# Copyright (C) 2002-2007 Python Software Foundation
# Author: Ben Gertzfield, Barry Warsaw
# Contact: email-sig@python.org
"""Header encoding and decoding functionality."""
from email.errors import HeaderParseError
from email import charset as _charset
Charset = _charset.Charset
USASCII = Charset('us-ascii')
# Match encoded-word strings in the form =?charset?q?Hello_World?=
(?P<charset>[^?]*?) # non-greedy up to the next ? is the charset
(?P<encoding>[qb]) # either a "q" or a "b", case insensitive
(?P<encoded>.*?) # non-greedy up to the next ?= is the encoded string
''', re.VERBOSE | re.IGNORECASE | re.MULTILINE)
# Field name regexp, including trailing colon, but not separating whitespace,
# according to RFC 2822. Character range is from tilde to exclamation mark.
fcre = re.compile(r'[\041-\176]+:$')
# Find a header embedded in a putative header value. Used to check for
# header injection attack.
_embedded_header = re.compile(r'\n[^ \t]+:')
_max_append = email.quoprimime._max_append
def decode_header(header):
"""Decode a message header value without converting charset.
Returns a list of (string, charset) pairs containing each of the decoded
parts of the header. Charset is None for non-encoded parts of the header,
otherwise a lower-case string containing the name of the character set
specified in the encoded string.
header may be a string that may or may not contain RFC2047 encoded words,
or it may be a Header object.
An email.errors.HeaderParseError may be raised when certain decoding error
occurs (e.g. a base64 decoding exception).
# If it is a Header object, we can just return the encoded chunks.
if hasattr(header, '_chunks'):
return [(_charset._encode(string, str(charset)), str(charset))
for string, charset in header._chunks]
# If no encoding, just return the header with no charset.
if not ecre.search(header):
# First step is to parse all the encoded parts into triplets of the form
# (encoded_string, encoding, charset). For unencoded strings, the last
# two parts will be None.
for line in header.splitlines():
unencoded = unencoded.lstrip()
words.append((unencoded, None, None))
charset = parts.pop(0).lower()
encoding = parts.pop(0).lower()
words.append((encoded, encoding, charset))
# Now loop over words and remove words that consist of whitespace
# between two encoded strings.
for n, w in enumerate(words):
if n>1 and w[1] and words[n-2][1] and words[n-1][0].isspace():
for d in reversed(droplist):
# The next step is to decode each encoded word by applying the reverse
# base64 or quopri transformation. decoded_words is now a list of the
# form (decoded_word, charset).
for encoded_string, encoding, charset in words:
# This is an unencoded word.
decoded_words.append((encoded_string, charset))
word = email.quoprimime.header_decode(encoded_string)
decoded_words.append((word, charset))
paderr = len(encoded_string) % 4 # Postel's law: add missing padding
encoded_string += '==='[:4 - paderr]
word = email.base64mime.decode(encoded_string)
raise HeaderParseError('Base64 decoding error')
decoded_words.append((word, charset))
raise AssertionError('Unexpected encoding: ' + encoding)
# Now convert all words to bytes and collapse consecutive runs of
# similarly encoded words.
last_word = last_charset = None
for word, charset in decoded_words:
if isinstance(word, str):
word = bytes(word, 'raw-unicode-escape')
elif charset != last_charset:
collapsed.append((last_word, last_charset))
elif last_charset is None:
last_word += BSPACE + word
collapsed.append((last_word, last_charset))
def make_header(decoded_seq, maxlinelen=None, header_name=None,
"""Create a Header from a sequence of pairs as returned by decode_header()
decode_header() takes a header value string and returns a sequence of
pairs of the format (decoded_string, charset) where charset is the string
name of the character set.
This function takes one of those sequence of pairs and returns a Header
instance. Optional maxlinelen, header_name, and continuation_ws are as in
h = Header(maxlinelen=maxlinelen, header_name=header_name,
continuation_ws=continuation_ws)
for s, charset in decoded_seq:
# None means us-ascii but we can simply pass it on to h.append()
if charset is not None and not isinstance(charset, Charset):
charset = Charset(charset)
def __init__(self, s=None, charset=None,
maxlinelen=None, header_name=None,
continuation_ws=' ', errors='strict'):
"""Create a MIME-compliant header that can contain many character sets.
Optional s is the initial header value. If None, the initial header
value is not set. You can later append to the header with .append()
method calls. s may be a byte string or a Unicode string, but see the
.append() documentation for semantics.
Optional charset serves two purposes: it has the same meaning as the
charset argument to the .append() method. It also sets the default
character set for all subsequent .append() calls that omit the charset
argument. If charset is not provided in the constructor, the us-ascii
charset is used both as s's initial charset and as the default for
subsequent .append() calls.
The maximum line length can be specified explicitly via maxlinelen. For
splitting the first line to a shorter value (to account for the field
header which isn't included in s, e.g. `Subject') pass in the name of
the field in header_name. The default maxlinelen is 78 as recommended
continuation_ws must be RFC 2822 compliant folding whitespace (usually
either a space or a hard tab) which will be prepended to continuation
errors is passed through to the .append() call.
elif not isinstance(charset, Charset):
charset = Charset(charset)
self._continuation_ws = continuation_ws
self.append(s, charset, errors)
self._maxlinelen = maxlinelen
# Take the separating colon and space into account.
self._headerlen = len(header_name) + 2
"""Return the string value of the header."""
for string, charset in self._chunks:
# We must preserve spaces between encoded and non-encoded word
# boundaries, which means for us we need to add a space when we go
# from a charset to None/us-ascii, or from None/us-ascii to a
# charset. Only do this for the second and subsequent chunks.
# Don't add a space if the None/us-ascii string already has
# a space (trailing or leading depending on transition)
if nextcs == _charset.UNKNOWN8BIT:
original_bytes = string.encode('ascii', 'surrogateescape')
string = original_bytes.decode('ascii', 'replace')
hasspace = string and self._nonctext(string[0])
if lastcs not in (None, 'us-ascii'):
if nextcs in (None, 'us-ascii') and not hasspace:
elif nextcs not in (None, 'us-ascii') and not lastspace:
lastspace = string and self._nonctext(string[-1])
return EMPTYSTRING.join(uchunks)
# Rich comparison operators for equality only. BAW: does it make sense to
# have or explicitly disable <, <=, >, >= operators?
# other may be a Header or a string. Both are fine so coerce
# ourselves to a unicode (of the unencoded header value), swap the
# args and do another comparison.
return other == str(self)
def append(self, s, charset=None, errors='strict'):
"""Append a string to the MIME header.
Optional charset, if given, should be a Charset instance or the name
of a character set (which will be converted to a Charset instance). A
value of None (the default) means that the charset given in the
s may be a byte string or a Unicode string. If it is a byte string
(i.e. isinstance(s, str) is false), then charset is the encoding of
that byte string, and a UnicodeError will be raised if the string
cannot be decoded with that charset. If s is a Unicode string, then
charset is a hint specifying the character set of the characters in
the string. In either case, when producing an RFC 2822 compliant
header using RFC 2047 rules, the string will be encoded using the
output codec of the charset. If the string cannot be encoded to the
output codec, a UnicodeError will be raised.
Optional `errors' is passed as the errors argument to the decode
call if s is a byte string.
elif not isinstance(charset, Charset):
charset = Charset(charset)
if not isinstance(s, str):
input_charset = charset.input_codec or 'us-ascii'
if input_charset == _charset.UNKNOWN8BIT:
s = s.decode('us-ascii', 'surrogateescape')
s = s.decode(input_charset, errors)
# Ensure that the bytes we're storing can be decoded to the output
# character set, otherwise an early error is raised.
output_charset = charset.output_codec or 'us-ascii'
if output_charset != _charset.UNKNOWN8BIT:
s.encode(output_charset, errors)
except UnicodeEncodeError:
if output_charset!='us-ascii':
self._chunks.append((s, charset))
"""True if string s is not a ctext character of RFC822.
return s.isspace() or s in ('(', ')', '\\')
def encode(self, splitchars=';, \t', maxlinelen=None, linesep='\n'):
r"""Encode a message header into an RFC-compliant format.
There are many issues involved in converting a given string for use in
an email header. Only certain character sets are readable in most
email clients, and as header strings can only contain a subset of
7-bit ASCII, care must be taken to properly convert and encode (with
Base64 or quoted-printable) header strings. In addition, there is a
75-character length limit on any given encoded header field, so
line-wrapping must be performed, even with double-byte character sets.
Optional maxlinelen specifies the maximum length of each generated
line, exclusive of the linesep string. Individual lines may be longer
than maxlinelen if a folding point cannot be found. The first line
will be shorter by the length of the header name plus ": " if a header
name was specified at Header construction time. The default value for
maxlinelen is determined at header construction time.
Optional splitchars is a string containing characters which should be
given extra weight by the splitting algorithm during normal header
wrapping. This is in very rough support of RFC 2822's `higher level
syntactic breaks': split points preceded by a splitchar are preferred
during line splitting, with the characters preferred in the order in
which they appear in the string. Space and tab may be included in the
string to indicate whether preference should be given to one over the
other as a split point when other split chars do not appear in the line
being split. Splitchars does not affect RFC 2047 encoded lines.
Optional linesep is a string to be used to separate the lines of
the value. The default value is the most useful for typical
Python applications, but it can be set to \r\n to produce RFC-compliant
line separators when needed.
maxlinelen = self._maxlinelen
# A maxlinelen of 0 means don't wrap. For all practical purposes,
# choosing a huge number here accomplishes that and makes the
# _ValueFormatter algorithm much simpler.
formatter = _ValueFormatter(self._headerlen, maxlinelen,
self._continuation_ws, splitchars)
hasspace = lastspace = None
for string, charset in self._chunks:
hasspace = string and self._nonctext(string[0])
if lastcs not in (None, 'us-ascii'):
if not hasspace or charset not in (None, 'us-ascii'):
formatter.add_transition()
elif charset not in (None, 'us-ascii') and not lastspace:
formatter.add_transition()
lastspace = string and self._nonctext(string[-1])
lines = string.splitlines()
formatter.feed('', lines[0], charset)
formatter.feed('', '', charset)
if charset.header_encoding is not None:
formatter.feed(self._continuation_ws, ' ' + line.lstrip(),
fws = line[:len(line)-len(sline)]
formatter.feed(fws, sline, charset)
formatter.add_transition()
value = formatter._str(linesep)
if _embedded_header.search(value):
raise HeaderParseError("header value appears to contain "
"an embedded header: {!r}".format(value))
# Step 1: Normalize the chunks so that all runs of identical charsets
# get collapsed into a single unicode string.
for string, charset in self._chunks:
if charset == last_charset:
last_chunk.append(string)
if last_charset is not None:
chunks.append((SPACE.join(last_chunk), last_charset))
chunks.append((SPACE.join(last_chunk), last_charset))
def __init__(self, headerlen, maxlen, continuation_ws, splitchars):
self._continuation_ws = continuation_ws
self._continuation_ws_len = len(continuation_ws)
self._splitchars = splitchars
self._current_line = _Accumulator(headerlen)
return linesep.join(self._lines)
end_of_line = self._current_line.pop()
if end_of_line != (' ', ''):
self._current_line.push(*end_of_line)
if len(self._current_line) > 0:
if self._current_line.is_onlyws():
self._lines[-1] += str(self._current_line)
self._lines.append(str(self._current_line))
self._current_line.reset()
def add_transition(self):
self._current_line.push(' ', '')
def feed(self, fws, string, charset):
# If the charset has no header encoding (i.e. it is an ASCII encoding)
# then we must split the header at the "highest level syntactic break"
# possible. Note that we don't have a lot of smarts about field
# syntax; we just try to break on semi-colons, then commas, then
# whitespace. Eventually, this should be pluggable.
if charset.header_encoding is None:
self._ascii_split(fws, string, self._splitchars)
# Otherwise, we're doing either a Base64 or a quoted-printable
# encoding which means we don't need to split the line on syntactic
# breaks. We can basically just find enough characters to fit on the
# current line, minus the RFC 2047 chrome. What makes this trickier
# though is that we have to split at octet boundaries, not character
# boundaries but it's only safe to split at character boundaries so at
# best we can only get close.
encoded_lines = charset.header_encode_lines(string, self._maxlengths())
# The first element extends the current line, but if it's None then
# nothing more fit on the current line so start a new line.
first_line = encoded_lines.pop(0)
# There are no encoded lines, so we're done.
if first_line is not None:
self._append_chunk(fws, first_line)
last_line = encoded_lines.pop()
# There was only one line.
self._current_line.push(self._continuation_ws, last_line)
# Everything else are full lines in themselves.
for line in encoded_lines:
self._lines.append(self._continuation_ws + line)
# The first line's length.
yield self._maxlen - len(self._current_line)
yield self._maxlen - self._continuation_ws_len
def _ascii_split(self, fws, string, splitchars):
# The RFC 2822 header folding algorithm is simple in principle but
# complex in practice. Lines may be folded any place where "folding
# white space" appears by inserting a linesep character in front of the
# FWS. The complication is that not all spaces or tabs qualify as FWS,
# and we are also supposed to prefer to break at "higher level
# syntactic breaks". We can't do either of these without intimate
# knowledge of the structure of structured headers, which we don't have
# here. So the best we can do here is prefer to break at the specified
# splitchars, and hope that we don't choose any spaces or tabs that
# aren't legal FWS. (This is at least better than the old algorithm,
# where we would sometimes *introduce* FWS after a splitchar, or the
# algorithm before that, where we would turn all white space runs into
# single spaces or tabs.)
parts = re.split("(["+FWS+"]+)", fws+string)
It is recommended that you Edit text format, this type of Fix handles quite a lot in one request
