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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 222 - (hide annotations) (download)
Fri Aug 17 10:48:51 2007 UTC (6 years, 8 months ago) by ph10
File size: 137413 byte(s)
Fix pcregrep -Mv looping bug.

1 nigel 3 ChangeLog for PCRE
2     ------------------
3    
4 ph10 221 Version 7.3 17-Aug-07
5 ph10 157 ---------------------
6    
7 ph10 189 1. In the rejigging of the build system that eventually resulted in 7.1, the
8     line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
9     brackets there is not right, since it causes compilers to look for an
10     installed pcre.h, not the version that is in the source that is being
11     compiled (which of course may be different). I have changed it back to:
12 ph10 197
13 ph10 189 #include "pcre.h"
14 ph10 197
15     I have a vague recollection that the change was concerned with compiling in
16     different directories, but in the new build system, that is taken care of
17     by the VPATH setting the Makefile.
18    
19 ph10 190 2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
20     when the subject happened to end in the byte 0x85 (e.g. if the last
21     character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
22     characters but of course it shouldn't be taken as a newline when it is part
23     of another character. The bug was that, for an unlimited repeat of . in
24     not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
25     characters when looking for a newline.
26 ph10 197
27     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
28    
29     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
30     in debug output.
31    
32 ph10 201 5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
33     long printing in the pcrecpp unittest when running under MinGW.
34 ph10 189
35 ph10 197 6. ESC_K was left out of the EBCDIC table.
36 ph10 189
37 ph10 197 7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
38     parentheses; I made it 1000, which seemed large enough. Unfortunately, the
39     limit also applies to "virtual nesting" when a pattern is recursive, and in
40     this case 1000 isn't so big. I have been able to remove this limit at the
41     expense of backing off one optimization in certain circumstances. Normally,
42     when pcre_exec() would call its internal match() function recursively and
43     immediately return the result unconditionally, it uses a "tail recursion"
44     feature to save stack. However, when a subpattern that can match an empty
45     string has an unlimited repetition quantifier, it no longer makes this
46     optimization. That gives it a stack frame in which to save the data for
47     checking that an empty string has been matched. Previously this was taken
48     from the 1000-entry workspace that had been reserved. So now there is no
49     explicit limit, but more stack is used.
50 ph10 189
51 ph10 200 8. Applied Daniel's patches to solve problems with the import/export magic
52     syntax that is required for Windows, and which was going wrong for the
53     pcreposix and pcrecpp parts of the library. These were overlooked when this
54     problem was solved for the main library.
55 ph10 197
56 ph10 202 9. There were some crude static tests to avoid integer overflow when computing
57     the size of patterns that contain repeated groups with explicit upper
58     limits. As the maximum quantifier is 65535, the maximum group length was
59     set at 30,000 so that the product of these two numbers did not overflow a
60     32-bit integer. However, it turns out that people want to use groups that
61     are longer than 30,000 bytes (though not repeat them that many times).
62     Change 7.0/17 (the refactoring of the way the pattern size is computed) has
63     made it possible to implement the integer overflow checks in a much more
64     dynamic way, which I have now done. The artificial limitation on group
65     length has been removed - we now have only the limit on the total length of
66     the compiled pattern, which depends on the LINK_SIZE setting.
67 ph10 208
68     10. Fixed a bug in the documentation for get/copy named substring when
69     duplicate names are permitted. If none of the named substrings are set, the
70     functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
71     empty string.
72    
73     11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
74     instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
75     because the ] is interpreted as the first data character and the
76     terminating ] is not found. PCRE has been made compatible with Perl in this
77     regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
78     cause memory overwriting.
79    
80 ph10 206 10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
81     string has been matched (to stop an infinite loop). It was not recognizing
82 ph10 208 a conditional subpattern that could match an empty string if that
83 ph10 206 subpattern was within another subpattern. For example, it looped when
84 ph10 208 trying to match (((?(1)X|))*) but it was OK with ((?(1)X|)*) where the
85 ph10 206 condition was not nested. This bug has been fixed.
86 ph10 208
87 ph10 207 12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
88     past the start of the subject in the presence of bytes with the top bit
89     set, for example "\x8aBCD".
90 ph10 211
91 ph10 210 13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
92     (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
93 ph10 200
94 ph10 211 14. Optimized (?!) to (*FAIL).
95 ph10 202
96 ph10 212 15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
97     This restricts code points to be within the range 0 to 0x10FFFF, excluding
98     the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
99     full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
100     does: it's just the validity check that is more restrictive.
101 ph10 220
102     16. Inserted checks for integer overflows during escape sequence (backslash)
103     processing, and also fixed erroneous offset values for syntax errors during
104     backslash processing.
105    
106 ph10 214 17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
107 ph10 220 for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
108    
109 ph10 215 18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
110     caused an overrun.
111 ph10 220
112     19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
113     something other than just ASCII characters) inside a group that had an
114     unlimited repeat caused a loop at compile time (while checking to see
115     whether the group could match an empty string).
116    
117 ph10 217 20. Debugging a pattern containing \p or \P could cause a crash. For example,
118     [\P{Any}] did so. (Error in the code for printing property names.)
119 ph10 210
120 ph10 220 21. An orphan \E inside a character class could cause a crash.
121    
122     22. A repeated capturing bracket such as (A)? could cause a wild memory
123 ph10 218 reference during compilation.
124 ph10 220
125     23. There are several functions in pcre_compile() that scan along a compiled
126     expression for various reasons (e.g. to see if it's fixed length for look
127 ph10 218 behind). There were bugs in these functions when a repeated \p or \P was
128 ph10 220 present in the pattern. These operators have additional parameters compared
129     with \d, etc, and these were not being taken into account when moving along
130 ph10 218 the compiled data. Specifically:
131 ph10 220
132     (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
133     length.
134    
135     (b) An item such as \pL+ within a repeated group could cause crashes or
136 ph10 218 loops.
137 ph10 220
138 ph10 218 (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
139 ph10 220 "reference to non-existent subpattern" error.
140    
141 ph10 221 (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
142    
143 ph10 219 24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
144 ph10 220 characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
145 ph10 211
146 ph10 222 25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
147 ph10 218
148 ph10 222
149 ph10 189 Version 7.2 19-Jun-07
150     ---------------------
151    
152 ph10 157 1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
153     which is apparently normally available under Windows.
154 ph10 159
155     2. Re-jig the pcregrep tests with different newline settings in an attempt
156     to make them independent of the local environment's newline setting.
157 ph10 157
158 ph10 160 3. Add code to configure.ac to remove -g from the CFLAGS default settings.
159 ph10 165
160 ph10 161 4. Some of the "internals" tests were previously cut out when the link size
161     was not 2, because the output contained actual offsets. The recent new
162 ph10 165 "Z" feature of pcretest means that these can be cut out, making the tests
163     usable with all link sizes.
164    
165 ph10 164 5. Implemented Stan Switzer's goto replacement for longjmp() when not using
166     stack recursion. This gives a massive performance boost under BSD, but just
167 ph10 165 a small improvement under Linux. However, it saves one field in the frame
168 ph10 164 in all cases.
169 ph10 172
170 ph10 166 6. Added more features from the forthcoming Perl 5.10:
171 ph10 172
172 ph10 166 (a) (?-n) (where n is a string of digits) is a relative subroutine or
173     recursion call. It refers to the nth most recently opened parentheses.
174 ph10 172
175 ph10 166 (b) (?+n) is also a relative subroutine call; it refers to the nth next
176 ph10 172 to be opened parentheses.
177    
178     (c) Conditions that refer to capturing parentheses can be specified
179 ph10 167 relatively, for example, (?(-2)... or (?(+3)...
180 ph10 172
181 ph10 168 (d) \K resets the start of the current match so that everything before
182 ph10 172 is not part of it.
183    
184 ph10 171 (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
185 ph10 172
186 ph10 171 (f) \g{name} is another synonym - part of Perl 5.10's unification of
187 ph10 172 reference syntax.
188 ph10 182
189 ph10 175 (g) (?| introduces a group in which the numbering of parentheses in each
190 ph10 182 alternative starts with the same number.
191 ph10 172
192 ph10 182 (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
193    
194 ph10 172 7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
195     PCRE_INFO_JCHANGED.
196    
197     8. A pattern such as (.*(.)?)* caused pcre_exec() to fail by either not
198     terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
199 ph10 170 for detecting groups that can match an empty string.
200 ph10 159
201 ph10 172 9. A pattern with a very large number of alternatives (more than several
202     hundred) was running out of internal workspace during the pre-compile
203     phase, where pcre_compile() figures out how much memory will be needed. A
204     bit of new cunning has reduced the workspace needed for groups with
205     alternatives. The 1000-alternative test pattern now uses 12 bytes of
206     workspace instead of running out of the 4096 that are available.
207 ph10 182
208 ph10 176 10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
209 ph10 172
210 ph10 179 11. Applied patch from Google to remove an optimization that didn't quite work.
211     The report of the bug said:
212 ph10 182
213 ph10 179 pcrecpp::RE("a*").FullMatch("aaa") matches, while
214     pcrecpp::RE("a*?").FullMatch("aaa") does not, and
215 ph10 182 pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
216 ph10 185
217 ph10 184 12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
218 ph10 185 it matched the wrong number of bytes.
219 ph10 172
220 ph10 179
221 ph10 155 Version 7.1 24-Apr-07
222 ph10 98 ---------------------
223    
224 ph10 111 1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
225 ph10 122 that is more "standard", making use of automake and other Autotools. There
226 ph10 99 is some re-arrangement of the files and adjustment of comments consequent
227     on this.
228 ph10 111
229     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
230     for recursive directory scanning broke on some systems because the files
231     are not scanned in any specific order and on different systems the order
232     was different. A call to "sort" has been inserted into RunGrepTest for the
233     approprate test as a short-term fix. In the longer term there may be an
234 ph10 100 alternative.
235 ph10 111
236 ph10 100 3. I had an email from Eric Raymond about problems translating some of PCRE's
237 ph10 111 man pages to HTML (despite the fact that I distribute HTML pages, some
238     people do their own conversions for various reasons). The problems
239     concerned the use of low-level troff macros .br and .in. I have therefore
240     removed all such uses from the man pages (some were redundant, some could
241 ph10 113 be replaced by .nf/.fi pairs). The 132html script that I use to generate
242     HTML has been updated to handle .nf/.fi and to complain if it encounters
243     .br or .in.
244 ph10 111
245 ph10 100 4. Updated comments in configure.ac that get placed in config.h.in and also
246 ph10 123 arranged for config.h to be included in the distribution, with the name
247 ph10 111 config.h.generic, for the benefit of those who have to compile without
248     Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
249    
250     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
251     Weber: (1) pcre_internal.h was missing some function renames; (2) updated
252 ph10 127 makevp.bat for the current PCRE, using the additional files
253 ph10 135 makevp_c.txt, makevp_l.txt, and pcregexp.pas.
254 ph10 111
255     6. A Windows user reported a minor discrepancy with test 2, which turned out
256     to be caused by a trailing space on an input line that had got lost in his
257 ph10 102 copy. The trailing space was an accident, so I've just removed it.
258 ph10 111
259 ph10 104 7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
260 ph10 111 that is needed.
261    
262 ph10 105 8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
263     as "const" (a) because they are and (b) because it helps the PHP
264     maintainers who have recently made a script to detect big data structures
265 ph10 111 in the php code that should be moved to the .rodata section. I remembered
266     to update Builducptable as well, so it won't revert if ucptable.h is ever
267 ph10 105 re-created.
268 ph10 111
269     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
270 ph10 107 pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
271 ph10 111 order to be able to cut out the UTF-8 tables in the latter when UTF-8
272     support is not required. This saves 1.5-2K of code, which is important in
273 ph10 107 some applications.
274 ph10 111
275 ph10 110 Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
276 ph10 111 so as not to refer to the tables, even though these functions will never be
277     called when UTF-8 support is disabled. Otherwise there are problems with a
278     shared library.
279    
280 ph10 118 10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
281    
282     (a) It was defining its arguments as char * instead of void *.
283    
284 ph10 123 (b) It was assuming that all moves were upwards in memory; this was true
285     a long time ago when I wrote it, but is no longer the case.
286    
287 ph10 118 The emulated memove() is provided for those environments that have neither
288 ph10 123 memmove() nor bcopy(). I didn't think anyone used it these days, but that
289 ph10 118 is clearly not the case, as these two bugs were recently reported.
290 ph10 123
291 ph10 111 11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
292 ph10 123 and Detrail to create the HTML documentation, the .txt form of the man
293     pages, and it removes trailing spaces from listed files. It also creates
294     pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
295     case, it wraps all the #defines with #ifndefs. This script should be run
296 ph10 111 before "make dist".
297 ph10 123
298 ph10 115 12. Fixed two fairly obscure bugs concerned with quantified caseless matching
299     with Unicode property support.
300 ph10 123
301     (a) For a maximizing quantifier, if the two different cases of the
302     character were of different lengths in their UTF-8 codings (there are
303     some cases like this - I found 11), and the matching function had to
304 ph10 115 back up over a mixture of the two cases, it incorrectly assumed they
305     were both the same length.
306 ph10 123
307     (b) When PCRE was configured to use the heap rather than the stack for
308     recursion during matching, it was not correctly preserving the data for
309     the other case of a UTF-8 character when checking ahead for a match
310     while processing a minimizing repeat. If the check also involved
311 ph10 115 matching a wide character, but failed, corruption could cause an
312     erroneous result when trying to check for a repeat of the original
313     character.
314 ph10 123
315 ph10 116 13. Some tidying changes to the testing mechanism:
316 ph10 98
317 ph10 116 (a) The RunTest script now detects the internal link size and whether there
318     is UTF-8 and UCP support by running ./pcretest -C instead of relying on
319 ph10 123 values substituted by "configure". (The RunGrepTest script already did
320     this for UTF-8.) The configure.ac script no longer substitutes the
321     relevant variables.
322    
323 ph10 116 (b) The debugging options /B and /D in pcretest show the compiled bytecode
324     with length and offset values. This means that the output is different
325     for different internal link sizes. Test 2 is skipped for link sizes
326     other than 2 because of this, bypassing the problem. Unfortunately,
327     there was also a test in test 3 (the locale tests) that used /B and
328 ph10 123 failed for link sizes other than 2. Rather than cut the whole test out,
329     I have added a new /Z option to pcretest that replaces the length and
330     offset values with spaces. This is now used to make test 3 independent
331 ph10 122 of link size. (Test 2 will be tidied up later.)
332 ph10 123
333     14. If erroroffset was passed as NULL to pcre_compile, it provoked a
334 ph10 122 segmentation fault instead of returning the appropriate error message.
335 ph10 134
336 ph10 131 15. In multiline mode when the newline sequence was set to "any", the pattern
337 ph10 134 ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
338     This doesn't seem right; it now treats the CRLF combination as the line
339     ending, and so does not match in that case. It's only a pattern such as ^$
340     that would hit this one: something like ^ABC$ would have failed after \r
341     and then tried again after \r\n.
342    
343 ph10 131 16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
344 ph10 134 in an attempt to make files that differ only in their line terminators
345     compare equal. This works on Linux.
346 ph10 142
347 ph10 141 17. Under certain error circumstances pcregrep might try to free random memory
348     as it exited. This is now fixed, thanks to valgrind.
349 ph10 142
350 ph10 141 19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
351 ph10 142 "abc\r\n\r\n", it found an unwanted second match after the second \r. This
352     was because its rules for how to advance for /g after matching an empty
353 ph10 143 string at the end of a line did not allow for this case. They now check for
354     it specially.
355 ph10 150
356     20. pcretest is supposed to handle patterns and data of any length, by
357     extending its buffers when necessary. It was getting this wrong when the
358 ph10 147 buffer for a data line had to be extended.
359 ph10 150
360 ph10 149 21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
361 ph10 150 CRLF as a newline sequence.
362 ph10 152
363 ph10 151 22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
364 ph10 152 out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
365     I have nevertheless tidied it up.
366 ph10 111
367 ph10 152 23. Added some casts to kill warnings from HP-UX ia64 compiler.
368 ph10 123
369 ph10 153 24. Added a man page for pcre-config.
370 ph10 152
371 ph10 153
372 nigel 93 Version 7.0 19-Dec-06
373     ---------------------
374    
375     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
376     moving to gcc 4.1.1.
377    
378     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
379     sys/time.h, which is documented as needed for this function. It doesn't
380     seem to matter on Linux, but it showed up on some releases of OS X.
381    
382     3. It seems that there are systems where bytes whose values are greater than
383     127 match isprint() in the "C" locale. The "C" locale should be the
384     default when a C program starts up. In most systems, only ASCII printing
385     characters match isprint(). This difference caused the output from pcretest
386     to vary, making some of the tests fail. I have changed pcretest so that:
387    
388     (a) When it is outputting text in the compiled version of a pattern, bytes
389     other than 32-126 are always shown as hex escapes.
390    
391     (b) When it is outputting text that is a matched part of a subject string,
392     it does the same, unless a different locale has been set for the match
393     (using the /L modifier). In this case, it uses isprint() to decide.
394    
395     4. Fixed a major bug that caused incorrect computation of the amount of memory
396     required for a compiled pattern when options that changed within the
397     pattern affected the logic of the preliminary scan that determines the
398     length. The relevant options are -x, and -i in UTF-8 mode. The result was
399     that the computed length was too small. The symptoms of this bug were
400     either the PCRE error "internal error: code overflow" from pcre_compile(),
401     or a glibc crash with a message such as "pcretest: free(): invalid next
402     size (fast)". Examples of patterns that provoked this bug (shown in
403     pcretest format) are:
404    
405     /(?-x: )/x
406     /(?x)(?-x: \s*#\s*)/
407     /((?i)[\x{c0}])/8
408     /(?i:[\x{c0}])/8
409    
410     HOWEVER: Change 17 below makes this fix obsolete as the memory computation
411     is now done differently.
412    
413     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
414     wrapper classes; (b) implement a new function in the C++ scanner that is
415     more efficient than the old way of doing things because it avoids levels of
416     recursion in the regex matching; (c) add a paragraph to the documentation
417     for the FullMatch() function.
418    
419     6. The escape sequence \n was being treated as whatever was defined as
420     "newline". Not only was this contrary to the documentation, which states
421     that \n is character 10 (hex 0A), but it also went horribly wrong when
422     "newline" was defined as CRLF. This has been fixed.
423    
424     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
425     was being set to -1 for the "end of line" case (supposedly a value that no
426     character can have). Though this value is never used (the check for end of
427     line is "zero bytes in current character"), it caused compiler complaints.
428     I've changed it to 0xffffffff.
429    
430     8. In pcre_version.c, the version string was being built by a sequence of
431     C macros that, in the event of PCRE_PRERELEASE being defined as an empty
432     string (as it is for production releases) called a macro with an empty
433     argument. The C standard says the result of this is undefined. The gcc
434     compiler treats it as an empty string (which was what was wanted) but it is
435     reported that Visual C gives an error. The source has been hacked around to
436     avoid this problem.
437    
438     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
439     builds of pcretest, and changed the call to _setmode() to use _O_BINARY
440     instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
441     of them did).
442    
443     10. Originally, pcretest opened its input and output without "b"; then I was
444     told that "b" was needed in some environments, so it was added for release
445     5.0 to both the input and output. (It makes no difference on Unix-like
446     systems.) Later I was told that it is wrong for the input on Windows. I've
447     now abstracted the modes into two macros, to make it easier to fiddle with
448     them, and removed "b" from the input mode under Windows.
449    
450     11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
451    
452     12. Added -help and --help to pcretest as an official way of being reminded
453     of the options.
454    
455     13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
456     and pcrecpp.cc because they annoy compilers at high warning levels.
457    
458     14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
459    
460     15. Fixed an occurrence of == in configure.ac that should have been = (shell
461     scripts are not C programs :-) and which was not noticed because it works
462     on Linux.
463    
464     16. pcretest is supposed to handle any length of pattern and data line (as one
465     line or as a continued sequence of lines) by extending its input buffer if
466     necessary. This feature was broken for very long pattern lines, leading to
467     a string of junk being passed to pcre_compile() if the pattern was longer
468     than about 50K.
469    
470     17. I have done a major re-factoring of the way pcre_compile() computes the
471     amount of memory needed for a compiled pattern. Previously, there was code
472     that made a preliminary scan of the pattern in order to do this. That was
473     OK when PCRE was new, but as the facilities have expanded, it has become
474     harder and harder to keep it in step with the real compile phase, and there
475     have been a number of bugs (see for example, 4 above). I have now found a
476     cunning way of running the real compile function in a "fake" mode that
477     enables it to compute how much memory it would need, while actually only
478     ever using a few hundred bytes of working memory and without too many
479     tests of the mode. This should make future maintenance and development
480     easier. A side effect of this work is that the limit of 200 on the nesting
481     depth of parentheses has been removed (though this was never a serious
482     limitation, I suspect). However, there is a downside: pcre_compile() now
483     runs more slowly than before (30% or more, depending on the pattern). I
484     hope this isn't a big issue. There is no effect on runtime performance.
485    
486     18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
487     newline (only possible for the last line of a file) and it was a
488     pattern that set a locale (followed by /Lsomething), pcretest crashed.
489    
490     19. Added additional timing features to pcretest. (1) The -tm option now times
491     matching only, not compiling. (2) Both -t and -tm can be followed, as a
492     separate command line item, by a number that specifies the number of
493     repeats to use when timing. The default is 50000; this gives better
494     precision, but takes uncomfortably long for very large patterns.
495    
496     20. Extended pcre_study() to be more clever in cases where a branch of a
497     subpattern has no definite first character. For example, (a*|b*)[cd] would
498     previously give no result from pcre_study(). Now it recognizes that the
499     first character must be a, b, c, or d.
500    
501     21. There was an incorrect error "recursive call could loop indefinitely" if
502     a subpattern (or the entire pattern) that was being tested for matching an
503     empty string contained only one non-empty item after a nested subpattern.
504     For example, the pattern (?>\x{100}*)\d(?R) provoked this error
505     incorrectly, because the \d was being skipped in the check.
506    
507     22. The pcretest program now has a new pattern option /B and a command line
508     option -b, which is equivalent to adding /B to every pattern. This causes
509     it to show the compiled bytecode, without the additional information that
510     -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
511     is the same as /B/I).
512    
513     23. A new optimization is now able automatically to treat some sequences such
514     as a*b as a*+b. More specifically, if something simple (such as a character
515     or a simple class like \d) has an unlimited quantifier, and is followed by
516     something that cannot possibly match the quantified thing, the quantifier
517     is automatically "possessified".
518    
519     24. A recursive reference to a subpattern whose number was greater than 39
520     went wrong under certain circumstances in UTF-8 mode. This bug could also
521     have affected the operation of pcre_study().
522    
523     25. Realized that a little bit of performance could be had by replacing
524     (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
525    
526     26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
527    
528     27. Possessive quantifiers such as a++ were previously implemented by turning
529     them into atomic groups such as ($>a+). Now they have their own opcodes,
530     which improves performance. This includes the automatically created ones
531     from 23 above.
532    
533     28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
534     lookahead was broken if it was not anchored. PCRE was mistakenly expecting
535     the first matched character to be a colon. This applied both to named and
536     numbered groups.
537    
538     29. The ucpinternal.h header file was missing its idempotency #ifdef.
539    
540     30. I was sent a "project" file called libpcre.a.dev which I understand makes
541     building PCRE on Windows easier, so I have included it in the distribution.
542    
543     31. There is now a check in pcretest against a ridiculously large number being
544     returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
545     loop, the loop is abandoned.
546    
547     32. Forward references to subpatterns in conditions such as (?(2)...) where
548     subpattern 2 is defined later cause pcre_compile() to search forwards in
549     the pattern for the relevant set of parentheses. This search went wrong
550     when there were unescaped parentheses in a character class, parentheses
551     escaped with \Q...\E, or parentheses in a #-comment in /x mode.
552    
553     33. "Subroutine" calls and backreferences were previously restricted to
554     referencing subpatterns earlier in the regex. This restriction has now
555     been removed.
556    
557     34. Added a number of extra features that are going to be in Perl 5.10. On the
558     whole, these are just syntactic alternatives for features that PCRE had
559     previously implemented using the Python syntax or my own invention. The
560     other formats are all retained for compatibility.
561    
562     (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
563     as (?P<name>...). The new forms, as well as being in Perl 5.10, are
564     also .NET compatible.
565    
566     (b) A recursion or subroutine call to a named group can now be defined as
567     (?&name) as well as (?P>name).
568    
569     (c) A backreference to a named group can now be defined as \k<name> or
570     \k'name' as well as (?P=name). The new forms, as well as being in Perl
571     5.10, are also .NET compatible.
572    
573     (d) A conditional reference to a named group can now use the syntax
574     (?(<name>) or (?('name') as well as (?(name).
575    
576     (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
577     groups (named and numbered) that are never evaluated inline, but can be
578     called as "subroutines" from elsewhere. In effect, the DEFINE condition
579     is always false. There may be only one alternative in such a group.
580    
581     (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
582     as the simple (?(R). The condition is true only if the most recent
583     recursion is that of the given number or name. It does not search out
584     through the entire recursion stack.
585    
586     (g) The escape \gN or \g{N} has been added, where N is a positive or
587     negative number, specifying an absolute or relative reference.
588    
589     35. Tidied to get rid of some further signed/unsigned compiler warnings and
590     some "unreachable code" warnings.
591    
592     36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
593     things, this adds five new scripts.
594    
595     37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
596     There were also incompatibilities regarding the handling of \Q..\E inside
597     character classes, for example with patterns like [\Qa\E-\Qz\E] where the
598     hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
599    
600     38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
601     matches an empty string, and forcibly breaks the loop. There were bugs in
602     this code in non-simple cases. For a pattern such as ^(a()*)* matched
603     against aaaa the result was just "a" rather than "aaaa", for example. Two
604     separate and independent bugs (that affected different cases) have been
605     fixed.
606    
607     39. Refactored the code to abolish the use of different opcodes for small
608     capturing bracket numbers. This is a tidy that I avoided doing when I
609     removed the limit on the number of capturing brackets for 3.5 back in 2001.
610     The new approach is not only tidier, it makes it possible to reduce the
611     memory needed to fix the previous bug (38).
612    
613     40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
614     sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
615     processing dot, circumflex, or dollar metacharacters, or #-comments in /x
616     mode.
617    
618     41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
619     report.
620    
621     42. Applied patch, originally from Ari Pollak, modified by Google, to allow
622     copy construction and assignment in the C++ wrapper.
623    
624     43. Updated pcregrep to support "--newline=any". In the process, I fixed a
625     couple of bugs that could have given wrong results in the "--newline=crlf"
626     case.
627    
628     44. Added a number of casts and did some reorganization of signed/unsigned int
629     variables following suggestions from Dair Grant. Also renamed the variable
630     "this" as "item" because it is a C++ keyword.
631    
632     45. Arranged for dftables to add
633    
634     #include "pcre_internal.h"
635    
636     to pcre_chartables.c because without it, gcc 4.x may remove the array
637     definition from the final binary if PCRE is built into a static library and
638     dead code stripping is activated.
639    
640     46. For an unanchored pattern, if a match attempt fails at the start of a
641     newline sequence, and the newline setting is CRLF or ANY, and the next two
642     characters are CRLF, advance by two characters instead of one.
643    
644    
645 nigel 91 Version 6.7 04-Jul-06
646     ---------------------
647    
648     1. In order to handle tests when input lines are enormously long, pcretest has
649     been re-factored so that it automatically extends its buffers when
650     necessary. The code is crude, but this _is_ just a test program. The
651     default size has been increased from 32K to 50K.
652    
653     2. The code in pcre_study() was using the value of the re argument before
654     testing it for NULL. (Of course, in any sensible call of the function, it
655     won't be NULL.)
656    
657     3. The memmove() emulation function in pcre_internal.h, which is used on
658     systems that lack both memmove() and bcopy() - that is, hardly ever -
659     was missing a "static" storage class specifier.
660    
661     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
662     containing an extended class (one that cannot be represented by a bitmap
663     because it contains high-valued characters or Unicode property items, e.g.
664     [\pZ]). Almost always one would set UTF-8 mode when processing such a
665     pattern, but PCRE should not loop if you do not (it no longer does).
666     [Detail: two cases were found: (a) a repeated subpattern containing an
667     extended class; (b) a recursive reference to a subpattern that followed a
668     previous extended class. It wasn't skipping over the extended class
669     correctly when UTF-8 mode was not set.]
670    
671     5. A negated single-character class was not being recognized as fixed-length
672     in lookbehind assertions such as (?<=[^f]), leading to an incorrect
673     compile error "lookbehind assertion is not fixed length".
674    
675     6. The RunPerlTest auxiliary script was showing an unexpected difference
676     between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
677     write a Perl script that can interpret lines of an input file either as
678     byte characters or as UTF-8, which is what "perltest" was being required to
679     do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
680     can't do is switch easily at run time between having the "use utf8;" pragma
681     or not. In the end, I fudged it by using the RunPerlTest script to insert
682     "use utf8;" explicitly for the UTF-8 tests.
683    
684     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
685     the end of the subject string, contrary to the documentation and to what
686     Perl does. This was true of both matching functions. Now it matches only at
687     the start of the subject and immediately after *internal* newlines.
688    
689     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
690     a pointer to an int instead of a pointer to an unsigned long int. This
691     caused problems on 64-bit systems.
692    
693     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
694     instance of the 'standard' template library not being so standard".
695    
696     10. There was no check on the number of named subpatterns nor the maximum
697     length of a subpattern name. The product of these values is used to compute
698     the size of the memory block for a compiled pattern. By supplying a very
699     long subpattern name and a large number of named subpatterns, the size
700     computation could be caused to overflow. This is now prevented by limiting
701     the length of names to 32 characters, and the number of named subpatterns
702     to 10,000.
703    
704     11. Subpatterns that are repeated with specific counts have to be replicated in
705     the compiled pattern. The size of memory for this was computed from the
706     length of the subpattern and the repeat count. The latter is limited to
707     65535, but there was no limit on the former, meaning that integer overflow
708     could in principle occur. The compiled length of a repeated subpattern is
709     now limited to 30,000 bytes in order to prevent this.
710    
711     12. Added the optional facility to have named substrings with the same name.
712    
713     13. Added the ability to use a named substring as a condition, using the
714     Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
715     are numbers (not recommended). Forward references are permitted.
716    
717     14. Added forward references in named backreferences (if you see what I mean).
718    
719     15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
720     pattern could run off the end of the subject. For example, the pattern
721     "(?s)(.{1,5})"8 did this with the subject "ab".
722    
723     16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
724     PCRE_CASELESS was set when matching characters that were quantified with ?
725     or *.
726    
727     17. A character class other than a single negated character that had a minimum
728     but no maximum quantifier - for example [ab]{6,} - was not handled
729     correctly by pce_dfa_exec(). It would match only one character.
730    
731     18. A valid (though odd) pattern that looked like a POSIX character
732     class but used an invalid character after [ (for example [[,abc,]]) caused
733     pcre_compile() to give the error "Failed: internal error: code overflow" or
734     in some cases to crash with a glibc free() error. This could even happen if
735     the pattern terminated after [[ but there just happened to be a sequence of
736     letters, a binary zero, and a closing ] in the memory that followed.
737    
738     19. Perl's treatment of octal escapes in the range \400 to \777 has changed
739     over the years. Originally (before any Unicode support), just the bottom 8
740     bits were taken. Thus, for example, \500 really meant \100. Nowadays the
741     output from "man perlunicode" includes this:
742    
743     The regular expression compiler produces polymorphic opcodes. That
744     is, the pattern adapts to the data and automatically switches to
745     the Unicode character scheme when presented with Unicode data--or
746     instead uses a traditional byte scheme when presented with byte
747     data.
748    
749     Sadly, a wide octal escape does not cause a switch, and in a string with
750     no other multibyte characters, these octal escapes are treated as before.
751     Thus, in Perl, the pattern /\500/ actually matches \100 but the pattern
752     /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
753     Unicode string.
754    
755     I have not perpetrated such confusion in PCRE. Up till now, it took just
756     the bottom 8 bits, as in old Perl. I have now made octal escapes with
757     values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
758     translate to the appropriate multibyte character.
759    
760     29. Applied some refactoring to reduce the number of warnings from Microsoft
761     and Borland compilers. This has included removing the fudge introduced
762     seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
763     a warning about an unused variable.
764    
765     21. PCRE has not included VT (character 0x0b) in the set of whitespace
766     characters since release 4.0, because Perl (from release 5.004) does not.
767     [Or at least, is documented not to: some releases seem to be in conflict
768     with the documentation.] However, when a pattern was studied with
769     pcre_study() and all its branches started with \s, PCRE still included VT
770     as a possible starting character. Of course, this did no harm; it just
771     caused an unnecessary match attempt.
772    
773     22. Removed a now-redundant internal flag bit that recorded the fact that case
774     dependency changed within the pattern. This was once needed for "required
775     byte" processing, but is no longer used. This recovers a now-scarce options
776     bit. Also moved the least significant internal flag bit to the most-
777     significant bit of the word, which was not previously used (hangover from
778     the days when it was an int rather than a uint) to free up another bit for
779     the future.
780    
781     23. Added support for CRLF line endings as well as CR and LF. As well as the
782     default being selectable at build time, it can now be changed at runtime
783     via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
784     specify that it is scanning data with non-default line endings.
785    
786     24. Changed the definition of CXXLINK to make it agree with the definition of
787     LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
788    
789     25. Applied Ian Taylor's patches to avoid using another stack frame for tail
790     recursions. This makes a big different to stack usage for some patterns.
791    
792     26. If a subpattern containing a named recursion or subroutine reference such
793     as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
794     the space required for the compiled pattern went wrong and gave too small a
795     value. Depending on the environment, this could lead to "Failed: internal
796     error: code overflow at offset 49" or "glibc detected double free or
797     corruption" errors.
798    
799     27. Applied patches from Google (a) to support the new newline modes and (b) to
800     advance over multibyte UTF-8 characters in GlobalReplace.
801    
802     28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
803     difference for some implementation of PCRE in some Windows version.
804    
805     29. Added some extra testing facilities to pcretest:
806    
807     \q<number> in a data line sets the "match limit" value
808     \Q<number> in a data line sets the "match recursion limt" value
809     -S <number> sets the stack size, where <number> is in megabytes
810    
811     The -S option isn't available for Windows.
812    
813    
814 nigel 89 Version 6.6 06-Feb-06
815     ---------------------
816    
817     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
818     in pcreposix.h. I have copied the definition from pcre.h.
819    
820     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
821     because pcre.h is no longer a built file.
822    
823     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
824     not normally included in the compiled code.
825    
826    
827 nigel 87 Version 6.5 01-Feb-06
828     ---------------------
829    
830     1. When using the partial match feature with pcre_dfa_exec(), it was not
831     anchoring the second and subsequent partial matches at the new starting
832     point. This could lead to incorrect results. For example, with the pattern
833     /1234/, partially matching against "123" and then "a4" gave a match.
834    
835     2. Changes to pcregrep:
836    
837     (a) All non-match returns from pcre_exec() were being treated as failures
838     to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
839     error message is output. Some extra information is given for the
840     PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
841     probably the only errors that are likely to be caused by users (by
842     specifying a regex that has nested indefinite repeats, for instance).
843     If there are more than 20 of these errors, pcregrep is abandoned.
844    
845     (b) A binary zero was treated as data while matching, but terminated the
846     output line if it was written out. This has been fixed: binary zeroes
847     are now no different to any other data bytes.
848    
849     (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
850     used to set a locale for matching. The --locale=xxxx long option has
851     been added (no short equivalent) to specify a locale explicitly on the
852     pcregrep command, overriding the environment variables.
853    
854     (d) When -B was used with -n, some line numbers in the output were one less
855     than they should have been.
856    
857     (e) Added the -o (--only-matching) option.
858    
859     (f) If -A or -C was used with -c (count only), some lines of context were
860     accidentally printed for the final match.
861    
862     (g) Added the -H (--with-filename) option.
863    
864     (h) The combination of options -rh failed to suppress file names for files
865     that were found from directory arguments.
866    
867     (i) Added the -D (--devices) and -d (--directories) options.
868    
869     (j) Added the -F (--fixed-strings) option.
870    
871     (k) Allow "-" to be used as a file name for -f as well as for a data file.
872    
873     (l) Added the --colo(u)r option.
874    
875     (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
876     is not present by default.
877    
878     3. A nasty bug was discovered in the handling of recursive patterns, that is,
879     items such as (?R) or (?1), when the recursion could match a number of
880     alternatives. If it matched one of the alternatives, but subsequently,
881     outside the recursion, there was a failure, the code tried to back up into
882     the recursion. However, because of the way PCRE is implemented, this is not
883     possible, and the result was an incorrect result from the match.
884    
885     In order to prevent this happening, the specification of recursion has
886     been changed so that all such subpatterns are automatically treated as
887     atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
888    
889     4. I had overlooked the fact that, in some locales, there are characters for
890     which isalpha() is true but neither isupper() nor islower() are true. In
891     the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
892     and ordfeminine) are like this. This affected the treatment of \w and \W
893     when they appeared in character classes, but not when they appeared outside
894     a character class. The bit map for "word" characters is now created
895     separately from the results of isalnum() instead of just taking it from the
896     upper, lower, and digit maps. (Plus the underscore character, of course.)
897    
898     5. The above bug also affected the handling of POSIX character classes such as
899     [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
900     permanent tables. Instead, the bit maps for such a class were previously
901     created as the appropriate unions of the upper, lower, and digit bitmaps.
902     Now they are created by subtraction from the [[:word:]] class, which has
903     its own bitmap.
904    
905     6. The [[:blank:]] character class matches horizontal, but not vertical space.
906     It is created by subtracting the vertical space characters (\x09, \x0a,
907     \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
908     subtraction was done in the overall bitmap for a character class, meaning
909     that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
910     be recognized. This bug has been fixed.
911    
912     7. Patches from the folks at Google:
913    
914     (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
915     real life, but is still worth protecting against".
916    
917     (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
918     regular expressions".
919    
920     (c) pcre_scanner.cc: avoid use of std::count() because not all systems
921     have it.
922    
923     (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
924     "configure" and the latter not, in order to fix a problem somebody had
925     with compiling the Arg class on HP-UX.
926    
927     (e) Improve the error-handling of the C++ wrapper a little bit.
928    
929     (f) New tests for checking recursion limiting.
930    
931     8. The pcre_memmove() function, which is used only if the environment does not
932     have a standard memmove() function (and is therefore rarely compiled),
933     contained two bugs: (a) use of int instead of size_t, and (b) it was not
934     returning a result (though PCRE never actually uses the result).
935    
936     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
937     large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
938     returned instead of calling malloc() with an overflowing number that would
939     most likely cause subsequent chaos.
940    
941     10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
942    
943     11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
944     with this option is matched, the nmatch and pmatch options of regexec() are
945     ignored.
946    
947     12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
948     provided in case anyone wants to the the POSIX interface with UTF-8
949     strings.
950    
951     13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
952     C++ linking (needed for some HP-UX environments).
953    
954     14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
955     (unused parameter) and in the pcre_printint() function (omitted "default"
956     switch label when the default is to do nothing).
957    
958     15. Added some code to make it possible, when PCRE is compiled as a C++
959     library, to replace subject pointers for pcre_exec() with a smart pointer
960     class, thus making it possible to process discontinuous strings.
961    
962     16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
963     much the same function. They were added by different people who were trying
964     to make PCRE easy to compile on non-Unix systems. It has been suggested
965     that PCRE_EXPORT be abolished now that there is more automatic apparatus
966     for compiling on Windows systems. I have therefore replaced it with
967     PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
968     defaults to "extern" for C or "extern C" for C++, which works fine on
969     Unix-like systems. It is now possible to override the value of PCRE_DATA_
970     SCOPE with something explicit in config.h. In addition:
971    
972     (a) pcreposix.h still had just "extern" instead of either of these macros;
973     I have replaced it with PCRE_DATA_SCOPE.
974    
975     (b) Functions such as _pcre_xclass(), which are internal to the library,
976     but external in the C sense, all had PCRE_EXPORT in their definitions.
977     This is apparently wrong for the Windows case, so I have removed it.
978     (It makes no difference on Unix-like systems.)
979    
980     17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
981     of recursive calls to match(). This is different to MATCH_LIMIT because
982     that limits the total number of calls to match(), not all of which increase
983     the depth of recursion. Limiting the recursion depth limits the amount of
984     stack (or heap if NO_RECURSE is set) that is used. The default can be set
985     when PCRE is compiled, and changed at run time. A patch from Google adds
986     this functionality to the C++ interface.
987    
988     18. Changes to the handling of Unicode character properties:
989    
990     (a) Updated the table to Unicode 4.1.0.
991    
992     (b) Recognize characters that are not in the table as "Cn" (undefined).
993    
994     (c) I revised the way the table is implemented to a much improved format
995     which includes recognition of ranges. It now supports the ranges that
996     are defined in UnicodeData.txt, and it also amalgamates other
997     characters into ranges. This has reduced the number of entries in the
998     table from around 16,000 to around 3,000, thus reducing its size
999     considerably. I realized I did not need to use a tree structure after
1000     all - a binary chop search is just as efficient. Having reduced the
1001     number of entries, I extended their size from 6 bytes to 8 bytes to
1002     allow for more data.
1003    
1004     (d) Added support for Unicode script names via properties such as \p{Han}.
1005    
1006     19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1007     matching that character.
1008    
1009     20. When matching a repeated Unicode property with a minimum greater than zero,
1010     (for example \pL{2,}), PCRE could look past the end of the subject if it
1011     reached it while seeking the minimum number of characters. This could
1012     happen only if some of the characters were more than one byte long, because
1013     there is a check for at least the minimum number of bytes.
1014    
1015     21. Refactored the implementation of \p and \P so as to be more general, to
1016     allow for more different types of property in future. This has changed the
1017     compiled form incompatibly. Anybody with saved compiled patterns that use
1018     \p or \P will have to recompile them.
1019    
1020     22. Added "Any" and "L&" to the supported property types.
1021    
1022     23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1023     but give a compile time error if the value is greater than 0xff.
1024    
1025     24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1026     accidentally not being installed or uninstalled.
1027    
1028     25. The pcre.h file was built from pcre.h.in, but the only changes that were
1029     made were to insert the current release number. This seemed silly, because
1030     it made things harder for people building PCRE on systems that don't run
1031     "configure". I have turned pcre.h into a distributed file, no longer built
1032     by "configure", with the version identification directly included. There is
1033     no longer a pcre.h.in file.
1034    
1035     However, this change necessitated a change to the pcre-config script as
1036     well. It is built from pcre-config.in, and one of the substitutions was the
1037     release number. I have updated configure.ac so that ./configure now finds
1038     the release number by grepping pcre.h.
1039    
1040     26. Added the ability to run the tests under valgrind.
1041    
1042    
1043 nigel 85 Version 6.4 05-Sep-05
1044     ---------------------
1045    
1046     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1047     "--" to be printed when multiple files were scanned, even when none of the
1048     -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1049     consider it to be a bug, and have restored the previous behaviour.
1050    
1051     2. A couple of code tidies to get rid of compiler warnings.
1052    
1053     3. The pcretest program used to cheat by referring to symbols in the library
1054     whose names begin with _pcre_. These are internal symbols that are not
1055     really supposed to be visible externally, and in some environments it is
1056     possible to suppress them. The cheating is now confined to including
1057     certain files from the library's source, which is a bit cleaner.
1058    
1059     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1060     file's purpose clearer.
1061    
1062     5. Reorganized pcre_ucp_findchar().
1063    
1064    
1065 nigel 83 Version 6.3 15-Aug-05
1066     ---------------------
1067    
1068     1. The file libpcre.pc.in did not have general read permission in the tarball.
1069    
1070     2. There were some problems when building without C++ support:
1071    
1072     (a) If C++ support was not built, "make install" and "make test" still
1073     tried to test it.
1074    
1075     (b) There were problems when the value of CXX was explicitly set. Some
1076     changes have been made to try to fix these, and ...
1077    
1078     (c) --disable-cpp can now be used to explicitly disable C++ support.
1079    
1080     (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1081     backslash in a target when C++ was disabled. This confuses some
1082     versions of "make", apparently. Using an intermediate variable solves
1083     this. (Same for CPP_LOBJ.)
1084    
1085     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1086     (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1087     necessary on certain architectures.
1088    
1089     4. Added a setting of -export-symbols-regex to the link command to remove
1090     those symbols that are exported in the C sense, but actually are local
1091     within the library, and not documented. Their names all begin with
1092     "_pcre_". This is not a perfect job, because (a) we have to except some
1093     symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1094     available (and never for static libraries). I have made a note to try to
1095     find a way round (a) in the future.
1096    
1097    
1098 nigel 81 Version 6.2 01-Aug-05
1099     ---------------------
1100    
1101     1. There was no test for integer overflow of quantifier values. A construction
1102     such as {1111111111111111} would give undefined results. What is worse, if
1103     a minimum quantifier for a parenthesized subpattern overflowed and became
1104     negative, the calculation of the memory size went wrong. This could have
1105     led to memory overwriting.
1106    
1107     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1108    
1109     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1110     operating environments where this matters.
1111    
1112     4. Applied Giuseppe Maxia's patch to add additional features for controlling
1113     PCRE options from within the C++ wrapper.
1114    
1115     5. Named capturing subpatterns were not being correctly counted when a pattern
1116     was compiled. This caused two problems: (a) If there were more than 100
1117     such subpatterns, the calculation of the memory needed for the whole
1118     compiled pattern went wrong, leading to an overflow error. (b) Numerical
1119     back references of the form \12, where the number was greater than 9, were
1120     not recognized as back references, even though there were sufficient
1121     previous subpatterns.
1122    
1123     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1124     versions of gcc, e.g. 2.95.4.
1125    
1126    
1127 nigel 79 Version 6.1 21-Jun-05
1128     ---------------------
1129    
1130     1. There was one reference to the variable "posix" in pcretest.c that was not
1131     surrounded by "#if !defined NOPOSIX".
1132    
1133     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1134     the cross-check on the old pcre_info() function, for the benefit of the
1135     cut-down version of PCRE that is currently imported into Exim.
1136    
1137     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1138     allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1139     patterns that start (?i) but I don't think that matters. The use of (?i) is
1140     just an example; this all applies to the other options as well.
1141    
1142     4. Since libtool seems to echo the compile commands it is issuing, the output
1143     from "make" can be reduced a bit by putting "@" in front of each libtool
1144     compile command.
1145    
1146     5. Patch from the folks at Google for configure.in to be a bit more thorough
1147     in checking for a suitable C++ installation before trying to compile the
1148     C++ stuff. This should fix a reported problem when a compiler was present,
1149     but no suitable headers.
1150    
1151     6. The man pages all had just "PCRE" as their title. I have changed them to
1152     be the relevant file name. I have also arranged that these names are
1153     retained in the file doc/pcre.txt, which is a concatenation in text format
1154     of all the man pages except the little individual ones for each function.
1155    
1156     7. The NON-UNIX-USE file had not been updated for the different set of source
1157     files that come with release 6. I also added a few comments about the C++
1158     wrapper.
1159    
1160    
1161 nigel 77 Version 6.0 07-Jun-05
1162     ---------------------
1163    
1164     1. Some minor internal re-organization to help with my DFA experiments.
1165    
1166     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1167     didn't matter for the library itself when fully configured, but did matter
1168     when compiling without UCP support, or within Exim, where the ucp files are
1169     not imported.
1170    
1171     3. Refactoring of the library code to split up the various functions into
1172     different source modules. The addition of the new DFA matching code (see
1173     below) to a single monolithic source would have made it really too
1174     unwieldy, quite apart from causing all the code to be include in a
1175     statically linked application, when only some functions are used. This is
1176     relevant even without the DFA addition now that patterns can be compiled in
1177     one application and matched in another.
1178    
1179     The downside of splitting up is that there have to be some external
1180     functions and data tables that are used internally in different modules of
1181     the library but which are not part of the API. These have all had their
1182     names changed to start with "_pcre_" so that they are unlikely to clash
1183     with other external names.
1184    
1185     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1186     a different (DFA) algorithm. Although it is slower than the original
1187     function, it does have some advantages for certain types of matching
1188     problem.
1189    
1190     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1191     including restarting after a partial match.
1192    
1193     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1194     defined when compiling for Windows was sent to me. I have put it into the
1195     code, though I have no means of testing or verifying it.
1196    
1197     7. Added the pcre_refcount() auxiliary function.
1198    
1199     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1200     match before or at the first newline in the subject string. In pcretest,
1201     the /f option on a pattern can be used to set this.
1202    
1203     9. A repeated \w when used in UTF-8 mode with characters greater than 256
1204     would behave wrongly. This has been present in PCRE since release 4.0.
1205    
1206     10. A number of changes to the pcregrep command:
1207    
1208     (a) Refactored how -x works; insert ^(...)$ instead of setting
1209     PCRE_ANCHORED and checking the length, in preparation for adding
1210     something similar for -w.
1211    
1212     (b) Added the -w (match as a word) option.
1213    
1214     (c) Refactored the way lines are read and buffered so as to have more
1215     than one at a time available.
1216    
1217     (d) Implemented a pcregrep test script.
1218    
1219     (e) Added the -M (multiline match) option. This allows patterns to match
1220     over several lines of the subject. The buffering ensures that at least
1221     8K, or the rest of the document (whichever is the shorter) is available
1222     for matching (and similarly the previous 8K for lookbehind assertions).
1223    
1224     (f) Changed the --help output so that it now says
1225    
1226     -w, --word-regex(p)
1227    
1228     instead of two lines, one with "regex" and the other with "regexp"
1229     because that confused at least one person since the short forms are the
1230     same. (This required a bit of code, as the output is generated
1231     automatically from a table. It wasn't just a text change.)
1232    
1233     (g) -- can be used to terminate pcregrep options if the next thing isn't an
1234     option but starts with a hyphen. Could be a pattern or a path name
1235     starting with a hyphen, for instance.
1236    
1237     (h) "-" can be given as a file name to represent stdin.
1238    
1239     (i) When file names are being printed, "(standard input)" is used for
1240     the standard input, for compatibility with GNU grep. Previously
1241     "<stdin>" was used.
1242    
1243     (j) The option --label=xxx can be used to supply a name to be used for
1244     stdin when file names are being printed. There is no short form.
1245    
1246     (k) Re-factored the options decoding logic because we are going to add
1247     two more options that take data. Such options can now be given in four
1248     different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1249    
1250     (l) Added the -A, -B, and -C options for requesting that lines of context
1251     around matches be printed.
1252    
1253     (m) Added the -L option to print the names of files that do not contain
1254     any matching lines, that is, the complement of -l.
1255    
1256     (n) The return code is 2 if any file cannot be opened, but pcregrep does
1257     continue to scan other files.
1258    
1259     (o) The -s option was incorrectly implemented. For compatibility with other
1260     greps, it now suppresses the error message for a non-existent or non-
1261     accessible file (but not the return code). There is a new option called
1262     -q that suppresses the output of matching lines, which was what -s was
1263     previously doing.
1264    
1265     (p) Added --include and --exclude options to specify files for inclusion
1266     and exclusion when recursing.
1267    
1268     11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1269     Hopefully, it now does.
1270    
1271     12. Missing cast in pcre_study().
1272    
1273     13. Added an "uninstall" target to the makefile.
1274    
1275     14. Replaced "extern" in the function prototypes in Makefile.in with
1276     "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1277     world, but is set differently for Windows.
1278    
1279     15. Added a second compiling function called pcre_compile2(). The only
1280     difference is that it has an extra argument, which is a pointer to an
1281     integer error code. When there is a compile-time failure, this is set
1282     non-zero, in addition to the error test pointer being set to point to an
1283     error message. The new argument may be NULL if no error number is required
1284     (but then you may as well call pcre_compile(), which is now just a
1285     wrapper). This facility is provided because some applications need a
1286     numeric error indication, but it has also enabled me to tidy up the way
1287     compile-time errors are handled in the POSIX wrapper.
1288    
1289     16. Added VPATH=.libs to the makefile; this should help when building with one
1290     prefix path and installing with another. (Or so I'm told by someone who
1291     knows more about this stuff than I do.)
1292    
1293     17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1294     passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1295     match everything, including newlines. This is not POSIX-compatible, but
1296     somebody wanted the feature. From pcretest it can be activated by using
1297     both the P and the s flags.
1298    
1299     18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1300    
1301     19. libpcre.pc was being incorrectly installed as executable.
1302    
1303     20. A couple of places in pcretest check for end-of-line by looking for '\n';
1304     it now also looks for '\r' so that it will work unmodified on Windows.
1305    
1306     21. Added Google's contributed C++ wrapper to the distribution.
1307    
1308     22. Added some untidy missing memory free() calls in pcretest, to keep
1309     Electric Fence happy when testing.
1310    
1311    
1312    
1313 nigel 75 Version 5.0 13-Sep-04
1314     ---------------------
1315    
1316     1. Internal change: literal characters are no longer packed up into items
1317     containing multiple characters in a single byte-string. Each character
1318     is now matched using a separate opcode. However, there may be more than one
1319     byte in the character in UTF-8 mode.
1320    
1321     2. The pcre_callout_block structure has two new fields: pattern_position and
1322     next_item_length. These contain the offset in the pattern to the next match
1323     item, and its length, respectively.
1324    
1325     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1326     insertion of callouts before each pattern item. Added the /C option to
1327     pcretest to make use of this.
1328    
1329     4. On the advice of a Windows user, the lines
1330    
1331     #if defined(_WIN32) || defined(WIN32)
1332     _setmode( _fileno( stdout ), 0x8000 );
1333     #endif /* defined(_WIN32) || defined(WIN32) */
1334    
1335     have been added to the source of pcretest. This apparently does useful
1336     magic in relation to line terminators.
1337    
1338     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1339     for the benefit of those environments where the "b" makes a difference.
1340    
1341     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1342     to know about it. I have put a hack into configure.in that adds in code
1343     to set GCC=yes if CC=icc. This seems to end up at a point in the
1344     generated configure script that is early enough to affect the setting of
1345     compiler options, which is what is needed, but I have no means of testing
1346     whether it really works. (The user who reported this had patched the
1347     generated configure script, which of course I cannot do.)
1348    
1349     LATER: After change 22 below (new libtool files), the configure script
1350     seems to know about icc (and also ecc). Therefore, I have commented out
1351     this hack in configure.in.
1352    
1353     7. Added support for pkg-config (2 patches were sent in).
1354    
1355     8. Negated POSIX character classes that used a combination of internal tables
1356     were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1357     [[:^ascii]]. Typically, they would match almost any characters. The other
1358     POSIX classes were not broken in this way.
1359    
1360     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1361     to find the match, as PCRE was deluded into thinking that the match had to
1362     start at the start point or following a newline. The same bug applied to
1363     patterns with negative forward assertions or any backward assertions
1364     preceding ".*" at the start, unless the pattern required a fixed first
1365     character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1366    
1367     10. In UTF-8 mode, when moving forwards in the subject after a failed match
1368     starting at the last subject character, bytes beyond the end of the subject
1369     string were read.
1370    
1371     11. Renamed the variable "class" as "classbits" to make life easier for C++
1372     users. (Previously there was a macro definition, but it apparently wasn't
1373     enough.)
1374    
1375     12. Added the new field "tables" to the extra data so that tables can be passed
1376     in at exec time, or the internal tables can be re-selected. This allows
1377     a compiled regex to be saved and re-used at a later time by a different
1378     program that might have everything at different addresses.
1379    
1380     13. Modified the pcre-config script so that, when run on Solaris, it shows a
1381     -R library as well as a -L library.
1382    
1383     14. The debugging options of pcretest (-d on the command line or D on a
1384     pattern) showed incorrect output for anything following an extended class
1385     that contained multibyte characters and which was followed by a quantifier.
1386    
1387     15. Added optional support for general category Unicode character properties
1388     via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1389     support. It adds about 90K to the size of the library. The meanings of the
1390     inbuilt class escapes such as \d and \s have NOT been changed.
1391    
1392     16. Updated pcredemo.c to include calls to free() to release the memory for the
1393     compiled pattern.
1394    
1395     17. The generated file chartables.c was being created in the source directory
1396     instead of in the building directory. This caused the build to fail if the
1397     source directory was different from the building directory, and was
1398     read-only.
1399    
1400     18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1401     file. No doubt somebody will tell me if they don't make sense... Also added
1402     Dan Mooney's comments about building on OpenVMS.
1403    
1404     19. Added support for partial matching via the PCRE_PARTIAL option for
1405     pcre_exec() and the \P data escape in pcretest.
1406    
1407     20. Extended pcretest with 3 new pattern features:
1408    
1409     (i) A pattern option of the form ">rest-of-line" causes pcretest to
1410     write the compiled pattern to the file whose name is "rest-of-line".
1411     This is a straight binary dump of the data, with the saved pointer to
1412     the character tables forced to be NULL. The study data, if any, is
1413     written too. After writing, pcretest reads a new pattern.
1414    
1415     (ii) If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1416     compiled pattern from the given file. There must not be any
1417     occurrences of "<" in the file name (pretty unlikely); if there are,
1418     pcretest will instead treat the initial "<" as a pattern delimiter.
1419     After reading in the pattern, pcretest goes on to read data lines as
1420     usual.
1421    
1422     (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1423     and 16-bit fields in a compiled pattern, to simulate a pattern that
1424     was compiled on a host of opposite endianness.
1425    
1426     21. The pcre-exec() function can now cope with patterns that were compiled on
1427     hosts of opposite endianness, with this restriction:
1428    
1429     As for any compiled expression that is saved and used later, the tables
1430     pointer field cannot be preserved; the extra_data field in the arguments
1431     to pcre_exec() should be used to pass in a tables address if a value
1432     other than the default internal tables were used at compile time.
1433    
1434     22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1435     now diagnosed as an error. Previously, most of the time, a negative number
1436     would have been treated as zero, but if in addition "ovector" was passed as
1437     NULL, a crash could occur.
1438    
1439     23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1440     new versions from the libtool 1.5 distribution (the last one is a copy of
1441     a file called libtool.m4). This seems to have fixed the need to patch
1442     "configure" to support Darwin 1.3 (which I used to do). However, I still
1443     had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1444     workstation).
1445    
1446     24. Changed the PCRE licence to be the more standard "BSD" licence.
1447    
1448    
1449 nigel 73 Version 4.5 01-Dec-03
1450     ---------------------
1451    
1452     1. There has been some re-arrangement of the code for the match() function so
1453     that it can be compiled in a version that does not call itself recursively.
1454     Instead, it keeps those local variables that need separate instances for
1455     each "recursion" in a frame on the heap, and gets/frees frames whenever it
1456     needs to "recurse". Keeping track of where control must go is done by means
1457     of setjmp/longjmp. The whole thing is implemented by a set of macros that
1458     hide most of the details from the main code, and operates only if
1459     NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1460     "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1461     operating.
1462    
1463     To make it easier for callers to provide specially tailored get/free
1464     functions for this usage, two new functions, pcre_stack_malloc, and
1465     pcre_stack_free, are used. They are always called in strict stacking order,
1466     and the size of block requested is always the same.
1467    
1468     The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1469     PCRE has been compiled to use the stack or the heap for recursion. The
1470     -C option of pcretest uses this to show which version is compiled.
1471    
1472     A new data escape \S, is added to pcretest; it causes the amounts of store
1473     obtained and freed by both kinds of malloc/free at match time to be added
1474     to the output.
1475    
1476     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1477     what's available on my current Linux desktop machine.
1478    
1479     3. When matching a UTF-8 string, the test for a valid string at the start has
1480     been extended. If start_offset is not zero, PCRE now checks that it points
1481     to a byte that is the start of a UTF-8 character. If not, it returns
1482     PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1483     this is necessary because there may be backward assertions in the pattern.
1484     When matching the same subject several times, it may save resources to use
1485     PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1486    
1487     4. The code for checking the validity of UTF-8 strings has been tightened so
1488     that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1489     containing "overlong sequences".
1490    
1491     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1492     I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1493     should have been "&", but it just so happened that all the cases this let
1494     through by mistake were picked up later in the function.
1495    
1496     6. I had used a variable called "isblank" - this is a C99 function, causing
1497     some compilers to warn. To avoid this, I renamed it (as "blankclass").
1498    
1499     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1500     prompting; (b) run "./pcretest /dev/null" at the start of the test script
1501     so the version is shown; (c) stop "make test" echoing "./RunTest".
1502    
1503     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1504    
1505     9. The prototype for memmove() for systems that don't have it was using
1506     size_t, but the inclusion of the header that defines size_t was later. I've
1507     moved the #includes for the C headers earlier to avoid this.
1508    
1509     10. Added some adjustments to the code to make it easier to compiler on certain
1510     special systems:
1511    
1512     (a) Some "const" qualifiers were missing.
1513     (b) Added the macro EXPORT before all exported functions; by default this
1514     is defined to be empty.
1515     (c) Changed the dftables auxiliary program (that builds chartables.c) so
1516     that it reads its output file name as an argument instead of writing
1517     to the standard output and assuming this can be redirected.
1518    
1519     11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
1520     class containing characters with values greater than 255, PCRE compilation
1521     went into a loop.
1522    
1523     12. A recursive reference to a subpattern that was within another subpattern
1524     that had a minimum quantifier of zero caused PCRE to crash. For example,
1525     (x(y(?2))z)? provoked this bug with a subject that got as far as the
1526     recursion. If the recursively-called subpattern itself had a zero repeat,
1527     that was OK.
1528    
1529     13. In pcretest, the buffer for reading a data line was set at 30K, but the
1530     buffer into which it was copied (for escape processing) was still set at
1531     1024, so long lines caused crashes.
1532    
1533     14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
1534     "internal error: code overflow...". This applied to any character class
1535     that was followed by a possessive quantifier.
1536    
1537     15. Modified the Makefile to add libpcre.la as a prerequisite for
1538     libpcreposix.la because I was told this is needed for a parallel build to
1539     work.
1540    
1541     16. If a pattern that contained .* following optional items at the start was
1542     studied, the wrong optimizing data was generated, leading to matching
1543     errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
1544     matching string must start with a or b or c. The correct conclusion for
1545     this pattern is that a match can start with any character.
1546    
1547    
1548 nigel 71 Version 4.4 13-Aug-03
1549     ---------------------
1550    
1551     1. In UTF-8 mode, a character class containing characters with values between
1552     127 and 255 was not handled correctly if the compiled pattern was studied.
1553     In fixing this, I have also improved the studying algorithm for such
1554     classes (slightly).
1555    
1556     2. Three internal functions had redundant arguments passed to them. Removal
1557     might give a very teeny performance improvement.
1558    
1559     3. Documentation bug: the value of the capture_top field in a callout is *one
1560     more than* the number of the hightest numbered captured substring.
1561    
1562     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
1563     in incorrectly linking with a previously installed version. They now link
1564     explicitly with libpcre.la.
1565    
1566     5. configure.in no longer needs to recognize Cygwin specially.
1567    
1568     6. A problem in pcre.in for Windows platforms is fixed.
1569    
1570     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
1571     pcretest, it used to include the size of the study block as part of its
1572     output. Unfortunately, the structure contains a field that has a different
1573     size on different hardware architectures. This meant that the tests that
1574     showed this size failed. As the block is currently always of a fixed size,
1575     this information isn't actually particularly useful in pcretest output, so
1576     I have just removed it.
1577    
1578     8. Three pre-processor statements accidentally did not start in column 1.
1579     Sadly, there are *still* compilers around that complain, even though
1580     standard C has not required this for well over a decade. Sigh.
1581    
1582     9. In pcretest, the code for checking callouts passed small integers in the
1583     callout_data field, which is a void * field. However, some picky compilers
1584     complained about the casts involved for this on 64-bit systems. Now
1585     pcretest passes the address of the small integer instead, which should get
1586     rid of the warnings.
1587    
1588     10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
1589     both compile and run time, and gives an error if an invalid UTF-8 sequence
1590     is found. There is a option for disabling this check in cases where the
1591     string is known to be correct and/or the maximum performance is wanted.
1592    
1593     11. In response to a bug report, I changed one line in Makefile.in from
1594    
1595     -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
1596     to
1597     -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
1598    
1599     to look similar to other lines, but I have no way of telling whether this
1600     is the right thing to do, as I do not use Windows. No doubt I'll get told
1601     if it's wrong...
1602    
1603    
1604 nigel 69 Version 4.3 21-May-03
1605     ---------------------
1606    
1607     1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
1608     Makefile.
1609    
1610     2. Some refactoring to improve the quality of the code:
1611    
1612     (i) The utf8_table... variables are now declared "const".
1613    
1614     (ii) The code for \cx, which used the "case flipping" table to upper case
1615     lower case letters, now just substracts 32. This is ASCII-specific,
1616     but the whole concept of \cx is ASCII-specific, so it seems
1617     reasonable.
1618    
1619     (iii) PCRE was using its character types table to recognize decimal and
1620     hexadecimal digits in the pattern. This is silly, because it handles
1621     only 0-9, a-f, and A-F, but the character types table is locale-
1622     specific, which means strange things might happen. A private
1623     table is now used for this - though it costs 256 bytes, a table is
1624     much faster than multiple explicit tests. Of course, the standard
1625     character types table is still used for matching digits in subject
1626     strings against \d.
1627    
1628     (iv) Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
1629     ending in _t are). So I've renamed it as ESC_tee.
1630    
1631     3. The first argument for regexec() in the POSIX wrapper should have been
1632     defined as "const".
1633    
1634     4. Changed pcretest to use malloc() for its buffers so that they can be
1635     Electric Fenced for debugging.
1636    
1637     5. There were several places in the code where, in UTF-8 mode, PCRE would try
1638     to read one or more bytes before the start of the subject string. Often this
1639     had no effect on PCRE's behaviour, but in some circumstances it could
1640     provoke a segmentation fault.
1641    
1642     6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
1643     to try to read one or more bytes before the start of the subject string.
1644    
1645     7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
1646     UTF-8 support could misbehave in various ways if the subject string
1647     contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
1648     area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
1649     back over UTF-8 characters.)
1650    
1651    
1652 nigel 67 Version 4.2 14-Apr-03
1653     ---------------------
1654    
1655     1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
1656    
1657     2. Changes to the building process, supplied by Ronald Landheer-Cieslak
1658     [ON_WINDOWS]: new variable, "#" on non-Windows platforms
1659     [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
1660     [WIN_PREFIX]: new variable, "cyg" for Cygwin
1661     * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
1662     and BUILD_EXEEXT
1663     Note: automatic setting of the BUILD variables is not yet working
1664     set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
1665     compile-time but not at link-time
1666     [LINK]: use for linking executables only
1667     make different versions for Windows and non-Windows
1668     [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
1669     libraries
1670     [LINK_FOR_BUILD]: new variable
1671     [OBJEXT]: use throughout
1672     [EXEEXT]: use throughout
1673     <winshared>: new target
1674     <wininstall>: new target
1675     <dftables.o>: use native compiler
1676     <dftables>: use native linker
1677     <install>: handle Windows platform correctly
1678     <clean>: ditto
1679     <check>: ditto
1680     copy DLL to top builddir before testing
1681    
1682     As part of these changes, -no-undefined was removed again. This was reported
1683     to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
1684     in any case.
1685    
1686     3. Some tidies to get rid of compiler warnings:
1687    
1688     . In the match_data structure, match_limit was an unsigned long int, whereas
1689     match_call_count was an int. I've made them both unsigned long ints.
1690    
1691     . In pcretest the fact that a const uschar * doesn't automatically cast to
1692     a void * provoked a warning.
1693    
1694     . Turning on some more compiler warnings threw up some "shadow" variables
1695     and a few more missing casts.
1696    
1697     4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
1698     option, a class that contained a single character with a value between 128
1699     and 255 (e.g. /[\xFF]/) caused PCRE to crash.
1700    
1701     5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
1702     option, a class that contained several characters, but with at least one
1703     whose value was between 128 and 255 caused PCRE to crash.
1704    
1705    
1706 nigel 65 Version 4.1 12-Mar-03
1707     ---------------------
1708 nigel 63
1709 nigel 65 1. Compiling with gcc -pedantic found a couple of places where casts were
1710     needed, and a string in dftables.c that was longer than standard compilers are
1711     required to support.
1712    
1713     2. Compiling with Sun's compiler found a few more places where the code could
1714     be tidied up in order to avoid warnings.
1715    
1716     3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
1717     first of these names is deprecated in the latest Autoconf in favour of the name
1718     CC_FOR_BUILD, because "host" is typically used to mean the system on which the
1719     compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
1720     analogy I have changed it to CFLAGS_FOR_BUILD.
1721    
1722     4. Added -no-undefined to the linking command in the Makefile, because this is
1723     apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
1724     linking step for the pcreposix library.
1725    
1726     5. PCRE was failing to diagnose the case of two named groups with the same
1727     name.
1728    
1729     6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
1730     literal character that is needed in the subject for a match, and scans along to
1731     ensure that it is present before embarking on the full matching process. This
1732     saves time in cases of nested unlimited repeats that are never going to match.
1733     Problem: the scan can take a lot of time if the subject is very long (e.g.
1734     megabytes), thus penalizing straightforward matches. It is now done only if the
1735     amount of subject to be scanned is less than 1000 bytes.
1736    
1737     7. A lesser problem with the same optimization is that it was recording the
1738     first character of an anchored pattern as "needed", thus provoking a search
1739     right along the subject, even when the first match of the pattern was going to
1740     fail. The "needed" character is now not set for anchored patterns, unless it
1741     follows something in the pattern that is of non-fixed length. Thus, it still
1742     fulfils its original purpose of finding quick non-matches in cases of nested
1743     unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
1744    
1745    
1746     Version 4.0 17-Feb-03
1747     ---------------------
1748    
1749 nigel 63 1. If a comment in an extended regex that started immediately after a meta-item
1750     extended to the end of string, PCRE compiled incorrect data. This could lead to
1751     all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
1752    
1753     2. Moved to autoconf 2.53 and libtool 1.4.2.
1754    
1755     3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
1756     the special perltest8 script is no longer needed - all the tests can be run
1757     from a single perltest script.
1758    
1759     4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
1760     by \s. It has now been removed in PCRE. This means it isn't recognized as
1761     whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
1762     class [:space:] *does* include VT, thereby creating a mess.
1763    
1764     5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
1765     space and tab.
1766    
1767     6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
1768     its new features into the main test script, reducing the number of scripts.
1769    
1770     7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
1771     were backward compatible, and made the (?i) apply to the whole pattern, as if
1772     /i were given. Now it behaves more logically, and applies the option setting
1773     only to what follows. PCRE has been changed to follow suit. However, if it
1774     finds options settings right at the start of the pattern, it extracts them into
1775     the global options, as before. Thus, they show up in the info data.
1776    
1777     8. Added support for the \Q...\E escape sequence. Characters in between are
1778     treated as literals. This is slightly different from Perl in that $ and @ are
1779     also handled as literals inside the quotes. In Perl, they will cause variable
1780     interpolation. Note the following examples:
1781    
1782     Pattern PCRE matches Perl matches
1783    
1784     \Qabc$xyz\E abc$xyz abc followed by the contents of $xyz
1785     \Qabc\$xyz\E abc\$xyz abc\$xyz
1786     \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
1787    
1788     For compatibility with Perl, \Q...\E sequences are recognized inside character
1789     classes as well as outside them.
1790    
1791     9. Re-organized 3 code statements in pcretest to avoid "overflow in
1792     floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
1793     (size_t) cast to one statement in pcretest and one in pcreposix to avoid
1794     signed/unsigned warnings.
1795    
1796     10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
1797     option for pcretest, so I've replaced it by a simple function that does just
1798     that job.
1799    
1800     11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
1801     "pcregrep -".
1802    
1803     12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
1804     Java package. This provides some syntactic sugar for simple cases of what my
1805     documentation calls "once-only subpatterns". A pattern such as x*+ is the same
1806     as (?>x*). In other words, if what is inside (?>...) is just a single repeated
1807     item, you can use this simplified notation. Note that only makes sense with
1808     greedy quantifiers. Consequently, the use of the possessive quantifier forces
1809     greediness, whatever the setting of the PCRE_UNGREEDY option.
1810    
1811     13. A change of greediness default within a pattern was not taking effect at
1812     the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
1813     subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
1814     was abstracted outside.
1815    
1816     14. PCRE now supports the \G assertion. It is true when the current matching
1817     position is at the start point of the match. This differs from \A when the
1818     starting offset is non-zero. Used with the /g option of pcretest (or similar
1819     code), it works in the same way as it does for Perl's /g option. If all
1820     alternatives of a regex begin with \G, the expression is anchored to the start
1821     match position, and the "anchored" flag is set in the compiled expression.
1822    
1823     15. Some bugs concerning the handling of certain option changes within patterns
1824     have been fixed. These applied to options other than (?ims). For example,
1825     "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
1826     been the other way round. Some of this was related to change 7 above.
1827    
1828     16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
1829     features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
1830     and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
1831     POSIX classes only within a class (e.g. /[[:alpha:]]/).
1832    
1833     17. Added support for Perl's \C escape. This matches one byte, even in UTF8
1834     mode. Unlike ".", it always matches newline, whatever the setting of
1835     PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
1836     assertions. Perl allows it, but it doesn't (in general) work because it can't
1837     calculate the length of the lookbehind. At least, that's the case for Perl
1838     5.8.0 - I've been told they are going to document that it doesn't work in
1839     future.
1840    
1841     18. Added an error diagnosis for escapes that PCRE does not support: these are
1842     \L, \l, \N, \P, \p, \U, \u, and \X.
1843    
1844     19. Although correctly diagnosing a missing ']' in a character class, PCRE was
1845     reading past the end of the pattern in cases such as /[abcd/.
1846    
1847     20. PCRE was getting more memory than necessary for patterns with classes that
1848     contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
1849    
1850     21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
1851     compiling PCRE for use with Virtual Pascal.
1852    
1853     22. Small fix to the Makefile to make it work properly if the build is done
1854     outside the source tree.
1855    
1856     23. Added a new extension: a condition to go with recursion. If a conditional
1857     subpattern starts with (?(R) the "true" branch is used if recursion has
1858     happened, whereas the "false" branch is used only at the top level.
1859    
1860     24. When there was a very long string of literal characters (over 255 bytes
1861     without UTF support, over 250 bytes with UTF support), the computation of how
1862     much memory was required could be incorrect, leading to segfaults or other
1863     strange effects.
1864    
1865     25. PCRE was incorrectly assuming anchoring (either to start of subject or to
1866     start of line for a non-DOTALL pattern) when a pattern started with (.*) and
1867     there was a subsequent back reference to those brackets. This meant that, for
1868     example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
1869     possible to check for precisely this case. All we can do is abandon the
1870     optimization if .* occurs inside capturing brackets when there are any back
1871     references whatsoever. (See below for a better fix that came later.)
1872    
1873     26. The handling of the optimization for finding the first character of a
1874     non-anchored pattern, and for finding a character that is required later in the
1875     match were failing in some cases. This didn't break the matching; it just
1876     failed to optimize when it could. The way this is done has been re-implemented.
1877    
1878     27. Fixed typo in error message for invalid (?R item (it said "(?p").
1879    
1880     28. Added a new feature that provides some of the functionality that Perl
1881     provides with (?{...}). The facility is termed a "callout". The way it is done
1882     in PCRE is for the caller to provide an optional function, by setting
1883     pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
1884     global variable. By default it is unset, which disables all calling out. To get
1885     the function called, the regex must include (?C) at appropriate points. This
1886     is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
1887     This provides a means of identifying different callout points. When PCRE
1888     reaches such a point in the regex, if pcre_callout has been set, the external
1889     function is called. It is provided with data in a structure called
1890     pcre_callout_block, which is defined in pcre.h. If the function returns 0,
1891     matching continues; if it returns a non-zero value, the match at the current
1892     point fails. However, backtracking will occur if possible. [This was changed
1893     later and other features added - see item 49 below.]
1894    
1895     29. pcretest is upgraded to test the callout functionality. It provides a
1896     callout function that displays information. By default, it shows the start of
1897     the match and the current position in the text. There are some new data escapes
1898     to vary what happens:
1899    
1900     \C+ in addition, show current contents of captured substrings
1901     \C- do not supply a callout function
1902     \C!n return 1 when callout number n is reached
1903     \C!n!m return 1 when callout number n is reached for the mth time
1904    
1905     30. If pcregrep was called with the -l option and just a single file name, it
1906     output "<stdin>" if a match was found, instead of the file name.
1907    
1908     31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
1909     slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
1910     pcre_exec(). This saves a malloc/free per call. The default value of
1911     POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
1912     when configuring.
1913    
1914     32. The default maximum size of a compiled pattern is 64K. There have been a
1915     few cases of people hitting this limit. The code now uses macros to handle the
1916     storing of links as offsets within the compiled pattern. It defaults to 2-byte
1917     links, but this can be changed to 3 or 4 bytes by --with-link-size when
1918     configuring. Tests 2 and 5 work only with 2-byte links because they output
1919     debugging information about compiled patterns.
1920    
1921     33. Internal code re-arrangements:
1922    
1923     (a) Moved the debugging function for printing out a compiled regex into
1924     its own source file (printint.c) and used #include to pull it into
1925     pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
1926     separate copies.
1927    
1928     (b) Defined the list of op-code names for debugging as a macro in
1929     internal.h so that it is next to the definition of the opcodes.
1930    
1931     (c) Defined a table of op-code lengths for simpler skipping along compiled
1932     code. This is again a macro in internal.h so that it is next to the
1933     definition of the opcodes.
1934    
1935     34. Added support for recursive calls to individual subpatterns, along the
1936     lines of Robin Houston's patch (but implemented somewhat differently).
1937    
1938     35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
1939     allow it to read and process whole directories in Win32. This code was
1940     contributed by Lionel Fourquaux; it has not been tested by me.
1941    
1942     36. Added support for named subpatterns. The Python syntax (?P<name>...) is
1943     used to name a group. Names consist of alphanumerics and underscores, and must
1944     be unique. Back references use the syntax (?P=name) and recursive calls use
1945     (?P>name) which is a PCRE extension to the Python extension. Groups still have
1946     numbers. The function pcre_fullinfo() can be used after compilation to extract
1947     a name/number map. There are three relevant calls:
1948    
1949     PCRE_INFO_NAMEENTRYSIZE yields the size of each entry in the map
1950     PCRE_INFO_NAMECOUNT yields the number of entries
1951     PCRE_INFO_NAMETABLE yields a pointer to the map.
1952    
1953     The map is a vector of fixed-size entries. The size of each entry depends on
1954     the length of the longest name used. The first two bytes of each entry are the
1955     group number, most significant byte first. There follows the corresponding
1956     name, zero terminated. The names are in alphabetical order.
1957    
1958     37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
1959     case instead of 255. Making it the same both with and without UTF-8 support
1960     means that the same test output works with both.
1961    
1962     38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
1963     calling malloc() with a zero argument.
1964    
1965     39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
1966     optimization. I've improved things by keeping a bitmap of backreferences with
1967     numbers 1-31 so that if .* occurs inside capturing brackets that are not in
1968     fact referenced, the optimization can be applied. It is unlikely that a
1969     relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
1970     the match to follow \n) will appear inside brackets with a number greater than
1971     31, but if it does, any back reference > 31 suppresses the optimization.
1972    
1973     40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
1974     of disabling numbered capturing parentheses. Any opening parenthesis that is
1975     not followed by ? behaves as if it were followed by ?: but named parentheses
1976     can still be used for capturing (and they will acquire numbers in the usual
1977     way).
1978    
1979     41. Redesigned the return codes from the match() function into yes/no/error so
1980     that errors can be passed back from deep inside the nested calls. A malloc
1981     failure while inside a recursive subpattern call now causes the
1982     PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
1983    
1984     42. It is now possible to set a limit on the number of times the match()
1985     function is called in a call to pcre_exec(). This facility makes it possible to
1986     limit the amount of recursion and backtracking, though not in a directly
1987     obvious way, because the match() function is used in a number of different
1988     circumstances. The count starts from zero for each position in the subject
1989     string (for non-anchored patterns). The default limit is, for compatibility, a
1990     large number, namely 10 000 000. You can change this in two ways:
1991    
1992     (a) When configuring PCRE before making, you can use --with-match-limit=n
1993     to set a default value for the compiled library.
1994    
1995     (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
1996     a different value is set. See 45 below.
1997    
1998     If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
1999    
2000     43. Added a new function pcre_config(int, void *) to enable run-time extraction
2001     of things that can be changed at compile time. The first argument specifies
2002     what is wanted and the second points to where the information is to be placed.
2003     The current list of available information is:
2004    
2005     PCRE_CONFIG_UTF8
2006    
2007     The output is an integer that is set to one if UTF-8 support is available;
2008     otherwise it is set to zero.
2009    
2010     PCRE_CONFIG_NEWLINE
2011    
2012     The output is an integer that it set to the value of the code that is used for
2013     newline. It is either LF (10) or CR (13).
2014    
2015     PCRE_CONFIG_LINK_SIZE
2016    
2017     The output is an integer that contains the number of bytes used for internal
2018     linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
2019    
2020     PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
2021    
2022     The output is an integer that contains the threshold above which the POSIX
2023     interface uses malloc() for output vectors. See item 31 above.
2024    
2025     PCRE_CONFIG_MATCH_LIMIT
2026    
2027     The output is an unsigned integer that contains the default limit of the number
2028     of match() calls in a pcre_exec() execution. See 42 above.
2029    
2030     44. pcretest has been upgraded by the addition of the -C option. This causes it
2031     to extract all the available output from the new pcre_config() function, and to
2032     output it. The program then exits immediately.
2033    
2034     45. A need has arisen to pass over additional data with calls to pcre_exec() in
2035     order to support additional features. One way would have been to define
2036     pcre_exec2() (for example) with extra arguments, but this would not have been
2037     extensible, and would also have required all calls to the original function to
2038     be mapped to the new one. Instead, I have chosen to extend the mechanism that
2039     is used for passing in "extra" data from pcre_study().
2040    
2041     The pcre_extra structure is now exposed and defined in pcre.h. It currently
2042     contains the following fields:
2043    
2044     flags a bitmap indicating which of the following fields are set
2045     study_data opaque data from pcre_study()
2046     match_limit a way of specifying a limit on match() calls for a specific
2047     call to pcre_exec()
2048     callout_data data for callouts (see 49 below)
2049    
2050     The flag bits are also defined in pcre.h, and are
2051    
2052     PCRE_EXTRA_STUDY_DATA
2053     PCRE_EXTRA_MATCH_LIMIT
2054     PCRE_EXTRA_CALLOUT_DATA
2055    
2056     The pcre_study() function now returns one of these new pcre_extra blocks, with
2057     the actual study data pointed to by the study_data field, and the
2058     PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
2059     before. That is, this change is entirely upwards-compatible and requires no
2060     change to existing code.
2061    
2062     If you want to pass in additional data to pcre_exec(), you can either place it
2063     in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
2064     block.
2065    
2066     46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
2067     data string contains the escape sequence \M, pcretest calls pcre_exec() several
2068     times with different match limits, until it finds the minimum value needed for
2069     pcre_exec() to complete. The value is then output. This can be instructive; for
2070     most simple matches the number is quite small, but for pathological cases it
2071     gets very large very quickly.
2072    
2073     47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
2074     returns the size of the data block pointed to by the study_data field in a
2075     pcre_extra block, that is, the value that was passed as the argument to
2076     pcre_malloc() when PCRE was getting memory in which to place the information
2077     created by pcre_study(). The fourth argument should point to a size_t variable.
2078     pcretest has been extended so that this information is shown after a successful
2079     pcre_study() call when information about the compiled regex is being displayed.
2080    
2081     48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
2082     because what follows is always an absolute path. (Later: it turns out that this
2083     is more than cosmetic for MinGW, because it doesn't like empty path
2084     components.)
2085    
2086     49. Some changes have been made to the callout feature (see 28 above):
2087    
2088     (i) A callout function now has three choices for what it returns:
2089    
2090     0 => success, carry on matching
2091     > 0 => failure at this point, but backtrack if possible
2092     < 0 => serious error, return this value from pcre_exec()
2093    
2094     Negative values should normally be chosen from the set of PCRE_ERROR_xxx
2095     values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
2096     "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
2097     use by callout functions. It will never be used by PCRE itself.
2098    
2099     (ii) The pcre_extra structure (see 45 above) has a void * field called
2100     callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
2101     pcre_callout_block structure has a field of the same name. The contents of
2102     the field passed in the pcre_extra structure are passed to the callout
2103     function in the corresponding field in the callout block. This makes it
2104     easier to use the same callout-containing regex from multiple threads. For
2105     testing, the pcretest program has a new data escape
2106    
2107     \C*n pass the number n (may be negative) as callout_data
2108    
2109     If the callout function in pcretest receives a non-zero value as
2110     callout_data, it returns that value.
2111    
2112     50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
2113     there were some redundant $(CFLAGS) in commands that are now specified as
2114     $(LINK), which already includes $(CFLAGS).
2115    
2116     51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
2117     has been compiled with UTF-8 support *and* pcre_compile() has been compiled
2118     with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
2119     one-byte characters throughout. Note that case-insensitive matching applies
2120     only to characters whose values are less than 256. PCRE doesn't support the
2121     notion of cases for higher-valued characters.
2122    
2123     (i) A character class whose characters are all within 0-255 is handled as
2124     a bit map, and the map is inverted for negative classes. Previously, a
2125     character > 255 always failed to match such a class; however it should
2126     match if the class was a negative one (e.g. [^ab]). This has been fixed.
2127    
2128     (ii) A negated character class with a single character < 255 is coded as
2129     "not this character" (OP_NOT). This wasn't working properly when the test
2130     character was multibyte, either singly or repeated.
2131    
2132     (iii) Repeats of multibyte characters are now handled correctly in UTF-8
2133     mode, for example: \x{100}{2,3}.
2134    
2135     (iv) The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
2136     singly or repeated) now correctly test multibyte characters. However,
2137     PCRE doesn't recognize any characters with values greater than 255 as
2138     digits, spaces, or word characters. Such characters always match \D, \S,
2139     and \W, and never match \d, \s, or \w.
2140    
2141     (v) Classes may now contain characters and character ranges with values
2142     greater than 255. For example: [ab\x{100}-\x{400}].
2143    
2144     (vi) pcregrep now has a --utf-8 option (synonym -u) which makes it call
2145     PCRE in UTF-8 mode.
2146    
2147     52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
2148     PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
2149     retained for backwards compatibility. (Note that LASTLITERAL is also a byte
2150     value.)
2151    
2152     53. The single man page has become too large. I have therefore split it up into
2153     a number of separate man pages. These also give rise to individual HTML pages;
2154     these are now put in a separate directory, and there is an index.html page that
2155     lists them all. Some hyperlinking between the pages has been installed.
2156    
2157     54. Added convenience functions for handling named capturing parentheses.
2158    
2159     55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2160     aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2161     true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2162     are faulted.
2163    
2164     56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2165     calling configure. These values are used when compiling the dftables.c program
2166     which is run to generate the source of the default character tables. They
2167     default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2168     you will need to set these values.
2169    
2170     57. Updated the building process for Windows DLL, as provided by Fred Cox.
2171    
2172    
2173     Version 3.9 02-Jan-02
2174 nigel 61 ---------------------
2175    
2176     1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2177    
2178     2. If --disable-static was given, the building process failed when trying to
2179     build pcretest and pcregrep. (For some reason it was using libtool to compile
2180     them, which is not right, as they aren't part of the library.)
2181    
2182    
2183 nigel 59 Version 3.8 18-Dec-01
2184     ---------------------
2185    
2186     1. The experimental UTF-8 code was completely screwed up. It was packing the
2187     bytes in the wrong order. How dumb can you get?
2188    
2189    
2190 nigel 57 Version 3.7 29-Oct-01
2191     ---------------------
2192    
2193     1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2194     This caused pcretest, when used on the test data, to segfault. Unfortunately,
2195     this didn't happen under Solaris 8, where I normally test things.
2196    
2197 nigel 59 2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2198     doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2199     isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2200     this fix an hour or so after the initial 3.7 release.)
2201 nigel 57
2202 nigel 59
2203 nigel 55 Version 3.6 23-Oct-01
2204     ---------------------
2205    
2206     1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2207     offsets passed as NULL with zero offset count.
2208    
2209     2. The config.guess and config.sub files had not been updated when I moved to
2210     the latest autoconf.
2211    
2212    
2213 nigel 53 Version 3.5 15-Aug-01
2214     ---------------------
2215 nigel 5
2216 nigel 53 1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2217     had been forgotten.
2218    
2219     2. By using declared but undefined structures, we can avoid using "void"
2220     definitions in pcre.h while keeping the internal definitions of the structures
2221     private.
2222    
2223     3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2224     user point of view, this means that both static and shared libraries are built
2225     by default, but this can be individually controlled. More of the work of
2226     handling this static/shared cases is now inside libtool instead of PCRE's make
2227     file.
2228    
2229     4. The pcretest utility is now installed along with pcregrep because it is
2230     useful for users (to test regexs) and by doing this, it automatically gets
2231     relinked by libtool. The documentation has been turned into a man page, so
2232     there are now .1, .txt, and .html versions in /doc.
2233    
2234     5. Upgrades to pcregrep:
2235     (i) Added long-form option names like gnu grep.
2236     (ii) Added --help to list all options with an explanatory phrase.
2237     (iii) Added -r, --recursive to recurse into sub-directories.
2238     (iv) Added -f, --file to read patterns from a file.
2239    
2240     6. pcre_exec() was referring to its "code" argument before testing that
2241     argument for NULL (and giving an error if it was NULL).
2242    
2243     7. Upgraded Makefile.in to allow for compiling in a different directory from
2244     the source directory.
2245    
2246     8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2247     options bits, the pointer it was passed was to an int instead of to an unsigned
2248     long int. This mattered only on 64-bit systems.
2249    
2250     9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2251     generated) instead of pcre.in, which it its source. Also made the same change
2252     in several of the .c files.
2253    
2254     10. A new release of gcc defines printf() as a macro, which broke pcretest
2255     because it had an ifdef in the middle of a string argument for printf(). Fixed
2256     by using separate calls to printf().
2257    
2258     11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2259     script, to force use of CR or LF instead of \n in the source. On non-Unix
2260     systems, the value can be set in config.h.
2261    
2262     12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2263     absolute limit. Changed the text of the error message to make this clear, and
2264     likewise updated the man page.
2265    
2266     13. The limit of 99 on the number of capturing subpatterns has been removed.
2267     The new limit is 65535, which I hope will not be a "real" limit.
2268    
2269    
2270 nigel 51 Version 3.4 22-Aug-00
2271     ---------------------
2272    
2273     1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2274    
2275     2. Diagnose condition (?(0) as an error instead of crashing on matching.
2276    
2277    
2278 nigel 49 Version 3.3 01-Aug-00
2279     ---------------------
2280    
2281     1. If an octal character was given, but the value was greater than \377, it
2282     was not getting masked to the least significant bits, as documented. This could
2283     lead to crashes in some systems.
2284    
2285     2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
2286     the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
2287    
2288     3. Added the functions pcre_free_substring() and pcre_free_substring_list().
2289     These just pass their arguments on to (pcre_free)(), but they are provided
2290     because some uses of PCRE bind it to non-C systems that can call its functions,
2291     but cannot call free() or pcre_free() directly.
2292    
2293     4. Add "make test" as a synonym for "make check". Corrected some comments in
2294     the Makefile.
2295    
2296     5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
2297     Makefile.
2298    
2299     6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
2300     command called pgrep for grepping around the active processes.
2301    
2302     7. Added the beginnings of support for UTF-8 character strings.
2303    
2304     8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
2305     RANLIB to ./ltconfig so that they are used by libtool. I think these are all
2306     the relevant ones. (AR is not passed because ./ltconfig does its own figuring
2307     out for the ar command.)
2308    
2309    
2310 nigel 47 Version 3.2 12-May-00
2311     ---------------------
2312    
2313     This is purely a bug fixing release.
2314    
2315     1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
2316     of ZA. This was just one example of several cases that could provoke this bug,
2317     which was introduced by change 9 of version 2.00. The code for breaking
2318     infinite loops after an iteration that matches an empty string was't working
2319     correctly.
2320    
2321     2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
2322     when matched against abbab (for example). After matching an empty string, it
2323     wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
2324     caused it to match further down the string than it should.
2325    
2326     3. The code contained an inclusion of sys/types.h. It isn't clear why this
2327     was there because it doesn't seem to be needed, and it causes trouble on some
2328     systems, as it is not a Standard C header. It has been removed.
2329    
2330     4. Made 4 silly changes to the source to avoid stupid compiler warnings that
2331     were reported on the Macintosh. The changes were from
2332    
2333     while ((c = *(++ptr)) != 0 && c != '\n');
2334     to
2335     while ((c = *(++ptr)) != 0 && c != '\n') ;
2336    
2337     Totally extraordinary, but if that's what it takes...
2338    
2339     5. PCRE is being used in one environment where neither memmove() nor bcopy() is
2340     available. Added HAVE_BCOPY and an autoconf test for it; if neither
2341     HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
2342     assumes the way PCRE uses memmove() (always moving upwards).
2343    
2344     6. PCRE is being used in one environment where strchr() is not available. There
2345     was only one use in pcre.c, and writing it out to avoid strchr() probably gives
2346     faster code anyway.
2347    
2348    
2349 nigel 45 Version 3.1 09-Feb-00
2350     ---------------------
2351    
2352     The only change in this release is the fixing of some bugs in Makefile.in for
2353     the "install" target:
2354    
2355     (1) It was failing to install pcreposix.h.
2356    
2357     (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
2358    
2359    
2360 nigel 43 Version 3.0 01-Feb-00
2361     ---------------------
2362 nigel 41
2363     1. Add support for the /+ modifier to perltest (to output $` like it does in
2364     pcretest).
2365    
2366     2. Add support for the /g modifier to perltest.
2367    
2368     3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
2369     matches null strings.
2370    
2371     4. Fix perltest so that it doesn't do unwanted things when fed an empty
2372     pattern. Perl treats empty patterns specially - it reuses the most recent
2373     pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
2374     effect.
2375    
2376     5. The POSIX interface was broken in that it was just handing over the POSIX
2377     captured string vector to pcre_exec(), but (since release 2.00) PCRE has
2378     required a bigger vector, with some working space on the end. This means that
2379     the POSIX wrapper now has to get and free some memory, and copy the results.
2380    
2381 nigel 43 6. Added some simple autoconf support, placing the test data and the
2382     documentation in separate directories, re-organizing some of the
2383     information files, and making it build pcre-config (a GNU standard). Also added
2384     libtool support for building PCRE as a shared library, which is now the
2385     default.
2386 nigel 41
2387 nigel 43 7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
2388     09 are not valid octal constants. Single digits will be used for minor values
2389     less than 10.
2390    
2391     8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
2392     existing programs that set these in the POSIX interface can use PCRE without
2393     modification.
2394    
2395     9. Added a new function, pcre_fullinfo() with an extensible interface. It can
2396     return all that pcre_info() returns, plus additional data. The pcre_info()
2397     function is retained for compatibility, but is considered to be obsolete.
2398    
2399     10. Added experimental recursion feature (?R) to handle one common case that
2400     Perl 5.6 will be able to do with (?p{...}).
2401    
2402     11. Added support for POSIX character classes like [:alpha:], which Perl is
2403     adopting.
2404    
2405    
2406 nigel 39 Version 2.08 31-Aug-99
2407     ----------------------
2408    
2409     1. When startoffset was not zero and the pattern began with ".*", PCRE was not
2410     trying to match at the startoffset position, but instead was moving forward to
2411     the next newline as if a previous match had failed.
2412    
2413     2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
2414     and could get into a loop if a null string was matched other than at the start
2415     of the subject.
2416    
2417     3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
2418     be distinguished at compile time, and for completeness also added PCRE_DATE.
2419    
2420     5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
2421     in GnuWin32 environments.
2422    
2423    
2424 nigel 37 Version 2.07 29-Jul-99
2425     ----------------------
2426    
2427     1. The documentation is now supplied in plain text form and HTML as well as in
2428     the form of man page sources.
2429    
2430     2. C++ compilers don't like assigning (void *) values to other pointer types.
2431     In particular this affects malloc(). Although there is no problem in Standard
2432     C, I've put in casts to keep C++ compilers happy.
2433    
2434     3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
2435     should be (const char *).
2436    
2437     4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
2438     be useful for non-Unix systems who don't want to bother with the POSIX stuff.
2439     However, I haven't made this a standard facility. The documentation doesn't
2440     mention it, and the Makefile doesn't support it.
2441    
2442     5. The Makefile now contains an "install" target, with editable destinations at
2443     the top of the file. The pcretest program is not installed.
2444    
2445     6. pgrep -V now gives the PCRE version number and date.
2446    
2447     7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
2448     causing the entire string to be ignored, instead of just the last character.
2449    
2450     8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
2451     non-matching string, it can take a very, very long time, even for strings of
2452     quite modest length, because of the nested recursion. PCRE now does better in
2453     some of these cases. It does this by remembering the last required literal
2454     character in the pattern, and pre-searching the subject to ensure it is present
2455     before running the real match. In other words, it applies a heuristic to detect
2456     some types of certain failure quickly, and in the above example, if presented
2457     with a string that has no trailing " it gives "no match" very quickly.
2458    
2459     9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
2460     other alternatives are tried instead.
2461    
2462    
2463 nigel 35 Version 2.06 09-Jun-99
2464     ----------------------
2465    
2466     1. Change pcretest's output for amount of store used to show just the code
2467     space, because the remainder (the data block) varies in size between 32-bit and
2468     64-bit systems.
2469    
2470     2. Added an extra argument to pcre_exec() to supply an offset in the subject to
2471     start matching at. This allows lookbehinds to work when searching for multiple
2472     occurrences in a string.
2473    
2474     3. Added additional options to pcretest for testing multiple occurrences:
2475    
2476     /+ outputs the rest of the string that follows a match
2477     /g loops for multiple occurrences, using the new startoffset argument
2478     /G loops for multiple occurrences by passing an incremented pointer
2479    
2480     4. PCRE wasn't doing the "first character" optimization for patterns starting
2481     with \b or \B, though it was doing it for other lookbehind assertions. That is,
2482     it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
2483     the letter 'x'. On long subject strings, this gives a significant speed-up.
2484    
2485    
2486 nigel 33 Version 2.05 21-Apr-99
2487     ----------------------
2488    
2489     1. Changed the type of magic_number from int to long int so that it works
2490     properly on 16-bit systems.
2491    
2492     2. Fixed a bug which caused patterns starting with .* not to work correctly
2493     when the subject string contained newline characters. PCRE was assuming
2494     anchoring for such patterns in all cases, which is not correct because .* will
2495     not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
2496     DOTALL is set at top level; otherwise it knows that patterns starting with .*
2497     must be retried after every newline in the subject.
2498    
2499    
2500 nigel 31 Version 2.04 18-Feb-99
2501     ----------------------
2502    
2503     1. For parenthesized subpatterns with repeats whose minimum was zero, the
2504     computation of the store needed to hold the pattern was incorrect (too large).
2505     If such patterns were nested a few deep, this could multiply and become a real
2506     problem.
2507    
2508     2. Added /M option to pcretest to show the memory requirement of a specific
2509     pattern. Made -m a synonym of -s (which does this globally) for compatibility.
2510    
2511     3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
2512     compiled in such a way that the backtracking after subsequent failure was
2513     pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
2514     ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
2515    
2516    
2517 nigel 29 Version 2.03 02-Feb-99
2518     ----------------------
2519    
2520     1. Fixed typo and small mistake in man page.
2521    
2522 nigel 31 2. Added 4th condition (GPL supersedes if conflict) and created separate
2523     LICENCE file containing the conditions.
2524 nigel 29
2525     3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
2526     Perl, that is the internal \ allows the delimiter to be included in the
2527     pattern. Locked out the use of \ as a delimiter. If \ immediately follows
2528     the final delimiter, add \ to the end of the pattern (to test the error).
2529    
2530     4. Added the convenience functions for extracting substrings after a successful
2531     match. Updated pcretest to make it able to test these functions.
2532    
2533    
2534 nigel 27 Version 2.02 14-Jan-99
2535     ----------------------
2536    
2537     1. Initialized the working variables associated with each extraction so that
2538     their saving and restoring doesn't refer to uninitialized store.
2539    
2540     2. Put dummy code into study.c in order to trick the optimizer of the IBM C
2541     compiler for OS/2 into generating correct code. Apparently IBM isn't going to
2542     fix the problem.
2543    
2544     3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
2545     calls, and wasn't printing the correct value for compiling calls. Increased the
2546     default value of LOOPREPEAT, and the number of significant figures in the
2547     times.
2548    
2549     4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
2550    
2551     5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
2552     a building problem on Windows NT with a FAT file system.
2553    
2554    
2555 nigel 25 Version 2.01 21-Oct-98
2556     ----------------------
2557    
2558     1. Changed the API for pcre_compile() to allow for the provision of a pointer
2559     to character tables built by pcre_maketables() in the current locale. If NULL
2560     is passed, the default tables are used.
2561    
2562    
2563 nigel 23 Version 2.00 24-Sep-98
2564 nigel 21 ----------------------
2565    
2566 nigel 23 1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
2567     it any more.
2568    
2569     2. Allow quantification of (?>) groups, and make it work correctly.
2570    
2571     3. The first character computation wasn't working for (?>) groups.
2572    
2573     4. Correct the implementation of \Z (it is permitted to match on the \n at the
2574     end of the subject) and add 5.005's \z, which really does match only at the
2575     very end of the subject.
2576    
2577     5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
2578    
2579     6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
2580     DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
2581     localized options. All options to pcre_study() were also removed.
2582    
2583     7. Add other new features from 5.005:
2584    
2585     $(?<= positive lookbehind
2586     $(?<! negative lookbehind
2587     (?imsx-imsx) added the unsetting capability
2588     such a setting is global if at outer level; local otherwise
2589     (?imsx-imsx:) non-capturing groups with option setting
2590     (?(cond)re|re) conditional pattern matching
2591    
2592     A backreference to itself in a repeated group matches the previous
2593     captured string.
2594    
2595     8. General tidying up of studying (both automatic and via "study")
2596     consequential on the addition of new assertions.
2597    
2598     9. As in 5.005, unlimited repeated groups that could match an empty substring
2599     are no longer faulted at compile time. Instead, the loop is forcibly broken at
2600     runtime if any iteration does actually match an empty substring.
2601    
2602     10. Include the RunTest script in the distribution.
2603    
2604     11. Added tests from the Perl 5.005_02 distribution. This showed up a few
2605     discrepancies, some of which were old and were also with respect to 5.004. They
2606     have now been fixed.
2607    
2608    
2609     Version 1.09 28-Apr-98
2610     ----------------------
2611    
2612 nigel 21 1. A negated single character class followed by a quantifier with a minimum
2613     value of one (e.g. [^x]{1,6} ) was not compiled correctly. This could lead to
2614     program crashes, or just wrong answers. This did not apply to negated classes
2615     containing more than one character, or to minima other than one.
2616    
2617    
2618 nigel 19 Version 1.08 27-Mar-98
2619     ----------------------
2620    
2621     1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
2622    
2623     2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
2624     latter must appear before anything that relies on it in the pattern.
2625    
2626    
2627 nigel 17 Version 1.07 16-Feb-98
2628     ----------------------
2629    
2630     1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
2631     repeat of a potentially empty string).
2632    
2633    
2634 nigel 15 Version 1.06 23-Jan-98
2635     ----------------------
2636    
2637     1. Added Markus Oberhumer's little patches for C++.
2638    
2639     2. Literal strings longer than 255 characters were broken.
2640    
2641    
2642 nigel 13 Version 1.05 23-Dec-97
2643     ----------------------
2644    
2645     1. Negated character classes containing more than one character were failing if
2646     PCRE_CASELESS was set at run time.
2647    
2648    
2649 nigel 11 Version 1.04 19-Dec-97
2650     ----------------------
2651    
2652     1. Corrected the man page, where some "const" qualifiers had been omitted.
2653    
2654     2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
2655     input syntax.
2656    
2657     3. Fixed memory leak which occurred when a regex with back references was
2658     matched with an offsets vector that wasn't big enough. The temporary memory
2659     that is used in this case wasn't being freed if the match failed.
2660    
2661     4. Tidied pcretest to ensure it frees memory that it gets.
2662    
2663     5. Temporary memory was being obtained in the case where the passed offsets
2664     vector was exactly big enough.
2665    
2666     6. Corrected definition of offsetof() from change 5 below.
2667    
2668     7. I had screwed up change 6 below and broken the rules for the use of
2669     setjmp(). Now fixed.
2670    
2671    
2672 nigel 9 Version 1.03 18-Dec-97
2673     ----------------------
2674    
2675     1. A erroneous regex with a missing opening parenthesis was correctly
2676     diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
2677     on some systems.
2678    
2679     2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
2680     it was reported that one broken compiler failed on the former because "code" is
2681     also an independent variable.
2682    
2683     3. The erroneous regex a[]b caused an array overrun reference.
2684    
2685     4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
2686     fail on data ending with that character. (It was going on too far, and checking
2687     the next character, typically a binary zero.) This was specific to the
2688     optimized code for single-character negative classes.
2689    
2690     5. Added a contributed patch from the TIN world which does the following:
2691    
2692     + Add an undef for memmove, in case the the system defines a macro for it.
2693    
2694     + Add a definition of offsetof(), in case there isn't one. (I don't know
2695     the reason behind this - offsetof() is part of the ANSI standard - but
2696     it does no harm).
2697    
2698     + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
2699     most of the places where whitespace preceded '#'. I have given up and
2700     allowed the remaining 2 cases to be at the margin.
2701    
2702     + Rename some variables in pcre to eliminate shadowing. This seems very
2703     pedantic, but does no harm, of course.
2704    
2705     6. Moved the call to setjmp() into its own function, to get rid of warnings
2706     from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
2707    
2708     7. Constructs such as \d{8,} were compiling into the equivalent of
2709 nigel 11 \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
2710 nigel 9 outcome, but in this particular case used more store than had been allocated,
2711     which caused the bug to be discovered because it threw up an internal error.
2712    
2713     8. The debugging code in both pcre and pcretest for outputting the compiled
2714     form of a regex was going wrong in the case of back references followed by
2715     curly-bracketed repeats.
2716    
2717    
2718 nigel 7 Version 1.02 12-Dec-97
2719     ----------------------
2720    
2721     1. Typos in pcre.3 and comments in the source fixed.
2722    
2723     2. Applied a contributed patch to get rid of places where it used to remove
2724     'const' from variables, and fixed some signed/unsigned and uninitialized
2725     variable warnings.
2726    
2727     3. Added the "runtest" target to Makefile.
2728    
2729     4. Set default compiler flag to -O2 rather than just -O.
2730    
2731    
2732 nigel 5 Version 1.01 19-Nov-97
2733     ----------------------
2734    
2735     1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
2736     like /([ab]*)*/, that is, for classes with more than one character in them.
2737    
2738     2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
2739     as /((?>a*))*/ (a PCRE_EXTRA facility).
2740    
2741    
2742     Version 1.00 18-Nov-97
2743     ----------------------
2744    
2745     1. Added compile-time macros to support systems such as SunOS4 which don't have
2746     memmove() or strerror() but have other things that can be used instead.
2747    
2748     2. Arranged that "make clean" removes the executables.
2749    
2750    
2751 nigel 3 Version 0.99 27-Oct-97
2752     ----------------------
2753    
2754     1. Fixed bug in code for optimizing classes with only one character. It was
2755     initializing a 32-byte map regardless, which could cause it to run off the end
2756     of the memory it had got.
2757    
2758     2. Added, conditional on PCRE_EXTRA, the proposed (?>REGEX) construction.
2759    
2760    
2761     Version 0.98 22-Oct-97
2762     ----------------------
2763    
2764     1. Fixed bug in code for handling temporary memory usage when there are more
2765     back references than supplied space in the ovector. This could cause segfaults.
2766    
2767    
2768     Version 0.97 21-Oct-97
2769     ----------------------
2770    
2771     1. Added the \X "cut" facility, conditional on PCRE_EXTRA.
2772    
2773     2. Optimized negated single characters not to use a bit map.
2774    
2775     3. Brought error texts together as macro definitions; clarified some of them;
2776     fixed one that was wrong - it said "range out of order" when it meant "invalid
2777     escape sequence".
2778    
2779     4. Changed some char * arguments to const char *.
2780    
2781     5. Added PCRE_NOTBOL and PCRE_NOTEOL (from POSIX).
2782    
2783     6. Added the POSIX-style API wrapper in pcreposix.a and testing facilities in
2784     pcretest.
2785    
2786    
2787     Version 0.96 16-Oct-97
2788     ----------------------
2789    
2790     1. Added a simple "pgrep" utility to the distribution.
2791    
2792     2. Fixed an incompatibility with Perl: "{" is now treated as a normal character
2793     unless it appears in one of the precise forms "{ddd}", "{ddd,}", or "{ddd,ddd}"
2794     where "ddd" means "one or more decimal digits".
2795    
2796     3. Fixed serious bug. If a pattern had a back reference, but the call to
2797     pcre_exec() didn't supply a large enough ovector to record the related
2798     identifying subpattern, the match always failed. PCRE now remembers the number
2799     of the largest back reference, and gets some temporary memory in which to save
2800     the offsets during matching if necessary, in order to ensure that
2801     backreferences always work.
2802    
2803     4. Increased the compatibility with Perl in a number of ways:
2804    
2805     (a) . no longer matches \n by default; an option PCRE_DOTALL is provided
2806     to request this handling. The option can be set at compile or exec time.
2807    
2808     (b) $ matches before a terminating newline by default; an option
2809     PCRE_DOLLAR_ENDONLY is provided to override this (but not in multiline
2810     mode). The option can be set at compile or exec time.
2811    
2812     (c) The handling of \ followed by a digit other than 0 is now supposed to be
2813     the same as Perl's. If the decimal number it represents is less than 10
2814     or there aren't that many previous left capturing parentheses, an octal
2815     escape is read. Inside a character class, it's always an octal escape,
2816     even if it is a single digit.
2817    
2818     (d) An escaped but undefined alphabetic character is taken as a literal,
2819     unless PCRE_EXTRA is set. Currently this just reserves the remaining
2820     escapes.
2821    
2822     (e) {0} is now permitted. (The previous item is removed from the compiled
2823     pattern).
2824    
2825     5. Changed all the names of code files so that the basic parts are no longer
2826     than 10 characters, and abolished the teeny "globals.c" file.
2827    
2828     6. Changed the handling of character classes; they are now done with a 32-byte
2829     bit map always.
2830    
2831     7. Added the -d and /D options to pcretest to make it possible to look at the
2832     internals of compilation without having to recompile pcre.
2833    
2834    
2835     Version 0.95 23-Sep-97
2836     ----------------------
2837    
2838     1. Fixed bug in pre-pass concerning escaped "normal" characters such as \x5c or
2839     \x20 at the start of a run of normal characters. These were being treated as
2840     real characters, instead of the source characters being re-checked.
2841    
2842    
2843     Version 0.94 18-Sep-97
2844     ----------------------
2845    
2846     1. The functions are now thread-safe, with the caveat that the global variables
2847     containing pointers to malloc() and free() or alternative functions are the
2848     same for all threads.
2849    
2850     2. Get pcre_study() to generate a bitmap of initial characters for non-
2851     anchored patterns when this is possible, and use it if passed to pcre_exec().
2852    
2853    
2854     Version 0.93 15-Sep-97
2855     ----------------------
2856    
2857     1. /(b)|(:+)/ was computing an incorrect first character.
2858    
2859     2. Add pcre_study() to the API and the passing of pcre_extra to pcre_exec(),
2860     but not actually doing anything yet.
2861    
2862     3. Treat "-" characters in classes that cannot be part of ranges as literals,
2863     as Perl does (e.g. [-az] or [az-]).
2864    
2865     4. Set the anchored flag if a branch starts with .* or .*? because that tests
2866     all possible positions.
2867    
2868     5. Split up into different modules to avoid including unneeded functions in a
2869     compiled binary. However, compile and exec are still in one module. The "study"
2870     function is split off.
2871    
2872     6. The character tables are now in a separate module whose source is generated
2873     by an auxiliary program - but can then be edited by hand if required. There are
2874     now no calls to isalnum(), isspace(), isdigit(), isxdigit(), tolower() or
2875     toupper() in the code.
2876    
2877     7. Turn the malloc/free funtions variables into pcre_malloc and pcre_free and
2878     make them global. Abolish the function for setting them, as the caller can now
2879     set them directly.
2880    
2881    
2882     Version 0.92 11-Sep-97
2883     ----------------------
2884    
2885     1. A repeat with a fixed maximum and a minimum of 1 for an ordinary character
2886     (e.g. /a{1,3}/) was broken (I mis-optimized it).
2887    
2888     2. Caseless matching was not working in character classes if the characters in
2889     the pattern were in upper case.
2890    
2891     3. Make ranges like [W-c] work in the same way as Perl for caseless matching.
2892    
2893     4. Make PCRE_ANCHORED public and accept as a compile option.
2894    
2895     5. Add an options word to pcre_exec() and accept PCRE_ANCHORED and
2896     PCRE_CASELESS at run time. Add escapes \A and \I to pcretest to cause it to
2897     pass them.
2898    
2899     6. Give an error if bad option bits passed at compile or run time.
2900    
2901     7. Add PCRE_MULTILINE at compile and exec time, and (?m) as well. Add \M to
2902     pcretest to cause it to pass that flag.
2903    
2904     8. Add pcre_info(), to get the number of identifying subpatterns, the stored
2905     options, and the first character, if set.
2906    
2907     9. Recognize C+ or C{n,m} where n >= 1 as providing a fixed starting character.
2908    
2909    
2910     Version 0.91 10-Sep-97
2911     ----------------------
2912    
2913     1. PCRE was failing to diagnose unlimited repeats of subpatterns that could
2914     match the empty string as in /(a*)*/. It was looping and ultimately crashing.
2915    
2916     2. PCRE was looping on encountering an indefinitely repeated back reference to
2917     a subpattern that had matched an empty string, e.g. /(a|)\1*/. It now does what
2918     Perl does - treats the match as successful.
2919    
2920     ****

Properties

Name Value
svn:eol-style native
svn:keywords "Author Date Id Revision Url"

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12