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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 217 - (hide annotations) (download)
Wed Aug 15 15:21:04 2007 UTC (6 years, 11 months ago) by ph10
File size: 136239 byte(s)
Crash in debug output when printing the name of a Unicode property.

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

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