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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 218 - (hide annotations) (download)
Thu Aug 16 10:13:23 2007 UTC (7 years, 1 month ago) by ph10
File size: 137184 byte(s)
Fix several compile bugs involving repeated \p or \P items.

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

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