This file is indexed.

/usr/share/perl5/Mail/SpamAssassin.pm is in spamassassin 3.4.2-0ubuntu0.14.04.1.

This file is owned by root:root, with mode 0o644.

The actual contents of the file can be viewed below.

   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
2239
2240
2241
2242
2243
2244
2245
2246
2247
2248
2249
2250
2251
2252
2253
2254
2255
2256
2257
2258
2259
2260
2261
2262
2263
2264
2265
2266
2267
2268
2269
2270
2271
2272
2273
2274
2275
2276
2277
2278
2279
2280
2281
2282
2283
2284
2285
2286
2287
2288
2289
2290
2291
2292
2293
2294
2295
2296
2297
2298
2299
2300
2301
2302
2303
2304
2305
2306
2307
2308
2309
# <@LICENSE>
# Licensed to the Apache Software Foundation (ASF) under one or more
# contributor license agreements.  See the NOTICE file distributed with
# this work for additional information regarding copyright ownership.
# The ASF licenses this file to you under the Apache License, Version 2.0
# (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.apache.org/licenses/LICENSE-2.0
# 
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# </@LICENSE>

=head1 NAME

Mail::SpamAssassin - Spam detector and markup engine

=head1 SYNOPSIS

  my $spamtest = Mail::SpamAssassin->new();
  my $mail = $spamtest->parse($message);
  my $status = $spamtest->check($mail);

  if ($status->is_spam()) {
    $message = $status->rewrite_mail();
  }
  else {
    ...
  }
  ...

  $status->finish();
  $mail->finish();
  $spamtest->finish();

=head1 DESCRIPTION

Mail::SpamAssassin is a module to identify spam using several methods
including text analysis, internet-based realtime blacklists, statistical
analysis, and internet-based hashing algorithms.

Using its rule base, it uses a wide range of heuristic tests on mail
headers and body text to identify "spam", also known as unsolicited bulk
email.  Once identified as spam, the mail can then be tagged as spam for
later filtering using the user's own mail user agent application or at
the mail transfer agent.

If you wish to use a command-line filter tool, try the C<spamassassin>
or the C<spamd>/C<spamc> tools provided.

=head1 METHODS

=over 4

=cut

package Mail::SpamAssassin;
use strict;
use warnings;
# use bytes;
use re 'taint';

require 5.006_001;

use Mail::SpamAssassin::Logger;
use Mail::SpamAssassin::Constants;
use Mail::SpamAssassin::Conf;
use Mail::SpamAssassin::Conf::SQL;
use Mail::SpamAssassin::Conf::LDAP;
use Mail::SpamAssassin::PerMsgStatus;
use Mail::SpamAssassin::Message;
use Mail::SpamAssassin::PluginHandler;
use Mail::SpamAssassin::DnsResolver;
use Mail::SpamAssassin::RegistryBoundaries;
use Mail::SpamAssassin::Util qw(untaint_var am_running_on_windows);
use Mail::SpamAssassin::Util::ScopedTimer;

use Errno qw(ENOENT EACCES);
use File::Basename;
use File::Path;
use File::Spec 0.8;
use Time::HiRes qw(time);
use Cwd;
use Config;

our $VERSION = "3.004002";      # update after release (same format as perl $])
#our $IS_DEVEL_BUILD = 1;        # 1 for devel build 
our $IS_DEVEL_BUILD = 0;        # 0 for release versions including rc & pre releases


# Used during the prerelease/release-candidate part of the official release
# process. If you hacked up your SA, you should add a version_tag to your .cf
# files; this variable should not be modified.
our @EXTRA_VERSION = qw();

our @ISA = qw();

# SUB_VERSION is now just <yyyy>-<mm>-<dd>
our $SUB_VERSION = 'svnunknown';
if ('$LastChangedDate: 2018-09-13 21:25:10 -0400 (Thu, 13 Sep 2018) $' =~ ':') {
  # Subversion keyword "$LastChangedDate: 2018-09-13 21:25:10 -0400 (Thu, 13 Sep 2018) $" has been successfully expanded.
  # Doesn't happen with automated launchpad builds:
  # https://bugs.launchpad.net/launchpad/+bug/780916
  $SUB_VERSION = (split(/\s+/,'$LastChangedDate: 2018-09-13 21:25:10 -0400 (Thu, 13 Sep 2018) $ updated by SVN'))[1];
}


if (defined $IS_DEVEL_BUILD && $IS_DEVEL_BUILD) {
  if ('$LastChangedRevision: 1840870 $' =~ ':') {
    # Subversion keyword "$LastChangedRevision: 1840870 $" has been successfully expanded.
    push(@EXTRA_VERSION, ('r' . qw{$LastChangedRevision: 1840870 $ updated by SVN}[1]));
  } else {
    push(@EXTRA_VERSION, ('r' . 'svnunknown'));
  }
}

sub Version {
  $VERSION =~ /^(\d+)\.(\d\d\d)(\d\d\d)$/;
  return join('-', sprintf("%d.%d.%d", $1, $2, $3), @EXTRA_VERSION);
}

our $HOME_URL = "http://spamassassin.apache.org/";

# note that the CWD takes priority.  This is required in case a user
# is testing a new version of SpamAssassin on a machine with an older
# version installed.  Unless you can come up with a fix for this that
# allows "make test" to work, don't change this.

our @default_rules_path = (
  '__local_state_dir__/__version__',
  '__def_rules_dir__',
  '__prefix__/share/spamassassin',
  '/usr/local/share/spamassassin',
  '/usr/share/spamassassin',
);

# first 3 are BSDish, latter 2 Linuxish
our @site_rules_path = (
  '__local_rules_dir__',
  '__prefix__/etc/mail/spamassassin',
  '__prefix__/etc/spamassassin',
  '/usr/local/etc/spamassassin',
  '/usr/pkg/etc/spamassassin',
  '/usr/etc/spamassassin',
  '/etc/mail/spamassassin',
  '/etc/spamassassin',
);

our @default_prefs_path = (
  '__local_rules_dir__/user_prefs.template',
  '__prefix__/etc/mail/spamassassin/user_prefs.template',
  '__prefix__/share/spamassassin/user_prefs.template',
  '__local_state_dir__/__version__/updates_spamassassin_org/user_prefs.template',
  '__def_rules_dir__/user_prefs.template',
  '/etc/spamassassin/user_prefs.template',
  '/etc/mail/spamassassin/user_prefs.template',
  '/usr/local/share/spamassassin/user_prefs.template',
  '/usr/share/spamassassin/user_prefs.template',
);

our @default_userprefs_path = (
  '~/.spamassassin/user_prefs',
);

our @default_userstate_dir = (
  '~/.spamassassin',
);

###########################################################################

=item $t = Mail::SpamAssassin->new( { opt => val, ... } )

Constructs a new C<Mail::SpamAssassin> object.  You may pass a hash
reference to the constructor which may contain the following attribute-
value pairs.

=over 4

=item debug

This is the debug options used to determine logging level.  It exists to
allow sections of debug messages (called "facilities") to be enabled or
disabled.  If this is a string, it is treated as a comma-delimited list
of the debug facilities.  If it's a hash reference, then the keys are
treated as the list of debug facilities and if it's a array reference,
then the elements are treated as the list of debug facilities.

There are also two special cases: (1) if the special case of "info" is
passed as a debug facility, then all informational messages are enabled;
(2) if the special case of "all" is passed as a debug facility, then all
debugging facilities are enabled.

=item rules_filename

The filename/directory to load spam-identifying rules from. (optional)

=item site_rules_filename

The filename/directory to load site-specific spam-identifying rules from.
(optional)

=item userprefs_filename

The filename to load preferences from. (optional)

=item userstate_dir

The directory user state is stored in. (optional)

=item config_tree_recurse

Set to C<1> to recurse through directories when reading configuration
files, instead of just reading a single level.  (optional, default 0)

=item config_text

The text of all rules and preferences.  If you prefer not to load the rules
from files, read them in yourself and set this instead.  As a result, this will
override the settings for C<rules_filename>, C<site_rules_filename>,
and C<userprefs_filename>.

=item pre_config_text

Similar to C<config_text>, this text is placed before config_text to allow an
override of config files.

=item post_config_text

Similar to C<config_text>, this text is placed after config_text to allow an
override of config files.

=item force_ipv4

If set to 1, DNS or other network tests will prefer IPv4 and not attempt
to use IPv6. Use if the existing tests for IPv6 availability produce
incorrect results or crashes.

=item force_ipv6

For symmetry with force_ipv4: if set to 1, DNS or other network tests
will prefer IPv6 and not attempt to use IPv4. Some plugins may disregard
this setting and use whatever protocol family they are comfortable with.

=item require_rules

If set to 1, init() will die if no valid rules could be loaded. This is the
default behaviour when called by C<spamassassin> or C<spamd>.

=item languages_filename

If you want to be able to use the language-guessing rule
C<UNWANTED_LANGUAGE_BODY>, and are using C<config_text> instead of
C<rules_filename>, C<site_rules_filename>, and C<userprefs_filename>, you will
need to set this.  It should be the path to the B<languages> file normally
found in the SpamAssassin B<rules> directory.

=item local_tests_only

If set to 1, no tests that require internet access will be performed. (default:
0)

=item need_tags

The option provides a way to avoid more expensive processing when it is known
in advance that some information will not be needed by a caller.

A value of the option can either be a string (a comma-delimited list of tag
names), or a reference to a list of individual tag names. A caller may provide
the list in advance, specifying his intention to later collect the information
through $pms->get_tag() calls. If a name of a tag starts with a 'NO' (case
insensitive), it shows that a caller will not be interested in such tag,
although there is no guarantee it would save any resources, nor that a tag
value will be empty. Currently no built-in tags start with 'NO'. A later
entry overrides previous one, e.g. ASN,NOASN,ASN,TIMING,NOASN is equivalent
to TIMING,NOASN.

For backward compatibility, all tags available as of version 3.2.4 will
be available by default (unless disabled by NOtag), even if not requested
through need_tags option. Future versions may provide new tags conditionally
available.

Currently the only tag that needs to be explicitly requested is 'TIMING'.
Not requesting it can save a millisecond or two - it mostly serves to
illustrate the usage of need_tags.

Example:
  need_tags =>    'TIMING,noLANGUAGES,RELAYCOUNTRY,ASN,noASNCIDR',
or:
  need_tags => [qw(TIMING noLANGUAGES RELAYCOUNTRY ASN noASNCIDR)],

=item ignore_site_cf_files

If set to 1, any rule files found in the C<site_rules_filename> directory will
be ignored.  *.pre files (used for loading plugins) found in the
C<site_rules_filename> directory will still be used. (default: 0)

=item dont_copy_prefs

If set to 1, the user preferences file will not be created if it doesn't
already exist. (default: 0)

=item save_pattern_hits

If set to 1, the patterns hit can be retrieved from the
C<Mail::SpamAssassin::PerMsgStatus> object.  Used for debugging.

=item home_dir_for_helpers

If set, the B<HOME> environment variable will be set to this value
when using test applications that require their configuration data,
such as Razor, Pyzor and DCC.

=item username

If set, the C<username> attribute will use this as the current user's name.
Otherwise, the default is taken from the runtime environment (ie. this process'
effective UID under UNIX).

=item skip_prng_reseeding

If skip_prng_reseeding is set to true, the SpamAssassin library will B<not>
call srand() to reseed a pseudo-random number generator (PRNG). The srand()
Perl function should be called during initialization of each child process,
soon after forking.

Prior to version 3.4.0, calling srand() was handled by the SpamAssassin
library.

This setting requires the caller to decide when to call srand().
This choice may be desired to preserve the entropy of a PRNG.  The default
value of skip_prng_reseeding is false to maintain backward compatibility. 

This option should only be set by a caller if it calls srand() upon spawning
child processes.  Unless you are certain you need it, leave this setting as
false.

NOTE: The skip_prng_reseeding feature is implemented in spamd as of 3.4.0
which allows spamd to call srand() right after forking a child process.

=back

If none of C<rules_filename>, C<site_rules_filename>, C<userprefs_filename>, or
C<config_text> is set, the C<Mail::SpamAssassin> module will search for the
configuration files in the usual installed locations using the below variable
definitions which can be passed in.

=over 4

=item PREFIX

Used as the root for certain directory paths such as:

  '__prefix__/etc/mail/spamassassin'
  '__prefix__/etc/spamassassin'

Defaults to "/usr".

=item DEF_RULES_DIR

Location where the default rules are installed.  Defaults to
"/usr/share/spamassassin".

=item LOCAL_RULES_DIR

Location where the local site rules are installed.  Defaults to
"/etc/spamassassin".

=item LOCAL_STATE_DIR

Location of the local state directory, mainly used for installing updates via
C<sa-update> and compiling rulesets to native code.  Defaults to
"/var/lib/spamassassin".

=back


=cut

# undocumented ctor settings: 
#
# - keep_config_parsing_metadata: used by build/listpromotable, default 0

sub new {
  my $class = shift;
  $class = ref($class) || $class;

  my $self = shift;
  if (!defined $self) { $self = { }; }
  bless ($self, $class);

  # basic backward compatibility; debug used to be a boolean.
  # translate that into 'all', which is what it meant before 3.1.0.
  if ($self->{debug} && $self->{debug} eq '1') {
    $self->{debug} = 'all';
  }

  # enable or disable debugging
  Mail::SpamAssassin::Logger::add_facilities($self->{debug});

  # first debugging information possibly printed should be the version
  dbg("generic: SpamAssassin version " . Version());

  # if the libs are installed in an alternate location, and the caller
  # didn't set PREFIX, we should have an estimated guess ready, values
  # substituted at 'make' time
  $self->{PREFIX}		||= '/usr';
  $self->{DEF_RULES_DIR}	||= '/usr/share/spamassassin';
  $self->{LOCAL_RULES_DIR}	||= '/etc/spamassassin';
  $self->{LOCAL_STATE_DIR}	||= '/var/lib/spamassassin';
  dbg("generic: Perl %s, %s", $], join(", ", map { $_ . '=' . $self->{$_} } 
      qw(PREFIX DEF_RULES_DIR LOCAL_RULES_DIR LOCAL_STATE_DIR)));

  $self->{needed_tags} = {};
  { my $ntags = $self->{need_tags};
    if (defined $ntags) {
      for my $t (ref $ntags ? @$ntags : split(/[, \s]+/,$ntags)) {
        $self->{needed_tags}->{$2} = !defined($1)  if $t =~ /^(NO)?(.+)\z/si;
      }
    }
  }
  if (would_log('dbg','timing') || $self->{needed_tags}->{TIMING}) {
    $self->timer_enable();
  }

  $self->{conf} ||= new Mail::SpamAssassin::Conf ($self);
  $self->{registryboundaries} = Mail::SpamAssassin::RegistryBoundaries->new ($self);
  $self->{plugins} = Mail::SpamAssassin::PluginHandler->new ($self);

  $self->{save_pattern_hits} ||= 0;

  # Make sure that we clean $PATH if we're tainted
  Mail::SpamAssassin::Util::clean_path_in_taint_mode();

  if (!defined $self->{username}) {
    $self->{username} = (Mail::SpamAssassin::Util::portable_getpwuid ($>))[0];
  }

  $self->create_locker();

  $self;
}

sub create_locker {
  my ($self) = @_;

  my $class;
  my $m = $self->{conf}->{lock_method};

  # let people choose what they want -- even if they may not work on their
  # OS.  (they could be using cygwin!)
  if ($m eq 'win32') { $class = 'Win32'; }
  elsif ($m eq 'flock') { $class = 'Flock'; }
  elsif ($m eq 'nfssafe') { $class = 'UnixNFSSafe'; }
  else {
    # OS-specific defaults
    if (am_running_on_windows()) {
      $class = 'Win32';
    } else {
      $class = 'UnixNFSSafe';
    }
  }

  # this could probably be made a little faster; for now I'm going
  # for slow but safe, by keeping in quotes
  eval '
    use Mail::SpamAssassin::Locker::'.$class.';
    $self->{locker} = new Mail::SpamAssassin::Locker::'.$class.' ($self);
    1;
  ' or do {
    my $eval_stat = $@ ne '' ? $@ : "errno=$!";  chomp $eval_stat;
    die "Mail::SpamAssassin::Locker::$class error: $eval_stat\n";
  };

  if (!defined $self->{locker}) { die "locker: oops! no locker"; }
}

###########################################################################

=item parse($message, $parse_now [, $suppl_attrib])

Parse will return a Mail::SpamAssassin::Message object with just the
headers parsed.  When calling this function, there are two optional
parameters that can be passed in: $message is either undef (which
will use STDIN), a scalar - a string containing an entire message,
a reference to such string, an array reference of the message with
one line per array element, or either a file glob or an IO::File object
which holds the entire contents of the message;  and $parse_now, which
specifies whether or not to create a MIME tree at parse time or later
as necessary.

The I<$parse_now> option, by default, is set to false (0).  This
allows SpamAssassin to not have to generate the tree of internal
data nodes if the information is not going to be used.  This is
handy, for instance, when running C<spamassassin -d>, which only
needs the pristine header and body which is always parsed and stored
by this function.

The optional last argument I<$suppl_attrib> provides a way for a caller
to pass additional information about a message to SpamAssassin. It is
either undef, or a ref to a hash where each key/value pair provides some
supplementary attribute of the message, typically information that cannot
be deduced from the message itself, or is hard to do so reliably, or would
represent unnecessary work for SpamAssassin to obtain it. The argument will
be stored to a Mail::SpamAssassin::Message object as 'suppl_attrib', thus
made available to the rest of the code as well as to plugins. The exact list
of attributes will evolve through time, any unknown attribute should be
ignored. Possible examples are: SMTP envelope information, a flag indicating
that a message as supplied by a caller was truncated due to size limit, an
already verified list of DKIM signature objects, or perhaps a list of rule
hits predetermined by a caller, which makes another possible way for a
caller to provide meta information (instead of having to insert made-up
header fields in order to pass information), or maybe just plain rule hits.

For more information, please see the C<Mail::SpamAssassin::Message>
and C<Mail::SpamAssassin::Message::Node> POD.

=cut

sub parse {
  my($self, $message, $parsenow, $suppl_attrib) = @_;

  my $start_time = time;
  $self->init(1);
  my $timer = $self->time_method("parse");

  my $master_deadline;
  # passed in at a function call
  if (ref $suppl_attrib && exists $suppl_attrib->{master_deadline}) {
    $master_deadline = $suppl_attrib->{master_deadline};  # may be undef
  }
  # found in a config file - overrides passed-in number if lower
  if ($self->{conf}->{time_limit}) {  # defined and nonzero
    my $time_limit_deadline = $start_time + $self->{conf}->{time_limit};
    if (!defined $master_deadline || $time_limit_deadline < $master_deadline) {
      $master_deadline = $time_limit_deadline;
    }
  }
  if (defined $master_deadline) {
    dbg("config: time limit %.1f s", $master_deadline - $start_time);
  }

  my $msg = Mail::SpamAssassin::Message->new({
    message=>$message, parsenow=>$parsenow,
    normalize=>$self->{conf}->{normalize_charset},
    master_deadline=>$master_deadline, suppl_attrib=>$suppl_attrib });

  # bug 5069: The goal here is to get rendering plugins to do things
  # like OCR, convert doc and pdf to text, etc, though it could be anything
  # that wants to process the message after it's been parsed.
  $self->call_plugins("post_message_parse", { message => $msg });

  return $msg;
}


###########################################################################

=item $status = $f->check ($mail)

Check a mail, encapsulated in a C<Mail::SpamAssassin::Message> object,
to determine if it is spam or not.

Returns a C<Mail::SpamAssassin::PerMsgStatus> object which can be
used to test or manipulate the mail message.

Note that the C<Mail::SpamAssassin> object can be re-used for further messages
without affecting this check; in OO terminology, the C<Mail::SpamAssassin>
object is a "factory".   However, if you do this, be sure to call the
C<finish()> method on the status objects when you're done with them.

=cut

sub check {
  my ($self, $mail_obj) = @_;

  $self->init(1);
  my $pms = Mail::SpamAssassin::PerMsgStatus->new($self, $mail_obj);
  $pms->check();
  dbg("timing: " . $self->timer_report())  if $self->{timer_enabled};
  $pms;
}

=item $status = $f->check_message_text ($mailtext)

Check a mail, encapsulated in a plain string C<$mailtext>, to determine if it
is spam or not.

Otherwise identical to C<check()> above.

=cut

sub check_message_text {
  my ($self, $mailtext) = @_;
  my $msg = $self->parse($mailtext, 1);
  my $result = $self->check($msg);

  # Kill off the metadata ...
  # Do _NOT_ call normal finish() here.  PerMsgStatus has a copy of
  # the message.  So killing it here will cause things like
  # rewrite_message() to fail. <grrr>
  #
  $msg->finish_metadata();

  return $result;
}

###########################################################################

=item $status = $f->learn ($mail, $id, $isspam, $forget)

Learn from a mail, encapsulated in a C<Mail::SpamAssassin::Message> object.

If C<$isspam> is set, the mail is assumed to be spam, otherwise it will
be learnt as non-spam.

If C<$forget> is set, the attributes of the mail will be removed from
both the non-spam and spam learning databases.

C<$id> is an optional message-identification string, used internally
to tag the message.  If it is C<undef>, the Message-Id of the message
will be used.  It should be unique to that message.

Returns a C<Mail::SpamAssassin::PerMsgLearner> object which can be used to
manipulate the learning process for each mail.

Note that the C<Mail::SpamAssassin> object can be re-used for further messages
without affecting this check; in OO terminology, the C<Mail::SpamAssassin>
object is a "factory".   However, if you do this, be sure to call the
C<finish()> method on the learner objects when you're done with them.

C<learn()> and C<check()> can be run using the same factory.  C<init_learner()>
must be called before using this method.

=cut

sub learn {
  my ($self, $mail_obj, $id, $isspam, $forget) = @_;
  local ($_);

  require Mail::SpamAssassin::PerMsgLearner;
  $self->init(1);
  my $msg = Mail::SpamAssassin::PerMsgLearner->new($self, $mail_obj);

  if ($forget) {
    dbg("learn: forgetting message");
    $msg->forget($id);
  } elsif ($isspam) {
    dbg("learn: learning spam");
    $msg->learn_spam($id);
  } else {
    dbg("learn: learning ham");
    $msg->learn_ham($id);
  }

  $msg;
}

###########################################################################

=item $f->init_learner ( [ { opt => val, ... } ] )

Initialise learning.  You may pass the following attribute-value pairs to this
method.

=over 4

=item caller_will_untie

Whether or not the code calling this method will take care of untie'ing
from the Bayes databases (by calling C<finish_learner()>) (optional, default 0).

=item force_expire

Should an expiration run be forced to occur immediately? (optional, default 0).

=item learn_to_journal

Should learning data be written to the journal, instead of directly to the
databases? (optional, default 0).

=item wait_for_lock

Whether or not to wait a long time for locks to complete (optional, default 0).

=item opportunistic_expire_check_only

During the opportunistic journal sync and expire check, don't actually do the
expire but report back whether or not it should occur (optional, default 0).

=item no_relearn

If doing a learn operation, and the message has already been learned as
the opposite type, don't re-learn the message.

=back

=cut

sub init_learner {
  my $self = shift;
  my $opts = shift;
  dbg("learn: initializing learner");

  # Make sure we're already initialized ...
  $self->init(1);

  my %kv = (
    'force_expire'			=> 'learn_force_expire',
    'learn_to_journal'			=> 'learn_to_journal',
    'caller_will_untie'			=> 'learn_caller_will_untie',
    'wait_for_lock'			=> 'learn_wait_for_lock',
    'opportunistic_expire_check_only'	=> 'opportunistic_expire_check_only',
    'no_relearn'			=> 'learn_no_relearn',
  );

  my %ret;

  # Set any other options that need setting ...
  while( my($k,$v) = each %kv ) {
    $ret{$k} = $self->{$v};
    if (exists $opts->{$k}) { $self->{$v} = $opts->{$k}; }
  }

  return \%ret;
}

###########################################################################

=item $f->rebuild_learner_caches ({ opt => val })

Rebuild any cache databases; should be called after the learning process.
Options include: C<verbose>, which will output diagnostics to C<stdout>
if set to 1.

=cut

sub rebuild_learner_caches {
  my $self = shift;
  my $opts = shift;
  $self->{bayes_scanner}->sync(1,1,$opts) if $self->{bayes_scanner};
  1;
}

=item $f->finish_learner ()

Finish learning.

=cut

sub finish_learner {
  my $self = shift;
  $self->{bayes_scanner}->force_close(1) if $self->{bayes_scanner};
  1;
}

=item $f->dump_bayes_db()

Dump the contents of the Bayes DB

=cut

sub dump_bayes_db {
  my($self,@opts) = @_;
  $self->{bayes_scanner}->dump_bayes_db(@opts) if $self->{bayes_scanner};
}

=item $f->signal_user_changed ( [ { opt => val, ... } ] )

Signals that the current user has changed (possibly using C<setuid>), meaning
that SpamAssassin should close any per-user databases it has open, and re-open
using ones appropriate for the new user.

Note that this should be called I<after> reading any per-user configuration, as
that data may override some paths opened in this method.  You may pass the
following attribute-value pairs:

=over 4

=item username

The username of the user.  This will be used for the C<username> attribute.

=item user_dir

A directory to use as a 'home directory' for the current user's data,
overriding the system default.  This directory must be readable and writable by
the process.  Note that the resulting C<userstate_dir> will be the
C<.spamassassin> subdirectory of this dir.

=item userstate_dir

A directory to use as a directory for the current user's data, overriding the
system default.  This directory must be readable and writable by the process.
The default is C<user_dir/.spamassassin>.

=back

=cut

sub signal_user_changed {
  my $self = shift;
  my $opts = shift;
  my $set = 0;

  my $timer = $self->time_method("signal_user_changed");
  dbg("info: user has changed");

  if (defined $opts && $opts->{username}) {
    $self->{username} = $opts->{username};
  } else {
    undef $self->{username};
  }
  if (defined $opts && $opts->{user_dir}) {
    $self->{user_dir} = $opts->{user_dir};
  } else {
    undef $self->{user_dir};
  }
  if (defined $opts && $opts->{userstate_dir}) {
    $self->{userstate_dir} = $opts->{userstate_dir};
  } else {
    undef $self->{userstate_dir};
  }

  # reopen bayes dbs for this user
  $self->{bayes_scanner}->finish() if $self->{bayes_scanner};
  if ($self->{conf}->{use_bayes}) {
      require Mail::SpamAssassin::Bayes;
      $self->{bayes_scanner} = new Mail::SpamAssassin::Bayes ($self);
  } else {
      delete $self->{bayes_scanner} if $self->{bayes_scanner};
  }

  # this user may have a different learn_to_journal setting, so reset appropriately
  $self->{'learn_to_journal'} = $self->{conf}->{bayes_learn_to_journal};

  $set |= 1 unless $self->{local_tests_only};
  $set |= 2 if $self->{bayes_scanner} && $self->{bayes_scanner}->is_scan_available() && $self->{conf}->{use_bayes_rules};

  $self->{conf}->set_score_set ($set);

  $self->call_plugins("signal_user_changed", {
		username => $self->{username},
		userstate_dir => $self->{userstate_dir},
		user_dir => $self->{user_dir},
	      });

  1;
}

###########################################################################

=item $f->report_as_spam ($mail, $options)

Report a mail, encapsulated in a C<Mail::SpamAssassin::Message> object, as
human-verified spam.  This will submit the mail message to live,
collaborative, spam-blocker databases, allowing other users to block this
message.

It will also submit the mail to SpamAssassin's Bayesian learner.

Options is an optional reference to a hash of options.  Currently these
can be:

=over 4

=item dont_report_to_dcc

Inhibits reporting of the spam to DCC.

=item dont_report_to_pyzor

Inhibits reporting of the spam to Pyzor.

=item dont_report_to_razor

Inhibits reporting of the spam to Razor.

=item dont_report_to_spamcop

Inhibits reporting of the spam to SpamCop.

=back

=cut

sub report_as_spam {
  my ($self, $mail, $options) = @_;
  local ($_);

  $self->init(1);
  my $timer = $self->time_method("report_as_spam");

  # learn as spam if enabled
  if ( $self->{conf}->{bayes_learn_during_report} ) {
    $self->learn ($mail, undef, 1, 0);
  }

  require Mail::SpamAssassin::Reporter;
  $mail = Mail::SpamAssassin::Reporter->new($self, $mail, $options);
  $mail->report();
}

###########################################################################

=item $f->revoke_as_spam ($mail, $options)

Revoke a mail, encapsulated in a C<Mail::SpamAssassin::Message> object, as
human-verified ham (non-spam).  This will revoke the mail message from live,
collaborative, spam-blocker databases, allowing other users to block this
message.

It will also submit the mail to SpamAssassin's Bayesian learner as nonspam.

Options is an optional reference to a hash of options.  Currently these
can be:

=over 4

=item dont_report_to_razor

Inhibits revoking of the spam to Razor.


=back

=cut

sub revoke_as_spam {
  my ($self, $mail, $options) = @_;
  local ($_);

  $self->init(1);
  my $timer = $self->time_method("revoke_as_spam");

  # learn as nonspam
  $self->learn ($mail, undef, 0, 0);

  require Mail::SpamAssassin::Reporter;
  $mail = Mail::SpamAssassin::Reporter->new($self, $mail, $options);
  $mail->revoke ();
}

###########################################################################

=item $f->add_address_to_whitelist ($addr, $cli_p)

Given a string containing an email address, add it to the automatic
whitelist database.

If $cli_p is set then underlying plugin may give visual feedback on additions/failures.

=cut

sub add_address_to_whitelist {
  my ($self, $addr, $cli_p) = @_;

  $self->call_plugins("whitelist_address", { address => $addr,
                                             cli_p => $cli_p });
}

###########################################################################

=item $f->add_all_addresses_to_whitelist ($mail, $cli_p)

Given a mail message, find as many addresses in the usual headers (To, Cc, From
etc.), and the message body, and add them to the automatic whitelist database.

If $cli_p is set then underlying plugin may give visual feedback on additions/failures.

=cut

sub add_all_addresses_to_whitelist {
  my ($self, $mail_obj, $cli_p) = @_;

  foreach my $addr ($self->find_all_addrs_in_mail ($mail_obj)) {
    $self->call_plugins("whitelist_address", { address => $addr,
                                               cli_p => $cli_p });
  }
}

###########################################################################

=item $f->remove_address_from_whitelist ($addr, $cli_p)

Given a string containing an email address, remove it from the automatic
whitelist database.

If $cli_p is set then underlying plugin may give visual feedback on additions/failures.

=cut

sub remove_address_from_whitelist {
  my ($self, $addr, $cli_p) = @_;

  $self->call_plugins("remove_address", { address => $addr,
                                          cli_p => $cli_p });
}

###########################################################################

=item $f->remove_all_addresses_from_whitelist ($mail, $cli_p)

Given a mail message, find as many addresses in the usual headers (To, Cc, From
etc.), and the message body, and remove them from the automatic whitelist
database.

If $cli_p is set then underlying plugin may give visual feedback on additions/failures.

=cut

sub remove_all_addresses_from_whitelist {
  my ($self, $mail_obj, $cli_p) = @_;

  foreach my $addr ($self->find_all_addrs_in_mail ($mail_obj)) {
    $self->call_plugins("remove_address", { address => $addr,
                                            cli_p => $cli_p });
  }
}

###########################################################################

=item $f->add_address_to_blacklist ($addr, $cli_p)

Given a string containing an email address, add it to the automatic
whitelist database with a high score, effectively blacklisting them.

If $cli_p is set then underlying plugin may give visual feedback on additions/failures.

=cut

sub add_address_to_blacklist {
  my ($self, $addr, $cli_p) = @_;
  $self->call_plugins("blacklist_address", { address => $addr,
                                             cli_p => $cli_p });
}

###########################################################################

=item $f->add_all_addresses_to_blacklist ($mail, $cli_p)

Given a mail message, find addresses in the From headers and add them to the
automatic whitelist database with a high score, effectively blacklisting them.

Note that To and Cc addresses are not used.

If $cli_p is set then underlying plugin may give visual feedback on additions/failures.

=cut

sub add_all_addresses_to_blacklist {
  my ($self, $mail_obj, $cli_p) = @_;

  $self->init(1);

  my @addrlist;
  my @hdrs = $mail_obj->get_header('From');
  if ($#hdrs >= 0) {
    push (@addrlist, $self->find_all_addrs_in_line (join (" ", @hdrs)));
  }

  foreach my $addr (@addrlist) {
    $self->call_plugins("blacklist_address", { address => $addr,
                                               cli_p => $cli_p });
  }

}

###########################################################################

=item $text = $f->remove_spamassassin_markup ($mail)

Returns the text of the message, with any SpamAssassin-added text (such
as the report, or X-Spam-Status headers) stripped.

Note that the B<$mail> object is not modified.

Warning: if the input message in B<$mail> contains a mixture of CR-LF
(Windows-style) and LF (UNIX-style) line endings, it will be "canonicalized"
to use one or the other consistently throughout.

=cut

sub remove_spamassassin_markup {
  my ($self, $mail_obj) = @_;
  local ($_);

  my $timer = $self->time_method("remove_spamassassin_markup");
  my $mbox = $mail_obj->get_mbox_separator() || '';

  dbg("markup: removing markup");

  # Go looking for a "report_safe" encapsulated message.  Abort out ASAP
  # if we have definitive proof it's not an encapsulated message.
  my $ct = $mail_obj->get_header("Content-Type") || '';
  if ( $ct =~ m!^\s*multipart/mixed;\s+boundary\s*=\s*["']?(.+?)["']?(?:;|$)!i ) {

    # Ok, this is a possible encapsulated message, search for the
    # appropriate mime part and deal with it if necessary.
    my $boundary = "\Q$1\E";
    my @msg = split(/^/,$mail_obj->get_pristine_body());

    my $flag = 0;
    $ct   = '';
    my $cd = '';
    for ( my $i = 0 ; $i <= $#msg ; $i++ ) {
      # only look at mime part headers
      next unless ( $msg[$i] =~ /^--$boundary\r?$/ || $flag );

      if ( $msg[$i] =~ /^\s*$/ ) {    # end of mime header

        # Ok, we found the encapsulated piece ...
	if ($ct =~ m@^(?:message/rfc822|text/plain);\s+x-spam-type=original@ ||
	    ($ct eq "message/rfc822" &&
	     $cd eq $self->{conf}->{'encapsulated_content_description'}))
        {
          splice @msg, 0, $i+1;  # remove the front part, including the blank line

          # find the end and chop it off
          for ( $i = 0 ; $i <= $#msg ; $i++ ) {
            if ( $msg[$i] =~ /^--$boundary/ ) {
              splice @msg, ($msg[$i-1] =~ /\S/ ? $i : $i-1);
	      # will remove the blank line (not sure it'll always be
	      # there) and everything below.  don't worry, the splice
	      # guarantees the for will stop ...
            }
          }

	  # Ok, we're done.  Return the rewritten message.
	  return join('', $mbox, @msg);
        }

        $flag = 0;
        $ct   = '';
        $cd   = '';
        next;
      }

      # Ok, we're in the mime header ...  Capture the appropriate headers...
      $flag = 1;
      if ( $msg[$i] =~ /^Content-Type:\s+(.+?)\s*$/i ) {
        $ct = $1;
      }
      elsif ( $msg[$i] =~ /^Content-Description:\s+(.+?)\s*$/i ) {
        $cd = $1;
      }
    }
  }

  # Ok, if we got here, the message wasn't a report_safe encapsulated message.
  # So treat it like a "report_safe 0" message.
  my $hdrs = $mail_obj->get_pristine_header();
  my $body = $mail_obj->get_pristine_body();

  # remove DOS line endings
  $hdrs =~ s/\r//gs;

  # unfold SA added headers, but not X-Spam-Prev headers ...
  $hdrs = "\n".$hdrs;   # simplifies regexp below
  1 while $hdrs =~ s/(\nX-Spam-(?!Prev).+?)\n[ \t]+(\S.*\n)/$1 $2/g;
  $hdrs =~ s/^\n//;

###########################################################################
  # Backward Compatibilty, pre 3.0.x.

  # deal with rewritten headers w/out X-Spam-Prev- versions ...
  $self->init(1);
  foreach my $header ( keys %{$self->{conf}->{rewrite_header}} ) {
    # let the 3.0 decoding do it...
    next if ($hdrs =~ /^X-Spam-Prev-$header:/im);

    dbg("markup: removing markup in $header");
    if ($header eq 'Subject') {
      my $tag = $self->{conf}->{rewrite_header}->{'Subject'};
      $tag = quotemeta($tag);
      $tag =~ s/_HITS_/\\d{2}\\.\\d{2}/g;
      $tag =~ s/_SCORE_/\\d{2}\\.\\d{2}/g;
      $tag =~ s/_REQD_/\\d{2}\\.\\d{2}/g;
      1 while $hdrs =~ s/^Subject: ${tag} /Subject: /gm;
    } else {
      $hdrs =~ s/^(${header}:[ \t].*?)\t\([^)]*\)$/$1/gm;
    }
  }

  # Now deal with report cleansing from 2.4x and previous.
  # possibly a blank line, "SPAM: ----.+", followed by "SPAM: stuff" lines,
  # followed by another "SPAM: ----.+" line, followed by a blank line.
  1 while ($body =~ s/^\n?SPAM: ----.+\n(?:SPAM:.*\n)*SPAM: ----.+\n\n//);
###########################################################################

  # 3.0 version -- support for previously-nonexistent Subject hdr.
  # ensure the Subject line didn't *really* contain "(nonexistent)" in
  # the original message!
  if ($hdrs =~ /^X-Spam-Prev-Subject:\s*\(nonexistent\)$/m
        && $hdrs !~ /^Subject:.*\(nonexistent\).*$/m)
  {
    $hdrs =~ s/(^|\n)X-Spam-Prev-Subject:\s*\(nonexistent\)\n/$1\n/s;
    $hdrs =~ s/(^|\n)Subject:\s*[ \t]*.*\n(?:\s+\S.*\n)*/$1\n/s;
  }

  # 3.0 version -- revert from X-Spam-Prev to original ...
  while ($hdrs =~ s/^X-Spam-Prev-(([^:]+:)[ \t]*.*\n(?:\s+\S.*\n)*)//m) {
    my($hdr, $name) = ($1,$2);

    # If the rewritten version doesn't exist, we should deal with it anyway...
    unless ($hdrs =~ s/^$name[ \t]*.*\n(?:\s+\S.*\n)*/$hdr/m) {
      $hdrs =~ s/\n\n/\n$hdr\n/;
    }
  }

  # remove any other X-Spam headers we added, will be unfolded
  $hdrs = "\n".$hdrs;   # simplifies regexp below
  1 while $hdrs =~ s/\nX-Spam-.*\n/\n/g;
  $hdrs =~ s/^\n//;

  # re-add DOS line endings
  if ($mail_obj->{line_ending} ne "\n") {
    $hdrs =~ s/\r?\n/$mail_obj->{line_ending}/gs;
  }

  # Put the whole thing back together ...
  return join ('', $mbox, $hdrs, $body);
}

###########################################################################

=item $f->read_scoreonly_config ($filename)

Read a configuration file and parse user preferences from it.

User preferences are as defined in the C<Mail::SpamAssassin::Conf> manual page.
In other words, they include scoring options, scores, whitelists and
blacklists, and so on, but do not include rule definitions, privileged
settings, etc. unless C<allow_user_rules> is enabled; and they never include
the administrator settings.

=cut

sub read_scoreonly_config {
  my ($self, $filename) = @_;

  my $timer = $self->time_method("read_scoreonly_config");
  local *IN;
  if (!open(IN,"<$filename")) {
    # the file may not exist; this should not be verbose
    dbg("config: read_scoreonly_config: cannot open \"$filename\": $!");
    return;
  }

  my($inbuf,$nread,$text); $text = '';
  while ( $nread=read(IN,$inbuf,16384) ) { $text .= $inbuf }
  defined $nread  or die "error reading $filename: $!";
  close IN  or die "error closing $filename: $!";
  undef $inbuf;

  $text = "file start $filename\n" . $text;
  # add an extra \n in case file did not end in one.
  $text .= "\nfile end $filename\n";

  $self->{conf}->{main} = $self;
  $self->{conf}->parse_scores_only ($text);
  $self->{conf}->finish_parsing(1);

  delete $self->{conf}->{main};	# to allow future GC'ing
}

###########################################################################

=item $f->load_scoreonly_sql ($username)

Read configuration paramaters from SQL database and parse scores from it.  This
will only take effect if the perl C<DBI> module is installed, and the
configuration parameters C<user_scores_dsn>, C<user_scores_sql_username>, and
C<user_scores_sql_password> are set correctly.

The username in C<$username> will also be used for the C<username> attribute of
the Mail::SpamAssassin object.

=cut

sub load_scoreonly_sql {
  my ($self, $username) = @_;

  my $timer = $self->time_method("load_scoreonly_sql");
  my $src = Mail::SpamAssassin::Conf::SQL->new ($self);
  $self->{username} = $username;
  unless ($src->load($username)) {
    return 0;
  }
  return 1;
}

###########################################################################

=item $f->load_scoreonly_ldap ($username)

Read configuration paramaters from an LDAP server and parse scores from it.
This will only take effect if the perl C<Net::LDAP> and C<URI> modules are
installed, and the configuration parameters C<user_scores_dsn>,
C<user_scores_ldap_username>, and C<user_scores_ldap_password> are set
correctly.

The username in C<$username> will also be used for the C<username> attribute of
the Mail::SpamAssassin object.

=cut

sub load_scoreonly_ldap {
  my ($self, $username) = @_;

  dbg("config: load_scoreonly_ldap($username)");
  my $timer = $self->time_method("load_scoreonly_ldap");
  my $src = Mail::SpamAssassin::Conf::LDAP->new ($self);
  $self->{username} = $username;
  $src->load($username);
}

###########################################################################

=item $f->set_persistent_address_list_factory ($factoryobj)

Set the persistent address list factory, used to create objects for the
automatic whitelist algorithm's persistent-storage back-end.  See
C<Mail::SpamAssassin::PersistentAddrList> for the API these factory objects
must implement, and the API the objects they produce must implement.

=cut

sub set_persistent_address_list_factory {
  my ($self, $fac) = @_;
  $self->{pers_addr_list_factory} = $fac;
}

###########################################################################

=item $f->compile_now ($use_user_prefs, $keep_userstate)

Compile all patterns, load all configuration files, and load all
possibly-required Perl modules.

Normally, Mail::SpamAssassin uses lazy evaluation where possible, but if you
plan to fork() or start a new perl interpreter thread to process a message,
this is suboptimal, as each process/thread will have to perform these actions.

Call this function in the master thread or process to perform the actions
straightaway, so that the sub-processes will not have to.

If C<$use_user_prefs> is 0, this will initialise the SpamAssassin
configuration without reading the per-user configuration file and it will
assume that you will call C<read_scoreonly_config> at a later point.

If C<$keep_userstate> is true, compile_now() will revert any configuration
options which have a default with I<__userstate__> in it post-init(),
and then re-change the option before returning.  This lets you change
I<$ENV{'HOME'}> to a temp directory, have compile_now() and create any
files there as necessary without disturbing the actual files as changed
by a configuration option.  By default, this is disabled.

=cut

sub compile_now {
  my ($self, $use_user_prefs, $deal_with_userstate) = @_;

  my $timer = $self->time_method("compile_now");

  # Backup default values which deal with userstate.
  # This is done so we can create any new files in, presumably, a temp dir.
  # see bug 2762 for more details.
  my %backup;
  if (defined $deal_with_userstate && $deal_with_userstate) {
    while(my($k,$v) = each %{$self->{conf}}) {
      $backup{$k} = $v if (defined $v && !ref($v) && $v =~/__userstate__/);
    }
  }

  $self->init($use_user_prefs);

  # if init() didn't change the value from default, forget about it.
  # if the value is different, remember the new version, and reset the default.
  while(my($k,$v) = each %backup) {
    if ($self->{conf}->{$k} eq $v) {
      delete $backup{$k};
    }
    else {
      my $backup = $backup{$k};
      $backup{$k} = $self->{conf}->{$k};
      $self->{conf}->{$k} = $backup;
    }
  }

  dbg("ignore: test message to precompile patterns and load modules");

  # tell plugins we are about to send a message for compiling purposes
  $self->call_plugins("compile_now_start",
		      { use_user_prefs => $use_user_prefs,
			keep_userstate => $deal_with_userstate});

  # note: this may incur network access. Good.  We want to make sure
  # as much as possible is preloaded!
  my @testmsg = ("From: ignore\@compiling.spamassassin.taint.org\n", 
    "Message-Id:  <".time."\@spamassassin_spamd_init>\n", "\n",
    "I need to make this message body somewhat long so TextCat preloads\n"x20);

  my $mail = $self->parse(\@testmsg, 1, { master_deadline => undef });
  my $status = Mail::SpamAssassin::PerMsgStatus->new($self, $mail,
                        { disable_auto_learning => 1 } );

  # We want to turn off the bayes rules for this test msg
  my $use_bayes_rules_value = $self->{conf}->{use_bayes_rules};
  $self->{conf}->{use_bayes_rules} = 0;
  $status->check();
  $self->{conf}->{use_bayes_rules} = $use_bayes_rules_value;
  $status->finish();
  $mail->finish();
  $self->finish_learner();

  $self->{conf}->free_uncompiled_rule_source();

  # load SQL modules now as well
  my $dsn = $self->{conf}->{user_scores_dsn};
  if ($dsn ne '') {
    if ($dsn =~ /^ldap:/i) {
      Mail::SpamAssassin::Conf::LDAP::load_modules();
    } else {
      Mail::SpamAssassin::Conf::SQL::load_modules();
    }
  }

  # make sure things are ready for scanning
  $self->{bayes_scanner}->force_close() if $self->{bayes_scanner};
  $self->call_plugins("compile_now_finish",
		      { use_user_prefs => $use_user_prefs,
			keep_userstate => $deal_with_userstate});

  # Reset any non-default values to the post-init() version.
  while(my($k,$v) = each %backup) {
    $self->{conf}->{$k} = $v;
  }

  # clear sed_path_cache
  delete $self->{conf}->{sed_path_cache};

  1;
}

###########################################################################

=item $f->debug_diagnostics ()

Output some diagnostic information, useful for debugging SpamAssassin
problems.

=cut

sub debug_diagnostics {
  my ($self) = @_;

  # load this class lazily, to avoid overhead when this method isn't
  # called.
  eval {
    require Mail::SpamAssassin::Util::DependencyInfo;
    dbg(Mail::SpamAssassin::Util::DependencyInfo::debug_diagnostics($self));
  };
}

###########################################################################

=item $failed = $f->lint_rules ()

Syntax-check the current set of rules.  Returns the number of 
syntax errors discovered, or 0 if the configuration is valid.

=cut

sub lint_rules {
  my ($self) = @_;

  dbg("ignore: using a test message to lint rules");
  my @testmsg = ("From: ignore\@compiling.spamassassin.taint.org\n", 
    "Subject: \n",
    "Message-Id:  <".CORE::time()."\@lint_rules>\n", "\n",
    "I need to make this message body somewhat long so TextCat preloads\n"x20);

  $self->{lint_rules} = $self->{conf}->{lint_rules} = 1;
  $self->{syntax_errors} = 0;

  my $olddcp = $self->{dont_copy_prefs};
  $self->{dont_copy_prefs} = 1;

  $self->init(1);
  $self->{syntax_errors} += $self->{conf}->{errors};

  $self->{dont_copy_prefs} = $olddcp;       # revert back to previous

  # bug 5048: override settings to ensure a faster lint
  $self->{'conf'}->{'use_auto_whitelist'} = 0;
  $self->{'conf'}->{'bayes_auto_learn'} = 0;

  my $mail = $self->parse(\@testmsg, 1, { master_deadline => undef });
  my $status = Mail::SpamAssassin::PerMsgStatus->new($self, $mail,
                        { disable_auto_learning => 1 } );
  $status->check();

  $self->{syntax_errors} += $status->{rule_errors};
  $status->finish();
  $mail->finish();
  dbg("timing: " . $self->timer_report())  if $self->{timer_enabled};
  return ($self->{syntax_errors});
}

###########################################################################

=item $f->finish()

Destroy this object, so that it will be garbage-collected once it
goes out of scope.  The object will no longer be usable after this
method is called.

=cut

sub finish {
  my ($self) = @_;

  $self->timer_start("finish");
  $self->call_plugins("finish_tests", { conf => $self->{conf},
                                        main => $self });

  $self->{conf}->finish(); delete $self->{conf};
  $self->{plugins}->finish(); delete $self->{plugins};

  if ($self->{bayes_scanner}) {
    $self->{bayes_scanner}->finish();
    delete $self->{bayes_scanner};
  }

  $self->{resolver}->finish()  if $self->{resolver};

  $self->timer_end("finish");
  %{$self} = ();
}

###########################################################################
# timers: bug 5356

sub timer_enable {
  my ($self) = @_;
  dbg("config: timing enabled")  if !$self->{timer_enabled};
  $self->{timer_enabled} = 1;
}

sub timer_disable {
  my ($self) = @_;
  dbg("config: timing disabled")  if $self->{timer_enabled};
  $self->{timer_enabled} = 0;
}

# discard all timers, start afresh
sub timer_reset {
  my ($self) = @_;
  delete $self->{timers};
  delete $self->{timers_order};
}

sub timer_start {
  my ($self, $name) = @_;

  return unless $self->{timer_enabled};
# dbg("timing: '$name' starting");

  if (!exists $self->{timers}->{$name}) {
    push @{$self->{timers_order}}, $name;
  }
  
  $self->{timers}->{$name}->{start} = Time::HiRes::time();
  # note that this will reset any existing, unstopped timer of that name;
  # that's ok
}

sub timer_end {
  my ($self, $name) = @_;
  return unless $self->{timer_enabled};

  my $t = $self->{timers}->{$name};
  $t->{end} = time;

  if (!$t->{start}) {
    warn "timer_end('$name') with no timer_start";
    return;
  }

  # add to any existing elapsed time for this event, since
  # we may call the same timer name multiple times -- this is ok,
  # as long as they are not nested
  my $dt = $t->{end} - $t->{start};
  $dt = 0  if $dt < 0;  # tolerate clock jumps, just in case
  if (defined $t->{elapsed}) { $t->{elapsed} += $dt }
  else { $t->{elapsed} = $dt }
}

sub time_method {
  my ($self, $name) = @_;
  return unless $self->{timer_enabled};
  return Mail::SpamAssassin::Util::ScopedTimer->new($self, $name);
}

sub timer_report {
  my ($self) = @_;

  my $earliest;
  my $latest;

  while (my($name,$h) = each(%{$self->{timers}})) {
  # dbg("timing: %s - %s", $name, join(", ",
  #     map { sprintf("%s => %s", $_, $h->{$_}) } keys(%$h)));
    my $start = $h->{start};
    if (defined $start && (!defined $earliest || $earliest > $start)) {
      $earliest = $start;
    }
    my $end = $h->{end};
    if (defined $end && (!defined $latest || $latest < $end)) {
      $latest = $end;
    }
    dbg("timing: start but no end: $name") if defined $start && !defined $end;
  }
  my $total =
    (!defined $latest || !defined $earliest) ? 0 : $latest - $earliest;
  my @str;
  foreach my $name (@{$self->{timers_order}}) {
    my $elapsed = $self->{timers}->{$name}->{elapsed} || 0;
    my $pc = $total <= 0 || $elapsed >= $total ? 100 : ($elapsed/$total)*100;
    my $fmt = $elapsed >= 0.005 ? "%.0f" : $elapsed >= 0.002 ? "%.1f" : "%.2f";
    push @str, sprintf("%s: $fmt (%.1f%%)", $name, $elapsed*1000, $pc);
  }

  return sprintf("total %.0f ms - %s", $total*1000, join(", ", @str));
}

###########################################################################
# non-public methods.

sub init {
  my ($self, $use_user_pref) = @_;

  # Allow init() to be called multiple times, but only run once.
  if (defined $self->{_initted}) {
    # If the PID changes, reseed the PRNG (if permitted) and the DNS ID counter
    if ($self->{_initted} != $$) {
      $self->{_initted} = $$;
      srand  if !$self->{skip_prng_reseeding};
      $self->{resolver}->reinit_post_fork();
    }
    return;
  }

  my $timer = $self->time_method("init");
  # Note that this PID has run init()
  $self->{_initted} = $$;

  #fix spamd reading root prefs file
  if (!defined $use_user_pref) {
    $use_user_pref = 1;
  }

  if (!defined $self->{config_text}) {
    $self->{config_text} = '';

    # read a file called "init.pre" in site rules dir *before* all others;
    # even the system config.
    my $siterules = $self->{site_rules_filename};
    $siterules ||= $self->first_existing_path (@site_rules_path);

    my $sysrules = $self->{rules_filename};
    $sysrules ||= $self->first_existing_path (@default_rules_path);

    if ($siterules) {
      $self->{config_text} .= $self->read_pre($siterules, 'site rules pre files');
    }
    else {
      warn "config: could not find site rules directory\n";
    }

    if ($sysrules) {
      $self->{config_text} .= $self->read_pre($sysrules, 'sys rules pre files');
    }
    else {
      warn "config: could not find sys rules directory\n";
    }

    if ($sysrules) {
      my $cftext = $self->read_cf($sysrules, 'default rules dir');
      if ($self->{require_rules} && $cftext !~ /\S/) {
        die "config: no rules were found!  Do you need to run 'sa-update'?\n";
      }
      $self->{config_text} .= $cftext;
    }

    if (!$self->{languages_filename}) {
      $self->{languages_filename} = $self->find_rule_support_file("languages");
    }

    if ($siterules && !$self->{ignore_site_cf_files}) {
      $self->{config_text} .= $self->read_cf($siterules, 'site rules dir');
    }

    if ( $use_user_pref != 0 ) {
      $self->get_and_create_userstate_dir();

      # user prefs file
      my $fname = $self->{userprefs_filename};
      $fname ||= $self->first_existing_path (@default_userprefs_path);

      if (!$self->{dont_copy_prefs}) {
        # bug 4932: if the userprefs path doesn't exist, we need to make it, so
        # just use the last entry in the array as the default path.
        $fname ||= $self->sed_path($default_userprefs_path[-1]);

        my $stat_errn = stat($fname) ? 0 : 0+$!;
        if ($stat_errn == 0 && -f _) {
          # exists and is a regular file, nothing to do
        } elsif ($stat_errn == 0) {
          warn "config: default user preference file $fname is not a regular file\n";
        } elsif ($stat_errn != ENOENT) {
          warn "config: default user preference file $fname not accessible: $!\n";
        } elsif (!$self->create_default_prefs($fname)) {
          warn "config: failed to create default user preference file $fname\n";
        }
      }

      $self->{config_text} .= $self->read_cf($fname, 'user prefs file');
    }
  }

  if ($self->{pre_config_text}) {
    $self->{config_text} = $self->{pre_config_text} . $self->{config_text};
  }
  if ($self->{post_config_text}) {
    $self->{config_text} .= $self->{post_config_text};
  }

  if ($self->{config_text} !~ /\S/) {
    my $m = "config: no configuration text or files found! do you need to run 'sa-update'?\n";
    if ($self->{require_rules}) {
      die $m;
    } else {
      warn $m;
    }
  }

  # Go and parse the config!
  $self->{conf}->{main} = $self;
  if (would_log('dbg', 'config_text') > 1) {
    dbg('config_text: '.$self->{config_text});
  }
  $self->{conf}->parse_rules ($self->{config_text});
  $self->{conf}->finish_parsing(0);
  delete $self->{conf}->{main};	# to allow future GC'ing

  undef $self->{config_text};   # ensure it's actually freed
  delete $self->{config_text};

  if ($self->{require_rules} && !$self->{conf}->found_any_rules()) {
    die "config: no rules were found!  Do you need to run 'sa-update'?\n";
  }

  # Initialize the Bayes subsystem
  if ($self->{conf}->{use_bayes}) {
      require Mail::SpamAssassin::Bayes;
      $self->{bayes_scanner} = new Mail::SpamAssassin::Bayes ($self);
  }
  $self->{'learn_to_journal'} = $self->{conf}->{bayes_learn_to_journal};

  # Figure out/set our initial scoreset
  my $set = 0;
  $set |= 1 unless $self->{local_tests_only};
  $set |= 2 if $self->{bayes_scanner} && $self->{bayes_scanner}->is_scan_available() && $self->{conf}->{use_bayes_rules};
  $self->{conf}->set_score_set ($set);

  if ($self->{only_these_rules}) {
    $self->{conf}->trim_rules($self->{only_these_rules});
  }

  if (!$self->{timer_enabled}) {
    # enable timing implicitly if _TIMING_ is used in add_header templates
    foreach my $hf_ref (@{$self->{conf}->{'headers_ham'}},
                        @{$self->{conf}->{'headers_spam'}}) {
      if ($hf_ref->[1] =~ /_TIMING_/) { $self->timer_enable(); last }
    }
  }

  # should be called only after configuration has been parsed
  $self->{resolver} = Mail::SpamAssassin::DnsResolver->new($self);

  # TODO -- open DNS cache etc. if necessary
}

sub read_cf {
  my ($self, $allpaths, $desc) = @_;
  return $self->_read_cf_pre($allpaths,$desc,\&get_cf_files_in_dir);
}

sub read_pre {
  my ($self, $allpaths, $desc) = @_;
  return $self->_read_cf_pre($allpaths,$desc,\&get_pre_files_in_dir);
}

sub _read_cf_pre {
  my ($self, $allpaths, $desc, $filelistmethod) = @_;

  return '' unless defined ($allpaths);

  my $txt = '';
  foreach my $path (split("\000", $allpaths)) 
  {
    dbg("config: using \"$path\" for $desc");

    my $stat_errn = stat($path) ? 0 : 0+$!;
    if ($stat_errn == ENOENT) {
      # no file or directory
    } elsif ($stat_errn != 0) {
      dbg("config: file or directory $path not accessible: $!");
    } elsif (-d _) {
      foreach my $file ($self->$filelistmethod($path)) {
        $txt .= read_cf_file($file);
      }
    } elsif (-f _ && -s _ && -r _) {
      $txt .= read_cf_file($path);
    }
  }

  return $txt;
}


sub read_cf_file {
  my($path) = @_;
  my $txt = '';

  local *IN;
  if (open (IN, "<".$path)) {

    my($inbuf,$nread); $txt = '';
    while ( $nread=read(IN,$inbuf,16384) ) { $txt .= $inbuf }
    defined $nread  or die "error reading $path: $!";
    close IN  or die "error closing $path: $!";
    undef $inbuf;

    $txt = "file start $path\n" . $txt;
    # add an extra \n in case file did not end in one.
    $txt .= "\nfile end $path\n";

    dbg("config: read file $path");
  }
  else {
    warn "config: cannot open \"$path\": $!\n";
  }

  return $txt;
}

sub get_and_create_userstate_dir {
  my ($self, $dir) = @_;

  my $fname;

  # If vpopmail is enabled then set fname to virtual homedir
  # precedence: dir, userstate_dir, derive from user_dir, system default
  if (defined $dir) {
    $fname = File::Spec->catdir ($dir, ".spamassassin");
  }
  elsif (defined $self->{userstate_dir}) {
    $fname = $self->{userstate_dir};
  }
  elsif (defined $self->{user_dir}) {
    $fname = File::Spec->catdir ($self->{user_dir}, ".spamassassin");
  }

  $fname ||= $self->first_existing_path (@default_userstate_dir);

  # bug 4932: use the last default_userstate_dir entry if none of the others
  # already exist
  $fname ||= $self->sed_path($default_userstate_dir[-1]);

  if (!$self->{dont_copy_prefs}) {
    dbg("config: using \"$fname\" for user state dir");
  }

  # if this is not a dir, not readable, or we are unable to create the dir,
  # this is not (yet) a serious error; in fact, it's not even worth
  # a warning at all times, so use dbg().  see bug 6268
  my $stat_errn = stat($fname) ? 0 : 0+$!;
  if ($stat_errn == 0 && !-d _) {
    dbg("config: $fname exists but is not a directory");
  } elsif ($stat_errn != 0 && $stat_errn != ENOENT) {
    dbg("config: error accessing $fname: $!");
  } else {  # does not exist, create it
    eval {
      mkpath($fname, 0, 0700);  1;
    } or do {
      my $eval_stat = $@ ne '' ? $@ : "errno=$!";  chomp $eval_stat;
      dbg("config: mkdir $fname failed: $eval_stat");
    };
  }

  $fname;
}

=item $fullpath = $f->find_rule_support_file ($filename)

Find a rule-support file, such as C<languages> or C<triplets.txt>,
in the system-wide rules directory, and return its full path if
it exists, or undef if it doesn't exist.

(This API was added in SpamAssassin 3.1.1.)

=cut

sub find_rule_support_file {
  my ($self, $filename) = @_;

  return $self->first_existing_path(
    map { my $p = $_; $p =~ s{$}{/$filename}; $p } @default_rules_path );
}

=item $f->create_default_prefs ($filename, $username [ , $userdir ] )

Copy default preferences file into home directory for later use and
modification, if it does not already exist and C<dont_copy_prefs> is
not set.

=cut

sub create_default_prefs {
  # $userdir will only exist if vpopmail config is enabled thru spamd
  # Its value will be the virtual user's maildir
  #
  my ($self, $fname, $user, $userdir) = @_;

  if ($self->{dont_copy_prefs}) {
    return(0);
  }

#  if ($userdir && $userdir ne $self->{user_dir}) {
#    warn "config: hooray! user_dirs don't match! '$userdir' vs '$self->{user_dir}'\n";
#  }

  my $stat_errn = stat($fname) ? 0 : 0+$!;
  if ($stat_errn == 0) {
    # fine, it already exists
  } elsif ($stat_errn != ENOENT) {
    dbg("config: cannot access user preferences file $fname: $!");
  } else {
    # Pass on the value of $userdir for virtual users in vpopmail
    # otherwise it is empty and the user's normal homedir is used
    $self->get_and_create_userstate_dir($userdir);

    # copy in the default one for later editing
    my $defprefs =
      $self->first_existing_path(@Mail::SpamAssassin::default_prefs_path);

    local(*IN,*OUT);
    $fname = Mail::SpamAssassin::Util::untaint_file_path($fname);
    if (!defined $defprefs) {
      warn "config: can not determine default prefs path\n";
    } elsif (!open(IN, "<$defprefs")) {
      warn "config: cannot open $defprefs: $!\n";
    } elsif (!open(OUT, ">$fname")) {
      warn "config: cannot create user preferences file $fname: $!\n";
    } else {
      # former code skipped lines beginning with '#* ', the following copy
      # procedure no longer does so, as it avoids reading line-by-line
      my($inbuf,$nread);
      while ( $nread=read(IN,$inbuf,16384) ) {
        print OUT $inbuf  or die "cannot write to $fname: $!";
      }
      defined $nread  or die "error reading $defprefs: $!";
      undef $inbuf;
      close OUT or die "error closing $fname: $!";
      close IN  or die "error closing $defprefs: $!";

      if (($< == 0) && ($> == 0) && defined($user)) { # chown it
        my ($uid,$gid) = (getpwnam(untaint_var($user)))[2,3];
        unless (chown($uid, $gid, $fname)) {
          warn "config: couldn't chown $fname to $uid:$gid for $user: $!\n";
        }
      }
      warn "config: created user preferences file: $fname\n";
      return(1);
    }
  }

  return(0);
}

###########################################################################

sub expand_name {
  my ($self, $name) = @_;
  my $home = $self->{user_dir} || $ENV{HOME} || '';

  if (am_running_on_windows()) {
    my $userprofile = $ENV{USERPROFILE} || '';

    return $userprofile if ($userprofile && $userprofile =~ m/^[a-z]\:[\/\\]/oi);
    return $userprofile if ($userprofile =~ m/^\\\\/o);

    return $home if ($home && $home =~ m/^[a-z]\:[\/\\]/oi);
    return $home if ($home =~ m/^\\\\/o);

    return '';
  } else {
    return $home if ($home && $home =~ /\//o);
    return (getpwnam($name))[7] if ($name ne '');
    return (getpwuid($>))[7];
  }
}

sub sed_path {
  my ($self, $path) = @_;
  return if !defined $path;

  if (exists($self->{conf}->{sed_path_cache}->{$path})) {
    return $self->{conf}->{sed_path_cache}->{$path};
  }

  my $orig_path = $path;

  $path =~ s/__local_rules_dir__/$self->{LOCAL_RULES_DIR} || ''/ges;
  $path =~ s/__local_state_dir__/$self->{LOCAL_STATE_DIR} || ''/ges;
  $path =~ s/__def_rules_dir__/$self->{DEF_RULES_DIR} || ''/ges;
  $path =~ s{__prefix__}{$self->{PREFIX} || $Config{prefix} || '/usr'}ges;
  $path =~ s{__userstate__}{$self->get_and_create_userstate_dir() || ''}ges;
  $path =~ s{__perl_major_ver__}{$self->get_perl_major_version()}ges;
  $path =~ s/__version__/${VERSION}/gs;
  $path =~ s/^\~([^\/]*)/$self->expand_name($1)/es;

  $path = Mail::SpamAssassin::Util::untaint_file_path ($path);
  $self->{conf}->{sed_path_cache}->{$orig_path} = $path;
  return $path;
}

sub get_perl_major_version {
  my $self = shift;
  $] =~ /^(\d\.\d\d\d)/ or die "bad perl ver $]";
  return $1;
}

sub first_existing_path {
  my $self = shift;
  my $path;
  foreach my $p (@_) {
    $path = $self->sed_path ($p);
    if (defined $path) {
      my($errn) = stat($path) ? 0 : 0+$!;
      if    ($errn == ENOENT) { }  # does not exist
      elsif ($errn) {  warn "config: path \"$path\" is inaccessible: $!\n" }
      else { return $path }
    }
  }
  return;
}

###########################################################################

sub get_cf_files_in_dir {
  my ($self, $dir) = @_;
  return $self->_get_cf_pre_files_in_dir($dir, 'cf');
}

sub get_pre_files_in_dir {
  my ($self, $dir) = @_;
  return $self->_get_cf_pre_files_in_dir($dir, 'pre');
}

sub _get_cf_pre_files_in_dir {
  my ($self, $dir, $type) = @_;

  if ($self->{config_tree_recurse}) {
    my @cfs;

    # use "eval" to avoid loading File::Find unless this is specified
    eval ' use File::Find qw();
      File::Find::find(
        { untaint => 1,
          follow => 1,
          wanted =>
            sub { push(@cfs, $File::Find::name) if /\.\Q$type\E$/i && -f $_ }
        }, $dir); 1;
    ' or do {
      my $eval_stat = $@ ne '' ? $@ : "errno=$!";  chomp $eval_stat;
      die "_get_cf_pre_files_in_dir error: $eval_stat";
    };
    @cfs = sort { $a cmp $b } @cfs;
    return @cfs;
  }
  else {
    opendir(SA_CF_DIR, $dir) or warn "config: cannot opendir $dir: $!\n";
    my @cfs = grep { $_ ne '.' && $_ ne '..' &&
                     /\.${type}$/i && -f "$dir/$_" } readdir(SA_CF_DIR);
    closedir SA_CF_DIR;

    return map { "$dir/$_" } sort { $a cmp $b } @cfs;
  }
}

###########################################################################

sub have_plugin {
  my ($self, $subname) = @_;

  # We could potentially get called after a finish(), so just return.
  return unless $self->{plugins};

  return $self->{plugins}->have_callback ($subname);
}

sub call_plugins {
  my $self = shift;

  # We could potentially get called after a finish(), so just return.
  return unless $self->{plugins};

  # safety net in case some plugin changes global settings, Bug 6218
  local $/ = $/;  # prevent underlying modules from changing the global $/

  my $subname = shift;
  return $self->{plugins}->callback($subname, @_);
}

###########################################################################

sub find_all_addrs_in_mail {
  my ($self, $mail_obj) = @_;

  $self->init(1);

  my @addrlist;
  foreach my $header (qw(To From Cc Reply-To Sender
  				Errors-To Mail-Followup-To))
  {
    my @hdrs = $mail_obj->get_header($header);
    if ($#hdrs < 0) { next; }
    push (@addrlist, $self->find_all_addrs_in_line(join (" ", @hdrs)));
  }

  # find addrs in body, too
  foreach my $line (@{$mail_obj->get_body()}) {
    push (@addrlist, $self->find_all_addrs_in_line($line));
  }

  my @ret;
  my %done;

  foreach (@addrlist) {
    s/^mailto://;       # from Outlook "forwarded" message
    next if defined ($done{$_}); $done{$_} = 1;
    push (@ret, $_);
  }

  @ret;
}

sub find_all_addrs_in_line {
  my ($self, $line) = @_;

  # a more permissive pattern based on "dot-atom" as per RFC2822
  my $ID_PATTERN   = '[-a-z0-9_\+\:\=\!\#\$\%\&\*\^\?\{\}\|\~\/\.]+';
  my $HOST_PATTERN = '[-a-z0-9_\+\:\/]+';

  my @addrs;
  my %seen;
  while ($line =~ s/(?:mailto:)?\s*
	      ($ID_PATTERN \@
	      $HOST_PATTERN(?:\.$HOST_PATTERN)+)//oix) 
  {
    my $addr = $1;
    $addr =~ s/^mailto://;
    next if (defined ($seen{$addr})); $seen{$addr} = 1;
    push (@addrs, $addr);
  }

  return @addrs;
}

###########################################################################

# sa_die -- used to die with a useful exit code.

sub sa_die {
  my $exitcode = shift;
  warn @_;
  exit $exitcode;
}

###########################################################################

=item $f->copy_config ( [ $source ], [ $dest ] )

Used for daemons to keep a persistent Mail::SpamAssassin object's
configuration correct if switching between users.  Pass an associative
array reference as either $source or $dest, and set the other to 'undef'
so that the object will use its current configuration.  i.e.:

  # create object w/ configuration
  my $spamtest = Mail::SpamAssassin->new( ... );

  # backup configuration to %conf_backup
  my %conf_backup;
  $spamtest->copy_config(undef, \%conf_backup) ||
    die "config: error returned from copy_config!\n";

  ... do stuff, perhaps modify the config, etc ...

  # reset the configuration back to the original
  $spamtest->copy_config(\%conf_backup, undef) ||
    die "config: error returned from copy_config!\n";

Note that the contents of the associative arrays should be considered
opaque by calling code.

=cut

sub copy_config {
  my ($self, $source, $dest) = @_;

  # At least one of either source or dest needs to be a hash reference ...
  unless ((defined $source && ref($source) eq 'HASH') ||
          (defined $dest && ref($dest) eq 'HASH'))
  {
    return 0;
  }

  my $timer = $self->time_method("copy_config");

  # let the Conf object itself do all the heavy lifting.  It's better
  # than having this class know all about that class' internals...
  if (defined $source) {
    dbg ("config: copying current conf from backup");
  }
  else {
    dbg ("config: copying current conf to backup");
  }
  return $self->{conf}->clone($source, $dest);
}

###########################################################################

=item @plugins = $f->get_loaded_plugins_list ( )

Return the list of plugins currently loaded by this SpamAssassin object's
configuration; each entry in the list is an object of type
C<Mail::SpamAssassin::Plugin>.

(This API was added in SpamAssassin 3.2.0.)

=cut

sub get_loaded_plugins_list {
  my ($self) = @_;
  return $self->{plugins}->get_loaded_plugins_list();
}

1;
__END__

###########################################################################

=back

=head1 PREREQUISITES

C<HTML::Parser>
C<Sys::Syslog>

=head1 MORE DOCUMENTATION

See also E<lt>http://spamassassin.apache.org/E<gt> and
E<lt>http://wiki.apache.org/spamassassin/E<gt> for more information.

=head1 SEE ALSO

Mail::SpamAssassin::Conf(3)
Mail::SpamAssassin::PerMsgStatus(3)
spamassassin(1)
sa-update(1)

=head1 BUGS

See E<lt>http://issues.apache.org/SpamAssassin/E<gt>

=head1 AUTHORS

The SpamAssassin(tm) Project E<lt>http://spamassassin.apache.org/E<gt>

=head1 COPYRIGHT

SpamAssassin is distributed under the Apache License, Version 2.0, as
described in the file C<LICENSE> included with the distribution.

=head1 AVAILABILITY

The latest version of this library is likely to be available from CPAN
as well as:

  E<lt>http://spamassassin.apache.org/E<gt>

=cut