1 /*
2  * "streamable kanji code filter and converter"
3  * Copyright (c) 1998-2002 HappySize, Inc. All rights reserved.
4  *
5  * LICENSE NOTICES
6  *
7  * This file is part of "streamable kanji code filter and converter",
8  * which is distributed under the terms of GNU Lesser General Public
9  * License (version 2) as published by the Free Software Foundation.
10  *
11  * This software is distributed in the hope that it will be useful,
12  * but WITHOUT ANY WARRANTY; without even the implied warranty of
13  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
14  * GNU Lesser General Public License for more details.
15  *
16  * You should have received a copy of the GNU Lesser General Public
17  * License along with "streamable kanji code filter and converter";
18  * if not, write to the Free Software Foundation, Inc., 59 Temple Place,
19  * Suite 330, Boston, MA  02111-1307  USA
20  *
21  * The author of this part: Haluk AKIN <halukakin@gmail.com>
22  *
23  */
24 /*
25  * The source code included in this files was separated from mbfilter_ru.c
26  * by moriyoshi koizumi <moriyoshi@php.net> on 4 dec 2002.
27  *
28  */
29 
30 #ifdef HAVE_CONFIG_H
31 #include "config.h"
32 #endif
33 
34 #include "mbfilter.h"
35 #include "mbfilter_cp1254.h"
36 #include "unicode_table_cp1254.h"
37 
38 static int mbfl_filt_ident_cp1254(int c, mbfl_identify_filter *filter);
39 
40 static const char *mbfl_encoding_cp1254_aliases[] = {"CP1254", "CP-1254", "WINDOWS-1254", NULL};
41 
42 const mbfl_encoding mbfl_encoding_cp1254 = {
43 	mbfl_no_encoding_cp1254,
44 	"Windows-1254",
45 	"Windows-1254",
46 	(const char *(*)[])&mbfl_encoding_cp1254_aliases,
47 	NULL,
48 	MBFL_ENCTYPE_SBCS
49 };
50 
51 const struct mbfl_identify_vtbl vtbl_identify_cp1254 = {
52 	mbfl_no_encoding_cp1254,
53 	mbfl_filt_ident_common_ctor,
54 	mbfl_filt_ident_common_dtor,
55 	mbfl_filt_ident_cp1254
56 };
57 
58 const struct mbfl_convert_vtbl vtbl_cp1254_wchar = {
59 	mbfl_no_encoding_cp1254,
60 	mbfl_no_encoding_wchar,
61 	mbfl_filt_conv_common_ctor,
62 	mbfl_filt_conv_common_dtor,
63 	mbfl_filt_conv_cp1254_wchar,
64 	mbfl_filt_conv_common_flush
65 };
66 
67 const struct mbfl_convert_vtbl vtbl_wchar_cp1254 = {
68 	mbfl_no_encoding_wchar,
69 	mbfl_no_encoding_cp1254,
70 	mbfl_filt_conv_common_ctor,
71 	mbfl_filt_conv_common_dtor,
72 	mbfl_filt_conv_wchar_cp1254,
73 	mbfl_filt_conv_common_flush
74 };
75 
76 #define CK(statement)	do { if ((statement) < 0) return (-1); } while (0)
77 
78 /*
79  * wchar => cp1254
80  */
81 int
mbfl_filt_conv_wchar_cp1254(int c,mbfl_convert_filter * filter)82 mbfl_filt_conv_wchar_cp1254(int c, mbfl_convert_filter *filter)
83 {
84 	int s, n;
85 
86 	if (c < 0x80) {
87 		s = c;
88 	} else {
89 		s = -1;
90 		n = cp1254_ucs_table_len-1;
91 		while (n >= 0) {
92 			if (c == cp1254_ucs_table[n] && c != 0xfffe) {
93 				s = cp1254_ucs_table_min + n;
94 				break;
95 			}
96 			n--;
97 		}
98 		if (s <= 0 && (c & ~MBFL_WCSPLANE_MASK) == MBFL_WCSPLANE_CP1254) {
99 			s = c & MBFL_WCSPLANE_MASK;
100 		}
101 	}
102 
103 	if (s >= 0) {
104 		CK((*filter->output_function)(s, filter->data));
105 	} else {
106 		if (filter->illegal_mode != MBFL_OUTPUTFILTER_ILLEGAL_MODE_NONE) {
107 			CK(mbfl_filt_conv_illegal_output(c, filter));
108 		}
109 	}
110 
111 	return c;
112 }
113 
114 /*
115  * cp1254 => wchar
116  */
117 int
mbfl_filt_conv_cp1254_wchar(int c,mbfl_convert_filter * filter)118 mbfl_filt_conv_cp1254_wchar(int c, mbfl_convert_filter *filter)
119 {
120 	int s;
121 
122 	if (c >= 0 && c < cp1254_ucs_table_min) {
123 		s = c;
124 	} else if (c >= cp1254_ucs_table_min && c < 0x100) {
125 		s = cp1254_ucs_table[c - cp1254_ucs_table_min];
126 		if (s <= 0) {
127 			s = c;
128 			s &= MBFL_WCSPLANE_MASK;
129 			s |= MBFL_WCSPLANE_CP1254;
130 		}
131 	} else {
132 		s = c;
133 		s &= MBFL_WCSGROUP_MASK;
134 		s |= MBFL_WCSGROUP_THROUGH;
135 	}
136 
137 	CK((*filter->output_function)(s, filter->data));
138 
139 	return c;
140 }
141 
142 /* We only distinguish the MS extensions to ISO-8859-1.
143  * Actually, this is pretty much a NO-OP, since the identification
144  * system doesn't allow us to discriminate between a positive match,
145  * a possible match and a definite non-match.
146  * The problem here is that cp1254 looks like SJIS for certain chars.
147  * */
mbfl_filt_ident_cp1254(int c,mbfl_identify_filter * filter)148 static int mbfl_filt_ident_cp1254(int c, mbfl_identify_filter *filter)
149 {
150 	if (c >= 0x80 && c < 0xff)
151 		filter->flag = 0;
152 	else
153 		filter->flag = 1; /* not it */
154 	return c;
155 }
156