summaryrefslogtreecommitdiffstats
path: root/chrome/browser/safe_browsing/protocol_parser.cc
blob: 398426c451817dd66e22fa9e43ccacd158745551 (plain)
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
// 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.
//
// Parse the data returned from the SafeBrowsing v2.1 protocol response.

// TODOv3(shess): Review these changes carefully.

#include "chrome/browser/safe_browsing/protocol_parser.h"

#include <stdint.h>
#include <stdlib.h>
#include <utility>

#include "base/format_macros.h"
#include "base/logging.h"
#include "base/macros.h"
#include "base/strings/string_number_conversions.h"
#include "base/strings/string_split.h"
#include "base/strings/stringprintf.h"
#include "base/sys_byteorder.h"
#include "base/time/time.h"
#include "build/build_config.h"
#include "chrome/browser/safe_browsing/safe_browsing_util.h"

namespace safe_browsing {

namespace {

// Helper class for scanning a buffer.
class BufferReader {
 public:
  BufferReader(const char* data, size_t length)
      : data_(data),
        length_(length) {
  }

  // Return info about remaining buffer data.
  size_t length() const {
    return length_;
  }
  const char* data() const {
    return data_;
  }
  bool empty() const {
    return length_ == 0;
  }

  // Remove |l| characters from the buffer.
  void Advance(size_t l) {
    DCHECK_LE(l, length());
    data_ += l;
    length_ -= l;
  }

  // Get a reference to data in the buffer.
  // TODO(shess): I'm not sure I like this.  Fill out a StringPiece instead?
  bool RefData(const void** pptr, size_t l) {
    if (length() < l) {
      Advance(length());  // poison
      return false;
    }

    *pptr = data();
    Advance(l);
    return true;
  }

  // Copy data out of the buffer.
  bool GetData(void* ptr, size_t l) {
    const void* buf_ptr;
    if (!RefData(&buf_ptr, l))
      return false;

    memcpy(ptr, buf_ptr, l);
    return true;
  }

  // Read a 32-bit integer in network byte order into a local uint32_t.
  bool GetNet32(uint32_t* i) {
    if (!GetData(i, sizeof(*i)))
      return false;

    *i = base::NetToHost32(*i);
    return true;
  }

  // Returns false if there is no data, otherwise fills |*line| with a reference
  // to the next line of data in the buffer.
  bool GetLine(base::StringPiece* line) {
    if (!length_)
      return false;

    // Find the end of the line, or the end of the input.
    size_t eol = 0;
    while (eol < length_ && data_[eol] != '\n') {
      ++eol;
    }
    line->set(data_, eol);
    Advance(eol);

    // Skip the newline if present.
    if (length_ && data_[0] == '\n')
      Advance(1);

    return true;
  }

  // Read out |c| colon-separated pieces from the next line.  The resulting
  // pieces point into the original data buffer.
  bool GetPieces(size_t c, std::vector<base::StringPiece>* pieces) {
    base::StringPiece line;
    if (!GetLine(&line))
      return false;

    // Find the parts separated by ':'.
    while (pieces->size() + 1 < c) {
      size_t colon_ofs = line.find(':');
      if (colon_ofs == base::StringPiece::npos) {
        Advance(length_);
        return false;
      }

      pieces->push_back(line.substr(0, colon_ofs));
      line.remove_prefix(colon_ofs + 1);
    }

    // The last piece runs to the end of the line.
    pieces->push_back(line);
    return true;
  }

 private:
  const char* data_;
  size_t length_;

  DISALLOW_COPY_AND_ASSIGN(BufferReader);
};

bool ParseGetHashMetadata(
    size_t hash_count,
    BufferReader* reader,
    std::vector<SBFullHashResult>* full_hashes) {
  for (size_t i = 0; i < hash_count; ++i) {
    base::StringPiece line;
    if (!reader->GetLine(&line))
      return false;

    size_t meta_data_len;
    if (!base::StringToSizeT(line, &meta_data_len))
      return false;

    const void* meta_data;
    if (!reader->RefData(&meta_data, meta_data_len))
      return false;

    if (full_hashes) {
      (*full_hashes)[full_hashes->size() - hash_count + i].metadata.assign(
          reinterpret_cast<const char*>(meta_data), meta_data_len);
    }
  }
  return true;
}

}  // namespace

// BODY          = CACHELIFETIME LF HASHENTRY* EOF
// CACHELIFETIME = DIGIT+
// HASHENTRY     = LISTNAME ":" HASHSIZE ":" NUMRESPONSES [":m"] LF
//                 HASHDATA (METADATALEN LF METADATA)*
// HASHSIZE      = DIGIT+                  # Length of each full hash
// NUMRESPONSES  = DIGIT+                  # Number of full hashes in HASHDATA
// HASHDATA      = <HASHSIZE*NUMRESPONSES number of unsigned bytes>
// METADATALEN   = DIGIT+
// METADATA      = <METADATALEN number of unsigned bytes>
bool ParseGetHash(const char* chunk_data,
                  size_t chunk_len,
                  base::TimeDelta* cache_lifetime,
                  std::vector<SBFullHashResult>* full_hashes) {
  full_hashes->clear();
  BufferReader reader(chunk_data, chunk_len);

  // Parse out cache lifetime.
  {
    base::StringPiece line;
    if (!reader.GetLine(&line))
      return false;

    int64_t cache_lifetime_seconds;
    if (!base::StringToInt64(line, &cache_lifetime_seconds))
      return false;

    // TODO(shess): Zero also doesn't make sense, but isn't clearly forbidden,
    // either.  Maybe there should be a threshold involved.
    if (cache_lifetime_seconds < 0)
      return false;

    *cache_lifetime = base::TimeDelta::FromSeconds(cache_lifetime_seconds);
  }

  while (!reader.empty()) {
    std::vector<base::StringPiece> cmd_parts;
    if (!reader.GetPieces(3, &cmd_parts))
      return false;

    SBFullHashResult full_hash;
    full_hash.list_id = GetListId(cmd_parts[0]);

    size_t hash_len;
    if (!base::StringToSizeT(cmd_parts[1], &hash_len))
      return false;

    // TODO(shess): Is this possible?  If not, why the length present?
    if (hash_len != sizeof(SBFullHash))
      return false;

    // Metadata is indicated by an optional ":m" at the end of the line.
    bool has_metadata = false;
    base::StringPiece hash_count_string = cmd_parts[2];
    size_t optional_colon = hash_count_string.find(':', 0);
    if (optional_colon != base::StringPiece::npos) {
      if (hash_count_string.substr(optional_colon) != ":m")
        return false;
      has_metadata = true;
      hash_count_string.remove_suffix(2);
    }

    size_t hash_count;
    if (!base::StringToSizeT(hash_count_string, &hash_count))
      return false;

    if (hash_len * hash_count > reader.length())
      return false;

    // Ignore hash results from lists we don't recognize.
    if (full_hash.list_id < 0) {
      reader.Advance(hash_len * hash_count);
      if (has_metadata && !ParseGetHashMetadata(hash_count, &reader, NULL))
        return false;
      continue;
    }

    for (size_t i = 0; i < hash_count; ++i) {
      if (!reader.GetData(&full_hash.hash, hash_len))
        return false;
      full_hashes->push_back(full_hash);
    }

    if (has_metadata && !ParseGetHashMetadata(hash_count, &reader, full_hashes))
      return false;
  }

  return reader.empty();
}

// BODY       = HEADER LF PREFIXES EOF
// HEADER     = PREFIXSIZE ":" LENGTH
// PREFIXSIZE = DIGIT+         # Size of each prefix in bytes
// LENGTH     = DIGIT+         # Size of PREFIXES in bytes
std::string FormatGetHash(const std::vector<SBPrefix>& prefixes) {
  std::string request;
  request.append(base::Uint64ToString(sizeof(SBPrefix)));
  request.append(":");
  request.append(base::Uint64ToString(sizeof(SBPrefix) * prefixes.size()));
  request.append("\n");

  // SBPrefix values are read without concern for byte order, so write back the
  // same way.
  for (size_t i = 0; i < prefixes.size(); ++i) {
    request.append(reinterpret_cast<const char*>(&prefixes[i]),
                   sizeof(SBPrefix));
  }

  return request;
}

bool ParseUpdate(const char* chunk_data,
                 size_t chunk_len,
                 size_t* next_update_sec,
                 bool* reset,
                 std::vector<SBChunkDelete>* deletes,
                 std::vector<ChunkUrl>* chunk_urls) {
  DCHECK(next_update_sec);
  DCHECK(deletes);
  DCHECK(chunk_urls);

  BufferReader reader(chunk_data, chunk_len);

  // Populated below.
  std::string list_name;

  while (!reader.empty()) {
    std::vector<base::StringPiece> pieces;
    if (!reader.GetPieces(2, &pieces))
      return false;

    base::StringPiece& command = pieces[0];

    // Differentiate on the first character of the command (which is usually
    // only one character, with the exception of the 'ad' and 'sd' commands).
    switch (command[0]) {
      case 'a':
      case 's': {
        // Must be either an 'ad' (add-del) or 'sd' (sub-del) chunk. We must
        // have also parsed the list name before getting here, or the add-del
        // or sub-del will have no context.
        if (list_name.empty() || (command != "ad" && command != "sd"))
          return false;
        SBChunkDelete chunk_delete;
        chunk_delete.is_sub_del = command[0] == 's';
        StringToRanges(pieces[1].as_string(), &chunk_delete.chunk_del);
        chunk_delete.list_name = list_name;
        deletes->push_back(chunk_delete);
        break;
      }

      case 'i':
        // The line providing the name of the list (i.e. 'goog-phish-shavar').
        list_name = pieces[1].as_string();
        break;

      case 'n':
        // The line providing the next earliest time (in seconds) to re-query.
        if (!base::StringToSizeT(pieces[1], next_update_sec))
          return false;
        break;

      case 'u': {
        ChunkUrl chunk_url;
        chunk_url.url = pieces[1].as_string();  // Skip the initial "u:".
        chunk_url.list_name = list_name;
        chunk_urls->push_back(chunk_url);
        break;
      }

      case 'r':
        if (pieces[1] != "pleasereset")
          return false;
        *reset = true;
        break;

      default:
        // According to the spec, we ignore commands we don't understand.
        // TODO(shess): Does this apply to r:unknown or n:not-integer?
        break;
    }
  }

  return true;
}

// BODY      = (UINT32 CHUNKDATA)+
// UINT32    = Unsigned 32-bit integer in network byte order
// CHUNKDATA = Encoded ChunkData protocol message
bool ParseChunk(const char* data,
                size_t length,
                std::vector<scoped_ptr<SBChunkData>>* chunks) {
  BufferReader reader(data, length);

  while (!reader.empty()) {
    uint32_t l = 0;
    if (!reader.GetNet32(&l) || l == 0 || l > reader.length())
      return false;

    const void* p = NULL;
    if (!reader.RefData(&p, l))
      return false;

    scoped_ptr<SBChunkData> chunk(new SBChunkData());
    if (!chunk->ParseFrom(reinterpret_cast<const unsigned char*>(p), l))
      return false;

    chunks->push_back(std::move(chunk));
  }

  DCHECK(reader.empty());
  return true;
}

// LIST      = LISTNAME ";" LISTINFO (":" LISTINFO)*
// LISTINFO  = CHUNKTYPE ":" CHUNKLIST
// CHUNKTYPE = "a" | "s"
// CHUNKLIST = (RANGE | NUMBER) ["," CHUNKLIST]
// NUMBER    = DIGIT+
// RANGE     = NUMBER "-" NUMBER
std::string FormatList(const SBListChunkRanges& list) {
  std::string formatted_results = list.name;
  formatted_results.append(";");

  if (!list.adds.empty())
    formatted_results.append("a:").append(list.adds);
  if (!list.adds.empty() && !list.subs.empty())
    formatted_results.append(":");
  if (!list.subs.empty())
    formatted_results.append("s:").append(list.subs);
  formatted_results.append("\n");

  return formatted_results;
}

}  // namespace safe_browsing