diff options
author | Alan Stern <stern@rowland.harvard.edu> | 2009-04-30 10:08:18 -0400 |
---|---|---|
committer | Greg Kroah-Hartman <gregkh@suse.de> | 2009-06-16 00:44:43 -0400 |
commit | 74675a58507e769beee7d949dbed788af3c4139d (patch) | |
tree | d4ae3cc06dbfadecf1eaf6ed0aef249fc87b07e6 /fs/nls/nls_base.c | |
parent | a853a3d4eb2edb066248a39f0634f6f5858816a0 (diff) |
NLS: update handling of Unicode
This patch (as1239) updates the kernel's treatment of Unicode. The
character-set conversion routines are well behind the current state of
the Unicode specification: They don't recognize the existence of code
points beyond plane 0 or of surrogate pairs in the UTF-16 encoding.
The old wchar_t 16-bit type is retained because it's still used in
lots of places. This shouldn't cause any new problems; if a
conversion now results in an invalid 16-bit code then before it must
have yielded an undefined code.
Difficult-to-read names like "utf_mbstowcs" are replaced with more
transparent names like "utf8s_to_utf16s" and the ordering of the
parameters is rationalized (buffer lengths come immediate after the
pointers they refer to, and the inputs precede the outputs).
Fortunately the low-level conversion routines are used in only a few
places; the interfaces to the higher-level uni2char and char2uni
methods have been left unchanged.
Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Acked-by: Clemens Ladisch <clemens@ladisch.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Diffstat (limited to 'fs/nls/nls_base.c')
-rw-r--r-- | fs/nls/nls_base.c | 164 |
1 files changed, 106 insertions, 58 deletions
diff --git a/fs/nls/nls_base.c b/fs/nls/nls_base.c index 750abf211e26..477d37d83b31 100644 --- a/fs/nls/nls_base.c +++ b/fs/nls/nls_base.c | |||
@@ -15,6 +15,7 @@ | |||
15 | #include <linux/errno.h> | 15 | #include <linux/errno.h> |
16 | #include <linux/kmod.h> | 16 | #include <linux/kmod.h> |
17 | #include <linux/spinlock.h> | 17 | #include <linux/spinlock.h> |
18 | #include <asm/byteorder.h> | ||
18 | 19 | ||
19 | static struct nls_table default_table; | 20 | static struct nls_table default_table; |
20 | static struct nls_table *tables = &default_table; | 21 | static struct nls_table *tables = &default_table; |
@@ -43,10 +44,17 @@ static const struct utf8_table utf8_table[] = | |||
43 | {0, /* end of table */} | 44 | {0, /* end of table */} |
44 | }; | 45 | }; |
45 | 46 | ||
46 | int | 47 | #define UNICODE_MAX 0x0010ffff |
47 | utf8_mbtowc(wchar_t *p, const __u8 *s, int n) | 48 | #define PLANE_SIZE 0x00010000 |
49 | |||
50 | #define SURROGATE_MASK 0xfffff800 | ||
51 | #define SURROGATE_PAIR 0x0000d800 | ||
52 | #define SURROGATE_LOW 0x00000400 | ||
53 | #define SURROGATE_BITS 0x000003ff | ||
54 | |||
55 | int utf8_to_utf32(const u8 *s, int len, unicode_t *pu) | ||
48 | { | 56 | { |
49 | long l; | 57 | unsigned long l; |
50 | int c0, c, nc; | 58 | int c0, c, nc; |
51 | const struct utf8_table *t; | 59 | const struct utf8_table *t; |
52 | 60 | ||
@@ -57,12 +65,13 @@ utf8_mbtowc(wchar_t *p, const __u8 *s, int n) | |||
57 | nc++; | 65 | nc++; |
58 | if ((c0 & t->cmask) == t->cval) { | 66 | if ((c0 & t->cmask) == t->cval) { |
59 | l &= t->lmask; | 67 | l &= t->lmask; |
60 | if (l < t->lval) | 68 | if (l < t->lval || l > UNICODE_MAX || |
69 | (l & SURROGATE_MASK) == SURROGATE_PAIR) | ||
61 | return -1; | 70 | return -1; |
62 | *p = l; | 71 | *pu = (unicode_t) l; |
63 | return nc; | 72 | return nc; |
64 | } | 73 | } |
65 | if (n <= nc) | 74 | if (len <= nc) |
66 | return -1; | 75 | return -1; |
67 | s++; | 76 | s++; |
68 | c = (*s ^ 0x80) & 0xFF; | 77 | c = (*s ^ 0x80) & 0xFF; |
@@ -72,76 +81,119 @@ utf8_mbtowc(wchar_t *p, const __u8 *s, int n) | |||
72 | } | 81 | } |
73 | return -1; | 82 | return -1; |
74 | } | 83 | } |
84 | EXPORT_SYMBOL(utf8_to_utf32); | ||
75 | 85 | ||
76 | int | 86 | int utf32_to_utf8(unicode_t u, u8 *s, int maxlen) |
77 | utf8_mbstowcs(wchar_t *pwcs, const __u8 *s, int n) | ||
78 | { | 87 | { |
79 | __u16 *op; | 88 | unsigned long l; |
80 | const __u8 *ip; | ||
81 | int size; | ||
82 | |||
83 | op = pwcs; | ||
84 | ip = s; | ||
85 | while (*ip && n > 0) { | ||
86 | if (*ip & 0x80) { | ||
87 | size = utf8_mbtowc(op, ip, n); | ||
88 | if (size == -1) { | ||
89 | /* Ignore character and move on */ | ||
90 | ip++; | ||
91 | n--; | ||
92 | } else { | ||
93 | op++; | ||
94 | ip += size; | ||
95 | n -= size; | ||
96 | } | ||
97 | } else { | ||
98 | *op++ = *ip++; | ||
99 | n--; | ||
100 | } | ||
101 | } | ||
102 | return (op - pwcs); | ||
103 | } | ||
104 | |||
105 | int | ||
106 | utf8_wctomb(__u8 *s, wchar_t wc, int maxlen) | ||
107 | { | ||
108 | long l; | ||
109 | int c, nc; | 89 | int c, nc; |
110 | const struct utf8_table *t; | 90 | const struct utf8_table *t; |
111 | 91 | ||
112 | if (!s) | 92 | if (!s) |
113 | return 0; | 93 | return 0; |
114 | 94 | ||
115 | l = wc; | 95 | l = u; |
96 | if (l > UNICODE_MAX || (l & SURROGATE_MASK) == SURROGATE_PAIR) | ||
97 | return -1; | ||
98 | |||
116 | nc = 0; | 99 | nc = 0; |
117 | for (t = utf8_table; t->cmask && maxlen; t++, maxlen--) { | 100 | for (t = utf8_table; t->cmask && maxlen; t++, maxlen--) { |
118 | nc++; | 101 | nc++; |
119 | if (l <= t->lmask) { | 102 | if (l <= t->lmask) { |
120 | c = t->shift; | 103 | c = t->shift; |
121 | *s = t->cval | (l >> c); | 104 | *s = (u8) (t->cval | (l >> c)); |
122 | while (c > 0) { | 105 | while (c > 0) { |
123 | c -= 6; | 106 | c -= 6; |
124 | s++; | 107 | s++; |
125 | *s = 0x80 | ((l >> c) & 0x3F); | 108 | *s = (u8) (0x80 | ((l >> c) & 0x3F)); |
126 | } | 109 | } |
127 | return nc; | 110 | return nc; |
128 | } | 111 | } |
129 | } | 112 | } |
130 | return -1; | 113 | return -1; |
131 | } | 114 | } |
115 | EXPORT_SYMBOL(utf32_to_utf8); | ||
132 | 116 | ||
133 | int | 117 | int utf8s_to_utf16s(const u8 *s, int len, wchar_t *pwcs) |
134 | utf8_wcstombs(__u8 *s, const wchar_t *pwcs, int maxlen) | ||
135 | { | 118 | { |
136 | const __u16 *ip; | 119 | u16 *op; |
137 | __u8 *op; | ||
138 | int size; | 120 | int size; |
121 | unicode_t u; | ||
122 | |||
123 | op = pwcs; | ||
124 | while (*s && len > 0) { | ||
125 | if (*s & 0x80) { | ||
126 | size = utf8_to_utf32(s, len, &u); | ||
127 | if (size < 0) { | ||
128 | /* Ignore character and move on */ | ||
129 | size = 1; | ||
130 | } else if (u >= PLANE_SIZE) { | ||
131 | u -= PLANE_SIZE; | ||
132 | *op++ = (wchar_t) (SURROGATE_PAIR | | ||
133 | ((u >> 10) & SURROGATE_BITS)); | ||
134 | *op++ = (wchar_t) (SURROGATE_PAIR | | ||
135 | SURROGATE_LOW | | ||
136 | (u & SURROGATE_BITS)); | ||
137 | } else { | ||
138 | *op++ = (wchar_t) u; | ||
139 | } | ||
140 | s += size; | ||
141 | len -= size; | ||
142 | } else { | ||
143 | *op++ = *s++; | ||
144 | len--; | ||
145 | } | ||
146 | } | ||
147 | return op - pwcs; | ||
148 | } | ||
149 | EXPORT_SYMBOL(utf8s_to_utf16s); | ||
150 | |||
151 | static inline unsigned long get_utf16(unsigned c, enum utf16_endian endian) | ||
152 | { | ||
153 | switch (endian) { | ||
154 | default: | ||
155 | return c; | ||
156 | case UTF16_LITTLE_ENDIAN: | ||
157 | return __le16_to_cpu(c); | ||
158 | case UTF16_BIG_ENDIAN: | ||
159 | return __be16_to_cpu(c); | ||
160 | } | ||
161 | } | ||
162 | |||
163 | int utf16s_to_utf8s(const wchar_t *pwcs, int len, enum utf16_endian endian, | ||
164 | u8 *s, int maxlen) | ||
165 | { | ||
166 | u8 *op; | ||
167 | int size; | ||
168 | unsigned long u, v; | ||
139 | 169 | ||
140 | op = s; | 170 | op = s; |
141 | ip = pwcs; | 171 | while (len > 0 && maxlen > 0) { |
142 | while (*ip && maxlen > 0) { | 172 | u = get_utf16(*pwcs, endian); |
143 | if (*ip > 0x7f) { | 173 | if (!u) |
144 | size = utf8_wctomb(op, *ip, maxlen); | 174 | break; |
175 | pwcs++; | ||
176 | len--; | ||
177 | if (u > 0x7f) { | ||
178 | if ((u & SURROGATE_MASK) == SURROGATE_PAIR) { | ||
179 | if (u & SURROGATE_LOW) { | ||
180 | /* Ignore character and move on */ | ||
181 | continue; | ||
182 | } | ||
183 | if (len <= 0) | ||
184 | break; | ||
185 | v = get_utf16(*pwcs, endian); | ||
186 | if ((v & SURROGATE_MASK) != SURROGATE_PAIR || | ||
187 | !(v & SURROGATE_LOW)) { | ||
188 | /* Ignore character and move on */ | ||
189 | continue; | ||
190 | } | ||
191 | u = PLANE_SIZE + ((u & SURROGATE_BITS) << 10) | ||
192 | + (v & SURROGATE_BITS); | ||
193 | pwcs++; | ||
194 | len--; | ||
195 | } | ||
196 | size = utf32_to_utf8(u, op, maxlen); | ||
145 | if (size == -1) { | 197 | if (size == -1) { |
146 | /* Ignore character and move on */ | 198 | /* Ignore character and move on */ |
147 | } else { | 199 | } else { |
@@ -149,13 +201,13 @@ utf8_wcstombs(__u8 *s, const wchar_t *pwcs, int maxlen) | |||
149 | maxlen -= size; | 201 | maxlen -= size; |
150 | } | 202 | } |
151 | } else { | 203 | } else { |
152 | *op++ = (__u8) *ip; | 204 | *op++ = (u8) u; |
153 | maxlen--; | 205 | maxlen--; |
154 | } | 206 | } |
155 | ip++; | ||
156 | } | 207 | } |
157 | return (op - s); | 208 | return op - s; |
158 | } | 209 | } |
210 | EXPORT_SYMBOL(utf16s_to_utf8s); | ||
159 | 211 | ||
160 | int register_nls(struct nls_table * nls) | 212 | int register_nls(struct nls_table * nls) |
161 | { | 213 | { |
@@ -467,9 +519,5 @@ EXPORT_SYMBOL(unregister_nls); | |||
467 | EXPORT_SYMBOL(unload_nls); | 519 | EXPORT_SYMBOL(unload_nls); |
468 | EXPORT_SYMBOL(load_nls); | 520 | EXPORT_SYMBOL(load_nls); |
469 | EXPORT_SYMBOL(load_nls_default); | 521 | EXPORT_SYMBOL(load_nls_default); |
470 | EXPORT_SYMBOL(utf8_mbtowc); | ||
471 | EXPORT_SYMBOL(utf8_mbstowcs); | ||
472 | EXPORT_SYMBOL(utf8_wctomb); | ||
473 | EXPORT_SYMBOL(utf8_wcstombs); | ||
474 | 522 | ||
475 | MODULE_LICENSE("Dual BSD/GPL"); | 523 | MODULE_LICENSE("Dual BSD/GPL"); |