trunk/doc/pcrebuild.3

.TH PCRE 3
.SH NAME
PCRE - Perl-compatible regular expressions
.SH "PCRE BUILD-TIME OPTIONS"
.rs
.sp
This document describes the optional features of PCRE that can be selected when
the library is compiled. They are all selected, or deselected, by providing
options to the \fBconfigure\fP script that is run before the \fBmake\fP
command. The complete list of options for \fBconfigure\fP (which includes the
standard ones such as the selection of the installation directory) can be
obtained by running
.sp
  ./configure --help
.sp
The following sections describe certain options whose names begin with --enable
or --disable. These settings specify changes to the defaults for the
\fBconfigure\fP command. Because of the way that \fBconfigure\fP works,
--enable and --disable always come in pairs, so the complementary option always
exists as well, but as it specifies the default, it is not described.
.
.SH "UTF-8 SUPPORT"
.rs
.sp
To build PCRE with support for UTF-8 character strings, add
.sp
  --enable-utf8
.sp
to the \fBconfigure\fP command. Of itself, this does not make PCRE treat
strings as UTF-8. As well as compiling PCRE with this option, you also have
have to set the PCRE_UTF8 option when you call the \fBpcre_compile()\fP
function.
.
.SH "UNICODE CHARACTER PROPERTY SUPPORT"
.rs
.sp
UTF-8 support allows PCRE to process character values greater than 255 in the
strings that it handles. On its own, however, it does not provide any
facilities for accessing the properties of such characters. If you want to be
able to use the pattern escapes \eP, \ep, and \eX, which refer to Unicode
character properties, you must add
.sp
  --enable-unicode-properties
.sp
to the \fBconfigure\fP command. This implies UTF-8 support, even if you have
not explicitly requested it.
.P
Including Unicode property support adds around 90K of tables to the PCRE
library, approximately doubling its size. Only the general category properties
such as \fILu\fP and \fINd\fP are supported. Details are given in the
.\" HREF
\fBpcrepattern\fP
.\"
documentation.
.
.SH "CODE VALUE OF NEWLINE"
.rs
.sp
By default, PCRE treats character 10 (linefeed) as the newline character. This
is the normal newline character on Unix-like systems. You can compile PCRE to
use character 13 (carriage return) instead by adding
.sp
  --enable-newline-is-cr
.sp
to the \fBconfigure\fP command. For completeness there is also a
--enable-newline-is-lf option, which explicitly specifies linefeed as the
newline character.
.
.SH "BUILDING SHARED AND STATIC LIBRARIES"
.rs
.sp
The PCRE building process uses \fBlibtool\fP to build both shared and static
Unix libraries by default. You can suppress one of these by adding one of
.sp
  --disable-shared
  --disable-static
.sp
to the \fBconfigure\fP command, as required.
.
.SH "POSIX MALLOC USAGE"
.rs
.sp
When PCRE is called through the POSIX interface (see the
.\" HREF
\fBpcreposix\fP
.\"
documentation), additional working storage is required for holding the pointers
to capturing substrings, because PCRE requires three integers per substring,
whereas the POSIX interface provides only two. If the number of expected
substrings is small, the wrapper function uses space on the stack, because this
is faster than using \fBmalloc()\fP for each call. The default threshold above
which the stack is no longer used is 10; it can be changed by adding a setting
such as
.sp
  --with-posix-malloc-threshold=20
.sp
to the \fBconfigure\fP command.
.
.SH "LIMITING PCRE RESOURCE USAGE"
.rs
.sp
Internally, PCRE has a function called \fBmatch()\fP, which it calls repeatedly
(possibly recursively) when matching a pattern with the \fBpcre_exec()\fP
function. By controlling the maximum number of times this function may be
called during a single matching operation, a limit can be placed on the
resources used by a single call to \fBpcre_exec()\fP. The limit can be changed
at run time, as described in the
.\" HREF
\fBpcreapi\fP
.\"
documentation. The default is 10 million, but this can be changed by adding a
setting such as
.sp
  --with-match-limit=500000
.sp
to the \fBconfigure\fP command. This setting has no effect on the
\fBpcre_dfa_exec()\fP matching function.
.
.SH "HANDLING VERY LARGE PATTERNS"
.rs
.sp
Within a compiled pattern, offset values are used to point from one part to
another (for example, from an opening parenthesis to an alternation
metacharacter). By default, two-byte values are used for these offsets, leading
to a maximum size for a compiled pattern of around 64K. This is sufficient to
handle all but the most gigantic patterns. Nevertheless, some people do want to
process enormous patterns, so it is possible to compile PCRE to use three-byte
or four-byte offsets by adding a setting such as
.sp
  --with-link-size=3
.sp
to the \fBconfigure\fP command. The value given must be 2, 3, or 4. Using
longer offsets slows down the operation of PCRE because it has to load
additional bytes when handling them.
.P
If you build PCRE with an increased link size, test 2 (and test 5 if you are
using UTF-8) will fail. Part of the output of these tests is a representation
of the compiled pattern, and this changes with the link size.
.
.SH "AVOIDING EXCESSIVE STACK USAGE"
.rs
.sp
When matching with the \fBpcre_exec()\fP function, PCRE implements backtracking
by making recursive calls to an internal function called \fBmatch()\fP. In
environments where the size of the stack is limited, this can severely limit
PCRE's operation. (The Unix environment does not usually suffer from this
problem.) An alternative approach that uses memory from the heap to remember
data, instead of using recursive function calls, has been implemented to work
round this problem. If you want to build a version of PCRE that works this way,
add
.sp
  --disable-stack-for-recursion
.sp
to the \fBconfigure\fP command. With this configuration, PCRE will use the
\fBpcre_stack_malloc\fP and \fBpcre_stack_free\fP variables to call memory
management functions. Separate functions are provided because the usage is very
predictable: the block sizes requested are always the same, and the blocks are
always freed in reverse order. A calling program might be able to implement
optimized functions that perform better than the standard \fBmalloc()\fP and
\fBfree()\fP functions. PCRE runs noticeably more slowly when built in this
way. This option affects only the \fBpcre_exec()\fP function; it is not
relevant for the the \fBpcre_dfa_exec()\fP function.
.
.SH "USING EBCDIC CODE"
.rs
.sp
PCRE assumes by default that it will run in an environment where the character
code is ASCII (or Unicode, which is a superset of ASCII). PCRE can, however, be
compiled to run in an EBCDIC environment by adding
.sp
  --enable-ebcdic
.sp
to the \fBconfigure\fP command.
.P
.in 0
Last updated: 28 February 2005
.br
Copyright (c) 1997-2005 University of Cambridge.
1	nigel	63	.TH PCRE 3
2			.SH NAME
3			PCRE - Perl-compatible regular expressions
4	nigel	75	.SH "PCRE BUILD-TIME OPTIONS"
5	nigel	63	.rs
6			.sp
7			This document describes the optional features of PCRE that can be selected when
8			the library is compiled. They are all selected, or deselected, by providing
9	nigel	75	options to the \fBconfigure\fP script that is run before the \fBmake\fP
10			command. The complete list of options for \fBconfigure\fP (which includes the
11	nigel	63	standard ones such as the selection of the installation directory) can be
12			obtained by running
13	nigel	75	.sp
14	nigel	63	./configure --help
15	nigel	75	.sp
16	nigel	63	The following sections describe certain options whose names begin with --enable
17			or --disable. These settings specify changes to the defaults for the
18	nigel	75	\fBconfigure\fP command. Because of the way that \fBconfigure\fP works,
19	nigel	63	--enable and --disable always come in pairs, so the complementary option always
20			exists as well, but as it specifies the default, it is not described.
21	nigel	75	.
22			.SH "UTF-8 SUPPORT"
23	nigel	63	.rs
24			.sp
25			To build PCRE with support for UTF-8 character strings, add
26	nigel	75	.sp
27	nigel	63	--enable-utf8
28	nigel	75	.sp
29			to the \fBconfigure\fP command. Of itself, this does not make PCRE treat
30	nigel	63	strings as UTF-8. As well as compiling PCRE with this option, you also have
31	nigel	75	have to set the PCRE_UTF8 option when you call the \fBpcre_compile()\fP
32	nigel	63	function.
33	nigel	75	.
34			.SH "UNICODE CHARACTER PROPERTY SUPPORT"
35	nigel	63	.rs
36			.sp
37	nigel	75	UTF-8 support allows PCRE to process character values greater than 255 in the
38			strings that it handles. On its own, however, it does not provide any
39			facilities for accessing the properties of such characters. If you want to be
40			able to use the pattern escapes \eP, \ep, and \eX, which refer to Unicode
41			character properties, you must add
42			.sp
43			--enable-unicode-properties
44			.sp
45			to the \fBconfigure\fP command. This implies UTF-8 support, even if you have
46			not explicitly requested it.
47			.P
48			Including Unicode property support adds around 90K of tables to the PCRE
49			library, approximately doubling its size. Only the general category properties
50			such as \fILu\fP and \fINd\fP are supported. Details are given in the
51			.\" HREF
52			\fBpcrepattern\fP
53			.\"
54			documentation.
55			.
56			.SH "CODE VALUE OF NEWLINE"
57			.rs
58			.sp
59	nigel	63	By default, PCRE treats character 10 (linefeed) as the newline character. This
60			is the normal newline character on Unix-like systems. You can compile PCRE to
61			use character 13 (carriage return) instead by adding
62	nigel	75	.sp
63	nigel	63	--enable-newline-is-cr
64	nigel	75	.sp
65			to the \fBconfigure\fP command. For completeness there is also a
66	nigel	63	--enable-newline-is-lf option, which explicitly specifies linefeed as the
67			newline character.
68	nigel	75	.
69			.SH "BUILDING SHARED AND STATIC LIBRARIES"
70	nigel	63	.rs
71			.sp
72	nigel	75	The PCRE building process uses \fBlibtool\fP to build both shared and static
73	nigel	63	Unix libraries by default. You can suppress one of these by adding one of
74	nigel	75	.sp
75	nigel	63	--disable-shared
76			--disable-static
77	nigel	75	.sp
78			to the \fBconfigure\fP command, as required.
79			.
80			.SH "POSIX MALLOC USAGE"
81	nigel	63	.rs
82			.sp
83	nigel	75	When PCRE is called through the POSIX interface (see the
84			.\" HREF
85			\fBpcreposix\fP
86			.\"
87	nigel	63	documentation), additional working storage is required for holding the pointers
88	nigel	75	to capturing substrings, because PCRE requires three integers per substring,
89	nigel	63	whereas the POSIX interface provides only two. If the number of expected
90			substrings is small, the wrapper function uses space on the stack, because this
91	nigel	75	is faster than using \fBmalloc()\fP for each call. The default threshold above
92	nigel	63	which the stack is no longer used is 10; it can be changed by adding a setting
93			such as
94	nigel	75	.sp
95	nigel	63	--with-posix-malloc-threshold=20
96	nigel	75	.sp
97			to the \fBconfigure\fP command.
98			.
99			.SH "LIMITING PCRE RESOURCE USAGE"
100	nigel	63	.rs
101			.sp
102	nigel	75	Internally, PCRE has a function called \fBmatch()\fP, which it calls repeatedly
103	nigel	77	(possibly recursively) when matching a pattern with the \fBpcre_exec()\fP
104			function. By controlling the maximum number of times this function may be
105			called during a single matching operation, a limit can be placed on the
106			resources used by a single call to \fBpcre_exec()\fP. The limit can be changed
107			at run time, as described in the
108	nigel	75	.\" HREF
109			\fBpcreapi\fP
110			.\"
111			documentation. The default is 10 million, but this can be changed by adding a
112			setting such as
113			.sp
114	nigel	63	--with-match-limit=500000
115	nigel	75	.sp
116	nigel	77	to the \fBconfigure\fP command. This setting has no effect on the
117			\fBpcre_dfa_exec()\fP matching function.
118	nigel	75	.
119			.SH "HANDLING VERY LARGE PATTERNS"
120	nigel	63	.rs
121			.sp
122			Within a compiled pattern, offset values are used to point from one part to
123			another (for example, from an opening parenthesis to an alternation
124	nigel	75	metacharacter). By default, two-byte values are used for these offsets, leading
125	nigel	63	to a maximum size for a compiled pattern of around 64K. This is sufficient to
126			handle all but the most gigantic patterns. Nevertheless, some people do want to
127			process enormous patterns, so it is possible to compile PCRE to use three-byte
128			or four-byte offsets by adding a setting such as
129	nigel	75	.sp
130	nigel	63	--with-link-size=3
131	nigel	75	.sp
132			to the \fBconfigure\fP command. The value given must be 2, 3, or 4. Using
133	nigel	63	longer offsets slows down the operation of PCRE because it has to load
134			additional bytes when handling them.
135	nigel	75	.P
136	nigel	63	If you build PCRE with an increased link size, test 2 (and test 5 if you are
137			using UTF-8) will fail. Part of the output of these tests is a representation
138			of the compiled pattern, and this changes with the link size.
139	nigel	75	.
140			.SH "AVOIDING EXCESSIVE STACK USAGE"
141	nigel	73	.rs
142			.sp
143	nigel	77	When matching with the \fBpcre_exec()\fP function, PCRE implements backtracking
144			by making recursive calls to an internal function called \fBmatch()\fP. In
145			environments where the size of the stack is limited, this can severely limit
146			PCRE's operation. (The Unix environment does not usually suffer from this
147			problem.) An alternative approach that uses memory from the heap to remember
148			data, instead of using recursive function calls, has been implemented to work
149			round this problem. If you want to build a version of PCRE that works this way,
150			add
151	nigel	75	.sp
152	nigel	73	--disable-stack-for-recursion
153	nigel	75	.sp
154			to the \fBconfigure\fP command. With this configuration, PCRE will use the
155			\fBpcre_stack_malloc\fP and \fBpcre_stack_free\fP variables to call memory
156	nigel	73	management functions. Separate functions are provided because the usage is very
157			predictable: the block sizes requested are always the same, and the blocks are
158			always freed in reverse order. A calling program might be able to implement
159	nigel	75	optimized functions that perform better than the standard \fBmalloc()\fP and
160			\fBfree()\fP functions. PCRE runs noticeably more slowly when built in this
161	nigel	77	way. This option affects only the \fBpcre_exec()\fP function; it is not
162			relevant for the the \fBpcre_dfa_exec()\fP function.
163	nigel	75	.
164			.SH "USING EBCDIC CODE"
165	nigel	73	.rs
166			.sp
167			PCRE assumes by default that it will run in an environment where the character
168	nigel	75	code is ASCII (or Unicode, which is a superset of ASCII). PCRE can, however, be
169	nigel	73	compiled to run in an EBCDIC environment by adding
170	nigel	75	.sp
171	nigel	73	--enable-ebcdic
172	nigel	75	.sp
173			to the \fBconfigure\fP command.
174			.P
175	nigel	63	.in 0
176	nigel	77	Last updated: 28 February 2005
177	nigel	63	.br
178	nigel	77	Copyright (c) 1997-2005 University of Cambridge.