base64mime.py revision 98297ee7815939b124156e438b22bd652d67b5db 
1# Copyright (C) 2002-2007 Python Software Foundation
2# Author: Ben Gertzfield
3# Contact: email-sig@python.org
4
5"""Base64 content transfer encoding per RFCs 2045-2047.
6
7This module handles the content transfer encoding method defined in RFC 2045
8to encode arbitrary 8-bit data using the three 8-bit bytes in four 7-bit
9characters encoding known as Base64.
10
11It is used in the MIME standards for email to attach images, audio, and text
12using some 8-bit character sets to messages.
13
14This module provides an interface to encode and decode both headers and bodies
15with Base64 encoding.
16
17RFC 2045 defines a method for including character set information in an
18`encoded-word' in a header.  This method is commonly used for 8-bit real names
19in To:, From:, Cc:, etc. fields, as well as Subject: lines.
20
21This module does not do the line wrapping or end-of-line character conversion
22necessary for proper internationalized headers; it only does dumb encoding and
23decoding.  To deal with the various line wrapping issues, use the email.Header
24module.
25"""
26
27__all__ = [
28    'body_decode',
29    'body_encode',
30    'decode',
31    'decodestring',
32    'encode',
33    'encodestring',
34    'header_encode',
35    'header_length',
36    ]
37
38import re
39
40from base64 import b64encode
41from binascii import b2a_base64, a2b_base64
42
43CRLF = '\r\n'
44NL = '\n'
45EMPTYSTRING = ''
46
47# See also Charset.py
48MISC_LEN = 7
49
50
51
52# Helpers
53def header_length(bytearray):
54    """Return the length of s when it is encoded with base64."""
55    groups_of_3, leftover = divmod(len(bytearray), 3)
56    # 4 bytes out for each 3 bytes (or nonzero fraction thereof) in.
57    n = groups_of_3 * 4
58    if leftover:
59        n += 4
60    return n
61
62
63
64def header_encode(header_bytes, charset='iso-8859-1'):
65    """Encode a single header line with Base64 encoding in a given charset.
66
67    charset names the character set to use to encode the header.  It defaults
68    to iso-8859-1.  Base64 encoding is defined in RFC 2045.
69    """
70    # Return empty headers unchanged
71    if not header_bytes:
72        return str(header_bytes)
73    encoded = b64encode(header_bytes).decode("ascii")
74    return '=?%s?b?%s?=' % (charset, encoded)
75
76
77
78def body_encode(s, maxlinelen=76, eol=NL):
79    """Encode a string with base64.
80
81    Each line will be wrapped at, at most, maxlinelen characters (defaults to
82    76 characters).
83
84    Each line of encoded text will end with eol, which defaults to "\\n".  Set
85    this to "\r\n" if you will be using the result of this function directly
86    in an email.
87    """
88    if not s:
89        return s
90
91    encvec = []
92    max_unencoded = maxlinelen * 3 // 4
93    for i in range(0, len(s), max_unencoded):
94        # BAW: should encode() inherit b2a_base64()'s dubious behavior in
95        # adding a newline to the encoded string?
96        enc = b2a_base64(s[i:i + max_unencoded]).decode("ascii")
97        if enc.endswith(NL) and eol != NL:
98            enc = enc[:-1] + eol
99        encvec.append(enc)
100    return EMPTYSTRING.join(encvec)
101
102
103
104def decode(string):
105    """Decode a raw base64 string, returning a bytes object.
106
107    This function does not parse a full MIME header value encoded with
108    base64 (like =?iso-8895-1?b?bmloISBuaWgh?=) -- please use the high
109    level email.Header class for that functionality.
110    """
111    if not string:
112        return bytes()
113    elif isinstance(string, str):
114        return a2b_base64(string.encode('raw-unicode-escape'))
115    else:
116        return a2b_base64(s)
117
118
119# For convenience and backwards compatibility w/ standard base64 module
120body_decode = decode
121decodestring = decode
122