ViewVC logotype

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log

Revision 224 - (hide annotations) (download)
Mon Aug 20 12:58:29 2007 UTC (7 years, 8 months ago) by ph10
File size: 137674 byte(s)
Fix loop for group with possible zero repeat containing \p or \P.

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


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

ViewVC Help
Powered by ViewVC 1.1.12