2 * Copyright (C) 2003 David Brownell
4 * This program is free software; you can redistribute it and/or modify
5 * it under the terms of the GNU Lesser General Public License as published
6 * by the Free Software Foundation; either version 2.1 of the License, or
7 * (at your option) any later version.
10 #include <linux/errno.h>
11 #include <linux/kernel.h>
12 #include <linux/list.h>
13 #include <linux/string.h>
14 #include <linux/device.h>
15 #include <linux/init.h>
17 #include <linux/usb/ch9.h>
18 #include <linux/usb/gadget.h>
20 #include <asm/unaligned.h>
23 static int utf8_to_utf16le(const char *s, __le16 *cp, unsigned len)
29 /* this insists on correct encodings, though not minimal ones.
30 * BUT it currently rejects legit 4-byte UTF-8 code points,
31 * which need surrogate pairs. (Unicode 3.1 can use them.)
33 while (len != 0 && (c = (u8) *s++) != 0) {
34 if (unlikely(c & 0x80)) {
36 // 00000yyyyyxxxxxx = 110yyyyy 10xxxxxx
37 if ((c & 0xe0) == 0xc0) {
38 uchar = (c & 0x1f) << 6;
41 if ((c & 0xc0) != 0xc0)
46 // 3-byte sequence (most CJKV characters):
47 // zzzzyyyyyyxxxxxx = 1110zzzz 10yyyyyy 10xxxxxx
48 } else if ((c & 0xf0) == 0xe0) {
49 uchar = (c & 0x0f) << 12;
52 if ((c & 0xc0) != 0xc0)
58 if ((c & 0xc0) != 0xc0)
63 /* no bogus surrogates */
64 if (0xd800 <= uchar && uchar <= 0xdfff)
67 // 4-byte sequence (surrogate pairs, currently rare):
68 // 11101110wwwwzzzzyy + 110111yyyyxxxxxx
69 // = 11110uuu 10uuzzzz 10yyyyyy 10xxxxxx
71 // FIXME accept the surrogate code points (only)
77 put_unaligned (cpu_to_le16 (uchar), cp++);
88 * usb_gadget_get_string - fill out a string descriptor
89 * @table: of c strings encoded using UTF-8
90 * @id: string id, from low byte of wValue in get string descriptor
91 * @buf: at least 256 bytes
93 * Finds the UTF-8 string matching the ID, and converts it into a
94 * string descriptor in utf16-le.
95 * Returns length of descriptor (always even) or negative errno
97 * If your driver needs stings in multiple languages, you'll probably
98 * "switch (wIndex) { ... }" in your ep0 string descriptor logic,
99 * using this routine after choosing which set of UTF-8 strings to use.
100 * Note that US-ASCII is a strict subset of UTF-8; any string bytes with
101 * the eighth bit set will be multibyte UTF-8 characters, not ISO-8859/1
102 * characters (which are also widely used in C strings).
105 usb_gadget_get_string (struct usb_gadget_strings *table, int id, u8 *buf)
107 struct usb_string *s;
110 /* descriptor 0 has the language id */
113 buf [1] = USB_DT_STRING;
114 buf [2] = (u8) table->language;
115 buf [3] = (u8) (table->language >> 8);
118 for (s = table->strings; s && s->s; s++)
122 /* unrecognized: stall. */
126 /* string descriptors have length, tag, then UTF16-LE text */
127 len = min ((size_t) 126, strlen (s->s));
128 memset (buf + 2, 0, 2 * len); /* zero all the bytes */
129 len = utf8_to_utf16le(s->s, (__le16 *)&buf[2], len);
132 buf [0] = (len + 1) * 2;
133 buf [1] = USB_DT_STRING;