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
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
1646
1647
1648
1649
1650
1651
1652
1653
1654
1655
1656
1657
1658
1659
1660
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
1676
1677
1678
1679
1680
1681
1682
1683
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
1699
1700
1701
1702
1703
1704
1705
1706
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
1793
1794
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
1808
1809
1810
1811
1812
1813
1814
1815
1816
1817
1818
1819
1820
1821
1822
1823
1824
1825
1826
1827
1828
1829
1830
1831
1832
1833
1834
1835
1836
1837
1838
1839
1840
1841
1842
1843
1844
1845
1846
1847
1848
1849
1850
1851
1852
1853
1854
1855
1856
1857
1858
1859
1860
1861
1862
1863
1864
1865
1866
1867
1868
1869
1870
1871
1872
1873
1874
1875
1876
1877
1878
1879
1880
1881
1882
1883
1884
1885
1886
1887
1888
1889
1890
1891
1892
1893
1894
1895
1896
1897
1898
1899
1900
1901
1902
1903
1904
1905
1906
1907
1908
1909
1910
1911
1912
1913
1914
1915
1916
1917
1918
1919
1920
1921
1922
1923
1924
1925
1926
1927
1928
1929
1930
1931
1932
1933
1934
1935
1936
1937
1938
1939
1940
1941
1942
1943
1944
1945
1946
1947
1948
1949
1950
1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
|
// Copyright (c) 2010 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.
// Portions of this code based on Mozilla:
// (netwerk/cookie/src/nsCookieService.cpp)
/* ***** BEGIN LICENSE BLOCK *****
* Version: MPL 1.1/GPL 2.0/LGPL 2.1
*
* The contents of this file are subject to the Mozilla Public License Version
* 1.1 (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
* http://www.mozilla.org/MPL/
*
* Software distributed under the License is distributed on an "AS IS" basis,
* WITHOUT WARRANTY OF ANY KIND, either express or implied. See the License
* for the specific language governing rights and limitations under the
* License.
*
* The Original Code is mozilla.org code.
*
* The Initial Developer of the Original Code is
* Netscape Communications Corporation.
* Portions created by the Initial Developer are Copyright (C) 2003
* the Initial Developer. All Rights Reserved.
*
* Contributor(s):
* Daniel Witte (dwitte@stanford.edu)
* Michiel van Leeuwen (mvl@exedo.nl)
*
* Alternatively, the contents of this file may be used under the terms of
* either the GNU General Public License Version 2 or later (the "GPL"), or
* the GNU Lesser General Public License Version 2.1 or later (the "LGPL"),
* in which case the provisions of the GPL or the LGPL are applicable instead
* of those above. If you wish to allow use of your version of this file only
* under the terms of either the GPL or the LGPL, and not to allow others to
* use your version of this file under the terms of the MPL, indicate your
* decision by deleting the provisions above and replace them with the notice
* and other provisions required by the GPL or the LGPL. If you do not delete
* the provisions above, a recipient may use your version of this file under
* the terms of any one of the MPL, the GPL or the LGPL.
*
* ***** END LICENSE BLOCK ***** */
#include "net/base/cookie_monster.h"
#include <algorithm>
#include "base/basictypes.h"
#include "base/format_macros.h"
#include "base/logging.h"
#include "base/metrics/histogram.h"
#include "base/scoped_ptr.h"
#include "base/string_tokenizer.h"
#include "base/string_util.h"
#include "base/stringprintf.h"
#include "googleurl/src/gurl.h"
#include "googleurl/src/url_canon.h"
#include "net/base/net_util.h"
#include "net/base/registry_controlled_domain.h"
using base::Time;
using base::TimeDelta;
using base::TimeTicks;
static const int kMinutesInTenYears = 10 * 365 * 24 * 60;
namespace net {
// See comments at declaration of these variables in cookie_monster.h
// for details.
const size_t CookieMonster::kDomainMaxCookies = 180;
const size_t CookieMonster::kDomainPurgeCookies = 30;
const size_t CookieMonster::kMaxCookies = 3300;
const size_t CookieMonster::kPurgeCookies = 300;
const int CookieMonster::kSafeFromGlobalPurgeDays = 30;
namespace {
// Default minimum delay after updating a cookie's LastAccessDate before we
// will update it again.
const int kDefaultAccessUpdateThresholdSeconds = 60;
// Comparator to sort cookies from highest creation date to lowest
// creation date.
struct OrderByCreationTimeDesc {
bool operator()(const CookieMonster::CookieMap::iterator& a,
const CookieMonster::CookieMap::iterator& b) const {
return a->second->CreationDate() > b->second->CreationDate();
}
};
} // namespace
// static
bool CookieMonster::enable_file_scheme_ = false;
// static
void CookieMonster::EnableFileScheme() {
enable_file_scheme_ = true;
}
CookieMonster::CookieMonster(PersistentCookieStore* store, Delegate* delegate)
: initialized_(false),
expiry_and_key_scheme_(expiry_and_key_default_),
store_(store),
last_access_threshold_(
TimeDelta::FromSeconds(kDefaultAccessUpdateThresholdSeconds)),
delegate_(delegate),
last_statistic_record_time_(Time::Now()) {
InitializeHistograms();
SetDefaultCookieableSchemes();
}
CookieMonster::~CookieMonster() {
DeleteAll(false);
}
// Initialize all histogram counter variables used in this class.
//
// Normal histogram usage involves using the macros defined in
// histogram.h, which automatically takes care of declaring these
// variables (as statics), initializing them, and accumulating into
// them, all from a single entry point. Unfortunately, that solution
// doesn't work for the CookieMonster, as it's vulnerable to races between
// separate threads executing the same functions and hence initializing the
// same static variables. There isn't a race danger in the histogram
// accumulation calls; they are written to be resilient to simultaneous
// calls from multiple threads.
//
// The solution taken here is to have per-CookieMonster instance
// variables that are constructed during CookieMonster construction.
// Note that these variables refer to the same underlying histogram,
// so we still race (but safely) with other CookieMonster instances
// for accumulation.
//
// To do this we've expanded out the individual histogram macros calls,
// with declarations of the variables in the class decl, initialization here
// (done from the class constructor) and direct calls to the accumulation
// methods where needed. The specific histogram macro calls on which the
// initialization is based are included in comments below.
void CookieMonster::InitializeHistograms() {
// From UMA_HISTOGRAM_CUSTOM_COUNTS
histogram_expiration_duration_minutes_ = base::Histogram::FactoryGet(
"Cookie.ExpirationDurationMinutes",
1, kMinutesInTenYears, 50,
base::Histogram::kUmaTargetedHistogramFlag);
histogram_between_access_interval_minutes_ = base::Histogram::FactoryGet(
"Cookie.BetweenAccessIntervalMinutes",
1, kMinutesInTenYears, 50,
base::Histogram::kUmaTargetedHistogramFlag);
histogram_evicted_last_access_minutes_ = base::Histogram::FactoryGet(
"Cookie.EvictedLastAccessMinutes",
1, kMinutesInTenYears, 50,
base::Histogram::kUmaTargetedHistogramFlag);
histogram_count_ = base::Histogram::FactoryGet(
"Cookie.Count", 1, 4000, 50,
base::Histogram::kUmaTargetedHistogramFlag);
histogram_domain_count_ = base::Histogram::FactoryGet(
"Cookie.DomainCount", 1, 4000, 50,
base::Histogram::kUmaTargetedHistogramFlag);
histogram_etldp1_count_ = base::Histogram::FactoryGet(
"Cookie.Etldp1Count", 1, 4000, 50,
base::Histogram::kUmaTargetedHistogramFlag);
histogram_domain_per_etldp1_count_ = base::Histogram::FactoryGet(
"Cookie.DomainPerEtldp1Count", 1, 4000, 50,
base::Histogram::kUmaTargetedHistogramFlag);
// From UMA_HISTOGRAM_COUNTS_10000 & UMA_HISTOGRAM_CUSTOM_COUNTS
histogram_number_duplicate_db_cookies_ = base::Histogram::FactoryGet(
"Net.NumDuplicateCookiesInDb", 1, 10000, 50,
base::Histogram::kUmaTargetedHistogramFlag);
// From UMA_HISTOGRAM_ENUMERATION
histogram_cookie_deletion_cause_ = base::LinearHistogram::FactoryGet(
"Cookie.DeletionCause", 1,
DELETE_COOKIE_LAST_ENTRY - 1, DELETE_COOKIE_LAST_ENTRY,
base::Histogram::kUmaTargetedHistogramFlag);
// From UMA_HISTOGRAM_{CUSTOM_,}TIMES
histogram_time_get_ = base::Histogram::FactoryTimeGet("Cookie.TimeGet",
base::TimeDelta::FromMilliseconds(1), base::TimeDelta::FromMinutes(1),
50, base::Histogram::kUmaTargetedHistogramFlag);
histogram_time_load_ = base::Histogram::FactoryTimeGet("Cookie.TimeLoad",
base::TimeDelta::FromMilliseconds(1), base::TimeDelta::FromMinutes(1),
50, base::Histogram::kUmaTargetedHistogramFlag);
}
void CookieMonster::InitStore() {
DCHECK(store_) << "Store must exist to initialize";
TimeTicks beginning_time(TimeTicks::Now());
// Initialize the store and sync in any saved persistent cookies. We don't
// care if it's expired, insert it so it can be garbage collected, removed,
// and sync'd.
std::vector<CanonicalCookie*> cookies;
// Reserve space for the maximum amount of cookies a database should have.
// This prevents multiple vector growth / copies as we append cookies.
cookies.reserve(kMaxCookies);
store_->Load(&cookies);
// Avoid ever letting cookies with duplicate creation times into the store;
// that way we don't have to worry about what sections of code are safe
// to call while it's in that state.
std::set<int64> creation_times;
// Presumably later than any access time in the store.
Time earliest_access_time;
for (std::vector<CanonicalCookie*>::const_iterator it = cookies.begin();
it != cookies.end(); ++it) {
int64 cookie_creation_time = (*it)->CreationDate().ToInternalValue();
if (creation_times.insert(cookie_creation_time).second) {
InternalInsertCookie(GetKey((*it)->Domain()), *it, false);
const Time cookie_access_time((*it)->LastAccessDate());
if (earliest_access_time.is_null() ||
cookie_access_time < earliest_access_time)
earliest_access_time = cookie_access_time;
} else {
LOG(ERROR) << base::StringPrintf("Found cookies with duplicate creation "
"times in backing store: "
"{name='%s', domain='%s', path='%s'}",
(*it)->Name().c_str(),
(*it)->Domain().c_str(),
(*it)->Path().c_str());
// We've been given ownership of the cookie and are throwing it
// away; reclaim the space.
delete (*it);
}
}
earliest_access_time_= earliest_access_time;
// After importing cookies from the PersistentCookieStore, verify that
// none of our other constraints are violated.
//
// In particular, the backing store might have given us duplicate cookies.
EnsureCookiesMapIsValid();
histogram_time_load_->AddTime(TimeTicks::Now() - beginning_time);
}
void CookieMonster::EnsureCookiesMapIsValid() {
lock_.AssertAcquired();
int num_duplicates_trimmed = 0;
// Iterate through all the of the cookies, grouped by host.
CookieMap::iterator prev_range_end = cookies_.begin();
while (prev_range_end != cookies_.end()) {
CookieMap::iterator cur_range_begin = prev_range_end;
const std::string key = cur_range_begin->first; // Keep a copy.
CookieMap::iterator cur_range_end = cookies_.upper_bound(key);
prev_range_end = cur_range_end;
// Ensure no equivalent cookies for this host.
num_duplicates_trimmed +=
TrimDuplicateCookiesForKey(key, cur_range_begin, cur_range_end);
}
// Record how many duplicates were found in the database.
// See InitializeHistograms() for details.
histogram_cookie_deletion_cause_->Add(num_duplicates_trimmed);
}
// Our strategy to find duplicates is:
// (1) Build a map from (cookiename, cookiepath) to
// {list of cookies with this signature, sorted by creation time}.
// (2) For each list with more than 1 entry, keep the cookie having the
// most recent creation time, and delete the others.
namespace {
// Two cookies are considered equivalent if they have the same domain,
// name, and path.
struct CookieSignature {
public:
CookieSignature(const std::string& name, const std::string& domain,
const std::string& path)
: name(name),
domain(domain),
path(path) {}
// To be a key for a map this class needs to be assignable, copyable,
// and have an operator<. The default assignment operator
// and copy constructor are exactly what we want.
bool operator<(const CookieSignature& cs) const {
// Name compare dominates, then domain, then path.
int diff = name.compare(cs.name);
if (diff != 0)
return diff < 0;
diff = domain.compare(cs.domain);
if (diff != 0)
return diff < 0;
return path.compare(cs.path) < 0;
}
std::string name;
std::string domain;
std::string path;
};
}
int CookieMonster::TrimDuplicateCookiesForKey(
const std::string& key,
CookieMap::iterator begin,
CookieMap::iterator end) {
lock_.AssertAcquired();
// Set of cookies ordered by creation time.
typedef std::set<CookieMap::iterator, OrderByCreationTimeDesc> CookieSet;
// Helper map we populate to find the duplicates.
typedef std::map<CookieSignature, CookieSet> EquivalenceMap;
EquivalenceMap equivalent_cookies;
// The number of duplicate cookies that have been found.
int num_duplicates = 0;
// Iterate through all of the cookies in our range, and insert them into
// the equivalence map.
for (CookieMap::iterator it = begin; it != end; ++it) {
DCHECK_EQ(key, it->first);
CanonicalCookie* cookie = it->second;
CookieSignature signature(cookie->Name(), cookie->Domain(),
cookie->Path());
CookieSet& set = equivalent_cookies[signature];
// We found a duplicate!
if (!set.empty())
num_duplicates++;
// We save the iterator into |cookies_| rather than the actual cookie
// pointer, since we may need to delete it later.
bool insert_success = set.insert(it).second;
DCHECK(insert_success) <<
"Duplicate creation times found in duplicate cookie name scan.";
}
// If there were no duplicates, we are done!
if (num_duplicates == 0)
return 0;
// Make sure we find everything below that we did above.
int num_duplicates_found = 0;
// Otherwise, delete all the duplicate cookies, both from our in-memory store
// and from the backing store.
for (EquivalenceMap::iterator it = equivalent_cookies.begin();
it != equivalent_cookies.end();
++it) {
const CookieSignature& signature = it->first;
CookieSet& dupes = it->second;
if (dupes.size() <= 1)
continue; // This cookiename/path has no duplicates.
num_duplicates_found += dupes.size() - 1;
// Since |dups| is sorted by creation time (descending), the first cookie
// is the most recent one, so we will keep it. The rest are duplicates.
dupes.erase(dupes.begin());
LOG(ERROR) << base::StringPrintf(
"Found %d duplicate cookies for host='%s', "
"with {name='%s', domain='%s', path='%s'}",
static_cast<int>(dupes.size()),
key.c_str(),
signature.name.c_str(),
signature.domain.c_str(),
signature.path.c_str());
// Remove all the cookies identified by |dupes|. It is valid to delete our
// list of iterators one at a time, since |cookies_| is a multimap (they
// don't invalidate existing iterators following deletion).
for (CookieSet::iterator dupes_it = dupes.begin();
dupes_it != dupes.end();
++dupes_it) {
InternalDeleteCookie(*dupes_it, true /*sync_to_store*/,
DELETE_COOKIE_DUPLICATE_IN_BACKING_STORE);
}
}
DCHECK_EQ(num_duplicates, num_duplicates_found);
return num_duplicates;
}
void CookieMonster::SetDefaultCookieableSchemes() {
// Note: file must be the last scheme.
static const char* kDefaultCookieableSchemes[] = { "http", "https", "file" };
int num_schemes = enable_file_scheme_ ? 3 : 2;
SetCookieableSchemes(kDefaultCookieableSchemes, num_schemes);
}
void CookieMonster::SetExpiryAndKeyScheme(ExpiryAndKeyScheme key_scheme) {
DCHECK(!initialized_);
expiry_and_key_scheme_ = key_scheme;
}
// The system resolution is not high enough, so we can have multiple
// set cookies that result in the same system time. When this happens, we
// increment by one Time unit. Let's hope computers don't get too fast.
Time CookieMonster::CurrentTime() {
return std::max(Time::Now(),
Time::FromInternalValue(last_time_seen_.ToInternalValue() + 1));
}
// Parse a cookie expiration time. We try to be lenient, but we need to
// assume some order to distinguish the fields. The basic rules:
// - The month name must be present and prefix the first 3 letters of the
// full month name (jan for January, jun for June).
// - If the year is <= 2 digits, it must occur after the day of month.
// - The time must be of the format hh:mm:ss.
// An average cookie expiration will look something like this:
// Sat, 15-Apr-17 21:01:22 GMT
Time CookieMonster::ParseCookieTime(const std::string& time_string) {
static const char* kMonths[] = { "jan", "feb", "mar", "apr", "may", "jun",
"jul", "aug", "sep", "oct", "nov", "dec" };
static const int kMonthsLen = arraysize(kMonths);
// We want to be pretty liberal, and support most non-ascii and non-digit
// characters as a delimiter. We can't treat : as a delimiter, because it
// is the delimiter for hh:mm:ss, and we want to keep this field together.
// We make sure to include - and +, since they could prefix numbers.
// If the cookie attribute came in in quotes (ex expires="XXX"), the quotes
// will be preserved, and we will get them here. So we make sure to include
// quote characters, and also \ for anything that was internally escaped.
static const char* kDelimiters = "\t !\"#$%&'()*+,-./;<=>?@[\\]^_`{|}~";
Time::Exploded exploded = {0};
StringTokenizer tokenizer(time_string, kDelimiters);
bool found_day_of_month = false;
bool found_month = false;
bool found_time = false;
bool found_year = false;
while (tokenizer.GetNext()) {
const std::string token = tokenizer.token();
DCHECK(!token.empty());
bool numerical = IsAsciiDigit(token[0]);
// String field
if (!numerical) {
if (!found_month) {
for (int i = 0; i < kMonthsLen; ++i) {
// Match prefix, so we could match January, etc
if (base::strncasecmp(token.c_str(), kMonths[i], 3) == 0) {
exploded.month = i + 1;
found_month = true;
break;
}
}
} else {
// If we've gotten here, it means we've already found and parsed our
// month, and we have another string, which we would expect to be the
// the time zone name. According to the RFC and my experiments with
// how sites format their expirations, we don't have much of a reason
// to support timezones. We don't want to ever barf on user input,
// but this DCHECK should pass for well-formed data.
// DCHECK(token == "GMT");
}
// Numeric field w/ a colon
} else if (token.find(':') != std::string::npos) {
if (!found_time &&
#ifdef COMPILER_MSVC
sscanf_s(
#else
sscanf(
#endif
token.c_str(), "%2u:%2u:%2u", &exploded.hour,
&exploded.minute, &exploded.second) == 3) {
found_time = true;
} else {
// We should only ever encounter one time-like thing. If we're here,
// it means we've found a second, which shouldn't happen. We keep
// the first. This check should be ok for well-formed input:
// NOTREACHED();
}
// Numeric field
} else {
// Overflow with atoi() is unspecified, so we enforce a max length.
if (!found_day_of_month && token.length() <= 2) {
exploded.day_of_month = atoi(token.c_str());
found_day_of_month = true;
} else if (!found_year && token.length() <= 5) {
exploded.year = atoi(token.c_str());
found_year = true;
} else {
// If we're here, it means we've either found an extra numeric field,
// or a numeric field which was too long. For well-formed input, the
// following check would be reasonable:
// NOTREACHED();
}
}
}
if (!found_day_of_month || !found_month || !found_time || !found_year) {
// We didn't find all of the fields we need. For well-formed input, the
// following check would be reasonable:
// NOTREACHED() << "Cookie parse expiration failed: " << time_string;
return Time();
}
// Normalize the year to expand abbreviated years to the full year.
if (exploded.year >= 69 && exploded.year <= 99)
exploded.year += 1900;
if (exploded.year >= 0 && exploded.year <= 68)
exploded.year += 2000;
// If our values are within their correct ranges, we got our time.
if (exploded.day_of_month >= 1 && exploded.day_of_month <= 31 &&
exploded.month >= 1 && exploded.month <= 12 &&
exploded.year >= 1601 && exploded.year <= 30827 &&
exploded.hour <= 23 && exploded.minute <= 59 && exploded.second <= 59) {
return Time::FromUTCExploded(exploded);
}
// One of our values was out of expected range. For well-formed input,
// the following check would be reasonable:
// NOTREACHED() << "Cookie exploded expiration failed: " << time_string;
return Time();
}
bool CookieMonster::DomainIsHostOnly(const std::string& domain_string) {
return (domain_string.empty() || domain_string[0] != '.');
}
// Returns the effective TLD+1 for a given host. This only makes sense for http
// and https schemes. For other schemes, the host will be returned unchanged
// (minus any leading period).
static std::string GetEffectiveDomain(const std::string& scheme,
const std::string& host) {
if (scheme == "http" || scheme == "https")
return RegistryControlledDomainService::GetDomainAndRegistry(host);
if (!CookieMonster::DomainIsHostOnly(host))
return host.substr(1);
return host;
}
// A wrapper around RegistryControlledDomainService::GetDomainAndRegistry
// to make clear we're creating a key for our local map. Here and
// in FindCookiesForHostAndDomain() are the only two places where
// we need to conditionalize based on key type.
//
// Note that this key algorithm explicitly ignores the scheme. This is
// because when we're entering cookies into the map from the backing store,
// we in general won't have the scheme at that point.
// In practical terms, this means that file cookies will be stored
// in the map either by an empty string or by UNC name (and will be
// limited by kMaxCookiesPerHost), and extension cookies will be stored
// based on the single extension id, as the extension id won't have the
// form of a DNS host and hence GetKey() will return it unchanged.
//
// Arguably the right thing to do here is to make the key
// algorithm dependent on the scheme, and make sure that the scheme is
// available everywhere the key must be obtained (specfically at backing
// store load time). This would require either changing the backing store
// database schema to include the scheme (far more trouble than it's worth), or
// separating out file cookies into their own CookieMonster instance and
// thus restricting each scheme to a single cookie monster (which might
// be worth it, but is still too much trouble to solve what is currently a
// non-problem).
std::string CookieMonster::GetKey(const std::string& domain) const {
if (expiry_and_key_scheme_ == EKS_DISCARD_RECENT_AND_PURGE_DOMAIN)
return domain;
std::string effective_domain(
RegistryControlledDomainService::GetDomainAndRegistry(domain));
if (effective_domain.empty())
effective_domain = domain;
if (!effective_domain.empty() && effective_domain[0] == '.')
return effective_domain.substr(1);
return effective_domain;
}
// Determine the actual cookie domain based on the domain string passed
// (if any) and the URL from which the cookie came.
// On success returns true, and sets cookie_domain to either a
// -host cookie domain (ex: "google.com")
// -domain cookie domain (ex: ".google.com")
static bool GetCookieDomainWithString(const GURL& url,
const std::string& domain_string,
std::string* result) {
const std::string url_host(url.host());
// If no domain was specified in the domain string, default to a host cookie.
// We match IE/Firefox in allowing a domain=IPADDR if it matches the url
// ip address hostname exactly. It should be treated as a host cookie.
if (domain_string.empty() ||
(url.HostIsIPAddress() && url_host == domain_string)) {
*result = url_host;
DCHECK(CookieMonster::DomainIsHostOnly(*result));
return true;
}
// Get the normalized domain specified in cookie line.
// Note: The RFC says we can reject a cookie if the domain
// attribute does not start with a dot. IE/FF/Safari however, allow a cookie
// of the form domain=my.domain.com, treating it the same as
// domain=.my.domain.com -- for compatibility we do the same here. Firefox
// also treats domain=.....my.domain.com like domain=.my.domain.com, but
// neither IE nor Safari do this, and we don't either.
url_canon::CanonHostInfo ignored;
std::string cookie_domain(net::CanonicalizeHost(domain_string, &ignored));
if (cookie_domain.empty())
return false;
if (cookie_domain[0] != '.')
cookie_domain = "." + cookie_domain;
// Ensure |url| and |cookie_domain| have the same domain+registry.
const std::string url_scheme(url.scheme());
const std::string url_domain_and_registry(
GetEffectiveDomain(url_scheme, url_host));
if (url_domain_and_registry.empty())
return false; // IP addresses/intranet hosts can't set domain cookies.
const std::string cookie_domain_and_registry(
GetEffectiveDomain(url_scheme, cookie_domain));
if (url_domain_and_registry != cookie_domain_and_registry)
return false; // Can't set a cookie on a different domain + registry.
// Ensure |url_host| is |cookie_domain| or one of its subdomains. Given that
// we know the domain+registry are the same from the above checks, this is
// basically a simple string suffix check.
if ((url_host.length() < cookie_domain.length()) ?
(cookie_domain != ("." + url_host)) :
url_host.compare(url_host.length() - cookie_domain.length(),
cookie_domain.length(), cookie_domain))
return false;
*result = cookie_domain;
return true;
}
// Determine the cookie domain to use for setting the specified cookie.
static bool GetCookieDomain(const GURL& url,
const CookieMonster::ParsedCookie& pc,
std::string* result) {
std::string domain_string;
if (pc.HasDomain())
domain_string = pc.Domain();
return GetCookieDomainWithString(url, domain_string, result);
}
static std::string CanonPathWithString(const GURL& url,
const std::string& path_string) {
// The RFC says the path should be a prefix of the current URL path.
// However, Mozilla allows you to set any path for compatibility with
// broken websites. We unfortunately will mimic this behavior. We try
// to be generous and accept cookies with an invalid path attribute, and
// default the path to something reasonable.
// The path was supplied in the cookie, we'll take it.
if (!path_string.empty() && path_string[0] == '/')
return path_string;
// The path was not supplied in the cookie or invalid, we will default
// to the current URL path.
// """Defaults to the path of the request URL that generated the
// Set-Cookie response, up to, but not including, the
// right-most /."""
// How would this work for a cookie on /? We will include it then.
const std::string& url_path = url.path();
size_t idx = url_path.find_last_of('/');
// The cookie path was invalid or a single '/'.
if (idx == 0 || idx == std::string::npos)
return std::string("/");
// Return up to the rightmost '/'.
return url_path.substr(0, idx);
}
static std::string CanonPath(const GURL& url,
const CookieMonster::ParsedCookie& pc) {
std::string path_string;
if (pc.HasPath())
path_string = pc.Path();
return CanonPathWithString(url, path_string);
}
static Time CanonExpiration(const CookieMonster::ParsedCookie& pc,
const Time& current,
const CookieOptions& options) {
if (options.force_session())
return Time();
// First, try the Max-Age attribute.
uint64 max_age = 0;
if (pc.HasMaxAge() &&
#ifdef COMPILER_MSVC
sscanf_s(
#else
sscanf(
#endif
pc.MaxAge().c_str(), " %" PRIu64, &max_age) == 1) {
return current + TimeDelta::FromSeconds(max_age);
}
// Try the Expires attribute.
if (pc.HasExpires())
return CookieMonster::ParseCookieTime(pc.Expires());
// Invalid or no expiration, persistent cookie.
return Time();
}
bool CookieMonster::HasCookieableScheme(const GURL& url) {
lock_.AssertAcquired();
// Make sure the request is on a cookie-able url scheme.
for (size_t i = 0; i < cookieable_schemes_.size(); ++i) {
// We matched a scheme.
if (url.SchemeIs(cookieable_schemes_[i].c_str())) {
// We've matched a supported scheme.
return true;
}
}
// The scheme didn't match any in our whitelist.
DVLOG(1) << "WARNING: Unsupported cookie scheme: " << url.scheme();
return false;
}
void CookieMonster::SetCookieableSchemes(
const char* schemes[], size_t num_schemes) {
AutoLock autolock(lock_);
// Cookieable Schemes must be set before first use of function.
DCHECK(!initialized_);
cookieable_schemes_.clear();
cookieable_schemes_.insert(cookieable_schemes_.end(),
schemes, schemes + num_schemes);
}
bool CookieMonster::SetCookieWithCreationTimeAndOptions(
const GURL& url,
const std::string& cookie_line,
const Time& creation_time_or_null,
const CookieOptions& options) {
lock_.AssertAcquired();
DVLOG(1) << "SetCookie() line: " << cookie_line;
Time creation_time = creation_time_or_null;
if (creation_time.is_null()) {
creation_time = CurrentTime();
last_time_seen_ = creation_time;
}
// Parse the cookie.
ParsedCookie pc(cookie_line);
if (!pc.IsValid()) {
DVLOG(1) << "WARNING: Couldn't parse cookie";
return false;
}
if (options.exclude_httponly() && pc.IsHttpOnly()) {
DVLOG(1) << "SetCookie() not setting httponly cookie";
return false;
}
std::string cookie_domain;
if (!GetCookieDomain(url, pc, &cookie_domain)) {
return false;
}
std::string cookie_path = CanonPath(url, pc);
scoped_ptr<CanonicalCookie> cc;
Time cookie_expires = CanonExpiration(pc, creation_time, options);
cc.reset(new CanonicalCookie(pc.Name(), pc.Value(), cookie_domain,
cookie_path,
pc.IsSecure(), pc.IsHttpOnly(),
creation_time, creation_time,
!cookie_expires.is_null(), cookie_expires));
if (!cc.get()) {
DVLOG(1) << "WARNING: Failed to allocate CanonicalCookie";
return false;
}
return SetCanonicalCookie(&cc, creation_time, options);
}
bool CookieMonster::SetCookieWithCreationTime(const GURL& url,
const std::string& cookie_line,
const base::Time& creation_time) {
AutoLock autolock(lock_);
if (!HasCookieableScheme(url)) {
return false;
}
InitIfNecessary();
return SetCookieWithCreationTimeAndOptions(url, cookie_line, creation_time,
CookieOptions());
}
bool CookieMonster::SetCookieWithDetails(
const GURL& url, const std::string& name, const std::string& value,
const std::string& domain, const std::string& path,
const base::Time& expiration_time, bool secure, bool http_only) {
AutoLock autolock(lock_);
if (!HasCookieableScheme(url))
return false;
InitIfNecessary();
Time creation_time = CurrentTime();
last_time_seen_ = creation_time;
scoped_ptr<CanonicalCookie> cc;
cc.reset(CanonicalCookie::Create(
url, name, value, domain, path,
creation_time, expiration_time,
secure, http_only));
if (!cc.get())
return false;
CookieOptions options;
options.set_include_httponly();
return SetCanonicalCookie(&cc, creation_time, options);
}
bool CookieMonster::SetCanonicalCookie(scoped_ptr<CanonicalCookie>* cc,
const Time& creation_time,
const CookieOptions& options) {
const std::string key(GetKey((*cc)->Domain()));
if (DeleteAnyEquivalentCookie(key, **cc, options.exclude_httponly())) {
DVLOG(1) << "SetCookie() not clobbering httponly cookie";
return false;
}
DVLOG(1) << "SetCookie() key: " << key << " cc: " << (*cc)->DebugString();
// Realize that we might be setting an expired cookie, and the only point
// was to delete the cookie which we've already done.
if (!(*cc)->IsExpired(creation_time)) {
// See InitializeHistograms() for details.
histogram_expiration_duration_minutes_->Add(
((*cc)->ExpiryDate() - creation_time).InMinutes());
InternalInsertCookie(key, cc->release(), true);
}
// We assume that hopefully setting a cookie will be less common than
// querying a cookie. Since setting a cookie can put us over our limits,
// make sure that we garbage collect... We can also make the assumption that
// if a cookie was set, in the common case it will be used soon after,
// and we will purge the expired cookies in GetCookies().
GarbageCollect(creation_time, key);
return true;
}
void CookieMonster::InternalInsertCookie(const std::string& key,
CanonicalCookie* cc,
bool sync_to_store) {
lock_.AssertAcquired();
if (cc->IsPersistent() && store_ && sync_to_store)
store_->AddCookie(*cc);
cookies_.insert(CookieMap::value_type(key, cc));
if (delegate_.get())
delegate_->OnCookieChanged(*cc, false);
}
void CookieMonster::InternalUpdateCookieAccessTime(CanonicalCookie* cc,
const Time& current) {
lock_.AssertAcquired();
// Based off the Mozilla code. When a cookie has been accessed recently,
// don't bother updating its access time again. This reduces the number of
// updates we do during pageload, which in turn reduces the chance our storage
// backend will hit its batch thresholds and be forced to update.
if ((current - cc->LastAccessDate()) < last_access_threshold_)
return;
// See InitializeHistograms() for details.
histogram_between_access_interval_minutes_->Add(
(current - cc->LastAccessDate()).InMinutes());
cc->SetLastAccessDate(current);
if (cc->IsPersistent() && store_)
store_->UpdateCookieAccessTime(*cc);
}
void CookieMonster::InternalDeleteCookie(CookieMap::iterator it,
bool sync_to_store,
DeletionCause deletion_cause) {
lock_.AssertAcquired();
// See InitializeHistograms() for details.
if (deletion_cause != DELETE_COOKIE_DONT_RECORD)
histogram_cookie_deletion_cause_->Add(deletion_cause);
CanonicalCookie* cc = it->second;
DVLOG(1) << "InternalDeleteCookie() cc: " << cc->DebugString();
if (cc->IsPersistent() && store_ && sync_to_store)
store_->DeleteCookie(*cc);
if (delegate_.get())
delegate_->OnCookieChanged(*cc, true);
cookies_.erase(it);
delete cc;
}
bool CookieMonster::DeleteAnyEquivalentCookie(const std::string& key,
const CanonicalCookie& ecc,
bool skip_httponly) {
lock_.AssertAcquired();
bool found_equivalent_cookie = false;
bool skipped_httponly = false;
for (CookieMapItPair its = cookies_.equal_range(key);
its.first != its.second; ) {
CookieMap::iterator curit = its.first;
CanonicalCookie* cc = curit->second;
++its.first;
if (ecc.IsEquivalent(*cc)) {
// We should never have more than one equivalent cookie, since they should
// overwrite each other.
CHECK(!found_equivalent_cookie) <<
"Duplicate equivalent cookies found, cookie store is corrupted.";
if (skip_httponly && cc->IsHttpOnly()) {
skipped_httponly = true;
} else {
InternalDeleteCookie(curit, true, DELETE_COOKIE_OVERWRITE);
}
found_equivalent_cookie = true;
}
}
return skipped_httponly;
}
static bool LRUCookieSorter(const CookieMonster::CookieMap::iterator& it1,
const CookieMonster::CookieMap::iterator& it2) {
// Cookies accessed less recently should be deleted first.
if (it1->second->LastAccessDate() != it2->second->LastAccessDate())
return it1->second->LastAccessDate() < it2->second->LastAccessDate();
// In rare cases we might have two cookies with identical last access times.
// To preserve the stability of the sort, in these cases prefer to delete
// older cookies over newer ones. CreationDate() is guaranteed to be unique.
return it1->second->CreationDate() < it2->second->CreationDate();
}
// Helper for GarbageCollection. If |cookie_its->size() > num_max|, remove the
// |num_max - num_purge| most recently accessed cookies from cookie_its.
// (In other words, leave the entries that are candidates for
// eviction in cookie_its.) The cookies returned will be in order sorted by
// access time, least recently accessed first. The access time of the least
// recently accessed entry not returned will be placed in
// |*lra_removed| if that pointer is set. FindLeastRecentlyAccessed
// returns false if no manipulation is done (because the list size is less
// than num_max), true otherwise.
static bool FindLeastRecentlyAccessed(
size_t num_max,
size_t num_purge,
Time* lra_removed,
std::vector<CookieMonster::CookieMap::iterator>* cookie_its) {
DCHECK_LE(num_purge, num_max);
if (cookie_its->size() > num_max) {
DVLOG(1) << "FindLeastRecentlyAccessed() Deep Garbage Collect.";
num_purge += cookie_its->size() - num_max;
DCHECK_GT(cookie_its->size(), num_purge);
// Add 1 so that we can get the last time left in the store.
std::partial_sort(cookie_its->begin(), cookie_its->begin() + num_purge + 1,
cookie_its->end(), LRUCookieSorter);
*lra_removed =
(*(cookie_its->begin() + num_purge))->second->LastAccessDate();
cookie_its->erase(cookie_its->begin() + num_purge, cookie_its->end());
return true;
}
return false;
}
int CookieMonster::GarbageCollectDeleteList(
const Time& current,
const Time& keep_accessed_after,
DeletionCause cause,
std::vector<CookieMap::iterator>& cookie_its) {
int num_deleted = 0;
for (std::vector<CookieMap::iterator>::iterator it = cookie_its.begin();
it != cookie_its.end(); it++) {
if (keep_accessed_after.is_null() ||
(*it)->second->LastAccessDate() < keep_accessed_after) {
histogram_evicted_last_access_minutes_->Add(
(current - (*it)->second->LastAccessDate()).InMinutes());
InternalDeleteCookie((*it), true, cause);
num_deleted++;
}
}
return num_deleted;
}
// Domain expiry behavior is unchanged by key/expiry scheme (the
// meaning of the key is different, but that's not visible to this
// routine). Global garbage collection is dependent on key/expiry
// scheme in that recently touched cookies are not saved if
// expiry_and_key_scheme_ == EKS_DISCARD_RECENT_AND_PURGE_DOMAIN.
int CookieMonster::GarbageCollect(const Time& current,
const std::string& key) {
lock_.AssertAcquired();
int num_deleted = 0;
// Collect garbage for this key.
if (cookies_.count(key) > kDomainMaxCookies) {
DVLOG(1) << "GarbageCollect() key: " << key;
std::vector<CookieMap::iterator> cookie_its;
num_deleted += GarbageCollectExpired(
current, cookies_.equal_range(key), &cookie_its);
base::Time oldest_removed;
if (FindLeastRecentlyAccessed(kDomainMaxCookies, kDomainPurgeCookies,
&oldest_removed, &cookie_its)) {
// Delete in two passes so we can figure out what we're nuking
// that would be kept at the global level.
int num_subject_to_global_purge =
GarbageCollectDeleteList(
current,
Time::Now() - TimeDelta::FromDays(kSafeFromGlobalPurgeDays),
DELETE_COOKIE_EVICTED_DOMAIN_PRE_SAFE,
cookie_its);
num_deleted += num_subject_to_global_purge;
// Correct because FindLeastRecentlyAccessed returns a sorted list.
cookie_its.erase(cookie_its.begin(),
cookie_its.begin() + num_subject_to_global_purge);
num_deleted +=
GarbageCollectDeleteList(
current,
Time(),
DELETE_COOKIE_EVICTED_DOMAIN_POST_SAFE,
cookie_its);
}
}
// Collect garbage for everything. With firefox style we want to
// preserve cookies touched in kSafeFromGlobalPurgeDays, otherwise
// not.
if (cookies_.size() > kMaxCookies &&
(expiry_and_key_scheme_ == EKS_DISCARD_RECENT_AND_PURGE_DOMAIN ||
earliest_access_time_ <
Time::Now() - TimeDelta::FromDays(kSafeFromGlobalPurgeDays))) {
DVLOG(1) << "GarbageCollect() everything";
std::vector<CookieMap::iterator> cookie_its;
base::Time oldest_left;
num_deleted += GarbageCollectExpired(
current, CookieMapItPair(cookies_.begin(), cookies_.end()),
&cookie_its);
if (FindLeastRecentlyAccessed(kMaxCookies, kPurgeCookies,
&oldest_left, &cookie_its)) {
Time oldest_safe_cookie(
expiry_and_key_scheme_ == EKS_KEEP_RECENT_AND_PURGE_ETLDP1 ?
(Time::Now() - TimeDelta::FromDays(kSafeFromGlobalPurgeDays)) :
Time::Now());
int num_evicted = GarbageCollectDeleteList(
current,
oldest_safe_cookie,
DELETE_COOKIE_EVICTED_GLOBAL,
cookie_its);
// If no cookies were preserved by the time limit, the global last
// access is set to the value returned from FindLeastRecentlyAccessed.
// If the time limit preserved some cookies, we use the last access of
// the oldest preserved cookie.
if (num_evicted == static_cast<int>(cookie_its.size())) {
earliest_access_time_ = oldest_left;
} else {
earliest_access_time_ =
(*(cookie_its.begin() + num_evicted))->second->LastAccessDate();
}
num_deleted += num_evicted;
}
}
return num_deleted;
}
int CookieMonster::GarbageCollectExpired(
const Time& current,
const CookieMapItPair& itpair,
std::vector<CookieMap::iterator>* cookie_its) {
lock_.AssertAcquired();
int num_deleted = 0;
for (CookieMap::iterator it = itpair.first, end = itpair.second; it != end;) {
CookieMap::iterator curit = it;
++it;
if (curit->second->IsExpired(current)) {
InternalDeleteCookie(curit, true, DELETE_COOKIE_EXPIRED);
++num_deleted;
} else if (cookie_its) {
cookie_its->push_back(curit);
}
}
return num_deleted;
}
int CookieMonster::DeleteAll(bool sync_to_store) {
AutoLock autolock(lock_);
if (sync_to_store)
InitIfNecessary();
int num_deleted = 0;
for (CookieMap::iterator it = cookies_.begin(); it != cookies_.end();) {
CookieMap::iterator curit = it;
++it;
InternalDeleteCookie(curit, sync_to_store,
sync_to_store ? DELETE_COOKIE_EXPLICIT :
DELETE_COOKIE_DONT_RECORD /* Destruction. */);
++num_deleted;
}
return num_deleted;
}
int CookieMonster::DeleteAllCreatedBetween(const Time& delete_begin,
const Time& delete_end,
bool sync_to_store) {
AutoLock autolock(lock_);
InitIfNecessary();
int num_deleted = 0;
for (CookieMap::iterator it = cookies_.begin(); it != cookies_.end();) {
CookieMap::iterator curit = it;
CanonicalCookie* cc = curit->second;
++it;
if (cc->CreationDate() >= delete_begin &&
(delete_end.is_null() || cc->CreationDate() < delete_end)) {
InternalDeleteCookie(curit, sync_to_store, DELETE_COOKIE_EXPLICIT);
++num_deleted;
}
}
return num_deleted;
}
int CookieMonster::DeleteAllCreatedAfter(const Time& delete_begin,
bool sync_to_store) {
return DeleteAllCreatedBetween(delete_begin, Time(), sync_to_store);
}
int CookieMonster::DeleteAllForHost(const GURL& url) {
AutoLock autolock(lock_);
InitIfNecessary();
if (!HasCookieableScheme(url))
return 0;
const std::string scheme(url.scheme());
const std::string host(url.host());
// We store host cookies in the store by their canonical host name;
// domain cookies are stored with a leading ".". So this is a pretty
// simple lookup and per-cookie delete.
int num_deleted = 0;
for (CookieMapItPair its = cookies_.equal_range(GetKey(host));
its.first != its.second;) {
CookieMap::iterator curit = its.first;
++its.first;
const CanonicalCookie* const cc = curit->second;
// Delete only on a match as a host cookie.
if (cc->IsHostCookie() && cc->IsDomainMatch(scheme, host)) {
num_deleted++;
InternalDeleteCookie(curit, true, DELETE_COOKIE_EXPLICIT);
}
}
return num_deleted;
}
bool CookieMonster::DeleteCanonicalCookie(const CanonicalCookie& cookie) {
AutoLock autolock(lock_);
InitIfNecessary();
for (CookieMapItPair its = cookies_.equal_range(GetKey(cookie.Domain()));
its.first != its.second; ++its.first) {
// The creation date acts as our unique index...
if (its.first->second->CreationDate() == cookie.CreationDate()) {
InternalDeleteCookie(its.first, true, DELETE_COOKIE_EXPLICIT);
return true;
}
}
return false;
}
// Mozilla sorts on the path length (longest first), and then it
// sorts by creation time (oldest first).
// The RFC says the sort order for the domain attribute is undefined.
static bool CookieSorter(CookieMonster::CanonicalCookie* cc1,
CookieMonster::CanonicalCookie* cc2) {
if (cc1->Path().length() == cc2->Path().length())
return cc1->CreationDate() < cc2->CreationDate();
return cc1->Path().length() > cc2->Path().length();
}
bool CookieMonster::SetCookieWithOptions(const GURL& url,
const std::string& cookie_line,
const CookieOptions& options) {
AutoLock autolock(lock_);
if (!HasCookieableScheme(url)) {
return false;
}
InitIfNecessary();
return SetCookieWithCreationTimeAndOptions(url, cookie_line, Time(), options);
}
std::string CookieMonster::GetCookiesWithOptions(const GURL& url,
const CookieOptions& options) {
AutoLock autolock(lock_);
InitIfNecessary();
if (!HasCookieableScheme(url)) {
return std::string();
}
TimeTicks start_time(TimeTicks::Now());
// Get the cookies for this host and its domain(s).
std::vector<CanonicalCookie*> cookies;
FindCookiesForHostAndDomain(url, options, true, &cookies);
std::sort(cookies.begin(), cookies.end(), CookieSorter);
std::string cookie_line;
for (std::vector<CanonicalCookie*>::const_iterator it = cookies.begin();
it != cookies.end(); ++it) {
if (it != cookies.begin())
cookie_line += "; ";
// In Mozilla if you set a cookie like AAAA, it will have an empty token
// and a value of AAAA. When it sends the cookie back, it will send AAAA,
// so we need to avoid sending =AAAA for a blank token value.
if (!(*it)->Name().empty())
cookie_line += (*it)->Name() + "=";
cookie_line += (*it)->Value();
}
histogram_time_get_->AddTime(TimeTicks::Now() - start_time);
DVLOG(1) << "GetCookies() result: " << cookie_line;
return cookie_line;
}
void CookieMonster::DeleteCookie(const GURL& url,
const std::string& cookie_name) {
AutoLock autolock(lock_);
InitIfNecessary();
if (!HasCookieableScheme(url))
return;
CookieOptions options;
options.set_include_httponly();
// Get the cookies for this host and its domain(s).
std::vector<CanonicalCookie*> cookies;
FindCookiesForHostAndDomain(url, options, true, &cookies);
std::set<CanonicalCookie*> matching_cookies;
for (std::vector<CanonicalCookie*>::const_iterator it = cookies.begin();
it != cookies.end(); ++it) {
if ((*it)->Name() != cookie_name)
continue;
if (url.path().find((*it)->Path()))
continue;
matching_cookies.insert(*it);
}
for (CookieMap::iterator it = cookies_.begin(); it != cookies_.end();) {
CookieMap::iterator curit = it;
++it;
if (matching_cookies.find(curit->second) != matching_cookies.end()) {
InternalDeleteCookie(curit, true, DELETE_COOKIE_EXPLICIT);
}
}
}
CookieMonster::CookieList CookieMonster::GetAllCookies() {
AutoLock autolock(lock_);
InitIfNecessary();
// This function is being called to scrape the cookie list for management UI
// or similar. We shouldn't show expired cookies in this list since it will
// just be confusing to users, and this function is called rarely enough (and
// is already slow enough) that it's OK to take the time to garbage collect
// the expired cookies now.
//
// Note that this does not prune cookies to be below our limits (if we've
// exceeded them) the way that calling GarbageCollect() would.
GarbageCollectExpired(Time::Now(),
CookieMapItPair(cookies_.begin(), cookies_.end()),
NULL);
// Copy the CanonicalCookie pointers from the map so that we can use the same
// sorter as elsewhere, then copy the result out.
std::vector<CanonicalCookie*> cookie_ptrs;
cookie_ptrs.reserve(cookies_.size());
for (CookieMap::iterator it = cookies_.begin(); it != cookies_.end(); ++it)
cookie_ptrs.push_back(it->second);
std::sort(cookie_ptrs.begin(), cookie_ptrs.end(), CookieSorter);
CookieList cookie_list;
cookie_list.reserve(cookie_ptrs.size());
for (std::vector<CanonicalCookie*>::const_iterator it = cookie_ptrs.begin();
it != cookie_ptrs.end(); ++it)
cookie_list.push_back(**it);
return cookie_list;
}
CookieMonster::CookieList CookieMonster::GetAllCookiesForURL(const GURL& url) {
AutoLock autolock(lock_);
InitIfNecessary();
CookieOptions options;
options.set_include_httponly();
std::vector<CanonicalCookie*> cookie_ptrs;
FindCookiesForHostAndDomain(url, options, false, &cookie_ptrs);
std::sort(cookie_ptrs.begin(), cookie_ptrs.end(), CookieSorter);
CookieList cookies;
for (std::vector<CanonicalCookie*>::const_iterator it = cookie_ptrs.begin();
it != cookie_ptrs.end(); it++)
cookies.push_back(**it);
return cookies;
}
void CookieMonster::FindCookiesForHostAndDomain(
const GURL& url,
const CookieOptions& options,
bool update_access_time,
std::vector<CanonicalCookie*>* cookies) {
lock_.AssertAcquired();
const Time current_time(CurrentTime());
// Probe to save statistics relatively frequently. We do it here rather
// than in the set path as many websites won't set cookies, and we
// want to collect statistics whenever the browser's being used.
RecordPeriodicStats(current_time);
if (expiry_and_key_scheme_ == EKS_DISCARD_RECENT_AND_PURGE_DOMAIN) {
// Can just dispatch to FindCookiesForKey
const std::string key(GetKey(url.host()));
FindCookiesForKey(key, url, options, current_time,
update_access_time, cookies);
} else {
// Need to probe for all domains that might have relevant
// cookies for us.
// Query for the full host, For example: 'a.c.blah.com'.
std::string key(GetKey(url.host()));
FindCookiesForKey(key, url, options, current_time, update_access_time,
cookies);
// See if we can search for domain cookies, i.e. if the host has a TLD + 1.
const std::string domain(GetEffectiveDomain(url.scheme(), key));
if (domain.empty())
return;
DCHECK_LE(domain.length(), key.length());
DCHECK_EQ(0, key.compare(key.length() - domain.length(), domain.length(),
domain));
// Walk through the string and query at the dot points (GURL should have
// canonicalized the dots, so this should be safe). Stop once we reach the
// domain + registry; we can't write cookies past this point, and with some
// registrars other domains can, in which case we don't want to read their
// cookies.
for (key = "." + key; key.length() > domain.length(); ) {
FindCookiesForKey(key, url, options, current_time, update_access_time,
cookies);
const size_t next_dot = key.find('.', 1); // Skip over leading dot.
key.erase(0, next_dot);
}
}
}
void CookieMonster::FindCookiesForKey(
const std::string& key,
const GURL& url,
const CookieOptions& options,
const Time& current,
bool update_access_time,
std::vector<CanonicalCookie*>* cookies) {
lock_.AssertAcquired();
const std::string scheme(url.scheme());
const std::string host(url.host());
bool secure = url.SchemeIsSecure();
for (CookieMapItPair its = cookies_.equal_range(key);
its.first != its.second; ) {
CookieMap::iterator curit = its.first;
CanonicalCookie* cc = curit->second;
++its.first;
// If the cookie is expired, delete it.
if (cc->IsExpired(current)) {
InternalDeleteCookie(curit, true, DELETE_COOKIE_EXPIRED);
continue;
}
// Filter out HttpOnly cookies, per options.
if (options.exclude_httponly() && cc->IsHttpOnly())
continue;
// Filter out secure cookies unless we're https.
if (!secure && cc->IsSecure())
continue;
// Filter out cookies that don't apply to this domain.
if (expiry_and_key_scheme_ == EKS_KEEP_RECENT_AND_PURGE_ETLDP1
&& !cc->IsDomainMatch(scheme, host))
continue;
if (!cc->IsOnPath(url.path()))
continue;
// Add this cookie to the set of matching cookies. Update the access
// time if we've been requested to do so.
if (update_access_time) {
InternalUpdateCookieAccessTime(cc, current);
}
cookies->push_back(cc);
}
}
// Test to see if stats should be recorded, and record them if so.
// The goal here is to get sampling for the average browser-hour of
// activity. We won't take samples when the web isn't being surfed,
// and when the web is being surfed, we'll take samples about every
// kRecordStatisticsIntervalSeconds.
// last_statistic_record_time_ is initialized to Now() rather than null
// in the constructor so that we won't take statistics right after
// startup, to avoid bias from browsers that are started but not used.
void CookieMonster::RecordPeriodicStats(const base::Time& current_time) {
const base::TimeDelta kRecordStatisticsIntervalTime(
base::TimeDelta::FromSeconds(kRecordStatisticsIntervalSeconds));
// If we've taken statistics recently, return.
if (current_time - last_statistic_record_time_ <=
kRecordStatisticsIntervalTime) {
return;
}
// See InitializeHistograms() for details.
histogram_count_->Add(cookies_.size());
// More detailed statistics on cookie counts at different granularities.
TimeTicks beginning_of_time(TimeTicks::Now());
for (CookieMap::const_iterator it_key = cookies_.begin();
it_key != cookies_.end(); ) {
const std::string& key(it_key->first);
int key_count = 0;
typedef std::map<std::string, unsigned int> DomainMap;
DomainMap domain_map;
CookieMapItPair its_cookies = cookies_.equal_range(key);
while (its_cookies.first != its_cookies.second) {
key_count++;
const std::string& cookie_domain(its_cookies.first->second->Domain());
domain_map[cookie_domain]++;
its_cookies.first++;
}
histogram_etldp1_count_->Add(key_count);
histogram_domain_per_etldp1_count_->Add(domain_map.size());
for (DomainMap::const_iterator domain_map_it = domain_map.begin();
domain_map_it != domain_map.end(); domain_map_it++)
histogram_domain_count_->Add(domain_map_it->second);
it_key = its_cookies.second;
}
DVLOG(1) << "Time for recording cookie stats (us): "
<< (TimeTicks::Now() - beginning_of_time).InMicroseconds();
last_statistic_record_time_ = current_time;
}
CookieMonster::ParsedCookie::ParsedCookie(const std::string& cookie_line)
: is_valid_(false),
path_index_(0),
domain_index_(0),
expires_index_(0),
maxage_index_(0),
secure_index_(0),
httponly_index_(0) {
if (cookie_line.size() > kMaxCookieSize) {
VLOG(1) << "Not parsing cookie, too large: " << cookie_line.size();
return;
}
ParseTokenValuePairs(cookie_line);
if (pairs_.size() > 0) {
is_valid_ = true;
SetupAttributes();
}
}
CookieMonster::ParsedCookie::~ParsedCookie() {
}
// Returns true if |c| occurs in |chars|
// TODO maybe make this take an iterator, could check for end also?
static inline bool CharIsA(const char c, const char* chars) {
return strchr(chars, c) != NULL;
}
// Seek the iterator to the first occurrence of a character in |chars|.
// Returns true if it hit the end, false otherwise.
static inline bool SeekTo(std::string::const_iterator* it,
const std::string::const_iterator& end,
const char* chars) {
for (; *it != end && !CharIsA(**it, chars); ++(*it)) {}
return *it == end;
}
// Seek the iterator to the first occurrence of a character not in |chars|.
// Returns true if it hit the end, false otherwise.
static inline bool SeekPast(std::string::const_iterator* it,
const std::string::const_iterator& end,
const char* chars) {
for (; *it != end && CharIsA(**it, chars); ++(*it)) {}
return *it == end;
}
static inline bool SeekBackPast(std::string::const_iterator* it,
const std::string::const_iterator& end,
const char* chars) {
for (; *it != end && CharIsA(**it, chars); --(*it)) {}
return *it == end;
}
const char CookieMonster::ParsedCookie::kTerminator[] = "\n\r\0";
const int CookieMonster::ParsedCookie::kTerminatorLen =
sizeof(kTerminator) - 1;
const char CookieMonster::ParsedCookie::kWhitespace[] = " \t";
const char CookieMonster::ParsedCookie::kValueSeparator[] = ";";
const char CookieMonster::ParsedCookie::kTokenSeparator[] = ";=";
std::string::const_iterator CookieMonster::ParsedCookie::FindFirstTerminator(
const std::string& s) {
std::string::const_iterator end = s.end();
size_t term_pos =
s.find_first_of(std::string(kTerminator, kTerminatorLen));
if (term_pos != std::string::npos) {
// We found a character we should treat as an end of string.
end = s.begin() + term_pos;
}
return end;
}
bool CookieMonster::ParsedCookie::ParseToken(
std::string::const_iterator* it,
const std::string::const_iterator& end,
std::string::const_iterator* token_start,
std::string::const_iterator* token_end) {
DCHECK(it && token_start && token_end);
std::string::const_iterator token_real_end;
// Seek past any whitespace before the "token" (the name).
// token_start should point at the first character in the token
if (SeekPast(it, end, kWhitespace))
return false; // No token, whitespace or empty.
*token_start = *it;
// Seek over the token, to the token separator.
// token_real_end should point at the token separator, i.e. '='.
// If it == end after the seek, we probably have a token-value.
SeekTo(it, end, kTokenSeparator);
token_real_end = *it;
// Ignore any whitespace between the token and the token separator.
// token_end should point after the last interesting token character,
// pointing at either whitespace, or at '=' (and equal to token_real_end).
if (*it != *token_start) { // We could have an empty token name.
--(*it); // Go back before the token separator.
// Skip over any whitespace to the first non-whitespace character.
SeekBackPast(it, *token_start, kWhitespace);
// Point after it.
++(*it);
}
*token_end = *it;
// Seek us back to the end of the token.
*it = token_real_end;
return true;
}
void CookieMonster::ParsedCookie::ParseValue(
std::string::const_iterator* it,
const std::string::const_iterator& end,
std::string::const_iterator* value_start,
std::string::const_iterator* value_end) {
DCHECK(it && value_start && value_end);
// Seek past any whitespace that might in-between the token and value.
SeekPast(it, end, kWhitespace);
// value_start should point at the first character of the value.
*value_start = *it;
// It is unclear exactly how quoted string values should be handled.
// Major browsers do different things, for example, Firefox supports
// semicolons embedded in a quoted value, while IE does not. Looking at
// the specs, RFC 2109 and 2965 allow for a quoted-string as the value.
// However, these specs were apparently written after browsers had
// implemented cookies, and they seem very distant from the reality of
// what is actually implemented and used on the web. The original spec
// from Netscape is possibly what is closest to the cookies used today.
// This spec didn't have explicit support for double quoted strings, and
// states that ; is not allowed as part of a value. We had originally
// implement the Firefox behavior (A="B;C"; -> A="B;C";). However, since
// there is no standard that makes sense, we decided to follow the behavior
// of IE and Safari, which is closer to the original Netscape proposal.
// This means that A="B;C" -> A="B;. This also makes the code much simpler
// and reduces the possibility for invalid cookies, where other browsers
// like Opera currently reject those invalid cookies (ex A="B" "C";).
// Just look for ';' to terminate ('=' allowed).
// We can hit the end, maybe they didn't terminate.
SeekTo(it, end, kValueSeparator);
// Will be pointed at the ; seperator or the end.
*value_end = *it;
// Ignore any unwanted whitespace after the value.
if (*value_end != *value_start) { // Could have an empty value
--(*value_end);
SeekBackPast(value_end, *value_start, kWhitespace);
++(*value_end);
}
}
std::string CookieMonster::ParsedCookie::ParseTokenString(
const std::string& token) {
std::string::const_iterator it = token.begin();
std::string::const_iterator end = FindFirstTerminator(token);
std::string::const_iterator token_start, token_end;
if (ParseToken(&it, end, &token_start, &token_end))
return std::string(token_start, token_end);
return std::string();
}
std::string CookieMonster::ParsedCookie::ParseValueString(
const std::string& value) {
std::string::const_iterator it = value.begin();
std::string::const_iterator end = FindFirstTerminator(value);
std::string::const_iterator value_start, value_end;
ParseValue(&it, end, &value_start, &value_end);
return std::string(value_start, value_end);
}
// Parse all token/value pairs and populate pairs_.
void CookieMonster::ParsedCookie::ParseTokenValuePairs(
const std::string& cookie_line) {
pairs_.clear();
// Ok, here we go. We should be expecting to be starting somewhere
// before the cookie line, not including any header name...
std::string::const_iterator start = cookie_line.begin();
std::string::const_iterator it = start;
// TODO Make sure we're stripping \r\n in the network code. Then we
// can log any unexpected terminators.
std::string::const_iterator end = FindFirstTerminator(cookie_line);
for (int pair_num = 0; pair_num < kMaxPairs && it != end; ++pair_num) {
TokenValuePair pair;
std::string::const_iterator token_start, token_end;
if (!ParseToken(&it, end, &token_start, &token_end))
break;
if (it == end || *it != '=') {
// We have a token-value, we didn't have any token name.
if (pair_num == 0) {
// For the first time around, we want to treat single values
// as a value with an empty name. (Mozilla bug 169091).
// IE seems to also have this behavior, ex "AAA", and "AAA=10" will
// set 2 different cookies, and setting "BBB" will then replace "AAA".
pair.first = "";
// Rewind to the beginning of what we thought was the token name,
// and let it get parsed as a value.
it = token_start;
} else {
// Any not-first attribute we want to treat a value as a
// name with an empty value... This is so something like
// "secure;" will get parsed as a Token name, and not a value.
pair.first = std::string(token_start, token_end);
}
} else {
// We have a TOKEN=VALUE.
pair.first = std::string(token_start, token_end);
++it; // Skip past the '='.
}
// OK, now try to parse a value.
std::string::const_iterator value_start, value_end;
ParseValue(&it, end, &value_start, &value_end);
// OK, we're finished with a Token/Value.
pair.second = std::string(value_start, value_end);
// From RFC2109: "Attributes (names) (attr) are case-insensitive."
if (pair_num != 0)
StringToLowerASCII(&pair.first);
pairs_.push_back(pair);
// We've processed a token/value pair, we're either at the end of
// the string or a ValueSeparator like ';', which we want to skip.
if (it != end)
++it;
}
}
void CookieMonster::ParsedCookie::SetupAttributes() {
static const char kPathTokenName[] = "path";
static const char kDomainTokenName[] = "domain";
static const char kExpiresTokenName[] = "expires";
static const char kMaxAgeTokenName[] = "max-age";
static const char kSecureTokenName[] = "secure";
static const char kHttpOnlyTokenName[] = "httponly";
// We skip over the first token/value, the user supplied one.
for (size_t i = 1; i < pairs_.size(); ++i) {
if (pairs_[i].first == kPathTokenName) {
path_index_ = i;
} else if (pairs_[i].first == kDomainTokenName) {
domain_index_ = i;
} else if (pairs_[i].first == kExpiresTokenName) {
expires_index_ = i;
} else if (pairs_[i].first == kMaxAgeTokenName) {
maxage_index_ = i;
} else if (pairs_[i].first == kSecureTokenName) {
secure_index_ = i;
} else if (pairs_[i].first == kHttpOnlyTokenName) {
httponly_index_ = i;
} else {
/* some attribute we don't know or don't care about. */
}
}
}
// Create a cookie-line for the cookie. For debugging only!
// If we want to use this for something more than debugging, we
// should rewrite it better...
std::string CookieMonster::ParsedCookie::DebugString() const {
std::string out;
for (PairList::const_iterator it = pairs_.begin();
it != pairs_.end(); ++it) {
out.append(it->first);
out.append("=");
out.append(it->second);
out.append("; ");
}
return out;
}
CookieMonster::CanonicalCookie::CanonicalCookie() {
}
CookieMonster::CanonicalCookie::CanonicalCookie(const std::string& name,
const std::string& value,
const std::string& domain,
const std::string& path,
bool secure,
bool httponly,
const base::Time& creation,
const base::Time& last_access,
bool has_expires,
const base::Time& expires)
: name_(name),
value_(value),
domain_(domain),
path_(path),
creation_date_(creation),
last_access_date_(last_access),
expiry_date_(expires),
has_expires_(has_expires),
secure_(secure),
httponly_(httponly) {
}
CookieMonster::CanonicalCookie::CanonicalCookie(const GURL& url,
const ParsedCookie& pc)
: name_(pc.Name()),
value_(pc.Value()),
path_(CanonPath(url, pc)),
creation_date_(Time::Now()),
last_access_date_(Time()),
has_expires_(pc.HasExpires()),
secure_(pc.IsSecure()),
httponly_(pc.IsHttpOnly()) {
if (has_expires_)
expiry_date_ = CanonExpiration(pc, creation_date_, CookieOptions());
// Do the best we can with the domain.
std::string cookie_domain;
std::string domain_string;
if (pc.HasDomain()) {
domain_string = pc.Domain();
}
bool result
= GetCookieDomainWithString(url, domain_string,
&cookie_domain);
// Caller is responsible for passing in good arguments.
DCHECK(result);
domain_ = cookie_domain;
}
CookieMonster::CookieMonster(PersistentCookieStore* store,
Delegate* delegate,
int last_access_threshold_milliseconds)
: initialized_(false),
expiry_and_key_scheme_(expiry_and_key_default_),
store_(store),
last_access_threshold_(base::TimeDelta::FromMilliseconds(
last_access_threshold_milliseconds)),
delegate_(delegate),
last_statistic_record_time_(base::Time::Now()) {
InitializeHistograms();
SetDefaultCookieableSchemes();
}
CookieMonster::CanonicalCookie::~CanonicalCookie() {
}
CookieMonster::CanonicalCookie* CookieMonster::CanonicalCookie::Create(
const GURL& url, const std::string& name, const std::string& value,
const std::string& domain, const std::string& path,
const base::Time& creation_time, const base::Time& expiration_time,
bool secure, bool http_only) {
// Expect valid attribute tokens and values, as defined by the ParsedCookie
// logic, otherwise don't create the cookie.
std::string parsed_name = ParsedCookie::ParseTokenString(name);
if (parsed_name != name)
return NULL;
std::string parsed_value = ParsedCookie::ParseValueString(value);
if (parsed_value != value)
return NULL;
std::string parsed_domain = ParsedCookie::ParseValueString(domain);
if (parsed_domain != domain)
return NULL;
std::string cookie_domain;
if (!GetCookieDomainWithString(url, parsed_domain, &cookie_domain))
return NULL;
std::string parsed_path = ParsedCookie::ParseValueString(path);
if (parsed_path != path)
return NULL;
std::string cookie_path = CanonPathWithString(url, parsed_path);
// Expect that the path was either not specified (empty), or is valid.
if (!parsed_path.empty() && cookie_path != parsed_path)
return NULL;
// Canonicalize path again to make sure it escapes characters as needed.
url_parse::Component path_component(0, cookie_path.length());
url_canon::RawCanonOutputT<char> canon_path;
url_parse::Component canon_path_component;
url_canon::CanonicalizePath(cookie_path.data(), path_component,
&canon_path, &canon_path_component);
cookie_path = std::string(canon_path.data() + canon_path_component.begin,
canon_path_component.len);
return new CanonicalCookie(parsed_name, parsed_value, cookie_domain,
cookie_path, secure, http_only,
creation_time, creation_time,
!expiration_time.is_null(), expiration_time);
}
bool CookieMonster::CanonicalCookie::IsOnPath(
const std::string& url_path) const {
// A zero length would be unsafe for our trailing '/' checks, and
// would also make no sense for our prefix match. The code that
// creates a CanonicalCookie should make sure the path is never zero length,
// but we double check anyway.
if (path_.empty())
return false;
// The Mozilla code broke it into 3 cases, if it's strings lengths
// are less than, equal, or greater. I think this is simpler:
// Make sure the cookie path is a prefix of the url path. If the
// url path is shorter than the cookie path, then the cookie path
// can't be a prefix.
if (url_path.find(path_) != 0)
return false;
// Now we know that url_path is >= cookie_path, and that cookie_path
// is a prefix of url_path. If they are the are the same length then
// they are identical, otherwise we need an additional check:
// In order to avoid in correctly matching a cookie path of /blah
// with a request path of '/blahblah/', we need to make sure that either
// the cookie path ends in a trailing '/', or that we prefix up to a '/'
// in the url path. Since we know that the url path length is greater
// than the cookie path length, it's safe to index one byte past.
if (path_.length() != url_path.length() &&
path_[path_.length() - 1] != '/' &&
url_path[path_.length()] != '/')
return false;
return true;
}
bool CookieMonster::CanonicalCookie::IsDomainMatch(
const std::string& scheme,
const std::string& host) const {
// Can domain match in two ways; as a domain cookie (where the cookie
// domain begins with ".") or as a host cookie (where it doesn't).
// Some consumers of the CookieMonster expect to set cookies on
// URLs like http://.strange.url. To retrieve cookies in this instance,
// we allow matching as a host cookie even when the domain_ starts with
// a period.
if (host == domain_)
return true;
// Domain cookie must have an initial ".". To match, it must be
// equal to url's host with initial period removed, or a suffix of
// it.
// Arguably this should only apply to "http" or "https" cookies, but
// extension cookie tests currently use the funtionality, and if we
// ever decide to implement that it should be done by preventing
// such cookies from being set.
if (domain_.empty() || domain_[0] != '.')
return false;
// The host with a "." prefixed.
if (domain_.compare(1, std::string::npos, host) == 0)
return true;
// A pure suffix of the host (ok since we know the domain already
// starts with a ".")
return (host.length() > domain_.length() &&
host.compare(host.length() - domain_.length(),
domain_.length(), domain_) == 0);
}
std::string CookieMonster::CanonicalCookie::DebugString() const {
return base::StringPrintf(
"name: %s value: %s domain: %s path: %s creation: %"
PRId64,
name_.c_str(), value_.c_str(),
domain_.c_str(), path_.c_str(),
static_cast<int64>(creation_date_.ToTimeT()));
}
} // namespace
|