summaryrefslogtreecommitdiffstats
path: root/Bugzilla/Util.pm
blob: bf072e88d51b818c166d371c6c614a09a6fd8396 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
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
# This Source Code Form is subject to the terms of the Mozilla Public
# License, v. 2.0. If a copy of the MPL was not distributed with this
# file, You can obtain one at http://mozilla.org/MPL/2.0/.
#
# This Source Code Form is "Incompatible With Secondary Licenses", as
# defined by the Mozilla Public License, v. 2.0.

package Bugzilla::Util;

use 5.10.1;
use strict;

use parent qw(Exporter);
@Bugzilla::Util::EXPORT = qw(trick_taint detaint_natural detaint_signed
                             html_quote url_quote xml_quote
                             css_class_quote html_light_quote
                             i_am_cgi correct_urlbase remote_ip validate_ip
                             do_ssl_redirect_if_required use_attachbase
                             diff_arrays on_main_db
                             trim wrap_hard wrap_comment find_wrap_point
                             format_time validate_date validate_time datetime_from
                             is_7bit_clean bz_crypt generate_random_password
                             validate_email_syntax check_email_syntax clean_text
                             get_text template_var disable_utf8
                             detect_encoding email_filter);

use Bugzilla::Constants;
use Bugzilla::RNG qw(irand);
use Bugzilla::Error;

use Date::Parse;
use Date::Format;
use Digest;
use Email::Address;
use List::Util qw(first);
use Scalar::Util qw(tainted blessed);
use Text::Wrap;
use Encode qw(encode decode resolve_alias);
use Encode::Guess;

sub trick_taint {
    require Carp;
    Carp::confess("Undef to trick_taint") unless defined $_[0];
    my $match = $_[0] =~ /^(.*)$/s;
    $_[0] = $match ? $1 : undef;
    return (defined($_[0]));
}

sub detaint_natural {
    my $match = $_[0] =~ /^(\d+)$/;
    $_[0] = $match ? int($1) : undef;
    return (defined($_[0]));
}

sub detaint_signed {
    my $match = $_[0] =~ /^([-+]?\d+)$/;
    # The "int()" call removes any leading plus sign.
    $_[0] = $match ? int($1) : undef;
    return (defined($_[0]));
}

# Bug 120030: Override html filter to obscure the '@' in user
#             visible strings.
# Bug 319331: Handle BiDi disruptions.
sub html_quote {
    my $var = shift;
    $var =~ s/&/&/g;
    $var =~ s/</&lt;/g;
    $var =~ s/>/&gt;/g;
    $var =~ s/"/&quot;/g;
    # Obscure '@'.
    $var =~ s/\@/\&#64;/g;
    if (Bugzilla->params->{'utf8'}) {
        # Remove the following characters because they're
        # influencing BiDi:
        # --------------------------------------------------------
        # |Code  |Name                      |UTF-8 representation|
        # |------|--------------------------|--------------------|
        # |U+202a|Left-To-Right Embedding   |0xe2 0x80 0xaa      |
        # |U+202b|Right-To-Left Embedding   |0xe2 0x80 0xab      |
        # |U+202c|Pop Directional Formatting|0xe2 0x80 0xac      |
        # |U+202d|Left-To-Right Override    |0xe2 0x80 0xad      |
        # |U+202e|Right-To-Left Override    |0xe2 0x80 0xae      |
        # --------------------------------------------------------
        #
        # The following are characters influencing BiDi, too, but
        # they can be spared from filtering because they don't
        # influence more than one character right or left:
        # --------------------------------------------------------
        # |Code  |Name                      |UTF-8 representation|
        # |------|--------------------------|--------------------|
        # |U+200e|Left-To-Right Mark        |0xe2 0x80 0x8e      |
        # |U+200f|Right-To-Left Mark        |0xe2 0x80 0x8f      |
        # --------------------------------------------------------
        $var =~ s/[\x{202a}-\x{202e}]//g;
    }
    return $var;
}

sub html_light_quote {
    my ($text) = @_;
    # admin/table.html.tmpl calls |FILTER html_light| many times.
    # There is no need to recreate the HTML::Scrubber object again and again.
    my $scrubber = Bugzilla->process_cache->{html_scrubber};

    # List of allowed HTML elements having no attributes.
    my @allow = qw(b strong em i u p br abbr acronym ins del cite code var
                   dfn samp kbd big small sub sup tt dd dt dl ul li ol
                   fieldset legend);

    if (!Bugzilla->feature('html_desc')) {
        my $safe = join('|', @allow);
        my $chr = chr(1);

        # First, escape safe elements.
        $text =~ s#<($safe)>#$chr$1$chr#go;
        $text =~ s#</($safe)>#$chr/$1$chr#go;
        # Now filter < and >.
        $text =~ s#<#&lt;#g;
        $text =~ s#>#&gt;#g;
        # Restore safe elements.
        $text =~ s#$chr/($safe)$chr#</$1>#go;
        $text =~ s#$chr($safe)$chr#<$1>#go;
        return $text;
    }
    elsif (!$scrubber) {
        # We can be less restrictive. We can accept elements with attributes.
        push(@allow, qw(a blockquote q span));

        # Allowed protocols.
        my $safe_protocols = join('|', SAFE_PROTOCOLS);
        my $protocol_regexp = qr{(^(?:$safe_protocols):|^[^:]+$)}i;

        # Deny all elements and attributes unless explicitly authorized.
        my @default = (0 => {
                             id    => 1,
                             name  => 1,
                             class => 1,
                             '*'   => 0, # Reject all other attributes.
                            }
                       );

        # Specific rules for allowed elements. If no specific rule is set
        # for a given element, then the default is used.
        my @rules = (a => {
                           href   => $protocol_regexp,
                           target => qr{^(?:_blank|_parent|_self|_top)$}i,
                           title  => 1,
                           id     => 1,
                           name   => 1,
                           class  => 1,
                           '*'    => 0, # Reject all other attributes.
                          },
                     blockquote => {
                                    cite => $protocol_regexp,
                                    id    => 1,
                                    name  => 1,
                                    class => 1,
                                    '*'  => 0, # Reject all other attributes.
                                   },
                     'q' => {
                             cite => $protocol_regexp,
                             id    => 1,
                             name  => 1,
                             class => 1,
                             '*'  => 0, # Reject all other attributes.
                          },
                    );

        Bugzilla->process_cache->{html_scrubber} = $scrubber =
          HTML::Scrubber->new(default => \@default,
                              allow   => \@allow,
                              rules   => \@rules,
                              comment => 0,
                              process => 0);
    }
    return $scrubber->scrub($text);
}

sub email_filter {
    my ($toencode) = @_;
    if (!Bugzilla->user->id) {
        my @emails = Email::Address->parse($toencode);
        if (scalar @emails) {
            my @hosts = map { quotemeta($_->host) } @emails;
            my $hosts_re = join('|', @hosts);
            $toencode =~ s/\@(?:$hosts_re)//g;
            return $toencode;
        }
    }
    return $toencode;
}

# This originally came from CGI.pm, by Lincoln D. Stein
sub url_quote {
    my ($toencode) = (@_);
    utf8::encode($toencode) # The below regex works only on bytes
        if Bugzilla->params->{'utf8'} && utf8::is_utf8($toencode);
    $toencode =~ s/([^a-zA-Z0-9_\-.])/uc sprintf("%%%02x",ord($1))/eg;
    return $toencode;
}

sub css_class_quote {
    my ($toencode) = (@_);
    $toencode =~ s#[ /]#_#g;
    $toencode =~ s/([^a-zA-Z0-9_\-.])/uc sprintf("&#x%x;",ord($1))/eg;
    return $toencode;
}

sub xml_quote {
    my ($var) = (@_);
    $var =~ s/\&/\&amp;/g;
    $var =~ s/</\&lt;/g;
    $var =~ s/>/\&gt;/g;
    $var =~ s/\"/\&quot;/g;
    $var =~ s/\'/\&apos;/g;
    
    # the following nukes characters disallowed by the XML 1.0
    # spec, Production 2.2. 1.0 declares that only the following 
    # are valid:
    # (#x9 | #xA | #xD | [#x20-#xD7FF] | [#xE000-#xFFFD] | [#x10000-#x10FFFF])
    $var =~ s/([\x{0001}-\x{0008}]|
               [\x{000B}-\x{000C}]|
               [\x{000E}-\x{001F}]|
               [\x{D800}-\x{DFFF}]|
               [\x{FFFE}-\x{FFFF}])//gx;
    return $var;
}

sub i_am_cgi {
    # I use SERVER_SOFTWARE because it's required to be
    # defined for all requests in the CGI spec.
    return exists $ENV{'SERVER_SOFTWARE'} ? 1 : 0;
}

# This exists as a separate function from Bugzilla::CGI::redirect_to_https
# because we don't want to create a CGI object during XML-RPC calls
# (doing so can mess up XML-RPC).
sub do_ssl_redirect_if_required {
    return if !i_am_cgi();
    return if !Bugzilla->params->{'ssl_redirect'};

    my $sslbase = Bugzilla->params->{'sslbase'};
    
    # If we're already running under SSL, never redirect.
    return if uc($ENV{HTTPS} || '') eq 'ON';
    # Never redirect if there isn't an sslbase.
    return if !$sslbase;
    Bugzilla->cgi->redirect_to_https();
}

sub correct_urlbase {
    my $ssl = Bugzilla->params->{'ssl_redirect'};
    my $urlbase = Bugzilla->params->{'urlbase'};
    my $sslbase = Bugzilla->params->{'sslbase'};

    if (!$sslbase) {
        return $urlbase;
    }
    elsif ($ssl) {
        return $sslbase;
    }
    else {
        # Return what the user currently uses.
        return (uc($ENV{HTTPS} || '') eq 'ON') ? $sslbase : $urlbase;
    }
}

sub remote_ip {
    my $ip = $ENV{'REMOTE_ADDR'} || '127.0.0.1';
    my @proxies = split(/[\s,]+/, Bugzilla->params->{'inbound_proxies'});

    # If the IP address is one of our trusted proxies, then we look at
    # the X-Forwarded-For header to determine the real remote IP address.
    if ($ENV{'HTTP_X_FORWARDED_FOR'} && first { $_ eq $ip } @proxies) {
        my @ips = split(/[\s,]+/, $ENV{'HTTP_X_FORWARDED_FOR'});
        # This header can contain several IP addresses. We want the
        # IP address of the machine which connected to our proxies as
        # all other IP addresses may be fake or internal ones.
        # Note that this may block a whole external proxy, but we have
        # no way to determine if this proxy is malicious or trustable.
        foreach my $remote_ip (reverse @ips) {
            if (!first { $_ eq $remote_ip } @proxies) {
                # Keep the original IP address if the remote IP is invalid.
                $ip = validate_ip($remote_ip) || $ip;
                last;
            }
        }
    }
    return $ip;
}

sub validate_ip {
    my $ip = shift;
    return is_ipv4($ip) || is_ipv6($ip);
}

# Copied from Data::Validate::IP::is_ipv4().
sub is_ipv4 {
    my $ip = shift;
    return unless defined $ip;

    my @octets = $ip =~ /^(\d{1,3})\.(\d{1,3})\.(\d{1,3})\.(\d{1,3})$/;
    return unless scalar(@octets) == 4;

    foreach my $octet (@octets) {
        return unless ($octet >= 0 && $octet <= 255 && $octet !~ /^0\d{1,2}$/);
    }

    # The IP address is valid and can now be detainted.
    return join('.', @octets);
}

# Copied from Data::Validate::IP::is_ipv6().
sub is_ipv6 {
    my $ip = shift;
    return unless defined $ip;

    # If there is a :: then there must be only one :: and the length
    # can be variable. Without it, the length must be 8 groups.
    my @chunks = split(':', $ip);

    # Need to check if the last chunk is an IPv4 address, if it is we
    # pop it off and exempt it from the normal IPv6 checking and stick
    # it back on at the end. If there is only one chunk and it's an IPv4
    # address, then it isn't an IPv6 address.
    my $ipv4;
    my $expected_chunks = 8;
    if (@chunks > 1 && is_ipv4($chunks[$#chunks])) {
        $ipv4 = pop(@chunks);
        $expected_chunks--;
    }

    my $empty = 0;
    # Workaround to handle trailing :: being valid.
    if ($ip =~ /[0-9a-f]{1,4}::$/) {
        $empty++;
    # Single trailing ':' is invalid.
    } elsif ($ip =~ /:$/) {
        return;
    }

    foreach my $chunk (@chunks) {
        return unless $chunk =~ /^[0-9a-f]{0,4}$/i;
        $empty++ if $chunk eq '';
    }
    # More than one :: block is bad, but if it starts with :: it will
    # look like two, so we need an exception.
    if ($empty == 2 && $ip =~ /^::/) {
        # This is ok
    } elsif ($empty > 1) {
        return;
    }

    push(@chunks, $ipv4) if $ipv4;
    # Need 8 chunks, or we need an empty section that could be filled
    # to represent the missing '0' sections.
    return unless (@chunks == $expected_chunks || @chunks < $expected_chunks && $empty);

    my $ipv6 = join(':', @chunks);
    # The IP address is valid and can now be detainted.
    trick_taint($ipv6);

    # Need to handle the exception of trailing :: being valid.
    return "${ipv6}::" if $ip =~ /::$/;
    return $ipv6;
}

sub use_attachbase {
    my $attachbase = Bugzilla->params->{'attachment_base'};
    return ($attachbase ne ''
            && $attachbase ne Bugzilla->params->{'urlbase'}
            && $attachbase ne Bugzilla->params->{'sslbase'}) ? 1 : 0;
}

sub diff_arrays {
    my ($old_ref, $new_ref, $attrib) = @_;
    $attrib ||= 'name';

    my (%counts, %pos);
    # We are going to alter the old array.
    my @old = @$old_ref;
    my $i = 0;

    # $counts{foo}-- means old, $counts{foo}++ means new.
    # If $counts{foo} becomes positive, then we are adding new items,
    # else we simply cancel one old existing item. Remaining items
    # in the old list have been removed.
    foreach (@old) {
        next unless defined $_;
        my $value = blessed($_) ? $_->$attrib : $_;
        $counts{$value}--;
        push @{$pos{$value}}, $i++;
    }
    my @added;
    foreach (@$new_ref) {
        next unless defined $_;
        my $value = blessed($_) ? $_->$attrib : $_;
        if (++$counts{$value} > 0) {
            # Ignore empty strings, but objects having an empty string
            # as attribute are fine.
            push(@added, $_) unless ($value eq '' && !blessed($_));
        }
        else {
            my $old_pos = shift @{$pos{$value}};
            $old[$old_pos] = undef;
        }
    }
    # Ignore canceled items as well as empty strings.
    my @removed = grep { defined $_ && $_ ne '' } @old;
    return (\@removed, \@added);
}

sub trim {
    my ($str) = @_;
    if ($str) {
      $str =~ s/^\s+//g;
      $str =~ s/\s+$//g;
    }
    return $str;
}

sub wrap_comment {
    my ($comment, $cols) = @_;
    my $wrappedcomment = "";

    # Use 'local', as recommended by Text::Wrap's perldoc.
    local $Text::Wrap::columns = $cols || COMMENT_COLS;
    # Make words that are longer than COMMENT_COLS not wrap.
    local $Text::Wrap::huge    = 'overflow';
    # Don't mess with tabs.
    local $Text::Wrap::unexpand = 0;

    # If the line starts with ">", don't wrap it. Otherwise, wrap.
    foreach my $line (split(/\r\n|\r|\n/, $comment)) {
      if ($line =~ qr/^>/) {
        $wrappedcomment .= ($line . "\n");
      }
      else {
        $wrappedcomment .= (wrap('', '', $line) . "\n");
      }
    }

    chomp($wrappedcomment); # Text::Wrap adds an extra newline at the end.
    return $wrappedcomment;
}

sub find_wrap_point {
    my ($string, $maxpos) = @_;
    if (!$string) { return 0 }
    if (length($string) < $maxpos) { return length($string) }
    my $wrappoint = rindex($string, ",", $maxpos); # look for comma
    if ($wrappoint <= 0) {  # can't find comma
        $wrappoint = rindex($string, " ", $maxpos); # look for space
        if ($wrappoint <= 0) {  # can't find space
            $wrappoint = rindex($string, "-", $maxpos); # look for hyphen
            if ($wrappoint <= 0) {  # can't find hyphen
                $wrappoint = $maxpos;  # just truncate it
            } else {
                $wrappoint++; # leave hyphen on the left side
            }
        }
    }
    return $wrappoint;
}

sub wrap_hard {
    my ($string, $columns) = @_;
    local $Text::Wrap::columns = $columns;
    local $Text::Wrap::unexpand = 0;
    local $Text::Wrap::huge = 'wrap';
    
    my $wrapped = wrap('', '', $string);
    chomp($wrapped);
    return $wrapped;
}

sub format_time {
    my ($date, $format, $timezone) = @_;

    # If $format is not set, try to guess the correct date format.
    if (!$format) {
        if (!ref $date
            && $date =~ /^(\d{4})[-\.](\d{2})[-\.](\d{2}) (\d{2}):(\d{2})(:(\d{2}))?$/) 
        {
            my $sec = $7;
            if (defined $sec) {
                $format = "%Y-%m-%d %T %Z";
            } else {
                $format = "%Y-%m-%d %R %Z";
            }
        } else {
            # Default date format. See DateTime for other formats available.
            $format = "%Y-%m-%d %R %Z";
        }
    }

    my $dt = ref $date ? $date : datetime_from($date, $timezone);
    $date = defined $dt ? $dt->strftime($format) : '';
    return trim($date);
}

sub datetime_from {
    my ($date, $timezone) = @_;

    # In the database, this is the "0" date.
    return undef if $date =~ /^0000/;

    # strptime($date) returns an empty array if $date has an invalid
    # date format.
    my @time = strptime($date);

    unless (scalar @time) {
        # If an unknown timezone is passed (such as MSK, for Moskow),
        # strptime() is unable to parse the date. We try again, but we first
        # remove the timezone.
        $date =~ s/\s+\S+$//;
        @time = strptime($date);
    }

    return undef if !@time;

    # strptime() counts years from 1900, and months from 0 (January).
    # We have to fix both values.
    my %args = (
        year   => $time[5] + 1900,
        month  => $time[4] + 1,
        day    => $time[3],
        hour   => $time[2],
        minute => $time[1],
        # DateTime doesn't like fractional seconds.
        # Also, sometimes seconds are undef.
        second => defined($time[0]) ? int($time[0]) : undef,
        # If a timezone was specified, use it. Otherwise, use the
        # local timezone.
        time_zone => Bugzilla->local_timezone->offset_as_string($time[6]) 
                     || Bugzilla->local_timezone,
    );

    # If something wasn't specified in the date, it's best to just not
    # pass it to DateTime at all. (This is important for doing datetime_from
    # on the deadline field, which is usually just a date with no time.)
    foreach my $arg (keys %args) {
        delete $args{$arg} if !defined $args{$arg};
    }

    # This module takes time to load and is only used here, so we
    # |require| it here rather than |use| it.
    require DateTime;
    my $dt = new DateTime(\%args);

    # Now display the date using the given timezone,
    # or the user's timezone if none is given.
    $dt->set_time_zone($timezone || Bugzilla->user->timezone);
    return $dt;
}

sub bz_crypt {
    my ($password, $salt) = @_;

    my $algorithm;
    if (!defined $salt) {
        # If you don't use a salt, then people can create tables of
        # hashes that map to particular passwords, and then break your
        # hashing very easily if they have a large-enough table of common
        # (or even uncommon) passwords. So we generate a unique salt for
        # each password in the database, and then just prepend it to
        # the hash.
        $salt = generate_random_password(PASSWORD_SALT_LENGTH);
        $algorithm = PASSWORD_DIGEST_ALGORITHM;
    }

    # We append the algorithm used to the string. This is good because then
    # we can change the algorithm being used, in the future, without 
    # disrupting the validation of existing passwords. Also, this tells
    # us if a password is using the old "crypt" method of hashing passwords,
    # because the algorithm will be missing from the string.
    if ($salt =~ /{([^}]+)}$/) {
        $algorithm = $1;
    }

    my $crypted_password;
    if (!$algorithm) {
        # Wide characters cause crypt to die
        if (Bugzilla->params->{'utf8'}) {
            utf8::encode($password) if utf8::is_utf8($password);
        }

        # Crypt the password.
        $crypted_password = crypt($password, $salt);
    }
    else {
        my $hasher = Digest->new($algorithm);
        # We only want to use the first characters of the salt, no
        # matter how long of a salt we may have been passed.
        $salt = substr($salt, 0, PASSWORD_SALT_LENGTH);
        $hasher->add($password, $salt);
        $crypted_password = $salt . $hasher->b64digest . "{$algorithm}";
    }

    # Return the crypted password.
    return $crypted_password;
}

# If you want to understand the security of strings generated by this
# function, here's a quick formula that will help you estimate:
# We pick from 62 characters, which is close to 64, which is 2^6.
# So 8 characters is (2^6)^8 == 2^48 combinations. Just multiply 6
# by the number of characters you generate, and that gets you the equivalent
# strength of the string in bits.
sub generate_random_password {
    my $size = shift || 10; # default to 10 chars if nothing specified
    return join("", map{ ('0'..'9','a'..'z','A'..'Z')[irand 62] } (1..$size));
}

sub validate_email_syntax {
    my ($addr) = @_;
    my $match = Bugzilla->params->{'emailregexp'};
    my $email = $addr . Bugzilla->params->{'emailsuffix'};
    # This regexp follows RFC 2822 section 3.4.1.
    my $addr_spec = $Email::Address::addr_spec;
    # RFC 2822 section 2.1 specifies that email addresses must
    # be made of US-ASCII characters only.
    # Email::Address::addr_spec doesn't enforce this.
    my $ret = ($addr =~ /$match/ && $email !~ /\P{ASCII}/ && $email =~ /^$addr_spec$/);
    if ($ret) {
        # We assume these checks to suffice to consider the address untainted.
        trick_taint($_[0]);
    }
    return $ret ? 1 : 0;
}

sub check_email_syntax {
    my ($addr) = @_;

    unless (validate_email_syntax(@_)) {
        my $email = $addr . Bugzilla->params->{'emailsuffix'};
        ThrowUserError('illegal_email_address', { addr => $email });
    }
}

sub validate_date {
    my ($date) = @_;
    my $date2;

    # $ts is undefined if the parser fails.
    my $ts = str2time($date);
    if ($ts) {
        $date2 = time2str("%Y-%m-%d", $ts);

        $date =~ s/(\d+)-0*(\d+?)-0*(\d+?)/$1-$2-$3/; 
        $date2 =~ s/(\d+)-0*(\d+?)-0*(\d+?)/$1-$2-$3/;
    }
    my $ret = ($ts && $date eq $date2);
    return $ret ? 1 : 0;
}

sub validate_time {
    my ($time) = @_;
    my $time2;

    # $ts is undefined if the parser fails.
    my $ts = str2time($time);
    if ($ts) {
        $time2 = time2str("%H:%M:%S", $ts);
        if ($time =~ /^(\d{1,2}):(\d\d)(?::(\d\d))?$/) {
            $time = sprintf("%02d:%02d:%02d", $1, $2, $3 || 0);
        }
    }
    my $ret = ($ts && $time eq $time2);
    return $ret ? 1 : 0;
}

sub is_7bit_clean {
    return $_[0] !~ /[^\x20-\x7E\x0A\x0D]/;
}

sub clean_text {
    my $dtext = shift;
    if ($dtext) {
        # change control characters into a space
        $dtext =~ s/[\x00-\x1F\x7F]+/ /g;
    }
    return trim($dtext);
}

sub on_main_db (&) {
    my $code = shift;
    my $original_dbh = Bugzilla->dbh;
    Bugzilla->request_cache->{dbh} = Bugzilla->dbh_main;
    $code->();
    Bugzilla->request_cache->{dbh} = $original_dbh;
}

sub get_text {
    my ($name, $vars) = @_;
    my $template = Bugzilla->template_inner;
    $vars ||= {};
    $vars->{'message'} = $name;
    my $message;
    $template->process('global/message.txt.tmpl', $vars, \$message)
      || ThrowTemplateError($template->error());

    # Remove the indenting that exists in messages.html.tmpl.
    $message =~ s/^    //gm;
    return $message;
}

sub template_var {
    my $name = shift;
    my $request_cache = Bugzilla->request_cache;
    my $cache = $request_cache->{util_template_var} ||= {};
    my $lang = $request_cache->{template_current_lang}->[0];
    return $cache->{$lang}->{$name} if defined $cache->{$lang};

    my $template = Bugzilla->template_inner($lang);
    my %vars;
    # Note: If we suddenly start needing a lot of template_var variables,
    # they should move into their own template, not field-descs.
    $template->process('global/field-descs.none.tmpl',
                       { vars => \%vars, in_template_var => 1 })
      || ThrowTemplateError($template->error());

    $cache->{$lang} = \%vars;
    return $vars{$name};
}

sub display_value {
    my ($field, $value) = @_;
    return template_var('value_descs')->{$field}->{$value} // $value;
}

sub disable_utf8 {
    if (Bugzilla->params->{'utf8'}) {
        binmode STDOUT, ':bytes'; # Turn off UTF8 encoding.
    }
}

use constant UTF8_ACCIDENTAL => qw(shiftjis big5-eten euc-kr euc-jp);

sub detect_encoding {
    my $data = shift;

    Bugzilla->feature('detect_charset')
      || ThrowUserError('feature_disabled', { feature => 'detect_charset' });

    require Encode::Detect::Detector;
    import Encode::Detect::Detector 'detect';

    my $encoding = detect($data);
    $encoding = resolve_alias($encoding) if $encoding;

    # Encode::Detect is bad at detecting certain charsets, but Encode::Guess
    # is better at them. Here's the details:

    # shiftjis, big5-eten, euc-kr, and euc-jp: (Encode::Detect
    # tends to accidentally mis-detect UTF-8 strings as being
    # these encodings.)
    if ($encoding && grep($_ eq $encoding, UTF8_ACCIDENTAL)) {
        $encoding = undef;
        my $decoder = guess_encoding($data, UTF8_ACCIDENTAL);
        $encoding = $decoder->name if ref $decoder;
    }

    # Encode::Detect sometimes mis-detects various ISO encodings as iso-8859-8,
    # or cp1255, but Encode::Guess can usually tell which one it is.
    if ($encoding && ($encoding eq 'iso-8859-8' || $encoding eq 'cp1255')) {
        my $decoded_as = _guess_iso($data, 'iso-8859-8', 
            # These are ordered this way because it gives the most 
            # accurate results.
            qw(cp1252 iso-8859-7 iso-8859-2));
        $encoding = $decoded_as if $decoded_as;
    }

    return $encoding;
}

# A helper for detect_encoding.
sub _guess_iso {
    my ($data, $versus, @isos) = (shift, shift, shift);

    my $encoding;
    foreach my $iso (@isos) {
        my $decoder = guess_encoding($data, ($iso, $versus));
        if (ref $decoder) {
            $encoding = $decoder->name if ref $decoder;
            last;
        }
    }
    return $encoding;
}

1;

__END__

=head1 NAME

Bugzilla::Util - Generic utility functions for bugzilla

=head1 SYNOPSIS

  use Bugzilla::Util;

  # Functions for dealing with variable tainting
  trick_taint($var);
  detaint_natural($var);
  detaint_signed($var);

  # Functions for quoting
  html_quote($var);
  url_quote($var);
  xml_quote($var);
  email_filter($var);

  # Functions that tell you about your environment
  my $is_cgi   = i_am_cgi();
  my $urlbase  = correct_urlbase();

  # Data manipulation
  ($removed, $added) = diff_arrays(\@old, \@new);

  # Functions for manipulating strings
  $val = trim(" abc ");
  $wrapped = wrap_comment($comment);

  # Functions for formatting time
  format_time($time);
  datetime_from($time, $timezone);

  # Cryptographic Functions
  $crypted_password = bz_crypt($password);
  $new_password = generate_random_password($password_length);

  # Validation Functions
  validate_email_syntax($email);
  check_email_syntax($email);
  validate_date($date);

  # DB-related functions
  on_main_db {
     ... code here ...
  };

=head1 DESCRIPTION

This package contains various utility functions which do not belong anywhere
else.

B<It is not intended as a general dumping group for something which
people feel might be useful somewhere, someday>. Do not add methods to this
package unless it is intended to be used for a significant number of files,
and it does not belong anywhere else.

=head1 FUNCTIONS

This package provides several types of routines:

=head2 Tainting

Several functions are available to deal with tainted variables. B<Use these
with care> to avoid security holes.

=over 4

=item C<trick_taint($val)>

Tricks perl into untainting a particular variable.

Use trick_taint() when you know that there is no way that the data
in a scalar can be tainted, but taint mode still bails on it.

B<WARNING!! Using this routine on data that really could be tainted defeats
the purpose of taint mode.  It should only be used on variables that have been
sanity checked in some way and have been determined to be OK.>

=item C<detaint_natural($num)>

This routine detaints a natural number. It returns a true value if the
value passed in was a valid natural number, else it returns false. You
B<MUST> check the result of this routine to avoid security holes.

=item C<detaint_signed($num)>

This routine detaints a signed integer. It returns a true value if the
value passed in was a valid signed integer, else it returns false. You
B<MUST> check the result of this routine to avoid security holes.

=back

=head2 Quoting

Some values may need to be quoted from perl. However, this should in general
be done in the template where possible.

=over 4

=item C<html_quote($val)>

Returns a value quoted for use in HTML, with &, E<lt>, E<gt>, E<34> and @ being
replaced with their appropriate HTML entities.  Also, Unicode BiDi controls are
deleted.

=item C<html_light_quote($val)>

Returns a string where only explicitly allowed HTML elements and attributes
are kept. All HTML elements and attributes not being in the whitelist are either
escaped (if HTML::Scrubber is not installed) or removed.

=item C<url_quote($val)>

Quotes characters so that they may be included as part of a url.

=item C<css_class_quote($val)>

Quotes characters so that they may be used as CSS class names. Spaces
and forward slashes are replaced by underscores.

=item C<xml_quote($val)>

This is similar to C<html_quote>, except that ' is escaped to &apos;. This
is kept separate from html_quote partly for compatibility with previous code
(for &apos;) and partly for future handling of non-ASCII characters.

=item C<email_filter>

Removes the hostname from email addresses in the string, if the user
currently viewing Bugzilla is logged out. If the user is logged-in,
this filter just returns the input string.

=back

=head2 Environment and Location

Functions returning information about your environment or location.

=over 4

=item C<i_am_cgi()>

Tells you whether or not you are being run as a CGI script in a web
server. For example, it would return false if the caller is running
in a command-line script.

=item C<correct_urlbase()>

Returns either the C<sslbase> or C<urlbase> parameter, depending on the
current setting for the C<ssl_redirect> parameter.

=item C<remote_ip()>

Returns the IP address of the remote client. If Bugzilla is behind
a trusted proxy, it will get the remote IP address by looking at the
X-Forwarded-For header.

=item C<validate_ip($ip)>

Returns the sanitized IP address if it is a valid IPv4 or IPv6 address,
else returns undef.

=item C<use_attachbase()>

Returns true if an alternate host is used to display attachments; false
otherwise.

=back

=head2 Data Manipulation

=over 4

=item C<diff_arrays(\@old, \@new)>

 Description: Takes two arrayrefs, and will tell you what it takes to 
              get from @old to @new.
 Params:      @old = array that you are changing from
              @new = array that you are changing to
 Returns:     A list of two arrayrefs. The first is a reference to an 
              array containing items that were removed from @old. The
              second is a reference to an array containing items
              that were added to @old. If both returned arrays are 
              empty, @old and @new contain the same values.

=back

=head2 String Manipulation

=over 4

=item C<trim($str)>

Removes any leading or trailing whitespace from a string. This routine does not
modify the existing string.

=item C<wrap_hard($string, $size)>

Wraps a string, so that a line is I<never> longer than C<$size>.
Returns the string, wrapped.

=item C<wrap_comment($comment)>

Takes a bug comment, and wraps it to the appropriate length. The length is
currently specified in C<Bugzilla::Constants::COMMENT_COLS>. Lines beginning
with ">" are assumed to be quotes, and they will not be wrapped.

The intended use of this function is to wrap comments that are about to be
displayed or emailed. Generally, wrapped text should not be stored in the
database.

=item C<find_wrap_point($string, $maxpos)>

Search for a comma, a whitespace or a hyphen to split $string, within the first
$maxpos characters. If none of them is found, just split $string at $maxpos.
The search starts at $maxpos and goes back to the beginning of the string.

=item C<is_7bit_clean($str)>

Returns true is the string contains only 7-bit characters (ASCII 32 through 126,
ASCII 10 (LineFeed) and ASCII 13 (Carrage Return).

=item C<disable_utf8()>

Disable utf8 on STDOUT (and display raw data instead).

=item C<detect_encoding($str)>

Guesses what encoding a given data is encoded in, returning the canonical name
of the detected encoding (which may be different from the MIME charset 
specification).

=item C<clean_text($str)>
Returns the parameter "cleaned" by exchanging non-printable characters with spaces.
Specifically characters (ASCII 0 through 31) and (ASCII 127) will become ASCII 32 (Space).

=item C<get_text>

=over

=item B<Description>

This is a method of getting localized strings within Bugzilla code.
Use this when you don't want to display a whole template, you just
want a particular string.

It uses the F<global/message.txt.tmpl> template to return a string.

=item B<Params>

=over

=item C<$message> - The identifier for the message.

=item C<$vars> - A hashref. Any variables you want to pass to the template.

=back

=item B<Returns>

A string.

=back


=item C<template_var>

This is a method of getting the value of a variable from a template in
Perl code. The available variables are in the C<global/field-descs.none.tmpl>
template. Just pass in the name of the variable that you want the value of.


=back

=head2 Formatting Time

=over 4

=item C<format_time($time)>

Takes a time and converts it to the desired format and timezone.
If no format is given, the routine guesses the correct one and returns
an empty array if it cannot. If no timezone is given, the user's timezone
is used, as defined in his preferences.

This routine is mainly called from templates to filter dates, see
"FILTER time" in L<Bugzilla::Template>.

=item C<datetime_from($time, $timezone)>

Returns a DateTime object given a date string. If the string is not in some
valid date format that C<strptime> understands, we return C<undef>.

You can optionally specify a timezone for the returned date. If not
specified, defaults to the currently-logged-in user's timezone, or
the Bugzilla server's local timezone if there isn't a logged-in user.

=back

=head2 Cryptography

=over 4

=item C<bz_crypt($password, $salt)>

Takes a string and returns a hashed (encrypted) value for it, using a
random salt. An optional salt string may also be passed in.

Please always use this function instead of the built-in perl C<crypt>
function, when checking or setting a password. Bugzilla does not use
C<crypt>.

=begin undocumented

Random salts are generated because the alternative is usually
to use the first two characters of the password itself, and since
the salt appears in plaintext at the beginning of the encrypted
password string this has the effect of revealing the first two
characters of the password to anyone who views the encrypted version.

=end undocumented

=item C<generate_random_password($password_length)>

Returns an alphanumeric string with the specified length
(10 characters by default). Use this function to generate passwords
and tokens.

=back

=head2 Validation

=over 4

=item C<validate_email_syntax($email)>

Do a syntax checking for a legal email address and returns 1 if
the check is successful, else returns 0.
Untaints C<$email> if successful.

=item C<check_email_syntax($email)>

Do a syntax checking for a legal email address and throws an error
if the check fails.
Untaints C<$email> if successful.

=item C<validate_date($date)>

Make sure the date has the correct format and returns 1 if
the check is successful, else returns 0.

=back

=head2 Database

=over

=item C<on_main_db>

Runs a block of code always on the main DB. Useful for when you're inside
a subroutine and need to do some writes to the database, but don't know
if Bugzilla is currently using the shadowdb or not. Used like:

 on_main_db {
     my $dbh = Bugzilla->dbh;
     $dbh->do("INSERT ...");
 }

=back

=head1 B<Methods in need of POD>

=over

=item do_ssl_redirect_if_required

=item validate_time

=item is_ipv4

=item is_ipv6

=item display_value

=back