# Copyright (C) 2002-2006 Python Software Foundation

[0] Fix | Delete

# Author: Ben Gertzfield

[1] Fix | Delete

# Contact: email-sig@python.org

[2] Fix | Delete

[3] Fix | Delete

"""Base64 content transfer encoding per RFCs 2045-2047.

[4] Fix | Delete

[5] Fix | Delete

This module handles the content transfer encoding method defined in RFC 2045

[6] Fix | Delete

to encode arbitrary 8-bit data using the three 8-bit bytes in four 7-bit

[7] Fix | Delete

characters encoding known as Base64.

[8] Fix | Delete

[9] Fix | Delete

It is used in the MIME standards for email to attach images, audio, and text

[10] Fix | Delete

using some 8-bit character sets to messages.

[11] Fix | Delete

[12] Fix | Delete

This module provides an interface to encode and decode both headers and bodies

[13] Fix | Delete

with Base64 encoding.

[14] Fix | Delete

[15] Fix | Delete

RFC 2045 defines a method for including character set information in an

[16] Fix | Delete

`encoded-word' in a header. This method is commonly used for 8-bit real names

[17] Fix | Delete

in To:, From:, Cc:, etc. fields, as well as Subject: lines.

[18] Fix | Delete

[19] Fix | Delete

This module does not do the line wrapping or end-of-line character conversion

[20] Fix | Delete

necessary for proper internationalized headers; it only does dumb encoding and

[21] Fix | Delete

decoding. To deal with the various line wrapping issues, use the email.header

[22] Fix | Delete

module.

[23] Fix | Delete

"""

[24] Fix | Delete

[25] Fix | Delete

__all__ = [

[26] Fix | Delete

'base64_len',

[27] Fix | Delete

'body_decode',

[28] Fix | Delete

'body_encode',

[29] Fix | Delete

'decode',

[30] Fix | Delete

'decodestring',

[31] Fix | Delete

'encode',

[32] Fix | Delete

'encodestring',

[33] Fix | Delete

'header_encode',

[34] Fix | Delete

]

[35] Fix | Delete

[36] Fix | Delete

[37] Fix | Delete

from binascii import b2a_base64, a2b_base64

[38] Fix | Delete

from email.utils import fix_eols

[39] Fix | Delete

[40] Fix | Delete

CRLF = '\r\n'

[41] Fix | Delete

NL = '\n'

[42] Fix | Delete

EMPTYSTRING = ''

[43] Fix | Delete

[44] Fix | Delete

# See also Charset.py

[45] Fix | Delete

MISC_LEN = 7

[46] Fix | Delete

[47] Fix | Delete

[48] Fix | Delete



[49] Fix | Delete

# Helpers

[50] Fix | Delete

def base64_len(s):

[51] Fix | Delete

"""Return the length of s when it is encoded with base64."""

[52] Fix | Delete

groups_of_3, leftover = divmod(len(s), 3)

[53] Fix | Delete

# 4 bytes out for each 3 bytes (or nonzero fraction thereof) in.

[54] Fix | Delete

# Thanks, Tim!

[55] Fix | Delete

n = groups_of_3 * 4

[56] Fix | Delete

if leftover:

[57] Fix | Delete

n += 4

[58] Fix | Delete

return n

[59] Fix | Delete

[60] Fix | Delete

[61] Fix | Delete



[62] Fix | Delete

def header_encode(header, charset='iso-8859-1', keep_eols=False,

[63] Fix | Delete

maxlinelen=76, eol=NL):

[64] Fix | Delete

"""Encode a single header line with Base64 encoding in a given charset.

[65] Fix | Delete

[66] Fix | Delete

Defined in RFC 2045, this Base64 encoding is identical to normal Base64

[67] Fix | Delete

encoding, except that each line must be intelligently wrapped (respecting

[68] Fix | Delete

the Base64 encoding), and subsequent lines must start with a space.

[69] Fix | Delete

[70] Fix | Delete

charset names the character set to use to encode the header. It defaults

[71] Fix | Delete

to iso-8859-1.

[72] Fix | Delete

[73] Fix | Delete

End-of-line characters (\\r, \\n, \\r\\n) will be automatically converted

[74] Fix | Delete

to the canonical email line separator \\r\\n unless the keep_eols

[75] Fix | Delete

parameter is True (the default is False).

[76] Fix | Delete

[77] Fix | Delete

Each line of the header will be terminated in the value of eol, which

[78] Fix | Delete

defaults to "\\n". Set this to "\\r\\n" if you are using the result of

[79] Fix | Delete

this function directly in email.

[80] Fix | Delete

[81] Fix | Delete

The resulting string will be in the form:

[82] Fix | Delete

[83] Fix | Delete

"=?charset?b?WW/5ciBtYXp66XLrIHf8eiBhIGhhbXBzdGHuciBBIFlv+XIgbWF6euly?=\\n

[84] Fix | Delete

=?charset?b?6yB3/HogYSBoYW1wc3Rh7nIgQkMgWW/5ciBtYXp66XLrIHf8eiBhIGhh?="

[85] Fix | Delete

[86] Fix | Delete

with each line wrapped at, at most, maxlinelen characters (defaults to 76

[87] Fix | Delete

characters).

[88] Fix | Delete

"""

[89] Fix | Delete

# Return empty headers unchanged

[90] Fix | Delete

if not header:

[91] Fix | Delete

return header

[92] Fix | Delete

[93] Fix | Delete

if not keep_eols:

[94] Fix | Delete

header = fix_eols(header)

[95] Fix | Delete

[96] Fix | Delete

# Base64 encode each line, in encoded chunks no greater than maxlinelen in

[97] Fix | Delete

# length, after the RFC chrome is added in.

[98] Fix | Delete

base64ed = []

[99] Fix | Delete

max_encoded = maxlinelen - len(charset) - MISC_LEN

[100] Fix | Delete

max_unencoded = max_encoded * 3 // 4

[101] Fix | Delete

[102] Fix | Delete

for i in range(0, len(header), max_unencoded):

[103] Fix | Delete

base64ed.append(b2a_base64(header[i:i+max_unencoded]))

[104] Fix | Delete

[105] Fix | Delete

# Now add the RFC chrome to each encoded chunk

[106] Fix | Delete

lines = []

[107] Fix | Delete

for line in base64ed:

[108] Fix | Delete

# Ignore the last character of each line if it is a newline

[109] Fix | Delete

if line.endswith(NL):

[110] Fix | Delete

line = line[:-1]

[111] Fix | Delete

# Add the chrome

[112] Fix | Delete

lines.append('=?%s?b?%s?=' % (charset, line))

[113] Fix | Delete

# Glue the lines together and return it. BAW: should we be able to

[114] Fix | Delete

# specify the leading whitespace in the joiner?

[115] Fix | Delete

joiner = eol + ' '

[116] Fix | Delete

return joiner.join(lines)

[117] Fix | Delete

[118] Fix | Delete

[119] Fix | Delete



[120] Fix | Delete

def encode(s, binary=True, maxlinelen=76, eol=NL):

[121] Fix | Delete

"""Encode a string with base64.

[122] Fix | Delete

[123] Fix | Delete

Each line will be wrapped at, at most, maxlinelen characters (defaults to

[124] Fix | Delete

76 characters).

[125] Fix | Delete

[126] Fix | Delete

If binary is False, end-of-line characters will be converted to the

[127] Fix | Delete

canonical email end-of-line sequence \\r\\n. Otherwise they will be left

[128] Fix | Delete

verbatim (this is the default).

[129] Fix | Delete

[130] Fix | Delete

Each line of encoded text will end with eol, which defaults to "\\n". Set

[131] Fix | Delete

this to "\\r\\n" if you will be using the result of this function directly

[132] Fix | Delete

in an email.

[133] Fix | Delete

"""

[134] Fix | Delete

if not s:

[135] Fix | Delete

return s

[136] Fix | Delete

[137] Fix | Delete

if not binary:

[138] Fix | Delete

s = fix_eols(s)

[139] Fix | Delete

[140] Fix | Delete

encvec = []

[141] Fix | Delete

max_unencoded = maxlinelen * 3 // 4

[142] Fix | Delete

for i in range(0, len(s), max_unencoded):

[143] Fix | Delete

# BAW: should encode() inherit b2a_base64()'s dubious behavior in

[144] Fix | Delete

# adding a newline to the encoded string?

[145] Fix | Delete

enc = b2a_base64(s[i:i + max_unencoded])

[146] Fix | Delete

if enc.endswith(NL) and eol != NL:

[147] Fix | Delete

enc = enc[:-1] + eol

[148] Fix | Delete

encvec.append(enc)

[149] Fix | Delete

return EMPTYSTRING.join(encvec)

[150] Fix | Delete

[151] Fix | Delete

[152] Fix | Delete

# For convenience and backwards compatibility w/ standard base64 module

[153] Fix | Delete

body_encode = encode

[154] Fix | Delete

encodestring = encode

[155] Fix | Delete

[156] Fix | Delete

[157] Fix | Delete



[158] Fix | Delete

def decode(s, convert_eols=None):

[159] Fix | Delete

"""Decode a raw base64 string.

[160] Fix | Delete

[161] Fix | Delete

If convert_eols is set to a string value, all canonical email linefeeds,

[162] Fix | Delete

e.g. "\\r\\n", in the decoded text will be converted to the value of

[163] Fix | Delete

convert_eols. os.linesep is a good choice for convert_eols if you are

[164] Fix | Delete

decoding a text attachment.

[165] Fix | Delete

[166] Fix | Delete

This function does not parse a full MIME header value encoded with

[167] Fix | Delete

base64 (like =?iso-8859-1?b?bmloISBuaWgh?=) -- please use the high

[168] Fix | Delete

level email.header class for that functionality.

[169] Fix | Delete

"""

[170] Fix | Delete

if not s:

[171] Fix | Delete

return s

[172] Fix | Delete

[173] Fix | Delete

dec = a2b_base64(s)

[174] Fix | Delete

if convert_eols:

[175] Fix | Delete

return dec.replace(CRLF, convert_eols)

[176] Fix | Delete

return dec

[177] Fix | Delete

[178] Fix | Delete

[179] Fix | Delete

# For convenience and backwards compatibility w/ standard base64 module

[180] Fix | Delete

body_decode = decode

[181] Fix | Delete

decodestring = decode

[182] Fix | Delete

[183] Fix | Delete