63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
|
if (entity[i] >= '0' && entity[i] <= '9')
c = (c * 10) + (entity[i] - '0');
else
return nil;
}
}
if ((i = of_string_unicode_to_utf8(c, buf)) == 0)
return nil;
buf[i] = 0;
return [OFString stringWithCString: buf
length: i];
}
@implementation OFString (XMLUnescaping)
- (OFString*)stringByXMLUnescaping
{
return [self stringByXMLUnescapingWithDelegate: nil];
}
- (OFString*)stringByXMLUnescapingWithDelegate:
(id <OFStringXMLUnescapingDelegate>)delegate
{
size_t i, last;
BOOL in_entity;
OFMutableString *ret;
last = 0;
in_entity = NO;
ret = [OFMutableString string];
((OFString*)ret)->isUTF8 = [self isUTF8];
for (i = 0; i < length; i++) {
if (!in_entity && string[i] == '&') {
[ret appendCStringWithoutUTF8Checking: string + last
length: i - last];
last = i + 1;
in_entity = YES;
} else if (in_entity && string[i] == ';') {
char *entity = string + last;
size_t len = i - last;
if (len == 2 && !memcmp(entity, "lt", 2))
[ret appendCStringWithoutUTF8Checking: "<"
length: 1];
else if (len == 2 && !memcmp(entity, "gt", 2))
[ret appendCStringWithoutUTF8Checking: ">"
length: 1];
else if (len == 4 && !memcmp(entity, "quot", 4))
[ret appendCStringWithoutUTF8Checking: "\""
length: 1];
else if (len == 4 && !memcmp(entity, "apos", 4))
[ret appendCStringWithoutUTF8Checking: "'"
length: 1];
else if (len == 3 && !memcmp(entity, "amp", 3))
[ret appendCStringWithoutUTF8Checking: "&"
length: 1];
else if (entity[0] == '#') {
OFAutoreleasePool *pool;
OFString *tmp;
pool = [[OFAutoreleasePool alloc] init];
tmp = parse_numeric_entity(entity, len);
if (tmp == nil)
@throw [OFInvalidEncodingException
newWithClass: isa];
[ret appendString: tmp];
[pool release];
} else if (delegate != nil) {
OFAutoreleasePool *pool;
OFString *n, *tmp;
pool = [[OFAutoreleasePool alloc] init];
n = [OFString stringWithCString: entity
length: len];
tmp = [delegate string: self
containsUnknownEntityNamed: n];
if (tmp == nil)
@throw [OFInvalidEncodingException
newWithClass: isa];
[ret appendString: tmp];
[pool release];
} else
@throw [OFInvalidEncodingException
newWithClass: isa];
last = i + 1;
in_entity = NO;
}
}
if (in_entity)
@throw [OFInvalidEncodingException newWithClass: isa];
[ret appendCStringWithoutUTF8Checking: string + last
length: i - last];
/*
* Class swizzle the string to be immutable. We declared the return type
* to be OFString*, so it can't be modified anyway. But not swizzling it
* would create a real copy each time -[copy] is called.
*/
ret->isa = [OFString class];
return ret;
}
#ifdef OF_HAVE_BLOCKS
- (OFString*)stringByXMLUnescapingWithBlock:
(of_string_xml_unescaping_block_t)block
{
size_t i, last;
BOOL in_entity;
OFMutableString *ret;
last = 0;
in_entity = NO;
ret = [OFMutableString string];
((OFString*)ret)->isUTF8 = [self isUTF8];
for (i = 0; i < length; i++) {
if (!in_entity && string[i] == '&') {
[ret appendCStringWithoutUTF8Checking: string + last
length: i - last];
last = i + 1;
in_entity = YES;
} else if (in_entity && string[i] == ';') {
char *entity = string + last;
size_t len = i - last;
if (len == 2 && !memcmp(entity, "lt", 2))
[ret appendCStringWithoutUTF8Checking: "<"
length: 1];
else if (len == 2 && !memcmp(entity, "gt", 2))
[ret appendCStringWithoutUTF8Checking: ">"
length: 1];
else if (len == 4 && !memcmp(entity, "quot", 4))
[ret appendCStringWithoutUTF8Checking: "\""
length: 1];
else if (len == 4 && !memcmp(entity, "apos", 4))
[ret appendCStringWithoutUTF8Checking: "'"
length: 1];
else if (len == 3 && !memcmp(entity, "amp", 3))
[ret appendCStringWithoutUTF8Checking: "&"
length: 1];
else if (entity[0] == '#') {
OFAutoreleasePool *pool;
OFString *tmp;
pool = [[OFAutoreleasePool alloc] init];
tmp = parse_numeric_entity(entity, len);
if (tmp == nil)
@throw [OFInvalidEncodingException
newWithClass: isa];
[ret appendString: tmp];
[pool release];
} else {
OFAutoreleasePool *pool;
OFString *n, *tmp;
pool = [[OFAutoreleasePool alloc] init];
n = [OFString stringWithCString: entity
length: len];
tmp = block(self, n);
if (tmp == nil)
@throw [OFInvalidEncodingException
newWithClass: isa];
[ret appendString: tmp];
[pool release];
}
last = i + 1;
in_entity = NO;
}
}
if (in_entity)
@throw [OFInvalidEncodingException newWithClass: isa];
[ret appendCStringWithoutUTF8Checking: string + last
length: i - last];
/*
* Class swizzle the string to be immutable. We declared the return type
|
|
|
|
|
|
|
|
|
|
|
|
>
|
>
|
|
|
>
|
|
|
|
|
|
|
|
|
|
|
>
|
>
|
|
|
>
|
>
|
|
|
|
|
|
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
|
if (entity[i] >= '0' && entity[i] <= '9')
c = (c * 10) + (entity[i] - '0');
else
return nil;
}
}
if ((i = of_string_unicode_to_utf8(c, buffer)) == 0)
return nil;
buffer[i] = 0;
return [OFString stringWithCString: buffer
length: i];
}
@implementation OFString (XMLUnescaping)
- (OFString*)stringByXMLUnescaping
{
return [self stringByXMLUnescapingWithDelegate: nil];
}
- (OFString*)stringByXMLUnescapingWithDelegate:
(id <OFStringXMLUnescapingDelegate>)delegate
{
size_t i, last;
BOOL inEntity;
OFMutableString *ret;
last = 0;
inEntity = NO;
ret = [OFMutableString string];
((OFString*)ret)->isUTF8 = [self isUTF8];
for (i = 0; i < length; i++) {
if (!inEntity && string[i] == '&') {
[ret appendCStringWithoutUTF8Checking: string + last
length: i - last];
last = i + 1;
inEntity = YES;
} else if (inEntity && string[i] == ';') {
char *entity = string + last;
size_t entityLength = i - last;
if (entityLength == 2 && !memcmp(entity, "lt", 2))
[ret appendCStringWithoutUTF8Checking: "<"
length: 1];
else if (entityLength == 2 && !memcmp(entity, "gt", 2))
[ret appendCStringWithoutUTF8Checking: ">"
length: 1];
else if (entityLength == 4 &&
!memcmp(entity, "quot", 4))
[ret appendCStringWithoutUTF8Checking: "\""
length: 1];
else if (entityLength == 4 &&
!memcmp(entity, "apos", 4))
[ret appendCStringWithoutUTF8Checking: "'"
length: 1];
else if (entityLength == 3 && !memcmp(entity, "amp", 3))
[ret appendCStringWithoutUTF8Checking: "&"
length: 1];
else if (entity[0] == '#') {
OFAutoreleasePool *pool;
OFString *tmp;
pool = [[OFAutoreleasePool alloc] init];
tmp = parse_numeric_entity(entity,
entityLength);
if (tmp == nil)
@throw [OFInvalidEncodingException
newWithClass: isa];
[ret appendString: tmp];
[pool release];
} else if (delegate != nil) {
OFAutoreleasePool *pool;
OFString *n, *tmp;
pool = [[OFAutoreleasePool alloc] init];
n = [OFString stringWithCString: entity
length: entityLength];
tmp = [delegate string: self
containsUnknownEntityNamed: n];
if (tmp == nil)
@throw [OFInvalidEncodingException
newWithClass: isa];
[ret appendString: tmp];
[pool release];
} else
@throw [OFInvalidEncodingException
newWithClass: isa];
last = i + 1;
inEntity = NO;
}
}
if (inEntity)
@throw [OFInvalidEncodingException newWithClass: isa];
[ret appendCStringWithoutUTF8Checking: string + last
length: i - last];
/*
* Class swizzle the string to be immutable. We declared the return type
* to be OFString*, so it can't be modified anyway. But not swizzling it
* would create a real copy each time -[copy] is called.
*/
ret->isa = [OFString class];
return ret;
}
#ifdef OF_HAVE_BLOCKS
- (OFString*)stringByXMLUnescapingWithBlock:
(of_string_xml_unescaping_block_t)block
{
size_t i, last;
BOOL inEntity;
OFMutableString *ret;
last = 0;
inEntity = NO;
ret = [OFMutableString string];
((OFString*)ret)->isUTF8 = [self isUTF8];
for (i = 0; i < length; i++) {
if (!inEntity && string[i] == '&') {
[ret appendCStringWithoutUTF8Checking: string + last
length: i - last];
last = i + 1;
inEntity = YES;
} else if (inEntity && string[i] == ';') {
char *entity = string + last;
size_t entityLength = i - last;
if (entityLength == 2 && !memcmp(entity, "lt", 2))
[ret appendCStringWithoutUTF8Checking: "<"
length: 1];
else if (entityLength == 2 && !memcmp(entity, "gt", 2))
[ret appendCStringWithoutUTF8Checking: ">"
length: 1];
else if (entityLength == 4 &&
!memcmp(entity, "quot", 4))
[ret appendCStringWithoutUTF8Checking: "\""
length: 1];
else if (entityLength == 4 &&
!memcmp(entity, "apos", 4))
[ret appendCStringWithoutUTF8Checking: "'"
length: 1];
else if (entityLength == 3 && !memcmp(entity, "amp", 3))
[ret appendCStringWithoutUTF8Checking: "&"
length: 1];
else if (entity[0] == '#') {
OFAutoreleasePool *pool;
OFString *tmp;
pool = [[OFAutoreleasePool alloc] init];
tmp = parse_numeric_entity(entity,
entityLength);
if (tmp == nil)
@throw [OFInvalidEncodingException
newWithClass: isa];
[ret appendString: tmp];
[pool release];
} else {
OFAutoreleasePool *pool;
OFString *entityString, *tmp;
pool = [[OFAutoreleasePool alloc] init];
entityString = [OFString
stringWithCString: entity
length: entityLength];
tmp = block(self, entityString);
if (tmp == nil)
@throw [OFInvalidEncodingException
newWithClass: isa];
[ret appendString: tmp];
[pool release];
}
last = i + 1;
inEntity = NO;
}
}
if (inEntity)
@throw [OFInvalidEncodingException newWithClass: isa];
[ret appendCStringWithoutUTF8Checking: string + last
length: i - last];
/*
* Class swizzle the string to be immutable. We declared the return type
|