/[pcre]/code/trunk/doc/pcreunicode.3
ViewVC logotype

Contents of /code/trunk/doc/pcreunicode.3

Parent Directory Parent Directory | Revision Log Revision Log


Revision 1107 - (show annotations)
Tue Oct 16 15:56:51 2012 UTC (6 years, 10 months ago) by chpe
File size: 12006 byte(s)
pcre32: docs: Document that non-character codepoints are excluded

Document that UTF-8/16/32 validity checks now exclude the non-character
code points, which are U+FDD0..U+FDEF and the last two code points in each
plane, U+??FFFE and U+??FFFF.
1 .TH PCREUNICODE 3 "25 September 2012" "PCRE 8.32"
2 .SH NAME
3 PCRE - Perl-compatible regular expressions
4 .SH "UTF-8, UTF-16, AND UNICODE PROPERTY SUPPORT"
5 .rs
6 .sp
7 From Release 8.30, in addition to its previous UTF-8 support, PCRE also
8 supports UTF-16 by means of a separate 16-bit library. This can be built as
9 well as, or instead of, the 8-bit library.
10 .P
11 From Release 8.32, in addition to its previous UTF-8 and UTF-16 support,
12 PCRE also supports UTF-32 by means of a separate 32-bit library. This can be
13 built as well as, or instead of, the 8-bit and 16-bit libraries.
14 .
15 .SH "UTF-8 SUPPORT"
16 .rs
17 .sp
18 In order process UTF-8 strings, you must build PCRE's 8-bit library with UTF
19 support, and, in addition, you must call
20 .\" HREF
21 \fBpcre_compile()\fP
22 .\"
23 with the PCRE_UTF8 option flag, or the pattern must start with the sequence
24 (*UTF8). When either of these is the case, both the pattern and any subject
25 strings that are matched against it are treated as UTF-8 strings instead of
26 strings of 1-byte characters.
27 .
28 .
29 .SH "UTF-16 SUPPORT"
30 .rs
31 .sp
32 In order process UTF-16 strings, you must build PCRE's 16-bit library with UTF
33 support, and, in addition, you must call
34 .\" HTML <a href="pcre_compile.html">
35 .\" </a>
36 \fBpcre16_compile()\fP
37 .\"
38 with the PCRE_UTF16 option flag, or the pattern must start with the sequence
39 (*UTF16). When either of these is the case, both the pattern and any subject
40 strings that are matched against it are treated as UTF-16 strings instead of
41 strings of 16-bit characters.
42 .
43 .
44 .SH "UTF-32 SUPPORT"
45 .rs
46 .sp
47 In order process UTF-32 strings, you must build PCRE's 32-bit library with UTF
48 support, and, in addition, you must call
49 .\" HTML <a href="pcre_compile.html">
50 .\" </a>
51 \fBpcre32_compile()\fP
52 .\"
53 with the PCRE_UTF32 option flag, or the pattern must start with the sequence
54 (*UTF32). When either of these is the case, both the pattern and any subject
55 strings that are matched against it are treated as UTF-32 strings instead of
56 strings of 32-bit characters.
57 .
58 .
59 .SH "UTF SUPPORT OVERHEAD"
60 .rs
61 .sp
62 If you compile PCRE with UTF support, but do not use it at run time, the
63 library will be a bit bigger, but the additional run time overhead is limited
64 to testing the PCRE_UTF[8|16|32] flag occasionally, so should not be very big.
65 .
66 .
67 .SH "UNICODE PROPERTY SUPPORT"
68 .rs
69 .sp
70 If PCRE is built with Unicode character property support (which implies UTF
71 support), the escape sequences \ep{..}, \eP{..}, and \eX can be used.
72 The available properties that can be tested are limited to the general
73 category properties such as Lu for an upper case letter or Nd for a decimal
74 number, the Unicode script names such as Arabic or Han, and the derived
75 properties Any and L&. A full list is given in the
76 .\" HREF
77 \fBpcrepattern\fP
78 .\"
79 documentation. Only the short names for properties are supported. For example,
80 \ep{L} matches a letter. Its Perl synonym, \ep{Letter}, is not supported.
81 Furthermore, in Perl, many properties may optionally be prefixed by "Is", for
82 compatibility with Perl 5.6. PCRE does not support this.
83 .
84 .
85 .\" HTML <a name="utf8strings"></a>
86 .SS "Validity of UTF-8 strings"
87 .rs
88 .sp
89 When you set the PCRE_UTF8 flag, the byte strings passed as patterns and
90 subjects are (by default) checked for validity on entry to the relevant
91 functions. The entire string is checked before any other processing takes
92 place. From release 7.3 of PCRE, the check is according the rules of RFC 3629,
93 which are themselves derived from the Unicode specification. Earlier releases
94 of PCRE followed the rules of RFC 2279, which allows the full range of 31-bit
95 values (0 to 0x7FFFFFFF). The current check allows only values in the range U+0
96 to U+10FFFF, excluding the surrogate area, and the non-characters.
97 .P
98 Excluded code points are the "Surrogate Area" of Unicode. They are reserved
99 for use by UTF-16, where they are used in pairs to encode codepoints with
100 values greater than 0xFFFF. The code points that are encoded by UTF-16 pairs
101 are available independently in the UTF-8 encoding. (In other words, the whole
102 surrogate thing is a fudge for UTF-16 which unfortunately messes up UTF-8.)
103 .P
104 Also excluded are the "Non-Characters" code points, which are U+FDD0 to U+FDEF
105 and the last two code points in each plane, U+??FFFE and U+??FFFF.
106 .P
107 If an invalid UTF-8 string is passed to PCRE, an error return is given. At
108 compile time, the only additional information is the offset to the first byte
109 of the failing character. The run-time functions \fBpcre_exec()\fP and
110 \fBpcre_dfa_exec()\fP also pass back this information, as well as a more
111 detailed reason code if the caller has provided memory in which to do this.
112 .P
113 In some situations, you may already know that your strings are valid, and
114 therefore want to skip these checks in order to improve performance, for
115 example in the case of a long subject string that is being scanned repeatedly
116 with different patterns. If you set the PCRE_NO_UTF8_CHECK flag at compile time
117 or at run time, PCRE assumes that the pattern or subject it is given
118 (respectively) contains only valid UTF-8 codes. In this case, it does not
119 diagnose an invalid UTF-8 string.
120 .P
121 If you pass an invalid UTF-8 string when PCRE_NO_UTF8_CHECK is set, what
122 happens depends on why the string is invalid. If the string conforms to the
123 "old" definition of UTF-8 (RFC 2279), it is processed as a string of characters
124 in the range 0 to 0x7FFFFFFF by \fBpcre_dfa_exec()\fP and the interpreted
125 version of \fBpcre_exec()\fP. In other words, apart from the initial validity
126 test, these functions (when in UTF-8 mode) handle strings according to the more
127 liberal rules of RFC 2279. However, the just-in-time (JIT) optimization for
128 \fBpcre_exec()\fP supports only RFC 3629. If you are using JIT optimization, or
129 if the string does not even conform to RFC 2279, the result is undefined. Your
130 program may crash.
131 .P
132 If you want to process strings of values in the full range 0 to 0x7FFFFFFF,
133 encoded in a UTF-8-like manner as per the old RFC, you can set
134 PCRE_NO_UTF8_CHECK to bypass the more restrictive test. However, in this
135 situation, you will have to apply your own validity check, and avoid the use of
136 JIT optimization.
137 .
138 .
139 .\" HTML <a name="utf16strings"></a>
140 .SS "Validity of UTF-16 strings"
141 .rs
142 .sp
143 When you set the PCRE_UTF16 flag, the strings of 16-bit data units that are
144 passed as patterns and subjects are (by default) checked for validity on entry
145 to the relevant functions. Values other than those in the surrogate range
146 U+D800 to U+DFFF are independent code points. Values in the surrogate range
147 must be used in pairs in the correct manner.
148 .P
149 Excluded are the "Non-Characters" code points, which are U+FDD0 to U+FDEF
150 and the last two code points in each plane, U+??FFFE and U+??FFFF.
151 .P
152 If an invalid UTF-16 string is passed to PCRE, an error return is given. At
153 compile time, the only additional information is the offset to the first data
154 unit of the failing character. The run-time functions \fBpcre16_exec()\fP and
155 \fBpcre16_dfa_exec()\fP also pass back this information, as well as a more
156 detailed reason code if the caller has provided memory in which to do this.
157 .P
158 In some situations, you may already know that your strings are valid, and
159 therefore want to skip these checks in order to improve performance. If you set
160 the PCRE_NO_UTF16_CHECK flag at compile time or at run time, PCRE assumes that
161 the pattern or subject it is given (respectively) contains only valid UTF-16
162 sequences. In this case, it does not diagnose an invalid UTF-16 string.
163 .
164 .
165 .\" HTML <a name="utf32strings"></a>
166 .SS "Validity of UTF-32 strings"
167 .rs
168 .sp
169 When you set the PCRE_UTF32 flag, the strings of 32-bit data units that are
170 passed as patterns and subjects are (by default) checked for validity on entry
171 to the relevant functions. This check allows only values in the range U+0
172 to U+10FFFF, excluding the surrogate area U+D800 to U+DFFF, and the
173 "Non-Characters" code points, which are U+FDD0 to U+FDEF and the last two
174 characters in each plane, U+??FFFE and U+??FFFF.
175 .P
176 If an invalid UTF-32 string is passed to PCRE, an error return is given. At
177 compile time, the only additional information is the offset to the first data
178 unit of the failing character. The run-time functions \fBpcre32_exec()\fP and
179 \fBpcre32_dfa_exec()\fP also pass back this information, as well as a more
180 detailed reason code if the caller has provided memory in which to do this.
181 .P
182 In some situations, you may already know that your strings are valid, and
183 therefore want to skip these checks in order to improve performance. If you set
184 the PCRE_NO_UTF32_CHECK flag at compile time or at run time, PCRE assumes that
185 the pattern or subject it is given (respectively) contains only valid UTF-32
186 sequences. In this case, it does not diagnose an invalid UTF-32 string.
187 .
188 .
189 .SS "General comments about UTF modes"
190 .rs
191 .sp
192 1. Codepoints less than 256 can be specified by either braced or unbraced
193 hexadecimal escape sequences (for example, \ex{b3} or \exb3). Larger values
194 have to use braced sequences.
195 .P
196 2. Octal numbers up to \e777 are recognized, and in UTF-8 mode, they match
197 two-byte characters for values greater than \e177.
198 .P
199 3. Repeat quantifiers apply to complete UTF characters, not to individual
200 data units, for example: \ex{100}{3}.
201 .P
202 4. The dot metacharacter matches one UTF character instead of a single data
203 unit.
204 .P
205 5. The escape sequence \eC can be used to match a single byte in UTF-8 mode, or
206 a single 16-bit data unit in UTF-16 mode, or a single 32-bit data unit in
207 UTF-32 mode, but its use can lead to some strange effects because it breaks up
208 multi-unit characters (see the description of \eC in the
209 .\" HREF
210 \fBpcrepattern\fP
211 .\"
212 documentation). The use of \eC is not supported in the alternative matching
213 function \fBpcre[16|32]_dfa_exec()\fP, nor is it supported in UTF mode by the JIT
214 optimization of \fBpcre[16|32]_exec()\fP. If JIT optimization is requested for a
215 UTF pattern that contains \eC, it will not succeed, and so the matching will
216 be carried out by the normal interpretive function.
217 .P
218 6. The character escapes \eb, \eB, \ed, \eD, \es, \eS, \ew, and \eW correctly
219 test characters of any code value, but, by default, the characters that PCRE
220 recognizes as digits, spaces, or word characters remain the same set as in
221 non-UTF mode, all with values less than 256. This remains true even when PCRE
222 is built to include Unicode property support, because to do otherwise would
223 slow down PCRE in many common cases. Note in particular that this applies to
224 \eb and \eB, because they are defined in terms of \ew and \eW. If you really
225 want to test for a wider sense of, say, "digit", you can use explicit Unicode
226 property tests such as \ep{Nd}. Alternatively, if you set the PCRE_UCP option,
227 the way that the character escapes work is changed so that Unicode properties
228 are used to determine which characters match. There are more details in the
229 section on
230 .\" HTML <a href="pcrepattern.html#genericchartypes">
231 .\" </a>
232 generic character types
233 .\"
234 in the
235 .\" HREF
236 \fBpcrepattern\fP
237 .\"
238 documentation.
239 .P
240 7. Similarly, characters that match the POSIX named character classes are all
241 low-valued characters, unless the PCRE_UCP option is set.
242 .P
243 8. However, the horizontal and vertical white space matching escapes (\eh, \eH,
244 \ev, and \eV) do match all the appropriate Unicode characters, whether or not
245 PCRE_UCP is set.
246 .P
247 9. Case-insensitive matching applies only to characters whose values are less
248 than 128, unless PCRE is built with Unicode property support. A few Unicode
249 characters such as Greek sigma have more than two codepoints that are
250 case-equivalent. Up to and including PCRE release 8.31, only one-to-one case
251 mappings were supported, but later releases (with Unicode property support) do
252 treat as case-equivalent all versions of characters such as Greek sigma.
253 .
254 .
255 .SH AUTHOR
256 .rs
257 .sp
258 .nf
259 Philip Hazel
260 University Computing Service
261 Cambridge CB2 3QH, England.
262 .fi
263 .
264 .
265 .SH REVISION
266 .rs
267 .sp
268 .nf
269 Last updated: 25 September 2012
270 Copyright (c) 1997-2012 University of Cambridge.
271 .fi

  ViewVC Help
Powered by ViewVC 1.1.5