dist/src/encoding.c

71  * ubuf, and the number of characters converted in ulen.
74 file_encoding(struct magic_set *ms, const unsigned char *buf, size_t nbytes, unichar **ubuf, size_t *ulen, const char **code, const char **code_mime, const char **type)
86 	mlen = (nbytes + 1) * sizeof((*ubuf)[0]);
87 	if ((*ubuf = CAST(unichar *, calloc((size_t)1, mlen))) == NULL) {
92 	if (looks_ascii(buf, nbytes, *ubuf, ulen)) {
96 	} else if (looks_utf8_with_BOM(buf, nbytes, *ubuf, ulen) > 0) {
100 	} else if (file_looks_utf8(buf, nbytes, *ubuf, ulen) > 1) {
105 	} else if ((ucs_type = looks_ucs16(buf, nbytes, *ubuf, ulen)) != 0) {
114 	} else if (looks_latin1(buf, nbytes, *ubuf, ulen)) {
118 	} else if (looks_extended(buf, nbytes, *ubuf, ulen)) {
125 		if (looks_ascii(nbuf, nbytes, *ubuf, ulen)) {
129 		} else if (looks_latin1(nbuf, nbytes, *ubuf, ulen)) {
227 looks_ascii(const unsigned char *buf, size_t nbytes, unichar *ubuf,
240 		ubuf[(*ulen)++] = buf[i];
247 looks_latin1(const unsigned char *buf, size_t nbytes, unichar *ubuf, size_t *ulen)
259 		ubuf[(*ulen)++] = buf[i];
266 looks_extended(const unsigned char *buf, size_t nbytes, unichar *ubuf,
279 		ubuf[(*ulen)++] = buf[i];
293  * If ubuf is non-NULL on entry, text is decoded into ubuf, *ulen;
294  * ubuf must be big enough!
297 file_looks_utf8(const unsigned char *buf, size_t nbytes, unichar *ubuf, size_t *ulen)
304 	if (ubuf)
317 			if (ubuf)
318 				ubuf[(*ulen)++] = buf[i];
353 			if (ubuf)
354 				ubuf[(*ulen)++] = c;
368 looks_utf8_with_BOM(const unsigned char *buf, size_t nbytes, unichar *ubuf,
372 		return file_looks_utf8(buf + 3, nbytes - 3, ubuf, ulen);
378 looks_ucs16(const unsigned char *buf, size_t nbytes, unichar *ubuf,
400 			ubuf[(*ulen)++] = buf[i + 1] + 256 * buf[i];
402 			ubuf[(*ulen)++] = buf[i] + 256 * buf[i + 1];
404 		if (ubuf[*ulen - 1] == 0xfffe)
406 		if (ubuf[*ulen - 1] < 128 &&
407 		    text_chars[(size_t)ubuf[*ulen - 1]] != T)