|
| 1 | +import re |
| 2 | +from binascii import b2a_base64, a2b_base64 |
| 3 | + |
| 4 | +ecre = re.compile(r''' |
| 5 | + =\? # literal =? |
| 6 | + (?P<charset>[^?]*?) # non-greedy up to the next ? is the charset |
| 7 | + \? # literal ? |
| 8 | + (?P<encoding>[qb]) # either a "q" or a "b", case insensitive |
| 9 | + \? # literal ? |
| 10 | + (?P<encoded>.*?) # non-greedy up to the next ?= is the encoded string |
| 11 | + \?= # literal ?= |
| 12 | + ''', re.VERBOSE | re.IGNORECASE) |
| 13 | + |
| 14 | +hqre = re.compile(r'^[-a-zA-Z0-9!*+/\[\]., ]+$') |
| 15 | + |
| 16 | +def base64_decode(s, convert_eols=None): |
| 17 | + """Decode a raw base64 string. |
| 18 | +
|
| 19 | + If convert_eols is set to a string value, all canonical email linefeeds, |
| 20 | + e.g. "\\r\\n", in the decoded text will be converted to the value of |
| 21 | + convert_eols. os.linesep is a good choice for convert_eols if you are |
| 22 | + decoding a text attachment. |
| 23 | +
|
| 24 | + This function does not parse a full MIME header value encoded with |
| 25 | + base64 (like =?iso-8895-1?b?bmloISBuaWgh?=) -- please use the high |
| 26 | + level email.Header class for that functionality. |
| 27 | +
|
| 28 | + Taken from 'email' module |
| 29 | + """ |
| 30 | + if not s: |
| 31 | + return s |
| 32 | + |
| 33 | + dec = a2b_base64(s) |
| 34 | + if convert_eols: |
| 35 | + return dec.replace(CRLF, convert_eols) |
| 36 | + return dec |
| 37 | + |
| 38 | +def unquote_match(match): |
| 39 | + """Turn a match in the form =AB to the ASCII character with value 0xab |
| 40 | +
|
| 41 | + Taken from 'email' module |
| 42 | + """ |
| 43 | + s = match.group(0) |
| 44 | + return chr(int(s[1:3], 16)) |
| 45 | + |
| 46 | +def qp_decode(s): |
| 47 | + """Decode a string encoded with RFC 2045 MIME header `Q' encoding. |
| 48 | +
|
| 49 | + This function does not parse a full MIME header value encoded with |
| 50 | + quoted-printable (like =?iso-8895-1?q?Hello_World?=) -- please use |
| 51 | + the high level email.Header class for that functionality. |
| 52 | +
|
| 53 | + Taken from 'email' module |
| 54 | + """ |
| 55 | + s = s.replace('_', ' ') |
| 56 | + return re.sub(r'=\w{2}', unquote_match, s) |
| 57 | + |
| 58 | +def _decode_header(header): |
| 59 | + """Decode a message header value without converting charset. |
| 60 | +
|
| 61 | + Returns a list of (decoded_string, charset) pairs containing each of the |
| 62 | + decoded parts of the header. Charset is None for non-encoded parts of the |
| 63 | + header, otherwise a lower-case string containing the name of the character |
| 64 | + set specified in the encoded string. |
| 65 | +
|
| 66 | + Taken from 'email' module |
| 67 | + """ |
| 68 | + # If no encoding, just return the header |
| 69 | + header = str(header) |
| 70 | + if not ecre.search(header): |
| 71 | + return [(header, None)] |
| 72 | + |
| 73 | + decoded = [] |
| 74 | + dec = '' |
| 75 | + for line in header.splitlines(): |
| 76 | + # This line might not have an encoding in it |
| 77 | + if not ecre.search(line): |
| 78 | + decoded.append((line, None)) |
| 79 | + continue |
| 80 | + |
| 81 | + parts = ecre.split(line) |
| 82 | + while parts: |
| 83 | + unenc = parts.pop(0) |
| 84 | + if unenc: |
| 85 | + if unenc.strip(): |
| 86 | + decoded.append((unenc, None)) |
| 87 | + if parts: |
| 88 | + charset, encoding = [s.lower() for s in parts[0:2]] |
| 89 | + encoded = parts[2] |
| 90 | + dec = '' |
| 91 | + if encoding == 'q': |
| 92 | + dec = qp_decode(encoded) |
| 93 | + elif encoding == 'b': |
| 94 | + dec = base64_decode(encoded) |
| 95 | + else: |
| 96 | + dec = encoded |
| 97 | + |
| 98 | + if decoded and decoded[-1][1] == charset: |
| 99 | + decoded[-1] = (decoded[-1][0] + dec, decoded[-1][1]) |
| 100 | + else: |
| 101 | + decoded.append((dec, charset)) |
| 102 | + del parts[0:3] |
| 103 | + return decoded |
| 104 | + |
| 105 | +def decode_header(hdr): |
| 106 | + """ Decodes rfc2822 encoded header and return utf-8 encoded string |
| 107 | + """ |
| 108 | + if not hdr: |
| 109 | + return None |
| 110 | + outs = u"" |
| 111 | + for section in _decode_header(hdr): |
| 112 | + charset = unaliasCharset(section[1]) |
| 113 | + outs += unicode(section[0], charset or 'iso-8859-1', 'replace') |
| 114 | + return outs.encode('utf-8') |
| 115 | + |
| 116 | +def encode_header(header): |
| 117 | + """ Will encode in quoted-printable encoding only if header |
| 118 | + contains non latin characters |
| 119 | + """ |
| 120 | + |
| 121 | + # Return empty headers unchanged |
| 122 | + if not header: |
| 123 | + return header |
| 124 | + |
| 125 | + global hqre |
| 126 | + # return plain header if it is not contains non-ascii characters |
| 127 | + if hqre.match(header): |
| 128 | + return header |
| 129 | + |
| 130 | + charset = 'utf-8' |
| 131 | + quoted = '' |
| 132 | + #max_encoded = 76 - len(charset) - 7 |
| 133 | + for c in header: |
| 134 | + # Space may be represented as _ instead of =20 for readability |
| 135 | + if c == ' ': |
| 136 | + quoted += '_' |
| 137 | + # These characters can be included verbatim |
| 138 | + elif hqre.match(c): |
| 139 | + quoted += c |
| 140 | + # Otherwise, replace with hex value like =E2 |
| 141 | + else: |
| 142 | + quoted += "=%02X" % ord(c) |
| 143 | + plain = 0 |
| 144 | + |
| 145 | + return '=?%s?q?%s?=' % (charset, quoted) |
| 146 | + |
| 147 | +def unaliasCharset(charset): |
| 148 | + if charset: |
| 149 | + return charset.lower().replace("windows-", 'cp') |
| 150 | + #return charset_table.get(charset.lower(), charset) |
| 151 | + return None |
| 152 | + |
| 153 | +def test(): |
| 154 | + print encode_header("Contrary, Mary") |
| 155 | + #print unaliasCharset('Windows-1251') |
| 156 | + |
| 157 | +if __name__ == '__main__': |
| 158 | + test() |
| 159 | + |
| 160 | +# vim: et |
0 commit comments