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

Diff of /code/trunk/ChangeLog

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

revision 5 by nigel, Sat Feb 24 21:38:05 2007 UTC revision 343 by ph10, Mon Apr 21 16:30:37 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    
102    Version 7.6 28-Jan-08
103    ---------------------
104    
105    1.  A character class containing a very large number of characters with
106        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
107        overflow.
108    
109    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
110        HAVE_LONG_LONG is not defined.
111    
112    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
113        bring them up to date and include new features. This patch includes:
114    
115        - Fixed PH's badly added libz and libbz2 support.
116        - Fixed a problem with static linking.
117        - Added pcredemo. [But later removed - see 7 below.]
118        - Fixed dftables problem and added an option.
119        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
120            HAVE_LONG_LONG.
121        - Added readline support for pcretest.
122        - Added an listing of the option settings after cmake has run.
123    
124    4.  A user submitted a patch to Makefile that makes it easy to create
125        "pcre.dll" under mingw when using Configure/Make. I added stuff to
126        Makefile.am that cause it to include this special target, without
127        affecting anything else. Note that the same mingw target plus all
128        the other distribution libraries and programs are now supported
129        when configuring with CMake (see 6 below) instead of with
130        Configure/Make.
131    
132    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
133        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
134        exported in the Windows port". It has not yet been confirmed that the patch
135        solves the problem, but it does no harm.
136    
137    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
138        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
139        with CMake, and also correct the comment about stack recursion.
140    
141    7.  Remove the automatic building of pcredemo from the ./configure system and
142        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
143        of a program that users should build themselves after PCRE is installed, so
144        building it automatically is not really right. What is more, it gave
145        trouble in some build environments.
146    
147    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
148    
149    
150    Version 7.5 10-Jan-08
151    ---------------------
152    
153    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
154        values in parens when parsing an RE using the C++ wrapper."
155    
156    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
157        Characters greater than 255 were excluded from the class instead of being
158        included.
159    
160    3.  The same bug as (2) above applied to negated POSIX classes such as
161        [:^space:].
162    
163    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
164        defined or documented. It seems to have been a typo for PCRE_STATIC, so
165        I have changed it.
166    
167    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
168        first named subpattern) and a construct such as (?&a) would reference the
169        first named subpattern whose name started with "a" (in other words, the
170        length check was missing). Both these problems are fixed. "Subpattern name
171        expected" is now given for (?&) (a zero-length name), and this patch also
172        makes it give the same error for \k'' (previously it complained that that
173        was a reference to a non-existent subpattern).
174    
175    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
176        this is right because (?- can be followed by option settings as well as by
177        digits. I have, however, made the messages clearer.
178    
179    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
180        than the number used in the conditional) now cause a compile-time error.
181        This is actually not compatible with Perl, which accepts such patterns, but
182        treats the conditional as always being FALSE (as PCRE used to), but it
183        seems to me that giving a diagnostic is better.
184    
185    8.  Change "alphameric" to the more common word "alphanumeric" in comments
186        and messages.
187    
188    9.  Fix two occurrences of "backslash" in comments that should have been
189        "backspace".
190    
191    10. Remove two redundant lines of code that can never be obeyed (their function
192        was moved elsewhere).
193    
194    11. The program that makes PCRE's Unicode character property table had a bug
195        which caused it to generate incorrect table entries for sequences of
196        characters that have the same character type, but are in different scripts.
197        It amalgamated them into a single range, with the script of the first of
198        them. In other words, some characters were in the wrong script. There were
199        thirteen such cases, affecting characters in the following ranges:
200    
201          U+002b0 - U+002c1
202          U+0060c - U+0060d
203          U+0061e - U+00612
204          U+0064b - U+0065e
205          U+0074d - U+0076d
206          U+01800 - U+01805
207          U+01d00 - U+01d77
208          U+01d9b - U+01dbf
209          U+0200b - U+0200f
210          U+030fc - U+030fe
211          U+03260 - U+0327f
212          U+0fb46 - U+0fbb1
213          U+10450 - U+1049d
214    
215    12. The -o option (show only the matching part of a line) for pcregrep was not
216        compatible with GNU grep in that, if there was more than one match in a
217        line, it showed only the first of them. It now behaves in the same way as
218        GNU grep.
219    
220    13. If the -o and -v options were combined for pcregrep, it printed a blank
221        line for every non-matching line. GNU grep prints nothing, and pcregrep now
222        does the same. The return code can be used to tell if there were any
223        non-matching lines.
224    
225    14. Added --file-offsets and --line-offsets to pcregrep.
226    
227    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
228        infinitely looping recursion. The bug was that positive lookaheads were not
229        being skipped when checking for a possible empty match (negative lookaheads
230        and both kinds of lookbehind were skipped).
231    
232    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
233        inclusion of <windows.h> to before rather than after the definition of
234        INVALID_FILE_ATTRIBUTES (patch from David Byron).
235    
236    17. Specifying a possessive quantifier with a specific limit for a Unicode
237        character property caused pcre_compile() to compile bad code, which led at
238        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
239        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
240        caused the error; without that there was no problem.
241    
242    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
243    
244    19. Added --enable-pcretest-libreadline.
245    
246    20. In pcrecpp.cc, the variable 'count' was incremented twice in
247        RE::GlobalReplace(). As a result, the number of replacements returned was
248        double what it should be. I removed one of the increments, but Craig sent a
249        later patch that removed the other one (the right fix) and added unit tests
250        that check the return values (which was not done before).
251    
252    21. Several CMake things:
253    
254        (1) Arranged that, when cmake is used on Unix, the libraries end up with
255            the names libpcre and libpcreposix, not just pcre and pcreposix.
256    
257        (2) The above change means that pcretest and pcregrep are now correctly
258            linked with the newly-built libraries, not previously installed ones.
259    
260        (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
261    
262    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
263        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
264        UTF-8 newline character). The key issue is that the pattern starts .*;
265        this means that the match must be either at the beginning, or after a
266        newline. The bug was in the code for advancing after a failed match and
267        checking that the new position followed a newline. It was not taking
268        account of UTF-8 characters correctly.
269    
270    23. PCRE was behaving differently from Perl in the way it recognized POSIX
271        character classes. PCRE was not treating the sequence [:...:] as a
272        character class unless the ... were all letters. Perl, however, seems to
273        allow any characters between [: and :], though of course it rejects as
274        unknown any "names" that contain non-letters, because all the known class
275        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
276        for example, whereas PCRE did not - it did not recognize a POSIX character
277        class. This seemed a bit dangerous, so the code has been changed to be
278        closer to Perl. The behaviour is not identical to Perl, because PCRE will
279        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
280        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
281        Perl does, and where it didn't before.
282    
283    24. Rewrite so as to remove the single use of %n from pcregrep because in some
284        Windows environments %n is disabled by default.
285    
286    
287    Version 7.4 21-Sep-07
288    ---------------------
289    
290    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
291        means that a class such as [\s] counted as "explicit reference to CR or
292        LF". That isn't really right - the whole point of the change was to try to
293        help when there was an actual mention of one of the two characters. So now
294        the change happens only if \r or \n (or a literal CR or LF) character is
295        encountered.
296    
297    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
298        of both had grown to the point where there were only 3 bits left.
299        Fortunately, there was spare space in the data structure, and so I have
300        moved the internal flags into a new 16-bit field to free up more option
301        bits.
302    
303    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
304        but did not set the internal JCHANGED flag - either of these is enough to
305        control the way the "get" function works - but the PCRE_INFO_JCHANGED
306        facility is supposed to tell if (?J) was ever used, so now (?J) at the
307        start sets both bits.
308    
309    4.  Added options (at build time, compile time, exec time) to change \R from
310        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
311    
312    5.  doc/pcresyntax.html was missing from the distribution.
313    
314    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
315        compatibility, even though it is no longer used.
316    
317    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
318        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
319        windows.h file is present (where different names are used). [This was
320        reversed later after testing - see 16 below.]
321    
322    8.  Changed all #include <config.h> to #include "config.h". There were also
323        some further <pcre.h> cases that I changed to "pcre.h".
324    
325    9.  When pcregrep was used with the --colour option, it missed the line ending
326        sequence off the lines that it output.
327    
328    10. It was pointed out to me that arrays of string pointers cause lots of
329        relocations when a shared library is dynamically loaded. A technique of
330        using a single long string with a table of offsets can drastically reduce
331        these. I have refactored PCRE in four places to do this. The result is
332        dramatic:
333    
334          Originally:                          290
335          After changing UCP table:            187
336          After changing error message table:   43
337          After changing table of "verbs"       36
338          After changing table of Posix names   22
339    
340        Thanks to the folks working on Gregex for glib for this insight.
341    
342    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
343        unicode-properties was also set.
344    
345    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
346    
347    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
348        checked only for CRLF.
349    
350    14. Added casts to pcretest.c to avoid compiler warnings.
351    
352    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
353    
354    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
355        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
356        entirely. This removes changes made in 7 above.
357    
358    17. The CMake files have been updated, and there is now more information about
359        building with CMake in the NON-UNIX-USE document.
360    
361    
362    Version 7.3 28-Aug-07
363    ---------------------
364    
365     1. In the rejigging of the build system that eventually resulted in 7.1, the
366        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
367        brackets there is not right, since it causes compilers to look for an
368        installed pcre.h, not the version that is in the source that is being
369        compiled (which of course may be different). I have changed it back to:
370    
371          #include "pcre.h"
372    
373        I have a vague recollection that the change was concerned with compiling in
374        different directories, but in the new build system, that is taken care of
375        by the VPATH setting the Makefile.
376    
377     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
378        when the subject happened to end in the byte 0x85 (e.g. if the last
379        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
380        characters but of course it shouldn't be taken as a newline when it is part
381        of another character. The bug was that, for an unlimited repeat of . in
382        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
383        characters when looking for a newline.
384    
385     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
386    
387     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
388        in debug output.
389    
390     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
391        long printing in the pcrecpp unittest when running under MinGW.
392    
393     6. ESC_K was left out of the EBCDIC table.
394    
395     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
396        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
397        limit also applies to "virtual nesting" when a pattern is recursive, and in
398        this case 1000 isn't so big. I have been able to remove this limit at the
399        expense of backing off one optimization in certain circumstances. Normally,
400        when pcre_exec() would call its internal match() function recursively and
401        immediately return the result unconditionally, it uses a "tail recursion"
402        feature to save stack. However, when a subpattern that can match an empty
403        string has an unlimited repetition quantifier, it no longer makes this
404        optimization. That gives it a stack frame in which to save the data for
405        checking that an empty string has been matched. Previously this was taken
406        from the 1000-entry workspace that had been reserved. So now there is no
407        explicit limit, but more stack is used.
408    
409     8. Applied Daniel's patches to solve problems with the import/export magic
410        syntax that is required for Windows, and which was going wrong for the
411        pcreposix and pcrecpp parts of the library. These were overlooked when this
412        problem was solved for the main library.
413    
414     9. There were some crude static tests to avoid integer overflow when computing
415        the size of patterns that contain repeated groups with explicit upper
416        limits. As the maximum quantifier is 65535, the maximum group length was
417        set at 30,000 so that the product of these two numbers did not overflow a
418        32-bit integer. However, it turns out that people want to use groups that
419        are longer than 30,000 bytes (though not repeat them that many times).
420        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
421        made it possible to implement the integer overflow checks in a much more
422        dynamic way, which I have now done. The artificial limitation on group
423        length has been removed - we now have only the limit on the total length of
424        the compiled pattern, which depends on the LINK_SIZE setting.
425    
426    10. Fixed a bug in the documentation for get/copy named substring when
427        duplicate names are permitted. If none of the named substrings are set, the
428        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
429        empty string.
430    
431    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
432        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
433        because the ] is interpreted as the first data character and the
434        terminating ] is not found. PCRE has been made compatible with Perl in this
435        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
436        cause memory overwriting.
437    
438    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
439        string has been matched (to stop an infinite loop). It was not recognizing
440        a conditional subpattern that could match an empty string if that
441        subpattern was within another subpattern. For example, it looped when
442        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
443        condition was not nested. This bug has been fixed.
444    
445    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
446        past the start of the subject in the presence of bytes with the top bit
447        set, for example "\x8aBCD".
448    
449    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
450        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
451    
452    14. Optimized (?!) to (*FAIL).
453    
454    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
455        This restricts code points to be within the range 0 to 0x10FFFF, excluding
456        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
457        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
458        does: it's just the validity check that is more restrictive.
459    
460    16. Inserted checks for integer overflows during escape sequence (backslash)
461        processing, and also fixed erroneous offset values for syntax errors during
462        backslash processing.
463    
464    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
465        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
466    
467    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
468        caused an overrun.
469    
470    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
471        something other than just ASCII characters) inside a group that had an
472        unlimited repeat caused a loop at compile time (while checking to see
473        whether the group could match an empty string).
474    
475    20. Debugging a pattern containing \p or \P could cause a crash. For example,
476        [\P{Any}] did so. (Error in the code for printing property names.)
477    
478    21. An orphan \E inside a character class could cause a crash.
479    
480    22. A repeated capturing bracket such as (A)? could cause a wild memory
481        reference during compilation.
482    
483    23. There are several functions in pcre_compile() that scan along a compiled
484        expression for various reasons (e.g. to see if it's fixed length for look
485        behind). There were bugs in these functions when a repeated \p or \P was
486        present in the pattern. These operators have additional parameters compared
487        with \d, etc, and these were not being taken into account when moving along
488        the compiled data. Specifically:
489    
490        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
491            length.
492    
493        (b) An item such as \pL+ within a repeated group could cause crashes or
494            loops.
495    
496        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
497            "reference to non-existent subpattern" error.
498    
499        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
500    
501    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
502        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
503    
504    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
505    
506    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
507        character were causing crashes (broken optimization).
508    
509    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
510        \p or \P) caused a compile-time loop.
511    
512    28. More problems have arisen in unanchored patterns when CRLF is a valid line
513        break. For example, the unstudied pattern [\r\n]A does not match the string
514        "\r\nA" because change 7.0/46 below moves the current point on by two
515        characters after failing to match at the start. However, the pattern \nA
516        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
517        the same is true. There doesn't seem any very clean way out of this, but
518        what I have chosen to do makes the common cases work: PCRE now takes note
519        of whether there can be an explicit match for \r or \n anywhere in the
520        pattern, and if so, 7.0/46 no longer applies. As part of this change,
521        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
522        pattern has explicit CR or LF references.
523    
524    29. Added (*CR) etc for changing newline setting at start of pattern.
525    
526    
527    Version 7.2 19-Jun-07
528    ---------------------
529    
530     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
531        which is apparently normally available under Windows.
532    
533     2. Re-jig the pcregrep tests with different newline settings in an attempt
534        to make them independent of the local environment's newline setting.
535    
536     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
537    
538     4. Some of the "internals" tests were previously cut out when the link size
539        was not 2, because the output contained actual offsets. The recent new
540        "Z" feature of pcretest means that these can be cut out, making the tests
541        usable with all link sizes.
542    
543     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
544        stack recursion. This gives a massive performance boost under BSD, but just
545        a small improvement under Linux. However, it saves one field in the frame
546        in all cases.
547    
548     6. Added more features from the forthcoming Perl 5.10:
549    
550        (a) (?-n) (where n is a string of digits) is a relative subroutine or
551            recursion call. It refers to the nth most recently opened parentheses.
552    
553        (b) (?+n) is also a relative subroutine call; it refers to the nth next
554            to be opened parentheses.
555    
556        (c) Conditions that refer to capturing parentheses can be specified
557            relatively, for example, (?(-2)... or (?(+3)...
558    
559        (d) \K resets the start of the current match so that everything before
560            is not part of it.
561    
562        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
563    
564        (f) \g{name} is another synonym - part of Perl 5.10's unification of
565            reference syntax.
566    
567        (g) (?| introduces a group in which the numbering of parentheses in each
568            alternative starts with the same number.
569    
570        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
571    
572     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
573        PCRE_INFO_JCHANGED.
574    
575     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
576        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
577        for detecting groups that can match an empty string.
578    
579     9. A pattern with a very large number of alternatives (more than several
580        hundred) was running out of internal workspace during the pre-compile
581        phase, where pcre_compile() figures out how much memory will be needed. A
582        bit of new cunning has reduced the workspace needed for groups with
583        alternatives. The 1000-alternative test pattern now uses 12 bytes of
584        workspace instead of running out of the 4096 that are available.
585    
586    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
587    
588    11. Applied patch from Google to remove an optimization that didn't quite work.
589        The report of the bug said:
590    
591          pcrecpp::RE("a*").FullMatch("aaa") matches, while
592          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
593          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
594    
595    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
596        it matched the wrong number of bytes.
597    
598    
599    Version 7.1 24-Apr-07
600    ---------------------
601    
602     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
603        that is more "standard", making use of automake and other Autotools. There
604        is some re-arrangement of the files and adjustment of comments consequent
605        on this.
606    
607     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
608        for recursive directory scanning broke on some systems because the files
609        are not scanned in any specific order and on different systems the order
610        was different. A call to "sort" has been inserted into RunGrepTest for the
611        approprate test as a short-term fix. In the longer term there may be an
612        alternative.
613    
614     3. I had an email from Eric Raymond about problems translating some of PCRE's
615        man pages to HTML (despite the fact that I distribute HTML pages, some
616        people do their own conversions for various reasons). The problems
617        concerned the use of low-level troff macros .br and .in. I have therefore
618        removed all such uses from the man pages (some were redundant, some could
619        be replaced by .nf/.fi pairs). The 132html script that I use to generate
620        HTML has been updated to handle .nf/.fi and to complain if it encounters
621        .br or .in.
622    
623     4. Updated comments in configure.ac that get placed in config.h.in and also
624        arranged for config.h to be included in the distribution, with the name
625        config.h.generic, for the benefit of those who have to compile without
626        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
627    
628     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
629        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
630        makevp.bat for the current PCRE, using the additional files
631        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
632    
633     6. A Windows user reported a minor discrepancy with test 2, which turned out
634        to be caused by a trailing space on an input line that had got lost in his
635        copy. The trailing space was an accident, so I've just removed it.
636    
637     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
638        that is needed.
639    
640     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
641        as "const" (a) because they are and (b) because it helps the PHP
642        maintainers who have recently made a script to detect big data structures
643        in the php code that should be moved to the .rodata section. I remembered
644        to update Builducptable as well, so it won't revert if ucptable.h is ever
645        re-created.
646    
647     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
648        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
649        order to be able to cut out the UTF-8 tables in the latter when UTF-8
650        support is not required. This saves 1.5-2K of code, which is important in
651        some applications.
652    
653        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
654        so as not to refer to the tables, even though these functions will never be
655        called when UTF-8 support is disabled. Otherwise there are problems with a
656        shared library.
657    
658    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
659    
660        (a) It was defining its arguments as char * instead of void *.
661    
662        (b) It was assuming that all moves were upwards in memory; this was true
663            a long time ago when I wrote it, but is no longer the case.
664    
665        The emulated memove() is provided for those environments that have neither
666        memmove() nor bcopy(). I didn't think anyone used it these days, but that
667        is clearly not the case, as these two bugs were recently reported.
668    
669    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
670        and Detrail to create the HTML documentation, the .txt form of the man
671        pages, and it removes trailing spaces from listed files. It also creates
672        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
673        case, it wraps all the #defines with #ifndefs. This script should be run
674        before "make dist".
675    
676    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
677        with Unicode property support.
678    
679        (a) For a maximizing quantifier, if the two different cases of the
680            character were of different lengths in their UTF-8 codings (there are
681            some cases like this - I found 11), and the matching function had to
682            back up over a mixture of the two cases, it incorrectly assumed they
683            were both the same length.
684    
685        (b) When PCRE was configured to use the heap rather than the stack for
686            recursion during matching, it was not correctly preserving the data for
687            the other case of a UTF-8 character when checking ahead for a match
688            while processing a minimizing repeat. If the check also involved
689            matching a wide character, but failed, corruption could cause an
690            erroneous result when trying to check for a repeat of the original
691            character.
692    
693    13. Some tidying changes to the testing mechanism:
694    
695        (a) The RunTest script now detects the internal link size and whether there
696            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
697            values substituted by "configure". (The RunGrepTest script already did
698            this for UTF-8.) The configure.ac script no longer substitutes the
699            relevant variables.
700    
701        (b) The debugging options /B and /D in pcretest show the compiled bytecode
702            with length and offset values. This means that the output is different
703            for different internal link sizes. Test 2 is skipped for link sizes
704            other than 2 because of this, bypassing the problem. Unfortunately,
705            there was also a test in test 3 (the locale tests) that used /B and
706            failed for link sizes other than 2. Rather than cut the whole test out,
707            I have added a new /Z option to pcretest that replaces the length and
708            offset values with spaces. This is now used to make test 3 independent
709            of link size. (Test 2 will be tidied up later.)
710    
711    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
712        segmentation fault instead of returning the appropriate error message.
713    
714    15. In multiline mode when the newline sequence was set to "any", the pattern
715        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
716        This doesn't seem right; it now treats the CRLF combination as the line
717        ending, and so does not match in that case. It's only a pattern such as ^$
718        that would hit this one: something like ^ABC$ would have failed after \r
719        and then tried again after \r\n.
720    
721    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
722        in an attempt to make files that differ only in their line terminators
723        compare equal. This works on Linux.
724    
725    17. Under certain error circumstances pcregrep might try to free random memory
726        as it exited. This is now fixed, thanks to valgrind.
727    
728    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
729        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
730        was because its rules for how to advance for /g after matching an empty
731        string at the end of a line did not allow for this case. They now check for
732        it specially.
733    
734    20. pcretest is supposed to handle patterns and data of any length, by
735        extending its buffers when necessary. It was getting this wrong when the
736        buffer for a data line had to be extended.
737    
738    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
739        CRLF as a newline sequence.
740    
741    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
742        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
743        I have nevertheless tidied it up.
744    
745    23. Added some casts to kill warnings from HP-UX ia64 compiler.
746    
747    24. Added a man page for pcre-config.
748    
749    
750    Version 7.0 19-Dec-06
751    ---------------------
752    
753     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
754        moving to gcc 4.1.1.
755    
756     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
757        sys/time.h, which is documented as needed for this function. It doesn't
758        seem to matter on Linux, but it showed up on some releases of OS X.
759    
760     3. It seems that there are systems where bytes whose values are greater than
761        127 match isprint() in the "C" locale. The "C" locale should be the
762        default when a C program starts up. In most systems, only ASCII printing
763        characters match isprint(). This difference caused the output from pcretest
764        to vary, making some of the tests fail. I have changed pcretest so that:
765    
766        (a) When it is outputting text in the compiled version of a pattern, bytes
767            other than 32-126 are always shown as hex escapes.
768    
769        (b) When it is outputting text that is a matched part of a subject string,
770            it does the same, unless a different locale has been set for the match
771            (using the /L modifier). In this case, it uses isprint() to decide.
772    
773     4. Fixed a major bug that caused incorrect computation of the amount of memory
774        required for a compiled pattern when options that changed within the
775        pattern affected the logic of the preliminary scan that determines the
776        length. The relevant options are -x, and -i in UTF-8 mode. The result was
777        that the computed length was too small. The symptoms of this bug were
778        either the PCRE error "internal error: code overflow" from pcre_compile(),
779        or a glibc crash with a message such as "pcretest: free(): invalid next
780        size (fast)". Examples of patterns that provoked this bug (shown in
781        pcretest format) are:
782    
783          /(?-x: )/x
784          /(?x)(?-x: \s*#\s*)/
785          /((?i)[\x{c0}])/8
786          /(?i:[\x{c0}])/8
787    
788        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
789        is now done differently.
790    
791     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
792        wrapper classes; (b) implement a new function in the C++ scanner that is
793        more efficient than the old way of doing things because it avoids levels of
794        recursion in the regex matching; (c) add a paragraph to the documentation
795        for the FullMatch() function.
796    
797     6. The escape sequence \n was being treated as whatever was defined as
798        "newline". Not only was this contrary to the documentation, which states
799        that \n is character 10 (hex 0A), but it also went horribly wrong when
800        "newline" was defined as CRLF. This has been fixed.
801    
802     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
803        was being set to -1 for the "end of line" case (supposedly a value that no
804        character can have). Though this value is never used (the check for end of
805        line is "zero bytes in current character"), it caused compiler complaints.
806        I've changed it to 0xffffffff.
807    
808     8. In pcre_version.c, the version string was being built by a sequence of
809        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
810        string (as it is for production releases) called a macro with an empty
811        argument. The C standard says the result of this is undefined. The gcc
812        compiler treats it as an empty string (which was what was wanted) but it is
813        reported that Visual C gives an error. The source has been hacked around to
814        avoid this problem.
815    
816     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
817        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
818        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
819        of them did).
820    
821    10. Originally, pcretest opened its input and output without "b"; then I was
822        told that "b" was needed in some environments, so it was added for release
823        5.0 to both the input and output. (It makes no difference on Unix-like
824        systems.) Later I was told that it is wrong for the input on Windows. I've
825        now abstracted the modes into two macros, to make it easier to fiddle with
826        them, and removed "b" from the input mode under Windows.
827    
828    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
829    
830    12. Added -help and --help to pcretest as an official way of being reminded
831        of the options.
832    
833    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
834        and pcrecpp.cc because they annoy compilers at high warning levels.
835    
836    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
837    
838    15. Fixed an occurrence of == in configure.ac that should have been = (shell
839        scripts are not C programs :-) and which was not noticed because it works
840        on Linux.
841    
842    16. pcretest is supposed to handle any length of pattern and data line (as one
843        line or as a continued sequence of lines) by extending its input buffer if
844        necessary. This feature was broken for very long pattern lines, leading to
845        a string of junk being passed to pcre_compile() if the pattern was longer
846        than about 50K.
847    
848    17. I have done a major re-factoring of the way pcre_compile() computes the
849        amount of memory needed for a compiled pattern. Previously, there was code
850        that made a preliminary scan of the pattern in order to do this. That was
851        OK when PCRE was new, but as the facilities have expanded, it has become
852        harder and harder to keep it in step with the real compile phase, and there
853        have been a number of bugs (see for example, 4 above). I have now found a
854        cunning way of running the real compile function in a "fake" mode that
855        enables it to compute how much memory it would need, while actually only
856        ever using a few hundred bytes of working memory and without too many
857        tests of the mode. This should make future maintenance and development
858        easier. A side effect of this work is that the limit of 200 on the nesting
859        depth of parentheses has been removed (though this was never a serious
860        limitation, I suspect). However, there is a downside: pcre_compile() now
861        runs more slowly than before (30% or more, depending on the pattern). I
862        hope this isn't a big issue. There is no effect on runtime performance.
863    
864    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
865        newline (only possible for the last line of a file) and it was a
866        pattern that set a locale (followed by /Lsomething), pcretest crashed.
867    
868    19. Added additional timing features to pcretest. (1) The -tm option now times
869        matching only, not compiling. (2) Both -t and -tm can be followed, as a
870        separate command line item, by a number that specifies the number of
871        repeats to use when timing. The default is 50000; this gives better
872        precision, but takes uncomfortably long for very large patterns.
873    
874    20. Extended pcre_study() to be more clever in cases where a branch of a
875        subpattern has no definite first character. For example, (a*|b*)[cd] would
876        previously give no result from pcre_study(). Now it recognizes that the
877        first character must be a, b, c, or d.
878    
879    21. There was an incorrect error "recursive call could loop indefinitely" if
880        a subpattern (or the entire pattern) that was being tested for matching an
881        empty string contained only one non-empty item after a nested subpattern.
882        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
883        incorrectly, because the \d was being skipped in the check.
884    
885    22. The pcretest program now has a new pattern option /B and a command line
886        option -b, which is equivalent to adding /B to every pattern. This causes
887        it to show the compiled bytecode, without the additional information that
888        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
889        is the same as /B/I).
890    
891    23. A new optimization is now able automatically to treat some sequences such
892        as a*b as a*+b. More specifically, if something simple (such as a character
893        or a simple class like \d) has an unlimited quantifier, and is followed by
894        something that cannot possibly match the quantified thing, the quantifier
895        is automatically "possessified".
896    
897    24. A recursive reference to a subpattern whose number was greater than 39
898        went wrong under certain circumstances in UTF-8 mode. This bug could also
899        have affected the operation of pcre_study().
900    
901    25. Realized that a little bit of performance could be had by replacing
902        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
903    
904    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
905    
906    27. Possessive quantifiers such as a++ were previously implemented by turning
907        them into atomic groups such as ($>a+). Now they have their own opcodes,
908        which improves performance. This includes the automatically created ones
909        from 23 above.
910    
911    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
912        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
913        the first matched character to be a colon. This applied both to named and
914        numbered groups.
915    
916    29. The ucpinternal.h header file was missing its idempotency #ifdef.
917    
918    30. I was sent a "project" file called libpcre.a.dev which I understand makes
919        building PCRE on Windows easier, so I have included it in the distribution.
920    
921    31. There is now a check in pcretest against a ridiculously large number being
922        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
923        loop, the loop is abandoned.
924    
925    32. Forward references to subpatterns in conditions such as (?(2)...) where
926        subpattern 2 is defined later cause pcre_compile() to search forwards in
927        the pattern for the relevant set of parentheses. This search went wrong
928        when there were unescaped parentheses in a character class, parentheses
929        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
930    
931    33. "Subroutine" calls and backreferences were previously restricted to
932        referencing subpatterns earlier in the regex. This restriction has now
933        been removed.
934    
935    34. Added a number of extra features that are going to be in Perl 5.10. On the
936        whole, these are just syntactic alternatives for features that PCRE had
937        previously implemented using the Python syntax or my own invention. The
938        other formats are all retained for compatibility.
939    
940        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
941            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
942            also .NET compatible.
943    
944        (b) A recursion or subroutine call to a named group can now be defined as
945            (?&name) as well as (?P>name).
946    
947        (c) A backreference to a named group can now be defined as \k<name> or
948            \k'name' as well as (?P=name). The new forms, as well as being in Perl
949            5.10, are also .NET compatible.
950    
951        (d) A conditional reference to a named group can now use the syntax
952            (?(<name>) or (?('name') as well as (?(name).
953    
954        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
955            groups (named and numbered) that are never evaluated inline, but can be
956            called as "subroutines" from elsewhere. In effect, the DEFINE condition
957            is always false. There may be only one alternative in such a group.
958    
959        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
960            as the simple (?(R). The condition is true only if the most recent
961            recursion is that of the given number or name. It does not search out
962            through the entire recursion stack.
963    
964        (g) The escape \gN or \g{N} has been added, where N is a positive or
965            negative number, specifying an absolute or relative reference.
966    
967    35. Tidied to get rid of some further signed/unsigned compiler warnings and
968        some "unreachable code" warnings.
969    
970    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
971        things, this adds five new scripts.
972    
973    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
974        There were also incompatibilities regarding the handling of \Q..\E inside
975        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
976        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
977    
978    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
979        matches an empty string, and forcibly breaks the loop. There were bugs in
980        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
981        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
982        separate and independent bugs (that affected different cases) have been
983        fixed.
984    
985    39. Refactored the code to abolish the use of different opcodes for small
986        capturing bracket numbers. This is a tidy that I avoided doing when I
987        removed the limit on the number of capturing brackets for 3.5 back in 2001.
988        The new approach is not only tidier, it makes it possible to reduce the
989        memory needed to fix the previous bug (38).
990    
991    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
992        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
993        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
994        mode.
995    
996    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
997        report.
998    
999    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1000        copy construction and assignment in the C++ wrapper.
1001    
1002    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1003        couple of bugs that could have given wrong results in the "--newline=crlf"
1004        case.
1005    
1006    44. Added a number of casts and did some reorganization of signed/unsigned int
1007        variables following suggestions from Dair Grant. Also renamed the variable
1008        "this" as "item" because it is a C++ keyword.
1009    
1010    45. Arranged for dftables to add
1011    
1012          #include "pcre_internal.h"
1013    
1014        to pcre_chartables.c because without it, gcc 4.x may remove the array
1015        definition from the final binary if PCRE is built into a static library and
1016        dead code stripping is activated.
1017    
1018    46. For an unanchored pattern, if a match attempt fails at the start of a
1019        newline sequence, and the newline setting is CRLF or ANY, and the next two
1020        characters are CRLF, advance by two characters instead of one.
1021    
1022    
1023    Version 6.7 04-Jul-06
1024    ---------------------
1025    
1026     1. In order to handle tests when input lines are enormously long, pcretest has
1027        been re-factored so that it automatically extends its buffers when
1028        necessary. The code is crude, but this _is_ just a test program. The
1029        default size has been increased from 32K to 50K.
1030    
1031     2. The code in pcre_study() was using the value of the re argument before
1032        testing it for NULL. (Of course, in any sensible call of the function, it
1033        won't be NULL.)
1034    
1035     3. The memmove() emulation function in pcre_internal.h, which is used on
1036        systems that lack both memmove() and bcopy() - that is, hardly ever -
1037        was missing a "static" storage class specifier.
1038    
1039     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1040        containing an extended class (one that cannot be represented by a bitmap
1041        because it contains high-valued characters or Unicode property items, e.g.
1042        [\pZ]). Almost always one would set UTF-8 mode when processing such a
1043        pattern, but PCRE should not loop if you do not (it no longer does).
1044        [Detail: two cases were found: (a) a repeated subpattern containing an
1045        extended class; (b) a recursive reference to a subpattern that followed a
1046        previous extended class. It wasn't skipping over the extended class
1047        correctly when UTF-8 mode was not set.]
1048    
1049     5. A negated single-character class was not being recognized as fixed-length
1050        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1051        compile error "lookbehind assertion is not fixed length".
1052    
1053     6. The RunPerlTest auxiliary script was showing an unexpected difference
1054        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1055        write a Perl script that can interpret lines of an input file either as
1056        byte characters or as UTF-8, which is what "perltest" was being required to
1057        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1058        can't do is switch easily at run time between having the "use utf8;" pragma
1059        or not. In the end, I fudged it by using the RunPerlTest script to insert
1060        "use utf8;" explicitly for the UTF-8 tests.
1061    
1062     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1063        the end of the subject string, contrary to the documentation and to what
1064        Perl does. This was true of both matching functions. Now it matches only at
1065        the start of the subject and immediately after *internal* newlines.
1066    
1067     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1068        a pointer to an int instead of a pointer to an unsigned long int. This
1069        caused problems on 64-bit systems.
1070    
1071     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1072        instance of the 'standard' template library not being so standard".
1073    
1074    10. There was no check on the number of named subpatterns nor the maximum
1075        length of a subpattern name. The product of these values is used to compute
1076        the size of the memory block for a compiled pattern. By supplying a very
1077        long subpattern name and a large number of named subpatterns, the size
1078        computation could be caused to overflow. This is now prevented by limiting
1079        the length of names to 32 characters, and the number of named subpatterns
1080        to 10,000.
1081    
1082    11. Subpatterns that are repeated with specific counts have to be replicated in
1083        the compiled pattern. The size of memory for this was computed from the
1084        length of the subpattern and the repeat count. The latter is limited to
1085        65535, but there was no limit on the former, meaning that integer overflow
1086        could in principle occur. The compiled length of a repeated subpattern is
1087        now limited to 30,000 bytes in order to prevent this.
1088    
1089    12. Added the optional facility to have named substrings with the same name.
1090    
1091    13. Added the ability to use a named substring as a condition, using the
1092        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1093        are numbers (not recommended). Forward references are permitted.
1094    
1095    14. Added forward references in named backreferences (if you see what I mean).
1096    
1097    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1098        pattern could run off the end of the subject. For example, the pattern
1099        "(?s)(.{1,5})"8 did this with the subject "ab".
1100    
1101    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1102        PCRE_CASELESS was set when matching characters that were quantified with ?
1103        or *.
1104    
1105    17. A character class other than a single negated character that had a minimum
1106        but no maximum quantifier - for example [ab]{6,} - was not handled
1107        correctly by pce_dfa_exec(). It would match only one character.
1108    
1109    18. A valid (though odd) pattern that looked like a POSIX character
1110        class but used an invalid character after [ (for example [[,abc,]]) caused
1111        pcre_compile() to give the error "Failed: internal error: code overflow" or
1112        in some cases to crash with a glibc free() error. This could even happen if
1113        the pattern terminated after [[ but there just happened to be a sequence of
1114        letters, a binary zero, and a closing ] in the memory that followed.
1115    
1116    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1117        over the years. Originally (before any Unicode support), just the bottom 8
1118        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1119        output from "man perlunicode" includes this:
1120    
1121          The regular expression compiler produces polymorphic opcodes.  That
1122          is, the pattern adapts to the data and automatically switches to
1123          the Unicode character scheme when presented with Unicode data--or
1124          instead uses a traditional byte scheme when presented with byte
1125          data.
1126    
1127        Sadly, a wide octal escape does not cause a switch, and in a string with
1128        no other multibyte characters, these octal escapes are treated as before.
1129        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
1130        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1131        Unicode string.
1132    
1133        I have not perpetrated such confusion in PCRE. Up till now, it took just
1134        the bottom 8 bits, as in old Perl. I have now made octal escapes with
1135        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1136        translate to the appropriate multibyte character.
1137    
1138    29. Applied some refactoring to reduce the number of warnings from Microsoft
1139        and Borland compilers. This has included removing the fudge introduced
1140        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1141        a warning about an unused variable.
1142    
1143    21. PCRE has not included VT (character 0x0b) in the set of whitespace
1144        characters since release 4.0, because Perl (from release 5.004) does not.
1145        [Or at least, is documented not to: some releases seem to be in conflict
1146        with the documentation.] However, when a pattern was studied with
1147        pcre_study() and all its branches started with \s, PCRE still included VT
1148        as a possible starting character. Of course, this did no harm; it just
1149        caused an unnecessary match attempt.
1150    
1151    22. Removed a now-redundant internal flag bit that recorded the fact that case
1152        dependency changed within the pattern. This was once needed for "required
1153        byte" processing, but is no longer used. This recovers a now-scarce options
1154        bit. Also moved the least significant internal flag bit to the most-
1155        significant bit of the word, which was not previously used (hangover from
1156        the days when it was an int rather than a uint) to free up another bit for
1157        the future.
1158    
1159    23. Added support for CRLF line endings as well as CR and LF. As well as the
1160        default being selectable at build time, it can now be changed at runtime
1161        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1162        specify that it is scanning data with non-default line endings.
1163    
1164    24. Changed the definition of CXXLINK to make it agree with the definition of
1165        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1166    
1167    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1168        recursions. This makes a big different to stack usage for some patterns.
1169    
1170    26. If a subpattern containing a named recursion or subroutine reference such
1171        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1172        the space required for the compiled pattern went wrong and gave too small a
1173        value. Depending on the environment, this could lead to "Failed: internal
1174        error: code overflow at offset 49" or "glibc detected double free or
1175        corruption" errors.
1176    
1177    27. Applied patches from Google (a) to support the new newline modes and (b) to
1178        advance over multibyte UTF-8 characters in GlobalReplace.
1179    
1180    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1181        difference for some implementation of PCRE in some Windows version.
1182    
1183    29. Added some extra testing facilities to pcretest:
1184    
1185        \q<number>   in a data line sets the "match limit" value
1186        \Q<number>   in a data line sets the "match recursion limt" value
1187        -S <number>  sets the stack size, where <number> is in megabytes
1188    
1189        The -S option isn't available for Windows.
1190    
1191    
1192    Version 6.6 06-Feb-06
1193    ---------------------
1194    
1195     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1196        in pcreposix.h. I have copied the definition from pcre.h.
1197    
1198     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1199        because pcre.h is no longer a built file.
1200    
1201     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1202        not normally included in the compiled code.
1203    
1204    
1205    Version 6.5 01-Feb-06
1206    ---------------------
1207    
1208     1. When using the partial match feature with pcre_dfa_exec(), it was not
1209        anchoring the second and subsequent partial matches at the new starting
1210        point. This could lead to incorrect results. For example, with the pattern
1211        /1234/, partially matching against "123" and then "a4" gave a match.
1212    
1213     2. Changes to pcregrep:
1214    
1215        (a) All non-match returns from pcre_exec() were being treated as failures
1216            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1217            error message is output. Some extra information is given for the
1218            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
1219            probably the only errors that are likely to be caused by users (by
1220            specifying a regex that has nested indefinite repeats, for instance).
1221            If there are more than 20 of these errors, pcregrep is abandoned.
1222    
1223        (b) A binary zero was treated as data while matching, but terminated the
1224            output line if it was written out. This has been fixed: binary zeroes
1225            are now no different to any other data bytes.
1226    
1227        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1228            used to set a locale for matching. The --locale=xxxx long option has
1229            been added (no short equivalent) to specify a locale explicitly on the
1230            pcregrep command, overriding the environment variables.
1231    
1232        (d) When -B was used with -n, some line numbers in the output were one less
1233            than they should have been.
1234    
1235        (e) Added the -o (--only-matching) option.
1236    
1237        (f) If -A or -C was used with -c (count only), some lines of context were
1238            accidentally printed for the final match.
1239    
1240        (g) Added the -H (--with-filename) option.
1241    
1242        (h) The combination of options -rh failed to suppress file names for files
1243            that were found from directory arguments.
1244    
1245        (i) Added the -D (--devices) and -d (--directories) options.
1246    
1247        (j) Added the -F (--fixed-strings) option.
1248    
1249        (k) Allow "-" to be used as a file name for -f as well as for a data file.
1250    
1251        (l) Added the --colo(u)r option.
1252    
1253        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1254            is not present by default.
1255    
1256     3. A nasty bug was discovered in the handling of recursive patterns, that is,
1257        items such as (?R) or (?1), when the recursion could match a number of
1258        alternatives. If it matched one of the alternatives, but subsequently,
1259        outside the recursion, there was a failure, the code tried to back up into
1260        the recursion. However, because of the way PCRE is implemented, this is not
1261        possible, and the result was an incorrect result from the match.
1262    
1263        In order to prevent this happening, the specification of recursion has
1264        been changed so that all such subpatterns are automatically treated as
1265        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1266    
1267     4. I had overlooked the fact that, in some locales, there are characters for
1268        which isalpha() is true but neither isupper() nor islower() are true. In
1269        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1270        and ordfeminine) are like this. This affected the treatment of \w and \W
1271        when they appeared in character classes, but not when they appeared outside
1272        a character class. The bit map for "word" characters is now created
1273        separately from the results of isalnum() instead of just taking it from the
1274        upper, lower, and digit maps. (Plus the underscore character, of course.)
1275    
1276     5. The above bug also affected the handling of POSIX character classes such as
1277        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1278        permanent tables. Instead, the bit maps for such a class were previously
1279        created as the appropriate unions of the upper, lower, and digit bitmaps.
1280        Now they are created by subtraction from the [[:word:]] class, which has
1281        its own bitmap.
1282    
1283     6. The [[:blank:]] character class matches horizontal, but not vertical space.
1284        It is created by subtracting the vertical space characters (\x09, \x0a,
1285        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1286        subtraction was done in the overall bitmap for a character class, meaning
1287        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1288        be recognized. This bug has been fixed.
1289    
1290     7. Patches from the folks at Google:
1291    
1292          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1293          real life, but is still worth protecting against".
1294    
1295          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1296          regular expressions".
1297    
1298          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1299          have it.
1300    
1301          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1302          "configure" and the latter not, in order to fix a problem somebody had
1303          with compiling the Arg class on HP-UX.
1304    
1305          (e) Improve the error-handling of the C++ wrapper a little bit.
1306    
1307          (f) New tests for checking recursion limiting.
1308    
1309     8. The pcre_memmove() function, which is used only if the environment does not
1310        have a standard memmove() function (and is therefore rarely compiled),
1311        contained two bugs: (a) use of int instead of size_t, and (b) it was not
1312        returning a result (though PCRE never actually uses the result).
1313    
1314     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1315        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1316        returned instead of calling malloc() with an overflowing number that would
1317        most likely cause subsequent chaos.
1318    
1319    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1320    
1321    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1322        with this option is matched, the nmatch and pmatch options of regexec() are
1323        ignored.
1324    
1325    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1326        provided in case anyone wants to the the POSIX interface with UTF-8
1327        strings.
1328    
1329    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1330        C++ linking (needed for some HP-UX environments).
1331    
1332    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1333        (unused parameter) and in the pcre_printint() function (omitted "default"
1334        switch label when the default is to do nothing).
1335    
1336    15. Added some code to make it possible, when PCRE is compiled as a C++
1337        library, to replace subject pointers for pcre_exec() with a smart pointer
1338        class, thus making it possible to process discontinuous strings.
1339    
1340    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1341        much the same function. They were added by different people who were trying
1342        to make PCRE easy to compile on non-Unix systems. It has been suggested
1343        that PCRE_EXPORT be abolished now that there is more automatic apparatus
1344        for compiling on Windows systems. I have therefore replaced it with
1345        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1346        defaults to "extern" for C or "extern C" for C++, which works fine on
1347        Unix-like systems. It is now possible to override the value of PCRE_DATA_
1348        SCOPE with something explicit in config.h. In addition:
1349    
1350        (a) pcreposix.h still had just "extern" instead of either of these macros;
1351            I have replaced it with PCRE_DATA_SCOPE.
1352    
1353        (b) Functions such as _pcre_xclass(), which are internal to the library,
1354            but external in the C sense, all had PCRE_EXPORT in their definitions.
1355            This is apparently wrong for the Windows case, so I have removed it.
1356            (It makes no difference on Unix-like systems.)
1357    
1358    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1359        of recursive calls to match(). This is different to MATCH_LIMIT because
1360        that limits the total number of calls to match(), not all of which increase
1361        the depth of recursion. Limiting the recursion depth limits the amount of
1362        stack (or heap if NO_RECURSE is set) that is used. The default can be set
1363        when PCRE is compiled, and changed at run time. A patch from Google adds
1364        this functionality to the C++ interface.
1365    
1366    18. Changes to the handling of Unicode character properties:
1367    
1368        (a) Updated the table to Unicode 4.1.0.
1369    
1370        (b) Recognize characters that are not in the table as "Cn" (undefined).
1371    
1372        (c) I revised the way the table is implemented to a much improved format
1373            which includes recognition of ranges. It now supports the ranges that
1374            are defined in UnicodeData.txt, and it also amalgamates other
1375            characters into ranges. This has reduced the number of entries in the
1376            table from around 16,000 to around 3,000, thus reducing its size
1377            considerably. I realized I did not need to use a tree structure after
1378            all - a binary chop search is just as efficient. Having reduced the
1379            number of entries, I extended their size from 6 bytes to 8 bytes to
1380            allow for more data.
1381    
1382        (d) Added support for Unicode script names via properties such as \p{Han}.
1383    
1384    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1385        matching that character.
1386    
1387    20. When matching a repeated Unicode property with a minimum greater than zero,
1388        (for example \pL{2,}), PCRE could look past the end of the subject if it
1389        reached it while seeking the minimum number of characters. This could
1390        happen only if some of the characters were more than one byte long, because
1391        there is a check for at least the minimum number of bytes.
1392    
1393    21. Refactored the implementation of \p and \P so as to be more general, to
1394        allow for more different types of property in future. This has changed the
1395        compiled form incompatibly. Anybody with saved compiled patterns that use
1396        \p or \P will have to recompile them.
1397    
1398    22. Added "Any" and "L&" to the supported property types.
1399    
1400    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1401        but give a compile time error if the value is greater than 0xff.
1402    
1403    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1404        accidentally not being installed or uninstalled.
1405    
1406    25. The pcre.h file was built from pcre.h.in, but the only changes that were
1407        made were to insert the current release number. This seemed silly, because
1408        it made things harder for people building PCRE on systems that don't run
1409        "configure". I have turned pcre.h into a distributed file, no longer built
1410        by "configure", with the version identification directly included. There is
1411        no longer a pcre.h.in file.
1412    
1413        However, this change necessitated a change to the pcre-config script as
1414        well. It is built from pcre-config.in, and one of the substitutions was the
1415        release number. I have updated configure.ac so that ./configure now finds
1416        the release number by grepping pcre.h.
1417    
1418    26. Added the ability to run the tests under valgrind.
1419    
1420    
1421    Version 6.4 05-Sep-05
1422    ---------------------
1423    
1424     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1425        "--" to be printed when multiple files were scanned, even when none of the
1426        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1427        consider it to be a bug, and have restored the previous behaviour.
1428    
1429     2. A couple of code tidies to get rid of compiler warnings.
1430    
1431     3. The pcretest program used to cheat by referring to symbols in the library
1432        whose names begin with _pcre_. These are internal symbols that are not
1433        really supposed to be visible externally, and in some environments it is
1434        possible to suppress them. The cheating is now confined to including
1435        certain files from the library's source, which is a bit cleaner.
1436    
1437     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1438        file's purpose clearer.
1439    
1440     5. Reorganized pcre_ucp_findchar().
1441    
1442    
1443    Version 6.3 15-Aug-05
1444    ---------------------
1445    
1446     1. The file libpcre.pc.in did not have general read permission in the tarball.
1447    
1448     2. There were some problems when building without C++ support:
1449    
1450        (a) If C++ support was not built, "make install" and "make test" still
1451            tried to test it.
1452    
1453        (b) There were problems when the value of CXX was explicitly set. Some
1454            changes have been made to try to fix these, and ...
1455    
1456        (c) --disable-cpp can now be used to explicitly disable C++ support.
1457    
1458        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1459            backslash in a target when C++ was disabled. This confuses some
1460            versions of "make", apparently. Using an intermediate variable solves
1461            this. (Same for CPP_LOBJ.)
1462    
1463     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1464        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1465        necessary on certain architectures.
1466    
1467     4. Added a setting of -export-symbols-regex to the link command to remove
1468        those symbols that are exported in the C sense, but actually are local
1469        within the library, and not documented. Their names all begin with
1470        "_pcre_". This is not a perfect job, because (a) we have to except some
1471        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1472        available (and never for static libraries). I have made a note to try to
1473        find a way round (a) in the future.
1474    
1475    
1476    Version 6.2 01-Aug-05
1477    ---------------------
1478    
1479     1. There was no test for integer overflow of quantifier values. A construction
1480        such as {1111111111111111} would give undefined results. What is worse, if
1481        a minimum quantifier for a parenthesized subpattern overflowed and became
1482        negative, the calculation of the memory size went wrong. This could have
1483        led to memory overwriting.
1484    
1485     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1486    
1487     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1488        operating environments where this matters.
1489    
1490     4. Applied Giuseppe Maxia's patch to add additional features for controlling
1491        PCRE options from within the C++ wrapper.
1492    
1493     5. Named capturing subpatterns were not being correctly counted when a pattern
1494        was compiled. This caused two problems: (a) If there were more than 100
1495        such subpatterns, the calculation of the memory needed for the whole
1496        compiled pattern went wrong, leading to an overflow error. (b) Numerical
1497        back references of the form \12, where the number was greater than 9, were
1498        not recognized as back references, even though there were sufficient
1499        previous subpatterns.
1500    
1501     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1502        versions of gcc, e.g. 2.95.4.
1503    
1504    
1505    Version 6.1 21-Jun-05
1506    ---------------------
1507    
1508     1. There was one reference to the variable "posix" in pcretest.c that was not
1509        surrounded by "#if !defined NOPOSIX".
1510    
1511     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1512        the cross-check on the old pcre_info() function, for the benefit of the
1513        cut-down version of PCRE that is currently imported into Exim.
1514    
1515     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1516        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1517        patterns that start (?i) but I don't think that matters. The use of (?i) is
1518        just an example; this all applies to the other options as well.
1519    
1520     4. Since libtool seems to echo the compile commands it is issuing, the output
1521        from "make" can be reduced a bit by putting "@" in front of each libtool
1522        compile command.
1523    
1524     5. Patch from the folks at Google for configure.in to be a bit more thorough
1525        in checking for a suitable C++ installation before trying to compile the
1526        C++ stuff. This should fix a reported problem when a compiler was present,
1527        but no suitable headers.
1528    
1529     6. The man pages all had just "PCRE" as their title. I have changed them to
1530        be the relevant file name. I have also arranged that these names are
1531        retained in the file doc/pcre.txt, which is a concatenation in text format
1532        of all the man pages except the little individual ones for each function.
1533    
1534     7. The NON-UNIX-USE file had not been updated for the different set of source
1535        files that come with release 6. I also added a few comments about the C++
1536        wrapper.
1537    
1538    
1539    Version 6.0 07-Jun-05
1540    ---------------------
1541    
1542     1. Some minor internal re-organization to help with my DFA experiments.
1543    
1544     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1545        didn't matter for the library itself when fully configured, but did matter
1546        when compiling without UCP support, or within Exim, where the ucp files are
1547        not imported.
1548    
1549     3. Refactoring of the library code to split up the various functions into
1550        different source modules. The addition of the new DFA matching code (see
1551        below) to a single monolithic source would have made it really too
1552        unwieldy, quite apart from causing all the code to be include in a
1553        statically linked application, when only some functions are used. This is
1554        relevant even without the DFA addition now that patterns can be compiled in
1555        one application and matched in another.
1556    
1557        The downside of splitting up is that there have to be some external
1558        functions and data tables that are used internally in different modules of
1559        the library but which are not part of the API. These have all had their
1560        names changed to start with "_pcre_" so that they are unlikely to clash
1561        with other external names.
1562    
1563     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1564        a different (DFA) algorithm. Although it is slower than the original
1565        function, it does have some advantages for certain types of matching
1566        problem.
1567    
1568     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1569        including restarting after a partial match.
1570    
1571     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1572        defined when compiling for Windows was sent to me. I have put it into the
1573        code, though I have no means of testing or verifying it.
1574    
1575     7. Added the pcre_refcount() auxiliary function.
1576    
1577     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1578        match before or at the first newline in the subject string. In pcretest,
1579        the /f option on a pattern can be used to set this.
1580    
1581     9. A repeated \w when used in UTF-8 mode with characters greater than 256
1582        would behave wrongly. This has been present in PCRE since release 4.0.
1583    
1584    10. A number of changes to the pcregrep command:
1585    
1586        (a) Refactored how -x works; insert ^(...)$ instead of setting
1587            PCRE_ANCHORED and checking the length, in preparation for adding
1588            something similar for -w.
1589    
1590        (b) Added the -w (match as a word) option.
1591    
1592        (c) Refactored the way lines are read and buffered so as to have more
1593            than one at a time available.
1594    
1595        (d) Implemented a pcregrep test script.
1596    
1597        (e) Added the -M (multiline match) option. This allows patterns to match
1598            over several lines of the subject. The buffering ensures that at least
1599            8K, or the rest of the document (whichever is the shorter) is available
1600            for matching (and similarly the previous 8K for lookbehind assertions).
1601    
1602        (f) Changed the --help output so that it now says
1603    
1604              -w, --word-regex(p)
1605    
1606            instead of two lines, one with "regex" and the other with "regexp"
1607            because that confused at least one person since the short forms are the
1608            same. (This required a bit of code, as the output is generated
1609            automatically from a table. It wasn't just a text change.)
1610    
1611        (g) -- can be used to terminate pcregrep options if the next thing isn't an
1612            option but starts with a hyphen. Could be a pattern or a path name
1613            starting with a hyphen, for instance.
1614    
1615        (h) "-" can be given as a file name to represent stdin.
1616    
1617        (i) When file names are being printed, "(standard input)" is used for
1618            the standard input, for compatibility with GNU grep. Previously
1619            "<stdin>" was used.
1620    
1621        (j) The option --label=xxx can be used to supply a name to be used for
1622            stdin when file names are being printed. There is no short form.
1623    
1624        (k) Re-factored the options decoding logic because we are going to add
1625            two more options that take data. Such options can now be given in four
1626            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1627    
1628        (l) Added the -A, -B, and -C options for requesting that lines of context
1629            around matches be printed.
1630    
1631        (m) Added the -L option to print the names of files that do not contain
1632            any matching lines, that is, the complement of -l.
1633    
1634        (n) The return code is 2 if any file cannot be opened, but pcregrep does
1635            continue to scan other files.
1636    
1637        (o) The -s option was incorrectly implemented. For compatibility with other
1638            greps, it now suppresses the error message for a non-existent or non-
1639            accessible file (but not the return code). There is a new option called
1640            -q that suppresses the output of matching lines, which was what -s was
1641            previously doing.
1642    
1643        (p) Added --include and --exclude options to specify files for inclusion
1644            and exclusion when recursing.
1645    
1646    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1647        Hopefully, it now does.
1648    
1649    12. Missing cast in pcre_study().
1650    
1651    13. Added an "uninstall" target to the makefile.
1652    
1653    14. Replaced "extern" in the function prototypes in Makefile.in with
1654        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1655        world, but is set differently for Windows.
1656    
1657    15. Added a second compiling function called pcre_compile2(). The only
1658        difference is that it has an extra argument, which is a pointer to an
1659        integer error code. When there is a compile-time failure, this is set
1660        non-zero, in addition to the error test pointer being set to point to an
1661        error message. The new argument may be NULL if no error number is required
1662        (but then you may as well call pcre_compile(), which is now just a
1663        wrapper). This facility is provided because some applications need a
1664        numeric error indication, but it has also enabled me to tidy up the way
1665        compile-time errors are handled in the POSIX wrapper.
1666    
1667    16. Added VPATH=.libs to the makefile; this should help when building with one
1668        prefix path and installing with another. (Or so I'm told by someone who
1669        knows more about this stuff than I do.)
1670    
1671    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1672        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1673        match everything, including newlines. This is not POSIX-compatible, but
1674        somebody wanted the feature. From pcretest it can be activated by using
1675        both the P and the s flags.
1676    
1677    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1678    
1679    19. libpcre.pc was being incorrectly installed as executable.
1680    
1681    20. A couple of places in pcretest check for end-of-line by looking for '\n';
1682        it now also looks for '\r' so that it will work unmodified on Windows.
1683    
1684    21. Added Google's contributed C++ wrapper to the distribution.
1685    
1686    22. Added some untidy missing memory free() calls in pcretest, to keep
1687        Electric Fence happy when testing.
1688    
1689    
1690    
1691    Version 5.0 13-Sep-04
1692    ---------------------
1693    
1694     1. Internal change: literal characters are no longer packed up into items
1695        containing multiple characters in a single byte-string. Each character
1696        is now matched using a separate opcode. However, there may be more than one
1697        byte in the character in UTF-8 mode.
1698    
1699     2. The pcre_callout_block structure has two new fields: pattern_position and
1700        next_item_length. These contain the offset in the pattern to the next match
1701        item, and its length, respectively.
1702    
1703     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1704        insertion of callouts before each pattern item. Added the /C option to
1705        pcretest to make use of this.
1706    
1707     4. On the advice of a Windows user, the lines
1708    
1709          #if defined(_WIN32) || defined(WIN32)
1710          _setmode( _fileno( stdout ), 0x8000 );
1711          #endif  /* defined(_WIN32) || defined(WIN32) */
1712    
1713        have been added to the source of pcretest. This apparently does useful
1714        magic in relation to line terminators.
1715    
1716     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1717        for the benefit of those environments where the "b" makes a difference.
1718    
1719     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1720        to know about it. I have put a hack into configure.in that adds in code
1721        to set GCC=yes if CC=icc. This seems to end up at a point in the
1722        generated configure script that is early enough to affect the setting of
1723        compiler options, which is what is needed, but I have no means of testing
1724        whether it really works. (The user who reported this had patched the
1725        generated configure script, which of course I cannot do.)
1726    
1727        LATER: After change 22 below (new libtool files), the configure script
1728        seems to know about icc (and also ecc). Therefore, I have commented out
1729        this hack in configure.in.
1730    
1731     7. Added support for pkg-config (2 patches were sent in).
1732    
1733     8. Negated POSIX character classes that used a combination of internal tables
1734        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1735        [[:^ascii]]. Typically, they would match almost any characters. The other
1736        POSIX classes were not broken in this way.
1737    
1738     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1739        to find the match, as PCRE was deluded into thinking that the match had to
1740        start at the start point or following a newline. The same bug applied to
1741        patterns with negative forward assertions or any backward assertions
1742        preceding ".*" at the start, unless the pattern required a fixed first
1743        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1744    
1745    10. In UTF-8 mode, when moving forwards in the subject after a failed match
1746        starting at the last subject character, bytes beyond the end of the subject
1747        string were read.
1748    
1749    11. Renamed the variable "class" as "classbits" to make life easier for C++
1750        users. (Previously there was a macro definition, but it apparently wasn't
1751        enough.)
1752    
1753    12. Added the new field "tables" to the extra data so that tables can be passed
1754        in at exec time, or the internal tables can be re-selected. This allows
1755        a compiled regex to be saved and re-used at a later time by a different
1756        program that might have everything at different addresses.
1757    
1758    13. Modified the pcre-config script so that, when run on Solaris, it shows a
1759        -R library as well as a -L library.
1760    
1761    14. The debugging options of pcretest (-d on the command line or D on a
1762        pattern) showed incorrect output for anything following an extended class
1763        that contained multibyte characters and which was followed by a quantifier.
1764    
1765    15. Added optional support for general category Unicode character properties
1766        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1767        support. It adds about 90K to the size of the library. The meanings of the
1768        inbuilt class escapes such as \d and \s have NOT been changed.
1769    
1770    16. Updated pcredemo.c to include calls to free() to release the memory for the
1771        compiled pattern.
1772    
1773    17. The generated file chartables.c was being created in the source directory
1774        instead of in the building directory. This caused the build to fail if the
1775        source directory was different from the building directory, and was
1776        read-only.
1777    
1778    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1779        file. No doubt somebody will tell me if they don't make sense... Also added
1780        Dan Mooney's comments about building on OpenVMS.
1781    
1782    19. Added support for partial matching via the PCRE_PARTIAL option for
1783        pcre_exec() and the \P data escape in pcretest.
1784    
1785    20. Extended pcretest with 3 new pattern features:
1786    
1787        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
1788              write the compiled pattern to the file whose name is "rest-of-line".
1789              This is a straight binary dump of the data, with the saved pointer to
1790              the character tables forced to be NULL. The study data, if any, is
1791              written too. After writing, pcretest reads a new pattern.
1792    
1793        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1794              compiled pattern from the given file. There must not be any
1795              occurrences of "<" in the file name (pretty unlikely); if there are,
1796              pcretest will instead treat the initial "<" as a pattern delimiter.
1797              After reading in the pattern, pcretest goes on to read data lines as
1798              usual.
1799    
1800        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1801              and 16-bit fields in a compiled pattern, to simulate a pattern that
1802              was compiled on a host of opposite endianness.
1803    
1804    21. The pcre-exec() function can now cope with patterns that were compiled on
1805        hosts of opposite endianness, with this restriction:
1806    
1807          As for any compiled expression that is saved and used later, the tables
1808          pointer field cannot be preserved; the extra_data field in the arguments
1809          to pcre_exec() should be used to pass in a tables address if a value
1810          other than the default internal tables were used at compile time.
1811    
1812    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1813        now diagnosed as an error. Previously, most of the time, a negative number
1814        would have been treated as zero, but if in addition "ovector" was passed as
1815        NULL, a crash could occur.
1816    
1817    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1818        new versions from the libtool 1.5 distribution (the last one is a copy of
1819        a file called libtool.m4). This seems to have fixed the need to patch
1820        "configure" to support Darwin 1.3 (which I used to do). However, I still
1821        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1822        workstation).
1823    
1824    24. Changed the PCRE licence to be the more standard "BSD" licence.
1825    
1826    
1827    Version 4.5 01-Dec-03
1828    ---------------------
1829    
1830     1. There has been some re-arrangement of the code for the match() function so
1831        that it can be compiled in a version that does not call itself recursively.
1832        Instead, it keeps those local variables that need separate instances for
1833        each "recursion" in a frame on the heap, and gets/frees frames whenever it
1834        needs to "recurse". Keeping track of where control must go is done by means
1835        of setjmp/longjmp. The whole thing is implemented by a set of macros that
1836        hide most of the details from the main code, and operates only if
1837        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1838        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1839        operating.
1840    
1841        To make it easier for callers to provide specially tailored get/free
1842        functions for this usage, two new functions, pcre_stack_malloc, and
1843        pcre_stack_free, are used. They are always called in strict stacking order,
1844        and the size of block requested is always the same.
1845    
1846        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1847        PCRE has been compiled to use the stack or the heap for recursion. The
1848        -C option of pcretest uses this to show which version is compiled.
1849    
1850        A new data escape \S, is added to pcretest; it causes the amounts of store
1851        obtained and freed by both kinds of malloc/free at match time to be added
1852        to the output.
1853    
1854     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1855        what's available on my current Linux desktop machine.
1856    
1857     3. When matching a UTF-8 string, the test for a valid string at the start has
1858        been extended. If start_offset is not zero, PCRE now checks that it points
1859        to a byte that is the start of a UTF-8 character. If not, it returns
1860        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1861        this is necessary because there may be backward assertions in the pattern.
1862        When matching the same subject several times, it may save resources to use
1863        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1864    
1865     4. The code for checking the validity of UTF-8 strings has been tightened so
1866        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1867        containing "overlong sequences".
1868    
1869     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1870        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1871        should have been "&", but it just so happened that all the cases this let
1872        through by mistake were picked up later in the function.
1873    
1874     6. I had used a variable called "isblank" - this is a C99 function, causing
1875        some compilers to warn. To avoid this, I renamed it (as "blankclass").
1876    
1877     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1878        prompting; (b) run "./pcretest /dev/null" at the start of the test script
1879        so the version is shown; (c) stop "make test" echoing "./RunTest".
1880    
1881     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1882    
1883     9. The prototype for memmove() for systems that don't have it was using
1884        size_t, but the inclusion of the header that defines size_t was later. I've
1885        moved the #includes for the C headers earlier to avoid this.
1886    
1887    10. Added some adjustments to the code to make it easier to compiler on certain
1888        special systems:
1889    
1890          (a) Some "const" qualifiers were missing.
1891          (b) Added the macro EXPORT before all exported functions; by default this
1892              is defined to be empty.
1893          (c) Changed the dftables auxiliary program (that builds chartables.c) so
1894              that it reads its output file name as an argument instead of writing
1895              to the standard output and assuming this can be redirected.
1896    
1897    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
1898        class containing characters with values greater than 255, PCRE compilation
1899        went into a loop.
1900    
1901    12. A recursive reference to a subpattern that was within another subpattern
1902        that had a minimum quantifier of zero caused PCRE to crash. For example,
1903        (x(y(?2))z)? provoked this bug with a subject that got as far as the
1904        recursion. If the recursively-called subpattern itself had a zero repeat,
1905        that was OK.
1906    
1907    13. In pcretest, the buffer for reading a data line was set at 30K, but the
1908        buffer into which it was copied (for escape processing) was still set at
1909        1024, so long lines caused crashes.
1910    
1911    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
1912        "internal error: code overflow...". This applied to any character class
1913        that was followed by a possessive quantifier.
1914    
1915    15. Modified the Makefile to add libpcre.la as a prerequisite for
1916        libpcreposix.la because I was told this is needed for a parallel build to
1917        work.
1918    
1919    16. If a pattern that contained .* following optional items at the start was
1920        studied, the wrong optimizing data was generated, leading to matching
1921        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
1922        matching string must start with a or b or c. The correct conclusion for
1923        this pattern is that a match can start with any character.
1924    
1925    
1926    Version 4.4 13-Aug-03
1927    ---------------------
1928    
1929     1. In UTF-8 mode, a character class containing characters with values between
1930        127 and 255 was not handled correctly if the compiled pattern was studied.
1931        In fixing this, I have also improved the studying algorithm for such
1932        classes (slightly).
1933    
1934     2. Three internal functions had redundant arguments passed to them. Removal
1935        might give a very teeny performance improvement.
1936    
1937     3. Documentation bug: the value of the capture_top field in a callout is *one
1938        more than* the number of the hightest numbered captured substring.
1939    
1940     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
1941        in incorrectly linking with a previously installed version. They now link
1942        explicitly with libpcre.la.
1943    
1944     5. configure.in no longer needs to recognize Cygwin specially.
1945    
1946     6. A problem in pcre.in for Windows platforms is fixed.
1947    
1948     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
1949        pcretest, it used to include the size of the study block as part of its
1950        output. Unfortunately, the structure contains a field that has a different
1951        size on different hardware architectures. This meant that the tests that
1952        showed this size failed. As the block is currently always of a fixed size,
1953        this information isn't actually particularly useful in pcretest output, so
1954        I have just removed it.
1955    
1956     8. Three pre-processor statements accidentally did not start in column 1.
1957        Sadly, there are *still* compilers around that complain, even though
1958        standard C has not required this for well over a decade. Sigh.
1959    
1960     9. In pcretest, the code for checking callouts passed small integers in the
1961        callout_data field, which is a void * field. However, some picky compilers
1962        complained about the casts involved for this on 64-bit systems. Now
1963        pcretest passes the address of the small integer instead, which should get
1964        rid of the warnings.
1965    
1966    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
1967        both compile and run time, and gives an error if an invalid UTF-8 sequence
1968        is found. There is a option for disabling this check in cases where the
1969        string is known to be correct and/or the maximum performance is wanted.
1970    
1971    11. In response to a bug report, I changed one line in Makefile.in from
1972    
1973            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
1974        to
1975            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
1976    
1977        to look similar to other lines, but I have no way of telling whether this
1978        is the right thing to do, as I do not use Windows. No doubt I'll get told
1979        if it's wrong...
1980    
1981    
1982    Version 4.3 21-May-03
1983    ---------------------
1984    
1985    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
1986       Makefile.
1987    
1988    2. Some refactoring to improve the quality of the code:
1989    
1990       (i)   The utf8_table... variables are now declared "const".
1991    
1992       (ii)  The code for \cx, which used the "case flipping" table to upper case
1993             lower case letters, now just substracts 32. This is ASCII-specific,
1994             but the whole concept of \cx is ASCII-specific, so it seems
1995             reasonable.
1996    
1997       (iii) PCRE was using its character types table to recognize decimal and
1998             hexadecimal digits in the pattern. This is silly, because it handles
1999             only 0-9, a-f, and A-F, but the character types table is locale-
2000             specific, which means strange things might happen. A private
2001             table is now used for this - though it costs 256 bytes, a table is
2002             much faster than multiple explicit tests. Of course, the standard
2003             character types table is still used for matching digits in subject
2004             strings against \d.
2005    
2006       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
2007             ending in _t are). So I've renamed it as ESC_tee.
2008    
2009    3. The first argument for regexec() in the POSIX wrapper should have been
2010       defined as "const".
2011    
2012    4. Changed pcretest to use malloc() for its buffers so that they can be
2013       Electric Fenced for debugging.
2014    
2015    5. There were several places in the code where, in UTF-8 mode, PCRE would try
2016       to read one or more bytes before the start of the subject string. Often this
2017       had no effect on PCRE's behaviour, but in some circumstances it could
2018       provoke a segmentation fault.
2019    
2020    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
2021       to try to read one or more bytes before the start of the subject string.
2022    
2023    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
2024       UTF-8 support could misbehave in various ways if the subject string
2025       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
2026       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
2027       back over UTF-8 characters.)
2028    
2029    
2030    Version 4.2 14-Apr-03
2031    ---------------------
2032    
2033    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
2034    
2035    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
2036         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
2037         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
2038         [WIN_PREFIX]: new variable, "cyg" for Cygwin
2039         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
2040           and BUILD_EXEEXT
2041         Note: automatic setting of the BUILD variables is not yet working
2042         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
2043           compile-time but not at link-time
2044         [LINK]: use for linking executables only
2045         make different versions for Windows and non-Windows
2046         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
2047           libraries
2048         [LINK_FOR_BUILD]: new variable
2049         [OBJEXT]: use throughout
2050         [EXEEXT]: use throughout
2051         <winshared>: new target
2052         <wininstall>: new target
2053         <dftables.o>: use native compiler
2054         <dftables>: use native linker
2055         <install>: handle Windows platform correctly
2056         <clean>: ditto
2057         <check>: ditto
2058         copy DLL to top builddir before testing
2059    
2060       As part of these changes, -no-undefined was removed again. This was reported
2061       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
2062       in any case.
2063    
2064    3. Some tidies to get rid of compiler warnings:
2065    
2066       . In the match_data structure, match_limit was an unsigned long int, whereas
2067         match_call_count was an int. I've made them both unsigned long ints.
2068    
2069       . In pcretest the fact that a const uschar * doesn't automatically cast to
2070         a void * provoked a warning.
2071    
2072       . Turning on some more compiler warnings threw up some "shadow" variables
2073         and a few more missing casts.
2074    
2075    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
2076       option, a class that contained a single character with a value between 128
2077       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
2078    
2079    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
2080       option, a class that contained several characters, but with at least one
2081       whose value was between 128 and 255 caused PCRE to crash.
2082    
2083    
2084    Version 4.1 12-Mar-03
2085    ---------------------
2086    
2087    1. Compiling with gcc -pedantic found a couple of places where casts were
2088    needed, and a string in dftables.c that was longer than standard compilers are
2089    required to support.
2090    
2091    2. Compiling with Sun's compiler found a few more places where the code could
2092    be tidied up in order to avoid warnings.
2093    
2094    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
2095    first of these names is deprecated in the latest Autoconf in favour of the name
2096    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
2097    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
2098    analogy I have changed it to CFLAGS_FOR_BUILD.
2099    
2100    4. Added -no-undefined to the linking command in the Makefile, because this is
2101    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
2102    linking step for the pcreposix library.
2103    
2104    5. PCRE was failing to diagnose the case of two named groups with the same
2105    name.
2106    
2107    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
2108    literal character that is needed in the subject for a match, and scans along to
2109    ensure that it is present before embarking on the full matching process. This
2110    saves time in cases of nested unlimited repeats that are never going to match.
2111    Problem: the scan can take a lot of time if the subject is very long (e.g.
2112    megabytes), thus penalizing straightforward matches. It is now done only if the
2113    amount of subject to be scanned is less than 1000 bytes.
2114    
2115    7. A lesser problem with the same optimization is that it was recording the
2116    first character of an anchored pattern as "needed", thus provoking a search
2117    right along the subject, even when the first match of the pattern was going to
2118    fail. The "needed" character is now not set for anchored patterns, unless it
2119    follows something in the pattern that is of non-fixed length. Thus, it still
2120    fulfils its original purpose of finding quick non-matches in cases of nested
2121    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
2122    
2123    
2124    Version 4.0 17-Feb-03
2125    ---------------------
2126    
2127    1. If a comment in an extended regex that started immediately after a meta-item
2128    extended to the end of string, PCRE compiled incorrect data. This could lead to
2129    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
2130    
2131    2. Moved to autoconf 2.53 and libtool 1.4.2.
2132    
2133    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
2134    the special perltest8 script is no longer needed - all the tests can be run
2135    from a single perltest script.
2136    
2137    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
2138    by \s. It has now been removed in PCRE. This means it isn't recognized as
2139    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
2140    class [:space:] *does* include VT, thereby creating a mess.
2141    
2142    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
2143    space and tab.
2144    
2145    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
2146    its new features into the main test script, reducing the number of scripts.
2147    
2148    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
2149    were backward compatible, and made the (?i) apply to the whole pattern, as if
2150    /i were given. Now it behaves more logically, and applies the option setting
2151    only to what follows. PCRE has been changed to follow suit. However, if it
2152    finds options settings right at the start of the pattern, it extracts them into
2153    the global options, as before. Thus, they show up in the info data.
2154    
2155    8. Added support for the \Q...\E escape sequence. Characters in between are
2156    treated as literals. This is slightly different from Perl in that $ and @ are
2157    also handled as literals inside the quotes. In Perl, they will cause variable
2158    interpolation. Note the following examples:
2159    
2160        Pattern            PCRE matches      Perl matches
2161    
2162        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
2163        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
2164        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
2165    
2166    For compatibility with Perl, \Q...\E sequences are recognized inside character
2167    classes as well as outside them.
2168    
2169    9. Re-organized 3 code statements in pcretest to avoid "overflow in
2170    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
2171    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
2172    signed/unsigned warnings.
2173    
2174    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
2175    option for pcretest, so I've replaced it by a simple function that does just
2176    that job.
2177    
2178    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
2179    "pcregrep -".
2180    
2181    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
2182    Java package. This provides some syntactic sugar for simple cases of what my
2183    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
2184    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
2185    item, you can use this simplified notation. Note that only makes sense with
2186    greedy quantifiers. Consequently, the use of the possessive quantifier forces
2187    greediness, whatever the setting of the PCRE_UNGREEDY option.
2188    
2189    13. A change of greediness default within a pattern was not taking effect at
2190    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
2191    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
2192    was abstracted outside.
2193    
2194    14. PCRE now supports the \G assertion. It is true when the current matching
2195    position is at the start point of the match. This differs from \A when the
2196    starting offset is non-zero. Used with the /g option of pcretest (or similar
2197    code), it works in the same way as it does for Perl's /g option. If all
2198    alternatives of a regex begin with \G, the expression is anchored to the start
2199    match position, and the "anchored" flag is set in the compiled expression.
2200    
2201    15. Some bugs concerning the handling of certain option changes within patterns
2202    have been fixed. These applied to options other than (?ims). For example,
2203    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
2204    been the other way round. Some of this was related to change 7 above.
2205    
2206    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
2207    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
2208    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
2209    POSIX classes only within a class (e.g. /[[:alpha:]]/).
2210    
2211    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
2212    mode. Unlike ".", it always matches newline, whatever the setting of
2213    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
2214    assertions. Perl allows it, but it doesn't (in general) work because it can't
2215    calculate the length of the lookbehind. At least, that's the case for Perl
2216    5.8.0 - I've been told they are going to document that it doesn't work in
2217    future.
2218    
2219    18. Added an error diagnosis for escapes that PCRE does not support: these are
2220    \L, \l, \N, \P, \p, \U, \u, and \X.
2221    
2222    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
2223    reading past the end of the pattern in cases such as /[abcd/.
2224    
2225    20. PCRE was getting more memory than necessary for patterns with classes that
2226    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
2227    
2228    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
2229    compiling PCRE for use with Virtual Pascal.
2230    
2231    22. Small fix to the Makefile to make it work properly if the build is done
2232    outside the source tree.
2233    
2234    23. Added a new extension: a condition to go with recursion. If a conditional
2235    subpattern starts with (?(R) the "true" branch is used if recursion has
2236    happened, whereas the "false" branch is used only at the top level.
2237    
2238    24. When there was a very long string of literal characters (over 255 bytes
2239    without UTF support, over 250 bytes with UTF support), the computation of how
2240    much memory was required could be incorrect, leading to segfaults or other
2241    strange effects.
2242    
2243    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
2244    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
2245    there was a subsequent back reference to those brackets. This meant that, for
2246    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
2247    possible to check for precisely this case. All we can do is abandon the
2248    optimization if .* occurs inside capturing brackets when there are any back
2249    references whatsoever. (See below for a better fix that came later.)
2250    
2251    26. The handling of the optimization for finding the first character of a
2252    non-anchored pattern, and for finding a character that is required later in the
2253    match were failing in some cases. This didn't break the matching; it just
2254    failed to optimize when it could. The way this is done has been re-implemented.
2255    
2256    27. Fixed typo in error message for invalid (?R item (it said "(?p").
2257    
2258    28. Added a new feature that provides some of the functionality that Perl
2259    provides with (?{...}). The facility is termed a "callout". The way it is done
2260    in PCRE is for the caller to provide an optional function, by setting
2261    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
2262    global variable. By default it is unset, which disables all calling out. To get
2263    the function called, the regex must include (?C) at appropriate points. This
2264    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
2265    This provides a means of identifying different callout points. When PCRE
2266    reaches such a point in the regex, if pcre_callout has been set, the external
2267    function is called. It is provided with data in a structure called
2268    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
2269    matching continues; if it returns a non-zero value, the match at the current
2270    point fails. However, backtracking will occur if possible. [This was changed
2271    later and other features added - see item 49 below.]
2272    
2273    29. pcretest is upgraded to test the callout functionality. It provides a
2274    callout function that displays information. By default, it shows the start of
2275    the match and the current position in the text. There are some new data escapes
2276    to vary what happens:
2277    
2278        \C+         in addition, show current contents of captured substrings
2279        \C-         do not supply a callout function
2280        \C!n        return 1 when callout number n is reached
2281        \C!n!m      return 1 when callout number n is reached for the mth time
2282    
2283    30. If pcregrep was called with the -l option and just a single file name, it
2284    output "<stdin>" if a match was found, instead of the file name.
2285    
2286    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
2287    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
2288    pcre_exec(). This saves a malloc/free per call. The default value of
2289    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
2290    when configuring.
2291    
2292    32. The default maximum size of a compiled pattern is 64K. There have been a
2293    few cases of people hitting this limit. The code now uses macros to handle the
2294    storing of links as offsets within the compiled pattern. It defaults to 2-byte
2295    links, but this can be changed to 3 or 4 bytes by --with-link-size when
2296    configuring. Tests 2 and 5 work only with 2-byte links because they output
2297    debugging information about compiled patterns.
2298    
2299    33. Internal code re-arrangements:
2300    
2301    (a) Moved the debugging function for printing out a compiled regex into
2302        its own source file (printint.c) and used #include to pull it into
2303        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
2304        separate copies.
2305    
2306    (b) Defined the list of op-code names for debugging as a macro in
2307        internal.h so that it is next to the definition of the opcodes.
2308    
2309    (c) Defined a table of op-code lengths for simpler skipping along compiled
2310        code. This is again a macro in internal.h so that it is next to the
2311        definition of the opcodes.
2312    
2313    34. Added support for recursive calls to individual subpatterns, along the
2314    lines of Robin Houston's patch (but implemented somewhat differently).
2315    
2316    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
2317    allow it to read and process whole directories in Win32. This code was
2318    contributed by Lionel Fourquaux; it has not been tested by me.
2319    
2320    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
2321    used to name a group. Names consist of alphanumerics and underscores, and must
2322    be unique. Back references use the syntax (?P=name) and recursive calls use
2323    (?P>name) which is a PCRE extension to the Python extension. Groups still have
2324    numbers. The function pcre_fullinfo() can be used after compilation to extract
2325    a name/number map. There are three relevant calls:
2326    
2327      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
2328      PCRE_INFO_NAMECOUNT            yields the number of entries
2329      PCRE_INFO_NAMETABLE            yields a pointer to the map.
2330    
2331    The map is a vector of fixed-size entries. The size of each entry depends on
2332    the length of the longest name used. The first two bytes of each entry are the
2333    group number, most significant byte first. There follows the corresponding
2334    name, zero terminated. The names are in alphabetical order.
2335    
2336    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
2337    case instead of 255. Making it the same both with and without UTF-8 support
2338    means that the same test output works with both.
2339    
2340    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
2341    calling malloc() with a zero argument.
2342    
2343    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
2344    optimization. I've improved things by keeping a bitmap of backreferences with
2345    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
2346    fact referenced, the optimization can be applied. It is unlikely that a
2347    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
2348    the match to follow \n) will appear inside brackets with a number greater than
2349    31, but if it does, any back reference > 31 suppresses the optimization.
2350    
2351    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
2352    of disabling numbered capturing parentheses. Any opening parenthesis that is
2353    not followed by ? behaves as if it were followed by ?: but named parentheses
2354    can still be used for capturing (and they will acquire numbers in the usual
2355    way).
2356    
2357    41. Redesigned the return codes from the match() function into yes/no/error so
2358    that errors can be passed back from deep inside the nested calls. A malloc
2359    failure while inside a recursive subpattern call now causes the
2360    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
2361    
2362    42. It is now possible to set a limit on the number of times the match()
2363    function is called in a call to pcre_exec(). This facility makes it possible to
2364    limit the amount of recursion and backtracking, though not in a directly
2365    obvious way, because the match() function is used in a number of different
2366    circumstances. The count starts from zero for each position in the subject
2367    string (for non-anchored patterns). The default limit is, for compatibility, a
2368    large number, namely 10 000 000. You can change this in two ways:
2369    
2370    (a) When configuring PCRE before making, you can use --with-match-limit=n
2371        to set a default value for the compiled library.
2372    
2373    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
2374        a different value is set. See 45 below.
2375    
2376    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
2377    
2378    43. Added a new function pcre_config(int, void *) to enable run-time extraction
2379    of things that can be changed at compile time. The first argument specifies
2380    what is wanted and the second points to where the information is to be placed.
2381    The current list of available information is:
2382    
2383      PCRE_CONFIG_UTF8
2384    
2385    The output is an integer that is set to one if UTF-8 support is available;
2386    otherwise it is set to zero.
2387    
2388      PCRE_CONFIG_NEWLINE
2389    
2390    The output is an integer that it set to the value of the code that is used for
2391    newline. It is either LF (10) or CR (13).
2392    
2393      PCRE_CONFIG_LINK_SIZE
2394    
2395    The output is an integer that contains the number of bytes used for internal
2396    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
2397    
2398      PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
2399    
2400    The output is an integer that contains the threshold above which the POSIX
2401    interface uses malloc() for output vectors. See item 31 above.
2402    
2403      PCRE_CONFIG_MATCH_LIMIT
2404    
2405    The output is an unsigned integer that contains the default limit of the number
2406    of match() calls in a pcre_exec() execution. See 42 above.
2407    
2408    44. pcretest has been upgraded by the addition of the -C option. This causes it
2409    to extract all the available output from the new pcre_config() function, and to
2410    output it. The program then exits immediately.
2411    
2412    45. A need has arisen to pass over additional data with calls to pcre_exec() in
2413    order to support additional features. One way would have been to define
2414    pcre_exec2() (for example) with extra arguments, but this would not have been
2415    extensible, and would also have required all calls to the original function to
2416    be mapped to the new one. Instead, I have chosen to extend the mechanism that
2417    is used for passing in "extra" data from pcre_study().
2418    
2419    The pcre_extra structure is now exposed and defined in pcre.h. It currently
2420    contains the following fields:
2421    
2422      flags         a bitmap indicating which of the following fields are set
2423      study_data    opaque data from pcre_study()
2424      match_limit   a way of specifying a limit on match() calls for a specific
2425                      call to pcre_exec()
2426      callout_data  data for callouts (see 49 below)
2427    
2428    The flag bits are also defined in pcre.h, and are
2429    
2430      PCRE_EXTRA_STUDY_DATA
2431      PCRE_EXTRA_MATCH_LIMIT
2432      PCRE_EXTRA_CALLOUT_DATA
2433    
2434    The pcre_study() function now returns one of these new pcre_extra blocks, with
2435    the actual study data pointed to by the study_data field, and the
2436    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
2437    before. That is, this change is entirely upwards-compatible and requires no
2438    change to existing code.
2439    
2440    If you want to pass in additional data to pcre_exec(), you can either place it
2441    in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
2442    block.
2443    
2444    46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
2445    data string contains the escape sequence \M, pcretest calls pcre_exec() several
2446    times with different match limits, until it finds the minimum value needed for
2447    pcre_exec() to complete. The value is then output. This can be instructive; for
2448    most simple matches the number is quite small, but for pathological cases it
2449    gets very large very quickly.
2450    
2451    47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
2452    returns the size of the data block pointed to by the study_data field in a
2453    pcre_extra block, that is, the value that was passed as the argument to
2454    pcre_malloc() when PCRE was getting memory in which to place the information
2455    created by pcre_study(). The fourth argument should point to a size_t variable.
2456    pcretest has been extended so that this information is shown after a successful
2457    pcre_study() call when information about the compiled regex is being displayed.
2458    
2459    48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
2460    because what follows is always an absolute path. (Later: it turns out that this
2461    is more than cosmetic for MinGW, because it doesn't like empty path
2462    components.)
2463    
2464    49. Some changes have been made to the callout feature (see 28 above):
2465    
2466    (i)  A callout function now has three choices for what it returns:
2467    
2468           0  =>  success, carry on matching
2469         > 0  =>  failure at this point, but backtrack if possible
2470         < 0  =>  serious error, return this value from pcre_exec()
2471    
2472         Negative values should normally be chosen from the set of PCRE_ERROR_xxx
2473         values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
2474         "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
2475         use by callout functions. It will never be used by PCRE itself.
2476    
2477    (ii) The pcre_extra structure (see 45 above) has a void * field called
2478         callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
2479         pcre_callout_block structure has a field of the same name. The contents of
2480         the field passed in the pcre_extra structure are passed to the callout
2481         function in the corresponding field in the callout block. This makes it
2482         easier to use the same callout-containing regex from multiple threads. For
2483         testing, the pcretest program has a new data escape
2484    
2485           \C*n        pass the number n (may be negative) as callout_data
2486    
2487         If the callout function in pcretest receives a non-zero value as
2488         callout_data, it returns that value.
2489    
2490    50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
2491    there were some redundant $(CFLAGS) in commands that are now specified as
2492    $(LINK), which already includes $(CFLAGS).
2493    
2494    51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
2495    has been compiled with UTF-8 support *and* pcre_compile() has been compiled
2496    with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
2497    one-byte characters throughout. Note that case-insensitive matching applies
2498    only to characters whose values are less than 256. PCRE doesn't support the
2499    notion of cases for higher-valued characters.
2500    
2501    (i)   A character class whose characters are all within 0-255 is handled as
2502          a bit map, and the map is inverted for negative classes. Previously, a
2503          character > 255 always failed to match such a class; however it should
2504          match if the class was a negative one (e.g. [^ab]). This has been fixed.
2505    
2506    (ii)  A negated character class with a single character < 255 is coded as
2507          "not this character" (OP_NOT). This wasn't working properly when the test
2508          character was multibyte, either singly or repeated.
2509    
2510    (iii) Repeats of multibyte characters are now handled correctly in UTF-8
2511          mode, for example: \x{100}{2,3}.
2512    
2513    (iv)  The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
2514          singly or repeated) now correctly test multibyte characters. However,
2515          PCRE doesn't recognize any characters with values greater than 255 as
2516          digits, spaces, or word characters. Such characters always match \D, \S,
2517          and \W, and never match \d, \s, or \w.
2518    
2519    (v)   Classes may now contain characters and character ranges with values
2520          greater than 255. For example: [ab\x{100}-\x{400}].
2521    
2522    (vi)  pcregrep now has a --utf-8 option (synonym -u) which makes it call
2523          PCRE in UTF-8 mode.
2524    
2525    52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
2526    PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
2527    retained for backwards compatibility. (Note that LASTLITERAL is also a byte
2528    value.)
2529    
2530    53. The single man page has become too large. I have therefore split it up into
2531    a number of separate man pages. These also give rise to individual HTML pages;
2532    these are now put in a separate directory, and there is an index.html page that
2533    lists them all. Some hyperlinking between the pages has been installed.
2534    
2535    54. Added convenience functions for handling named capturing parentheses.
2536    
2537    55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2538    aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2539    true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2540    are faulted.
2541    
2542    56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2543    calling configure. These values are used when compiling the dftables.c program
2544    which is run to generate the source of the default character tables. They
2545    default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2546    you will need to set these values.
2547    
2548    57. Updated the building process for Windows DLL, as provided by Fred Cox.
2549    
2550    
2551    Version 3.9 02-Jan-02
2552    ---------------------
2553    
2554    1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2555    
2556    2. If --disable-static was given, the building process failed when trying to
2557    build pcretest and pcregrep. (For some reason it was using libtool to compile
2558    them, which is not right, as they aren't part of the library.)
2559    
2560    
2561    Version 3.8 18-Dec-01
2562    ---------------------
2563    
2564    1. The experimental UTF-8 code was completely screwed up. It was packing the
2565    bytes in the wrong order. How dumb can you get?
2566    
2567    
2568    Version 3.7 29-Oct-01
2569    ---------------------
2570    
2571    1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2572    This caused pcretest, when used on the test data, to segfault. Unfortunately,
2573    this didn't happen under Solaris 8, where I normally test things.
2574    
2575    2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2576    doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2577    isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2578    this fix an hour or so after the initial 3.7 release.)
2579    
2580    
2581    Version 3.6 23-Oct-01
2582    ---------------------
2583    
2584    1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2585    offsets passed as NULL with zero offset count.
2586    
2587    2. The config.guess and config.sub files had not been updated when I moved to
2588    the latest autoconf.
2589    
2590    
2591    Version 3.5 15-Aug-01
2592    ---------------------
2593    
2594    1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2595    had been forgotten.
2596    
2597    2. By using declared but undefined structures, we can avoid using "void"
2598    definitions in pcre.h while keeping the internal definitions of the structures
2599    private.
2600    
2601    3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2602    user point of view, this means that both static and shared libraries are built
2603    by default, but this can be individually controlled. More of the work of
2604    handling this static/shared cases is now inside libtool instead of PCRE's make
2605    file.
2606    
2607    4. The pcretest utility is now installed along with pcregrep because it is
2608    useful for users (to test regexs) and by doing this, it automatically gets
2609    relinked by libtool. The documentation has been turned into a man page, so
2610    there are now .1, .txt, and .html versions in /doc.
2611    
2612    5. Upgrades to pcregrep:
2613       (i)   Added long-form option names like gnu grep.
2614       (ii)  Added --help to list all options with an explanatory phrase.
2615       (iii) Added -r, --recursive to recurse into sub-directories.
2616       (iv)  Added -f, --file to read patterns from a file.
2617    
2618    6. pcre_exec() was referring to its "code" argument before testing that
2619    argument for NULL (and giving an error if it was NULL).
2620    
2621    7. Upgraded Makefile.in to allow for compiling in a different directory from
2622    the source directory.
2623    
2624    8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2625    options bits, the pointer it was passed was to an int instead of to an unsigned
2626    long int. This mattered only on 64-bit systems.
2627    
2628    9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2629    generated) instead of pcre.in, which it its source. Also made the same change
2630    in several of the .c files.
2631    
2632    10. A new release of gcc defines printf() as a macro, which broke pcretest
2633    because it had an ifdef in the middle of a string argument for printf(). Fixed
2634    by using separate calls to printf().
2635    
2636    11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2637    script, to force use of CR or LF instead of \n in the source. On non-Unix
2638    systems, the value can be set in config.h.
2639    
2640    12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2641    absolute limit. Changed the text of the error message to make this clear, and
2642    likewise updated the man page.
2643    
2644    13. The limit of 99 on the number of capturing subpatterns has been removed.
2645    The new limit is 65535, which I hope will not be a "real" limit.
2646    
2647    
2648    Version 3.4 22-Aug-00
2649    ---------------------
2650    
2651    1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2652    
2653    2. Diagnose condition (?(0) as an error instead of crashing on matching.
2654    
2655    
2656    Version 3.3 01-Aug-00
2657    ---------------------
2658    
2659    1. If an octal character was given, but the value was greater than \377, it
2660    was not getting masked to the least significant bits, as documented. This could
2661    lead to crashes in some systems.
2662    
2663    2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
2664    the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
2665    
2666    3. Added the functions pcre_free_substring() and pcre_free_substring_list().
2667    These just pass their arguments on to (pcre_free)(), but they are provided
2668    because some uses of PCRE bind it to non-C systems that can call its functions,
2669    but cannot call free() or pcre_free() directly.
2670    
2671    4. Add "make test" as a synonym for "make check". Corrected some comments in
2672    the Makefile.
2673    
2674    5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
2675    Makefile.
2676    
2677    6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
2678    command called pgrep for grepping around the active processes.
2679    
2680    7. Added the beginnings of support for UTF-8 character strings.
2681    
2682    8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
2683    RANLIB to ./ltconfig so that they are used by libtool. I think these are all
2684    the relevant ones. (AR is not passed because ./ltconfig does its own figuring
2685    out for the ar command.)
2686    
2687    
2688    Version 3.2 12-May-00
2689    ---------------------
2690    
2691    This is purely a bug fixing release.
2692    
2693    1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
2694    of ZA. This was just one example of several cases that could provoke this bug,
2695    which was introduced by change 9 of version 2.00. The code for breaking
2696    infinite loops after an iteration that matches an empty string was't working
2697    correctly.
2698    
2699    2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
2700    when matched against abbab (for example). After matching an empty string, it
2701    wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
2702    caused it to match further down the string than it should.
2703    
2704    3. The code contained an inclusion of sys/types.h. It isn't clear why this
2705    was there because it doesn't seem to be needed, and it causes trouble on some
2706    systems, as it is not a Standard C header. It has been removed.
2707    
2708    4. Made 4 silly changes to the source to avoid stupid compiler warnings that
2709    were reported on the Macintosh. The changes were from
2710    
2711      while ((c = *(++ptr)) != 0 && c != '\n');
2712    to
2713      while ((c = *(++ptr)) != 0 && c != '\n') ;
2714    
2715    Totally extraordinary, but if that's what it takes...
2716    
2717    5. PCRE is being used in one environment where neither memmove() nor bcopy() is
2718    available. Added HAVE_BCOPY and an autoconf test for it; if neither
2719    HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
2720    assumes the way PCRE uses memmove() (always moving upwards).
2721    
2722    6. PCRE is being used in one environment where strchr() is not available. There
2723    was only one use in pcre.c, and writing it out to avoid strchr() probably gives
2724    faster code anyway.
2725    
2726    
2727    Version 3.1 09-Feb-00
2728    ---------------------
2729    
2730    The only change in this release is the fixing of some bugs in Makefile.in for
2731    the "install" target:
2732    
2733    (1) It was failing to install pcreposix.h.
2734    
2735    (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
2736    
2737    
2738    Version 3.0 01-Feb-00
2739    ---------------------
2740    
2741    1. Add support for the /+ modifier to perltest (to output $` like it does in
2742    pcretest).
2743    
2744    2. Add support for the /g modifier to perltest.
2745    
2746    3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
2747    matches null strings.
2748    
2749    4. Fix perltest so that it doesn't do unwanted things when fed an empty
2750    pattern. Perl treats empty patterns specially - it reuses the most recent
2751    pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
2752    effect.
2753    
2754    5. The POSIX interface was broken in that it was just handing over the POSIX
2755    captured string vector to pcre_exec(), but (since release 2.00) PCRE has
2756    required a bigger vector, with some working space on the end. This means that
2757    the POSIX wrapper now has to get and free some memory, and copy the results.
2758    
2759    6. Added some simple autoconf support, placing the test data and the
2760    documentation in separate directories, re-organizing some of the
2761    information files, and making it build pcre-config (a GNU standard). Also added
2762    libtool support for building PCRE as a shared library, which is now the
2763    default.
2764    
2765    7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
2766    09 are not valid octal constants. Single digits will be used for minor values
2767    less than 10.
2768    
2769    8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
2770    existing programs that set these in the POSIX interface can use PCRE without
2771    modification.
2772    
2773    9. Added a new function, pcre_fullinfo() with an extensible interface. It can
2774    return all that pcre_info() returns, plus additional data. The pcre_info()
2775    function is retained for compatibility, but is considered to be obsolete.
2776    
2777    10. Added experimental recursion feature (?R) to handle one common case that
2778    Perl 5.6 will be able to do with (?p{...}).
2779    
2780    11. Added support for POSIX character classes like [:alpha:], which Perl is
2781    adopting.
2782    
2783    
2784    Version 2.08 31-Aug-99
2785    ----------------------
2786    
2787    1. When startoffset was not zero and the pattern began with ".*", PCRE was not
2788    trying to match at the startoffset position, but instead was moving forward to
2789    the next newline as if a previous match had failed.
2790    
2791    2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
2792    and could get into a loop if a null string was matched other than at the start
2793    of the subject.
2794    
2795    3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
2796    be distinguished at compile time, and for completeness also added PCRE_DATE.
2797    
2798    5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
2799    in GnuWin32 environments.
2800    
2801    
2802    Version 2.07 29-Jul-99
2803    ----------------------
2804    
2805    1. The documentation is now supplied in plain text form and HTML as well as in
2806    the form of man page sources.
2807    
2808    2. C++ compilers don't like assigning (void *) values to other pointer types.
2809    In particular this affects malloc(). Although there is no problem in Standard
2810    C, I've put in casts to keep C++ compilers happy.
2811    
2812    3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
2813    should be (const char *).
2814    
2815    4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
2816    be useful for non-Unix systems who don't want to bother with the POSIX stuff.
2817    However, I haven't made this a standard facility. The documentation doesn't
2818    mention it, and the Makefile doesn't support it.
2819    
2820    5. The Makefile now contains an "install" target, with editable destinations at
2821    the top of the file. The pcretest program is not installed.
2822    
2823    6. pgrep -V now gives the PCRE version number and date.
2824    
2825    7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
2826    causing the entire string to be ignored, instead of just the last character.
2827    
2828    8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
2829    non-matching string, it can take a very, very long time, even for strings of
2830    quite modest length, because of the nested recursion. PCRE now does better in
2831    some of these cases. It does this by remembering the last required literal
2832    character in the pattern, and pre-searching the subject to ensure it is present
2833    before running the real match. In other words, it applies a heuristic to detect
2834    some types of certain failure quickly, and in the above example, if presented
2835    with a string that has no trailing " it gives "no match" very quickly.
2836    
2837    9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
2838    other alternatives are tried instead.
2839    
2840    
2841    Version 2.06 09-Jun-99
2842    ----------------------
2843    
2844    1. Change pcretest's output for amount of store used to show just the code
2845    space, because the remainder (the data block) varies in size between 32-bit and
2846    64-bit systems.
2847    
2848    2. Added an extra argument to pcre_exec() to supply an offset in the subject to
2849    start matching at. This allows lookbehinds to work when searching for multiple
2850    occurrences in a string.
2851    
2852    3. Added additional options to pcretest for testing multiple occurrences:
2853    
2854       /+   outputs the rest of the string that follows a match
2855       /g   loops for multiple occurrences, using the new startoffset argument
2856       /G   loops for multiple occurrences by passing an incremented pointer
2857    
2858    4. PCRE wasn't doing the "first character" optimization for patterns starting
2859    with \b or \B, though it was doing it for other lookbehind assertions. That is,
2860    it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
2861    the letter 'x'. On long subject strings, this gives a significant speed-up.
2862    
2863    
2864    Version 2.05 21-Apr-99
2865    ----------------------
2866    
2867    1. Changed the type of magic_number from int to long int so that it works
2868    properly on 16-bit systems.
2869    
2870    2. Fixed a bug which caused patterns starting with .* not to work correctly
2871    when the subject string contained newline characters. PCRE was assuming
2872    anchoring for such patterns in all cases, which is not correct because .* will
2873    not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
2874    DOTALL is set at top level; otherwise it knows that patterns starting with .*
2875    must be retried after every newline in the subject.
2876    
2877    
2878    Version 2.04 18-Feb-99
2879    ----------------------
2880    
2881    1. For parenthesized subpatterns with repeats whose minimum was zero, the
2882    computation of the store needed to hold the pattern was incorrect (too large).
2883    If such patterns were nested a few deep, this could multiply and become a real
2884    problem.
2885    
2886    2. Added /M option to pcretest to show the memory requirement of a specific
2887    pattern. Made -m a synonym of -s (which does this globally) for compatibility.
2888    
2889    3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
2890    compiled in such a way that the backtracking after subsequent failure was
2891    pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
2892    ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
2893    
2894    
2895    Version 2.03 02-Feb-99
2896    ----------------------
2897    
2898    1. Fixed typo and small mistake in man page.
2899    
2900    2. Added 4th condition (GPL supersedes if conflict) and created separate
2901    LICENCE file containing the conditions.
2902    
2903    3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
2904    Perl, that is the internal \ allows the delimiter to be included in the
2905    pattern. Locked out the use of \ as a delimiter. If \ immediately follows
2906    the final delimiter, add \ to the end of the pattern (to test the error).
2907    
2908    4. Added the convenience functions for extracting substrings after a successful
2909    match. Updated pcretest to make it able to test these functions.
2910    
2911    
2912    Version 2.02 14-Jan-99
2913    ----------------------
2914    
2915    1. Initialized the working variables associated with each extraction so that
2916    their saving and restoring doesn't refer to uninitialized store.
2917    
2918    2. Put dummy code into study.c in order to trick the optimizer of the IBM C
2919    compiler for OS/2 into generating correct code. Apparently IBM isn't going to
2920    fix the problem.
2921    
2922    3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
2923    calls, and wasn't printing the correct value for compiling calls. Increased the
2924    default value of LOOPREPEAT, and the number of significant figures in the
2925    times.
2926    
2927    4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
2928    
2929    5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
2930    a building problem on Windows NT with a FAT file system.
2931    
2932    
2933    Version 2.01 21-Oct-98
2934    ----------------------
2935    
2936    1. Changed the API for pcre_compile() to allow for the provision of a pointer
2937    to character tables built by pcre_maketables() in the current locale. If NULL
2938    is passed, the default tables are used.
2939    
2940    
2941    Version 2.00 24-Sep-98
2942    ----------------------
2943    
2944    1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
2945    it any more.
2946    
2947    2. Allow quantification of (?>) groups, and make it work correctly.
2948    
2949    3. The first character computation wasn't working for (?>) groups.
2950    
2951    4. Correct the implementation of \Z (it is permitted to match on the \n at the
2952    end of the subject) and add 5.005's \z, which really does match only at the
2953    very end of the subject.
2954    
2955    5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
2956    
2957    6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
2958    DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
2959    localized options. All options to pcre_study() were also removed.
2960    
2961    7. Add other new features from 5.005:
2962    
2963       $(?<=           positive lookbehind
2964       $(?<!           negative lookbehind
2965       (?imsx-imsx)    added the unsetting capability
2966                       such a setting is global if at outer level; local otherwise
2967       (?imsx-imsx:)   non-capturing groups with option setting
2968       (?(cond)re|re)  conditional pattern matching
2969    
2970       A backreference to itself in a repeated group matches the previous
2971       captured string.
2972    
2973    8. General tidying up of studying (both automatic and via "study")
2974    consequential on the addition of new assertions.
2975    
2976    9. As in 5.005, unlimited repeated groups that could match an empty substring
2977    are no longer faulted at compile time. Instead, the loop is forcibly broken at
2978    runtime if any iteration does actually match an empty substring.
2979    
2980    10. Include the RunTest script in the distribution.
2981    
2982    11. Added tests from the Perl 5.005_02 distribution. This showed up a few
2983    discrepancies, some of which were old and were also with respect to 5.004. They
2984    have now been fixed.
2985    
2986    
2987    Version 1.09 28-Apr-98
2988    ----------------------
2989    
2990    1. A negated single character class followed by a quantifier with a minimum
2991    value of one (e.g.  [^x]{1,6}  ) was not compiled correctly. This could lead to
2992    program crashes, or just wrong answers. This did not apply to negated classes
2993    containing more than one character, or to minima other than one.
2994    
2995    
2996    Version 1.08 27-Mar-98
2997    ----------------------
2998    
2999    1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
3000    
3001    2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
3002    latter must appear before anything that relies on it in the pattern.
3003    
3004    
3005    Version 1.07 16-Feb-98
3006    ----------------------
3007    
3008    1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
3009    repeat of a potentially empty string).
3010    
3011    
3012    Version 1.06 23-Jan-98
3013    ----------------------
3014    
3015    1. Added Markus Oberhumer's little patches for C++.
3016    
3017    2. Literal strings longer than 255 characters were broken.
3018    
3019    
3020    Version 1.05 23-Dec-97
3021    ----------------------
3022    
3023    1. Negated character classes containing more than one character were failing if
3024    PCRE_CASELESS was set at run time.
3025    
3026    
3027    Version 1.04 19-Dec-97
3028    ----------------------
3029    
3030    1. Corrected the man page, where some "const" qualifiers had been omitted.
3031    
3032    2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
3033    input syntax.
3034    
3035    3. Fixed memory leak which occurred when a regex with back references was
3036    matched with an offsets vector that wasn't big enough. The temporary memory
3037    that is used in this case wasn't being freed if the match failed.
3038    
3039    4. Tidied pcretest to ensure it frees memory that it gets.
3040    
3041    5. Temporary memory was being obtained in the case where the passed offsets
3042    vector was exactly big enough.
3043    
3044    6. Corrected definition of offsetof() from change 5 below.
3045    
3046    7. I had screwed up change 6 below and broken the rules for the use of
3047    setjmp(). Now fixed.
3048    
3049    
3050    Version 1.03 18-Dec-97
3051    ----------------------
3052    
3053    1. A erroneous regex with a missing opening parenthesis was correctly
3054    diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
3055    on some systems.
3056    
3057    2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
3058    it was reported that one broken compiler failed on the former because "code" is
3059    also an independent variable.
3060    
3061    3. The erroneous regex a[]b caused an array overrun reference.
3062    
3063    4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
3064    fail on data ending with that character. (It was going on too far, and checking
3065    the next character, typically a binary zero.) This was specific to the
3066    optimized code for single-character negative classes.
3067    
3068    5. Added a contributed patch from the TIN world which does the following:
3069    
3070      + Add an undef for memmove, in case the the system defines a macro for it.
3071    
3072      + Add a definition of offsetof(), in case there isn't one. (I don't know
3073        the reason behind this - offsetof() is part of the ANSI standard - but
3074        it does no harm).
3075    
3076      + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
3077        most of the places where whitespace preceded '#'. I have given up and
3078        allowed the remaining 2 cases to be at the margin.
3079    
3080      + Rename some variables in pcre to eliminate shadowing. This seems very
3081        pedantic, but does no harm, of course.
3082    
3083    6. Moved the call to setjmp() into its own function, to get rid of warnings
3084    from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
3085    
3086    7. Constructs such as \d{8,} were compiling into the equivalent of
3087    \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
3088    outcome, but in this particular case used more store than had been allocated,
3089    which caused the bug to be discovered because it threw up an internal error.
3090    
3091    8. The debugging code in both pcre and pcretest for outputting the compiled
3092    form of a regex was going wrong in the case of back references followed by
3093    curly-bracketed repeats.
3094    
3095    
3096    Version 1.02 12-Dec-97
3097    ----------------------
3098    
3099    1. Typos in pcre.3 and comments in the source fixed.
3100    
3101    2. Applied a contributed patch to get rid of places where it used to remove
3102    'const' from variables, and fixed some signed/unsigned and uninitialized
3103    variable warnings.
3104    
3105    3. Added the "runtest" target to Makefile.
3106    
3107    4. Set default compiler flag to -O2 rather than just -O.
3108    
3109    
3110  Version 1.01 19-Nov-97  Version 1.01 19-Nov-97
3111  ----------------------  ----------------------

Legend:
Removed from v.5  
changed lines
  Added in v.343

  ViewVC Help
Powered by ViewVC 1.1.5