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

Diff of /code/trunk/ChangeLog

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

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

Legend:
Removed from v.29  
changed lines
  Added in v.323

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12