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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 11 - (show annotations) (download)
Sat Feb 24 21:38:17 2007 UTC (7 years, 8 months ago) by nigel
File size: 10030 byte(s)
Load pcre-1.04 into code/trunk.

1 ChangeLog for PCRE
2 ------------------
3
4
5 Version 1.04 19-Dec-97
6 ----------------------
7
8 1. Corrected the man page, where some "const" qualifiers had been omitted.
9
10 2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
11 input syntax.
12
13 3. Fixed memory leak which occurred when a regex with back references was
14 matched with an offsets vector that wasn't big enough. The temporary memory
15 that is used in this case wasn't being freed if the match failed.
16
17 4. Tidied pcretest to ensure it frees memory that it gets.
18
19 5. Temporary memory was being obtained in the case where the passed offsets
20 vector was exactly big enough.
21
22 6. Corrected definition of offsetof() from change 5 below.
23
24 7. I had screwed up change 6 below and broken the rules for the use of
25 setjmp(). Now fixed.
26
27
28 Version 1.03 18-Dec-97
29 ----------------------
30
31 1. A erroneous regex with a missing opening parenthesis was correctly
32 diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
33 on some systems.
34
35 2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
36 it was reported that one broken compiler failed on the former because "code" is
37 also an independent variable.
38
39 3. The erroneous regex a[]b caused an array overrun reference.
40
41 4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
42 fail on data ending with that character. (It was going on too far, and checking
43 the next character, typically a binary zero.) This was specific to the
44 optimized code for single-character negative classes.
45
46 5. Added a contributed patch from the TIN world which does the following:
47
48 + Add an undef for memmove, in case the the system defines a macro for it.
49
50 + Add a definition of offsetof(), in case there isn't one. (I don't know
51 the reason behind this - offsetof() is part of the ANSI standard - but
52 it does no harm).
53
54 + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
55 most of the places where whitespace preceded '#'. I have given up and
56 allowed the remaining 2 cases to be at the margin.
57
58 + Rename some variables in pcre to eliminate shadowing. This seems very
59 pedantic, but does no harm, of course.
60
61 6. Moved the call to setjmp() into its own function, to get rid of warnings
62 from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
63
64 7. Constructs such as \d{8,} were compiling into the equivalent of
65 \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
66 outcome, but in this particular case used more store than had been allocated,
67 which caused the bug to be discovered because it threw up an internal error.
68
69 8. The debugging code in both pcre and pcretest for outputting the compiled
70 form of a regex was going wrong in the case of back references followed by
71 curly-bracketed repeats.
72
73
74 Version 1.02 12-Dec-97
75 ----------------------
76
77 1. Typos in pcre.3 and comments in the source fixed.
78
79 2. Applied a contributed patch to get rid of places where it used to remove
80 'const' from variables, and fixed some signed/unsigned and uninitialized
81 variable warnings.
82
83 3. Added the "runtest" target to Makefile.
84
85 4. Set default compiler flag to -O2 rather than just -O.
86
87
88 Version 1.01 19-Nov-97
89 ----------------------
90
91 1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
92 like /([ab]*)*/, that is, for classes with more than one character in them.
93
94 2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
95 as /((?>a*))*/ (a PCRE_EXTRA facility).
96
97
98 Version 1.00 18-Nov-97
99 ----------------------
100
101 1. Added compile-time macros to support systems such as SunOS4 which don't have
102 memmove() or strerror() but have other things that can be used instead.
103
104 2. Arranged that "make clean" removes the executables.
105
106
107 Version 0.99 27-Oct-97
108 ----------------------
109
110 1. Fixed bug in code for optimizing classes with only one character. It was
111 initializing a 32-byte map regardless, which could cause it to run off the end
112 of the memory it had got.
113
114 2. Added, conditional on PCRE_EXTRA, the proposed (?>REGEX) construction.
115
116
117 Version 0.98 22-Oct-97
118 ----------------------
119
120 1. Fixed bug in code for handling temporary memory usage when there are more
121 back references than supplied space in the ovector. This could cause segfaults.
122
123
124 Version 0.97 21-Oct-97
125 ----------------------
126
127 1. Added the \X "cut" facility, conditional on PCRE_EXTRA.
128
129 2. Optimized negated single characters not to use a bit map.
130
131 3. Brought error texts together as macro definitions; clarified some of them;
132 fixed one that was wrong - it said "range out of order" when it meant "invalid
133 escape sequence".
134
135 4. Changed some char * arguments to const char *.
136
137 5. Added PCRE_NOTBOL and PCRE_NOTEOL (from POSIX).
138
139 6. Added the POSIX-style API wrapper in pcreposix.a and testing facilities in
140 pcretest.
141
142
143 Version 0.96 16-Oct-97
144 ----------------------
145
146 1. Added a simple "pgrep" utility to the distribution.
147
148 2. Fixed an incompatibility with Perl: "{" is now treated as a normal character
149 unless it appears in one of the precise forms "{ddd}", "{ddd,}", or "{ddd,ddd}"
150 where "ddd" means "one or more decimal digits".
151
152 3. Fixed serious bug. If a pattern had a back reference, but the call to
153 pcre_exec() didn't supply a large enough ovector to record the related
154 identifying subpattern, the match always failed. PCRE now remembers the number
155 of the largest back reference, and gets some temporary memory in which to save
156 the offsets during matching if necessary, in order to ensure that
157 backreferences always work.
158
159 4. Increased the compatibility with Perl in a number of ways:
160
161 (a) . no longer matches \n by default; an option PCRE_DOTALL is provided
162 to request this handling. The option can be set at compile or exec time.
163
164 (b) $ matches before a terminating newline by default; an option
165 PCRE_DOLLAR_ENDONLY is provided to override this (but not in multiline
166 mode). The option can be set at compile or exec time.
167
168 (c) The handling of \ followed by a digit other than 0 is now supposed to be
169 the same as Perl's. If the decimal number it represents is less than 10
170 or there aren't that many previous left capturing parentheses, an octal
171 escape is read. Inside a character class, it's always an octal escape,
172 even if it is a single digit.
173
174 (d) An escaped but undefined alphabetic character is taken as a literal,
175 unless PCRE_EXTRA is set. Currently this just reserves the remaining
176 escapes.
177
178 (e) {0} is now permitted. (The previous item is removed from the compiled
179 pattern).
180
181 5. Changed all the names of code files so that the basic parts are no longer
182 than 10 characters, and abolished the teeny "globals.c" file.
183
184 6. Changed the handling of character classes; they are now done with a 32-byte
185 bit map always.
186
187 7. Added the -d and /D options to pcretest to make it possible to look at the
188 internals of compilation without having to recompile pcre.
189
190
191 Version 0.95 23-Sep-97
192 ----------------------
193
194 1. Fixed bug in pre-pass concerning escaped "normal" characters such as \x5c or
195 \x20 at the start of a run of normal characters. These were being treated as
196 real characters, instead of the source characters being re-checked.
197
198
199 Version 0.94 18-Sep-97
200 ----------------------
201
202 1. The functions are now thread-safe, with the caveat that the global variables
203 containing pointers to malloc() and free() or alternative functions are the
204 same for all threads.
205
206 2. Get pcre_study() to generate a bitmap of initial characters for non-
207 anchored patterns when this is possible, and use it if passed to pcre_exec().
208
209
210 Version 0.93 15-Sep-97
211 ----------------------
212
213 1. /(b)|(:+)/ was computing an incorrect first character.
214
215 2. Add pcre_study() to the API and the passing of pcre_extra to pcre_exec(),
216 but not actually doing anything yet.
217
218 3. Treat "-" characters in classes that cannot be part of ranges as literals,
219 as Perl does (e.g. [-az] or [az-]).
220
221 4. Set the anchored flag if a branch starts with .* or .*? because that tests
222 all possible positions.
223
224 5. Split up into different modules to avoid including unneeded functions in a
225 compiled binary. However, compile and exec are still in one module. The "study"
226 function is split off.
227
228 6. The character tables are now in a separate module whose source is generated
229 by an auxiliary program - but can then be edited by hand if required. There are
230 now no calls to isalnum(), isspace(), isdigit(), isxdigit(), tolower() or
231 toupper() in the code.
232
233 7. Turn the malloc/free funtions variables into pcre_malloc and pcre_free and
234 make them global. Abolish the function for setting them, as the caller can now
235 set them directly.
236
237
238 Version 0.92 11-Sep-97
239 ----------------------
240
241 1. A repeat with a fixed maximum and a minimum of 1 for an ordinary character
242 (e.g. /a{1,3}/) was broken (I mis-optimized it).
243
244 2. Caseless matching was not working in character classes if the characters in
245 the pattern were in upper case.
246
247 3. Make ranges like [W-c] work in the same way as Perl for caseless matching.
248
249 4. Make PCRE_ANCHORED public and accept as a compile option.
250
251 5. Add an options word to pcre_exec() and accept PCRE_ANCHORED and
252 PCRE_CASELESS at run time. Add escapes \A and \I to pcretest to cause it to
253 pass them.
254
255 6. Give an error if bad option bits passed at compile or run time.
256
257 7. Add PCRE_MULTILINE at compile and exec time, and (?m) as well. Add \M to
258 pcretest to cause it to pass that flag.
259
260 8. Add pcre_info(), to get the number of identifying subpatterns, the stored
261 options, and the first character, if set.
262
263 9. Recognize C+ or C{n,m} where n >= 1 as providing a fixed starting character.
264
265
266 Version 0.91 10-Sep-97
267 ----------------------
268
269 1. PCRE was failing to diagnose unlimited repeats of subpatterns that could
270 match the empty string as in /(a*)*/. It was looping and ultimately crashing.
271
272 2. PCRE was looping on encountering an indefinitely repeated back reference to
273 a subpattern that had matched an empty string, e.g. /(a|)\1*/. It now does what
274 Perl does - treats the match as successful.
275
276 ****

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12