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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 232 - (hide annotations) (download)
Tue Sep 11 11:18:50 2007 UTC (7 years ago) by ph10
File size: 139879 byte(s)
doc/pcresyntax.html was missing from the distribution.

1 nigel 3 ChangeLog for PCRE
2     ------------------
3    
4 ph10 230 Version 7.4 10-Sep-07
5     ---------------------
6    
7     1. Change 7.3/28 was implemented for classes by looking at the bitmap. This
8 ph10 231 means that a class such as [\s] counted as "explicit reference to CR or
9     LF". That isn't really right - the whole point of the change was to try to
10     help when there was an actual mention of one of the two characters. So now
11     the change happens only if \r or \n (or a literal CR or LF) character is
12 ph10 230 encountered.
13 ph10 231
14     2. The 32-bit options word was also used for 6 internal flags, but the numbers
15     of both had grown to the point where there were only 3 bits left.
16     Fortunately, there was spare space in the data structure, and so I have
17     moved the internal flags into a new 16-bit field to free up more option
18 ph10 230 bits.
19 ph10 231
20     3. The appearance of (?J) at the start of a pattern set the DUPNAMES option,
21     but did not set the internal JCHANGED flag - either of these is enough to
22     control the way the "get" function works - but the PCRE_INFO_JCHANGED
23     facility is supposed to tell if (?J) was ever used, so now (?J) at the
24 ph10 230 start sets both bits.
25    
26 ph10 231 4. Added options (at build time, compile time, exec time) to change \R from
27     matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
28 ph10 232
29     5. doc/pcresyntax.html was missing from the distribution.
30 ph10 230
31 ph10 231
32 ph10 228 Version 7.3 28-Aug-07
33 ph10 157 ---------------------
34    
35 ph10 189 1. In the rejigging of the build system that eventually resulted in 7.1, the
36     line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
37     brackets there is not right, since it causes compilers to look for an
38     installed pcre.h, not the version that is in the source that is being
39     compiled (which of course may be different). I have changed it back to:
40 ph10 197
41 ph10 189 #include "pcre.h"
42 ph10 197
43     I have a vague recollection that the change was concerned with compiling in
44     different directories, but in the new build system, that is taken care of
45     by the VPATH setting the Makefile.
46    
47 ph10 190 2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
48     when the subject happened to end in the byte 0x85 (e.g. if the last
49     character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
50     characters but of course it shouldn't be taken as a newline when it is part
51     of another character. The bug was that, for an unlimited repeat of . in
52     not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
53     characters when looking for a newline.
54 ph10 197
55     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
56    
57     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
58     in debug output.
59    
60 ph10 201 5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
61     long printing in the pcrecpp unittest when running under MinGW.
62 ph10 189
63 ph10 197 6. ESC_K was left out of the EBCDIC table.
64 ph10 189
65 ph10 197 7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
66     parentheses; I made it 1000, which seemed large enough. Unfortunately, the
67     limit also applies to "virtual nesting" when a pattern is recursive, and in
68     this case 1000 isn't so big. I have been able to remove this limit at the
69     expense of backing off one optimization in certain circumstances. Normally,
70     when pcre_exec() would call its internal match() function recursively and
71     immediately return the result unconditionally, it uses a "tail recursion"
72     feature to save stack. However, when a subpattern that can match an empty
73     string has an unlimited repetition quantifier, it no longer makes this
74     optimization. That gives it a stack frame in which to save the data for
75     checking that an empty string has been matched. Previously this was taken
76     from the 1000-entry workspace that had been reserved. So now there is no
77     explicit limit, but more stack is used.
78 ph10 189
79 ph10 200 8. Applied Daniel's patches to solve problems with the import/export magic
80     syntax that is required for Windows, and which was going wrong for the
81     pcreposix and pcrecpp parts of the library. These were overlooked when this
82     problem was solved for the main library.
83 ph10 197
84 ph10 202 9. There were some crude static tests to avoid integer overflow when computing
85     the size of patterns that contain repeated groups with explicit upper
86     limits. As the maximum quantifier is 65535, the maximum group length was
87     set at 30,000 so that the product of these two numbers did not overflow a
88     32-bit integer. However, it turns out that people want to use groups that
89     are longer than 30,000 bytes (though not repeat them that many times).
90     Change 7.0/17 (the refactoring of the way the pattern size is computed) has
91     made it possible to implement the integer overflow checks in a much more
92     dynamic way, which I have now done. The artificial limitation on group
93     length has been removed - we now have only the limit on the total length of
94     the compiled pattern, which depends on the LINK_SIZE setting.
95 ph10 208
96     10. Fixed a bug in the documentation for get/copy named substring when
97     duplicate names are permitted. If none of the named substrings are set, the
98     functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
99     empty string.
100    
101     11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
102     instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
103     because the ] is interpreted as the first data character and the
104     terminating ] is not found. PCRE has been made compatible with Perl in this
105     regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
106     cause memory overwriting.
107    
108 ph10 206 10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
109     string has been matched (to stop an infinite loop). It was not recognizing
110 ph10 208 a conditional subpattern that could match an empty string if that
111 ph10 206 subpattern was within another subpattern. For example, it looped when
112 ph10 208 trying to match (((?(1)X|))*) but it was OK with ((?(1)X|)*) where the
113 ph10 206 condition was not nested. This bug has been fixed.
114 ph10 208
115 ph10 207 12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
116     past the start of the subject in the presence of bytes with the top bit
117     set, for example "\x8aBCD".
118 ph10 211
119 ph10 210 13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
120     (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
121 ph10 200
122 ph10 211 14. Optimized (?!) to (*FAIL).
123 ph10 202
124 ph10 212 15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
125     This restricts code points to be within the range 0 to 0x10FFFF, excluding
126     the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
127     full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
128     does: it's just the validity check that is more restrictive.
129 ph10 220
130     16. Inserted checks for integer overflows during escape sequence (backslash)
131     processing, and also fixed erroneous offset values for syntax errors during
132     backslash processing.
133    
134 ph10 214 17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
135 ph10 220 for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
136    
137 ph10 215 18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
138     caused an overrun.
139 ph10 220
140     19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
141     something other than just ASCII characters) inside a group that had an
142     unlimited repeat caused a loop at compile time (while checking to see
143     whether the group could match an empty string).
144    
145 ph10 217 20. Debugging a pattern containing \p or \P could cause a crash. For example,
146     [\P{Any}] did so. (Error in the code for printing property names.)
147 ph10 210
148 ph10 220 21. An orphan \E inside a character class could cause a crash.
149    
150     22. A repeated capturing bracket such as (A)? could cause a wild memory
151 ph10 218 reference during compilation.
152 ph10 220
153     23. There are several functions in pcre_compile() that scan along a compiled
154     expression for various reasons (e.g. to see if it's fixed length for look
155 ph10 218 behind). There were bugs in these functions when a repeated \p or \P was
156 ph10 220 present in the pattern. These operators have additional parameters compared
157     with \d, etc, and these were not being taken into account when moving along
158 ph10 218 the compiled data. Specifically:
159 ph10 220
160     (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
161     length.
162    
163     (b) An item such as \pL+ within a repeated group could cause crashes or
164 ph10 218 loops.
165 ph10 220
166 ph10 218 (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
167 ph10 220 "reference to non-existent subpattern" error.
168    
169 ph10 221 (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
170    
171 ph10 219 24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
172 ph10 220 characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
173 ph10 211
174 ph10 222 25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
175 ph10 218
176 ph10 227 26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
177 ph10 223 character were causing crashes (broken optimization).
178 ph10 227
179 ph10 224 27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
180 ph10 227 \p or \P) caused a compile-time loop.
181    
182 ph10 226 28. More problems have arisen in unanchored patterns when CRLF is a valid line
183     break. For example, the unstudied pattern [\r\n]A does not match the string
184     "\r\nA" because change 7.0/46 below moves the current point on by two
185     characters after failing to match at the start. However, the pattern \nA
186 ph10 227 *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
187     the same is true. There doesn't seem any very clean way out of this, but
188     what I have chosen to do makes the common cases work: PCRE now takes note
189     of whether there can be an explicit match for \r or \n anywhere in the
190     pattern, and if so, 7.0/46 no longer applies. As part of this change,
191     there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
192     pattern has explicit CR or LF references.
193 ph10 222
194 ph10 227 29. Added (*CR) etc for changing newline setting at start of pattern.
195 ph10 223
196 ph10 227
197 ph10 189 Version 7.2 19-Jun-07
198     ---------------------
199    
200 ph10 157 1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
201     which is apparently normally available under Windows.
202 ph10 159
203     2. Re-jig the pcregrep tests with different newline settings in an attempt
204     to make them independent of the local environment's newline setting.
205 ph10 157
206 ph10 160 3. Add code to configure.ac to remove -g from the CFLAGS default settings.
207 ph10 165
208 ph10 161 4. Some of the "internals" tests were previously cut out when the link size
209     was not 2, because the output contained actual offsets. The recent new
210 ph10 165 "Z" feature of pcretest means that these can be cut out, making the tests
211     usable with all link sizes.
212    
213 ph10 164 5. Implemented Stan Switzer's goto replacement for longjmp() when not using
214     stack recursion. This gives a massive performance boost under BSD, but just
215 ph10 165 a small improvement under Linux. However, it saves one field in the frame
216 ph10 164 in all cases.
217 ph10 172
218 ph10 166 6. Added more features from the forthcoming Perl 5.10:
219 ph10 172
220 ph10 166 (a) (?-n) (where n is a string of digits) is a relative subroutine or
221     recursion call. It refers to the nth most recently opened parentheses.
222 ph10 172
223 ph10 166 (b) (?+n) is also a relative subroutine call; it refers to the nth next
224 ph10 172 to be opened parentheses.
225    
226     (c) Conditions that refer to capturing parentheses can be specified
227 ph10 167 relatively, for example, (?(-2)... or (?(+3)...
228 ph10 172
229 ph10 168 (d) \K resets the start of the current match so that everything before
230 ph10 172 is not part of it.
231    
232 ph10 171 (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
233 ph10 172
234 ph10 171 (f) \g{name} is another synonym - part of Perl 5.10's unification of
235 ph10 172 reference syntax.
236 ph10 182
237 ph10 175 (g) (?| introduces a group in which the numbering of parentheses in each
238 ph10 182 alternative starts with the same number.
239 ph10 172
240 ph10 182 (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
241    
242 ph10 172 7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
243     PCRE_INFO_JCHANGED.
244    
245     8. A pattern such as (.*(.)?)* caused pcre_exec() to fail by either not
246     terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
247 ph10 170 for detecting groups that can match an empty string.
248 ph10 159
249 ph10 172 9. A pattern with a very large number of alternatives (more than several
250     hundred) was running out of internal workspace during the pre-compile
251     phase, where pcre_compile() figures out how much memory will be needed. A
252     bit of new cunning has reduced the workspace needed for groups with
253     alternatives. The 1000-alternative test pattern now uses 12 bytes of
254     workspace instead of running out of the 4096 that are available.
255 ph10 182
256 ph10 176 10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
257 ph10 172
258 ph10 179 11. Applied patch from Google to remove an optimization that didn't quite work.
259     The report of the bug said:
260 ph10 182
261 ph10 179 pcrecpp::RE("a*").FullMatch("aaa") matches, while
262     pcrecpp::RE("a*?").FullMatch("aaa") does not, and
263 ph10 182 pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
264 ph10 185
265 ph10 184 12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
266 ph10 185 it matched the wrong number of bytes.
267 ph10 172
268 ph10 179
269 ph10 155 Version 7.1 24-Apr-07
270 ph10 98 ---------------------
271    
272 ph10 111 1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
273 ph10 122 that is more "standard", making use of automake and other Autotools. There
274 ph10 99 is some re-arrangement of the files and adjustment of comments consequent
275     on this.
276 ph10 111
277     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
278     for recursive directory scanning broke on some systems because the files
279     are not scanned in any specific order and on different systems the order
280     was different. A call to "sort" has been inserted into RunGrepTest for the
281     approprate test as a short-term fix. In the longer term there may be an
282 ph10 100 alternative.
283 ph10 111
284 ph10 100 3. I had an email from Eric Raymond about problems translating some of PCRE's
285 ph10 111 man pages to HTML (despite the fact that I distribute HTML pages, some
286     people do their own conversions for various reasons). The problems
287     concerned the use of low-level troff macros .br and .in. I have therefore
288     removed all such uses from the man pages (some were redundant, some could
289 ph10 113 be replaced by .nf/.fi pairs). The 132html script that I use to generate
290     HTML has been updated to handle .nf/.fi and to complain if it encounters
291     .br or .in.
292 ph10 111
293 ph10 100 4. Updated comments in configure.ac that get placed in config.h.in and also
294 ph10 123 arranged for config.h to be included in the distribution, with the name
295 ph10 111 config.h.generic, for the benefit of those who have to compile without
296     Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
297    
298     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
299     Weber: (1) pcre_internal.h was missing some function renames; (2) updated
300 ph10 127 makevp.bat for the current PCRE, using the additional files
301 ph10 135 makevp_c.txt, makevp_l.txt, and pcregexp.pas.
302 ph10 111
303     6. A Windows user reported a minor discrepancy with test 2, which turned out
304     to be caused by a trailing space on an input line that had got lost in his
305 ph10 102 copy. The trailing space was an accident, so I've just removed it.
306 ph10 111
307 ph10 104 7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
308 ph10 111 that is needed.
309    
310 ph10 105 8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
311     as "const" (a) because they are and (b) because it helps the PHP
312     maintainers who have recently made a script to detect big data structures
313 ph10 111 in the php code that should be moved to the .rodata section. I remembered
314     to update Builducptable as well, so it won't revert if ucptable.h is ever
315 ph10 105 re-created.
316 ph10 111
317     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
318 ph10 107 pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
319 ph10 111 order to be able to cut out the UTF-8 tables in the latter when UTF-8
320     support is not required. This saves 1.5-2K of code, which is important in
321 ph10 107 some applications.
322 ph10 111
323 ph10 110 Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
324 ph10 111 so as not to refer to the tables, even though these functions will never be
325     called when UTF-8 support is disabled. Otherwise there are problems with a
326     shared library.
327    
328 ph10 118 10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
329    
330     (a) It was defining its arguments as char * instead of void *.
331    
332 ph10 123 (b) It was assuming that all moves were upwards in memory; this was true
333     a long time ago when I wrote it, but is no longer the case.
334    
335 ph10 118 The emulated memove() is provided for those environments that have neither
336 ph10 123 memmove() nor bcopy(). I didn't think anyone used it these days, but that
337 ph10 118 is clearly not the case, as these two bugs were recently reported.
338 ph10 123
339 ph10 111 11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
340 ph10 123 and Detrail to create the HTML documentation, the .txt form of the man
341     pages, and it removes trailing spaces from listed files. It also creates
342     pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
343     case, it wraps all the #defines with #ifndefs. This script should be run
344 ph10 111 before "make dist".
345 ph10 123
346 ph10 115 12. Fixed two fairly obscure bugs concerned with quantified caseless matching
347     with Unicode property support.
348 ph10 123
349     (a) For a maximizing quantifier, if the two different cases of the
350     character were of different lengths in their UTF-8 codings (there are
351     some cases like this - I found 11), and the matching function had to
352 ph10 115 back up over a mixture of the two cases, it incorrectly assumed they
353     were both the same length.
354 ph10 123
355     (b) When PCRE was configured to use the heap rather than the stack for
356     recursion during matching, it was not correctly preserving the data for
357     the other case of a UTF-8 character when checking ahead for a match
358     while processing a minimizing repeat. If the check also involved
359 ph10 115 matching a wide character, but failed, corruption could cause an
360     erroneous result when trying to check for a repeat of the original
361     character.
362 ph10 123
363 ph10 116 13. Some tidying changes to the testing mechanism:
364 ph10 98
365 ph10 116 (a) The RunTest script now detects the internal link size and whether there
366     is UTF-8 and UCP support by running ./pcretest -C instead of relying on
367 ph10 123 values substituted by "configure". (The RunGrepTest script already did
368     this for UTF-8.) The configure.ac script no longer substitutes the
369     relevant variables.
370    
371 ph10 116 (b) The debugging options /B and /D in pcretest show the compiled bytecode
372     with length and offset values. This means that the output is different
373     for different internal link sizes. Test 2 is skipped for link sizes
374     other than 2 because of this, bypassing the problem. Unfortunately,
375     there was also a test in test 3 (the locale tests) that used /B and
376 ph10 123 failed for link sizes other than 2. Rather than cut the whole test out,
377     I have added a new /Z option to pcretest that replaces the length and
378     offset values with spaces. This is now used to make test 3 independent
379 ph10 122 of link size. (Test 2 will be tidied up later.)
380 ph10 123
381     14. If erroroffset was passed as NULL to pcre_compile, it provoked a
382 ph10 122 segmentation fault instead of returning the appropriate error message.
383 ph10 134
384 ph10 131 15. In multiline mode when the newline sequence was set to "any", the pattern
385 ph10 134 ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
386     This doesn't seem right; it now treats the CRLF combination as the line
387     ending, and so does not match in that case. It's only a pattern such as ^$
388     that would hit this one: something like ^ABC$ would have failed after \r
389     and then tried again after \r\n.
390    
391 ph10 131 16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
392 ph10 134 in an attempt to make files that differ only in their line terminators
393     compare equal. This works on Linux.
394 ph10 142
395 ph10 141 17. Under certain error circumstances pcregrep might try to free random memory
396     as it exited. This is now fixed, thanks to valgrind.
397 ph10 142
398 ph10 141 19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
399 ph10 142 "abc\r\n\r\n", it found an unwanted second match after the second \r. This
400     was because its rules for how to advance for /g after matching an empty
401 ph10 143 string at the end of a line did not allow for this case. They now check for
402     it specially.
403 ph10 150
404     20. pcretest is supposed to handle patterns and data of any length, by
405     extending its buffers when necessary. It was getting this wrong when the
406 ph10 147 buffer for a data line had to be extended.
407 ph10 150
408 ph10 149 21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
409 ph10 150 CRLF as a newline sequence.
410 ph10 152
411 ph10 151 22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
412 ph10 152 out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
413     I have nevertheless tidied it up.
414 ph10 111
415 ph10 152 23. Added some casts to kill warnings from HP-UX ia64 compiler.
416 ph10 123
417 ph10 153 24. Added a man page for pcre-config.
418 ph10 152
419 ph10 153
420 nigel 93 Version 7.0 19-Dec-06
421     ---------------------
422    
423     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
424     moving to gcc 4.1.1.
425    
426     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
427     sys/time.h, which is documented as needed for this function. It doesn't
428     seem to matter on Linux, but it showed up on some releases of OS X.
429    
430     3. It seems that there are systems where bytes whose values are greater than
431     127 match isprint() in the "C" locale. The "C" locale should be the
432     default when a C program starts up. In most systems, only ASCII printing
433     characters match isprint(). This difference caused the output from pcretest
434     to vary, making some of the tests fail. I have changed pcretest so that:
435    
436     (a) When it is outputting text in the compiled version of a pattern, bytes
437     other than 32-126 are always shown as hex escapes.
438    
439     (b) When it is outputting text that is a matched part of a subject string,
440     it does the same, unless a different locale has been set for the match
441     (using the /L modifier). In this case, it uses isprint() to decide.
442    
443     4. Fixed a major bug that caused incorrect computation of the amount of memory
444     required for a compiled pattern when options that changed within the
445     pattern affected the logic of the preliminary scan that determines the
446     length. The relevant options are -x, and -i in UTF-8 mode. The result was
447     that the computed length was too small. The symptoms of this bug were
448     either the PCRE error "internal error: code overflow" from pcre_compile(),
449     or a glibc crash with a message such as "pcretest: free(): invalid next
450     size (fast)". Examples of patterns that provoked this bug (shown in
451     pcretest format) are:
452    
453     /(?-x: )/x
454     /(?x)(?-x: \s*#\s*)/
455     /((?i)[\x{c0}])/8
456     /(?i:[\x{c0}])/8
457    
458     HOWEVER: Change 17 below makes this fix obsolete as the memory computation
459     is now done differently.
460    
461     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
462     wrapper classes; (b) implement a new function in the C++ scanner that is
463     more efficient than the old way of doing things because it avoids levels of
464     recursion in the regex matching; (c) add a paragraph to the documentation
465     for the FullMatch() function.
466    
467     6. The escape sequence \n was being treated as whatever was defined as
468     "newline". Not only was this contrary to the documentation, which states
469     that \n is character 10 (hex 0A), but it also went horribly wrong when
470     "newline" was defined as CRLF. This has been fixed.
471    
472     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
473     was being set to -1 for the "end of line" case (supposedly a value that no
474     character can have). Though this value is never used (the check for end of
475     line is "zero bytes in current character"), it caused compiler complaints.
476     I've changed it to 0xffffffff.
477    
478     8. In pcre_version.c, the version string was being built by a sequence of
479     C macros that, in the event of PCRE_PRERELEASE being defined as an empty
480     string (as it is for production releases) called a macro with an empty
481     argument. The C standard says the result of this is undefined. The gcc
482     compiler treats it as an empty string (which was what was wanted) but it is
483     reported that Visual C gives an error. The source has been hacked around to
484     avoid this problem.
485    
486     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
487     builds of pcretest, and changed the call to _setmode() to use _O_BINARY
488     instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
489     of them did).
490    
491     10. Originally, pcretest opened its input and output without "b"; then I was
492     told that "b" was needed in some environments, so it was added for release
493     5.0 to both the input and output. (It makes no difference on Unix-like
494     systems.) Later I was told that it is wrong for the input on Windows. I've
495     now abstracted the modes into two macros, to make it easier to fiddle with
496     them, and removed "b" from the input mode under Windows.
497    
498     11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
499    
500     12. Added -help and --help to pcretest as an official way of being reminded
501     of the options.
502    
503     13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
504     and pcrecpp.cc because they annoy compilers at high warning levels.
505    
506     14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
507    
508     15. Fixed an occurrence of == in configure.ac that should have been = (shell
509     scripts are not C programs :-) and which was not noticed because it works
510     on Linux.
511    
512     16. pcretest is supposed to handle any length of pattern and data line (as one
513     line or as a continued sequence of lines) by extending its input buffer if
514     necessary. This feature was broken for very long pattern lines, leading to
515     a string of junk being passed to pcre_compile() if the pattern was longer
516     than about 50K.
517    
518     17. I have done a major re-factoring of the way pcre_compile() computes the
519     amount of memory needed for a compiled pattern. Previously, there was code
520     that made a preliminary scan of the pattern in order to do this. That was
521     OK when PCRE was new, but as the facilities have expanded, it has become
522     harder and harder to keep it in step with the real compile phase, and there
523     have been a number of bugs (see for example, 4 above). I have now found a
524     cunning way of running the real compile function in a "fake" mode that
525     enables it to compute how much memory it would need, while actually only
526     ever using a few hundred bytes of working memory and without too many
527     tests of the mode. This should make future maintenance and development
528     easier. A side effect of this work is that the limit of 200 on the nesting
529     depth of parentheses has been removed (though this was never a serious
530     limitation, I suspect). However, there is a downside: pcre_compile() now
531     runs more slowly than before (30% or more, depending on the pattern). I
532     hope this isn't a big issue. There is no effect on runtime performance.
533    
534     18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
535     newline (only possible for the last line of a file) and it was a
536     pattern that set a locale (followed by /Lsomething), pcretest crashed.
537    
538     19. Added additional timing features to pcretest. (1) The -tm option now times
539     matching only, not compiling. (2) Both -t and -tm can be followed, as a
540     separate command line item, by a number that specifies the number of
541     repeats to use when timing. The default is 50000; this gives better
542     precision, but takes uncomfortably long for very large patterns.
543    
544     20. Extended pcre_study() to be more clever in cases where a branch of a
545     subpattern has no definite first character. For example, (a*|b*)[cd] would
546     previously give no result from pcre_study(). Now it recognizes that the
547     first character must be a, b, c, or d.
548    
549     21. There was an incorrect error "recursive call could loop indefinitely" if
550     a subpattern (or the entire pattern) that was being tested for matching an
551     empty string contained only one non-empty item after a nested subpattern.
552     For example, the pattern (?>\x{100}*)\d(?R) provoked this error
553     incorrectly, because the \d was being skipped in the check.
554    
555     22. The pcretest program now has a new pattern option /B and a command line
556     option -b, which is equivalent to adding /B to every pattern. This causes
557     it to show the compiled bytecode, without the additional information that
558     -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
559     is the same as /B/I).
560    
561     23. A new optimization is now able automatically to treat some sequences such
562     as a*b as a*+b. More specifically, if something simple (such as a character
563     or a simple class like \d) has an unlimited quantifier, and is followed by
564     something that cannot possibly match the quantified thing, the quantifier
565     is automatically "possessified".
566    
567     24. A recursive reference to a subpattern whose number was greater than 39
568     went wrong under certain circumstances in UTF-8 mode. This bug could also
569     have affected the operation of pcre_study().
570    
571     25. Realized that a little bit of performance could be had by replacing
572     (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
573    
574     26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
575    
576     27. Possessive quantifiers such as a++ were previously implemented by turning
577     them into atomic groups such as ($>a+). Now they have their own opcodes,
578     which improves performance. This includes the automatically created ones
579     from 23 above.
580    
581     28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
582     lookahead was broken if it was not anchored. PCRE was mistakenly expecting
583     the first matched character to be a colon. This applied both to named and
584     numbered groups.
585    
586     29. The ucpinternal.h header file was missing its idempotency #ifdef.
587    
588     30. I was sent a "project" file called libpcre.a.dev which I understand makes
589     building PCRE on Windows easier, so I have included it in the distribution.
590    
591     31. There is now a check in pcretest against a ridiculously large number being
592     returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
593     loop, the loop is abandoned.
594    
595     32. Forward references to subpatterns in conditions such as (?(2)...) where
596     subpattern 2 is defined later cause pcre_compile() to search forwards in
597     the pattern for the relevant set of parentheses. This search went wrong
598     when there were unescaped parentheses in a character class, parentheses
599     escaped with \Q...\E, or parentheses in a #-comment in /x mode.
600    
601     33. "Subroutine" calls and backreferences were previously restricted to
602     referencing subpatterns earlier in the regex. This restriction has now
603     been removed.
604    
605     34. Added a number of extra features that are going to be in Perl 5.10. On the
606     whole, these are just syntactic alternatives for features that PCRE had
607     previously implemented using the Python syntax or my own invention. The
608     other formats are all retained for compatibility.
609    
610     (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
611     as (?P<name>...). The new forms, as well as being in Perl 5.10, are
612     also .NET compatible.
613    
614     (b) A recursion or subroutine call to a named group can now be defined as
615     (?&name) as well as (?P>name).
616    
617     (c) A backreference to a named group can now be defined as \k<name> or
618     \k'name' as well as (?P=name). The new forms, as well as being in Perl
619     5.10, are also .NET compatible.
620    
621     (d) A conditional reference to a named group can now use the syntax
622     (?(<name>) or (?('name') as well as (?(name).
623    
624     (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
625     groups (named and numbered) that are never evaluated inline, but can be
626     called as "subroutines" from elsewhere. In effect, the DEFINE condition
627     is always false. There may be only one alternative in such a group.
628    
629     (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
630     as the simple (?(R). The condition is true only if the most recent
631     recursion is that of the given number or name. It does not search out
632     through the entire recursion stack.
633    
634     (g) The escape \gN or \g{N} has been added, where N is a positive or
635     negative number, specifying an absolute or relative reference.
636    
637     35. Tidied to get rid of some further signed/unsigned compiler warnings and
638     some "unreachable code" warnings.
639    
640     36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
641     things, this adds five new scripts.
642    
643     37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
644     There were also incompatibilities regarding the handling of \Q..\E inside
645     character classes, for example with patterns like [\Qa\E-\Qz\E] where the
646     hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
647    
648     38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
649     matches an empty string, and forcibly breaks the loop. There were bugs in
650     this code in non-simple cases. For a pattern such as ^(a()*)* matched
651     against aaaa the result was just "a" rather than "aaaa", for example. Two
652     separate and independent bugs (that affected different cases) have been
653     fixed.
654    
655     39. Refactored the code to abolish the use of different opcodes for small
656     capturing bracket numbers. This is a tidy that I avoided doing when I
657     removed the limit on the number of capturing brackets for 3.5 back in 2001.
658     The new approach is not only tidier, it makes it possible to reduce the
659     memory needed to fix the previous bug (38).
660    
661     40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
662     sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
663     processing dot, circumflex, or dollar metacharacters, or #-comments in /x
664     mode.
665    
666     41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
667     report.
668    
669     42. Applied patch, originally from Ari Pollak, modified by Google, to allow
670     copy construction and assignment in the C++ wrapper.
671    
672     43. Updated pcregrep to support "--newline=any". In the process, I fixed a
673     couple of bugs that could have given wrong results in the "--newline=crlf"
674     case.
675    
676     44. Added a number of casts and did some reorganization of signed/unsigned int
677     variables following suggestions from Dair Grant. Also renamed the variable
678     "this" as "item" because it is a C++ keyword.
679    
680     45. Arranged for dftables to add
681    
682     #include "pcre_internal.h"
683    
684     to pcre_chartables.c because without it, gcc 4.x may remove the array
685     definition from the final binary if PCRE is built into a static library and
686     dead code stripping is activated.
687    
688     46. For an unanchored pattern, if a match attempt fails at the start of a
689     newline sequence, and the newline setting is CRLF or ANY, and the next two
690     characters are CRLF, advance by two characters instead of one.
691    
692    
693 nigel 91 Version 6.7 04-Jul-06
694     ---------------------
695    
696     1. In order to handle tests when input lines are enormously long, pcretest has
697     been re-factored so that it automatically extends its buffers when
698     necessary. The code is crude, but this _is_ just a test program. The
699     default size has been increased from 32K to 50K.
700    
701     2. The code in pcre_study() was using the value of the re argument before
702     testing it for NULL. (Of course, in any sensible call of the function, it
703     won't be NULL.)
704    
705     3. The memmove() emulation function in pcre_internal.h, which is used on
706     systems that lack both memmove() and bcopy() - that is, hardly ever -
707     was missing a "static" storage class specifier.
708    
709     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
710     containing an extended class (one that cannot be represented by a bitmap
711     because it contains high-valued characters or Unicode property items, e.g.
712     [\pZ]). Almost always one would set UTF-8 mode when processing such a
713     pattern, but PCRE should not loop if you do not (it no longer does).
714     [Detail: two cases were found: (a) a repeated subpattern containing an
715     extended class; (b) a recursive reference to a subpattern that followed a
716     previous extended class. It wasn't skipping over the extended class
717     correctly when UTF-8 mode was not set.]
718    
719     5. A negated single-character class was not being recognized as fixed-length
720     in lookbehind assertions such as (?<=[^f]), leading to an incorrect
721     compile error "lookbehind assertion is not fixed length".
722    
723     6. The RunPerlTest auxiliary script was showing an unexpected difference
724     between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
725     write a Perl script that can interpret lines of an input file either as
726     byte characters or as UTF-8, which is what "perltest" was being required to
727     do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
728     can't do is switch easily at run time between having the "use utf8;" pragma
729     or not. In the end, I fudged it by using the RunPerlTest script to insert
730     "use utf8;" explicitly for the UTF-8 tests.
731    
732     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
733     the end of the subject string, contrary to the documentation and to what
734     Perl does. This was true of both matching functions. Now it matches only at
735     the start of the subject and immediately after *internal* newlines.
736    
737     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
738     a pointer to an int instead of a pointer to an unsigned long int. This
739     caused problems on 64-bit systems.
740    
741     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
742     instance of the 'standard' template library not being so standard".
743    
744     10. There was no check on the number of named subpatterns nor the maximum
745     length of a subpattern name. The product of these values is used to compute
746     the size of the memory block for a compiled pattern. By supplying a very
747     long subpattern name and a large number of named subpatterns, the size
748     computation could be caused to overflow. This is now prevented by limiting
749     the length of names to 32 characters, and the number of named subpatterns
750     to 10,000.
751    
752     11. Subpatterns that are repeated with specific counts have to be replicated in
753     the compiled pattern. The size of memory for this was computed from the
754     length of the subpattern and the repeat count. The latter is limited to
755     65535, but there was no limit on the former, meaning that integer overflow
756     could in principle occur. The compiled length of a repeated subpattern is
757     now limited to 30,000 bytes in order to prevent this.
758    
759     12. Added the optional facility to have named substrings with the same name.
760    
761     13. Added the ability to use a named substring as a condition, using the
762     Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
763     are numbers (not recommended). Forward references are permitted.
764    
765     14. Added forward references in named backreferences (if you see what I mean).
766    
767     15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
768     pattern could run off the end of the subject. For example, the pattern
769     "(?s)(.{1,5})"8 did this with the subject "ab".
770    
771     16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
772     PCRE_CASELESS was set when matching characters that were quantified with ?
773     or *.
774    
775     17. A character class other than a single negated character that had a minimum
776     but no maximum quantifier - for example [ab]{6,} - was not handled
777     correctly by pce_dfa_exec(). It would match only one character.
778    
779     18. A valid (though odd) pattern that looked like a POSIX character
780     class but used an invalid character after [ (for example [[,abc,]]) caused
781     pcre_compile() to give the error "Failed: internal error: code overflow" or
782     in some cases to crash with a glibc free() error. This could even happen if
783     the pattern terminated after [[ but there just happened to be a sequence of
784     letters, a binary zero, and a closing ] in the memory that followed.
785    
786     19. Perl's treatment of octal escapes in the range \400 to \777 has changed
787     over the years. Originally (before any Unicode support), just the bottom 8
788     bits were taken. Thus, for example, \500 really meant \100. Nowadays the
789     output from "man perlunicode" includes this:
790    
791     The regular expression compiler produces polymorphic opcodes. That
792     is, the pattern adapts to the data and automatically switches to
793     the Unicode character scheme when presented with Unicode data--or
794     instead uses a traditional byte scheme when presented with byte
795     data.
796    
797     Sadly, a wide octal escape does not cause a switch, and in a string with
798     no other multibyte characters, these octal escapes are treated as before.
799     Thus, in Perl, the pattern /\500/ actually matches \100 but the pattern
800     /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
801     Unicode string.
802    
803     I have not perpetrated such confusion in PCRE. Up till now, it took just
804     the bottom 8 bits, as in old Perl. I have now made octal escapes with
805     values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
806     translate to the appropriate multibyte character.
807    
808     29. Applied some refactoring to reduce the number of warnings from Microsoft
809     and Borland compilers. This has included removing the fudge introduced
810     seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
811     a warning about an unused variable.
812    
813     21. PCRE has not included VT (character 0x0b) in the set of whitespace
814     characters since release 4.0, because Perl (from release 5.004) does not.
815     [Or at least, is documented not to: some releases seem to be in conflict
816     with the documentation.] However, when a pattern was studied with
817     pcre_study() and all its branches started with \s, PCRE still included VT
818     as a possible starting character. Of course, this did no harm; it just
819     caused an unnecessary match attempt.
820    
821     22. Removed a now-redundant internal flag bit that recorded the fact that case
822     dependency changed within the pattern. This was once needed for "required
823     byte" processing, but is no longer used. This recovers a now-scarce options
824     bit. Also moved the least significant internal flag bit to the most-
825     significant bit of the word, which was not previously used (hangover from
826     the days when it was an int rather than a uint) to free up another bit for
827     the future.
828    
829     23. Added support for CRLF line endings as well as CR and LF. As well as the
830     default being selectable at build time, it can now be changed at runtime
831     via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
832     specify that it is scanning data with non-default line endings.
833    
834     24. Changed the definition of CXXLINK to make it agree with the definition of
835     LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
836    
837     25. Applied Ian Taylor's patches to avoid using another stack frame for tail
838     recursions. This makes a big different to stack usage for some patterns.
839    
840     26. If a subpattern containing a named recursion or subroutine reference such
841     as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
842     the space required for the compiled pattern went wrong and gave too small a
843     value. Depending on the environment, this could lead to "Failed: internal
844     error: code overflow at offset 49" or "glibc detected double free or
845     corruption" errors.
846    
847     27. Applied patches from Google (a) to support the new newline modes and (b) to
848     advance over multibyte UTF-8 characters in GlobalReplace.
849    
850     28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
851     difference for some implementation of PCRE in some Windows version.
852    
853     29. Added some extra testing facilities to pcretest:
854    
855     \q<number> in a data line sets the "match limit" value
856     \Q<number> in a data line sets the "match recursion limt" value
857     -S <number> sets the stack size, where <number> is in megabytes
858    
859     The -S option isn't available for Windows.
860    
861    
862 nigel 89 Version 6.6 06-Feb-06
863     ---------------------
864    
865     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
866     in pcreposix.h. I have copied the definition from pcre.h.
867    
868     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
869     because pcre.h is no longer a built file.
870    
871     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
872     not normally included in the compiled code.
873    
874    
875 nigel 87 Version 6.5 01-Feb-06
876     ---------------------
877    
878     1. When using the partial match feature with pcre_dfa_exec(), it was not
879     anchoring the second and subsequent partial matches at the new starting
880     point. This could lead to incorrect results. For example, with the pattern
881     /1234/, partially matching against "123" and then "a4" gave a match.
882    
883     2. Changes to pcregrep:
884    
885     (a) All non-match returns from pcre_exec() were being treated as failures
886     to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
887     error message is output. Some extra information is given for the
888     PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
889     probably the only errors that are likely to be caused by users (by
890     specifying a regex that has nested indefinite repeats, for instance).
891     If there are more than 20 of these errors, pcregrep is abandoned.
892    
893     (b) A binary zero was treated as data while matching, but terminated the
894     output line if it was written out. This has been fixed: binary zeroes
895     are now no different to any other data bytes.
896    
897     (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
898     used to set a locale for matching. The --locale=xxxx long option has
899     been added (no short equivalent) to specify a locale explicitly on the
900     pcregrep command, overriding the environment variables.
901    
902     (d) When -B was used with -n, some line numbers in the output were one less
903     than they should have been.
904    
905     (e) Added the -o (--only-matching) option.
906    
907     (f) If -A or -C was used with -c (count only), some lines of context were
908     accidentally printed for the final match.
909    
910     (g) Added the -H (--with-filename) option.
911    
912     (h) The combination of options -rh failed to suppress file names for files
913     that were found from directory arguments.
914    
915     (i) Added the -D (--devices) and -d (--directories) options.
916    
917     (j) Added the -F (--fixed-strings) option.
918    
919     (k) Allow "-" to be used as a file name for -f as well as for a data file.
920    
921     (l) Added the --colo(u)r option.
922    
923     (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
924     is not present by default.
925    
926     3. A nasty bug was discovered in the handling of recursive patterns, that is,
927     items such as (?R) or (?1), when the recursion could match a number of
928     alternatives. If it matched one of the alternatives, but subsequently,
929     outside the recursion, there was a failure, the code tried to back up into
930     the recursion. However, because of the way PCRE is implemented, this is not
931     possible, and the result was an incorrect result from the match.
932    
933     In order to prevent this happening, the specification of recursion has
934     been changed so that all such subpatterns are automatically treated as
935     atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
936    
937     4. I had overlooked the fact that, in some locales, there are characters for
938     which isalpha() is true but neither isupper() nor islower() are true. In
939     the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
940     and ordfeminine) are like this. This affected the treatment of \w and \W
941     when they appeared in character classes, but not when they appeared outside
942     a character class. The bit map for "word" characters is now created
943     separately from the results of isalnum() instead of just taking it from the
944     upper, lower, and digit maps. (Plus the underscore character, of course.)
945    
946     5. The above bug also affected the handling of POSIX character classes such as
947     [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
948     permanent tables. Instead, the bit maps for such a class were previously
949     created as the appropriate unions of the upper, lower, and digit bitmaps.
950     Now they are created by subtraction from the [[:word:]] class, which has
951     its own bitmap.
952    
953     6. The [[:blank:]] character class matches horizontal, but not vertical space.
954     It is created by subtracting the vertical space characters (\x09, \x0a,
955     \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
956     subtraction was done in the overall bitmap for a character class, meaning
957     that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
958     be recognized. This bug has been fixed.
959    
960     7. Patches from the folks at Google:
961    
962     (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
963     real life, but is still worth protecting against".
964    
965     (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
966     regular expressions".
967    
968     (c) pcre_scanner.cc: avoid use of std::count() because not all systems
969     have it.
970    
971     (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
972     "configure" and the latter not, in order to fix a problem somebody had
973     with compiling the Arg class on HP-UX.
974    
975     (e) Improve the error-handling of the C++ wrapper a little bit.
976    
977     (f) New tests for checking recursion limiting.
978    
979     8. The pcre_memmove() function, which is used only if the environment does not
980     have a standard memmove() function (and is therefore rarely compiled),
981     contained two bugs: (a) use of int instead of size_t, and (b) it was not
982     returning a result (though PCRE never actually uses the result).
983    
984     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
985     large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
986     returned instead of calling malloc() with an overflowing number that would
987     most likely cause subsequent chaos.
988    
989     10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
990    
991     11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
992     with this option is matched, the nmatch and pmatch options of regexec() are
993     ignored.
994    
995     12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
996     provided in case anyone wants to the the POSIX interface with UTF-8
997     strings.
998    
999     13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1000     C++ linking (needed for some HP-UX environments).
1001    
1002     14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1003     (unused parameter) and in the pcre_printint() function (omitted "default"
1004     switch label when the default is to do nothing).
1005    
1006     15. Added some code to make it possible, when PCRE is compiled as a C++
1007     library, to replace subject pointers for pcre_exec() with a smart pointer
1008     class, thus making it possible to process discontinuous strings.
1009    
1010     16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1011     much the same function. They were added by different people who were trying
1012     to make PCRE easy to compile on non-Unix systems. It has been suggested
1013     that PCRE_EXPORT be abolished now that there is more automatic apparatus
1014     for compiling on Windows systems. I have therefore replaced it with
1015     PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1016     defaults to "extern" for C or "extern C" for C++, which works fine on
1017     Unix-like systems. It is now possible to override the value of PCRE_DATA_
1018     SCOPE with something explicit in config.h. In addition:
1019    
1020     (a) pcreposix.h still had just "extern" instead of either of these macros;
1021     I have replaced it with PCRE_DATA_SCOPE.
1022    
1023     (b) Functions such as _pcre_xclass(), which are internal to the library,
1024     but external in the C sense, all had PCRE_EXPORT in their definitions.
1025     This is apparently wrong for the Windows case, so I have removed it.
1026     (It makes no difference on Unix-like systems.)
1027    
1028     17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1029     of recursive calls to match(). This is different to MATCH_LIMIT because
1030     that limits the total number of calls to match(), not all of which increase
1031     the depth of recursion. Limiting the recursion depth limits the amount of
1032     stack (or heap if NO_RECURSE is set) that is used. The default can be set
1033     when PCRE is compiled, and changed at run time. A patch from Google adds
1034     this functionality to the C++ interface.
1035    
1036     18. Changes to the handling of Unicode character properties:
1037    
1038     (a) Updated the table to Unicode 4.1.0.
1039    
1040     (b) Recognize characters that are not in the table as "Cn" (undefined).
1041    
1042     (c) I revised the way the table is implemented to a much improved format
1043     which includes recognition of ranges. It now supports the ranges that
1044     are defined in UnicodeData.txt, and it also amalgamates other
1045     characters into ranges. This has reduced the number of entries in the
1046     table from around 16,000 to around 3,000, thus reducing its size
1047     considerably. I realized I did not need to use a tree structure after
1048     all - a binary chop search is just as efficient. Having reduced the
1049     number of entries, I extended their size from 6 bytes to 8 bytes to
1050     allow for more data.
1051    
1052     (d) Added support for Unicode script names via properties such as \p{Han}.
1053    
1054     19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1055     matching that character.
1056    
1057     20. When matching a repeated Unicode property with a minimum greater than zero,
1058     (for example \pL{2,}), PCRE could look past the end of the subject if it
1059     reached it while seeking the minimum number of characters. This could
1060     happen only if some of the characters were more than one byte long, because
1061     there is a check for at least the minimum number of bytes.
1062    
1063     21. Refactored the implementation of \p and \P so as to be more general, to
1064     allow for more different types of property in future. This has changed the
1065     compiled form incompatibly. Anybody with saved compiled patterns that use
1066     \p or \P will have to recompile them.
1067    
1068     22. Added "Any" and "L&" to the supported property types.
1069    
1070     23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1071     but give a compile time error if the value is greater than 0xff.
1072    
1073     24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1074     accidentally not being installed or uninstalled.
1075    
1076     25. The pcre.h file was built from pcre.h.in, but the only changes that were
1077     made were to insert the current release number. This seemed silly, because
1078     it made things harder for people building PCRE on systems that don't run
1079     "configure". I have turned pcre.h into a distributed file, no longer built
1080     by "configure", with the version identification directly included. There is
1081     no longer a pcre.h.in file.
1082    
1083     However, this change necessitated a change to the pcre-config script as
1084     well. It is built from pcre-config.in, and one of the substitutions was the
1085     release number. I have updated configure.ac so that ./configure now finds
1086     the release number by grepping pcre.h.
1087    
1088     26. Added the ability to run the tests under valgrind.
1089    
1090    
1091 nigel 85 Version 6.4 05-Sep-05
1092     ---------------------
1093    
1094     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1095     "--" to be printed when multiple files were scanned, even when none of the
1096     -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1097     consider it to be a bug, and have restored the previous behaviour.
1098    
1099     2. A couple of code tidies to get rid of compiler warnings.
1100    
1101     3. The pcretest program used to cheat by referring to symbols in the library
1102     whose names begin with _pcre_. These are internal symbols that are not
1103     really supposed to be visible externally, and in some environments it is
1104     possible to suppress them. The cheating is now confined to including
1105     certain files from the library's source, which is a bit cleaner.
1106    
1107     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1108     file's purpose clearer.
1109    
1110     5. Reorganized pcre_ucp_findchar().
1111    
1112    
1113 nigel 83 Version 6.3 15-Aug-05
1114     ---------------------
1115    
1116     1. The file libpcre.pc.in did not have general read permission in the tarball.
1117    
1118     2. There were some problems when building without C++ support:
1119    
1120     (a) If C++ support was not built, "make install" and "make test" still
1121     tried to test it.
1122    
1123     (b) There were problems when the value of CXX was explicitly set. Some
1124     changes have been made to try to fix these, and ...
1125    
1126     (c) --disable-cpp can now be used to explicitly disable C++ support.
1127    
1128     (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1129     backslash in a target when C++ was disabled. This confuses some
1130     versions of "make", apparently. Using an intermediate variable solves
1131     this. (Same for CPP_LOBJ.)
1132    
1133     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1134     (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1135     necessary on certain architectures.
1136    
1137     4. Added a setting of -export-symbols-regex to the link command to remove
1138     those symbols that are exported in the C sense, but actually are local
1139     within the library, and not documented. Their names all begin with
1140     "_pcre_". This is not a perfect job, because (a) we have to except some
1141     symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1142     available (and never for static libraries). I have made a note to try to
1143     find a way round (a) in the future.
1144    
1145    
1146 nigel 81 Version 6.2 01-Aug-05
1147     ---------------------
1148    
1149     1. There was no test for integer overflow of quantifier values. A construction
1150     such as {1111111111111111} would give undefined results. What is worse, if
1151     a minimum quantifier for a parenthesized subpattern overflowed and became
1152     negative, the calculation of the memory size went wrong. This could have
1153     led to memory overwriting.
1154    
1155     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1156    
1157     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1158     operating environments where this matters.
1159    
1160     4. Applied Giuseppe Maxia's patch to add additional features for controlling
1161     PCRE options from within the C++ wrapper.
1162    
1163     5. Named capturing subpatterns were not being correctly counted when a pattern
1164     was compiled. This caused two problems: (a) If there were more than 100
1165     such subpatterns, the calculation of the memory needed for the whole
1166     compiled pattern went wrong, leading to an overflow error. (b) Numerical
1167     back references of the form \12, where the number was greater than 9, were
1168     not recognized as back references, even though there were sufficient
1169     previous subpatterns.
1170    
1171     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1172     versions of gcc, e.g. 2.95.4.
1173    
1174    
1175 nigel 79 Version 6.1 21-Jun-05
1176     ---------------------
1177    
1178     1. There was one reference to the variable "posix" in pcretest.c that was not
1179     surrounded by "#if !defined NOPOSIX".
1180    
1181     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1182     the cross-check on the old pcre_info() function, for the benefit of the
1183     cut-down version of PCRE that is currently imported into Exim.
1184    
1185     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1186     allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1187     patterns that start (?i) but I don't think that matters. The use of (?i) is
1188     just an example; this all applies to the other options as well.
1189    
1190     4. Since libtool seems to echo the compile commands it is issuing, the output
1191     from "make" can be reduced a bit by putting "@" in front of each libtool
1192     compile command.
1193    
1194     5. Patch from the folks at Google for configure.in to be a bit more thorough
1195     in checking for a suitable C++ installation before trying to compile the
1196     C++ stuff. This should fix a reported problem when a compiler was present,
1197     but no suitable headers.
1198    
1199     6. The man pages all had just "PCRE" as their title. I have changed them to
1200     be the relevant file name. I have also arranged that these names are
1201     retained in the file doc/pcre.txt, which is a concatenation in text format
1202     of all the man pages except the little individual ones for each function.
1203    
1204     7. The NON-UNIX-USE file had not been updated for the different set of source
1205     files that come with release 6. I also added a few comments about the C++
1206     wrapper.
1207    
1208    
1209 nigel 77 Version 6.0 07-Jun-05
1210     ---------------------
1211    
1212     1. Some minor internal re-organization to help with my DFA experiments.
1213    
1214     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1215     didn't matter for the library itself when fully configured, but did matter
1216     when compiling without UCP support, or within Exim, where the ucp files are
1217     not imported.
1218    
1219     3. Refactoring of the library code to split up the various functions into
1220     different source modules. The addition of the new DFA matching code (see
1221     below) to a single monolithic source would have made it really too
1222     unwieldy, quite apart from causing all the code to be include in a
1223     statically linked application, when only some functions are used. This is
1224     relevant even without the DFA addition now that patterns can be compiled in
1225     one application and matched in another.
1226    
1227     The downside of splitting up is that there have to be some external
1228     functions and data tables that are used internally in different modules of
1229     the library but which are not part of the API. These have all had their
1230     names changed to start with "_pcre_" so that they are unlikely to clash
1231     with other external names.
1232    
1233     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1234     a different (DFA) algorithm. Although it is slower than the original
1235     function, it does have some advantages for certain types of matching
1236     problem.
1237    
1238     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1239     including restarting after a partial match.
1240    
1241     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1242     defined when compiling for Windows was sent to me. I have put it into the
1243     code, though I have no means of testing or verifying it.
1244    
1245     7. Added the pcre_refcount() auxiliary function.
1246    
1247     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1248     match before or at the first newline in the subject string. In pcretest,
1249     the /f option on a pattern can be used to set this.
1250    
1251     9. A repeated \w when used in UTF-8 mode with characters greater than 256
1252     would behave wrongly. This has been present in PCRE since release 4.0.
1253    
1254     10. A number of changes to the pcregrep command:
1255    
1256     (a) Refactored how -x works; insert ^(...)$ instead of setting
1257     PCRE_ANCHORED and checking the length, in preparation for adding
1258     something similar for -w.
1259    
1260     (b) Added the -w (match as a word) option.
1261    
1262     (c) Refactored the way lines are read and buffered so as to have more
1263     than one at a time available.
1264    
1265     (d) Implemented a pcregrep test script.
1266    
1267     (e) Added the -M (multiline match) option. This allows patterns to match
1268     over several lines of the subject. The buffering ensures that at least
1269     8K, or the rest of the document (whichever is the shorter) is available
1270     for matching (and similarly the previous 8K for lookbehind assertions).
1271    
1272     (f) Changed the --help output so that it now says
1273    
1274     -w, --word-regex(p)
1275    
1276     instead of two lines, one with "regex" and the other with "regexp"
1277     because that confused at least one person since the short forms are the
1278     same. (This required a bit of code, as the output is generated
1279     automatically from a table. It wasn't just a text change.)
1280    
1281     (g) -- can be used to terminate pcregrep options if the next thing isn't an
1282     option but starts with a hyphen. Could be a pattern or a path name
1283     starting with a hyphen, for instance.
1284    
1285     (h) "-" can be given as a file name to represent stdin.
1286    
1287     (i) When file names are being printed, "(standard input)" is used for
1288     the standard input, for compatibility with GNU grep. Previously
1289     "<stdin>" was used.
1290    
1291     (j) The option --label=xxx can be used to supply a name to be used for
1292     stdin when file names are being printed. There is no short form.
1293    
1294     (k) Re-factored the options decoding logic because we are going to add
1295     two more options that take data. Such options can now be given in four
1296     different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1297    
1298     (l) Added the -A, -B, and -C options for requesting that lines of context
1299     around matches be printed.
1300    
1301     (m) Added the -L option to print the names of files that do not contain
1302     any matching lines, that is, the complement of -l.
1303    
1304     (n) The return code is 2 if any file cannot be opened, but pcregrep does
1305     continue to scan other files.
1306    
1307     (o) The -s option was incorrectly implemented. For compatibility with other
1308     greps, it now suppresses the error message for a non-existent or non-
1309     accessible file (but not the return code). There is a new option called
1310     -q that suppresses the output of matching lines, which was what -s was
1311     previously doing.
1312    
1313     (p) Added --include and --exclude options to specify files for inclusion
1314     and exclusion when recursing.
1315    
1316     11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1317     Hopefully, it now does.
1318    
1319     12. Missing cast in pcre_study().
1320    
1321     13. Added an "uninstall" target to the makefile.
1322    
1323     14. Replaced "extern" in the function prototypes in Makefile.in with
1324     "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1325     world, but is set differently for Windows.
1326    
1327     15. Added a second compiling function called pcre_compile2(). The only
1328     difference is that it has an extra argument, which is a pointer to an
1329     integer error code. When there is a compile-time failure, this is set
1330     non-zero, in addition to the error test pointer being set to point to an
1331     error message. The new argument may be NULL if no error number is required
1332     (but then you may as well call pcre_compile(), which is now just a
1333     wrapper). This facility is provided because some applications need a
1334     numeric error indication, but it has also enabled me to tidy up the way
1335     compile-time errors are handled in the POSIX wrapper.
1336    
1337     16. Added VPATH=.libs to the makefile; this should help when building with one
1338     prefix path and installing with another. (Or so I'm told by someone who
1339     knows more about this stuff than I do.)
1340    
1341     17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1342     passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1343     match everything, including newlines. This is not POSIX-compatible, but
1344     somebody wanted the feature. From pcretest it can be activated by using
1345     both the P and the s flags.
1346    
1347     18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1348    
1349     19. libpcre.pc was being incorrectly installed as executable.
1350    
1351     20. A couple of places in pcretest check for end-of-line by looking for '\n';
1352     it now also looks for '\r' so that it will work unmodified on Windows.
1353    
1354     21. Added Google's contributed C++ wrapper to the distribution.
1355    
1356     22. Added some untidy missing memory free() calls in pcretest, to keep
1357     Electric Fence happy when testing.
1358    
1359    
1360    
1361 nigel 75 Version 5.0 13-Sep-04
1362     ---------------------
1363    
1364     1. Internal change: literal characters are no longer packed up into items
1365     containing multiple characters in a single byte-string. Each character
1366     is now matched using a separate opcode. However, there may be more than one
1367     byte in the character in UTF-8 mode.
1368    
1369     2. The pcre_callout_block structure has two new fields: pattern_position and
1370     next_item_length. These contain the offset in the pattern to the next match
1371     item, and its length, respectively.
1372    
1373     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1374     insertion of callouts before each pattern item. Added the /C option to
1375     pcretest to make use of this.
1376    
1377     4. On the advice of a Windows user, the lines
1378    
1379     #if defined(_WIN32) || defined(WIN32)
1380     _setmode( _fileno( stdout ), 0x8000 );
1381     #endif /* defined(_WIN32) || defined(WIN32) */
1382    
1383     have been added to the source of pcretest. This apparently does useful
1384     magic in relation to line terminators.
1385    
1386     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1387     for the benefit of those environments where the "b" makes a difference.
1388    
1389     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1390     to know about it. I have put a hack into configure.in that adds in code
1391     to set GCC=yes if CC=icc. This seems to end up at a point in the
1392     generated configure script that is early enough to affect the setting of
1393     compiler options, which is what is needed, but I have no means of testing
1394     whether it really works. (The user who reported this had patched the
1395     generated configure script, which of course I cannot do.)
1396    
1397     LATER: After change 22 below (new libtool files), the configure script
1398     seems to know about icc (and also ecc). Therefore, I have commented out
1399     this hack in configure.in.
1400    
1401     7. Added support for pkg-config (2 patches were sent in).
1402    
1403     8. Negated POSIX character classes that used a combination of internal tables
1404     were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1405     [[:^ascii]]. Typically, they would match almost any characters. The other
1406     POSIX classes were not broken in this way.
1407    
1408     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1409     to find the match, as PCRE was deluded into thinking that the match had to
1410     start at the start point or following a newline. The same bug applied to
1411     patterns with negative forward assertions or any backward assertions
1412     preceding ".*" at the start, unless the pattern required a fixed first
1413     character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1414    
1415     10. In UTF-8 mode, when moving forwards in the subject after a failed match
1416     starting at the last subject character, bytes beyond the end of the subject
1417     string were read.
1418    
1419     11. Renamed the variable "class" as "classbits" to make life easier for C++
1420     users. (Previously there was a macro definition, but it apparently wasn't
1421     enough.)
1422    
1423     12. Added the new field "tables" to the extra data so that tables can be passed
1424     in at exec time, or the internal tables can be re-selected. This allows
1425     a compiled regex to be saved and re-used at a later time by a different
1426     program that might have everything at different addresses.
1427    
1428     13. Modified the pcre-config script so that, when run on Solaris, it shows a
1429     -R library as well as a -L library.
1430    
1431     14. The debugging options of pcretest (-d on the command line or D on a
1432     pattern) showed incorrect output for anything following an extended class
1433     that contained multibyte characters and which was followed by a quantifier.
1434    
1435     15. Added optional support for general category Unicode character properties
1436     via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1437     support. It adds about 90K to the size of the library. The meanings of the
1438     inbuilt class escapes such as \d and \s have NOT been changed.
1439    
1440     16. Updated pcredemo.c to include calls to free() to release the memory for the
1441     compiled pattern.
1442    
1443     17. The generated file chartables.c was being created in the source directory
1444     instead of in the building directory. This caused the build to fail if the
1445     source directory was different from the building directory, and was
1446     read-only.
1447    
1448     18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1449     file. No doubt somebody will tell me if they don't make sense... Also added
1450     Dan Mooney's comments about building on OpenVMS.
1451    
1452     19. Added support for partial matching via the PCRE_PARTIAL option for
1453     pcre_exec() and the \P data escape in pcretest.
1454    
1455     20. Extended pcretest with 3 new pattern features:
1456    
1457     (i) A pattern option of the form ">rest-of-line" causes pcretest to
1458     write the compiled pattern to the file whose name is "rest-of-line".
1459     This is a straight binary dump of the data, with the saved pointer to
1460     the character tables forced to be NULL. The study data, if any, is
1461     written too. After writing, pcretest reads a new pattern.
1462    
1463     (ii) If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1464     compiled pattern from the given file. There must not be any
1465     occurrences of "<" in the file name (pretty unlikely); if there are,
1466     pcretest will instead treat the initial "<" as a pattern delimiter.
1467     After reading in the pattern, pcretest goes on to read data lines as
1468     usual.
1469    
1470     (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1471     and 16-bit fields in a compiled pattern, to simulate a pattern that
1472     was compiled on a host of opposite endianness.
1473    
1474     21. The pcre-exec() function can now cope with patterns that were compiled on
1475     hosts of opposite endianness, with this restriction:
1476    
1477     As for any compiled expression that is saved and used later, the tables
1478     pointer field cannot be preserved; the extra_data field in the arguments
1479     to pcre_exec() should be used to pass in a tables address if a value
1480     other than the default internal tables were used at compile time.
1481    
1482     22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1483     now diagnosed as an error. Previously, most of the time, a negative number
1484     would have been treated as zero, but if in addition "ovector" was passed as
1485     NULL, a crash could occur.
1486    
1487     23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1488     new versions from the libtool 1.5 distribution (the last one is a copy of
1489     a file called libtool.m4). This seems to have fixed the need to patch
1490     "configure" to support Darwin 1.3 (which I used to do). However, I still
1491     had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1492     workstation).
1493    
1494     24. Changed the PCRE licence to be the more standard "BSD" licence.
1495    
1496    
1497 nigel 73 Version 4.5 01-Dec-03
1498     ---------------------
1499    
1500     1. There has been some re-arrangement of the code for the match() function so
1501     that it can be compiled in a version that does not call itself recursively.
1502     Instead, it keeps those local variables that need separate instances for
1503     each "recursion" in a frame on the heap, and gets/frees frames whenever it
1504     needs to "recurse". Keeping track of where control must go is done by means
1505     of setjmp/longjmp. The whole thing is implemented by a set of macros that
1506     hide most of the details from the main code, and operates only if
1507     NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1508     "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1509     operating.
1510    
1511     To make it easier for callers to provide specially tailored get/free
1512     functions for this usage, two new functions, pcre_stack_malloc, and
1513     pcre_stack_free, are used. They are always called in strict stacking order,
1514     and the size of block requested is always the same.
1515    
1516     The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1517     PCRE has been compiled to use the stack or the heap for recursion. The
1518     -C option of pcretest uses this to show which version is compiled.
1519    
1520     A new data escape \S, is added to pcretest; it causes the amounts of store
1521     obtained and freed by both kinds of malloc/free at match time to be added
1522     to the output.
1523    
1524     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1525     what's available on my current Linux desktop machine.
1526    
1527     3. When matching a UTF-8 string, the test for a valid string at the start has
1528     been extended. If start_offset is not zero, PCRE now checks that it points
1529     to a byte that is the start of a UTF-8 character. If not, it returns
1530     PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1531     this is necessary because there may be backward assertions in the pattern.
1532     When matching the same subject several times, it may save resources to use
1533     PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1534    
1535     4. The code for checking the validity of UTF-8 strings has been tightened so
1536     that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1537     containing "overlong sequences".
1538    
1539     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1540     I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1541     should have been "&", but it just so happened that all the cases this let
1542     through by mistake were picked up later in the function.
1543    
1544     6. I had used a variable called "isblank" - this is a C99 function, causing
1545     some compilers to warn. To avoid this, I renamed it (as "blankclass").
1546    
1547     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1548     prompting; (b) run "./pcretest /dev/null" at the start of the test script
1549     so the version is shown; (c) stop "make test" echoing "./RunTest".
1550    
1551     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1552    
1553     9. The prototype for memmove() for systems that don't have it was using
1554     size_t, but the inclusion of the header that defines size_t was later. I've
1555     moved the #includes for the C headers earlier to avoid this.
1556    
1557     10. Added some adjustments to the code to make it easier to compiler on certain
1558     special systems:
1559    
1560     (a) Some "const" qualifiers were missing.
1561     (b) Added the macro EXPORT before all exported functions; by default this
1562     is defined to be empty.
1563     (c) Changed the dftables auxiliary program (that builds chartables.c) so
1564     that it reads its output file name as an argument instead of writing
1565     to the standard output and assuming this can be redirected.
1566    
1567     11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
1568     class containing characters with values greater than 255, PCRE compilation
1569     went into a loop.
1570    
1571     12. A recursive reference to a subpattern that was within another subpattern
1572     that had a minimum quantifier of zero caused PCRE to crash. For example,
1573     (x(y(?2))z)? provoked this bug with a subject that got as far as the
1574     recursion. If the recursively-called subpattern itself had a zero repeat,
1575     that was OK.
1576    
1577     13. In pcretest, the buffer for reading a data line was set at 30K, but the
1578     buffer into which it was copied (for escape processing) was still set at
1579     1024, so long lines caused crashes.
1580    
1581     14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
1582     "internal error: code overflow...". This applied to any character class
1583     that was followed by a possessive quantifier.
1584    
1585     15. Modified the Makefile to add libpcre.la as a prerequisite for
1586     libpcreposix.la because I was told this is needed for a parallel build to
1587     work.
1588    
1589     16. If a pattern that contained .* following optional items at the start was
1590     studied, the wrong optimizing data was generated, leading to matching
1591     errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
1592     matching string must start with a or b or c. The correct conclusion for
1593     this pattern is that a match can start with any character.
1594    
1595    
1596 nigel 71 Version 4.4 13-Aug-03
1597     ---------------------
1598    
1599     1. In UTF-8 mode, a character class containing characters with values between
1600     127 and 255 was not handled correctly if the compiled pattern was studied.
1601     In fixing this, I have also improved the studying algorithm for such
1602     classes (slightly).
1603    
1604     2. Three internal functions had redundant arguments passed to them. Removal
1605     might give a very teeny performance improvement.
1606    
1607     3. Documentation bug: the value of the capture_top field in a callout is *one
1608     more than* the number of the hightest numbered captured substring.
1609    
1610     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
1611     in incorrectly linking with a previously installed version. They now link
1612     explicitly with libpcre.la.
1613    
1614     5. configure.in no longer needs to recognize Cygwin specially.
1615    
1616     6. A problem in pcre.in for Windows platforms is fixed.
1617    
1618     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
1619     pcretest, it used to include the size of the study block as part of its
1620     output. Unfortunately, the structure contains a field that has a different
1621     size on different hardware architectures. This meant that the tests that
1622     showed this size failed. As the block is currently always of a fixed size,
1623     this information isn't actually particularly useful in pcretest output, so
1624     I have just removed it.
1625    
1626     8. Three pre-processor statements accidentally did not start in column 1.
1627     Sadly, there are *still* compilers around that complain, even though
1628     standard C has not required this for well over a decade. Sigh.
1629    
1630     9. In pcretest, the code for checking callouts passed small integers in the
1631     callout_data field, which is a void * field. However, some picky compilers
1632     complained about the casts involved for this on 64-bit systems. Now
1633     pcretest passes the address of the small integer instead, which should get
1634     rid of the warnings.
1635    
1636     10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
1637     both compile and run time, and gives an error if an invalid UTF-8 sequence
1638     is found. There is a option for disabling this check in cases where the
1639     string is known to be correct and/or the maximum performance is wanted.
1640    
1641     11. In response to a bug report, I changed one line in Makefile.in from
1642    
1643     -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
1644     to
1645     -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
1646    
1647     to look similar to other lines, but I have no way of telling whether this
1648     is the right thing to do, as I do not use Windows. No doubt I'll get told
1649     if it's wrong...
1650    
1651    
1652 nigel 69 Version 4.3 21-May-03
1653     ---------------------
1654    
1655     1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
1656     Makefile.
1657    
1658     2. Some refactoring to improve the quality of the code:
1659    
1660     (i) The utf8_table... variables are now declared "const".
1661    
1662     (ii) The code for \cx, which used the "case flipping" table to upper case
1663     lower case letters, now just substracts 32. This is ASCII-specific,
1664     but the whole concept of \cx is ASCII-specific, so it seems
1665     reasonable.
1666    
1667     (iii) PCRE was using its character types table to recognize decimal and
1668     hexadecimal digits in the pattern. This is silly, because it handles
1669     only 0-9, a-f, and A-F, but the character types table is locale-
1670     specific, which means strange things might happen. A private
1671     table is now used for this - though it costs 256 bytes, a table is
1672     much faster than multiple explicit tests. Of course, the standard
1673     character types table is still used for matching digits in subject
1674     strings against \d.
1675    
1676     (iv) Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
1677     ending in _t are). So I've renamed it as ESC_tee.
1678    
1679     3. The first argument for regexec() in the POSIX wrapper should have been
1680     defined as "const".
1681    
1682     4. Changed pcretest to use malloc() for its buffers so that they can be
1683     Electric Fenced for debugging.
1684    
1685     5. There were several places in the code where, in UTF-8 mode, PCRE would try
1686     to read one or more bytes before the start of the subject string. Often this
1687     had no effect on PCRE's behaviour, but in some circumstances it could
1688     provoke a segmentation fault.
1689    
1690     6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
1691     to try to read one or more bytes before the start of the subject string.
1692    
1693     7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
1694     UTF-8 support could misbehave in various ways if the subject string
1695     contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
1696     area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
1697     back over UTF-8 characters.)
1698    
1699    
1700 nigel 67 Version 4.2 14-Apr-03
1701     ---------------------
1702    
1703     1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
1704    
1705     2. Changes to the building process, supplied by Ronald Landheer-Cieslak
1706     [ON_WINDOWS]: new variable, "#" on non-Windows platforms
1707     [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
1708     [WIN_PREFIX]: new variable, "cyg" for Cygwin
1709     * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
1710     and BUILD_EXEEXT
1711     Note: automatic setting of the BUILD variables is not yet working
1712     set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
1713     compile-time but not at link-time
1714     [LINK]: use for linking executables only
1715     make different versions for Windows and non-Windows
1716     [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
1717     libraries
1718     [LINK_FOR_BUILD]: new variable
1719     [OBJEXT]: use throughout
1720     [EXEEXT]: use throughout
1721     <winshared>: new target
1722     <wininstall>: new target
1723     <dftables.o>: use native compiler
1724     <dftables>: use native linker
1725     <install>: handle Windows platform correctly
1726     <clean>: ditto
1727     <check>: ditto
1728     copy DLL to top builddir before testing
1729    
1730     As part of these changes, -no-undefined was removed again. This was reported
1731     to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
1732     in any case.
1733    
1734     3. Some tidies to get rid of compiler warnings:
1735    
1736     . In the match_data structure, match_limit was an unsigned long int, whereas
1737     match_call_count was an int. I've made them both unsigned long ints.
1738    
1739     . In pcretest the fact that a const uschar * doesn't automatically cast to
1740     a void * provoked a warning.
1741    
1742     . Turning on some more compiler warnings threw up some "shadow" variables
1743     and a few more missing casts.
1744    
1745     4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
1746     option, a class that contained a single character with a value between 128
1747     and 255 (e.g. /[\xFF]/) caused PCRE to crash.
1748    
1749     5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
1750     option, a class that contained several characters, but with at least one
1751     whose value was between 128 and 255 caused PCRE to crash.
1752    
1753    
1754 nigel 65 Version 4.1 12-Mar-03
1755     ---------------------
1756 nigel 63
1757 nigel 65 1. Compiling with gcc -pedantic found a couple of places where casts were
1758     needed, and a string in dftables.c that was longer than standard compilers are
1759     required to support.
1760    
1761     2. Compiling with Sun's compiler found a few more places where the code could
1762     be tidied up in order to avoid warnings.
1763    
1764     3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
1765     first of these names is deprecated in the latest Autoconf in favour of the name
1766     CC_FOR_BUILD, because "host" is typically used to mean the system on which the
1767     compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
1768     analogy I have changed it to CFLAGS_FOR_BUILD.
1769    
1770     4. Added -no-undefined to the linking command in the Makefile, because this is
1771     apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
1772     linking step for the pcreposix library.
1773    
1774     5. PCRE was failing to diagnose the case of two named groups with the same
1775     name.
1776    
1777     6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
1778     literal character that is needed in the subject for a match, and scans along to
1779     ensure that it is present before embarking on the full matching process. This
1780     saves time in cases of nested unlimited repeats that are never going to match.
1781     Problem: the scan can take a lot of time if the subject is very long (e.g.
1782     megabytes), thus penalizing straightforward matches. It is now done only if the
1783     amount of subject to be scanned is less than 1000 bytes.
1784    
1785     7. A lesser problem with the same optimization is that it was recording the
1786     first character of an anchored pattern as "needed", thus provoking a search
1787     right along the subject, even when the first match of the pattern was going to
1788     fail. The "needed" character is now not set for anchored patterns, unless it
1789     follows something in the pattern that is of non-fixed length. Thus, it still
1790     fulfils its original purpose of finding quick non-matches in cases of nested
1791     unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
1792    
1793    
1794     Version 4.0 17-Feb-03
1795     ---------------------
1796    
1797 nigel 63 1. If a comment in an extended regex that started immediately after a meta-item
1798     extended to the end of string, PCRE compiled incorrect data. This could lead to
1799     all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
1800    
1801     2. Moved to autoconf 2.53 and libtool 1.4.2.
1802    
1803     3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
1804     the special perltest8 script is no longer needed - all the tests can be run
1805     from a single perltest script.
1806    
1807     4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
1808     by \s. It has now been removed in PCRE. This means it isn't recognized as
1809     whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
1810     class [:space:] *does* include VT, thereby creating a mess.
1811    
1812     5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
1813     space and tab.
1814    
1815     6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
1816     its new features into the main test script, reducing the number of scripts.
1817    
1818     7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
1819     were backward compatible, and made the (?i) apply to the whole pattern, as if
1820     /i were given. Now it behaves more logically, and applies the option setting
1821     only to what follows. PCRE has been changed to follow suit. However, if it
1822     finds options settings right at the start of the pattern, it extracts them into
1823     the global options, as before. Thus, they show up in the info data.
1824    
1825     8. Added support for the \Q...\E escape sequence. Characters in between are
1826     treated as literals. This is slightly different from Perl in that $ and @ are
1827     also handled as literals inside the quotes. In Perl, they will cause variable
1828     interpolation. Note the following examples:
1829    
1830     Pattern PCRE matches Perl matches
1831    
1832     \Qabc$xyz\E abc$xyz abc followed by the contents of $xyz
1833     \Qabc\$xyz\E abc\$xyz abc\$xyz
1834     \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
1835    
1836     For compatibility with Perl, \Q...\E sequences are recognized inside character
1837     classes as well as outside them.
1838    
1839     9. Re-organized 3 code statements in pcretest to avoid "overflow in
1840     floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
1841     (size_t) cast to one statement in pcretest and one in pcreposix to avoid
1842     signed/unsigned warnings.
1843    
1844     10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
1845     option for pcretest, so I've replaced it by a simple function that does just
1846     that job.
1847    
1848     11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
1849     "pcregrep -".
1850    
1851     12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
1852     Java package. This provides some syntactic sugar for simple cases of what my
1853     documentation calls "once-only subpatterns". A pattern such as x*+ is the same
1854     as (?>x*). In other words, if what is inside (?>...) is just a single repeated
1855     item, you can use this simplified notation. Note that only makes sense with
1856     greedy quantifiers. Consequently, the use of the possessive quantifier forces
1857     greediness, whatever the setting of the PCRE_UNGREEDY option.
1858    
1859     13. A change of greediness default within a pattern was not taking effect at
1860     the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
1861     subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
1862     was abstracted outside.
1863    
1864     14. PCRE now supports the \G assertion. It is true when the current matching
1865     position is at the start point of the match. This differs from \A when the
1866     starting offset is non-zero. Used with the /g option of pcretest (or similar
1867     code), it works in the same way as it does for Perl's /g option. If all
1868     alternatives of a regex begin with \G, the expression is anchored to the start
1869     match position, and the "anchored" flag is set in the compiled expression.
1870    
1871     15. Some bugs concerning the handling of certain option changes within patterns
1872     have been fixed. These applied to options other than (?ims). For example,
1873     "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
1874     been the other way round. Some of this was related to change 7 above.
1875    
1876     16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
1877     features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
1878     and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
1879     POSIX classes only within a class (e.g. /[[:alpha:]]/).
1880    
1881     17. Added support for Perl's \C escape. This matches one byte, even in UTF8
1882     mode. Unlike ".", it always matches newline, whatever the setting of
1883     PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
1884     assertions. Perl allows it, but it doesn't (in general) work because it can't
1885     calculate the length of the lookbehind. At least, that's the case for Perl
1886     5.8.0 - I've been told they are going to document that it doesn't work in
1887     future.
1888    
1889     18. Added an error diagnosis for escapes that PCRE does not support: these are
1890     \L, \l, \N, \P, \p, \U, \u, and \X.
1891    
1892     19. Although correctly diagnosing a missing ']' in a character class, PCRE was
1893     reading past the end of the pattern in cases such as /[abcd/.
1894    
1895     20. PCRE was getting more memory than necessary for patterns with classes that
1896     contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
1897    
1898     21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
1899     compiling PCRE for use with Virtual Pascal.
1900    
1901     22. Small fix to the Makefile to make it work properly if the build is done
1902     outside the source tree.
1903    
1904     23. Added a new extension: a condition to go with recursion. If a conditional
1905     subpattern starts with (?(R) the "true" branch is used if recursion has
1906     happened, whereas the "false" branch is used only at the top level.
1907    
1908     24. When there was a very long string of literal characters (over 255 bytes
1909     without UTF support, over 250 bytes with UTF support), the computation of how
1910     much memory was required could be incorrect, leading to segfaults or other
1911     strange effects.
1912    
1913     25. PCRE was incorrectly assuming anchoring (either to start of subject or to
1914     start of line for a non-DOTALL pattern) when a pattern started with (.*) and
1915     there was a subsequent back reference to those brackets. This meant that, for
1916     example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
1917     possible to check for precisely this case. All we can do is abandon the
1918     optimization if .* occurs inside capturing brackets when there are any back
1919     references whatsoever. (See below for a better fix that came later.)
1920    
1921     26. The handling of the optimization for finding the first character of a
1922     non-anchored pattern, and for finding a character that is required later in the
1923     match were failing in some cases. This didn't break the matching; it just
1924     failed to optimize when it could. The way this is done has been re-implemented.
1925    
1926     27. Fixed typo in error message for invalid (?R item (it said "(?p").
1927    
1928     28. Added a new feature that provides some of the functionality that Perl
1929     provides with (?{...}). The facility is termed a "callout". The way it is done
1930     in PCRE is for the caller to provide an optional function, by setting
1931     pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
1932     global variable. By default it is unset, which disables all calling out. To get
1933     the function called, the regex must include (?C) at appropriate points. This
1934     is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
1935     This provides a means of identifying different callout points. When PCRE
1936     reaches such a point in the regex, if pcre_callout has been set, the external
1937     function is called. It is provided with data in a structure called
1938     pcre_callout_block, which is defined in pcre.h. If the function returns 0,
1939     matching continues; if it returns a non-zero value, the match at the current
1940     point fails. However, backtracking will occur if possible. [This was changed
1941     later and other features added - see item 49 below.]
1942    
1943     29. pcretest is upgraded to test the callout functionality. It provides a
1944     callout function that displays information. By default, it shows the start of
1945     the match and the current position in the text. There are some new data escapes
1946     to vary what happens:
1947    
1948     \C+ in addition, show current contents of captured substrings
1949     \C- do not supply a callout function
1950     \C!n return 1 when callout number n is reached
1951     \C!n!m return 1 when callout number n is reached for the mth time
1952    
1953     30. If pcregrep was called with the -l option and just a single file name, it
1954     output "<stdin>" if a match was found, instead of the file name.
1955    
1956     31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
1957     slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
1958     pcre_exec(). This saves a malloc/free per call. The default value of
1959     POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
1960     when configuring.
1961    
1962     32. The default maximum size of a compiled pattern is 64K. There have been a
1963     few cases of people hitting this limit. The code now uses macros to handle the
1964     storing of links as offsets within the compiled pattern. It defaults to 2-byte
1965     links, but this can be changed to 3 or 4 bytes by --with-link-size when
1966     configuring. Tests 2 and 5 work only with 2-byte links because they output
1967     debugging information about compiled patterns.
1968    
1969     33. Internal code re-arrangements:
1970    
1971     (a) Moved the debugging function for printing out a compiled regex into
1972     its own source file (printint.c) and used #include to pull it into
1973     pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
1974     separate copies.
1975    
1976     (b) Defined the list of op-code names for debugging as a macro in
1977     internal.h so that it is next to the definition of the opcodes.
1978    
1979     (c) Defined a table of op-code lengths for simpler skipping along compiled
1980     code. This is again a macro in internal.h so that it is next to the
1981     definition of the opcodes.
1982    
1983     34. Added support for recursive calls to individual subpatterns, along the
1984     lines of Robin Houston's patch (but implemented somewhat differently).
1985    
1986     35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
1987     allow it to read and process whole directories in Win32. This code was
1988     contributed by Lionel Fourquaux; it has not been tested by me.
1989    
1990     36. Added support for named subpatterns. The Python syntax (?P<name>...) is
1991     used to name a group. Names consist of alphanumerics and underscores, and must
1992     be unique. Back references use the syntax (?P=name) and recursive calls use
1993     (?P>name) which is a PCRE extension to the Python extension. Groups still have
1994     numbers. The function pcre_fullinfo() can be used after compilation to extract
1995     a name/number map. There are three relevant calls:
1996    
1997     PCRE_INFO_NAMEENTRYSIZE yields the size of each entry in the map
1998     PCRE_INFO_NAMECOUNT yields the number of entries
1999     PCRE_INFO_NAMETABLE yields a pointer to the map.
2000    
2001     The map is a vector of fixed-size entries. The size of each entry depends on
2002     the length of the longest name used. The first two bytes of each entry are the
2003     group number, most significant byte first. There follows the corresponding
2004     name, zero terminated. The names are in alphabetical order.
2005    
2006     37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
2007     case instead of 255. Making it the same both with and without UTF-8 support
2008     means that the same test output works with both.
2009    
2010     38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
2011     calling malloc() with a zero argument.
2012    
2013     39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
2014     optimization. I've improved things by keeping a bitmap of backreferences with
2015     numbers 1-31 so that if .* occurs inside capturing brackets that are not in
2016     fact referenced, the optimization can be applied. It is unlikely that a
2017     relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
2018     the match to follow \n) will appear inside brackets with a number greater than
2019     31, but if it does, any back reference > 31 suppresses the optimization.
2020    
2021     40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
2022     of disabling numbered capturing parentheses. Any opening parenthesis that is
2023     not followed by ? behaves as if it were followed by ?: but named parentheses
2024     can still be used for capturing (and they will acquire numbers in the usual
2025     way).
2026    
2027     41. Redesigned the return codes from the match() function into yes/no/error so
2028     that errors can be passed back from deep inside the nested calls. A malloc
2029     failure while inside a recursive subpattern call now causes the
2030     PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
2031    
2032     42. It is now possible to set a limit on the number of times the match()
2033     function is called in a call to pcre_exec(). This facility makes it possible to
2034     limit the amount of recursion and backtracking, though not in a directly
2035     obvious way, because the match() function is used in a number of different
2036     circumstances. The count starts from zero for each position in the subject
2037     string (for non-anchored patterns). The default limit is, for compatibility, a
2038     large number, namely 10 000 000. You can change this in two ways:
2039    
2040     (a) When configuring PCRE before making, you can use --with-match-limit=n
2041     to set a default value for the compiled library.
2042    
2043     (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
2044     a different value is set. See 45 below.
2045    
2046     If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
2047    
2048     43. Added a new function pcre_config(int, void *) to enable run-time extraction
2049     of things that can be changed at compile time. The first argument specifies
2050     what is wanted and the second points to where the information is to be placed.
2051     The current list of available information is:
2052    
2053     PCRE_CONFIG_UTF8
2054    
2055     The output is an integer that is set to one if UTF-8 support is available;
2056     otherwise it is set to zero.
2057    
2058     PCRE_CONFIG_NEWLINE
2059    
2060     The output is an integer that it set to the value of the code that is used for
2061     newline. It is either LF (10) or CR (13).
2062    
2063     PCRE_CONFIG_LINK_SIZE
2064    
2065     The output is an integer that contains the number of bytes used for internal
2066     linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
2067    
2068     PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
2069    
2070     The output is an integer that contains the threshold above which the POSIX
2071     interface uses malloc() for output vectors. See item 31 above.
2072    
2073     PCRE_CONFIG_MATCH_LIMIT
2074    
2075     The output is an unsigned integer that contains the default limit of the number
2076     of match() calls in a pcre_exec() execution. See 42 above.
2077    
2078     44. pcretest has been upgraded by the addition of the -C option. This causes it
2079     to extract all the available output from the new pcre_config() function, and to
2080     output it. The program then exits immediately.
2081    
2082     45. A need has arisen to pass over additional data with calls to pcre_exec() in
2083     order to support additional features. One way would have been to define
2084     pcre_exec2() (for example) with extra arguments, but this would not have been
2085     extensible, and would also have required all calls to the original function to
2086     be mapped to the new one. Instead, I have chosen to extend the mechanism that
2087     is used for passing in "extra" data from pcre_study().
2088    
2089     The pcre_extra structure is now exposed and defined in pcre.h. It currently
2090     contains the following fields:
2091    
2092     flags a bitmap indicating which of the following fields are set
2093     study_data opaque data from pcre_study()
2094     match_limit a way of specifying a limit on match() calls for a specific
2095     call to pcre_exec()
2096     callout_data data for callouts (see 49 below)
2097    
2098     The flag bits are also defined in pcre.h, and are
2099    
2100     PCRE_EXTRA_STUDY_DATA
2101     PCRE_EXTRA_MATCH_LIMIT
2102     PCRE_EXTRA_CALLOUT_DATA
2103    
2104     The pcre_study() function now returns one of these new pcre_extra blocks, with
2105     the actual study data pointed to by the study_data field, and the
2106     PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
2107     before. That is, this change is entirely upwards-compatible and requires no
2108     change to existing code.
2109    
2110     If you want to pass in additional data to pcre_exec(), you can either place it
2111     in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
2112     block.
2113    
2114     46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
2115     data string contains the escape sequence \M, pcretest calls pcre_exec() several
2116     times with different match limits, until it finds the minimum value needed for
2117     pcre_exec() to complete. The value is then output. This can be instructive; for
2118     most simple matches the number is quite small, but for pathological cases it
2119     gets very large very quickly.
2120    
2121     47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
2122     returns the size of the data block pointed to by the study_data field in a
2123     pcre_extra block, that is, the value that was passed as the argument to
2124     pcre_malloc() when PCRE was getting memory in which to place the information
2125     created by pcre_study(). The fourth argument should point to a size_t variable.
2126     pcretest has been extended so that this information is shown after a successful
2127     pcre_study() call when information about the compiled regex is being displayed.
2128    
2129     48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
2130     because what follows is always an absolute path. (Later: it turns out that this
2131     is more than cosmetic for MinGW, because it doesn't like empty path
2132     components.)
2133    
2134     49. Some changes have been made to the callout feature (see 28 above):
2135    
2136     (i) A callout function now has three choices for what it returns:
2137    
2138     0 => success, carry on matching
2139     > 0 => failure at this point, but backtrack if possible
2140     < 0 => serious error, return this value from pcre_exec()
2141    
2142     Negative values should normally be chosen from the set of PCRE_ERROR_xxx
2143     values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
2144     "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
2145     use by callout functions. It will never be used by PCRE itself.
2146    
2147     (ii) The pcre_extra structure (see 45 above) has a void * field called
2148     callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
2149     pcre_callout_block structure has a field of the same name. The contents of
2150     the field passed in the pcre_extra structure are passed to the callout
2151     function in the corresponding field in the callout block. This makes it
2152     easier to use the same callout-containing regex from multiple threads. For
2153     testing, the pcretest program has a new data escape
2154    
2155     \C*n pass the number n (may be negative) as callout_data
2156    
2157     If the callout function in pcretest receives a non-zero value as
2158     callout_data, it returns that value.
2159    
2160     50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
2161     there were some redundant $(CFLAGS) in commands that are now specified as
2162     $(LINK), which already includes $(CFLAGS).
2163    
2164     51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
2165     has been compiled with UTF-8 support *and* pcre_compile() has been compiled
2166     with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
2167     one-byte characters throughout. Note that case-insensitive matching applies
2168     only to characters whose values are less than 256. PCRE doesn't support the
2169     notion of cases for higher-valued characters.
2170    
2171     (i) A character class whose characters are all within 0-255 is handled as
2172     a bit map, and the map is inverted for negative classes. Previously, a
2173     character > 255 always failed to match such a class; however it should
2174     match if the class was a negative one (e.g. [^ab]). This has been fixed.
2175    
2176     (ii) A negated character class with a single character < 255 is coded as
2177     "not this character" (OP_NOT). This wasn't working properly when the test
2178     character was multibyte, either singly or repeated.
2179    
2180     (iii) Repeats of multibyte characters are now handled correctly in UTF-8
2181     mode, for example: \x{100}{2,3}.
2182    
2183     (iv) The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
2184     singly or repeated) now correctly test multibyte characters. However,
2185     PCRE doesn't recognize any characters with values greater than 255 as
2186     digits, spaces, or word characters. Such characters always match \D, \S,
2187     and \W, and never match \d, \s, or \w.
2188    
2189     (v) Classes may now contain characters and character ranges with values
2190     greater than 255. For example: [ab\x{100}-\x{400}].
2191    
2192     (vi) pcregrep now has a --utf-8 option (synonym -u) which makes it call
2193     PCRE in UTF-8 mode.
2194    
2195     52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
2196     PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
2197     retained for backwards compatibility. (Note that LASTLITERAL is also a byte
2198     value.)
2199    
2200     53. The single man page has become too large. I have therefore split it up into
2201     a number of separate man pages. These also give rise to individual HTML pages;
2202     these are now put in a separate directory, and there is an index.html page that
2203     lists them all. Some hyperlinking between the pages has been installed.
2204    
2205     54. Added convenience functions for handling named capturing parentheses.
2206    
2207     55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2208     aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2209     true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2210     are faulted.
2211    
2212     56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2213     calling configure. These values are used when compiling the dftables.c program
2214     which is run to generate the source of the default character tables. They
2215     default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2216     you will need to set these values.
2217    
2218     57. Updated the building process for Windows DLL, as provided by Fred Cox.
2219    
2220    
2221     Version 3.9 02-Jan-02
2222 nigel 61 ---------------------
2223    
2224     1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2225    
2226     2. If --disable-static was given, the building process failed when trying to
2227     build pcretest and pcregrep. (For some reason it was using libtool to compile
2228     them, which is not right, as they aren't part of the library.)
2229    
2230    
2231 nigel 59 Version 3.8 18-Dec-01
2232     ---------------------
2233    
2234     1. The experimental UTF-8 code was completely screwed up. It was packing the
2235     bytes in the wrong order. How dumb can you get?
2236    
2237    
2238 nigel 57 Version 3.7 29-Oct-01
2239     ---------------------
2240    
2241     1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2242     This caused pcretest, when used on the test data, to segfault. Unfortunately,
2243     this didn't happen under Solaris 8, where I normally test things.
2244    
2245 nigel 59 2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2246     doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2247     isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2248     this fix an hour or so after the initial 3.7 release.)
2249 nigel 57
2250 nigel 59
2251 nigel 55 Version 3.6 23-Oct-01
2252     ---------------------
2253    
2254     1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2255     offsets passed as NULL with zero offset count.
2256    
2257     2. The config.guess and config.sub files had not been updated when I moved to
2258     the latest autoconf.
2259    
2260    
2261 nigel 53 Version 3.5 15-Aug-01
2262     ---------------------
2263 nigel 5
2264 nigel 53 1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2265     had been forgotten.
2266    
2267     2. By using declared but undefined structures, we can avoid using "void"
2268     definitions in pcre.h while keeping the internal definitions of the structures
2269     private.
2270    
2271     3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2272     user point of view, this means that both static and shared libraries are built
2273     by default, but this can be individually controlled. More of the work of
2274     handling this static/shared cases is now inside libtool instead of PCRE's make
2275     file.
2276    
2277     4. The pcretest utility is now installed along with pcregrep because it is
2278     useful for users (to test regexs) and by doing this, it automatically gets
2279     relinked by libtool. The documentation has been turned into a man page, so
2280     there are now .1, .txt, and .html versions in /doc.
2281    
2282     5. Upgrades to pcregrep:
2283     (i) Added long-form option names like gnu grep.
2284     (ii) Added --help to list all options with an explanatory phrase.
2285     (iii) Added -r, --recursive to recurse into sub-directories.
2286     (iv) Added -f, --file to read patterns from a file.
2287    
2288     6. pcre_exec() was referring to its "code" argument before testing that
2289     argument for NULL (and giving an error if it was NULL).
2290    
2291     7. Upgraded Makefile.in to allow for compiling in a different directory from
2292     the source directory.
2293    
2294     8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2295     options bits, the pointer it was passed was to an int instead of to an unsigned
2296     long int. This mattered only on 64-bit systems.
2297    
2298     9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2299     generated) instead of pcre.in, which it its source. Also made the same change
2300     in several of the .c files.
2301    
2302     10. A new release of gcc defines printf() as a macro, which broke pcretest
2303     because it had an ifdef in the middle of a string argument for printf(). Fixed
2304     by using separate calls to printf().
2305    
2306     11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2307     script, to force use of CR or LF instead of \n in the source. On non-Unix
2308     systems, the value can be set in config.h.
2309    
2310     12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2311     absolute limit. Changed the text of the error message to make this clear, and
2312     likewise updated the man page.
2313    
2314     13. The limit of 99 on the number of capturing subpatterns has been removed.
2315     The new limit is 65535, which I hope will not be a "real" limit.
2316    
2317    
2318 nigel 51 Version 3.4 22-Aug-00
2319     ---------------------
2320    
2321     1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2322    
2323     2. Diagnose condition (?(0) as an error instead of crashing on matching.
2324    
2325    
2326 nigel 49 Version 3.3 01-Aug-00
2327     ---------------------
2328    
2329     1. If an octal character was given, but the value was greater than \377, it
2330     was not getting masked to the least significant bits, as documented. This could
2331     lead to crashes in some systems.
2332    
2333     2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
2334     the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
2335    
2336     3. Added the functions pcre_free_substring() and pcre_free_substring_list().
2337     These just pass their arguments on to (pcre_free)(), but they are provided
2338     because some uses of PCRE bind it to non-C systems that can call its functions,
2339     but cannot call free() or pcre_free() directly.
2340    
2341     4. Add "make test" as a synonym for "make check". Corrected some comments in
2342     the Makefile.
2343    
2344     5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
2345     Makefile.
2346    
2347     6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
2348     command called pgrep for grepping around the active processes.
2349    
2350     7. Added the beginnings of support for UTF-8 character strings.
2351    
2352     8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
2353     RANLIB to ./ltconfig so that they are used by libtool. I think these are all
2354     the relevant ones. (AR is not passed because ./ltconfig does its own figuring
2355     out for the ar command.)
2356    
2357    
2358 nigel 47 Version 3.2 12-May-00
2359     ---------------------
2360    
2361     This is purely a bug fixing release.
2362    
2363     1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
2364     of ZA. This was just one example of several cases that could provoke this bug,
2365     which was introduced by change 9 of version 2.00. The code for breaking
2366     infinite loops after an iteration that matches an empty string was't working
2367     correctly.
2368    
2369     2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
2370     when matched against abbab (for example). After matching an empty string, it
2371     wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
2372     caused it to match further down the string than it should.
2373    
2374     3. The code contained an inclusion of sys/types.h. It isn't clear why this
2375     was there because it doesn't seem to be needed, and it causes trouble on some
2376     systems, as it is not a Standard C header. It has been removed.
2377    
2378     4. Made 4 silly changes to the source to avoid stupid compiler warnings that
2379     were reported on the Macintosh. The changes were from
2380    
2381     while ((c = *(++ptr)) != 0 && c != '\n');
2382     to
2383     while ((c = *(++ptr)) != 0 && c != '\n') ;
2384    
2385     Totally extraordinary, but if that's what it takes...
2386    
2387     5. PCRE is being used in one environment where neither memmove() nor bcopy() is
2388     available. Added HAVE_BCOPY and an autoconf test for it; if neither
2389     HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
2390     assumes the way PCRE uses memmove() (always moving upwards).
2391    
2392     6. PCRE is being used in one environment where strchr() is not available. There
2393     was only one use in pcre.c, and writing it out to avoid strchr() probably gives
2394     faster code anyway.
2395    
2396    
2397 nigel 45 Version 3.1 09-Feb-00
2398     ---------------------
2399    
2400     The only change in this release is the fixing of some bugs in Makefile.in for
2401     the "install" target:
2402    
2403     (1) It was failing to install pcreposix.h.
2404    
2405     (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
2406    
2407    
2408 nigel 43 Version 3.0 01-Feb-00
2409     ---------------------
2410 nigel 41
2411     1. Add support for the /+ modifier to perltest (to output $` like it does in
2412     pcretest).
2413    
2414     2. Add support for the /g modifier to perltest.
2415    
2416     3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
2417     matches null strings.
2418    
2419     4. Fix perltest so that it doesn't do unwanted things when fed an empty
2420     pattern. Perl treats empty patterns specially - it reuses the most recent
2421     pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
2422     effect.
2423    
2424     5. The POSIX interface was broken in that it was just handing over the POSIX
2425     captured string vector to pcre_exec(), but (since release 2.00) PCRE has
2426     required a bigger vector, with some working space on the end. This means that
2427     the POSIX wrapper now has to get and free some memory, and copy the results.
2428    
2429 nigel 43 6. Added some simple autoconf support, placing the test data and the
2430     documentation in separate directories, re-organizing some of the
2431     information files, and making it build pcre-config (a GNU standard). Also added
2432     libtool support for building PCRE as a shared library, which is now the
2433     default.
2434 nigel 41
2435 nigel 43 7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
2436     09 are not valid octal constants. Single digits will be used for minor values
2437     less than 10.
2438    
2439     8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
2440     existing programs that set these in the POSIX interface can use PCRE without
2441     modification.
2442    
2443     9. Added a new function, pcre_fullinfo() with an extensible interface. It can
2444     return all that pcre_info() returns, plus additional data. The pcre_info()
2445     function is retained for compatibility, but is considered to be obsolete.
2446    
2447     10. Added experimental recursion feature (?R) to handle one common case that
2448     Perl 5.6 will be able to do with (?p{...}).
2449    
2450     11. Added support for POSIX character classes like [:alpha:], which Perl is
2451     adopting.
2452    
2453    
2454 nigel 39 Version 2.08 31-Aug-99
2455     ----------------------
2456    
2457     1. When startoffset was not zero and the pattern began with ".*", PCRE was not
2458     trying to match at the startoffset position, but instead was moving forward to
2459     the next newline as if a previous match had failed.
2460    
2461     2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
2462     and could get into a loop if a null string was matched other than at the start
2463     of the subject.
2464    
2465     3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
2466     be distinguished at compile time, and for completeness also added PCRE_DATE.
2467    
2468     5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
2469     in GnuWin32 environments.
2470    
2471    
2472 nigel 37 Version 2.07 29-Jul-99
2473     ----------------------
2474    
2475     1. The documentation is now supplied in plain text form and HTML as well as in
2476     the form of man page sources.
2477    
2478     2. C++ compilers don't like assigning (void *) values to other pointer types.
2479     In particular this affects malloc(). Although there is no problem in Standard
2480     C, I've put in casts to keep C++ compilers happy.
2481    
2482     3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
2483     should be (const char *).
2484    
2485     4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
2486     be useful for non-Unix systems who don't want to bother with the POSIX stuff.
2487     However, I haven't made this a standard facility. The documentation doesn't
2488     mention it, and the Makefile doesn't support it.
2489    
2490     5. The Makefile now contains an "install" target, with editable destinations at
2491     the top of the file. The pcretest program is not installed.
2492    
2493     6. pgrep -V now gives the PCRE version number and date.
2494    
2495     7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
2496     causing the entire string to be ignored, instead of just the last character.
2497    
2498     8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
2499     non-matching string, it can take a very, very long time, even for strings of
2500     quite modest length, because of the nested recursion. PCRE now does better in
2501     some of these cases. It does this by remembering the last required literal
2502     character in the pattern, and pre-searching the subject to ensure it is present
2503     before running the real match. In other words, it applies a heuristic to detect
2504     some types of certain failure quickly, and in the above example, if presented
2505     with a string that has no trailing " it gives "no match" very quickly.
2506    
2507     9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
2508     other alternatives are tried instead.
2509    
2510    
2511 nigel 35 Version 2.06 09-Jun-99
2512     ----------------------
2513    
2514     1. Change pcretest's output for amount of store used to show just the code
2515     space, because the remainder (the data block) varies in size between 32-bit and
2516     64-bit systems.
2517    
2518     2. Added an extra argument to pcre_exec() to supply an offset in the subject to
2519     start matching at. This allows lookbehinds to work when searching for multiple
2520     occurrences in a string.
2521    
2522     3. Added additional options to pcretest for testing multiple occurrences:
2523    
2524     /+ outputs the rest of the string that follows a match
2525     /g loops for multiple occurrences, using the new startoffset argument
2526     /G loops for multiple occurrences by passing an incremented pointer
2527    
2528     4. PCRE wasn't doing the "first character" optimization for patterns starting
2529     with \b or \B, though it was doing it for other lookbehind assertions. That is,
2530     it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
2531     the letter 'x'. On long subject strings, this gives a significant speed-up.
2532    
2533    
2534 nigel 33 Version 2.05 21-Apr-99
2535     ----------------------
2536    
2537     1. Changed the type of magic_number from int to long int so that it works
2538     properly on 16-bit systems.
2539    
2540     2. Fixed a bug which caused patterns starting with .* not to work correctly
2541     when the subject string contained newline characters. PCRE was assuming
2542     anchoring for such patterns in all cases, which is not correct because .* will
2543     not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
2544     DOTALL is set at top level; otherwise it knows that patterns starting with .*
2545     must be retried after every newline in the subject.
2546    
2547    
2548 nigel 31 Version 2.04 18-Feb-99
2549     ----------------------
2550    
2551     1. For parenthesized subpatterns with repeats whose minimum was zero, the
2552     computation of the store needed to hold the pattern was incorrect (too large).
2553     If such patterns were nested a few deep, this could multiply and become a real
2554     problem.
2555    
2556     2. Added /M option to pcretest to show the memory requirement of a specific
2557     pattern. Made -m a synonym of -s (which does this globally) for compatibility.
2558    
2559     3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
2560     compiled in such a way that the backtracking after subsequent failure was
2561     pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
2562     ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
2563    
2564    
2565 nigel 29 Version 2.03 02-Feb-99
2566     ----------------------
2567    
2568     1. Fixed typo and small mistake in man page.
2569    
2570 nigel 31 2. Added 4th condition (GPL supersedes if conflict) and created separate
2571     LICENCE file containing the conditions.
2572 nigel 29
2573     3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
2574     Perl, that is the internal \ allows the delimiter to be included in the
2575     pattern. Locked out the use of \ as a delimiter. If \ immediately follows
2576     the final delimiter, add \ to the end of the pattern (to test the error).
2577    
2578     4. Added the convenience functions for extracting substrings after a successful
2579     match. Updated pcretest to make it able to test these functions.
2580    
2581    
2582 nigel 27 Version 2.02 14-Jan-99
2583     ----------------------
2584    
2585     1. Initialized the working variables associated with each extraction so that
2586     their saving and restoring doesn't refer to uninitialized store.
2587    
2588     2. Put dummy code into study.c in order to trick the optimizer of the IBM C
2589     compiler for OS/2 into generating correct code. Apparently IBM isn't going to
2590     fix the problem.
2591    
2592     3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
2593     calls, and wasn't printing the correct value for compiling calls. Increased the
2594     default value of LOOPREPEAT, and the number of significant figures in the
2595     times.
2596    
2597     4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
2598    
2599     5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
2600     a building problem on Windows NT with a FAT file system.
2601    
2602    
2603 nigel 25 Version 2.01 21-Oct-98
2604     ----------------------
2605    
2606     1. Changed the API for pcre_compile() to allow for the provision of a pointer
2607     to character tables built by pcre_maketables() in the current locale. If NULL
2608     is passed, the default tables are used.
2609    
2610    
2611 nigel 23 Version 2.00 24-Sep-98
2612 nigel 21 ----------------------
2613    
2614 nigel 23 1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
2615     it any more.
2616    
2617     2. Allow quantification of (?>) groups, and make it work correctly.
2618    
2619     3. The first character computation wasn't working for (?>) groups.
2620    
2621     4. Correct the implementation of \Z (it is permitted to match on the \n at the
2622     end of the subject) and add 5.005's \z, which really does match only at the
2623     very end of the subject.
2624    
2625     5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
2626    
2627     6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
2628     DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
2629     localized options. All options to pcre_study() were also removed.
2630    
2631     7. Add other new features from 5.005:
2632    
2633     $(?<= positive lookbehind
2634     $(?<! negative lookbehind
2635     (?imsx-imsx) added the unsetting capability
2636     such a setting is global if at outer level; local otherwise
2637     (?imsx-imsx:) non-capturing groups with option setting
2638     (?(cond)re|re) conditional pattern matching
2639    
2640     A backreference to itself in a repeated group matches the previous
2641     captured string.
2642    
2643     8. General tidying up of studying (both automatic and via "study")
2644     consequential on the addition of new assertions.
2645    
2646     9. As in 5.005, unlimited repeated groups that could match an empty substring
2647     are no longer faulted at compile time. Instead, the loop is forcibly broken at
2648     runtime if any iteration does actually match an empty substring.
2649    
2650     10. Include the RunTest script in the distribution.
2651    
2652     11. Added tests from the Perl 5.005_02 distribution. This showed up a few
2653     discrepancies, some of which were old and were also with respect to 5.004. They
2654     have now been fixed.
2655    
2656    
2657     Version 1.09 28-Apr-98
2658     ----------------------
2659    
2660 nigel 21 1. A negated single character class followed by a quantifier with a minimum
2661     value of one (e.g. [^x]{1,6} ) was not compiled correctly. This could lead to
2662     program crashes, or just wrong answers. This did not apply to negated classes
2663     containing more than one character, or to minima other than one.
2664    
2665    
2666 nigel 19 Version 1.08 27-Mar-98
2667     ----------------------
2668    
2669     1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
2670    
2671     2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
2672     latter must appear before anything that relies on it in the pattern.
2673    
2674    
2675 nigel 17 Version 1.07 16-Feb-98
2676     ----------------------
2677    
2678     1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
2679     repeat of a potentially empty string).
2680    
2681    
2682 nigel 15 Version 1.06 23-Jan-98
2683     ----------------------
2684    
2685     1. Added Markus Oberhumer's little patches for C++.
2686    
2687     2. Literal strings longer than 255 characters were broken.
2688    
2689    
2690 nigel 13 Version 1.05 23-Dec-97
2691     ----------------------
2692    
2693     1. Negated character classes containing more than one character were failing if
2694     PCRE_CASELESS was set at run time.
2695    
2696    
2697 nigel 11 Version 1.04 19-Dec-97
2698     ----------------------
2699    
2700     1. Corrected the man page, where some "const" qualifiers had been omitted.
2701    
2702     2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
2703     input syntax.
2704    
2705     3. Fixed memory leak which occurred when a regex with back references was
2706     matched with an offsets vector that wasn't big enough. The temporary memory
2707     that is used in this case wasn't being freed if the match failed.
2708    
2709     4. Tidied pcretest to ensure it frees memory that it gets.
2710    
2711     5. Temporary memory was being obtained in the case where the passed offsets
2712     vector was exactly big enough.
2713    
2714     6. Corrected definition of offsetof() from change 5 below.
2715    
2716     7. I had screwed up change 6 below and broken the rules for the use of
2717     setjmp(). Now fixed.
2718    
2719    
2720 nigel 9 Version 1.03 18-Dec-97
2721     ----------------------
2722    
2723     1. A erroneous regex with a missing opening parenthesis was correctly
2724     diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
2725     on some systems.
2726    
2727     2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
2728     it was reported that one broken compiler failed on the former because "code" is
2729     also an independent variable.
2730    
2731     3. The erroneous regex a[]b caused an array overrun reference.
2732    
2733     4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
2734     fail on data ending with that character. (It was going on too far, and checking
2735     the next character, typically a binary zero.) This was specific to the
2736     optimized code for single-character negative classes.
2737    
2738     5. Added a contributed patch from the TIN world which does the following:
2739    
2740     + Add an undef for memmove, in case the the system defines a macro for it.
2741    
2742     + Add a definition of offsetof(), in case there isn't one. (I don't know
2743     the reason behind this - offsetof() is part of the ANSI standard - but
2744     it does no harm).
2745    
2746     + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
2747     most of the places where whitespace preceded '#'. I have given up and
2748     allowed the remaining 2 cases to be at the margin.
2749    
2750     + Rename some variables in pcre to eliminate shadowing. This seems very
2751     pedantic, but does no harm, of course.
2752    
2753     6. Moved the call to setjmp() into its own function, to get rid of warnings
2754     from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
2755    
2756     7. Constructs such as \d{8,} were compiling into the equivalent of
2757 nigel 11 \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
2758 nigel 9 outcome, but in this particular case used more store than had been allocated,
2759     which caused the bug to be discovered because it threw up an internal error.
2760    
2761     8. The debugging code in both pcre and pcretest for outputting the compiled
2762     form of a regex was going wrong in the case of back references followed by
2763     curly-bracketed repeats.
2764    
2765    
2766 nigel 7 Version 1.02 12-Dec-97
2767     ----------------------
2768    
2769     1. Typos in pcre.3 and comments in the source fixed.
2770    
2771     2. Applied a contributed patch to get rid of places where it used to remove
2772     'const' from variables, and fixed some signed/unsigned and uninitialized
2773     variable warnings.
2774    
2775     3. Added the "runtest" target to Makefile.
2776    
2777     4. Set default compiler flag to -O2 rather than just -O.
2778    
2779    
2780 nigel 5 Version 1.01 19-Nov-97
2781     ----------------------
2782    
2783     1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
2784     like /([ab]*)*/, that is, for classes with more than one character in them.
2785    
2786     2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
2787     as /((?>a*))*/ (a PCRE_EXTRA facility).
2788    
2789    
2790     Version 1.00 18-Nov-97
2791     ----------------------
2792    
2793     1. Added compile-time macros to support systems such as SunOS4 which don't have
2794     memmove() or strerror() but have other things that can be used instead.
2795    
2796     2. Arranged that "make clean" removes the executables.
2797    
2798    
2799 nigel 3 Version 0.99 27-Oct-97
2800     ----------------------
2801    
2802     1. Fixed bug in code for optimizing classes with only one character. It was
2803     initializing a 32-byte map regardless, which could cause it to run off the end
2804     of the memory it had got.
2805    
2806     2. Added, conditional on PCRE_EXTRA, the proposed (?>REGEX) construction.
2807    
2808    
2809     Version 0.98 22-Oct-97
2810     ----------------------
2811    
2812     1. Fixed bug in code for handling temporary memory usage when there are more
2813     back references than supplied space in the ovector. This could cause segfaults.
2814    
2815    
2816     Version 0.97 21-Oct-97
2817     ----------------------
2818    
2819     1. Added the \X "cut" facility, conditional on PCRE_EXTRA.
2820    
2821     2. Optimized negated single characters not to use a bit map.
2822    
2823     3. Brought error texts together as macro definitions; clarified some of them;
2824     fixed one that was wrong - it said "range out of order" when it meant "invalid
2825     escape sequence".
2826    
2827     4. Changed some char * arguments to const char *.
2828    
2829     5. Added PCRE_NOTBOL and PCRE_NOTEOL (from POSIX).
2830    
2831     6. Added the POSIX-style API wrapper in pcreposix.a and testing facilities in
2832     pcretest.
2833    
2834    
2835     Version 0.96 16-Oct-97
2836     ----------------------
2837    
2838     1. Added a simple "pgrep" utility to the distribution.
2839    
2840     2. Fixed an incompatibility with Perl: "{" is now treated as a normal character
2841     unless it appears in one of the precise forms "{ddd}", "{ddd,}", or "{ddd,ddd}"
2842     where "ddd" means "one or more decimal digits".
2843    
2844     3. Fixed serious bug. If a pattern had a back reference, but the call to
2845     pcre_exec() didn't supply a large enough ovector to record the related
2846     identifying subpattern, the match always failed. PCRE now remembers the number
2847     of the largest back reference, and gets some temporary memory in which to save
2848     the offsets during matching if necessary, in order to ensure that
2849     backreferences always work.
2850    
2851     4. Increased the compatibility with Perl in a number of ways:
2852    
2853     (a) . no longer matches \n by default; an option PCRE_DOTALL is provided
2854     to request this handling. The option can be set at compile or exec time.
2855    
2856     (b) $ matches before a terminating newline by default; an option
2857     PCRE_DOLLAR_ENDONLY is provided to override this (but not in multiline
2858     mode). The option can be set at compile or exec time.
2859    
2860     (c) The handling of \ followed by a digit other than 0 is now supposed to be
2861     the same as Perl's. If the decimal number it represents is less than 10
2862     or there aren't that many previous left capturing parentheses, an octal
2863     escape is read. Inside a character class, it's always an octal escape,
2864     even if it is a single digit.
2865    
2866     (d) An escaped but undefined alphabetic character is taken as a literal,
2867     unless PCRE_EXTRA is set. Currently this just reserves the remaining
2868     escapes.
2869    
2870     (e) {0} is now permitted. (The previous item is removed from the compiled
2871     pattern).
2872    
2873     5. Changed all the names of code files so that the basic parts are no longer
2874     than 10 characters, and abolished the teeny "globals.c" file.
2875    
2876     6. Changed the handling of character classes; they are now done with a 32-byte
2877     bit map always.
2878    
2879     7. Added the -d and /D options to pcretest to make it possible to look at the
2880     internals of compilation without having to recompile pcre.
2881    
2882    
2883     Version 0.95 23-Sep-97
2884     ----------------------
2885    
2886     1. Fixed bug in pre-pass concerning escaped "normal" characters such as \x5c or
2887     \x20 at the start of a run of normal characters. These were being treated as
2888     real characters, instead of the source characters being re-checked.
2889    
2890    
2891     Version 0.94 18-Sep-97
2892     ----------------------
2893    
2894     1. The functions are now thread-safe, with the caveat that the global variables
2895     containing pointers to malloc() and free() or alternative functions are the
2896     same for all threads.
2897    
2898     2. Get pcre_study() to generate a bitmap of initial characters for non-
2899     anchored patterns when this is possible, and use it if passed to pcre_exec().
2900    
2901    
2902     Version 0.93 15-Sep-97
2903     ----------------------
2904    
2905     1. /(b)|(:+)/ was computing an incorrect first character.
2906    
2907     2. Add pcre_study() to the API and the passing of pcre_extra to pcre_exec(),
2908     but not actually doing anything yet.
2909    
2910     3. Treat "-" characters in classes that cannot be part of ranges as literals,
2911     as Perl does (e.g. [-az] or [az-]).
2912    
2913     4. Set the anchored flag if a branch starts with .* or .*? because that tests
2914     all possible positions.
2915    
2916     5. Split up into different modules to avoid including unneeded functions in a
2917     compiled binary. However, compile and exec are still in one module. The "study"
2918     function is split off.
2919    
2920     6. The character tables are now in a separate module whose source is generated
2921     by an auxiliary program - but can then be edited by hand if required. There are
2922     now no calls to isalnum(), isspace(), isdigit(), isxdigit(), tolower() or
2923     toupper() in the code.
2924    
2925     7. Turn the malloc/free funtions variables into pcre_malloc and pcre_free and
2926     make them global. Abolish the function for setting them, as the caller can now
2927     set them directly.
2928    
2929    
2930     Version 0.92 11-Sep-97
2931     ----------------------
2932    
2933     1. A repeat with a fixed maximum and a minimum of 1 for an ordinary character
2934     (e.g. /a{1,3}/) was broken (I mis-optimized it).
2935    
2936     2. Caseless matching was not working in character classes if the characters in
2937     the pattern were in upper case.
2938    
2939     3. Make ranges like [W-c] work in the same way as Perl for caseless matching.
2940    
2941     4. Make PCRE_ANCHORED public and accept as a compile option.
2942    
2943     5. Add an options word to pcre_exec() and accept PCRE_ANCHORED and
2944     PCRE_CASELESS at run time. Add escapes \A and \I to pcretest to cause it to
2945     pass them.
2946    
2947     6. Give an error if bad option bits passed at compile or run time.
2948    
2949     7. Add PCRE_MULTILINE at compile and exec time, and (?m) as well. Add \M to
2950     pcretest to cause it to pass that flag.
2951    
2952     8. Add pcre_info(), to get the number of identifying subpatterns, the stored
2953     options, and the first character, if set.
2954    
2955     9. Recognize C+ or C{n,m} where n >= 1 as providing a fixed starting character.
2956    
2957    
2958     Version 0.91 10-Sep-97
2959     ----------------------
2960    
2961     1. PCRE was failing to diagnose unlimited repeats of subpatterns that could
2962     match the empty string as in /(a*)*/. It was looping and ultimately crashing.
2963    
2964     2. PCRE was looping on encountering an indefinitely repeated back reference to
2965     a subpattern that had matched an empty string, e.g. /(a|)\1*/. It now does what
2966     Perl does - treats the match as successful.
2967    
2968     ****

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