Commit | Line | Data |
6fba102d |
1 | package MIME::Base64; |
2 | |
0a362e9d |
3 | use strict; |
4 | use vars qw(@ISA @EXPORT $VERSION); |
5 | |
6 | require Exporter; |
e1839706 |
7 | @ISA = qw(Exporter); |
0a362e9d |
8 | @EXPORT = qw(encode_base64 decode_base64); |
9 | |
dfed8d37 |
10 | $VERSION = '3.08'; |
0a362e9d |
11 | |
e1839706 |
12 | require XSLoader; |
13 | XSLoader::load('MIME::Base64', $VERSION); |
0a362e9d |
14 | |
15 | *encode = \&encode_base64; |
16 | *decode = \&decode_base64; |
17 | |
18 | 1; |
19 | |
20 | __END__ |
21 | |
6fba102d |
22 | =head1 NAME |
23 | |
24 | MIME::Base64 - Encoding and decoding of base64 strings |
25 | |
26 | =head1 SYNOPSIS |
27 | |
28 | use MIME::Base64; |
29 | |
30 | $encoded = encode_base64('Aladdin:open sesame'); |
31 | $decoded = decode_base64($encoded); |
32 | |
33 | =head1 DESCRIPTION |
34 | |
691d66bd |
35 | This module provides functions to encode and decode strings into and from the |
36 | base64 encoding specified in RFC 2045 - I<MIME (Multipurpose Internet |
37 | Mail Extensions)>. The base64 encoding is designed to represent |
6fba102d |
38 | arbitrary sequences of octets in a form that need not be humanly |
39 | readable. A 65-character subset ([A-Za-z0-9+/=]) of US-ASCII is used, |
40 | enabling 6 bits to be represented per printable character. |
41 | |
42 | The following functions are provided: |
43 | |
44 | =over 4 |
45 | |
6a63fb82 |
46 | =item encode_base64($str) |
47 | |
48 | =item encode_base64($str, $eol); |
6fba102d |
49 | |
50 | Encode data by calling the encode_base64() function. The first |
691d66bd |
51 | argument is the string to encode. The second argument is the |
52 | line-ending sequence to use. It is optional and defaults to "\n". The |
6fba102d |
53 | returned encoded string is broken into lines of no more than 76 |
54 | characters each and it will end with $eol unless it is empty. Pass an |
55 | empty string as second argument if you do not want the encoded string |
691d66bd |
56 | to be broken into lines. |
6fba102d |
57 | |
58 | =item decode_base64($str) |
59 | |
60 | Decode a base64 string by calling the decode_base64() function. This |
61 | function takes a single argument which is the string to decode and |
62 | returns the decoded data. |
63 | |
691d66bd |
64 | Any character not part of the 65-character base64 subset is |
65 | silently ignored. Characters occurring after a '=' padding character |
6fba102d |
66 | are never decoded. |
67 | |
8be5f608 |
68 | If the length of the string to decode, after ignoring |
691d66bd |
69 | non-base64 chars, is not a multiple of 4 or if padding occurs too early, |
6fba102d |
70 | then a warning is generated if perl is running under C<-w>. |
71 | |
72 | =back |
73 | |
691d66bd |
74 | If you prefer not to import these routines into your namespace, you can |
6fba102d |
75 | call them as: |
76 | |
77 | use MIME::Base64 (); |
78 | $encoded = MIME::Base64::encode($decoded); |
79 | $decoded = MIME::Base64::decode($encoded); |
80 | |
81 | =head1 DIAGNOSTICS |
82 | |
691d66bd |
83 | The following warnings can be generated if perl is invoked with the |
6fba102d |
84 | C<-w> switch: |
85 | |
86 | =over 4 |
87 | |
88 | =item Premature end of base64 data |
89 | |
90 | The number of characters to decode is not a multiple of 4. Legal |
91 | base64 data should be padded with one or two "=" characters to make |
4e2e3d15 |
92 | its length a multiple of 4. The decoded result will be the same |
93 | whether the padding is present or not. |
6fba102d |
94 | |
95 | =item Premature padding of base64 data |
96 | |
97 | The '=' padding character occurs as the first or second character |
98 | in a base64 quartet. |
99 | |
100 | =back |
101 | |
e1839706 |
102 | The following exception can be raised: |
103 | |
104 | =over 4 |
105 | |
106 | =item Wide character in subroutine entry |
107 | |
108 | The string passed to encode_base64() contains characters with code |
109 | above 255. The base64 encoding is only defined for single-byte |
110 | characters. Use the Encode module to select the byte encoding you |
111 | want. |
112 | |
113 | =back |
114 | |
6fba102d |
115 | =head1 EXAMPLES |
116 | |
117 | If you want to encode a large file, you should encode it in chunks |
118 | that are a multiple of 57 bytes. This ensures that the base64 lines |
119 | line up and that you do not end up with padding in the middle. 57 |
120 | bytes of data fills one complete base64 line (76 == 57*4/3): |
121 | |
122 | use MIME::Base64 qw(encode_base64); |
123 | |
124 | open(FILE, "/var/log/wtmp") or die "$!"; |
125 | while (read(FILE, $buf, 60*57)) { |
126 | print encode_base64($buf); |
127 | } |
128 | |
129 | or if you know you have enough memory |
130 | |
131 | use MIME::Base64 qw(encode_base64); |
132 | local($/) = undef; # slurp |
133 | print encode_base64(<STDIN>); |
134 | |
135 | The same approach as a command line: |
136 | |
137 | perl -MMIME::Base64 -0777 -ne 'print encode_base64($_)' <file |
138 | |
691d66bd |
139 | Decoding does not need slurp mode if every line contains a multiple |
140 | of four base64 chars: |
6fba102d |
141 | |
142 | perl -MMIME::Base64 -ne 'print decode_base64($_)' <file |
143 | |
e1839706 |
144 | Perl v5.8 and better allow extended Unicode characters in strings. |
145 | Such strings cannot be encoded directly, as the base64 |
146 | encoding is only defined for single-byte characters. The solution is |
147 | to use the Encode module to select the byte encoding you want. For |
148 | example: |
149 | |
150 | use MIME::Base64 qw(encode_base64); |
151 | use Encode qw(encode); |
152 | |
153 | $encoded = encode_base64(encode("UTF-8", "\x{FFFF}\n")); |
154 | print $encoded; |
155 | |
6fba102d |
156 | =head1 COPYRIGHT |
157 | |
691d66bd |
158 | Copyright 1995-1999, 2001-2004 Gisle Aas. |
6fba102d |
159 | |
160 | This library is free software; you can redistribute it and/or |
161 | modify it under the same terms as Perl itself. |
162 | |
163 | Distantly based on LWP::Base64 written by Martijn Koster |
164 | <m.koster@nexor.co.uk> and Joerg Reichelt <j.reichelt@nexor.co.uk> and |
165 | code posted to comp.lang.perl <3pd2lp$6gf@wsinti07.win.tue.nl> by Hans |
166 | Mulder <hansm@wsinti07.win.tue.nl> |
167 | |
691d66bd |
168 | The XS implementation uses code from metamail. Copyright 1991 Bell |
6fba102d |
169 | Communications Research, Inc. (Bellcore) |
170 | |
8be5f608 |
171 | =head1 SEE ALSO |
172 | |
173 | L<MIME::QuotedPrint> |
174 | |
6fba102d |
175 | =cut |