1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
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
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
|
// Copyright (c) 2012 The Chromium Authors. All rights reserved.
// Use of this source code is governed by a BSD-style license that can be
// found in the LICENSE file.
#include "webkit/glue/multipart_response_delegate.h"
#include "base/logging.h"
#include "base/string_number_conversions.h"
#include "base/string_util.h"
#include "net/base/net_util.h"
#include "net/http/http_util.h"
#include "third_party/WebKit/Source/Platform/chromium/public/WebHTTPHeaderVisitor.h"
#include "third_party/WebKit/Source/Platform/chromium/public/WebString.h"
#include "third_party/WebKit/Source/Platform/chromium/public/WebURL.h"
#include "third_party/WebKit/Source/Platform/chromium/public/WebURLLoaderClient.h"
using WebKit::WebHTTPHeaderVisitor;
using WebKit::WebString;
using WebKit::WebURLLoader;
using WebKit::WebURLLoaderClient;
using WebKit::WebURLResponse;
namespace webkit_glue {
namespace {
// The list of response headers that we do not copy from the original
// response when generating a WebURLResponse for a MIME payload.
const char* kReplaceHeaders[] = {
"content-type",
"content-length",
"content-disposition",
"content-range",
"range",
"set-cookie"
};
class HeaderCopier : public WebHTTPHeaderVisitor {
public:
HeaderCopier(WebURLResponse* response)
: response_(response) {
}
virtual void visitHeader(const WebString& name, const WebString& value) {
const std::string& name_utf8 = name.utf8();
for (size_t i = 0; i < arraysize(kReplaceHeaders); ++i) {
if (LowerCaseEqualsASCII(name_utf8, kReplaceHeaders[i]))
return;
}
response_->setHTTPHeaderField(name, value);
}
private:
WebURLResponse* response_;
};
} // namespace
MultipartResponseDelegate::MultipartResponseDelegate(
WebURLLoaderClient* client,
WebURLLoader* loader,
const WebURLResponse& response,
const std::string& boundary)
: client_(client),
loader_(loader),
original_response_(response),
encoded_data_length_(0),
boundary_("--"),
first_received_data_(true),
processing_headers_(false),
stop_sending_(false),
has_sent_first_response_(false) {
// Some servers report a boundary prefixed with "--". See bug 5786.
if (StartsWithASCII(boundary, "--", true)) {
boundary_.assign(boundary);
} else {
boundary_.append(boundary);
}
}
void MultipartResponseDelegate::OnReceivedData(const char* data,
int data_len,
int encoded_data_length) {
// stop_sending_ means that we've already received the final boundary token.
// The server should stop sending us data at this point, but if it does, we
// just throw it away.
if (stop_sending_)
return;
data_.append(data, data_len);
encoded_data_length_ += encoded_data_length;
if (first_received_data_) {
// Some servers don't send a boundary token before the first chunk of
// data. We handle this case anyway (Gecko does too).
first_received_data_ = false;
// Eat leading \r\n
int pos = PushOverLine(data_, 0);
if (pos)
data_ = data_.substr(pos);
if (data_.length() < boundary_.length() + 2) {
// We don't have enough data yet to make a boundary token. Just wait
// until the next chunk of data arrives.
first_received_data_ = true;
return;
}
if (0 != data_.compare(0, boundary_.length(), boundary_)) {
data_ = boundary_ + "\n" + data_;
}
}
DCHECK(!first_received_data_);
// Headers
if (processing_headers_) {
// Eat leading \r\n
int pos = PushOverLine(data_, 0);
if (pos)
data_ = data_.substr(pos);
if (ParseHeaders()) {
// Successfully parsed headers.
processing_headers_ = false;
} else {
// Get more data before trying again.
return;
}
}
DCHECK(!processing_headers_);
size_t boundary_pos;
while ((boundary_pos = FindBoundary()) != std::string::npos) {
if (client_) {
// Strip out trailing \n\r characters in the buffer preceding the
// boundary on the same lines as Firefox.
size_t data_length = boundary_pos;
if (boundary_pos > 0 && data_[boundary_pos - 1] == '\n') {
data_length--;
if (boundary_pos > 1 && data_[boundary_pos - 2] == '\r') {
data_length--;
}
}
if (data_length > 0) {
// Send the last data chunk.
client_->didReceiveData(loader_,
data_.data(),
static_cast<int>(data_length),
encoded_data_length_);
encoded_data_length_ = 0;
}
}
size_t boundary_end_pos = boundary_pos + boundary_.length();
if (boundary_end_pos < data_.length() && '-' == data_[boundary_end_pos]) {
// This was the last boundary so we can stop processing.
stop_sending_ = true;
data_.clear();
return;
}
// We can now throw out data up through the boundary
int offset = PushOverLine(data_, boundary_end_pos);
data_ = data_.substr(boundary_end_pos + offset);
// Ok, back to parsing headers
if (!ParseHeaders()) {
processing_headers_ = true;
break;
}
}
// At this point, we should send over any data we have, but keep enough data
// buffered to handle a boundary that may have been truncated.
if (!processing_headers_ && data_.length() > boundary_.length()) {
// If the last character is a new line character, go ahead and just send
// everything we have buffered. This matches an optimization in Gecko.
int send_length = data_.length() - boundary_.length();
if (data_[data_.length() - 1] == '\n')
send_length = data_.length();
if (client_)
client_->didReceiveData(loader_,
data_.data(),
send_length,
encoded_data_length_);
data_ = data_.substr(send_length);
encoded_data_length_ = 0;
}
}
void MultipartResponseDelegate::OnCompletedRequest() {
// If we have any pending data and we're not in a header, go ahead and send
// it to WebCore.
if (!processing_headers_ && !data_.empty() && !stop_sending_ && client_) {
client_->didReceiveData(loader_,
data_.data(),
static_cast<int>(data_.length()),
encoded_data_length_);
encoded_data_length_ = 0;
}
}
int MultipartResponseDelegate::PushOverLine(const std::string& data,
size_t pos) {
int offset = 0;
if (pos < data.length() && (data[pos] == '\r' || data[pos] == '\n')) {
++offset;
if (pos + 1 < data.length() && data[pos + 1] == '\n')
++offset;
}
return offset;
}
bool MultipartResponseDelegate::ParseHeaders() {
int line_feed_increment = 1;
// Grab the headers being liberal about line endings.
size_t line_start_pos = 0;
size_t line_end_pos = data_.find('\n');
while (line_end_pos != std::string::npos) {
// Handle CRLF
if (line_end_pos > line_start_pos && data_[line_end_pos - 1] == '\r') {
line_feed_increment = 2;
--line_end_pos;
} else {
line_feed_increment = 1;
}
if (line_start_pos == line_end_pos) {
// A blank line, end of headers
line_end_pos += line_feed_increment;
break;
}
// Find the next header line.
line_start_pos = line_end_pos + line_feed_increment;
line_end_pos = data_.find('\n', line_start_pos);
}
// Truncated in the middle of a header, stop parsing.
if (line_end_pos == std::string::npos)
return false;
// Eat headers
std::string headers("\n");
headers.append(data_, 0, line_end_pos);
data_ = data_.substr(line_end_pos);
// Create a WebURLResponse based on the original set of headers + the
// replacement headers. We only replace the same few headers that gecko
// does. See netwerk/streamconv/converters/nsMultiMixedConv.cpp.
std::string content_type = net::GetSpecificHeader(headers, "content-type");
std::string mime_type;
std::string charset;
bool has_charset = false;
net::HttpUtil::ParseContentType(content_type, &mime_type, &charset,
&has_charset, NULL);
WebURLResponse response(original_response_.url());
response.setMIMEType(WebString::fromUTF8(mime_type));
response.setTextEncodingName(WebString::fromUTF8(charset));
HeaderCopier copier(&response);
original_response_.visitHTTPHeaderFields(&copier);
for (size_t i = 0; i < arraysize(kReplaceHeaders); ++i) {
std::string name(kReplaceHeaders[i]);
std::string value = net::GetSpecificHeader(headers, name);
if (!value.empty()) {
response.setHTTPHeaderField(WebString::fromUTF8(name),
WebString::fromUTF8(value));
}
}
// To avoid recording every multipart load as a separate visit in
// the history database, we want to keep track of whether the response
// is part of a multipart payload. We do want to record the first visit,
// so we only set isMultipartPayload to true after the first visit.
response.setIsMultipartPayload(has_sent_first_response_);
has_sent_first_response_ = true;
// Send the response!
if (client_)
client_->didReceiveResponse(loader_, response);
return true;
}
// Boundaries are supposed to be preceeded with --, but it looks like gecko
// doesn't require the dashes to exist. See nsMultiMixedConv::FindToken.
size_t MultipartResponseDelegate::FindBoundary() {
size_t boundary_pos = data_.find(boundary_);
if (boundary_pos != std::string::npos) {
// Back up over -- for backwards compat
// TODO(tc): Don't we only want to do this once? Gecko code doesn't seem
// to care.
if (boundary_pos >= 2) {
if ('-' == data_[boundary_pos - 1] && '-' == data_[boundary_pos - 2]) {
boundary_pos -= 2;
boundary_ = "--" + boundary_;
}
}
}
return boundary_pos;
}
bool MultipartResponseDelegate::ReadMultipartBoundary(
const WebURLResponse& response,
std::string* multipart_boundary) {
std::string content_type =
response.httpHeaderField(WebString::fromUTF8("Content-Type")).utf8();
size_t boundary_start_offset = content_type.find("boundary=");
if (boundary_start_offset == std::string::npos)
return false;
boundary_start_offset += strlen("boundary=");
size_t boundary_end_offset = content_type.find(';', boundary_start_offset);
if (boundary_end_offset == std::string::npos)
boundary_end_offset = content_type.length();
size_t boundary_length = boundary_end_offset - boundary_start_offset;
*multipart_boundary =
content_type.substr(boundary_start_offset, boundary_length);
// The byte range response can have quoted boundary strings. This is legal
// as per MIME specifications. Individual data fragements however don't
// contain quoted boundary strings.
TrimString(*multipart_boundary, "\"", multipart_boundary);
return true;
}
bool MultipartResponseDelegate::ReadContentRanges(
const WebURLResponse& response,
int64* content_range_lower_bound,
int64* content_range_upper_bound,
int64* content_range_instance_size) {
std::string content_range = response.httpHeaderField("Content-Range").utf8();
if (content_range.empty()) {
content_range = response.httpHeaderField("Range").utf8();
}
if (content_range.empty()) {
DLOG(WARNING) << "Failed to read content range from response.";
return false;
}
size_t byte_range_lower_bound_start_offset = content_range.find(" ");
if (byte_range_lower_bound_start_offset == std::string::npos) {
return false;
}
// Skip over the initial space.
byte_range_lower_bound_start_offset++;
// Find the lower bound.
size_t byte_range_lower_bound_end_offset =
content_range.find("-", byte_range_lower_bound_start_offset);
if (byte_range_lower_bound_end_offset == std::string::npos) {
return false;
}
size_t byte_range_lower_bound_characters =
byte_range_lower_bound_end_offset - byte_range_lower_bound_start_offset;
std::string byte_range_lower_bound =
content_range.substr(byte_range_lower_bound_start_offset,
byte_range_lower_bound_characters);
// Find the upper bound.
size_t byte_range_upper_bound_start_offset =
byte_range_lower_bound_end_offset + 1;
size_t byte_range_upper_bound_end_offset =
content_range.find("/", byte_range_upper_bound_start_offset);
if (byte_range_upper_bound_end_offset == std::string::npos) {
return false;
}
size_t byte_range_upper_bound_characters =
byte_range_upper_bound_end_offset - byte_range_upper_bound_start_offset;
std::string byte_range_upper_bound =
content_range.substr(byte_range_upper_bound_start_offset,
byte_range_upper_bound_characters);
// Find the instance size.
size_t byte_range_instance_size_start_offset =
byte_range_upper_bound_end_offset + 1;
size_t byte_range_instance_size_end_offset =
content_range.length();
size_t byte_range_instance_size_characters =
byte_range_instance_size_end_offset -
byte_range_instance_size_start_offset;
std::string byte_range_instance_size =
content_range.substr(byte_range_instance_size_start_offset,
byte_range_instance_size_characters);
if (!base::StringToInt64(byte_range_lower_bound, content_range_lower_bound))
return false;
if (!base::StringToInt64(byte_range_upper_bound, content_range_upper_bound))
return false;
if (!base::StringToInt64(byte_range_instance_size,
content_range_instance_size)) {
return false;
}
return true;
}
} // namespace webkit_glue
|