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

Diff of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log | View Patch Patch

revision 3 by nigel, Sat Feb 24 21:38:01 2007 UTC revision 39 by nigel, Sat Feb 24 21:39:13 2007 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4    
5    Version 2.08 31-Aug-99
6    ----------------------
7    
8    1. When startoffset was not zero and the pattern began with ".*", PCRE was not
9    trying to match at the startoffset position, but instead was moving forward to
10    the next newline as if a previous match had failed.
11    
12    2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
13    and could get into a loop if a null string was matched other than at the start
14    of the subject.
15    
16    3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
17    be distinguished at compile time, and for completeness also added PCRE_DATE.
18    
19    5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
20    in GnuWin32 environments.
21    
22    
23    Version 2.07 29-Jul-99
24    ----------------------
25    
26    1. The documentation is now supplied in plain text form and HTML as well as in
27    the form of man page sources.
28    
29    2. C++ compilers don't like assigning (void *) values to other pointer types.
30    In particular this affects malloc(). Although there is no problem in Standard
31    C, I've put in casts to keep C++ compilers happy.
32    
33    3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
34    should be (const char *).
35    
36    4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
37    be useful for non-Unix systems who don't want to bother with the POSIX stuff.
38    However, I haven't made this a standard facility. The documentation doesn't
39    mention it, and the Makefile doesn't support it.
40    
41    5. The Makefile now contains an "install" target, with editable destinations at
42    the top of the file. The pcretest program is not installed.
43    
44    6. pgrep -V now gives the PCRE version number and date.
45    
46    7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
47    causing the entire string to be ignored, instead of just the last character.
48    
49    8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
50    non-matching string, it can take a very, very long time, even for strings of
51    quite modest length, because of the nested recursion. PCRE now does better in
52    some of these cases. It does this by remembering the last required literal
53    character in the pattern, and pre-searching the subject to ensure it is present
54    before running the real match. In other words, it applies a heuristic to detect
55    some types of certain failure quickly, and in the above example, if presented
56    with a string that has no trailing " it gives "no match" very quickly.
57    
58    9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
59    other alternatives are tried instead.
60    
61    
62    Version 2.06 09-Jun-99
63    ----------------------
64    
65    1. Change pcretest's output for amount of store used to show just the code
66    space, because the remainder (the data block) varies in size between 32-bit and
67    64-bit systems.
68    
69    2. Added an extra argument to pcre_exec() to supply an offset in the subject to
70    start matching at. This allows lookbehinds to work when searching for multiple
71    occurrences in a string.
72    
73    3. Added additional options to pcretest for testing multiple occurrences:
74    
75       /+   outputs the rest of the string that follows a match
76       /g   loops for multiple occurrences, using the new startoffset argument
77       /G   loops for multiple occurrences by passing an incremented pointer
78    
79    4. PCRE wasn't doing the "first character" optimization for patterns starting
80    with \b or \B, though it was doing it for other lookbehind assertions. That is,
81    it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
82    the letter 'x'. On long subject strings, this gives a significant speed-up.
83    
84    
85    Version 2.05 21-Apr-99
86    ----------------------
87    
88    1. Changed the type of magic_number from int to long int so that it works
89    properly on 16-bit systems.
90    
91    2. Fixed a bug which caused patterns starting with .* not to work correctly
92    when the subject string contained newline characters. PCRE was assuming
93    anchoring for such patterns in all cases, which is not correct because .* will
94    not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
95    DOTALL is set at top level; otherwise it knows that patterns starting with .*
96    must be retried after every newline in the subject.
97    
98    
99    Version 2.04 18-Feb-99
100    ----------------------
101    
102    1. For parenthesized subpatterns with repeats whose minimum was zero, the
103    computation of the store needed to hold the pattern was incorrect (too large).
104    If such patterns were nested a few deep, this could multiply and become a real
105    problem.
106    
107    2. Added /M option to pcretest to show the memory requirement of a specific
108    pattern. Made -m a synonym of -s (which does this globally) for compatibility.
109    
110    3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
111    compiled in such a way that the backtracking after subsequent failure was
112    pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
113    ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
114    
115    
116    Version 2.03 02-Feb-99
117    ----------------------
118    
119    1. Fixed typo and small mistake in man page.
120    
121    2. Added 4th condition (GPL supersedes if conflict) and created separate
122    LICENCE file containing the conditions.
123    
124    3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
125    Perl, that is the internal \ allows the delimiter to be included in the
126    pattern. Locked out the use of \ as a delimiter. If \ immediately follows
127    the final delimiter, add \ to the end of the pattern (to test the error).
128    
129    4. Added the convenience functions for extracting substrings after a successful
130    match. Updated pcretest to make it able to test these functions.
131    
132    
133    Version 2.02 14-Jan-99
134    ----------------------
135    
136    1. Initialized the working variables associated with each extraction so that
137    their saving and restoring doesn't refer to uninitialized store.
138    
139    2. Put dummy code into study.c in order to trick the optimizer of the IBM C
140    compiler for OS/2 into generating correct code. Apparently IBM isn't going to
141    fix the problem.
142    
143    3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
144    calls, and wasn't printing the correct value for compiling calls. Increased the
145    default value of LOOPREPEAT, and the number of significant figures in the
146    times.
147    
148    4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
149    
150    5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
151    a building problem on Windows NT with a FAT file system.
152    
153    
154    Version 2.01 21-Oct-98
155    ----------------------
156    
157    1. Changed the API for pcre_compile() to allow for the provision of a pointer
158    to character tables built by pcre_maketables() in the current locale. If NULL
159    is passed, the default tables are used.
160    
161    
162    Version 2.00 24-Sep-98
163    ----------------------
164    
165    1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
166    it any more.
167    
168    2. Allow quantification of (?>) groups, and make it work correctly.
169    
170    3. The first character computation wasn't working for (?>) groups.
171    
172    4. Correct the implementation of \Z (it is permitted to match on the \n at the
173    end of the subject) and add 5.005's \z, which really does match only at the
174    very end of the subject.
175    
176    5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
177    
178    6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
179    DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
180    localized options. All options to pcre_study() were also removed.
181    
182    7. Add other new features from 5.005:
183    
184       $(?<=           positive lookbehind
185       $(?<!           negative lookbehind
186       (?imsx-imsx)    added the unsetting capability
187                       such a setting is global if at outer level; local otherwise
188       (?imsx-imsx:)   non-capturing groups with option setting
189       (?(cond)re|re)  conditional pattern matching
190    
191       A backreference to itself in a repeated group matches the previous
192       captured string.
193    
194    8. General tidying up of studying (both automatic and via "study")
195    consequential on the addition of new assertions.
196    
197    9. As in 5.005, unlimited repeated groups that could match an empty substring
198    are no longer faulted at compile time. Instead, the loop is forcibly broken at
199    runtime if any iteration does actually match an empty substring.
200    
201    10. Include the RunTest script in the distribution.
202    
203    11. Added tests from the Perl 5.005_02 distribution. This showed up a few
204    discrepancies, some of which were old and were also with respect to 5.004. They
205    have now been fixed.
206    
207    
208    Version 1.09 28-Apr-98
209    ----------------------
210    
211    1. A negated single character class followed by a quantifier with a minimum
212    value of one (e.g.  [^x]{1,6}  ) was not compiled correctly. This could lead to
213    program crashes, or just wrong answers. This did not apply to negated classes
214    containing more than one character, or to minima other than one.
215    
216    
217    Version 1.08 27-Mar-98
218    ----------------------
219    
220    1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
221    
222    2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
223    latter must appear before anything that relies on it in the pattern.
224    
225    
226    Version 1.07 16-Feb-98
227    ----------------------
228    
229    1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
230    repeat of a potentially empty string).
231    
232    
233    Version 1.06 23-Jan-98
234    ----------------------
235    
236    1. Added Markus Oberhumer's little patches for C++.
237    
238    2. Literal strings longer than 255 characters were broken.
239    
240    
241    Version 1.05 23-Dec-97
242    ----------------------
243    
244    1. Negated character classes containing more than one character were failing if
245    PCRE_CASELESS was set at run time.
246    
247    
248    Version 1.04 19-Dec-97
249    ----------------------
250    
251    1. Corrected the man page, where some "const" qualifiers had been omitted.
252    
253    2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
254    input syntax.
255    
256    3. Fixed memory leak which occurred when a regex with back references was
257    matched with an offsets vector that wasn't big enough. The temporary memory
258    that is used in this case wasn't being freed if the match failed.
259    
260    4. Tidied pcretest to ensure it frees memory that it gets.
261    
262    5. Temporary memory was being obtained in the case where the passed offsets
263    vector was exactly big enough.
264    
265    6. Corrected definition of offsetof() from change 5 below.
266    
267    7. I had screwed up change 6 below and broken the rules for the use of
268    setjmp(). Now fixed.
269    
270    
271    Version 1.03 18-Dec-97
272    ----------------------
273    
274    1. A erroneous regex with a missing opening parenthesis was correctly
275    diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
276    on some systems.
277    
278    2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
279    it was reported that one broken compiler failed on the former because "code" is
280    also an independent variable.
281    
282    3. The erroneous regex a[]b caused an array overrun reference.
283    
284    4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
285    fail on data ending with that character. (It was going on too far, and checking
286    the next character, typically a binary zero.) This was specific to the
287    optimized code for single-character negative classes.
288    
289    5. Added a contributed patch from the TIN world which does the following:
290    
291      + Add an undef for memmove, in case the the system defines a macro for it.
292    
293      + Add a definition of offsetof(), in case there isn't one. (I don't know
294        the reason behind this - offsetof() is part of the ANSI standard - but
295        it does no harm).
296    
297      + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
298        most of the places where whitespace preceded '#'. I have given up and
299        allowed the remaining 2 cases to be at the margin.
300    
301      + Rename some variables in pcre to eliminate shadowing. This seems very
302        pedantic, but does no harm, of course.
303    
304    6. Moved the call to setjmp() into its own function, to get rid of warnings
305    from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
306    
307    7. Constructs such as \d{8,} were compiling into the equivalent of
308    \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
309    outcome, but in this particular case used more store than had been allocated,
310    which caused the bug to be discovered because it threw up an internal error.
311    
312    8. The debugging code in both pcre and pcretest for outputting the compiled
313    form of a regex was going wrong in the case of back references followed by
314    curly-bracketed repeats.
315    
316    
317    Version 1.02 12-Dec-97
318    ----------------------
319    
320    1. Typos in pcre.3 and comments in the source fixed.
321    
322    2. Applied a contributed patch to get rid of places where it used to remove
323    'const' from variables, and fixed some signed/unsigned and uninitialized
324    variable warnings.
325    
326    3. Added the "runtest" target to Makefile.
327    
328    4. Set default compiler flag to -O2 rather than just -O.
329    
330    
331    Version 1.01 19-Nov-97
332    ----------------------
333    
334    1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
335    like /([ab]*)*/, that is, for classes with more than one character in them.
336    
337    2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
338    as /((?>a*))*/ (a PCRE_EXTRA facility).
339    
340    
341    Version 1.00 18-Nov-97
342    ----------------------
343    
344    1. Added compile-time macros to support systems such as SunOS4 which don't have
345    memmove() or strerror() but have other things that can be used instead.
346    
347    2. Arranged that "make clean" removes the executables.
348    
349    
350  Version 0.99 27-Oct-97  Version 0.99 27-Oct-97
351  ----------------------  ----------------------
352    

Legend:
Removed from v.3  
changed lines
  Added in v.39

  ViewVC Help
Powered by ViewVC 1.1.5