Shortcuts on this page
r m x toggle line displays
j k next/prev highlighted chunk
0 (zero) top of page
1 (one) first highlighted chunk
1# Copyright (C) 2002-2007 Python Software Foundation
2# Author: Ben Gertzfield
3# Contact: email-sig@python.org
5"""Base64 content transfer encoding per RFCs 2045-2047.
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.
11It is used in the MIME standards for email to attach images, audio, and text
12using some 8-bit character sets to messages.
14This module provides an interface to encode and decode both headers and bodies
15with Base64 encoding.
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.
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"""
26from __future__ import unicode_literals
27from __future__ import division
28from __future__ import absolute_import
29from future.builtins import range
30from future.builtins import bytes
31from future.builtins import str
33__all__ = [
34 'body_decode',
35 'body_encode',
36 'decode',
37 'decodestring',
38 'header_encode',
39 'header_length',
40 ]
43from base64 import b64encode
44from binascii import b2a_base64, a2b_base64
46CRLF = '\r\n'
47NL = '\n'
48EMPTYSTRING = ''
50# See also Charset.py
51MISC_LEN = 7
54# Helpers
55def header_length(bytearray):
56 """Return the length of s when it is encoded with base64."""
57 groups_of_3, leftover = divmod(len(bytearray), 3)
58 # 4 bytes out for each 3 bytes (or nonzero fraction thereof) in.
59 n = groups_of_3 * 4
60 if leftover:
61 n += 4
62 return n
65def header_encode(header_bytes, charset='iso-8859-1'):
66 """Encode a single header line with Base64 encoding in a given charset.
68 charset names the character set to use to encode the header. It defaults
69 to iso-8859-1. Base64 encoding is defined in RFC 2045.
70 """
71 if not header_bytes:
72 return ""
73 if isinstance(header_bytes, str):
74 header_bytes = header_bytes.encode(charset)
75 encoded = b64encode(header_bytes).decode("ascii")
76 return '=?%s?b?%s?=' % (charset, encoded)
79def body_encode(s, maxlinelen=76, eol=NL):
80 r"""Encode a string with base64.
82 Each line will be wrapped at, at most, maxlinelen characters (defaults to
83 76 characters).
85 Each line of encoded text will end with eol, which defaults to "\n". Set
86 this to "\r\n" if you will be using the result of this function directly
87 in an email.
88 """
89 if not s:
90 return s
92 encvec = []
93 max_unencoded = maxlinelen * 3 // 4
94 for i in range(0, len(s), max_unencoded):
95 # BAW: should encode() inherit b2a_base64()'s dubious behavior in
96 # adding a newline to the encoded string?
97 enc = b2a_base64(s[i:i + max_unencoded]).decode("ascii")
98 if enc.endswith(NL) and eol != NL:
99 enc = enc[:-1] + eol
100 encvec.append(enc)
101 return EMPTYSTRING.join(encvec)
104def decode(string):
105 """Decode a raw base64 string, returning a bytes object.
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(string)
119# For convenience and backwards compatibility w/ standard base64 module
120body_decode = decode
121decodestring = decode