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
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2039
2040
2041
2042
2043
2044
2045
2046
2047
2048
2049
2050
2051
2052
2053
2054
2055
2056
2057
2058
2059
2060
2061
2062
2063
2064
2065
2066
2067
2068
2069
2070
2071
2072
2073
2074
2075
2076
2077
2078
2079
2080
2081
2082
2083
2084
2085
2086
2087
2088
2089
2090
2091
2092
2093
2094
2095
2096
2097
2098
2099
2100
2101
2102
2103
2104
2105
2106
2107
2108
2109
2110
2111
2112
2113
2114
2115
2116
2117
2118
2119
2120
2121
2122
2123
2124
2125
2126
2127
2128
2129
2130
2131
2132
2133
2134
2135
2136
2137
2138
2139
2140
2141
2142
2143
2144
2145
2146
2147
2148
2149
2150
2151
2152
2153
2154
2155
2156
2157
2158
2159
2160
2161
2162
2163
2164
2165
2166
2167
2168
2169
2170
2171
2172
2173
2174
2175
2176
2177
2178
2179
2180
2181
2182
2183
2184
2185
2186
2187
2188
2189
2190
2191
2192
2193
2194
2195
2196
2197
2198
2199
2200
2201
2202
2203
2204
2205
2206
2207
2208
2209
2210
2211
2212
2213
2214
2215
2216
2217
2218
2219
2220
2221
2222
2223
2224
2225
2226
2227
2228
2229
2230
2231
2232
2233
2234
2235
2236
2237
2238
|
/*-
* Copyright (c) 2003-2007 Tim Kientzle
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
*
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR(S) ``AS IS'' AND ANY EXPRESS OR
* IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
* OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
* IN NO EVENT SHALL THE AUTHOR(S) BE LIABLE FOR ANY DIRECT, INDIRECT,
* INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
* NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
* DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
* THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
* (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
* THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*/
#include "archive_platform.h"
__FBSDID("$FreeBSD$");
#ifdef HAVE_ERRNO_H
#include <errno.h>
#endif
#include <stddef.h>
/* #include <stdint.h> */ /* See archive_platform.h */
#ifdef HAVE_STDLIB_H
#include <stdlib.h>
#endif
#ifdef HAVE_STRING_H
#include <string.h>
#endif
/* Obtain suitable wide-character manipulation functions. */
#ifdef HAVE_WCHAR_H
#include <wchar.h>
#else
/* Good enough for equality testing, which is all we need. */
static int wcscmp(const wchar_t *s1, const wchar_t *s2)
{
int diff = *s1 - *s2;
while (*s1 && diff == 0)
diff = (int)*++s1 - (int)*++s2;
return diff;
}
/* Good enough for equality testing, which is all we need. */
static int wcsncmp(const wchar_t *s1, const wchar_t *s2, size_t n)
{
int diff = *s1 - *s2;
while (*s1 && diff == 0 && n-- > 0)
diff = (int)*++s1 - (int)*++s2;
return diff;
}
static size_t wcslen(const wchar_t *s)
{
const wchar_t *p = s;
while (*p)
p++;
return p - s;
}
#endif
#include "archive.h"
#include "archive_entry.h"
#include "archive_private.h"
#include "archive_read_private.h"
/*
* Layout of POSIX 'ustar' tar header.
*/
struct archive_entry_header_ustar {
char name[100];
char mode[8];
char uid[8];
char gid[8];
char size[12];
char mtime[12];
char checksum[8];
char typeflag[1];
char linkname[100]; /* "old format" header ends here */
char magic[6]; /* For POSIX: "ustar\0" */
char version[2]; /* For POSIX: "00" */
char uname[32];
char gname[32];
char rdevmajor[8];
char rdevminor[8];
char prefix[155];
};
/*
* Structure of GNU tar header
*/
struct gnu_sparse {
char offset[12];
char numbytes[12];
};
struct archive_entry_header_gnutar {
char name[100];
char mode[8];
char uid[8];
char gid[8];
char size[12];
char mtime[12];
char checksum[8];
char typeflag[1];
char linkname[100];
char magic[8]; /* "ustar \0" (note blank/blank/null at end) */
char uname[32];
char gname[32];
char rdevmajor[8];
char rdevminor[8];
char atime[12];
char ctime[12];
char offset[12];
char longnames[4];
char unused[1];
struct gnu_sparse sparse[4];
char isextended[1];
char realsize[12];
/*
* GNU doesn't use POSIX 'prefix' field; they use the 'L' (longname)
* entry instead.
*/
};
/*
* Data specific to this format.
*/
struct sparse_block {
struct sparse_block *next;
off_t offset;
off_t remaining;
};
struct tar {
struct archive_string acl_text;
struct archive_string entry_name;
struct archive_string entry_linkname;
struct archive_string entry_uname;
struct archive_string entry_gname;
struct archive_string longlink;
struct archive_string longname;
struct archive_string pax_header;
struct archive_string pax_global;
struct archive_string line;
wchar_t *pax_entry;
size_t pax_entry_length;
int header_recursion_depth;
off_t entry_bytes_remaining;
off_t entry_offset;
off_t entry_padding;
off_t realsize;
struct sparse_block *sparse_list;
struct sparse_block *sparse_last;
int64_t sparse_offset;
int64_t sparse_numbytes;
int sparse_gnu_major;
int sparse_gnu_minor;
char sparse_gnu_pending;
};
static size_t UTF8_mbrtowc(wchar_t *pwc, const char *s, size_t n);
static int archive_block_is_null(const unsigned char *p);
static char *base64_decode(const wchar_t *, size_t, size_t *);
static void gnu_add_sparse_entry(struct tar *,
off_t offset, off_t remaining);
static int gnu_sparse_old_read(struct archive_read *, struct tar *,
const struct archive_entry_header_gnutar *header);
static void gnu_sparse_old_parse(struct tar *,
const struct gnu_sparse *sparse, int length);
static int gnu_sparse_01_parse(struct tar *, const wchar_t *);
static ssize_t gnu_sparse_10_read(struct archive_read *, struct tar *);
static int header_Solaris_ACL(struct archive_read *, struct tar *,
struct archive_entry *, const void *);
static int header_common(struct archive_read *, struct tar *,
struct archive_entry *, const void *);
static int header_old_tar(struct archive_read *, struct tar *,
struct archive_entry *, const void *);
static int header_pax_extensions(struct archive_read *, struct tar *,
struct archive_entry *, const void *);
static int header_pax_global(struct archive_read *, struct tar *,
struct archive_entry *, const void *h);
static int header_longlink(struct archive_read *, struct tar *,
struct archive_entry *, const void *h);
static int header_longname(struct archive_read *, struct tar *,
struct archive_entry *, const void *h);
static int header_volume(struct archive_read *, struct tar *,
struct archive_entry *, const void *h);
static int header_ustar(struct archive_read *, struct tar *,
struct archive_entry *, const void *h);
static int header_gnutar(struct archive_read *, struct tar *,
struct archive_entry *, const void *h);
static int archive_read_format_tar_bid(struct archive_read *);
static int archive_read_format_tar_cleanup(struct archive_read *);
static int archive_read_format_tar_read_data(struct archive_read *a,
const void **buff, size_t *size, off_t *offset);
static int archive_read_format_tar_skip(struct archive_read *a);
static int archive_read_format_tar_read_header(struct archive_read *,
struct archive_entry *);
static int checksum(struct archive_read *, const void *);
static int pax_attribute(struct tar *, struct archive_entry *,
wchar_t *key, wchar_t *value);
static int pax_header(struct archive_read *, struct tar *,
struct archive_entry *, char *attr);
static void pax_time(const wchar_t *, int64_t *sec, long *nanos);
static ssize_t readline(struct archive_read *, struct tar *, const char **);
static int read_body_to_string(struct archive_read *, struct tar *,
struct archive_string *, const void *h);
static int64_t tar_atol(const char *, unsigned);
static int64_t tar_atol10(const wchar_t *, unsigned);
static int64_t tar_atol256(const char *, unsigned);
static int64_t tar_atol8(const char *, unsigned);
static int tar_read_header(struct archive_read *, struct tar *,
struct archive_entry *);
static int tohex(int c);
static char *url_decode(const char *);
static int utf8_decode(wchar_t *, const char *, size_t length);
static char *wide_to_narrow(const wchar_t *wval);
int
archive_read_support_format_gnutar(struct archive *a)
{
return (archive_read_support_format_tar(a));
}
int
archive_read_support_format_tar(struct archive *_a)
{
struct archive_read *a = (struct archive_read *)_a;
struct tar *tar;
int r;
tar = (struct tar *)malloc(sizeof(*tar));
if (tar == NULL) {
archive_set_error(&a->archive, ENOMEM,
"Can't allocate tar data");
return (ARCHIVE_FATAL);
}
memset(tar, 0, sizeof(*tar));
r = __archive_read_register_format(a, tar,
archive_read_format_tar_bid,
archive_read_format_tar_read_header,
archive_read_format_tar_read_data,
archive_read_format_tar_skip,
archive_read_format_tar_cleanup);
if (r != ARCHIVE_OK)
free(tar);
return (ARCHIVE_OK);
}
static int
archive_read_format_tar_cleanup(struct archive_read *a)
{
struct tar *tar;
struct sparse_block *p;
tar = (struct tar *)(a->format->data);
while (tar->sparse_list != NULL) {
p = tar->sparse_list;
tar->sparse_list = p->next;
free(p);
}
archive_string_free(&tar->acl_text);
archive_string_free(&tar->entry_name);
archive_string_free(&tar->entry_linkname);
archive_string_free(&tar->entry_uname);
archive_string_free(&tar->entry_gname);
archive_string_free(&tar->line);
archive_string_free(&tar->pax_global);
archive_string_free(&tar->pax_header);
free(tar->pax_entry);
free(tar);
(a->format->data) = NULL;
return (ARCHIVE_OK);
}
static int
archive_read_format_tar_bid(struct archive_read *a)
{
int bid;
ssize_t bytes_read;
const void *h;
const struct archive_entry_header_ustar *header;
/*
* If we're already reading a non-tar file, don't
* bother to bid.
*/
if (a->archive.archive_format != 0 &&
(a->archive.archive_format & ARCHIVE_FORMAT_BASE_MASK) !=
ARCHIVE_FORMAT_TAR)
return (0);
bid = 0;
/*
* If we're already reading a tar format, start the bid at 1 as
* a failsafe.
*/
if ((a->archive.archive_format & ARCHIVE_FORMAT_BASE_MASK) ==
ARCHIVE_FORMAT_TAR)
bid++;
/* Now let's look at the actual header and see if it matches. */
if (a->decompressor->read_ahead != NULL)
bytes_read = (a->decompressor->read_ahead)(a, &h, 512);
else
bytes_read = 0; /* Empty file. */
if (bytes_read < 0)
return (ARCHIVE_FATAL);
if (bytes_read == 0 && bid > 0) {
/* An archive without a proper end-of-archive marker. */
/* Hold our nose and bid 1 anyway. */
return (1);
}
if (bytes_read < 512) {
/* If it's a new archive, then just return a zero bid. */
if (bid == 0)
return (0);
/*
* If we already know this is a tar archive,
* then we have a problem.
*/
archive_set_error(&a->archive, ARCHIVE_ERRNO_FILE_FORMAT,
"Truncated tar archive");
return (ARCHIVE_FATAL);
}
/* If it's an end-of-archive mark, we can handle it. */
if ((*(const char *)h) == 0 && archive_block_is_null((const unsigned char *)h)) {
/* If it's a known tar file, end-of-archive is definite. */
if ((a->archive.archive_format & ARCHIVE_FORMAT_BASE_MASK) ==
ARCHIVE_FORMAT_TAR)
return (512);
/* Empty archive? */
return (1);
}
/* If it's not an end-of-archive mark, it must have a valid checksum.*/
if (!checksum(a, h))
return (0);
bid += 48; /* Checksum is usually 6 octal digits. */
header = (const struct archive_entry_header_ustar *)h;
/* Recognize POSIX formats. */
if ((memcmp(header->magic, "ustar\0", 6) == 0)
&&(memcmp(header->version, "00", 2)==0))
bid += 56;
/* Recognize GNU tar format. */
if ((memcmp(header->magic, "ustar ", 6) == 0)
&&(memcmp(header->version, " \0", 2)==0))
bid += 56;
/* Type flag must be null, digit or A-Z, a-z. */
if (header->typeflag[0] != 0 &&
!( header->typeflag[0] >= '0' && header->typeflag[0] <= '9') &&
!( header->typeflag[0] >= 'A' && header->typeflag[0] <= 'Z') &&
!( header->typeflag[0] >= 'a' && header->typeflag[0] <= 'z') )
return (0);
bid += 2; /* 6 bits of variation in an 8-bit field leaves 2 bits. */
/* Sanity check: Look at first byte of mode field. */
switch (255 & (unsigned)header->mode[0]) {
case 0: case 255:
/* Base-256 value: No further verification possible! */
break;
case ' ': /* Not recommended, but not illegal, either. */
break;
case '0': case '1': case '2': case '3':
case '4': case '5': case '6': case '7':
/* Octal Value. */
/* TODO: Check format of remainder of this field. */
break;
default:
/* Not a valid mode; bail out here. */
return (0);
}
/* TODO: Sanity test uid/gid/size/mtime/rdevmajor/rdevminor fields. */
return (bid);
}
/*
* The function invoked by archive_read_header(). This
* just sets up a few things and then calls the internal
* tar_read_header() function below.
*/
static int
archive_read_format_tar_read_header(struct archive_read *a,
struct archive_entry *entry)
{
/*
* When converting tar archives to cpio archives, it is
* essential that each distinct file have a distinct inode
* number. To simplify this, we keep a static count here to
* assign fake dev/inode numbers to each tar entry. Note that
* pax format archives may overwrite this with something more
* useful.
*
* Ideally, we would track every file read from the archive so
* that we could assign the same dev/ino pair to hardlinks,
* but the memory required to store a complete lookup table is
* probably not worthwhile just to support the relatively
* obscure tar->cpio conversion case.
*/
static int default_inode;
static int default_dev;
struct tar *tar;
struct sparse_block *sp;
const char *p;
int r;
size_t l;
ssize_t size;
/* Assign default device/inode values. */
archive_entry_set_dev(entry, 1 + default_dev); /* Don't use zero. */
archive_entry_set_ino(entry, ++default_inode); /* Don't use zero. */
/* Limit generated st_ino number to 16 bits. */
if (default_inode >= 0xffff) {
++default_dev;
default_inode = 0;
}
tar = (struct tar *)(a->format->data);
tar->entry_offset = 0;
while (tar->sparse_list != NULL) {
sp = tar->sparse_list;
tar->sparse_list = sp->next;
free(sp);
}
tar->sparse_last = NULL;
r = tar_read_header(a, tar, entry);
/*
* Yuck. See comments for gnu_sparse_10_read for why this
* is here and not in _read_data where it "should" go.
*/
if (tar->sparse_gnu_pending
&& tar->sparse_gnu_major == 1
&& tar->sparse_gnu_minor == 0) {
tar->sparse_gnu_pending = 0;
/* Read initial sparse map. */
size = gnu_sparse_10_read(a, tar);
if (size < 0)
return (size);
tar->entry_bytes_remaining -= size;
tar->entry_padding += size;
}
/*
* "non-sparse" files are really just sparse files with
* a single block.
*/
if (tar->sparse_list == NULL)
gnu_add_sparse_entry(tar, 0, tar->entry_bytes_remaining);
tar->realsize = archive_entry_size(entry);
if (r == ARCHIVE_OK) {
/*
* "Regular" entry with trailing '/' is really
* directory: This is needed for certain old tar
* variants and even for some broken newer ones.
*/
p = archive_entry_pathname(entry);
l = strlen(p);
if (archive_entry_filetype(entry) == AE_IFREG
&& p[l-1] == '/')
archive_entry_set_filetype(entry, AE_IFDIR);
}
return (r);
}
static int
archive_read_format_tar_read_data(struct archive_read *a,
const void **buff, size_t *size, off_t *offset)
{
ssize_t bytes_read;
struct tar *tar;
struct sparse_block *p;
tar = (struct tar *)(a->format->data);
if (tar->sparse_gnu_pending) {
if (tar->sparse_gnu_major == 1 && tar->sparse_gnu_minor == 0) {
/*
* <sigh> We should parse the sparse data
* here, but have to parse it as part of the
* header because of a bug in GNU tar 1.16.1.
*/
} else {
*size = 0;
*offset = 0;
archive_set_error(&a->archive, ARCHIVE_ERRNO_MISC,
"Unrecognized GNU sparse file format");
return (ARCHIVE_WARN);
}
tar->sparse_gnu_pending = 0;
}
/* Remove exhausted entries from sparse list. */
while (tar->sparse_list != NULL &&
tar->sparse_list->remaining == 0) {
p = tar->sparse_list;
tar->sparse_list = p->next;
free(p);
}
/* If we're at end of file, return EOF. */
if (tar->sparse_list == NULL || tar->entry_bytes_remaining == 0) {
if ((a->decompressor->skip)(a, tar->entry_padding) < 0)
return (ARCHIVE_FATAL);
tar->entry_padding = 0;
*buff = NULL;
*size = 0;
*offset = tar->realsize;
return (ARCHIVE_EOF);
}
bytes_read = (a->decompressor->read_ahead)(a, buff, 1);
if (bytes_read == 0) {
archive_set_error(&a->archive, ARCHIVE_ERRNO_MISC,
"Truncated tar archive");
return (ARCHIVE_FATAL);
}
if (bytes_read < 0)
return (ARCHIVE_FATAL);
if (bytes_read > tar->entry_bytes_remaining)
bytes_read = tar->entry_bytes_remaining;
/* Don't read more than is available in the
* current sparse block. */
if (tar->sparse_list->remaining < bytes_read)
bytes_read = tar->sparse_list->remaining;
*size = bytes_read;
*offset = tar->sparse_list->offset;
tar->sparse_list->remaining -= bytes_read;
tar->sparse_list->offset += bytes_read;
tar->entry_bytes_remaining -= bytes_read;
(a->decompressor->consume)(a, bytes_read);
return (ARCHIVE_OK);
}
static int
archive_read_format_tar_skip(struct archive_read *a)
{
off_t bytes_skipped;
struct tar* tar;
struct sparse_block *p;
tar = (struct tar *)(a->format->data);
/*
* Compression layer skip functions are required to either skip the
* length requested or fail, so we can rely upon the entire entry
* plus padding being skipped.
*/
bytes_skipped = (a->decompressor->skip)(a, tar->entry_bytes_remaining +
tar->entry_padding);
if (bytes_skipped < 0)
return (ARCHIVE_FATAL);
tar->entry_bytes_remaining = 0;
tar->entry_padding = 0;
/* Free the sparse list. */
while (tar->sparse_list != NULL) {
p = tar->sparse_list;
tar->sparse_list = p->next;
free(p);
}
tar->sparse_last = NULL;
return (ARCHIVE_OK);
}
/*
* This function recursively interprets all of the headers associated
* with a single entry.
*/
static int
tar_read_header(struct archive_read *a, struct tar *tar,
struct archive_entry *entry)
{
ssize_t bytes;
int err;
const void *h;
const struct archive_entry_header_ustar *header;
/* Read 512-byte header record */
bytes = (a->decompressor->read_ahead)(a, &h, 512);
if (bytes < 512) {
/*
* If we're here, it's becase the _bid function accepted
* this file. So just call a short read end-of-archive
* and be done with it.
*/
return (ARCHIVE_EOF);
}
(a->decompressor->consume)(a, 512);
/* Check for end-of-archive mark. */
if (((*(const char *)h)==0) && archive_block_is_null((const unsigned char *)h)) {
/* Try to consume a second all-null record, as well. */
bytes = (a->decompressor->read_ahead)(a, &h, 512);
if (bytes > 0)
(a->decompressor->consume)(a, bytes);
archive_set_error(&a->archive, 0, NULL);
return (ARCHIVE_EOF);
}
/*
* Note: If the checksum fails and we return ARCHIVE_RETRY,
* then the client is likely to just retry. This is a very
* crude way to search for the next valid header!
*
* TODO: Improve this by implementing a real header scan.
*/
if (!checksum(a, h)) {
archive_set_error(&a->archive, EINVAL, "Damaged tar archive");
return (ARCHIVE_RETRY); /* Retryable: Invalid header */
}
if (++tar->header_recursion_depth > 32) {
archive_set_error(&a->archive, EINVAL, "Too many special headers");
return (ARCHIVE_WARN);
}
/* Determine the format variant. */
header = (const struct archive_entry_header_ustar *)h;
switch(header->typeflag[0]) {
case 'A': /* Solaris tar ACL */
a->archive.archive_format = ARCHIVE_FORMAT_TAR_PAX_INTERCHANGE;
a->archive.archive_format_name = "Solaris tar";
err = header_Solaris_ACL(a, tar, entry, h);
break;
case 'g': /* POSIX-standard 'g' header. */
a->archive.archive_format = ARCHIVE_FORMAT_TAR_PAX_INTERCHANGE;
a->archive.archive_format_name = "POSIX pax interchange format";
err = header_pax_global(a, tar, entry, h);
break;
case 'K': /* Long link name (GNU tar, others) */
err = header_longlink(a, tar, entry, h);
break;
case 'L': /* Long filename (GNU tar, others) */
err = header_longname(a, tar, entry, h);
break;
case 'V': /* GNU volume header */
err = header_volume(a, tar, entry, h);
break;
case 'X': /* Used by SUN tar; same as 'x'. */
a->archive.archive_format = ARCHIVE_FORMAT_TAR_PAX_INTERCHANGE;
a->archive.archive_format_name =
"POSIX pax interchange format (Sun variant)";
err = header_pax_extensions(a, tar, entry, h);
break;
case 'x': /* POSIX-standard 'x' header. */
a->archive.archive_format = ARCHIVE_FORMAT_TAR_PAX_INTERCHANGE;
a->archive.archive_format_name = "POSIX pax interchange format";
err = header_pax_extensions(a, tar, entry, h);
break;
default:
if (memcmp(header->magic, "ustar \0", 8) == 0) {
a->archive.archive_format = ARCHIVE_FORMAT_TAR_GNUTAR;
a->archive.archive_format_name = "GNU tar format";
err = header_gnutar(a, tar, entry, h);
} else if (memcmp(header->magic, "ustar", 5) == 0) {
if (a->archive.archive_format != ARCHIVE_FORMAT_TAR_PAX_INTERCHANGE) {
a->archive.archive_format = ARCHIVE_FORMAT_TAR_USTAR;
a->archive.archive_format_name = "POSIX ustar format";
}
err = header_ustar(a, tar, entry, h);
} else {
a->archive.archive_format = ARCHIVE_FORMAT_TAR;
a->archive.archive_format_name = "tar (non-POSIX)";
err = header_old_tar(a, tar, entry, h);
}
}
--tar->header_recursion_depth;
return (err);
}
/*
* Return true if block checksum is correct.
*/
static int
checksum(struct archive_read *a, const void *h)
{
const unsigned char *bytes;
const struct archive_entry_header_ustar *header;
int check, i, sum;
(void)a; /* UNUSED */
bytes = (const unsigned char *)h;
header = (const struct archive_entry_header_ustar *)h;
/*
* Test the checksum. Note that POSIX specifies _unsigned_
* bytes for this calculation.
*/
sum = tar_atol(header->checksum, sizeof(header->checksum));
check = 0;
for (i = 0; i < 148; i++)
check += (unsigned char)bytes[i];
for (; i < 156; i++)
check += 32;
for (; i < 512; i++)
check += (unsigned char)bytes[i];
if (sum == check)
return (1);
/*
* Repeat test with _signed_ bytes, just in case this archive
* was created by an old BSD, Solaris, or HP-UX tar with a
* broken checksum calculation.
*/
check = 0;
for (i = 0; i < 148; i++)
check += (signed char)bytes[i];
for (; i < 156; i++)
check += 32;
for (; i < 512; i++)
check += (signed char)bytes[i];
if (sum == check)
return (1);
return (0);
}
/*
* Return true if this block contains only nulls.
*/
static int
archive_block_is_null(const unsigned char *p)
{
unsigned i;
for (i = 0; i < ARCHIVE_BYTES_PER_RECORD / sizeof(*p); i++)
if (*p++)
return (0);
return (1);
}
/*
* Interpret 'A' Solaris ACL header
*/
static int
header_Solaris_ACL(struct archive_read *a, struct tar *tar,
struct archive_entry *entry, const void *h)
{
int err, err2;
char *p;
wchar_t *wp;
err = read_body_to_string(a, tar, &(tar->acl_text), h);
err2 = tar_read_header(a, tar, entry);
err = err_combine(err, err2);
/* XXX Ensure p doesn't overrun acl_text */
/* Skip leading octal number. */
/* XXX TODO: Parse the octal number and sanity-check it. */
p = tar->acl_text.s;
while (*p != '\0')
p++;
p++;
wp = (wchar_t *)malloc((strlen(p) + 1) * sizeof(wchar_t));
if (wp != NULL) {
utf8_decode(wp, p, strlen(p));
err2 = __archive_entry_acl_parse_w(entry, wp,
ARCHIVE_ENTRY_ACL_TYPE_ACCESS);
err = err_combine(err, err2);
free(wp);
}
return (err);
}
/*
* Interpret 'K' long linkname header.
*/
static int
header_longlink(struct archive_read *a, struct tar *tar,
struct archive_entry *entry, const void *h)
{
int err, err2;
err = read_body_to_string(a, tar, &(tar->longlink), h);
err2 = tar_read_header(a, tar, entry);
if (err == ARCHIVE_OK && err2 == ARCHIVE_OK) {
/* Set symlink if symlink already set, else hardlink. */
archive_entry_set_link(entry, tar->longlink.s);
}
return (err_combine(err, err2));
}
/*
* Interpret 'L' long filename header.
*/
static int
header_longname(struct archive_read *a, struct tar *tar,
struct archive_entry *entry, const void *h)
{
int err, err2;
err = read_body_to_string(a, tar, &(tar->longname), h);
/* Read and parse "real" header, then override name. */
err2 = tar_read_header(a, tar, entry);
if (err == ARCHIVE_OK && err2 == ARCHIVE_OK)
archive_entry_set_pathname(entry, tar->longname.s);
return (err_combine(err, err2));
}
/*
* Interpret 'V' GNU tar volume header.
*/
static int
header_volume(struct archive_read *a, struct tar *tar,
struct archive_entry *entry, const void *h)
{
(void)h;
/* Just skip this and read the next header. */
return (tar_read_header(a, tar, entry));
}
/*
* Read body of an archive entry into an archive_string object.
*/
static int
read_body_to_string(struct archive_read *a, struct tar *tar,
struct archive_string *as, const void *h)
{
off_t size, padded_size;
ssize_t bytes_read, bytes_to_copy;
const struct archive_entry_header_ustar *header;
const void *src;
char *dest;
(void)tar; /* UNUSED */
header = (const struct archive_entry_header_ustar *)h;
size = tar_atol(header->size, sizeof(header->size));
/* Read the body into the string. */
archive_string_ensure(as, size+1);
padded_size = (size + 511) & ~ 511;
dest = as->s;
while (padded_size > 0) {
bytes_read = (a->decompressor->read_ahead)(a, &src, padded_size);
if (bytes_read < 0)
return (ARCHIVE_FATAL);
if (bytes_read > padded_size)
bytes_read = padded_size;
(a->decompressor->consume)(a, bytes_read);
bytes_to_copy = bytes_read;
if ((off_t)bytes_to_copy > size)
bytes_to_copy = (ssize_t)size;
memcpy(dest, src, bytes_to_copy);
dest += bytes_to_copy;
size -= bytes_to_copy;
padded_size -= bytes_read;
}
*dest = '\0';
return (ARCHIVE_OK);
}
/*
* Parse out common header elements.
*
* This would be the same as header_old_tar, except that the
* filename is handled slightly differently for old and POSIX
* entries (POSIX entries support a 'prefix'). This factoring
* allows header_old_tar and header_ustar
* to handle filenames differently, while still putting most of the
* common parsing into one place.
*/
static int
header_common(struct archive_read *a, struct tar *tar,
struct archive_entry *entry, const void *h)
{
const struct archive_entry_header_ustar *header;
char tartype;
(void)a; /* UNUSED */
header = (const struct archive_entry_header_ustar *)h;
if (header->linkname[0])
archive_strncpy(&(tar->entry_linkname), header->linkname,
sizeof(header->linkname));
else
archive_string_empty(&(tar->entry_linkname));
/* Parse out the numeric fields (all are octal) */
archive_entry_set_mode(entry, tar_atol(header->mode, sizeof(header->mode)));
archive_entry_set_uid(entry, tar_atol(header->uid, sizeof(header->uid)));
archive_entry_set_gid(entry, tar_atol(header->gid, sizeof(header->gid)));
tar->entry_bytes_remaining = tar_atol(header->size, sizeof(header->size));
archive_entry_set_size(entry, tar->entry_bytes_remaining);
archive_entry_set_mtime(entry, tar_atol(header->mtime, sizeof(header->mtime)), 0);
/* Handle the tar type flag appropriately. */
tartype = header->typeflag[0];
switch (tartype) {
case '1': /* Hard link */
archive_entry_set_hardlink(entry, tar->entry_linkname.s);
/*
* The following may seem odd, but: Technically, tar
* does not store the file type for a "hard link"
* entry, only the fact that it is a hard link. So, I
* leave the type zero normally. But, pax interchange
* format allows hard links to have data, which
* implies that the underlying entry is a regular
* file.
*/
if (archive_entry_size(entry) > 0)
archive_entry_set_filetype(entry, AE_IFREG);
/*
* A tricky point: Traditionally, tar readers have
* ignored the size field when reading hardlink
* entries, and some writers put non-zero sizes even
* though the body is empty. POSIX.1-2001 broke with
* this tradition by permitting hardlink entries to
* store valid bodies in pax interchange format, but
* not in ustar format. Since there is no hard and
* fast way to distinguish pax interchange from
* earlier archives (the 'x' and 'g' entries are
* optional, after all), we need a heuristic. Here, I
* use the bid function to test whether or not there's
* a valid header following. Of course, if we know
* this is pax interchange format, then we must obey
* the size.
*
* This heuristic will only fail for a pax interchange
* archive that is storing hardlink bodies, no pax
* extended attribute entries have yet occurred, and
* we encounter a hardlink entry for a file that is
* itself an uncompressed tar archive.
*/
if (archive_entry_size(entry) > 0 &&
a->archive.archive_format != ARCHIVE_FORMAT_TAR_PAX_INTERCHANGE &&
archive_read_format_tar_bid(a) > 50) {
archive_entry_set_size(entry, 0);
tar->entry_bytes_remaining = 0;
}
case '2': /* Symlink */
archive_entry_set_filetype(entry, AE_IFLNK);
archive_entry_set_size(entry, 0);
tar->entry_bytes_remaining = 0;
archive_entry_set_symlink(entry, tar->entry_linkname.s);
break;
case '3': /* Character device */
archive_entry_set_filetype(entry, AE_IFCHR);
archive_entry_set_size(entry, 0);
tar->entry_bytes_remaining = 0;
break;
case '4': /* Block device */
archive_entry_set_filetype(entry, AE_IFBLK);
archive_entry_set_size(entry, 0);
tar->entry_bytes_remaining = 0;
break;
case '5': /* Dir */
archive_entry_set_filetype(entry, AE_IFDIR);
archive_entry_set_size(entry, 0);
tar->entry_bytes_remaining = 0;
break;
case '6': /* FIFO device */
archive_entry_set_filetype(entry, AE_IFIFO);
archive_entry_set_size(entry, 0);
tar->entry_bytes_remaining = 0;
break;
case 'D': /* GNU incremental directory type */
/*
* No special handling is actually required here.
* It might be nice someday to preprocess the file list and
* provide it to the client, though.
*/
archive_entry_set_filetype(entry, AE_IFDIR);
break;
case 'M': /* GNU "Multi-volume" (remainder of file from last archive)*/
/*
* As far as I can tell, this is just like a regular file
* entry, except that the contents should be _appended_ to
* the indicated file at the indicated offset. This may
* require some API work to fully support.
*/
break;
case 'N': /* Old GNU "long filename" entry. */
/* The body of this entry is a script for renaming
* previously-extracted entries. Ugh. It will never
* be supported by libarchive. */
archive_entry_set_filetype(entry, AE_IFREG);
break;
case 'S': /* GNU sparse files */
/*
* Sparse files are really just regular files with
* sparse information in the extended area.
*/
/* FALLTHROUGH */
default: /* Regular file and non-standard types */
/*
* Per POSIX: non-recognized types should always be
* treated as regular files.
*/
archive_entry_set_filetype(entry, AE_IFREG);
break;
}
return (0);
}
/*
* Parse out header elements for "old-style" tar archives.
*/
static int
header_old_tar(struct archive_read *a, struct tar *tar,
struct archive_entry *entry, const void *h)
{
const struct archive_entry_header_ustar *header;
/* Copy filename over (to ensure null termination). */
header = (const struct archive_entry_header_ustar *)h;
archive_strncpy(&(tar->entry_name), header->name, sizeof(header->name));
archive_entry_set_pathname(entry, tar->entry_name.s);
/* Grab rest of common fields */
header_common(a, tar, entry, h);
tar->entry_padding = 0x1ff & (-tar->entry_bytes_remaining);
return (0);
}
/*
* Parse a file header for a pax extended archive entry.
*/
static int
header_pax_global(struct archive_read *a, struct tar *tar,
struct archive_entry *entry, const void *h)
{
int err, err2;
err = read_body_to_string(a, tar, &(tar->pax_global), h);
err2 = tar_read_header(a, tar, entry);
return (err_combine(err, err2));
}
static int
header_pax_extensions(struct archive_read *a, struct tar *tar,
struct archive_entry *entry, const void *h)
{
int err, err2;
read_body_to_string(a, tar, &(tar->pax_header), h);
/* Parse the next header. */
err = tar_read_header(a, tar, entry);
/*
* TODO: Parse global/default options into 'entry' struct here
* before handling file-specific options.
*
* This design (parse standard header, then overwrite with pax
* extended attribute data) usually works well, but isn't ideal;
* it would be better to parse the pax extended attributes first
* and then skip any fields in the standard header that were
* defined in the pax header.
*/
err2 = pax_header(a, tar, entry, tar->pax_header.s);
err = err_combine(err, err2);
tar->entry_padding = 0x1ff & (-tar->entry_bytes_remaining);
return (err);
}
/*
* Parse a file header for a Posix "ustar" archive entry. This also
* handles "pax" or "extended ustar" entries.
*/
static int
header_ustar(struct archive_read *a, struct tar *tar,
struct archive_entry *entry, const void *h)
{
const struct archive_entry_header_ustar *header;
struct archive_string *as;
header = (const struct archive_entry_header_ustar *)h;
/* Copy name into an internal buffer to ensure null-termination. */
as = &(tar->entry_name);
if (header->prefix[0]) {
archive_strncpy(as, header->prefix, sizeof(header->prefix));
if (as->s[archive_strlen(as) - 1] != '/')
archive_strappend_char(as, '/');
archive_strncat(as, header->name, sizeof(header->name));
} else
archive_strncpy(as, header->name, sizeof(header->name));
archive_entry_set_pathname(entry, as->s);
/* Handle rest of common fields. */
header_common(a, tar, entry, h);
/* Handle POSIX ustar fields. */
archive_strncpy(&(tar->entry_uname), header->uname,
sizeof(header->uname));
archive_entry_set_uname(entry, tar->entry_uname.s);
archive_strncpy(&(tar->entry_gname), header->gname,
sizeof(header->gname));
archive_entry_set_gname(entry, tar->entry_gname.s);
/* Parse out device numbers only for char and block specials. */
if (header->typeflag[0] == '3' || header->typeflag[0] == '4') {
archive_entry_set_rdevmajor(entry,
tar_atol(header->rdevmajor, sizeof(header->rdevmajor)));
archive_entry_set_rdevminor(entry,
tar_atol(header->rdevminor, sizeof(header->rdevminor)));
}
tar->entry_padding = 0x1ff & (-tar->entry_bytes_remaining);
return (0);
}
/*
* Parse the pax extended attributes record.
*
* Returns non-zero if there's an error in the data.
*/
static int
pax_header(struct archive_read *a, struct tar *tar,
struct archive_entry *entry, char *attr)
{
size_t attr_length, l, line_length;
char *line, *p;
wchar_t *key, *wp, *value;
int err, err2;
attr_length = strlen(attr);
err = ARCHIVE_OK;
while (attr_length > 0) {
/* Parse decimal length field at start of line. */
line_length = 0;
l = attr_length;
line = p = attr; /* Record start of line. */
while (l>0) {
if (*p == ' ') {
p++;
l--;
break;
}
if (*p < '0' || *p > '9')
return (-1);
line_length *= 10;
line_length += *p - '0';
if (line_length > 999999) {
archive_set_error(&a->archive, ARCHIVE_ERRNO_MISC,
"Rejecting pax extended attribute > 1MB");
return (ARCHIVE_WARN);
}
p++;
l--;
}
if (line_length > attr_length)
return (0);
/* Ensure pax_entry buffer is big enough. */
if (tar->pax_entry_length <= line_length) {
wchar_t *old_entry = tar->pax_entry;
if (tar->pax_entry_length <= 0)
tar->pax_entry_length = 1024;
while (tar->pax_entry_length <= line_length + 1)
tar->pax_entry_length *= 2;
old_entry = tar->pax_entry;
tar->pax_entry = (wchar_t *)realloc(tar->pax_entry,
tar->pax_entry_length * sizeof(wchar_t));
if (tar->pax_entry == NULL) {
free(old_entry);
archive_set_error(&a->archive, ENOMEM,
"No memory");
return (ARCHIVE_FATAL);
}
}
/* Decode UTF-8 to wchar_t, null-terminate result. */
if (utf8_decode(tar->pax_entry, p,
line_length - (p - attr) - 1)) {
archive_set_error(&a->archive, ARCHIVE_ERRNO_MISC,
"Invalid UTF8 character in pax extended attribute");
err = err_combine(err, ARCHIVE_WARN);
}
/* Null-terminate 'key' value. */
wp = key = tar->pax_entry;
if (key[0] == L'=')
return (-1);
while (*wp && *wp != L'=')
++wp;
if (*wp == L'\0') {
archive_set_error(&a->archive, ARCHIVE_ERRNO_MISC,
"Invalid pax extended attributes");
return (ARCHIVE_WARN);
}
*wp = 0;
/* Identify null-terminated 'value' portion. */
value = wp + 1;
/* Identify this attribute and set it in the entry. */
err2 = pax_attribute(tar, entry, key, value);
err = err_combine(err, err2);
/* Skip to next line */
attr += line_length;
attr_length -= line_length;
}
return (err);
}
static int
pax_attribute_xattr(struct archive_entry *entry,
wchar_t *name, wchar_t *value)
{
char *name_decoded, *name_narrow;
void *value_decoded;
size_t value_len;
if (wcslen(name) < 18 || (wcsncmp(name, L"LIBARCHIVE.xattr.", 17)) != 0)
return 3;
name += 17;
/* URL-decode name */
name_narrow = wide_to_narrow(name);
if (name_narrow == NULL)
return 2;
name_decoded = url_decode(name_narrow);
free(name_narrow);
if (name_decoded == NULL)
return 2;
/* Base-64 decode value */
value_decoded = base64_decode(value, wcslen(value), &value_len);
if (value_decoded == NULL) {
free(name_decoded);
return 1;
}
archive_entry_xattr_add_entry(entry, name_decoded,
value_decoded, value_len);
free(name_decoded);
free(value_decoded);
return 0;
}
/*
* Parse a single key=value attribute. key/value pointers are
* assumed to point into reasonably long-lived storage.
*
* Note that POSIX reserves all-lowercase keywords. Vendor-specific
* extensions should always have keywords of the form "VENDOR.attribute"
* In particular, it's quite feasible to support many different
* vendor extensions here. I'm using "LIBARCHIVE" for extensions
* unique to this library.
*
* Investigate other vendor-specific extensions and see if
* any of them look useful.
*/
static int
pax_attribute(struct tar *tar, struct archive_entry *entry,
wchar_t *key, wchar_t *value)
{
int64_t s;
long n;
switch (key[0]) {
case 'G':
/* GNU "0.0" sparse pax format. */
if (wcscmp(key, L"GNU.sparse.numblocks") == 0) {
tar->sparse_offset = -1;
tar->sparse_numbytes = -1;
tar->sparse_gnu_major = 0;
tar->sparse_gnu_minor = 0;
}
if (wcscmp(key, L"GNU.sparse.offset") == 0) {
tar->sparse_offset = tar_atol10(value, wcslen(value));
if (tar->sparse_numbytes != -1) {
gnu_add_sparse_entry(tar,
tar->sparse_offset, tar->sparse_numbytes);
tar->sparse_offset = -1;
tar->sparse_numbytes = -1;
}
}
if (wcscmp(key, L"GNU.sparse.numbytes") == 0) {
tar->sparse_numbytes = tar_atol10(value, wcslen(value));
if (tar->sparse_numbytes != -1) {
gnu_add_sparse_entry(tar,
tar->sparse_offset, tar->sparse_numbytes);
tar->sparse_offset = -1;
tar->sparse_numbytes = -1;
}
}
if (wcscmp(key, L"GNU.sparse.size") == 0)
archive_entry_set_size(entry,
tar_atol10(value, wcslen(value)));
/* GNU "0.1" sparse pax format. */
if (wcscmp(key, L"GNU.sparse.map") == 0) {
tar->sparse_gnu_major = 0;
tar->sparse_gnu_minor = 1;
if (gnu_sparse_01_parse(tar, value) != ARCHIVE_OK)
return (ARCHIVE_WARN);
}
/* GNU "1.0" sparse pax format */
if (wcscmp(key, L"GNU.sparse.major") == 0) {
tar->sparse_gnu_major = tar_atol10(value, wcslen(value));
tar->sparse_gnu_pending = 1;
}
if (wcscmp(key, L"GNU.sparse.minor") == 0) {
tar->sparse_gnu_minor = tar_atol10(value, wcslen(value));
tar->sparse_gnu_pending = 1;
}
if (wcscmp(key, L"GNU.sparse.name") == 0)
archive_entry_copy_pathname_w(entry, value);
if (wcscmp(key, L"GNU.sparse.realsize") == 0)
archive_entry_set_size(entry,
tar_atol10(value, wcslen(value)));
break;
case 'L':
/* Our extensions */
/* TODO: Handle arbitrary extended attributes... */
/*
if (strcmp(key, "LIBARCHIVE.xxxxxxx")==0)
archive_entry_set_xxxxxx(entry, value);
*/
if (wcsncmp(key, L"LIBARCHIVE.xattr.", 17)==0)
pax_attribute_xattr(entry, key, value);
break;
case 'S':
/* We support some keys used by the "star" archiver */
if (wcscmp(key, L"SCHILY.acl.access")==0)
__archive_entry_acl_parse_w(entry, value,
ARCHIVE_ENTRY_ACL_TYPE_ACCESS);
else if (wcscmp(key, L"SCHILY.acl.default")==0)
__archive_entry_acl_parse_w(entry, value,
ARCHIVE_ENTRY_ACL_TYPE_DEFAULT);
else if (wcscmp(key, L"SCHILY.devmajor")==0)
archive_entry_set_rdevmajor(entry, tar_atol10(value, wcslen(value)));
else if (wcscmp(key, L"SCHILY.devminor")==0)
archive_entry_set_rdevminor(entry, tar_atol10(value, wcslen(value)));
else if (wcscmp(key, L"SCHILY.fflags")==0)
archive_entry_copy_fflags_text_w(entry, value);
else if (wcscmp(key, L"SCHILY.dev")==0)
archive_entry_set_dev(entry, tar_atol10(value, wcslen(value)));
else if (wcscmp(key, L"SCHILY.ino")==0)
archive_entry_set_ino(entry, tar_atol10(value, wcslen(value)));
else if (wcscmp(key, L"SCHILY.nlink")==0)
archive_entry_set_nlink(entry, tar_atol10(value, wcslen(value)));
break;
case 'a':
if (wcscmp(key, L"atime")==0) {
pax_time(value, &s, &n);
archive_entry_set_atime(entry, s, n);
}
break;
case 'c':
if (wcscmp(key, L"ctime")==0) {
pax_time(value, &s, &n);
archive_entry_set_ctime(entry, s, n);
} else if (wcscmp(key, L"charset")==0) {
/* TODO: Publish charset information in entry. */
} else if (wcscmp(key, L"comment")==0) {
/* TODO: Publish comment in entry. */
}
break;
case 'g':
if (wcscmp(key, L"gid")==0)
archive_entry_set_gid(entry, tar_atol10(value, wcslen(value)));
else if (wcscmp(key, L"gname")==0)
archive_entry_copy_gname_w(entry, value);
break;
case 'l':
/* pax interchange doesn't distinguish hardlink vs. symlink. */
if (wcscmp(key, L"linkpath")==0) {
if (archive_entry_hardlink(entry))
archive_entry_copy_hardlink_w(entry, value);
else
archive_entry_copy_symlink_w(entry, value);
}
break;
case 'm':
if (wcscmp(key, L"mtime")==0) {
pax_time(value, &s, &n);
archive_entry_set_mtime(entry, s, n);
}
break;
case 'p':
if (wcscmp(key, L"path")==0)
archive_entry_copy_pathname_w(entry, value);
break;
case 'r':
/* POSIX has reserved 'realtime.*' */
break;
case 's':
/* POSIX has reserved 'security.*' */
/* Someday: if (wcscmp(key, L"security.acl")==0) { ... } */
if (wcscmp(key, L"size")==0) {
tar->entry_bytes_remaining = tar_atol10(value, wcslen(value));
archive_entry_set_size(entry, tar->entry_bytes_remaining);
}
tar->entry_bytes_remaining = 0;
break;
case 'u':
if (wcscmp(key, L"uid")==0)
archive_entry_set_uid(entry, tar_atol10(value, wcslen(value)));
else if (wcscmp(key, L"uname")==0)
archive_entry_copy_uname_w(entry, value);
break;
}
return (0);
}
/*
* parse a decimal time value, which may include a fractional portion
*/
static void
pax_time(const wchar_t *p, int64_t *ps, long *pn)
{
char digit;
int64_t s;
unsigned long l;
int sign;
int64_t limit, last_digit_limit;
limit = INT64_MAX / 10;
last_digit_limit = INT64_MAX % 10;
s = 0;
sign = 1;
if (*p == '-') {
sign = -1;
p++;
}
while (*p >= '0' && *p <= '9') {
digit = *p - '0';
if (s > limit ||
(s == limit && digit > last_digit_limit)) {
s = UINT64_MAX;
break;
}
s = (s * 10) + digit;
++p;
}
*ps = s * sign;
/* Calculate nanoseconds. */
*pn = 0;
if (*p != '.')
return;
l = 100000000UL;
do {
++p;
if (*p >= '0' && *p <= '9')
*pn += (*p - '0') * l;
else
break;
} while (l /= 10);
}
/*
* Parse GNU tar header
*/
static int
header_gnutar(struct archive_read *a, struct tar *tar,
struct archive_entry *entry, const void *h)
{
const struct archive_entry_header_gnutar *header;
(void)a;
/*
* GNU header is like POSIX ustar, except 'prefix' is
* replaced with some other fields. This also means the
* filename is stored as in old-style archives.
*/
/* Grab fields common to all tar variants. */
header_common(a, tar, entry, h);
/* Copy filename over (to ensure null termination). */
header = (const struct archive_entry_header_gnutar *)h;
archive_strncpy(&(tar->entry_name), header->name,
sizeof(header->name));
archive_entry_set_pathname(entry, tar->entry_name.s);
/* Fields common to ustar and GNU */
/* XXX Can the following be factored out since it's common
* to ustar and gnu tar? Is it okay to move it down into
* header_common, perhaps? */
archive_strncpy(&(tar->entry_uname),
header->uname, sizeof(header->uname));
archive_entry_set_uname(entry, tar->entry_uname.s);
archive_strncpy(&(tar->entry_gname),
header->gname, sizeof(header->gname));
archive_entry_set_gname(entry, tar->entry_gname.s);
/* Parse out device numbers only for char and block specials */
if (header->typeflag[0] == '3' || header->typeflag[0] == '4') {
archive_entry_set_rdevmajor(entry,
tar_atol(header->rdevmajor, sizeof(header->rdevmajor)));
archive_entry_set_rdevminor(entry,
tar_atol(header->rdevminor, sizeof(header->rdevminor)));
} else
archive_entry_set_rdev(entry, 0);
tar->entry_padding = 0x1ff & (-tar->entry_bytes_remaining);
/* Grab GNU-specific fields. */
archive_entry_set_atime(entry,
tar_atol(header->atime, sizeof(header->atime)), 0);
archive_entry_set_ctime(entry,
tar_atol(header->ctime, sizeof(header->ctime)), 0);
if (header->realsize[0] != 0) {
archive_entry_set_size(entry,
tar_atol(header->realsize, sizeof(header->realsize)));
}
if (header->sparse[0].offset[0] != 0) {
gnu_sparse_old_read(a, tar, header);
} else {
if (header->isextended[0] != 0) {
/* XXX WTF? XXX */
}
}
return (0);
}
static void
gnu_add_sparse_entry(struct tar *tar, off_t offset, off_t remaining)
{
struct sparse_block *p;
p = (struct sparse_block *)malloc(sizeof(*p));
if (p == NULL)
__archive_errx(1, "Out of memory");
memset(p, 0, sizeof(*p));
if (tar->sparse_last != NULL)
tar->sparse_last->next = p;
else
tar->sparse_list = p;
tar->sparse_last = p;
p->offset = offset;
p->remaining = remaining;
}
/*
* GNU tar old-format sparse data.
*
* GNU old-format sparse data is stored in a fixed-field
* format. Offset/size values are 11-byte octal fields (same
* format as 'size' field in ustart header). These are
* stored in the header, allocating subsequent header blocks
* as needed. Extending the header in this way is a pretty
* severe POSIX violation; this design has earned GNU tar a
* lot of criticism.
*/
static int
gnu_sparse_old_read(struct archive_read *a, struct tar *tar,
const struct archive_entry_header_gnutar *header)
{
ssize_t bytes_read;
const void *data;
struct extended {
struct gnu_sparse sparse[21];
char isextended[1];
char padding[7];
};
const struct extended *ext;
gnu_sparse_old_parse(tar, header->sparse, 4);
if (header->isextended[0] == 0)
return (ARCHIVE_OK);
do {
bytes_read = (a->decompressor->read_ahead)(a, &data, 512);
if (bytes_read < 0)
return (ARCHIVE_FATAL);
if (bytes_read < 512) {
archive_set_error(&a->archive, ARCHIVE_ERRNO_FILE_FORMAT,
"Truncated tar archive "
"detected while reading sparse file data");
return (ARCHIVE_FATAL);
}
(a->decompressor->consume)(a, 512);
ext = (const struct extended *)data;
gnu_sparse_old_parse(tar, ext->sparse, 21);
} while (ext->isextended[0] != 0);
if (tar->sparse_list != NULL)
tar->entry_offset = tar->sparse_list->offset;
return (ARCHIVE_OK);
}
static void
gnu_sparse_old_parse(struct tar *tar,
const struct gnu_sparse *sparse, int length)
{
while (length > 0 && sparse->offset[0] != 0) {
gnu_add_sparse_entry(tar,
tar_atol(sparse->offset, sizeof(sparse->offset)),
tar_atol(sparse->numbytes, sizeof(sparse->numbytes)));
sparse++;
length--;
}
}
/*
* GNU tar sparse format 0.0
*
* Beginning with GNU tar 1.15, sparse files are stored using
* information in the pax extended header. The GNU tar maintainers
* have gone through a number of variations in the process of working
* out this scheme; furtunately, they're all numbered.
*
* Sparse format 0.0 uses attribute GNU.sparse.numblocks to store the
* number of blocks, and GNU.sparse.offset/GNU.sparse.numbytes to
* store offset/size for each block. The repeated instances of these
* latter fields violate the pax specification (which frowns on
* duplicate keys), so this format was quickly replaced.
*/
/*
* GNU tar sparse format 0.1
*
* This version replaced the offset/numbytes attributes with
* a single "map" attribute that stored a list of integers. This
* format had two problems: First, the "map" attribute could be very
* long, which caused problems for some implementations. More
* importantly, the sparse data was lost when extracted by archivers
* that didn't recognize this extension.
*/
static int
gnu_sparse_01_parse(struct tar *tar, const wchar_t *p)
{
const wchar_t *e;
off_t offset = -1, size = -1;
for (;;) {
e = p;
while (*e != '\0' && *e != ',') {
if (*e < '0' || *e > '9')
return (ARCHIVE_WARN);
e++;
}
if (offset < 0) {
offset = tar_atol10(p, e - p);
if (offset < 0)
return (ARCHIVE_WARN);
} else {
size = tar_atol10(p, e - p);
if (size < 0)
return (ARCHIVE_WARN);
gnu_add_sparse_entry(tar, offset, size);
offset = -1;
}
if (*e == '\0')
return (ARCHIVE_OK);
p = e + 1;
}
}
/*
* GNU tar sparse format 1.0
*
* The idea: The offset/size data is stored as a series of base-10
* ASCII numbers prepended to the file data, so that dearchivers that
* don't support this format will extract the block map along with the
* data and a separate post-process can restore the sparseness.
*
* Unfortunately, GNU tar 1.16 adds bogus padding to the end of the
* entry that depends on the size of the map; this means we have to
* parse the sparse map when we read the header (otherwise, entry_skip
* will fail). This is why sparse_10_read is called from read_header
* above, instead of at the beginning of read_data, where it "should"
* go.
*
* This variant also replaced GNU.sparse.size with GNU.sparse.realsize
* and introduced the GNU.sparse.major/GNU.sparse.minor attributes.
*/
/*
* Read the next line from the input, and parse it as a decimal
* integer followed by '\n'. Returns positive integer value or
* negative on error.
*/
static int64_t
gnu_sparse_10_atol(struct archive_read *a, struct tar *tar,
ssize_t *total_read)
{
int64_t l, limit, last_digit_limit;
const char *p;
ssize_t bytes_read;
int base, digit;
base = 10;
limit = INT64_MAX / base;
last_digit_limit = INT64_MAX % base;
bytes_read = readline(a, tar, &p);
if (bytes_read <= 0)
return (ARCHIVE_FATAL);
*total_read += bytes_read;
l = 0;
while (bytes_read > 0) {
if (*p == '\n')
return (l);
if (*p < '0' || *p >= '0' + base)
return (ARCHIVE_WARN);
digit = *p - '0';
if (l > limit || (l == limit && digit > last_digit_limit))
l = UINT64_MAX; /* Truncate on overflow. */
else
l = (l * base) + digit;
p++;
bytes_read--;
}
/* TODO: Error message. */
return (ARCHIVE_WARN);
}
/*
* Returns number of bytes consumed to read the sparse block data.
*/
static ssize_t
gnu_sparse_10_read(struct archive_read *a, struct tar *tar)
{
ssize_t bytes_read = 0;
int entries;
off_t offset, size, to_skip;
/* Parse entries. */
entries = gnu_sparse_10_atol(a, tar, &bytes_read);
if (entries < 0)
return (ARCHIVE_FATAL);
/* Parse the individual entries. */
while (entries-- > 0) {
/* Parse offset/size */
offset = gnu_sparse_10_atol(a, tar, &bytes_read);
if (offset < 0)
return (ARCHIVE_FATAL);
size = gnu_sparse_10_atol(a, tar, &bytes_read);
if (size < 0)
return (ARCHIVE_FATAL);
/* Add a new sparse entry. */
gnu_add_sparse_entry(tar, offset, size);
}
/* Skip rest of block... */
to_skip = 0x1ff & -bytes_read;
if (to_skip != (a->decompressor->skip)(a, to_skip))
return (ARCHIVE_FATAL);
return (bytes_read + to_skip);
}
/*-
* Convert text->integer.
*
* Traditional tar formats (including POSIX) specify base-8 for
* all of the standard numeric fields. This is a significant limitation
* in practice:
* = file size is limited to 8GB
* = rdevmajor and rdevminor are limited to 21 bits
* = uid/gid are limited to 21 bits
*
* There are two workarounds for this:
* = pax extended headers, which use variable-length string fields
* = GNU tar and STAR both allow either base-8 or base-256 in
* most fields. The high bit is set to indicate base-256.
*
* On read, this implementation supports both extensions.
*/
static int64_t
tar_atol(const char *p, unsigned char_cnt)
{
/*
* Technically, GNU tar considers a field to be in base-256
* only if the first byte is 0xff or 0x80.
*/
if (*p & 0x80)
return (tar_atol256(p, char_cnt));
return (tar_atol8(p, char_cnt));
}
/*
* Note that this implementation does not (and should not!) obey
* locale settings; you cannot simply substitute strtol here, since
* it does obey locale.
*/
static int64_t
tar_atol8(const char *p, unsigned char_cnt)
{
int64_t l, limit, last_digit_limit;
int digit, sign, base;
base = 8;
limit = INT64_MAX / base;
last_digit_limit = INT64_MAX % base;
while (*p == ' ' || *p == '\t')
p++;
if (*p == '-') {
sign = -1;
p++;
} else
sign = 1;
l = 0;
digit = *p - '0';
while (digit >= 0 && digit < base && char_cnt-- > 0) {
if (l>limit || (l == limit && digit > last_digit_limit)) {
l = UINT64_MAX; /* Truncate on overflow. */
break;
}
l = (l * base) + digit;
digit = *++p - '0';
}
return (sign < 0) ? -l : l;
}
/*
* Note that this implementation does not (and should not!) obey
* locale settings; you cannot simply substitute strtol here, since
* it does obey locale.
*/
static int64_t
tar_atol10(const wchar_t *p, unsigned char_cnt)
{
int64_t l, limit, last_digit_limit;
int base, digit, sign;
base = 10;
limit = INT64_MAX / base;
last_digit_limit = INT64_MAX % base;
while (*p == ' ' || *p == '\t')
p++;
if (*p == '-') {
sign = -1;
p++;
} else
sign = 1;
l = 0;
digit = *p - '0';
while (digit >= 0 && digit < base && char_cnt-- > 0) {
if (l > limit || (l == limit && digit > last_digit_limit)) {
l = UINT64_MAX; /* Truncate on overflow. */
break;
}
l = (l * base) + digit;
digit = *++p - '0';
}
return (sign < 0) ? -l : l;
}
/*
* Parse a base-256 integer. This is just a straight signed binary
* value in big-endian order, except that the high-order bit is
* ignored. Remember that "int64_t" may or may not be exactly 64
* bits; the implementation here tries to avoid making any assumptions
* about the actual size of an int64_t. It does assume we're using
* twos-complement arithmetic, though.
*/
static int64_t
tar_atol256(const char *_p, unsigned char_cnt)
{
int64_t l, upper_limit, lower_limit;
const unsigned char *p = (const unsigned char *)_p;
upper_limit = INT64_MAX / 256;
lower_limit = INT64_MIN / 256;
/* Pad with 1 or 0 bits, depending on sign. */
if ((0x40 & *p) == 0x40)
l = (int64_t)-1;
else
l = 0;
l = (l << 6) | (0x3f & *p++);
while (--char_cnt > 0) {
if (l > upper_limit) {
l = INT64_MAX; /* Truncate on overflow */
break;
} else if (l < lower_limit) {
l = INT64_MIN;
break;
}
l = (l << 8) | (0xff & (int64_t)*p++);
}
return (l);
}
/*
* Returns length of line (including trailing newline)
* or negative on error. 'start' argument is updated to
* point to first character of line. This avoids copying
* when possible.
*/
static ssize_t
readline(struct archive_read *a, struct tar *tar, const char **start)
{
ssize_t bytes_read;
ssize_t total_size = 0;
const void *t;
const char *s;
void *p;
bytes_read = (a->decompressor->read_ahead)(a, &t, 1);
if (bytes_read <= 0)
return (ARCHIVE_FATAL);
s = t; /* Start of line? */
p = memchr(t, '\n', bytes_read);
/* If we found '\n' in the read buffer, return pointer to that. */
if (p != NULL) {
bytes_read = 1 + ((const char *)p) - s;
(a->decompressor->consume)(a, bytes_read);
*start = s;
return (bytes_read);
}
/* Otherwise, we need to accumulate in a line buffer. */
for (;;) {
archive_string_ensure(&tar->line, total_size + bytes_read);
memcpy(tar->line.s + total_size, t, bytes_read);
(a->decompressor->consume)(a, bytes_read);
total_size += bytes_read;
/* Read some more. */
bytes_read = (a->decompressor->read_ahead)(a, &t, 1);
if (bytes_read <= 0)
return (ARCHIVE_FATAL);
s = t; /* Start of line? */
p = memchr(t, '\n', bytes_read);
/* If we found '\n', finish the line. */
if (p != NULL) {
bytes_read = 1 + ((const char *)p) - s;
(a->decompressor->consume)(a, bytes_read);
*start = tar->line.s;
return (total_size + bytes_read);
}
}
}
static int
utf8_decode(wchar_t *dest, const char *src, size_t length)
{
size_t n;
int err;
err = 0;
while (length > 0) {
n = UTF8_mbrtowc(dest, src, length);
if (n == 0)
break;
dest++;
src += n;
length -= n;
}
*dest++ = L'\0';
return (err);
}
/*
* Copied and simplified from FreeBSD libc/locale.
*/
static size_t
UTF8_mbrtowc(wchar_t *pwc, const char *s, size_t n)
{
int ch, i, len, mask;
unsigned long wch;
if (s == NULL || n == 0 || pwc == NULL)
return (0);
/*
* Determine the number of octets that make up this character from
* the first octet, and a mask that extracts the interesting bits of
* the first octet.
*/
ch = (unsigned char)*s;
if ((ch & 0x80) == 0) {
mask = 0x7f;
len = 1;
} else if ((ch & 0xe0) == 0xc0) {
mask = 0x1f;
len = 2;
} else if ((ch & 0xf0) == 0xe0) {
mask = 0x0f;
len = 3;
} else if ((ch & 0xf8) == 0xf0) {
mask = 0x07;
len = 4;
} else if ((ch & 0xfc) == 0xf8) {
mask = 0x03;
len = 5;
} else if ((ch & 0xfe) == 0xfc) {
mask = 0x01;
len = 6;
} else {
/* Invalid first byte; convert to '?' */
*pwc = '?';
return (1);
}
if (n < (size_t)len) {
/* Invalid first byte; convert to '?' */
*pwc = '?';
return (1);
}
/*
* Decode the octet sequence representing the character in chunks
* of 6 bits, most significant first.
*/
wch = (unsigned char)*s++ & mask;
i = len;
while (--i != 0) {
if ((*s & 0xc0) != 0x80) {
/* Invalid intermediate byte; consume one byte and
* emit '?' */
*pwc = '?';
return (1);
}
wch <<= 6;
wch |= *s++ & 0x3f;
}
/* Assign the value to the output; out-of-range values
* just get truncated. */
*pwc = (wchar_t)wch;
#ifdef WCHAR_MAX
/*
* If platform has WCHAR_MAX, we can do something
* more sensible with out-of-range values.
*/
if (wch >= WCHAR_MAX)
*pwc = '?';
#endif
/* Return number of bytes input consumed: 0 for end-of-string. */
return (wch == L'\0' ? 0 : len);
}
/*
* base64_decode - Base64 decode
*
* This accepts most variations of base-64 encoding, including:
* * with or without line breaks
* * with or without the final group padded with '=' or '_' characters
* (The most economical Base-64 variant does not pad the last group and
* omits line breaks; RFC1341 used for MIME requires both.)
*/
static char *
base64_decode(const wchar_t *src, size_t len, size_t *out_len)
{
static const unsigned char digits[64] = {
'A','B','C','D','E','F','G','H','I','J','K','L','M','N',
'O','P','Q','R','S','T','U','V','W','X','Y','Z','a','b',
'c','d','e','f','g','h','i','j','k','l','m','n','o','p',
'q','r','s','t','u','v','w','x','y','z','0','1','2','3',
'4','5','6','7','8','9','+','/' };
static unsigned char decode_table[128];
char *out, *d;
/* If the decode table is not yet initialized, prepare it. */
if (decode_table[digits[1]] != 1) {
size_t i;
memset(decode_table, 0xff, sizeof(decode_table));
for (i = 0; i < sizeof(digits); i++)
decode_table[digits[i]] = i;
}
/* Allocate enough space to hold the entire output. */
/* Note that we may not use all of this... */
out = (char *)malloc((len * 3 + 3) / 4);
if (out == NULL) {
*out_len = 0;
return (NULL);
}
d = out;
while (len > 0) {
/* Collect the next group of (up to) four characters. */
int v = 0;
int group_size = 0;
while (group_size < 4 && len > 0) {
/* '=' or '_' padding indicates final group. */
if (*src == '=' || *src == '_') {
len = 0;
break;
}
/* Skip illegal characters (including line breaks) */
if (*src > 127 || *src < 32
|| decode_table[*src] == 0xff) {
len--;
src++;
continue;
}
v <<= 6;
v |= decode_table[*src++];
len --;
group_size++;
}
/* Align a short group properly. */
v <<= 6 * (4 - group_size);
/* Unpack the group we just collected. */
switch (group_size) {
case 4: d[2] = v & 0xff;
/* FALLTHROUGH */
case 3: d[1] = (v >> 8) & 0xff;
/* FALLTHROUGH */
case 2: d[0] = (v >> 16) & 0xff;
break;
case 1: /* this is invalid! */
break;
}
d += group_size * 3 / 4;
}
*out_len = d - out;
return (out);
}
/*
* This is a little tricky because the C99 standard wcstombs()
* function returns the number of bytes that were converted,
* not the number that should be converted. As a result,
* we can never accurately size the output buffer (without
* doing a tedious output size calculation in advance).
* This approach (try a conversion, then try again if it fails)
* will almost always succeed on the first try, and is thus
* much faster, at the cost of sometimes requiring multiple
* passes while we expand the buffer.
*/
static char *
wide_to_narrow(const wchar_t *wval)
{
int converted_length;
/* Guess an output buffer size and try the conversion. */
int alloc_length = wcslen(wval) * 3;
char *mbs_val = (char *)malloc(alloc_length + 1);
if (mbs_val == NULL)
return (NULL);
converted_length = wcstombs(mbs_val, wval, alloc_length);
/* If we exhausted the buffer, resize and try again. */
while (converted_length >= alloc_length) {
free(mbs_val);
alloc_length *= 2;
mbs_val = (char *)malloc(alloc_length + 1);
if (mbs_val == NULL)
return (NULL);
converted_length = wcstombs(mbs_val, wval, alloc_length);
}
/* Ensure a trailing null and return the final string. */
mbs_val[alloc_length] = '\0';
return (mbs_val);
}
static char *
url_decode(const char *in)
{
char *out, *d;
const char *s;
out = (char *)malloc(strlen(in) + 1);
if (out == NULL)
return (NULL);
for (s = in, d = out; *s != '\0'; ) {
if (*s == '%') {
/* Try to convert % escape */
int digit1 = tohex(s[1]);
int digit2 = tohex(s[2]);
if (digit1 >= 0 && digit2 >= 0) {
/* Looks good, consume three chars */
s += 3;
/* Convert output */
*d++ = ((digit1 << 4) | digit2);
continue;
}
/* Else fall through and treat '%' as normal char */
}
*d++ = *s++;
}
*d = '\0';
return (out);
}
static int
tohex(int c)
{
if (c >= '0' && c <= '9')
return (c - '0');
else if (c >= 'A' && c <= 'F')
return (c - 'A' + 10);
else if (c >= 'a' && c <= 'f')
return (c - 'a' + 10);
else
return (-1);
}
|