1 | /*
|
---|
2 | * MultiByteToWideChar implementation
|
---|
3 | *
|
---|
4 | * Copyright 2000 Alexandre Julliard
|
---|
5 | */
|
---|
6 |
|
---|
7 | #include <string.h>
|
---|
8 |
|
---|
9 | #include "winnls.h"
|
---|
10 | #include "wine/unicode.h"
|
---|
11 |
|
---|
12 | /* get the decomposition of a Unicode char */
|
---|
13 | static int get_decomposition( WCHAR src, WCHAR *dst, unsigned int dstlen )
|
---|
14 | {
|
---|
15 | extern const WCHAR unicode_decompose_table[];
|
---|
16 | const WCHAR *ptr = unicode_decompose_table;
|
---|
17 | int res;
|
---|
18 |
|
---|
19 | *dst = src;
|
---|
20 | ptr = unicode_decompose_table + ptr[src >> 8];
|
---|
21 | ptr = unicode_decompose_table + ptr[(src >> 4) & 0x0f] + 2 * (src & 0x0f);
|
---|
22 | if (!*ptr) return 1;
|
---|
23 | if (dstlen <= 1) return 0;
|
---|
24 | /* apply the decomposition recursively to the first char */
|
---|
25 | if ((res = get_decomposition( *ptr, dst, dstlen-1 ))) dst[res++] = ptr[1];
|
---|
26 | return res;
|
---|
27 | }
|
---|
28 |
|
---|
29 | /* check src string for invalid chars; return non-zero if invalid char found */
|
---|
30 | static inline int check_invalid_chars_sbcs( const struct sbcs_table *table,
|
---|
31 | const unsigned char *src, unsigned int srclen )
|
---|
32 | {
|
---|
33 | const WCHAR * const cp2uni = table->cp2uni;
|
---|
34 | while (srclen)
|
---|
35 | {
|
---|
36 | if (cp2uni[*src] == table->info.def_unicode_char && *src != table->info.def_char)
|
---|
37 | break;
|
---|
38 | src++;
|
---|
39 | srclen--;
|
---|
40 | }
|
---|
41 | return srclen;
|
---|
42 | }
|
---|
43 |
|
---|
44 | /* mbstowcs for single-byte code page */
|
---|
45 | /* all lengths are in characters, not bytes */
|
---|
46 | static inline int mbstowcs_sbcs( const struct sbcs_table *table,
|
---|
47 | const unsigned char *src, unsigned int srclen,
|
---|
48 | WCHAR *dst, unsigned int dstlen )
|
---|
49 | {
|
---|
50 | const WCHAR * const cp2uni = table->cp2uni;
|
---|
51 | int ret = srclen;
|
---|
52 |
|
---|
53 | if (dstlen < srclen)
|
---|
54 | {
|
---|
55 | /* buffer too small: fill it up to dstlen and return error */
|
---|
56 | srclen = dstlen;
|
---|
57 | ret = -1;
|
---|
58 | }
|
---|
59 |
|
---|
60 | for (;;)
|
---|
61 | {
|
---|
62 | switch(srclen)
|
---|
63 | {
|
---|
64 | default:
|
---|
65 | case 16: dst[15] = cp2uni[src[15]];
|
---|
66 | case 15: dst[14] = cp2uni[src[14]];
|
---|
67 | case 14: dst[13] = cp2uni[src[13]];
|
---|
68 | case 13: dst[12] = cp2uni[src[12]];
|
---|
69 | case 12: dst[11] = cp2uni[src[11]];
|
---|
70 | case 11: dst[10] = cp2uni[src[10]];
|
---|
71 | case 10: dst[9] = cp2uni[src[9]];
|
---|
72 | case 9: dst[8] = cp2uni[src[8]];
|
---|
73 | case 8: dst[7] = cp2uni[src[7]];
|
---|
74 | case 7: dst[6] = cp2uni[src[6]];
|
---|
75 | case 6: dst[5] = cp2uni[src[5]];
|
---|
76 | case 5: dst[4] = cp2uni[src[4]];
|
---|
77 | case 4: dst[3] = cp2uni[src[3]];
|
---|
78 | case 3: dst[2] = cp2uni[src[2]];
|
---|
79 | case 2: dst[1] = cp2uni[src[1]];
|
---|
80 | case 1: dst[0] = cp2uni[src[0]];
|
---|
81 | case 0: break;
|
---|
82 | }
|
---|
83 | if (srclen < 16) return ret;
|
---|
84 | dst += 16;
|
---|
85 | src += 16;
|
---|
86 | srclen -= 16;
|
---|
87 | }
|
---|
88 | }
|
---|
89 |
|
---|
90 | /* mbstowcs for single-byte code page with char decomposition */
|
---|
91 | static int mbstowcs_sbcs_decompose( const struct sbcs_table *table,
|
---|
92 | const unsigned char *src, unsigned int srclen,
|
---|
93 | WCHAR *dst, unsigned int dstlen )
|
---|
94 | {
|
---|
95 | const WCHAR * const cp2uni = table->cp2uni;
|
---|
96 | unsigned int len;
|
---|
97 |
|
---|
98 | if (!dstlen) /* compute length */
|
---|
99 | {
|
---|
100 | WCHAR dummy[4]; /* no decomposition is larger than 4 chars */
|
---|
101 | for (len = 0; srclen; srclen--, src++)
|
---|
102 | len += get_decomposition( cp2uni[*src], dummy, 4 );
|
---|
103 | return len;
|
---|
104 | }
|
---|
105 |
|
---|
106 | for (len = dstlen; srclen && len; srclen--, src++)
|
---|
107 | {
|
---|
108 | int res = get_decomposition( cp2uni[*src], dst, len );
|
---|
109 | if (!res) break;
|
---|
110 | len -= res;
|
---|
111 | dst += res;
|
---|
112 | }
|
---|
113 | if (srclen) return -1; /* overflow */
|
---|
114 | return dstlen - len;
|
---|
115 | }
|
---|
116 |
|
---|
117 | /* query necessary dst length for src string */
|
---|
118 | static inline int get_length_dbcs( const struct dbcs_table *table,
|
---|
119 | const unsigned char *src, unsigned int srclen )
|
---|
120 | {
|
---|
121 | const unsigned char * const cp2uni_lb = table->cp2uni_leadbytes;
|
---|
122 | int len;
|
---|
123 |
|
---|
124 | for (len = 0; srclen; srclen--, src++, len++)
|
---|
125 | {
|
---|
126 | if (cp2uni_lb[*src])
|
---|
127 | {
|
---|
128 | if (!--srclen) break; /* partial char, ignore it */
|
---|
129 | src++;
|
---|
130 | }
|
---|
131 | }
|
---|
132 | return len;
|
---|
133 | }
|
---|
134 |
|
---|
135 | /* check src string for invalid chars; return non-zero if invalid char found */
|
---|
136 | static inline int check_invalid_chars_dbcs( const struct dbcs_table *table,
|
---|
137 | const unsigned char *src, unsigned int srclen )
|
---|
138 | {
|
---|
139 | const WCHAR * const cp2uni = table->cp2uni;
|
---|
140 | const unsigned char * const cp2uni_lb = table->cp2uni_leadbytes;
|
---|
141 |
|
---|
142 | while (srclen)
|
---|
143 | {
|
---|
144 | unsigned char off = cp2uni_lb[*src];
|
---|
145 | if (off) /* multi-byte char */
|
---|
146 | {
|
---|
147 | if (srclen == 1) break; /* partial char, error */
|
---|
148 | if (cp2uni[(off << 8) + src[1]] == table->info.def_unicode_char &&
|
---|
149 | ((src[0] << 8) | src[1]) != table->info.def_char) break;
|
---|
150 | src++;
|
---|
151 | srclen--;
|
---|
152 | }
|
---|
153 | else if (cp2uni[*src] == table->info.def_unicode_char &&
|
---|
154 | *src != table->info.def_char) break;
|
---|
155 | src++;
|
---|
156 | srclen--;
|
---|
157 | }
|
---|
158 | return srclen;
|
---|
159 | }
|
---|
160 |
|
---|
161 | /* mbstowcs for double-byte code page */
|
---|
162 | /* all lengths are in characters, not bytes */
|
---|
163 | static inline int mbstowcs_dbcs( const struct dbcs_table *table,
|
---|
164 | const unsigned char *src, unsigned int srclen,
|
---|
165 | WCHAR *dst, unsigned int dstlen )
|
---|
166 | {
|
---|
167 | const WCHAR * const cp2uni = table->cp2uni;
|
---|
168 | const unsigned char * const cp2uni_lb = table->cp2uni_leadbytes;
|
---|
169 | unsigned int len;
|
---|
170 |
|
---|
171 | if (!dstlen) return get_length_dbcs( table, src, srclen );
|
---|
172 |
|
---|
173 | for (len = dstlen; srclen && len; len--, srclen--, src++, dst++)
|
---|
174 | {
|
---|
175 | unsigned char off = cp2uni_lb[*src];
|
---|
176 | if (off)
|
---|
177 | {
|
---|
178 | if (!--srclen) break; /* partial char, ignore it */
|
---|
179 | src++;
|
---|
180 | *dst = cp2uni[(off << 8) + *src];
|
---|
181 | }
|
---|
182 | else *dst = cp2uni[*src];
|
---|
183 | }
|
---|
184 | if (srclen) return -1; /* overflow */
|
---|
185 | return dstlen - len;
|
---|
186 | }
|
---|
187 |
|
---|
188 |
|
---|
189 | /* mbstowcs for double-byte code page with character decomposition */
|
---|
190 | static int mbstowcs_dbcs_decompose( const struct dbcs_table *table,
|
---|
191 | const unsigned char *src, unsigned int srclen,
|
---|
192 | WCHAR *dst, unsigned int dstlen )
|
---|
193 | {
|
---|
194 | const WCHAR * const cp2uni = table->cp2uni;
|
---|
195 | const unsigned char * const cp2uni_lb = table->cp2uni_leadbytes;
|
---|
196 | unsigned int len;
|
---|
197 | WCHAR ch;
|
---|
198 | int res;
|
---|
199 |
|
---|
200 | if (!dstlen) /* compute length */
|
---|
201 | {
|
---|
202 | WCHAR dummy[4]; /* no decomposition is larger than 4 chars */
|
---|
203 | for (len = 0; srclen; srclen--, src++)
|
---|
204 | {
|
---|
205 | unsigned char off = cp2uni_lb[*src];
|
---|
206 | if (off)
|
---|
207 | {
|
---|
208 | if (!--srclen) break; /* partial char, ignore it */
|
---|
209 | src++;
|
---|
210 | ch = cp2uni[(off << 8) + *src];
|
---|
211 | }
|
---|
212 | else ch = cp2uni[*src];
|
---|
213 | len += get_decomposition( ch, dummy, 4 );
|
---|
214 | }
|
---|
215 | return len;
|
---|
216 | }
|
---|
217 |
|
---|
218 | for (len = dstlen; srclen && len; srclen--, src++)
|
---|
219 | {
|
---|
220 | unsigned char off = cp2uni_lb[*src];
|
---|
221 | if (off)
|
---|
222 | {
|
---|
223 | if (!--srclen) break; /* partial char, ignore it */
|
---|
224 | src++;
|
---|
225 | ch = cp2uni[(off << 8) + *src];
|
---|
226 | }
|
---|
227 | else ch = cp2uni[*src];
|
---|
228 | if (!(res = get_decomposition( ch, dst, len ))) break;
|
---|
229 | dst += res;
|
---|
230 | len -= res;
|
---|
231 | }
|
---|
232 | if (srclen) return -1; /* overflow */
|
---|
233 | return dstlen - len;
|
---|
234 | }
|
---|
235 |
|
---|
236 |
|
---|
237 | /* return -1 on dst buffer overflow, -2 on invalid input char */
|
---|
238 | int cp_mbstowcs( const union cptable *table, int flags,
|
---|
239 | const char *src, int srclen,
|
---|
240 | WCHAR *dst, int dstlen )
|
---|
241 | {
|
---|
242 | if (table->info.char_size == 1)
|
---|
243 | {
|
---|
244 | if (flags & MB_ERR_INVALID_CHARS)
|
---|
245 | {
|
---|
246 | if (check_invalid_chars_sbcs( &table->sbcs, (const unsigned char *)src, srclen )) return -2;
|
---|
247 | }
|
---|
248 | if (!(flags & MB_COMPOSITE))
|
---|
249 | {
|
---|
250 | if (!dstlen) return srclen;
|
---|
251 | return mbstowcs_sbcs( &table->sbcs, (const unsigned char *)src, srclen, dst, dstlen );
|
---|
252 | }
|
---|
253 | return mbstowcs_sbcs_decompose( &table->sbcs, (const unsigned char *)src, srclen, dst, dstlen );
|
---|
254 | }
|
---|
255 | else /* mbcs */
|
---|
256 | {
|
---|
257 | if (flags & MB_ERR_INVALID_CHARS)
|
---|
258 | {
|
---|
259 | if (check_invalid_chars_dbcs( &table->dbcs, (const unsigned char *)src, srclen )) return -2;
|
---|
260 | }
|
---|
261 | if (!(flags & MB_COMPOSITE))
|
---|
262 | return mbstowcs_dbcs( &table->dbcs, (const unsigned char *)src, srclen, dst, dstlen );
|
---|
263 | else
|
---|
264 | return mbstowcs_dbcs_decompose( &table->dbcs, (const unsigned char *)src, srclen, dst, dstlen );
|
---|
265 | }
|
---|
266 | }
|
---|