trunk/doc/pcreunicode.3

.TH PCREUNICODE 3 "14 April 2012" "PCRE 8.30"
.SH NAME
PCRE - Perl-compatible regular expressions
.SH "UTF-8, UTF-16, AND UNICODE PROPERTY SUPPORT"
.rs
.sp
From Release 8.30, in addition to its previous UTF-8 support, PCRE also
supports UTF-16 by means of a separate 16-bit library. This can be built as
well as, or instead of, the 8-bit library.
.
.
.SH "UTF-8 SUPPORT"
.rs
.sp
In order process UTF-8 strings, you must build PCRE's 8-bit library with UTF
support, and, in addition, you must call
.\" HREF
\fBpcre_compile()\fP
.\"
with the PCRE_UTF8 option flag, or the pattern must start with the sequence
(*UTF8). When either of these is the case, both the pattern and any subject
strings that are matched against it are treated as UTF-8 strings instead of
strings of 1-byte characters.
.
.
.SH "UTF-16 SUPPORT"
.rs
.sp
In order process UTF-16 strings, you must build PCRE's 16-bit library with UTF
support, and, in addition, you must call
.\" HTML <a href="pcre_compile.html">
.\" </a>
\fBpcre16_compile()\fP
.\"
with the PCRE_UTF16 option flag, or the pattern must start with the sequence
(*UTF16). When either of these is the case, both the pattern and any subject
strings that are matched against it are treated as UTF-16 strings instead of
strings of 16-bit characters.
.
.
.SH "UTF SUPPORT OVERHEAD"
.rs
.sp
If you compile PCRE with UTF support, but do not use it at run time, the
library will be a bit bigger, but the additional run time overhead is limited
to testing the PCRE_UTF8/16 flag occasionally, so should not be very big.
.
.
.SH "UNICODE PROPERTY SUPPORT"
.rs
.sp
If PCRE is built with Unicode character property support (which implies UTF
support), the escape sequences \ep{..}, \eP{..}, and \eX can be used.
The available properties that can be tested are limited to the general
category properties such as Lu for an upper case letter or Nd for a decimal
number, the Unicode script names such as Arabic or Han, and the derived
properties Any and L&. A full list is given in the
.\" HREF
\fBpcrepattern\fP
.\"
documentation. Only the short names for properties are supported. For example,
\ep{L} matches a letter. Its Perl synonym, \ep{Letter}, is not supported.
Furthermore, in Perl, many properties may optionally be prefixed by "Is", for
compatibility with Perl 5.6. PCRE does not support this.
.
.
.\" HTML <a name="utf8strings"></a>
.SS "Validity of UTF-8 strings"
.rs
.sp
When you set the PCRE_UTF8 flag, the byte strings passed as patterns and
subjects are (by default) checked for validity on entry to the relevant
functions. The entire string is checked before any other processing takes
place. From release 7.3 of PCRE, the check is according the rules of RFC 3629,
which are themselves derived from the Unicode specification. Earlier releases
of PCRE followed the rules of RFC 2279, which allows the full range of 31-bit
values (0 to 0x7FFFFFFF). The current check allows only values in the range U+0
to U+10FFFF, excluding U+D800 to U+DFFF.
.P
The excluded code points are the "Surrogate Area" of Unicode. They are reserved
for use by UTF-16, where they are used in pairs to encode codepoints with
values greater than 0xFFFF. The code points that are encoded by UTF-16 pairs
are available independently in the UTF-8 encoding. (In other words, the whole
surrogate thing is a fudge for UTF-16 which unfortunately messes up UTF-8.)
.P
If an invalid UTF-8 string is passed to PCRE, an error return is given. At
compile time, the only additional information is the offset to the first byte
of the failing character. The run-time functions \fBpcre_exec()\fP and
\fBpcre_dfa_exec()\fP also pass back this information, as well as a more
detailed reason code if the caller has provided memory in which to do this.
.P
In some situations, you may already know that your strings are valid, and
therefore want to skip these checks in order to improve performance, for
example in the case of a long subject string that is being scanned repeatedly
with different patterns. If you set the PCRE_NO_UTF8_CHECK flag at compile time
or at run time, PCRE assumes that the pattern or subject it is given
(respectively) contains only valid UTF-8 codes. In this case, it does not
diagnose an invalid UTF-8 string.
.P
If you pass an invalid UTF-8 string when PCRE_NO_UTF8_CHECK is set, what
happens depends on why the string is invalid. If the string conforms to the
"old" definition of UTF-8 (RFC 2279), it is processed as a string of characters
in the range 0 to 0x7FFFFFFF by \fBpcre_dfa_exec()\fP and the interpreted
version of \fBpcre_exec()\fP. In other words, apart from the initial validity
test, these functions (when in UTF-8 mode) handle strings according to the more
liberal rules of RFC 2279. However, the just-in-time (JIT) optimization for
\fBpcre_exec()\fP supports only RFC 3629. If you are using JIT optimization, or
if the string does not even conform to RFC 2279, the result is undefined. Your
program may crash.
.P
If you want to process strings of values in the full range 0 to 0x7FFFFFFF,
encoded in a UTF-8-like manner as per the old RFC, you can set
PCRE_NO_UTF8_CHECK to bypass the more restrictive test. However, in this
situation, you will have to apply your own validity check, and avoid the use of
JIT optimization.
.
.
.\" HTML <a name="utf16strings"></a>
.SS "Validity of UTF-16 strings"
.rs
.sp
When you set the PCRE_UTF16 flag, the strings of 16-bit data units that are
passed as patterns and subjects are (by default) checked for validity on entry
to the relevant functions. Values other than those in the surrogate range
U+D800 to U+DFFF are independent code points. Values in the surrogate range
must be used in pairs in the correct manner.
.P
If an invalid UTF-16 string is passed to PCRE, an error return is given. At
compile time, the only additional information is the offset to the first data
unit of the failing character. The run-time functions \fBpcre16_exec()\fP and
\fBpcre16_dfa_exec()\fP also pass back this information, as well as a more
detailed reason code if the caller has provided memory in which to do this.
.P
In some situations, you may already know that your strings are valid, and
therefore want to skip these checks in order to improve performance. If you set
the PCRE_NO_UTF16_CHECK flag at compile time or at run time, PCRE assumes that
the pattern or subject it is given (respectively) contains only valid UTF-16
sequences. In this case, it does not diagnose an invalid UTF-16 string.
.
.
.SS "General comments about UTF modes"
.rs
.sp
1. Codepoints less than 256 can be specified by either braced or unbraced
hexadecimal escape sequences (for example, \ex{b3} or \exb3). Larger values
have to use braced sequences.
.P
2. Octal numbers up to \e777 are recognized, and in UTF-8 mode, they match
two-byte characters for values greater than \e177.
.P
3. Repeat quantifiers apply to complete UTF characters, not to individual
data units, for example: \ex{100}{3}.
.P
4. The dot metacharacter matches one UTF character instead of a single data
unit.
.P
5. The escape sequence \eC can be used to match a single byte in UTF-8 mode, or
a single 16-bit data unit in UTF-16 mode, but its use can lead to some strange
effects because it breaks up multi-unit characters (see the description of \eC
in the
.\" HREF
\fBpcrepattern\fP
.\"
documentation). The use of \eC is not supported in the alternative matching
function \fBpcre[16]_dfa_exec()\fP, nor is it supported in UTF mode by the JIT
optimization of \fBpcre[16]_exec()\fP. If JIT optimization is requested for a
UTF pattern that contains \eC, it will not succeed, and so the matching will
be carried out by the normal interpretive function.
.P
6. The character escapes \eb, \eB, \ed, \eD, \es, \eS, \ew, and \eW correctly
test characters of any code value, but, by default, the characters that PCRE
recognizes as digits, spaces, or word characters remain the same set as in
non-UTF mode, all with values less than 256. This remains true even when PCRE
is built to include Unicode property support, because to do otherwise would
slow down PCRE in many common cases. Note in particular that this applies to
\eb and \eB, because they are defined in terms of \ew and \eW. If you really
want to test for a wider sense of, say, "digit", you can use explicit Unicode
property tests such as \ep{Nd}. Alternatively, if you set the PCRE_UCP option,
the way that the character escapes work is changed so that Unicode properties
are used to determine which characters match. There are more details in the
section on
.\" HTML <a href="pcrepattern.html#genericchartypes">
.\" </a>
generic character types
.\"
in the
.\" HREF
\fBpcrepattern\fP
.\"
documentation.
.P
7. Similarly, characters that match the POSIX named character classes are all
low-valued characters, unless the PCRE_UCP option is set.
.P
8. However, the horizontal and vertical white space matching escapes (\eh, \eH,
\ev, and \eV) do match all the appropriate Unicode characters, whether or not
PCRE_UCP is set.
.P
9. Case-insensitive matching applies only to characters whose values are less
than 128, unless PCRE is built with Unicode property support. Even when Unicode
property support is available, PCRE still uses its own character tables when
checking the case of low-valued characters, so as not to degrade performance.
The Unicode property information is used only for characters with higher
values. Furthermore, PCRE supports case-insensitive matching only when there is
a one-to-one mapping between a letter's cases. There are a small number of
many-to-one mappings in Unicode; these are not supported by PCRE.
.
.
.SH AUTHOR
.rs
.sp
.nf
Philip Hazel
University Computing Service
Cambridge CB2 3QH, England.
.fi
.
.
.SH REVISION
.rs
.sp
.nf
Last updated: 14 April 2012
Copyright (c) 1997-2012 University of Cambridge.
.fi
1	ph10	959	.TH PCREUNICODE 3 "14 April 2012" "PCRE 8.30"
2	ph10	678	.SH NAME
3			PCRE - Perl-compatible regular expressions
4	ph10	868	.SH "UTF-8, UTF-16, AND UNICODE PROPERTY SUPPORT"
5	ph10	678	.rs
6			.sp
7	ph10	868	From Release 8.30, in addition to its previous UTF-8 support, PCRE also
8	ph10	903	supports UTF-16 by means of a separate 16-bit library. This can be built as
9	ph10	868	well as, or instead of, the 8-bit library.
10			.
11			.
12			.SH "UTF-8 SUPPORT"
13			.rs
14			.sp
15			In order process UTF-8 strings, you must build PCRE's 8-bit library with UTF
16			support, and, in addition, you must call
17	ph10	678	.\" HREF
18			\fBpcre_compile()\fP
19			.\"
20			with the PCRE_UTF8 option flag, or the pattern must start with the sequence
21			(*UTF8). When either of these is the case, both the pattern and any subject
22			strings that are matched against it are treated as UTF-8 strings instead of
23	ph10	868	strings of 1-byte characters.
24			.
25			.
26			.SH "UTF-16 SUPPORT"
27			.rs
28			.sp
29			In order process UTF-16 strings, you must build PCRE's 16-bit library with UTF
30			support, and, in addition, you must call
31	ph10	878	.\" HTML <a href="pcre_compile.html">
32			.\" </a>
33	ph10	868	\fBpcre16_compile()\fP
34			.\"
35			with the PCRE_UTF16 option flag, or the pattern must start with the sequence
36			(*UTF16). When either of these is the case, both the pattern and any subject
37			strings that are matched against it are treated as UTF-16 strings instead of
38			strings of 16-bit characters.
39			.
40			.
41			.SH "UTF SUPPORT OVERHEAD"
42			.rs
43			.sp
44			If you compile PCRE with UTF support, but do not use it at run time, the
45	ph10	678	library will be a bit bigger, but the additional run time overhead is limited
46	ph10	868	to testing the PCRE_UTF8/16 flag occasionally, so should not be very big.
47			.
48			.
49			.SH "UNICODE PROPERTY SUPPORT"
50			.rs
51			.sp
52			If PCRE is built with Unicode character property support (which implies UTF
53			support), the escape sequences \ep{..}, \eP{..}, and \eX can be used.
54	ph10	678	The available properties that can be tested are limited to the general
55			category properties such as Lu for an upper case letter or Nd for a decimal
56			number, the Unicode script names such as Arabic or Han, and the derived
57			properties Any and L&. A full list is given in the
58			.\" HREF
59			\fBpcrepattern\fP
60			.\"
61			documentation. Only the short names for properties are supported. For example,
62			\ep{L} matches a letter. Its Perl synonym, \ep{Letter}, is not supported.
63			Furthermore, in Perl, many properties may optionally be prefixed by "Is", for
64			compatibility with Perl 5.6. PCRE does not support this.
65			.
66			.
67			.\" HTML <a name="utf8strings"></a>
68			.SS "Validity of UTF-8 strings"
69			.rs
70			.sp
71	ph10	868	When you set the PCRE_UTF8 flag, the byte strings passed as patterns and
72			subjects are (by default) checked for validity on entry to the relevant
73	ph10	959	functions. The entire string is checked before any other processing takes
74			place. From release 7.3 of PCRE, the check is according the rules of RFC 3629,
75			which are themselves derived from the Unicode specification. Earlier releases
76			of PCRE followed the rules of RFC 2279, which allows the full range of 31-bit
77			values (0 to 0x7FFFFFFF). The current check allows only values in the range U+0
78			to U+10FFFF, excluding U+D800 to U+DFFF.
79	ph10	678	.P
80	ph10	868	The excluded code points are the "Surrogate Area" of Unicode. They are reserved
81	ph10	903	for use by UTF-16, where they are used in pairs to encode codepoints with
82	ph10	868	values greater than 0xFFFF. The code points that are encoded by UTF-16 pairs
83			are available independently in the UTF-8 encoding. (In other words, the whole
84			surrogate thing is a fudge for UTF-16 which unfortunately messes up UTF-8.)
85	ph10	678	.P
86			If an invalid UTF-8 string is passed to PCRE, an error return is given. At
87			compile time, the only additional information is the offset to the first byte
88	ph10	968	of the failing character. The run-time functions \fBpcre_exec()\fP and
89	ph10	678	\fBpcre_dfa_exec()\fP also pass back this information, as well as a more
90			detailed reason code if the caller has provided memory in which to do this.
91			.P
92			In some situations, you may already know that your strings are valid, and
93	ph10	959	therefore want to skip these checks in order to improve performance, for
94			example in the case of a long subject string that is being scanned repeatedly
95			with different patterns. If you set the PCRE_NO_UTF8_CHECK flag at compile time
96			or at run time, PCRE assumes that the pattern or subject it is given
97			(respectively) contains only valid UTF-8 codes. In this case, it does not
98			diagnose an invalid UTF-8 string.
99	ph10	678	.P
100			If you pass an invalid UTF-8 string when PCRE_NO_UTF8_CHECK is set, what
101			happens depends on why the string is invalid. If the string conforms to the
102			"old" definition of UTF-8 (RFC 2279), it is processed as a string of characters
103	ph10	683	in the range 0 to 0x7FFFFFFF by \fBpcre_dfa_exec()\fP and the interpreted
104			version of \fBpcre_exec()\fP. In other words, apart from the initial validity
105			test, these functions (when in UTF-8 mode) handle strings according to the more
106			liberal rules of RFC 2279. However, the just-in-time (JIT) optimization for
107			\fBpcre_exec()\fP supports only RFC 3629. If you are using JIT optimization, or
108			if the string does not even conform to RFC 2279, the result is undefined. Your
109			program may crash.
110	ph10	678	.P
111			If you want to process strings of values in the full range 0 to 0x7FFFFFFF,
112			encoded in a UTF-8-like manner as per the old RFC, you can set
113			PCRE_NO_UTF8_CHECK to bypass the more restrictive test. However, in this
114	ph10	691	situation, you will have to apply your own validity check, and avoid the use of
115	ph10	683	JIT optimization.
116	ph10	678	.
117			.
118	ph10	868	.\" HTML <a name="utf16strings"></a>
119			.SS "Validity of UTF-16 strings"
120	ph10	678	.rs
121			.sp
122	ph10	868	When you set the PCRE_UTF16 flag, the strings of 16-bit data units that are
123			passed as patterns and subjects are (by default) checked for validity on entry
124			to the relevant functions. Values other than those in the surrogate range
125			U+D800 to U+DFFF are independent code points. Values in the surrogate range
126			must be used in pairs in the correct manner.
127	ph10	678	.P
128	ph10	868	If an invalid UTF-16 string is passed to PCRE, an error return is given. At
129			compile time, the only additional information is the offset to the first data
130	ph10	968	unit of the failing character. The run-time functions \fBpcre16_exec()\fP and
131	ph10	868	\fBpcre16_dfa_exec()\fP also pass back this information, as well as a more
132			detailed reason code if the caller has provided memory in which to do this.
133	ph10	678	.P
134	ph10	868	In some situations, you may already know that your strings are valid, and
135			therefore want to skip these checks in order to improve performance. If you set
136			the PCRE_NO_UTF16_CHECK flag at compile time or at run time, PCRE assumes that
137			the pattern or subject it is given (respectively) contains only valid UTF-16
138			sequences. In this case, it does not diagnose an invalid UTF-16 string.
139			.
140			.
141			.SS "General comments about UTF modes"
142			.rs
143			.sp
144			1. Codepoints less than 256 can be specified by either braced or unbraced
145			hexadecimal escape sequences (for example, \ex{b3} or \exb3). Larger values
146			have to use braced sequences.
147	ph10	678	.P
148	ph10	868	2. Octal numbers up to \e777 are recognized, and in UTF-8 mode, they match
149			two-byte characters for values greater than \e177.
150	ph10	678	.P
151	ph10	868	3. Repeat quantifiers apply to complete UTF characters, not to individual
152			data units, for example: \ex{100}{3}.
153			.P
154	ph10	903	4. The dot metacharacter matches one UTF character instead of a single data
155	ph10	868	unit.
156			.P
157			5. The escape sequence \eC can be used to match a single byte in UTF-8 mode, or
158			a single 16-bit data unit in UTF-16 mode, but its use can lead to some strange
159			effects because it breaks up multi-unit characters (see the description of \eC
160			in the
161	ph10	737	.\" HREF
162			\fBpcrepattern\fP
163			.\"
164			documentation). The use of \eC is not supported in the alternative matching
165	ph10	868	function \fBpcre[16]_dfa_exec()\fP, nor is it supported in UTF mode by the JIT
166			optimization of \fBpcre[16]_exec()\fP. If JIT optimization is requested for a
167			UTF pattern that contains \eC, it will not succeed, and so the matching will
168			be carried out by the normal interpretive function.
169	ph10	678	.P
170			6. The character escapes \eb, \eB, \ed, \eD, \es, \eS, \ew, and \eW correctly
171			test characters of any code value, but, by default, the characters that PCRE
172	ph10	903	recognizes as digits, spaces, or word characters remain the same set as in
173	ph10	868	non-UTF mode, all with values less than 256. This remains true even when PCRE
174			is built to include Unicode property support, because to do otherwise would
175			slow down PCRE in many common cases. Note in particular that this applies to
176			\eb and \eB, because they are defined in terms of \ew and \eW. If you really
177			want to test for a wider sense of, say, "digit", you can use explicit Unicode
178			property tests such as \ep{Nd}. Alternatively, if you set the PCRE_UCP option,
179			the way that the character escapes work is changed so that Unicode properties
180			are used to determine which characters match. There are more details in the
181			section on
182	ph10	678	.\" HTML <a href="pcrepattern.html#genericchartypes">
183			.\" </a>
184			generic character types
185			.\"
186			in the
187			.\" HREF
188			\fBpcrepattern\fP
189			.\"
190			documentation.
191			.P
192			7. Similarly, characters that match the POSIX named character classes are all
193			low-valued characters, unless the PCRE_UCP option is set.
194			.P
195	ph10	968	8. However, the horizontal and vertical white space matching escapes (\eh, \eH,
196	ph10	678	\ev, and \eV) do match all the appropriate Unicode characters, whether or not
197			PCRE_UCP is set.
198			.P
199			9. Case-insensitive matching applies only to characters whose values are less
200			than 128, unless PCRE is built with Unicode property support. Even when Unicode
201			property support is available, PCRE still uses its own character tables when
202			checking the case of low-valued characters, so as not to degrade performance.
203			The Unicode property information is used only for characters with higher
204			values. Furthermore, PCRE supports case-insensitive matching only when there is
205			a one-to-one mapping between a letter's cases. There are a small number of
206			many-to-one mappings in Unicode; these are not supported by PCRE.
207			.
208			.
209			.SH AUTHOR
210			.rs
211			.sp
212			.nf
213			Philip Hazel
214			University Computing Service
215			Cambridge CB2 3QH, England.
216			.fi
217			.
218			.
219			.SH REVISION
220			.rs
221			.sp
222			.nf
223	ph10	959	Last updated: 14 April 2012
224	ph10	868	Copyright (c) 1997-2012 University of Cambridge.
225	ph10	678	.fi