1
/* Copyright 2000-2005 The Apache Software Foundation or its licensors, as
4
* Licensed under the Apache License, Version 2.0 (the "License");
5
* you may not use this file except in compliance with the License.
6
* You may obtain a copy of the License at
8
* http://www.apache.org/licenses/LICENSE-2.0
10
* Unless required by applicable law or agreed to in writing, software
11
* distributed under the License is distributed on an "AS IS" BASIS,
12
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
13
* See the License for the specific language governing permissions and
14
* limitations under the License.
18
#include "apr_private.h"
19
#include "apr_errno.h"
20
#include "apr_arch_utf8.h"
22
/* Implement the design principal specified by RFC 2718 2.2.5
23
* Guidelines for new URL Schemes - within the APR.
25
* Since many architectures support unicode, and UCS2 is the most
26
* efficient storage used by those archictures, these functions
27
* exist to validate a UCS string. It is up to the operating system
28
* to determine the validitity of the string in the context of it's
29
* native language support. File systems that support filename
30
* characters of 0x80-0xff but have no support of Unicode will find
31
* this function useful only for validating the character sequences
32
* and rejecting poorly encoded strings, if RFC 2718 2.2.5 naming is
35
* from RFC 2279 UTF-8, a transformation format of ISO 10646
37
* UCS-4 range (hex.) UTF-8 octet sequence (binary)
38
* 1:2 0000 0000-0000 007F 0xxxxxxx
39
* 2:2 0000 0080-0000 07FF 110XXXXx 10xxxxxx
40
* 3:2 0000 0800-0000 FFFF 1110XXXX 10Xxxxxx 10xxxxxx
41
* 4:4 0001 0000-001F FFFF 11110zXX 10XXxxxx 10xxxxxx 10xxxxxx
42
* inv 0020 0000-03FF FFFF 111110XX 10XXXxxx 10xxxxxx 10xxxxxx 10xxxxxx
43
* inv 0400 0000-7FFF FFFF 1111110X 10XXXXxx 10xxxxxx 10xxxxxx 10xxxxxx 10xxxxxx
45
* One of the X values must be one for the encoding length to be legit.
46
* Neither the z bit, nor the final two forms, are used for ucs-2
48
* "Pairs of UCS-2 values between D800 and DFFF (surrogate pairs in
49
* Unicode parlance), being actually UCS-4 characters transformed
50
* through UTF-16, need special treatment: the UTF-16 transformation
51
* must be undone, yielding a UCS-4 character that is then transformed
54
* from RFC2781 UTF-16: the compressed ISO 10646 encoding bitmask
57
* U' = 000000000000yyyyyyyyyyxxxxxxxxxx
58
* W1 = 110110yyyyyyyyyy
59
* W2 = 110111xxxxxxxxxx
61
* apr_conv_utf8_to_ucs2 out bytes:sizeof(in) * 1 <= Req <= sizeof(in) * 2
63
* apr_conv_ucs2_to_utf8 out words:sizeof(in) / 2 <= Req <= sizeof(in) * 3 / 2
66
APR_DECLARE(apr_status_t) apr_conv_utf8_to_ucs2(const char *in,
71
apr_int64_t newch, mask;
72
apr_size_t expect, eating;
75
while (*inbytes && *outwords)
77
ch = (unsigned char)(*in++);
79
/* US-ASCII-7 plain text
87
if ((ch & 0300) != 0300) {
88
/* Multibyte Continuation is out of place
94
/* Multibyte Sequence Lead Character
96
* Compute the expected bytes while adjusting
97
* or lead byte and leading zeros mask.
101
while ((ch & mask) == mask) {
103
if (++expect > 3) /* (truly 5 for ucs-4) */
108
if (*inbytes <= expect)
109
return APR_INCOMPLETE;
110
/* Reject values of excessive leading 0 bits
111
* utf-8 _demands_ the shortest possible byte length
118
/* Reject values of excessive leading 0 bits
120
if (!newch && !((unsigned char)*in & 0077 & (mask << 1)))
123
/* Reject values D800-DFFF when not utf16 encoded
124
* (may not be an appropriate restriction for ucs-4)
126
if (newch == 0015 && ((unsigned char)*in & 0040))
129
else if (expect == 3) {
130
/* Short circuit values > 110000
134
if (newch == 4 && ((unsigned char)*in & 0060))
138
/* Where the boolean (expect > 2) is true, we will need
139
* an extra word for the output.
141
if (*outwords < (apr_size_t)(expect > 2) + 1)
142
break; /* buffer full */
145
/* Multibyte Continuation must be legal */
146
if (((ch = (unsigned char)*(in++)) & 0300) != 0200)
149
newch |= (ch & 0077);
152
/* newch is now a true ucs-4 character
154
* now we need to fold to ucs-2
159
*(out++) = (apr_wchar_t) newch;
165
*(out++) = (apr_wchar_t) (0xD800 | (newch >> 10));
166
*(out++) = (apr_wchar_t) (0xDC00 | (newch & 0x03FF));
171
/* Buffer full 'errors' aren't errors, the client must inspect both
172
* the inbytes and outwords values
177
APR_DECLARE(apr_status_t) apr_conv_ucs2_to_utf8(const apr_wchar_t *in,
180
apr_size_t *outbytes)
182
apr_int64_t newch, require;
187
while (*inwords && *outbytes)
189
ch = (unsigned short)(*in++);
194
*(out++) = (unsigned char) ch;
198
if ((ch & 0xFC00) == 0xDC00) {
199
/* Invalid Leading ucs-2 Multiword Continuation Character
203
if ((ch & 0xFC00) == 0xD800) {
204
/* Leading ucs-2 Multiword Character
207
/* Missing ucs-2 Multiword Continuation Character
209
return APR_INCOMPLETE;
211
if (((unsigned short)(*in) & 0xFC00) != 0xDC00) {
212
/* Invalid ucs-2 Multiword Continuation Character
216
newch = (ch & 0x03FF) << 10 | ((unsigned short)(*in++) & 0x03FF);
220
/* ucs-2 Single Word Character
224
/* Determine the absolute minimum utf-8 bytes required
226
require = newch >> 11;
229
require >>= 5, ++need;
230
if (need >= *outbytes)
231
break; /* Insufficient buffer */
232
*inwords -= (need > 2) + 1;
233
*outbytes -= need + 1;
234
/* Compute the utf-8 characters in last to first order,
235
* calculating the lead character length bits along the way.
242
*(--invout) = (unsigned char)(0200 | (newch & 0077));
245
/* Compute the lead utf-8 character and move the dest offset
247
*(--invout) = (unsigned char)(ch | newch);
250
/* Buffer full 'errors' aren't errors, the client must inspect both
251
* the inwords and outbytes values