
2916    
2917    Version 2.02 14-Jan-99
2918    ----------------------
2919    
2920    1. Initialized the working variables associated with each extraction so that
2921    their saving and restoring doesn't refer to uninitialized store.
2922    
2923    2. Put dummy code into study.c in order to trick the optimizer of the IBM C
2924    compiler for OS/2 into generating correct code. Apparently IBM isn't going to
2925    fix the problem.
2926    
2927    3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
2928    calls, and wasn't printing the correct value for compiling calls. Increased the
2929    default value of LOOPREPEAT, and the number of significant figures in the
2930    times.
2931    
2932    4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
2933    
2934    5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
2935    a building problem on Windows NT with a FAT file system.
2936    
2937    
2938    Version 2.01 21-Oct-98
2939    ----------------------
2940    
2941    1. Changed the API for pcre_compile() to allow for the provision of a pointer
2942    to character tables built by pcre_maketables() in the current locale. If NULL
2943    is passed, the default tables are used.
2944    
2945    
2946    Version 2.00 24-Sep-98
2947    ----------------------
2948    
2949    1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
2950    it any more.
2951    
2952    2. Allow quantification of (?>) groups, and make it work correctly.
2953    
2954    3. The first character computation wasn't working for (?>) groups.
2955    
2956    4. Correct the implementation of \Z (it is permitted to match on the \n at the
2957    end of the subject) and add 5.005's \z, which really does match only at the
2958    very end of the subject.
2959    
2960    5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
2961    
2962    6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
2963    DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
2964    localized options. All options to pcre_study() were also removed.
2965    
2966    7. Add other new features from 5.005:
2967    
2968       $(?<=           positive lookbehind
2969       $(?<!           negative lookbehind
2970       (?imsx-imsx)    added the unsetting capability
2971                       such a setting is global if at outer level; local otherwise
2972       (?imsx-imsx:)   non-capturing groups with option setting
2973       (?(cond)re|re)  conditional pattern matching
2974    
2975       A backreference to itself in a repeated group matches the previous
2976       captured string.
2977    
2978    8. General tidying up of studying (both automatic and via "study")
2979    consequential on the addition of new assertions.
2980    
2981    9. As in 5.005, unlimited repeated groups that could match an empty substring
2982    are no longer faulted at compile time. Instead, the loop is forcibly broken at
2983    runtime if any iteration does actually match an empty substring.
2984    
2985    10. Include the RunTest script in the distribution.
2986    
2987    11. Added tests from the Perl 5.005_02 distribution. This showed up a few
2988    discrepancies, some of which were old and were also with respect to 5.004. They
2989    have now been fixed.
2990    
2991    
2992    Version 1.09 28-Apr-98
2993    ----------------------
2994    
2995    1. A negated single character class followed by a quantifier with a minimum
2996    value of one (e.g.  [^x]{1,6}  ) was not compiled correctly. This could lead to
2997    program crashes, or just wrong answers. This did not apply to negated classes
2998    containing more than one character, or to minima other than one.
2999    
3000    
3001    Version 1.08 27-Mar-98
3002    ----------------------
3003    
3004    1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
3005    
3006    2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
3007    latter must appear before anything that relies on it in the pattern.
3008    
3009    
3010    Version 1.07 16-Feb-98
3011    ----------------------
3012    
3013    1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
3014    repeat of a potentially empty string).
3015    
3016    
3017  Version 1.06 23-Jan-98  Version 1.06 23-Jan-98
3018  ----------------------  ----------------------

Legend:
Removed from v.15  
changed lines
  Added in v.344

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12