ViewVC logotype

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log

Revision 223 - (hide annotations) (download)
Mon Aug 20 11:07:53 2007 UTC (7 years, 8 months ago) by ph10
File size: 137546 byte(s)
Fix loop for classes containing \p or \P and just one ascii character.

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


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

ViewVC Help
Powered by ViewVC 1.1.12