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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 5 - (show annotations) (download)
Sat Feb 24 21:38:05 2007 UTC (7 years, 1 month ago) by nigel
File size: 6872 byte(s)
Load pcre-1.01 into code/trunk.

1 ChangeLog for PCRE
2 ------------------
3
4
5 Version 1.01 19-Nov-97
6 ----------------------
7
8 1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
9 like /([ab]*)*/, that is, for classes with more than one character in them.
10
11 2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
12 as /((?>a*))*/ (a PCRE_EXTRA facility).
13
14
15 Version 1.00 18-Nov-97
16 ----------------------
17
18 1. Added compile-time macros to support systems such as SunOS4 which don't have
19 memmove() or strerror() but have other things that can be used instead.
20
21 2. Arranged that "make clean" removes the executables.
22
23
24 Version 0.99 27-Oct-97
25 ----------------------
26
27 1. Fixed bug in code for optimizing classes with only one character. It was
28 initializing a 32-byte map regardless, which could cause it to run off the end
29 of the memory it had got.
30
31 2. Added, conditional on PCRE_EXTRA, the proposed (?>REGEX) construction.
32
33
34 Version 0.98 22-Oct-97
35 ----------------------
36
37 1. Fixed bug in code for handling temporary memory usage when there are more
38 back references than supplied space in the ovector. This could cause segfaults.
39
40
41 Version 0.97 21-Oct-97
42 ----------------------
43
44 1. Added the \X "cut" facility, conditional on PCRE_EXTRA.
45
46 2. Optimized negated single characters not to use a bit map.
47
48 3. Brought error texts together as macro definitions; clarified some of them;
49 fixed one that was wrong - it said "range out of order" when it meant "invalid
50 escape sequence".
51
52 4. Changed some char * arguments to const char *.
53
54 5. Added PCRE_NOTBOL and PCRE_NOTEOL (from POSIX).
55
56 6. Added the POSIX-style API wrapper in pcreposix.a and testing facilities in
57 pcretest.
58
59
60 Version 0.96 16-Oct-97
61 ----------------------
62
63 1. Added a simple "pgrep" utility to the distribution.
64
65 2. Fixed an incompatibility with Perl: "{" is now treated as a normal character
66 unless it appears in one of the precise forms "{ddd}", "{ddd,}", or "{ddd,ddd}"
67 where "ddd" means "one or more decimal digits".
68
69 3. Fixed serious bug. If a pattern had a back reference, but the call to
70 pcre_exec() didn't supply a large enough ovector to record the related
71 identifying subpattern, the match always failed. PCRE now remembers the number
72 of the largest back reference, and gets some temporary memory in which to save
73 the offsets during matching if necessary, in order to ensure that
74 backreferences always work.
75
76 4. Increased the compatibility with Perl in a number of ways:
77
78 (a) . no longer matches \n by default; an option PCRE_DOTALL is provided
79 to request this handling. The option can be set at compile or exec time.
80
81 (b) $ matches before a terminating newline by default; an option
82 PCRE_DOLLAR_ENDONLY is provided to override this (but not in multiline
83 mode). The option can be set at compile or exec time.
84
85 (c) The handling of \ followed by a digit other than 0 is now supposed to be
86 the same as Perl's. If the decimal number it represents is less than 10
87 or there aren't that many previous left capturing parentheses, an octal
88 escape is read. Inside a character class, it's always an octal escape,
89 even if it is a single digit.
90
91 (d) An escaped but undefined alphabetic character is taken as a literal,
92 unless PCRE_EXTRA is set. Currently this just reserves the remaining
93 escapes.
94
95 (e) {0} is now permitted. (The previous item is removed from the compiled
96 pattern).
97
98 5. Changed all the names of code files so that the basic parts are no longer
99 than 10 characters, and abolished the teeny "globals.c" file.
100
101 6. Changed the handling of character classes; they are now done with a 32-byte
102 bit map always.
103
104 7. Added the -d and /D options to pcretest to make it possible to look at the
105 internals of compilation without having to recompile pcre.
106
107
108 Version 0.95 23-Sep-97
109 ----------------------
110
111 1. Fixed bug in pre-pass concerning escaped "normal" characters such as \x5c or
112 \x20 at the start of a run of normal characters. These were being treated as
113 real characters, instead of the source characters being re-checked.
114
115
116 Version 0.94 18-Sep-97
117 ----------------------
118
119 1. The functions are now thread-safe, with the caveat that the global variables
120 containing pointers to malloc() and free() or alternative functions are the
121 same for all threads.
122
123 2. Get pcre_study() to generate a bitmap of initial characters for non-
124 anchored patterns when this is possible, and use it if passed to pcre_exec().
125
126
127 Version 0.93 15-Sep-97
128 ----------------------
129
130 1. /(b)|(:+)/ was computing an incorrect first character.
131
132 2. Add pcre_study() to the API and the passing of pcre_extra to pcre_exec(),
133 but not actually doing anything yet.
134
135 3. Treat "-" characters in classes that cannot be part of ranges as literals,
136 as Perl does (e.g. [-az] or [az-]).
137
138 4. Set the anchored flag if a branch starts with .* or .*? because that tests
139 all possible positions.
140
141 5. Split up into different modules to avoid including unneeded functions in a
142 compiled binary. However, compile and exec are still in one module. The "study"
143 function is split off.
144
145 6. The character tables are now in a separate module whose source is generated
146 by an auxiliary program - but can then be edited by hand if required. There are
147 now no calls to isalnum(), isspace(), isdigit(), isxdigit(), tolower() or
148 toupper() in the code.
149
150 7. Turn the malloc/free funtions variables into pcre_malloc and pcre_free and
151 make them global. Abolish the function for setting them, as the caller can now
152 set them directly.
153
154
155 Version 0.92 11-Sep-97
156 ----------------------
157
158 1. A repeat with a fixed maximum and a minimum of 1 for an ordinary character
159 (e.g. /a{1,3}/) was broken (I mis-optimized it).
160
161 2. Caseless matching was not working in character classes if the characters in
162 the pattern were in upper case.
163
164 3. Make ranges like [W-c] work in the same way as Perl for caseless matching.
165
166 4. Make PCRE_ANCHORED public and accept as a compile option.
167
168 5. Add an options word to pcre_exec() and accept PCRE_ANCHORED and
169 PCRE_CASELESS at run time. Add escapes \A and \I to pcretest to cause it to
170 pass them.
171
172 6. Give an error if bad option bits passed at compile or run time.
173
174 7. Add PCRE_MULTILINE at compile and exec time, and (?m) as well. Add \M to
175 pcretest to cause it to pass that flag.
176
177 8. Add pcre_info(), to get the number of identifying subpatterns, the stored
178 options, and the first character, if set.
179
180 9. Recognize C+ or C{n,m} where n >= 1 as providing a fixed starting character.
181
182
183 Version 0.91 10-Sep-97
184 ----------------------
185
186 1. PCRE was failing to diagnose unlimited repeats of subpatterns that could
187 match the empty string as in /(a*)*/. It was looping and ultimately crashing.
188
189 2. PCRE was looping on encountering an indefinitely repeated back reference to
190 a subpattern that had matched an empty string, e.g. /(a|)\1*/. It now does what
191 Perl does - treats the match as successful.
192
193 ****

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12