Commit | Line | Data |
---|---|---|
9ee028d2 PJ |
1 | #!/usr/bin/perl |
2 | ||
3 | # b2m.pl - Script to convert a Babyl file to an mbox file | |
4 | ||
d1e4cefd GM |
5 | # Copyright (C) 2002, 2003, 2004, 2005, 2006, 2007 |
6 | # Free Software Foundation, Inc. | |
7 | ||
9ee028d2 PJ |
8 | # This program is free software; you can redistribute it and/or modify |
9 | # it under the terms of the GNU General Public License as published by | |
10 | # the Free Software Foundation; either version 2 of the License, or | |
11 | # (at your option) any later version. | |
12 | ||
13 | # This program is distributed in the hope that it will be useful, but | |
14 | # WITHOUT ANY WARRANTY; without even the implied warranty of | |
15 | # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU | |
16 | # General Public License for more details. | |
17 | ||
18 | # You should have received a copy of the GNU General Public License | |
19 | # along with this program; if not, write to the Free Software | |
364c38d3 | 20 | # Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 |
9ee028d2 PJ |
21 | # USA. |
22 | ||
23 | # Maintained by Jonathan Kamens <jik@kamens.brookline.ma.us>. | |
24 | ||
25 | # Requires CPAN modules: MailTools (for Mail::Address), TimeDate (for | |
26 | # Date::Parse). | |
27 | ||
28 | use warnings; | |
29 | use strict; | |
30 | use File::Basename; | |
31 | use Getopt::Long; | |
32 | use Mail::Address; | |
33 | use Date::Parse; | |
34 | ||
35 | my($whoami) = basename $0; | |
364c38d3 | 36 | my($version) = '$Revision$'; |
9ee028d2 PJ |
37 | my($usage) = "Usage: $whoami [--help] [--version] [--[no]full-headers] [Babyl-file] |
38 | \tBy default, full headers are printed.\n"; | |
39 | ||
40 | my($opt_help, $opt_version); | |
41 | my($opt_full_headers) = 1; | |
42 | ||
43 | die $usage if (! GetOptions( | |
44 | 'help' => \$opt_help, | |
45 | 'version' => \$opt_version, | |
46 | 'full-headers!' => \$opt_full_headers, | |
47 | )); | |
48 | ||
49 | if ($opt_help) { | |
50 | print $usage; | |
51 | exit; | |
52 | } | |
53 | elsif ($opt_version) { | |
54 | print "$whoami version: $version\n"; | |
55 | exit; | |
56 | } | |
57 | ||
58 | die $usage if (@ARGV > 1); | |
59 | ||
60 | $/ = "\n\037"; | |
61 | ||
62 | if (<> !~ /^BABYL OPTIONS:/) { | |
63 | die "$whoami: $ARGV is not a Babyl file\n$usage"; | |
64 | } | |
65 | ||
66 | while (<>) { | |
67 | my($msg_num) = $. - 1; | |
85271e41 | 68 | my($labels, $pruned, $full_header, $header); |
9ee028d2 PJ |
69 | my($from_line, $from_addr); |
70 | my($time); | |
71 | ||
72 | # This will strip the initial form feed, any whitespace that may | |
73 | # be following it, and then a newline | |
74 | s/^\s+//; | |
75 | # This will strip the ^_ off of the end of the message | |
76 | s/\037$//; | |
77 | ||
78 | if (! s/(.*)\n//) { | |
79 | malformatted: | |
80 | warn "$whoami: message $msg_num in $ARGV is malformatted\n"; | |
81 | next; | |
82 | } | |
83 | $labels = $1; | |
84 | ||
85271e41 | 85 | # Strip the integer indicating whether the header is pruned |
177c0ea7 | 86 | $labels =~ s/^(\d+)[,\s]*//; |
85271e41 PJ |
87 | $pruned = $1; |
88 | ||
89 | s/(?:((?:.+\n)+)\n*)?\*\*\* EOOH \*\*\*\n+// || goto malformatted; | |
9ee028d2 PJ |
90 | $full_header = $1; |
91 | ||
92 | if (s/((?:.+\n)+)\n+//) { | |
93 | $header = $1; | |
94 | } | |
95 | else { | |
96 | # Message has no body | |
97 | $header = $_; | |
98 | $_ = ''; | |
99 | } | |
100 | ||
85271e41 PJ |
101 | # "$pruned eq '0'" is different from "! $pruned". We want to make |
102 | # sure that we found a valid label line which explicitly indicated | |
103 | # that the header was not pruned. | |
104 | if ((! $full_header) || ($pruned eq '0')) { | |
9ee028d2 PJ |
105 | $full_header = $header; |
106 | } | |
107 | ||
649fc2c5 PJ |
108 | # End message with two newlines (some mbox parsers require a blank |
109 | # line before the next "From " line). | |
110 | s/\s+$/\n\n/; | |
9ee028d2 PJ |
111 | |
112 | # Quote "^From " | |
113 | s/(^|\n)From /$1>From /g; | |
114 | ||
9ee028d2 PJ |
115 | # Strip extra commas and whitespace from the end |
116 | $labels =~ s/[,\s]+$//; | |
117 | # Now collapse extra commas and whitespace in the remaining label string | |
118 | $labels =~ s/[,\s]+/, /g; | |
177c0ea7 | 119 | |
9ee028d2 PJ |
120 | foreach my $rmail_header qw(summary-line x-coding-system) { |
121 | $full_header =~ s/(^|\n)$rmail_header:.*\n/$1/i; | |
122 | } | |
123 | ||
124 | if ($full_header =~ s/(^|\n)mail-from:\s*(From .*)\n/$1/i) { | |
125 | ($from_line = $2) =~ s/\s*$/\n/; | |
126 | } | |
127 | else { | |
128 | foreach my $addr_header qw(return-path from really-from sender) { | |
c89ed6ef | 129 | if ($full_header =~ /(?:^|\n)$addr_header:\s*(.*\n(?:\B.*\n)*)/i) { |
9ee028d2 PJ |
130 | my($addr) = Mail::Address->parse($1); |
131 | $from_addr = $addr->address($addr); | |
132 | last; | |
133 | } | |
134 | } | |
135 | ||
136 | if (! $from_addr) { | |
137 | $from_addr = "Babyl_to_mail_by_$whoami\@localhost"; | |
138 | } | |
139 | ||
140 | if ($full_header =~ /(?:^|\n)date:\s*(\S.*\S)/i) { | |
141 | $time = str2time($1); | |
142 | } | |
143 | ||
144 | if (! $time) { | |
145 | # No Date header or we failed to parse it | |
146 | $time = time; | |
147 | } | |
148 | ||
149 | $from_line = "From " . $from_addr . " " . localtime($time) . "\n"; | |
150 | } | |
151 | ||
152 | print($from_line, ($opt_full_headers ? $full_header : $header), | |
153 | ($labels ? "X-Babyl-Labels: $labels\n" : ""), "\n", | |
154 | $_) || die "$whoami: error writing to stdout: $!\n"; | |
155 | } | |
156 | ||
157 | close(STDOUT) || die "$whoami: Error closing stdout: $!\n"; | |
ab5796a9 MB |
158 | |
159 | # arch-tag: 8c7c8ab0-721c-46d7-ba3e-139801240aa8 |