Linux Audio

Check our new training course

Loading...
v4.10.11
  1/*
  2 * unicode.c
  3 *
  4 * PURPOSE
  5 *	Routines for converting between UTF-8 and OSTA Compressed Unicode.
  6 *      Also handles filename mangling
  7 *
  8 * DESCRIPTION
  9 *	OSTA Compressed Unicode is explained in the OSTA UDF specification.
 10 *		http://www.osta.org/
 11 *	UTF-8 is explained in the IETF RFC XXXX.
 12 *		ftp://ftp.internic.net/rfc/rfcxxxx.txt
 13 *
 14 * COPYRIGHT
 15 *	This file is distributed under the terms of the GNU General Public
 16 *	License (GPL). Copies of the GPL can be obtained from:
 17 *		ftp://prep.ai.mit.edu/pub/gnu/GPL
 18 *	Each contributing author retains all rights to their own work.
 19 */
 20
 21#include "udfdecl.h"
 22
 23#include <linux/kernel.h>
 24#include <linux/string.h>	/* for memset */
 25#include <linux/nls.h>
 26#include <linux/crc-itu-t.h>
 27#include <linux/slab.h>
 28
 29#include "udf_sb.h"
 30
 31static int udf_uni2char_utf8(wchar_t uni,
 32			     unsigned char *out,
 33			     int boundlen)
 34{
 35	int u_len = 0;
 
 36
 37	if (boundlen <= 0)
 38		return -ENAMETOOLONG;
 
 
 39
 40	if (uni < 0x80) {
 41		out[u_len++] = (unsigned char)uni;
 42	} else if (uni < 0x800) {
 43		if (boundlen < 2)
 44			return -ENAMETOOLONG;
 45		out[u_len++] = (unsigned char)(0xc0 | (uni >> 6));
 46		out[u_len++] = (unsigned char)(0x80 | (uni & 0x3f));
 47	} else {
 48		if (boundlen < 3)
 49			return -ENAMETOOLONG;
 50		out[u_len++] = (unsigned char)(0xe0 | (uni >> 12));
 51		out[u_len++] = (unsigned char)(0x80 | ((uni >> 6) & 0x3f));
 52		out[u_len++] = (unsigned char)(0x80 | (uni & 0x3f));
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 53	}
 54	return u_len;
 
 
 55}
 56
 57static int udf_char2uni_utf8(const unsigned char *in,
 58			     int boundlen,
 59			     wchar_t *uni)
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 60{
 61	unsigned int utf_char;
 62	unsigned char c;
 63	int utf_cnt, u_len;
 64
 65	utf_char = 0;
 66	utf_cnt = 0;
 67	for (u_len = 0; u_len < boundlen;) {
 68		c = in[u_len++];
 
 
 
 
 
 
 69
 70		/* Complete a multi-byte UTF-8 character */
 71		if (utf_cnt) {
 72			utf_char = (utf_char << 6) | (c & 0x3f);
 73			if (--utf_cnt)
 74				continue;
 75		} else {
 76			/* Check for a multi-byte UTF-8 character */
 77			if (c & 0x80) {
 78				/* Start a multi-byte UTF-8 character */
 79				if ((c & 0xe0) == 0xc0) {
 80					utf_char = c & 0x1f;
 81					utf_cnt = 1;
 82				} else if ((c & 0xf0) == 0xe0) {
 83					utf_char = c & 0x0f;
 84					utf_cnt = 2;
 85				} else if ((c & 0xf8) == 0xf0) {
 86					utf_char = c & 0x07;
 87					utf_cnt = 3;
 88				} else if ((c & 0xfc) == 0xf8) {
 89					utf_char = c & 0x03;
 90					utf_cnt = 4;
 91				} else if ((c & 0xfe) == 0xfc) {
 92					utf_char = c & 0x01;
 93					utf_cnt = 5;
 94				} else {
 95					utf_cnt = -1;
 96					break;
 97				}
 98				continue;
 99			} else {
100				/* Single byte UTF-8 character (most common) */
101				utf_char = c;
102			}
103		}
104		*uni = utf_char;
105		break;
106	}
107	if (utf_cnt) {
108		*uni = '?';
109		return -EINVAL;
110	}
111	return u_len;
112}
113
114#define ILLEGAL_CHAR_MARK	'_'
115#define EXT_MARK		'.'
116#define CRC_MARK		'#'
117#define EXT_SIZE		5
118/* Number of chars we need to store generated CRC to make filename unique */
119#define CRC_LEN			5
120
121static int udf_name_conv_char(uint8_t *str_o, int str_o_max_len,
122			      int *str_o_idx,
123			      const uint8_t *str_i, int str_i_max_len,
124			      int *str_i_idx,
125			      int u_ch, int *needsCRC,
126			      int (*conv_f)(wchar_t, unsigned char *, int),
127			      int translate)
128{
129	uint32_t c;
130	int illChar = 0;
131	int len, gotch = 0;
132
133	for (; (!gotch) && (*str_i_idx < str_i_max_len); *str_i_idx += u_ch) {
134		if (*str_o_idx >= str_o_max_len) {
135			*needsCRC = 1;
136			return gotch;
137		}
138
139		/* Expand OSTA compressed Unicode to Unicode */
140		c = str_i[*str_i_idx];
141		if (u_ch > 1)
142			c = (c << 8) | str_i[*str_i_idx + 1];
143
144		if (translate && (c == '/' || c == 0))
145			illChar = 1;
146		else if (illChar)
147			break;
148		else
149			gotch = 1;
150	}
151	if (illChar) {
152		*needsCRC = 1;
153		c = ILLEGAL_CHAR_MARK;
154		gotch = 1;
155	}
156	if (gotch) {
157		len = conv_f(c, &str_o[*str_o_idx], str_o_max_len - *str_o_idx);
158		/* Valid character? */
159		if (len >= 0)
160			*str_o_idx += len;
161		else if (len == -ENAMETOOLONG) {
162			*needsCRC = 1;
163			gotch = 0;
164		} else {
165			str_o[(*str_o_idx)++] = '?';
166			*needsCRC = 1;
167		}
168	}
169	return gotch;
 
 
 
170}
171
172static int udf_name_from_CS0(uint8_t *str_o, int str_max_len,
173			     const uint8_t *ocu, int ocu_len,
174			     int (*conv_f)(wchar_t, unsigned char *, int),
175			     int translate)
176{
177	uint32_t c;
178	uint8_t cmp_id;
179	int idx, len;
180	int u_ch;
181	int needsCRC = 0;
182	int ext_i_len, ext_max_len;
183	int str_o_len = 0;	/* Length of resulting output */
184	int ext_o_len = 0;	/* Extension output length */
185	int ext_crc_len = 0;	/* Extension output length if used with CRC */
186	int i_ext = -1;		/* Extension position in input buffer */
187	int o_crc = 0;		/* Rightmost possible output pos for CRC+ext */
188	unsigned short valueCRC;
189	uint8_t ext[EXT_SIZE * NLS_MAX_CHARSET_SIZE + 1];
190	uint8_t crc[CRC_LEN];
191
192	if (str_max_len <= 0)
193		return 0;
194
 
195	if (ocu_len == 0) {
196		memset(str_o, 0, str_max_len);
197		return 0;
198	}
199
200	cmp_id = ocu[0];
201	if (cmp_id != 8 && cmp_id != 16) {
202		memset(str_o, 0, str_max_len);
203		pr_err("unknown compression code (%d)\n", cmp_id);
204		return -EINVAL;
205	}
206	u_ch = cmp_id >> 3;
207
208	ocu++;
209	ocu_len--;
210
211	if (ocu_len % u_ch) {
212		pr_err("incorrect filename length (%d)\n", ocu_len + 1);
213		return -EINVAL;
214	}
215
216	if (translate) {
217		/* Look for extension */
218		for (idx = ocu_len - u_ch, ext_i_len = 0;
219		     (idx >= 0) && (ext_i_len < EXT_SIZE);
220		     idx -= u_ch, ext_i_len++) {
221			c = ocu[idx];
222			if (u_ch > 1)
223				c = (c << 8) | ocu[idx + 1];
224
225			if (c == EXT_MARK) {
226				if (ext_i_len)
227					i_ext = idx;
228				break;
229			}
230		}
231		if (i_ext >= 0) {
232			/* Convert extension */
233			ext_max_len = min_t(int, sizeof(ext), str_max_len);
234			ext[ext_o_len++] = EXT_MARK;
235			idx = i_ext + u_ch;
236			while (udf_name_conv_char(ext, ext_max_len, &ext_o_len,
237						  ocu, ocu_len, &idx,
238						  u_ch, &needsCRC,
239						  conv_f, translate)) {
240				if ((ext_o_len + CRC_LEN) < str_max_len)
241					ext_crc_len = ext_o_len;
242			}
243		}
244	}
245
246	idx = 0;
247	while (1) {
248		if (translate && (idx == i_ext)) {
249			if (str_o_len > (str_max_len - ext_o_len))
250				needsCRC = 1;
251			break;
252		}
253
254		if (!udf_name_conv_char(str_o, str_max_len, &str_o_len,
255					ocu, ocu_len, &idx,
256					u_ch, &needsCRC, conv_f, translate))
257			break;
258
259		if (translate &&
260		    (str_o_len <= (str_max_len - ext_o_len - CRC_LEN)))
261			o_crc = str_o_len;
262	}
263
264	if (translate) {
265		if (str_o_len <= 2 && str_o[0] == '.' &&
266		    (str_o_len == 1 || str_o[1] == '.'))
267			needsCRC = 1;
268		if (needsCRC) {
269			str_o_len = o_crc;
270			valueCRC = crc_itu_t(0, ocu, ocu_len);
271			crc[0] = CRC_MARK;
272			crc[1] = hex_asc_upper_hi(valueCRC >> 8);
273			crc[2] = hex_asc_upper_lo(valueCRC >> 8);
274			crc[3] = hex_asc_upper_hi(valueCRC);
275			crc[4] = hex_asc_upper_lo(valueCRC);
276			len = min_t(int, CRC_LEN, str_max_len - str_o_len);
277			memcpy(&str_o[str_o_len], crc, len);
278			str_o_len += len;
279			ext_o_len = ext_crc_len;
280		}
281		if (ext_o_len > 0) {
282			memcpy(&str_o[str_o_len], ext, ext_o_len);
283			str_o_len += ext_o_len;
284		}
285	}
 
286
287	return str_o_len;
288}
289
290static int udf_name_to_CS0(uint8_t *ocu, int ocu_max_len,
291			   const uint8_t *str_i, int str_len,
292			   int (*conv_f)(const unsigned char *, int, wchar_t *))
293{
294	int i, len;
295	unsigned int max_val;
296	wchar_t uni_char;
297	int u_len, u_ch;
298
299	if (ocu_max_len <= 0)
300		return 0;
301
302	memset(ocu, 0, ocu_max_len);
303	ocu[0] = 8;
304	max_val = 0xff;
305	u_ch = 1;
306
307try_again:
308	u_len = 1;
309	for (i = 0; i < str_len; i++) {
310		/* Name didn't fit? */
311		if (u_len + u_ch > ocu_max_len)
312			return 0;
313		len = conv_f(&str_i[i], str_len - i, &uni_char);
314		if (!len)
315			continue;
316		/* Invalid character, deal with it */
317		if (len < 0) {
318			len = 1;
319			uni_char = '?';
320		}
321
322		if (uni_char > max_val) {
323			max_val = 0xffff;
324			ocu[0] = 0x10;
325			u_ch = 2;
326			goto try_again;
327		}
328
329		if (max_val == 0xffff)
330			ocu[u_len++] = (uint8_t)(uni_char >> 8);
331		ocu[u_len++] = (uint8_t)(uni_char & 0xff);
332		i += len - 1;
333	}
334
335	return u_len;
 
336}
337
338int udf_dstrCS0toUTF8(uint8_t *utf_o, int o_len,
339		      const uint8_t *ocu_i, int i_len)
340{
341	int s_len = 0;
 
 
 
 
 
342
343	if (i_len > 0) {
344		s_len = ocu_i[i_len - 1];
345		if (s_len >= i_len) {
346			pr_err("incorrect dstring lengths (%d/%d)\n",
347			       s_len, i_len);
348			return -EINVAL;
 
 
 
 
 
 
349		}
350	}
 
 
 
 
 
 
 
 
351
352	return udf_name_from_CS0(utf_o, o_len, ocu_i, s_len,
353				 udf_uni2char_utf8, 0);
 
 
 
 
 
354}
355
356int udf_get_filename(struct super_block *sb, const uint8_t *sname, int slen,
357		     uint8_t *dname, int dlen)
358{
359	int (*conv_f)(wchar_t, unsigned char *, int);
360	int ret;
361
362	if (!slen)
363		return -EIO;
364
365	if (dlen <= 0)
366		return 0;
367
368	if (UDF_QUERY_FLAG(sb, UDF_FLAG_UTF8)) {
369		conv_f = udf_uni2char_utf8;
 
 
370	} else if (UDF_QUERY_FLAG(sb, UDF_FLAG_NLS_MAP)) {
371		conv_f = UDF_SB(sb)->s_nls_map->uni2char;
 
 
 
372	} else
373		BUG();
374
375	ret = udf_name_from_CS0(dname, dlen, sname, slen, conv_f, 1);
376	/* Zero length filename isn't valid... */
377	if (ret == 0)
378		ret = -EINVAL;
379	return ret;
380}
381
382int udf_put_filename(struct super_block *sb, const uint8_t *sname, int slen,
383		     uint8_t *dname, int dlen)
 
 
 
 
 
 
384{
385	int (*conv_f)(const unsigned char *, int, wchar_t *);
 
 
 
 
386
387	if (UDF_QUERY_FLAG(sb, UDF_FLAG_UTF8)) {
388		conv_f = udf_char2uni_utf8;
389	} else if (UDF_QUERY_FLAG(sb, UDF_FLAG_NLS_MAP)) {
390		conv_f = UDF_SB(sb)->s_nls_map->char2uni;
391	} else
392		BUG();
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
393
394	return udf_name_to_CS0(dname, dlen, sname, slen, conv_f);
395}
396
v3.1
  1/*
  2 * unicode.c
  3 *
  4 * PURPOSE
  5 *	Routines for converting between UTF-8 and OSTA Compressed Unicode.
  6 *      Also handles filename mangling
  7 *
  8 * DESCRIPTION
  9 *	OSTA Compressed Unicode is explained in the OSTA UDF specification.
 10 *		http://www.osta.org/
 11 *	UTF-8 is explained in the IETF RFC XXXX.
 12 *		ftp://ftp.internic.net/rfc/rfcxxxx.txt
 13 *
 14 * COPYRIGHT
 15 *	This file is distributed under the terms of the GNU General Public
 16 *	License (GPL). Copies of the GPL can be obtained from:
 17 *		ftp://prep.ai.mit.edu/pub/gnu/GPL
 18 *	Each contributing author retains all rights to their own work.
 19 */
 20
 21#include "udfdecl.h"
 22
 23#include <linux/kernel.h>
 24#include <linux/string.h>	/* for memset */
 25#include <linux/nls.h>
 26#include <linux/crc-itu-t.h>
 27#include <linux/slab.h>
 28
 29#include "udf_sb.h"
 30
 31static int udf_translate_to_linux(uint8_t *, uint8_t *, int, uint8_t *, int);
 32
 33static int udf_char_to_ustr(struct ustr *dest, const uint8_t *src, int strlen)
 34{
 35	if ((!dest) || (!src) || (!strlen) || (strlen > UDF_NAME_LEN - 2))
 36		return 0;
 37
 38	memset(dest, 0, sizeof(struct ustr));
 39	memcpy(dest->u_name, src, strlen);
 40	dest->u_cmpID = 0x08;
 41	dest->u_len = strlen;
 42
 43	return strlen;
 44}
 45
 46/*
 47 * udf_build_ustr
 48 */
 49int udf_build_ustr(struct ustr *dest, dstring *ptr, int size)
 50{
 51	int usesize;
 52
 53	if (!dest || !ptr || !size)
 54		return -1;
 55	BUG_ON(size < 2);
 56
 57	usesize = min_t(size_t, ptr[size - 1], sizeof(dest->u_name));
 58	usesize = min(usesize, size - 2);
 59	dest->u_cmpID = ptr[0];
 60	dest->u_len = usesize;
 61	memcpy(dest->u_name, ptr + 1, usesize);
 62	memset(dest->u_name + usesize, 0, sizeof(dest->u_name) - usesize);
 63
 64	return 0;
 65}
 66
 67/*
 68 * udf_build_ustr_exact
 69 */
 70static int udf_build_ustr_exact(struct ustr *dest, dstring *ptr, int exactsize)
 71{
 72	if ((!dest) || (!ptr) || (!exactsize))
 73		return -1;
 74
 75	memset(dest, 0, sizeof(struct ustr));
 76	dest->u_cmpID = ptr[0];
 77	dest->u_len = exactsize - 1;
 78	memcpy(dest->u_name, ptr + 1, exactsize - 1);
 79
 80	return 0;
 81}
 82
 83/*
 84 * udf_ocu_to_utf8
 85 *
 86 * PURPOSE
 87 *	Convert OSTA Compressed Unicode to the UTF-8 equivalent.
 88 *
 89 * PRE-CONDITIONS
 90 *	utf			Pointer to UTF-8 output buffer.
 91 *	ocu			Pointer to OSTA Compressed Unicode input buffer
 92 *				of size UDF_NAME_LEN bytes.
 93 * 				both of type "struct ustr *"
 94 *
 95 * POST-CONDITIONS
 96 *	<return>		Zero on success.
 97 *
 98 * HISTORY
 99 *	November 12, 1997 - Andrew E. Mileski
100 *	Written, tested, and released.
101 */
102int udf_CS0toUTF8(struct ustr *utf_o, const struct ustr *ocu_i)
103{
104	const uint8_t *ocu;
105	uint8_t cmp_id, ocu_len;
106	int i;
107
108	ocu_len = ocu_i->u_len;
109	if (ocu_len == 0) {
110		memset(utf_o, 0, sizeof(struct ustr));
111		return 0;
112	}
113
114	cmp_id = ocu_i->u_cmpID;
115	if (cmp_id != 8 && cmp_id != 16) {
116		memset(utf_o, 0, sizeof(struct ustr));
117		printk(KERN_ERR "udf: unknown compression code (%d) stri=%s\n",
118		       cmp_id, ocu_i->u_name);
119		return 0;
120	}
121
122	ocu = ocu_i->u_name;
123	utf_o->u_len = 0;
124	for (i = 0; (i < ocu_len) && (utf_o->u_len <= (UDF_NAME_LEN - 3));) {
125
126		/* Expand OSTA compressed Unicode to Unicode */
127		uint32_t c = ocu[i++];
128		if (cmp_id == 16)
129			c = (c << 8) | ocu[i++];
130
131		/* Compress Unicode to UTF-8 */
132		if (c < 0x80U)
133			utf_o->u_name[utf_o->u_len++] = (uint8_t)c;
134		else if (c < 0x800U) {
135			utf_o->u_name[utf_o->u_len++] =
136						(uint8_t)(0xc0 | (c >> 6));
137			utf_o->u_name[utf_o->u_len++] =
138						(uint8_t)(0x80 | (c & 0x3f));
139		} else {
140			utf_o->u_name[utf_o->u_len++] =
141						(uint8_t)(0xe0 | (c >> 12));
142			utf_o->u_name[utf_o->u_len++] =
143						(uint8_t)(0x80 |
144							  ((c >> 6) & 0x3f));
145			utf_o->u_name[utf_o->u_len++] =
146						(uint8_t)(0x80 | (c & 0x3f));
147		}
148	}
149	utf_o->u_cmpID = 8;
150
151	return utf_o->u_len;
152}
153
154/*
155 *
156 * udf_utf8_to_ocu
157 *
158 * PURPOSE
159 *	Convert UTF-8 to the OSTA Compressed Unicode equivalent.
160 *
161 * DESCRIPTION
162 *	This routine is only called by udf_lookup().
163 *
164 * PRE-CONDITIONS
165 *	ocu			Pointer to OSTA Compressed Unicode output
166 *				buffer of size UDF_NAME_LEN bytes.
167 *	utf			Pointer to UTF-8 input buffer.
168 *	utf_len			Length of UTF-8 input buffer in bytes.
169 *
170 * POST-CONDITIONS
171 *	<return>		Zero on success.
172 *
173 * HISTORY
174 *	November 12, 1997 - Andrew E. Mileski
175 *	Written, tested, and released.
176 */
177static int udf_UTF8toCS0(dstring *ocu, struct ustr *utf, int length)
178{
179	unsigned c, i, max_val, utf_char;
 
180	int utf_cnt, u_len;
181
182	memset(ocu, 0, sizeof(dstring) * length);
183	ocu[0] = 8;
184	max_val = 0xffU;
185
186try_again:
187	u_len = 0U;
188	utf_char = 0U;
189	utf_cnt = 0U;
190	for (i = 0U; i < utf->u_len; i++) {
191		c = (uint8_t)utf->u_name[i];
192
193		/* Complete a multi-byte UTF-8 character */
194		if (utf_cnt) {
195			utf_char = (utf_char << 6) | (c & 0x3fU);
196			if (--utf_cnt)
197				continue;
198		} else {
199			/* Check for a multi-byte UTF-8 character */
200			if (c & 0x80U) {
201				/* Start a multi-byte UTF-8 character */
202				if ((c & 0xe0U) == 0xc0U) {
203					utf_char = c & 0x1fU;
204					utf_cnt = 1;
205				} else if ((c & 0xf0U) == 0xe0U) {
206					utf_char = c & 0x0fU;
207					utf_cnt = 2;
208				} else if ((c & 0xf8U) == 0xf0U) {
209					utf_char = c & 0x07U;
210					utf_cnt = 3;
211				} else if ((c & 0xfcU) == 0xf8U) {
212					utf_char = c & 0x03U;
213					utf_cnt = 4;
214				} else if ((c & 0xfeU) == 0xfcU) {
215					utf_char = c & 0x01U;
216					utf_cnt = 5;
217				} else {
218					goto error_out;
 
219				}
220				continue;
221			} else {
222				/* Single byte UTF-8 character (most common) */
223				utf_char = c;
224			}
225		}
 
 
 
 
 
 
 
 
 
226
227		/* Choose no compression if necessary */
228		if (utf_char > max_val) {
229			if (max_val == 0xffU) {
230				max_val = 0xffffU;
231				ocu[0] = (uint8_t)0x10U;
232				goto try_again;
233			}
234			goto error_out;
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
235		}
236
237		if (max_val == 0xffffU)
238			ocu[++u_len] = (uint8_t)(utf_char >> 8);
239		ocu[++u_len] = (uint8_t)(utf_char & 0xffU);
 
 
 
 
 
 
 
 
 
 
 
 
 
240	}
241
242	if (utf_cnt) {
243error_out:
244		ocu[++u_len] = '?';
245		printk(KERN_DEBUG "udf: bad UTF-8 character\n");
 
 
 
 
 
 
 
246	}
247
248	ocu[length - 1] = (uint8_t)u_len + 1;
249
250	return u_len + 1;
251}
252
253static int udf_CS0toNLS(struct nls_table *nls, struct ustr *utf_o,
254			const struct ustr *ocu_i)
 
 
255{
256	const uint8_t *ocu;
257	uint8_t cmp_id, ocu_len;
258	int i, len;
 
 
 
 
 
 
 
 
 
 
 
259
 
 
260
261	ocu_len = ocu_i->u_len;
262	if (ocu_len == 0) {
263		memset(utf_o, 0, sizeof(struct ustr));
264		return 0;
265	}
266
267	cmp_id = ocu_i->u_cmpID;
268	if (cmp_id != 8 && cmp_id != 16) {
269		memset(utf_o, 0, sizeof(struct ustr));
270		printk(KERN_ERR "udf: unknown compression code (%d) stri=%s\n",
271		       cmp_id, ocu_i->u_name);
272		return 0;
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
273	}
274
275	ocu = ocu_i->u_name;
276	utf_o->u_len = 0;
277	for (i = 0; (i < ocu_len) && (utf_o->u_len <= (UDF_NAME_LEN - 3));) {
278		/* Expand OSTA compressed Unicode to Unicode */
279		uint32_t c = ocu[i++];
280		if (cmp_id == 16)
281			c = (c << 8) | ocu[i++];
282
283		len = nls->uni2char(c, &utf_o->u_name[utf_o->u_len],
284				    UDF_NAME_LEN - utf_o->u_len);
285		/* Valid character? */
286		if (len >= 0)
287			utf_o->u_len += len;
288		else
289			utf_o->u_name[utf_o->u_len++] = '?';
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
290	}
291	utf_o->u_cmpID = 8;
292
293	return utf_o->u_len;
294}
295
296static int udf_NLStoCS0(struct nls_table *nls, dstring *ocu, struct ustr *uni,
297			int length)
 
298{
299	int len;
300	unsigned i, max_val;
301	uint16_t uni_char;
302	int u_len;
 
 
 
303
304	memset(ocu, 0, sizeof(dstring) * length);
305	ocu[0] = 8;
306	max_val = 0xffU;
 
307
308try_again:
309	u_len = 0U;
310	for (i = 0U; i < uni->u_len; i++) {
311		len = nls->char2uni(&uni->u_name[i], uni->u_len - i, &uni_char);
 
 
 
312		if (!len)
313			continue;
314		/* Invalid character, deal with it */
315		if (len < 0) {
316			len = 1;
317			uni_char = '?';
318		}
319
320		if (uni_char > max_val) {
321			max_val = 0xffffU;
322			ocu[0] = (uint8_t)0x10U;
 
323			goto try_again;
324		}
325
326		if (max_val == 0xffffU)
327			ocu[++u_len] = (uint8_t)(uni_char >> 8);
328		ocu[++u_len] = (uint8_t)(uni_char & 0xffU);
329		i += len - 1;
330	}
331
332	ocu[length - 1] = (uint8_t)u_len + 1;
333	return u_len + 1;
334}
335
336int udf_get_filename(struct super_block *sb, uint8_t *sname, uint8_t *dname,
337		     int flen)
338{
339	struct ustr *filename, *unifilename;
340	int len = 0;
341
342	filename = kmalloc(sizeof(struct ustr), GFP_NOFS);
343	if (!filename)
344		return 0;
345
346	unifilename = kmalloc(sizeof(struct ustr), GFP_NOFS);
347	if (!unifilename)
348		goto out1;
349
350	if (udf_build_ustr_exact(unifilename, sname, flen))
351		goto out2;
352
353	if (UDF_QUERY_FLAG(sb, UDF_FLAG_UTF8)) {
354		if (!udf_CS0toUTF8(filename, unifilename)) {
355			udf_debug("Failed in udf_get_filename: sname = %s\n",
356				  sname);
357			goto out2;
358		}
359	} else if (UDF_QUERY_FLAG(sb, UDF_FLAG_NLS_MAP)) {
360		if (!udf_CS0toNLS(UDF_SB(sb)->s_nls_map, filename,
361				  unifilename)) {
362			udf_debug("Failed in udf_get_filename: sname = %s\n",
363				  sname);
364			goto out2;
365		}
366	} else
367		goto out2;
368
369	len = udf_translate_to_linux(dname, filename->u_name, filename->u_len,
370				     unifilename->u_name, unifilename->u_len);
371out2:
372	kfree(unifilename);
373out1:
374	kfree(filename);
375	return len;
376}
377
378int udf_put_filename(struct super_block *sb, const uint8_t *sname,
379		     uint8_t *dname, int flen)
380{
381	struct ustr unifilename;
382	int namelen;
 
 
 
383
384	if (!udf_char_to_ustr(&unifilename, sname, flen))
385		return 0;
386
387	if (UDF_QUERY_FLAG(sb, UDF_FLAG_UTF8)) {
388		namelen = udf_UTF8toCS0(dname, &unifilename, UDF_NAME_LEN);
389		if (!namelen)
390			return 0;
391	} else if (UDF_QUERY_FLAG(sb, UDF_FLAG_NLS_MAP)) {
392		namelen = udf_NLStoCS0(UDF_SB(sb)->s_nls_map, dname,
393					&unifilename, UDF_NAME_LEN);
394		if (!namelen)
395			return 0;
396	} else
397		return 0;
398
399	return namelen;
 
 
 
 
400}
401
402#define ILLEGAL_CHAR_MARK	'_'
403#define EXT_MARK		'.'
404#define CRC_MARK		'#'
405#define EXT_SIZE 		5
406
407static int udf_translate_to_linux(uint8_t *newName, uint8_t *udfName,
408				  int udfLen, uint8_t *fidName,
409				  int fidNameLen)
410{
411	int index, newIndex = 0, needsCRC = 0;
412	int extIndex = 0, newExtIndex = 0, hasExt = 0;
413	unsigned short valueCRC;
414	uint8_t curr;
415	const uint8_t hexChar[] = "0123456789ABCDEF";
416
417	if (udfName[0] == '.' &&
418	    (udfLen == 1 || (udfLen == 2 && udfName[1] == '.'))) {
419		needsCRC = 1;
420		newIndex = udfLen;
421		memcpy(newName, udfName, udfLen);
422	} else {
423		for (index = 0; index < udfLen; index++) {
424			curr = udfName[index];
425			if (curr == '/' || curr == 0) {
426				needsCRC = 1;
427				curr = ILLEGAL_CHAR_MARK;
428				while (index + 1 < udfLen &&
429						(udfName[index + 1] == '/' ||
430						 udfName[index + 1] == 0))
431					index++;
432			}
433			if (curr == EXT_MARK &&
434					(udfLen - index - 1) <= EXT_SIZE) {
435				if (udfLen == index + 1)
436					hasExt = 0;
437				else {
438					hasExt = 1;
439					extIndex = index;
440					newExtIndex = newIndex;
441				}
442			}
443			if (newIndex < 256)
444				newName[newIndex++] = curr;
445			else
446				needsCRC = 1;
447		}
448	}
449	if (needsCRC) {
450		uint8_t ext[EXT_SIZE];
451		int localExtIndex = 0;
452
453		if (hasExt) {
454			int maxFilenameLen;
455			for (index = 0;
456			     index < EXT_SIZE && extIndex + index + 1 < udfLen;
457			     index++) {
458				curr = udfName[extIndex + index + 1];
459
460				if (curr == '/' || curr == 0) {
461					needsCRC = 1;
462					curr = ILLEGAL_CHAR_MARK;
463					while (extIndex + index + 2 < udfLen &&
464					      (index + 1 < EXT_SIZE &&
465						(udfName[extIndex + index + 2] == '/' ||
466						 udfName[extIndex + index + 2] == 0)))
467						index++;
468				}
469				ext[localExtIndex++] = curr;
470			}
471			maxFilenameLen = 250 - localExtIndex;
472			if (newIndex > maxFilenameLen)
473				newIndex = maxFilenameLen;
474			else
475				newIndex = newExtIndex;
476		} else if (newIndex > 250)
477			newIndex = 250;
478		newName[newIndex++] = CRC_MARK;
479		valueCRC = crc_itu_t(0, fidName, fidNameLen);
480		newName[newIndex++] = hexChar[(valueCRC & 0xf000) >> 12];
481		newName[newIndex++] = hexChar[(valueCRC & 0x0f00) >> 8];
482		newName[newIndex++] = hexChar[(valueCRC & 0x00f0) >> 4];
483		newName[newIndex++] = hexChar[(valueCRC & 0x000f)];
484
485		if (hasExt) {
486			newName[newIndex++] = EXT_MARK;
487			for (index = 0; index < localExtIndex; index++)
488				newName[newIndex++] = ext[index];
489		}
490	}
491
492	return newIndex;
493}