The Tor Browser / file comparison

comparison: netwerk/streamconv/converters/nsDirIndexParser.cpp

netwerk/streamconv/converters/nsDirIndexParser.cpp

branch
TOR_BUG_9701
changeset 9
a63d609f5ebe
equal deleted inserted replaced
-1:000000000000 0:fe91278d4178
1 /* -*- Mode: C++; tab-width: 2; indent-tabs-mode: nil; c-basic-offset: 2 -*- */
2 /* This Source Code Form is subject to the terms of the Mozilla Public
3 * License, v. 2.0. If a copy of the MPL was not distributed with this
4 * file, You can obtain one at http://mozilla.org/MPL/2.0/. */
5
6 /* This parsing code originally lived in xpfe/components/directory/ - bbaetz */
7
8 #include "mozilla/ArrayUtils.h"
9
10 #include "prprf.h"
11
12 #include "nsDirIndexParser.h"
13 #include "nsEscape.h"
14 #include "nsIInputStream.h"
15 #include "nsCRT.h"
16 #include "mozilla/dom/FallbackEncoding.h"
17 #include "nsITextToSubURI.h"
18 #include "nsIDirIndex.h"
19 #include "nsServiceManagerUtils.h"
20
21 using namespace mozilla;
22
23 NS_IMPL_ISUPPORTS(nsDirIndexParser,
24 nsIRequestObserver,
25 nsIStreamListener,
26 nsIDirIndexParser)
27
28 nsDirIndexParser::nsDirIndexParser() {
29 }
30
31 nsresult
32 nsDirIndexParser::Init() {
33 mLineStart = 0;
34 mHasDescription = false;
35 mFormat = nullptr;
36 mozilla::dom::FallbackEncoding::FromLocale(mEncoding);
37
38 nsresult rv;
39 // XXX not threadsafe
40 if (gRefCntParser++ == 0)
41 rv = CallGetService(NS_ITEXTTOSUBURI_CONTRACTID, &gTextToSubURI);
42 else
43 rv = NS_OK;
44
45 return rv;
46 }
47
48 nsDirIndexParser::~nsDirIndexParser() {
49 delete[] mFormat;
50 // XXX not threadsafe
51 if (--gRefCntParser == 0) {
52 NS_IF_RELEASE(gTextToSubURI);
53 }
54 }
55
56 NS_IMETHODIMP
57 nsDirIndexParser::SetListener(nsIDirIndexListener* aListener) {
58 mListener = aListener;
59 return NS_OK;
60 }
61
62 NS_IMETHODIMP
63 nsDirIndexParser::GetListener(nsIDirIndexListener** aListener) {
64 NS_IF_ADDREF(*aListener = mListener.get());
65 return NS_OK;
66 }
67
68 NS_IMETHODIMP
69 nsDirIndexParser::GetComment(char** aComment) {
70 *aComment = ToNewCString(mComment);
71
72 if (!*aComment)
73 return NS_ERROR_OUT_OF_MEMORY;
74
75 return NS_OK;
76 }
77
78 NS_IMETHODIMP
79 nsDirIndexParser::SetEncoding(const char* aEncoding) {
80 mEncoding.Assign(aEncoding);
81 return NS_OK;
82 }
83
84 NS_IMETHODIMP
85 nsDirIndexParser::GetEncoding(char** aEncoding) {
86 *aEncoding = ToNewCString(mEncoding);
87
88 if (!*aEncoding)
89 return NS_ERROR_OUT_OF_MEMORY;
90
91 return NS_OK;
92 }
93
94 NS_IMETHODIMP
95 nsDirIndexParser::OnStartRequest(nsIRequest* aRequest, nsISupports* aCtxt) {
96 return NS_OK;
97 }
98
99 NS_IMETHODIMP
100 nsDirIndexParser::OnStopRequest(nsIRequest *aRequest, nsISupports *aCtxt,
101 nsresult aStatusCode) {
102 // Finish up
103 if (mBuf.Length() > (uint32_t) mLineStart) {
104 ProcessData(aRequest, aCtxt);
105 }
106
107 return NS_OK;
108 }
109
110 nsDirIndexParser::Field
111 nsDirIndexParser::gFieldTable[] = {
112 { "Filename", FIELD_FILENAME },
113 { "Description", FIELD_DESCRIPTION },
114 { "Content-Length", FIELD_CONTENTLENGTH },
115 { "Last-Modified", FIELD_LASTMODIFIED },
116 { "Content-Type", FIELD_CONTENTTYPE },
117 { "File-Type", FIELD_FILETYPE },
118 { nullptr, FIELD_UNKNOWN }
119 };
120
121 nsrefcnt nsDirIndexParser::gRefCntParser = 0;
122 nsITextToSubURI *nsDirIndexParser::gTextToSubURI;
123
124 nsresult
125 nsDirIndexParser::ParseFormat(const char* aFormatStr) {
126 // Parse a "200" format line, and remember the fields and their
127 // ordering in mFormat. Multiple 200 lines stomp on each other.
128
129 // Lets find out how many elements we have.
130 // easier to do this then realloc
131 const char* pos = aFormatStr;
132 unsigned int num = 0;
133 do {
134 while (*pos && nsCRT::IsAsciiSpace(char16_t(*pos)))
135 ++pos;
136
137 ++num;
138 // There are a maximum of six allowed header fields (doubled plus
139 // terminator, just in case) -- Bug 443299
140 if (num > (2 * ArrayLength(gFieldTable)))
141 return NS_ERROR_UNEXPECTED;
142
143 if (! *pos)
144 break;
145
146 while (*pos && !nsCRT::IsAsciiSpace(char16_t(*pos)))
147 ++pos;
148
149 } while (*pos);
150
151 delete[] mFormat;
152 mFormat = new int[num+1];
153 // Prevent nullptr Deref - Bug 443299
154 if (mFormat == nullptr)
155 return NS_ERROR_OUT_OF_MEMORY;
156 mFormat[num] = -1;
157
158 int formatNum=0;
159 do {
160 while (*aFormatStr && nsCRT::IsAsciiSpace(char16_t(*aFormatStr)))
161 ++aFormatStr;
162
163 if (! *aFormatStr)
164 break;
165
166 nsAutoCString name;
167 int32_t len = 0;
168 while (aFormatStr[len] && !nsCRT::IsAsciiSpace(char16_t(aFormatStr[len])))
169 ++len;
170 name.SetCapacity(len + 1);
171 name.Append(aFormatStr, len);
172 aFormatStr += len;
173
174 // Okay, we're gonna monkey with the nsStr. Bold!
175 name.SetLength(nsUnescapeCount(name.BeginWriting()));
176
177 // All tokens are case-insensitive - http://www.mozilla.org/projects/netlib/dirindexformat.html
178 if (name.LowerCaseEqualsLiteral("description"))
179 mHasDescription = true;
180
181 for (Field* i = gFieldTable; i->mName; ++i) {
182 if (name.EqualsIgnoreCase(i->mName)) {
183 mFormat[formatNum] = i->mType;
184 ++formatNum;
185 break;
186 }
187 }
188
189 } while (*aFormatStr);
190
191 return NS_OK;
192 }
193
194 nsresult
195 nsDirIndexParser::ParseData(nsIDirIndex *aIdx, char* aDataStr) {
196 // Parse a "201" data line, using the field ordering specified in
197 // mFormat.
198
199 if (!mFormat) {
200 // Ignore if we haven't seen a format yet.
201 return NS_OK;
202 }
203
204 nsresult rv = NS_OK;
205
206 nsAutoCString filename;
207
208 for (int32_t i = 0; mFormat[i] != -1; ++i) {
209 // If we've exhausted the data before we run out of fields, just
210 // bail.
211 if (! *aDataStr)
212 break;
213
214 while (*aDataStr && nsCRT::IsAsciiSpace(*aDataStr))
215 ++aDataStr;
216
217 char *value = aDataStr;
218
219 if (*aDataStr == '"' || *aDataStr == '\'') {
220 // it's a quoted string. snarf everything up to the next quote character
221 const char quotechar = *(aDataStr++);
222 ++value;
223 while (*aDataStr && *aDataStr != quotechar)
224 ++aDataStr;
225 *aDataStr++ = '\0';
226
227 if (! aDataStr) {
228 NS_WARNING("quoted value not terminated");
229 }
230 } else {
231 // it's unquoted. snarf until we see whitespace.
232 value = aDataStr;
233 while (*aDataStr && (!nsCRT::IsAsciiSpace(*aDataStr)))
234 ++aDataStr;
235 *aDataStr++ = '\0';
236 }
237
238 fieldType t = fieldType(mFormat[i]);
239 switch (t) {
240 case FIELD_FILENAME: {
241 // don't unescape at this point, so that UnEscapeAndConvert() can
242 filename = value;
243
244 bool success = false;
245
246 nsAutoString entryuri;
247
248 if (gTextToSubURI) {
249 char16_t *result = nullptr;
250 if (NS_SUCCEEDED(rv = gTextToSubURI->UnEscapeAndConvert(mEncoding.get(), filename.get(),
251 &result)) && (result)) {
252 if (*result) {
253 aIdx->SetLocation(filename.get());
254 if (!mHasDescription)
255 aIdx->SetDescription(result);
256 success = true;
257 }
258 NS_Free(result);
259 } else {
260 NS_WARNING("UnEscapeAndConvert error");
261 }
262 }
263
264 if (!success) {
265 // if unsuccessfully at charset conversion, then
266 // just fallback to unescape'ing in-place
267 // XXX - this shouldn't be using UTF8, should it?
268 // when can we fail to get the service, anyway? - bbaetz
269 aIdx->SetLocation(filename.get());
270 if (!mHasDescription) {
271 aIdx->SetDescription(NS_ConvertUTF8toUTF16(value).get());
272 }
273 }
274 }
275 break;
276 case FIELD_DESCRIPTION:
277 nsUnescape(value);
278 aIdx->SetDescription(NS_ConvertUTF8toUTF16(value).get());
279 break;
280 case FIELD_CONTENTLENGTH:
281 {
282 int64_t len;
283 int32_t status = PR_sscanf(value, "%lld", &len);
284 if (status == 1)
285 aIdx->SetSize(len);
286 else
287 aIdx->SetSize(UINT64_MAX); // UINT64_MAX means unknown
288 }
289 break;
290 case FIELD_LASTMODIFIED:
291 {
292 PRTime tm;
293 nsUnescape(value);
294 if (PR_ParseTimeString(value, false, &tm) == PR_SUCCESS) {
295 aIdx->SetLastModified(tm);
296 }
297 }
298 break;
299 case FIELD_CONTENTTYPE:
300 aIdx->SetContentType(value);
301 break;
302 case FIELD_FILETYPE:
303 // unescape in-place
304 nsUnescape(value);
305 if (!nsCRT::strcasecmp(value, "directory")) {
306 aIdx->SetType(nsIDirIndex::TYPE_DIRECTORY);
307 } else if (!nsCRT::strcasecmp(value, "file")) {
308 aIdx->SetType(nsIDirIndex::TYPE_FILE);
309 } else if (!nsCRT::strcasecmp(value, "symbolic-link")) {
310 aIdx->SetType(nsIDirIndex::TYPE_SYMLINK);
311 } else {
312 aIdx->SetType(nsIDirIndex::TYPE_UNKNOWN);
313 }
314 break;
315 case FIELD_UNKNOWN:
316 // ignore
317 break;
318 }
319 }
320
321 return NS_OK;
322 }
323
324 NS_IMETHODIMP
325 nsDirIndexParser::OnDataAvailable(nsIRequest *aRequest, nsISupports *aCtxt,
326 nsIInputStream *aStream,
327 uint64_t aSourceOffset,
328 uint32_t aCount) {
329 if (aCount < 1)
330 return NS_OK;
331
332 int32_t len = mBuf.Length();
333
334 // Ensure that our mBuf has capacity to hold the data we're about to
335 // read.
336 if (!mBuf.SetLength(len + aCount, fallible_t()))
337 return NS_ERROR_OUT_OF_MEMORY;
338
339 // Now read the data into our buffer.
340 nsresult rv;
341 uint32_t count;
342 rv = aStream->Read(mBuf.BeginWriting() + len, aCount, &count);
343 if (NS_FAILED(rv)) return rv;
344
345 // Set the string's length according to the amount of data we've read.
346 // Note: we know this to work on nsCString. This isn't guaranteed to
347 // work on other strings.
348 mBuf.SetLength(len + count);
349
350 return ProcessData(aRequest, aCtxt);
351 }
352
353 nsresult
354 nsDirIndexParser::ProcessData(nsIRequest *aRequest, nsISupports *aCtxt) {
355 if (!mListener)
356 return NS_ERROR_FAILURE;
357
358 int32_t numItems = 0;
359
360 while(true) {
361 ++numItems;
362
363 int32_t eol = mBuf.FindCharInSet("\n\r", mLineStart);
364 if (eol < 0) break;
365 mBuf.SetCharAt(char16_t('\0'), eol);
366
367 const char *line = mBuf.get() + mLineStart;
368
369 int32_t lineLen = eol - mLineStart;
370 mLineStart = eol + 1;
371
372 if (lineLen >= 4) {
373 nsresult rv;
374 const char *buf = line;
375
376 if (buf[0] == '1') {
377 if (buf[1] == '0') {
378 if (buf[2] == '0' && buf[3] == ':') {
379 // 100. Human-readable comment line. Ignore
380 } else if (buf[2] == '1' && buf[3] == ':') {
381 // 101. Human-readable information line.
382 mComment.Append(buf + 4);
383
384 char *value = ((char *)buf) + 4;
385 nsUnescape(value);
386 mListener->OnInformationAvailable(aRequest, aCtxt, NS_ConvertUTF8toUTF16(value));
387
388 } else if (buf[2] == '2' && buf[3] == ':') {
389 // 102. Human-readable information line, HTML.
390 mComment.Append(buf + 4);
391 }
392 }
393 } else if (buf[0] == '2') {
394 if (buf[1] == '0') {
395 if (buf[2] == '0' && buf[3] == ':') {
396 // 200. Define field names
397 rv = ParseFormat(buf + 4);
398 if (NS_FAILED(rv)) {
399 return rv;
400 }
401 } else if (buf[2] == '1' && buf[3] == ':') {
402 // 201. Field data
403 nsCOMPtr<nsIDirIndex> idx = do_CreateInstance("@mozilla.org/dirIndex;1",&rv);
404 if (NS_FAILED(rv))
405 return rv;
406
407 rv = ParseData(idx, ((char *)buf) + 4);
408 if (NS_FAILED(rv)) {
409 return rv;
410 }
411
412 mListener->OnIndexAvailable(aRequest, aCtxt, idx);
413 }
414 }
415 } else if (buf[0] == '3') {
416 if (buf[1] == '0') {
417 if (buf[2] == '0' && buf[3] == ':') {
418 // 300. Self-referring URL
419 } else if (buf[2] == '1' && buf[3] == ':') {
420 // 301. OUR EXTENSION - encoding
421 int i = 4;
422 while (buf[i] && nsCRT::IsAsciiSpace(buf[i]))
423 ++i;
424
425 if (buf[i])
426 SetEncoding(buf+i);
427 }
428 }
429 }
430 }
431 }
432
433 return NS_OK;
434 }

Mercurial Repository: The Tor Browser

Europalab SCM Repository Server

mercurial