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

Diff of /code/trunk/ChangeLog

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

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

Legend:
Removed from v.61  
changed lines
  Added in v.335

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12