ViewVC logotype

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

Parent Directory Parent Directory | Revision Log Revision Log

Revision 737 - (hide annotations) (download)
Wed Oct 19 17:37:29 2011 UTC (3 years, 6 months ago) by ph10
File size: 7718 byte(s)
Add more about \C to documentation.

1 ph10 678 .TH PCREUNICODE 3
2     .SH NAME
3     PCRE - Perl-compatible regular expressions
5     .rs
6     .sp
7     In order process UTF-8 strings, you must build PCRE to include UTF-8 support in
8     the code, and, in addition, you must call
9     .\" HREF
10     \fBpcre_compile()\fP
11     .\"
12     with the PCRE_UTF8 option flag, or the pattern must start with the sequence
13     (*UTF8). When either of these is the case, both the pattern and any subject
14     strings that are matched against it are treated as UTF-8 strings instead of
15 ph10 691 strings of 1-byte characters. PCRE does not support any other formats (in
16 ph10 678 particular, it does not support UTF-16).
17     .P
18     If you compile PCRE with UTF-8 support, but do not use it at run time, the
19     library will be a bit bigger, but the additional run time overhead is limited
20     to testing the PCRE_UTF8 flag occasionally, so should not be very big.
21     .P
22     If PCRE is built with Unicode character property support (which implies UTF-8
23     support), the escape sequences \ep{..}, \eP{..}, and \eX are supported.
24     The available properties that can be tested are limited to the general
25     category properties such as Lu for an upper case letter or Nd for a decimal
26     number, the Unicode script names such as Arabic or Han, and the derived
27     properties Any and L&. A full list is given in the
28     .\" HREF
29     \fBpcrepattern\fP
30     .\"
31     documentation. Only the short names for properties are supported. For example,
32     \ep{L} matches a letter. Its Perl synonym, \ep{Letter}, is not supported.
33     Furthermore, in Perl, many properties may optionally be prefixed by "Is", for
34     compatibility with Perl 5.6. PCRE does not support this.
35     .
36     .
37     .\" HTML <a name="utf8strings"></a>
38     .SS "Validity of UTF-8 strings"
39     .rs
40     .sp
41     When you set the PCRE_UTF8 flag, the strings passed as patterns and subjects
42     are (by default) checked for validity on entry to the relevant functions. From
43     release 7.3 of PCRE, the check is according the rules of RFC 3629, which are
44     themselves derived from the Unicode specification. Earlier releases of PCRE
45     followed the rules of RFC 2279, which allows the full range of 31-bit values (0
46     to 0x7FFFFFFF). The current check allows only values in the range U+0 to
47     U+10FFFF, excluding U+D800 to U+DFFF.
48     .P
49     The excluded code points are the "Low Surrogate Area" of Unicode, of which the
50     Unicode Standard says this: "The Low Surrogate Area does not contain any
51     character assignments, consequently no character code charts or namelists are
52     provided for this area. Surrogates are reserved for use with UTF-16 and then
53     must be used in pairs." The code points that are encoded by UTF-16 pairs are
54     available as independent code points in the UTF-8 encoding. (In other words,
55     the whole surrogate thing is a fudge for UTF-16 which unfortunately messes up
56     UTF-8.)
57     .P
58     If an invalid UTF-8 string is passed to PCRE, an error return is given. At
59     compile time, the only additional information is the offset to the first byte
60     of the failing character. The runtime functions \fBpcre_exec()\fP and
61     \fBpcre_dfa_exec()\fP also pass back this information, as well as a more
62     detailed reason code if the caller has provided memory in which to do this.
63     .P
64     In some situations, you may already know that your strings are valid, and
65     therefore want to skip these checks in order to improve performance. If you set
66     the PCRE_NO_UTF8_CHECK flag at compile time or at run time, PCRE assumes that
67     the pattern or subject it is given (respectively) contains only valid UTF-8
68     codes. In this case, it does not diagnose an invalid UTF-8 string.
69     .P
70     If you pass an invalid UTF-8 string when PCRE_NO_UTF8_CHECK is set, what
71     happens depends on why the string is invalid. If the string conforms to the
72     "old" definition of UTF-8 (RFC 2279), it is processed as a string of characters
73 ph10 683 in the range 0 to 0x7FFFFFFF by \fBpcre_dfa_exec()\fP and the interpreted
74     version of \fBpcre_exec()\fP. In other words, apart from the initial validity
75     test, these functions (when in UTF-8 mode) handle strings according to the more
76     liberal rules of RFC 2279. However, the just-in-time (JIT) optimization for
77     \fBpcre_exec()\fP supports only RFC 3629. If you are using JIT optimization, or
78     if the string does not even conform to RFC 2279, the result is undefined. Your
79     program may crash.
80 ph10 678 .P
81     If you want to process strings of values in the full range 0 to 0x7FFFFFFF,
82     encoded in a UTF-8-like manner as per the old RFC, you can set
83     PCRE_NO_UTF8_CHECK to bypass the more restrictive test. However, in this
84 ph10 691 situation, you will have to apply your own validity check, and avoid the use of
85 ph10 683 JIT optimization.
86 ph10 678 .
87     .
88     .SS "General comments about UTF-8 mode"
89     .rs
90     .sp
91     1. An unbraced hexadecimal escape sequence (such as \exb3) matches a two-byte
92     UTF-8 character if the value is greater than 127.
93     .P
94     2. Octal numbers up to \e777 are recognized, and match two-byte UTF-8
95     characters for values greater than \e177.
96     .P
97     3. Repeat quantifiers apply to complete UTF-8 characters, not to individual
98     bytes, for example: \ex{100}{3}.
99     .P
100     4. The dot metacharacter matches one UTF-8 character instead of a single byte.
101     .P
102     5. The escape sequence \eC can be used to match a single byte in UTF-8 mode,
103 ph10 737 but its use can lead to some strange effects because it breaks up multibyte
104     characters (see the description of \eC in the
105     .\" HREF
106     \fBpcrepattern\fP
107     .\"
108     documentation). The use of \eC is not supported in the alternative matching
109     function \fBpcre_dfa_exec()\fP, nor is it supported in UTF-8 mode by the JIT
110     optimization of \fBpcre_exec()\fP. If JIT optimization is requested for a UTF-8
111     pattern that contains \eC, it will not succeed, and so the matching will be
112     carried out by the normal interpretive function.
113 ph10 678 .P
114     6. The character escapes \eb, \eB, \ed, \eD, \es, \eS, \ew, and \eW correctly
115     test characters of any code value, but, by default, the characters that PCRE
116     recognizes as digits, spaces, or word characters remain the same set as before,
117     all with values less than 256. This remains true even when PCRE is built to
118     include Unicode property support, because to do otherwise would slow down PCRE
119     in many common cases. Note in particular that this applies to \eb and \eB,
120     because they are defined in terms of \ew and \eW. If you really want to test
121     for a wider sense of, say, "digit", you can use explicit Unicode property tests
122     such as \ep{Nd}. Alternatively, if you set the PCRE_UCP option, the way that
123     the character escapes work is changed so that Unicode properties are used to
124     determine which characters match. There are more details in the section on
125     .\" HTML <a href="pcrepattern.html#genericchartypes">
126     .\" </a>
127     generic character types
128     .\"
129     in the
130     .\" HREF
131     \fBpcrepattern\fP
132     .\"
133     documentation.
134     .P
135     7. Similarly, characters that match the POSIX named character classes are all
136     low-valued characters, unless the PCRE_UCP option is set.
137     .P
138     8. However, the horizontal and vertical whitespace matching escapes (\eh, \eH,
139     \ev, and \eV) do match all the appropriate Unicode characters, whether or not
140     PCRE_UCP is set.
141     .P
142     9. Case-insensitive matching applies only to characters whose values are less
143     than 128, unless PCRE is built with Unicode property support. Even when Unicode
144     property support is available, PCRE still uses its own character tables when
145     checking the case of low-valued characters, so as not to degrade performance.
146     The Unicode property information is used only for characters with higher
147     values. Furthermore, PCRE supports case-insensitive matching only when there is
148     a one-to-one mapping between a letter's cases. There are a small number of
149     many-to-one mappings in Unicode; these are not supported by PCRE.
150     .
151     .
152     .SH AUTHOR
153     .rs
154     .sp
155     .nf
156     Philip Hazel
157     University Computing Service
158     Cambridge CB2 3QH, England.
159     .fi
160     .
161     .
162     .SH REVISION
163     .rs
164     .sp
165     .nf
166 ph10 737 Last updated: 19 October 2011
167 ph10 678 Copyright (c) 1997-2011 University of Cambridge.
168     .fi

ViewVC Help
Powered by ViewVC 1.1.12