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

Diff of /code/trunk/ChangeLog

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

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

Legend:
Removed from v.43  
changed lines
  Added in v.287

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12