code/trunk/pcre_ucp_searchfuncs.c

/*************************************************
*      Perl-Compatible Regular Expressions       *
*************************************************/

/* PCRE is a library of functions to support regular expressions whose syntax
and semantics are as close as possible to those of the Perl 5 language.

                       Written by Philip Hazel
           Copyright (c) 1997-2008 University of Cambridge

-----------------------------------------------------------------------------
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are met:

    * Redistributions of source code must retain the above copyright notice,
      this list of conditions and the following disclaimer.

    * Redistributions in binary form must reproduce the above copyright
      notice, this list of conditions and the following disclaimer in the
      documentation and/or other materials provided with the distribution.

    * Neither the name of the University of Cambridge nor the names of its
      contributors may be used to endorse or promote products derived from
      this software without specific prior written permission.

THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
POSSIBILITY OF SUCH DAMAGE.
-----------------------------------------------------------------------------
*/


/* This module contains code for searching the table of Unicode character
properties. */

#ifdef HAVE_CONFIG_H
#include "config.h"
#endif

#include "pcre_internal.h"

#include "ucp.h"               /* Category definitions */
#include "ucpinternal.h"       /* Internal table details */
#include "ucptable.h"          /* The table itself */


/* Table to translate from particular type value to the general value. */

static const int ucp_gentype[] = {
  ucp_C, ucp_C, ucp_C, ucp_C, ucp_C,  /* Cc, Cf, Cn, Co, Cs */
  ucp_L, ucp_L, ucp_L, ucp_L, ucp_L,  /* Ll, Lu, Lm, Lo, Lt */
  ucp_M, ucp_M, ucp_M,                /* Mc, Me, Mn */
  ucp_N, ucp_N, ucp_N,                /* Nd, Nl, No */
  ucp_P, ucp_P, ucp_P, ucp_P, ucp_P,  /* Pc, Pd, Pe, Pf, Pi */
  ucp_P, ucp_P,                       /* Ps, Po */
  ucp_S, ucp_S, ucp_S, ucp_S,         /* Sc, Sk, Sm, So */
  ucp_Z, ucp_Z, ucp_Z                 /* Zl, Zp, Zs */
};


/*************************************************
*         Search table and return type           *
*************************************************/

/* Three values are returned: the category is ucp_C, ucp_L, etc. The detailed
character type is ucp_Lu, ucp_Nd, etc. The script is ucp_Latin, etc.

Arguments:
  c           the character value
  type_ptr    the detailed character type is returned here
  script_ptr  the script is returned here

Returns:      the character type category
*/

int
_pcre_ucp_findprop(const unsigned int c, int *type_ptr, int *script_ptr)
{
int bot = 0;
int top = sizeof(ucp_table)/sizeof(cnode);
int mid;

/* The table is searched using a binary chop. You might think that using
intermediate variables to hold some of the common expressions would speed
things up, but tests with gcc 3.4.4 on Linux showed that, on the contrary, it
makes things a lot slower. */

for (;;)
  {
  if (top <= bot)
    {
    *type_ptr = ucp_Cn;
    *script_ptr = ucp_Common;
    return ucp_C;
    }
  mid = (bot + top) >> 1;
  if (c == (ucp_table[mid].f0 & f0_charmask)) break;
  if (c < (ucp_table[mid].f0 & f0_charmask)) top = mid;
  else
    {
    if ((ucp_table[mid].f0 & f0_rangeflag) != 0 &&
        c <= (ucp_table[mid].f0 & f0_charmask) +
             (ucp_table[mid].f1 & f1_rangemask)) break;
    bot = mid + 1;
    }
  }

/* Found an entry in the table. Set the script and detailed type values, and
return the general type. */

*script_ptr = (ucp_table[mid].f0 & f0_scriptmask) >> f0_scriptshift;
*type_ptr = (ucp_table[mid].f1 & f1_typemask) >> f1_typeshift;

return ucp_gentype[*type_ptr];
}


/*************************************************
*       Search table and return other case       *
*************************************************/

/* If the given character is a letter, and there is another case for the
letter, return the other case. Otherwise, return -1.

Arguments:
  c           the character value

Returns:      the other case or NOTACHAR if none
*/

unsigned int
_pcre_ucp_othercase(const unsigned int c)
{
int bot = 0;
int top = sizeof(ucp_table)/sizeof(cnode);
int mid, offset;

/* The table is searched using a binary chop. You might think that using
intermediate variables to hold some of the common expressions would speed
things up, but tests with gcc 3.4.4 on Linux showed that, on the contrary, it
makes things a lot slower. */

for (;;)
  {
  if (top <= bot) return -1;
  mid = (bot + top) >> 1;
  if (c == (ucp_table[mid].f0 & f0_charmask)) break;
  if (c < (ucp_table[mid].f0 & f0_charmask)) top = mid;
  else
    {
    if ((ucp_table[mid].f0 & f0_rangeflag) != 0 &&
        c <= (ucp_table[mid].f0 & f0_charmask) +
             (ucp_table[mid].f1 & f1_rangemask)) break;
    bot = mid + 1;
    }
  }

/* Found an entry in the table. Return NOTACHAR for a range entry. Otherwise
return the other case if there is one, else NOTACHAR. */

if ((ucp_table[mid].f0 & f0_rangeflag) != 0) return NOTACHAR;

offset = ucp_table[mid].f1 & f1_casemask;
if ((offset & f1_caseneg) != 0) offset |= f1_caseneg;
return (offset == 0)? NOTACHAR : c + offset;
}


/* End of pcre_ucp_searchfuncs.c */
1	/*************************************************
2	* Perl-Compatible Regular Expressions *
3	*************************************************/
4
5	/* PCRE is a library of functions to support regular expressions whose syntax
6	and semantics are as close as possible to those of the Perl 5 language.
7
8	Written by Philip Hazel
9	Copyright (c) 1997-2008 University of Cambridge
10
11	-----------------------------------------------------------------------------
12	Redistribution and use in source and binary forms, with or without
13	modification, are permitted provided that the following conditions are met:
14
15	* Redistributions of source code must retain the above copyright notice,
16	this list of conditions and the following disclaimer.
17
18	* Redistributions in binary form must reproduce the above copyright
19	notice, this list of conditions and the following disclaimer in the
20	documentation and/or other materials provided with the distribution.
21
22	* Neither the name of the University of Cambridge nor the names of its
23	contributors may be used to endorse or promote products derived from
24	this software without specific prior written permission.
25
26	THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
27	AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
28	IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
29	ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
30	LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
31	CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
32	SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
33	INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
34	CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
35	ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
36	POSSIBILITY OF SUCH DAMAGE.
37	-----------------------------------------------------------------------------
38	*/
39
40
41	/* This module contains code for searching the table of Unicode character
42	properties. */
43
44	#ifdef HAVE_CONFIG_H
45	#include "config.h"
46	#endif
47
48	#include "pcre_internal.h"
49
50	#include "ucp.h" /* Category definitions */
51	#include "ucpinternal.h" /* Internal table details */
52	#include "ucptable.h" /* The table itself */
53
54
55	/* Table to translate from particular type value to the general value. */
56
57	static const int ucp_gentype[] = {
58	ucp_C, ucp_C, ucp_C, ucp_C, ucp_C, /* Cc, Cf, Cn, Co, Cs */
59	ucp_L, ucp_L, ucp_L, ucp_L, ucp_L, /* Ll, Lu, Lm, Lo, Lt */
60	ucp_M, ucp_M, ucp_M, /* Mc, Me, Mn */
61	ucp_N, ucp_N, ucp_N, /* Nd, Nl, No */
62	ucp_P, ucp_P, ucp_P, ucp_P, ucp_P, /* Pc, Pd, Pe, Pf, Pi */
63	ucp_P, ucp_P, /* Ps, Po */
64	ucp_S, ucp_S, ucp_S, ucp_S, /* Sc, Sk, Sm, So */
65	ucp_Z, ucp_Z, ucp_Z /* Zl, Zp, Zs */
66	};
67
68
69
70	/*************************************************
71	* Search table and return type *
72	*************************************************/
73
74	/* Three values are returned: the category is ucp_C, ucp_L, etc. The detailed
75	character type is ucp_Lu, ucp_Nd, etc. The script is ucp_Latin, etc.
76
77	Arguments:
78	c the character value
79	type_ptr the detailed character type is returned here
80	script_ptr the script is returned here
81
82	Returns: the character type category
83	*/
84
85	int
86	_pcre_ucp_findprop(const unsigned int c, int type_ptr, int script_ptr)
87	{
88	int bot = 0;
89	int top = sizeof(ucp_table)/sizeof(cnode);
90	int mid;
91
92	/* The table is searched using a binary chop. You might think that using
93	intermediate variables to hold some of the common expressions would speed
94	things up, but tests with gcc 3.4.4 on Linux showed that, on the contrary, it
95	makes things a lot slower. */
96
97	for (;;)
98	{
99	if (top <= bot)
100	{
101	*type_ptr = ucp_Cn;
102	*script_ptr = ucp_Common;
103	return ucp_C;
104	}
105	mid = (bot + top) >> 1;
106	if (c == (ucp_table[mid].f0 & f0_charmask)) break;
107	if (c < (ucp_table[mid].f0 & f0_charmask)) top = mid;
108	else
109	{
110	if ((ucp_table[mid].f0 & f0_rangeflag) != 0 &&
111	c <= (ucp_table[mid].f0 & f0_charmask) +
112	(ucp_table[mid].f1 & f1_rangemask)) break;
113	bot = mid + 1;
114	}
115	}
116
117	/* Found an entry in the table. Set the script and detailed type values, and
118	return the general type. */
119
120	*script_ptr = (ucp_table[mid].f0 & f0_scriptmask) >> f0_scriptshift;
121	*type_ptr = (ucp_table[mid].f1 & f1_typemask) >> f1_typeshift;
122
123	return ucp_gentype[*type_ptr];
124	}
125
126
127
128	/*************************************************
129	* Search table and return other case *
130	*************************************************/
131
132	/* If the given character is a letter, and there is another case for the
133	letter, return the other case. Otherwise, return -1.
134
135	Arguments:
136	c the character value
137
138	Returns: the other case or NOTACHAR if none
139	*/
140
141	unsigned int
142	_pcre_ucp_othercase(const unsigned int c)
143	{
144	int bot = 0;
145	int top = sizeof(ucp_table)/sizeof(cnode);
146	int mid, offset;
147
148	/* The table is searched using a binary chop. You might think that using
149	intermediate variables to hold some of the common expressions would speed
150	things up, but tests with gcc 3.4.4 on Linux showed that, on the contrary, it
151	makes things a lot slower. */
152
153	for (;;)
154	{
155	if (top <= bot) return -1;
156	mid = (bot + top) >> 1;
157	if (c == (ucp_table[mid].f0 & f0_charmask)) break;
158	if (c < (ucp_table[mid].f0 & f0_charmask)) top = mid;
159	else
160	{
161	if ((ucp_table[mid].f0 & f0_rangeflag) != 0 &&
162	c <= (ucp_table[mid].f0 & f0_charmask) +
163	(ucp_table[mid].f1 & f1_rangemask)) break;
164	bot = mid + 1;
165	}
166	}
167
168	/* Found an entry in the table. Return NOTACHAR for a range entry. Otherwise
169	return the other case if there is one, else NOTACHAR. */
170
171	if ((ucp_table[mid].f0 & f0_rangeflag) != 0) return NOTACHAR;
172
173	offset = ucp_table[mid].f1 & f1_casemask;
174	if ((offset & f1_caseneg) != 0) offset \|= f1_caseneg;
175	return (offset == 0)? NOTACHAR : c + offset;
176	}
177
178
179	/* End of pcre_ucp_searchfuncs.c */
Name	Value
svn:eol-style	native
svn:keywords	"Author Date Id Revision Url"