EmbeddedPkg: Extend NvVarStoreFormattedLib LIBRARY_CLASS

[mirror_edk2.git] / AppPkg / Applications / Python / Python-2.7.2 / Lib / email / header.py

# Copyright (C) 2002-2006 Python Software Foundation\r
# Author: Ben Gertzfield, Barry Warsaw\r
# Contact: email-sig@python.org\r
\r
"""Header encoding and decoding functionality."""\r
\r
__all__ = [\r
    'Header',\r
    'decode_header',\r
    'make_header',\r
    ]\r
\r
import re\r
import binascii\r
\r
import email.quoprimime\r
import email.base64mime\r
\r
from email.errors import HeaderParseError\r
from email.charset import Charset\r
\r
NL = '\n'\r
SPACE = ' '\r
USPACE = u' '\r
SPACE8 = ' ' * 8\r
UEMPTYSTRING = u''\r
\r
MAXLINELEN = 76\r
\r
USASCII = Charset('us-ascii')\r
UTF8 = Charset('utf-8')\r
\r
# Match encoded-word strings in the form =?charset?q?Hello_World?=\r
ecre = re.compile(r'''\r
  =\?                   # literal =?\r
  (?P<charset>[^?]*?)   # non-greedy up to the next ? is the charset\r
  \?                    # literal ?\r
  (?P<encoding>[qb])    # either a "q" or a "b", case insensitive\r
  \?                    # literal ?\r
  (?P<encoded>.*?)      # non-greedy up to the next ?= is the encoded string\r
  \?=                   # literal ?=\r
  (?=[ \t]|$)           # whitespace or the end of the string\r
  ''', re.VERBOSE | re.IGNORECASE | re.MULTILINE)\r
\r
# Field name regexp, including trailing colon, but not separating whitespace,\r
# according to RFC 2822.  Character range is from tilde to exclamation mark.\r
# For use with .match()\r
fcre = re.compile(r'[\041-\176]+:$')\r
\r
# Find a header embedded in a putative header value.  Used to check for\r
# header injection attack.\r
_embeded_header = re.compile(r'\n[^ \t]+:')\r
\r
\r
\f\r
# Helpers\r
_max_append = email.quoprimime._max_append\r
\r
\r
\f\r
def decode_header(header):\r
    """Decode a message header value without converting charset.\r
\r
    Returns a list of (decoded_string, charset) pairs containing each of the\r
    decoded parts of the header.  Charset is None for non-encoded parts of the\r
    header, otherwise a lower-case string containing the name of the character\r
    set specified in the encoded string.\r
\r
    An email.errors.HeaderParseError may be raised when certain decoding error\r
    occurs (e.g. a base64 decoding exception).\r
    """\r
    # If no encoding, just return the header\r
    header = str(header)\r
    if not ecre.search(header):\r
        return [(header, None)]\r
    decoded = []\r
    dec = ''\r
    for line in header.splitlines():\r
        # This line might not have an encoding in it\r
        if not ecre.search(line):\r
            decoded.append((line, None))\r
            continue\r
        parts = ecre.split(line)\r
        while parts:\r
            unenc = parts.pop(0).strip()\r
            if unenc:\r
                # Should we continue a long line?\r
                if decoded and decoded[-1][1] is None:\r
                    decoded[-1] = (decoded[-1][0] + SPACE + unenc, None)\r
                else:\r
                    decoded.append((unenc, None))\r
            if parts:\r
                charset, encoding = [s.lower() for s in parts[0:2]]\r
                encoded = parts[2]\r
                dec = None\r
                if encoding == 'q':\r
                    dec = email.quoprimime.header_decode(encoded)\r
                elif encoding == 'b':\r
                    paderr = len(encoded) % 4   # Postel's law: add missing padding\r
                    if paderr:\r
                        encoded += '==='[:4 - paderr]\r
                    try:\r
                        dec = email.base64mime.decode(encoded)\r
                    except binascii.Error:\r
                        # Turn this into a higher level exception.  BAW: Right\r
                        # now we throw the lower level exception away but\r
                        # when/if we get exception chaining, we'll preserve it.\r
                        raise HeaderParseError\r
                if dec is None:\r
                    dec = encoded\r
\r
                if decoded and decoded[-1][1] == charset:\r
                    decoded[-1] = (decoded[-1][0] + dec, decoded[-1][1])\r
                else:\r
                    decoded.append((dec, charset))\r
            del parts[0:3]\r
    return decoded\r
\r
\r
\f\r
def make_header(decoded_seq, maxlinelen=None, header_name=None,\r
                continuation_ws=' '):\r
    """Create a Header from a sequence of pairs as returned by decode_header()\r
\r
    decode_header() takes a header value string and returns a sequence of\r
    pairs of the format (decoded_string, charset) where charset is the string\r
    name of the character set.\r
\r
    This function takes one of those sequence of pairs and returns a Header\r
    instance.  Optional maxlinelen, header_name, and continuation_ws are as in\r
    the Header constructor.\r
    """\r
    h = Header(maxlinelen=maxlinelen, header_name=header_name,\r
               continuation_ws=continuation_ws)\r
    for s, charset in decoded_seq:\r
        # None means us-ascii but we can simply pass it on to h.append()\r
        if charset is not None and not isinstance(charset, Charset):\r
            charset = Charset(charset)\r
        h.append(s, charset)\r
    return h\r
\r
\r
\f\r
class Header:\r
    def __init__(self, s=None, charset=None,\r
                 maxlinelen=None, header_name=None,\r
                 continuation_ws=' ', errors='strict'):\r
        """Create a MIME-compliant header that can contain many character sets.\r
\r
        Optional s is the initial header value.  If None, the initial header\r
        value is not set.  You can later append to the header with .append()\r
        method calls.  s may be a byte string or a Unicode string, but see the\r
        .append() documentation for semantics.\r
\r
        Optional charset serves two purposes: it has the same meaning as the\r
        charset argument to the .append() method.  It also sets the default\r
        character set for all subsequent .append() calls that omit the charset\r
        argument.  If charset is not provided in the constructor, the us-ascii\r
        charset is used both as s's initial charset and as the default for\r
        subsequent .append() calls.\r
\r
        The maximum line length can be specified explicit via maxlinelen.  For\r
        splitting the first line to a shorter value (to account for the field\r
        header which isn't included in s, e.g. `Subject') pass in the name of\r
        the field in header_name.  The default maxlinelen is 76.\r
\r
        continuation_ws must be RFC 2822 compliant folding whitespace (usually\r
        either a space or a hard tab) which will be prepended to continuation\r
        lines.\r
\r
        errors is passed through to the .append() call.\r
        """\r
        if charset is None:\r
            charset = USASCII\r
        if not isinstance(charset, Charset):\r
            charset = Charset(charset)\r
        self._charset = charset\r
        self._continuation_ws = continuation_ws\r
        cws_expanded_len = len(continuation_ws.replace('\t', SPACE8))\r
        # BAW: I believe `chunks' and `maxlinelen' should be non-public.\r
        self._chunks = []\r
        if s is not None:\r
            self.append(s, charset, errors)\r
        if maxlinelen is None:\r
            maxlinelen = MAXLINELEN\r
        if header_name is None:\r
            # We don't know anything about the field header so the first line\r
            # is the same length as subsequent lines.\r
            self._firstlinelen = maxlinelen\r
        else:\r
            # The first line should be shorter to take into account the field\r
            # header.  Also subtract off 2 extra for the colon and space.\r
            self._firstlinelen = maxlinelen - len(header_name) - 2\r
        # Second and subsequent lines should subtract off the length in\r
        # columns of the continuation whitespace prefix.\r
        self._maxlinelen = maxlinelen - cws_expanded_len\r
\r
    def __str__(self):\r
        """A synonym for self.encode()."""\r
        return self.encode()\r
\r
    def __unicode__(self):\r
        """Helper for the built-in unicode function."""\r
        uchunks = []\r
        lastcs = None\r
        for s, charset in self._chunks:\r
            # We must preserve spaces between encoded and non-encoded word\r
            # boundaries, which means for us we need to add a space when we go\r
            # from a charset to None/us-ascii, or from None/us-ascii to a\r
            # charset.  Only do this for the second and subsequent chunks.\r
            nextcs = charset\r
            if uchunks:\r
                if lastcs not in (None, 'us-ascii'):\r
                    if nextcs in (None, 'us-ascii'):\r
                        uchunks.append(USPACE)\r
                        nextcs = None\r
                elif nextcs not in (None, 'us-ascii'):\r
                    uchunks.append(USPACE)\r
            lastcs = nextcs\r
            uchunks.append(unicode(s, str(charset)))\r
        return UEMPTYSTRING.join(uchunks)\r
\r
    # Rich comparison operators for equality only.  BAW: does it make sense to\r
    # have or explicitly disable <, <=, >, >= operators?\r
    def __eq__(self, other):\r
        # other may be a Header or a string.  Both are fine so coerce\r
        # ourselves to a string, swap the args and do another comparison.\r
        return other == self.encode()\r
\r
    def __ne__(self, other):\r
        return not self == other\r
\r
    def append(self, s, charset=None, errors='strict'):\r
        """Append a string to the MIME header.\r
\r
        Optional charset, if given, should be a Charset instance or the name\r
        of a character set (which will be converted to a Charset instance).  A\r
        value of None (the default) means that the charset given in the\r
        constructor is used.\r
\r
        s may be a byte string or a Unicode string.  If it is a byte string\r
        (i.e. isinstance(s, str) is true), then charset is the encoding of\r
        that byte string, and a UnicodeError will be raised if the string\r
        cannot be decoded with that charset.  If s is a Unicode string, then\r
        charset is a hint specifying the character set of the characters in\r
        the string.  In this case, when producing an RFC 2822 compliant header\r
        using RFC 2047 rules, the Unicode string will be encoded using the\r
        following charsets in order: us-ascii, the charset hint, utf-8.  The\r
        first character set not to provoke a UnicodeError is used.\r
\r
        Optional `errors' is passed as the third argument to any unicode() or\r
        ustr.encode() call.\r
        """\r
        if charset is None:\r
            charset = self._charset\r
        elif not isinstance(charset, Charset):\r
            charset = Charset(charset)\r
        # If the charset is our faux 8bit charset, leave the string unchanged\r
        if charset != '8bit':\r
            # We need to test that the string can be converted to unicode and\r
            # back to a byte string, given the input and output codecs of the\r
            # charset.\r
            if isinstance(s, str):\r
                # Possibly raise UnicodeError if the byte string can't be\r
                # converted to a unicode with the input codec of the charset.\r
                incodec = charset.input_codec or 'us-ascii'\r
                ustr = unicode(s, incodec, errors)\r
                # Now make sure that the unicode could be converted back to a\r
                # byte string with the output codec, which may be different\r
                # than the iput coded.  Still, use the original byte string.\r
                outcodec = charset.output_codec or 'us-ascii'\r
                ustr.encode(outcodec, errors)\r
            elif isinstance(s, unicode):\r
                # Now we have to be sure the unicode string can be converted\r
                # to a byte string with a reasonable output codec.  We want to\r
                # use the byte string in the chunk.\r
                for charset in USASCII, charset, UTF8:\r
                    try:\r
                        outcodec = charset.output_codec or 'us-ascii'\r
                        s = s.encode(outcodec, errors)\r
                        break\r
                    except UnicodeError:\r
                        pass\r
                else:\r
                    assert False, 'utf-8 conversion failed'\r
        self._chunks.append((s, charset))\r
\r
    def _split(self, s, charset, maxlinelen, splitchars):\r
        # Split up a header safely for use with encode_chunks.\r
        splittable = charset.to_splittable(s)\r
        encoded = charset.from_splittable(splittable, True)\r
        elen = charset.encoded_header_len(encoded)\r
        # If the line's encoded length first, just return it\r
        if elen <= maxlinelen:\r
            return [(encoded, charset)]\r
        # If we have undetermined raw 8bit characters sitting in a byte\r
        # string, we really don't know what the right thing to do is.  We\r
        # can't really split it because it might be multibyte data which we\r
        # could break if we split it between pairs.  The least harm seems to\r
        # be to not split the header at all, but that means they could go out\r
        # longer than maxlinelen.\r
        if charset == '8bit':\r
            return [(s, charset)]\r
        # BAW: I'm not sure what the right test here is.  What we're trying to\r
        # do is be faithful to RFC 2822's recommendation that ($2.2.3):\r
        #\r
        # "Note: Though structured field bodies are defined in such a way that\r
        #  folding can take place between many of the lexical tokens (and even\r
        #  within some of the lexical tokens), folding SHOULD be limited to\r
        #  placing the CRLF at higher-level syntactic breaks."\r
        #\r
        # For now, I can only imagine doing this when the charset is us-ascii,\r
        # although it's possible that other charsets may also benefit from the\r
        # higher-level syntactic breaks.\r
        elif charset == 'us-ascii':\r
            return self._split_ascii(s, charset, maxlinelen, splitchars)\r
        # BAW: should we use encoded?\r
        elif elen == len(s):\r
            # We can split on _maxlinelen boundaries because we know that the\r
            # encoding won't change the size of the string\r
            splitpnt = maxlinelen\r
            first = charset.from_splittable(splittable[:splitpnt], False)\r
            last = charset.from_splittable(splittable[splitpnt:], False)\r
        else:\r
            # Binary search for split point\r
            first, last = _binsplit(splittable, charset, maxlinelen)\r
        # first is of the proper length so just wrap it in the appropriate\r
        # chrome.  last must be recursively split.\r
        fsplittable = charset.to_splittable(first)\r
        fencoded = charset.from_splittable(fsplittable, True)\r
        chunk = [(fencoded, charset)]\r
        return chunk + self._split(last, charset, self._maxlinelen, splitchars)\r
\r
    def _split_ascii(self, s, charset, firstlen, splitchars):\r
        chunks = _split_ascii(s, firstlen, self._maxlinelen,\r
                              self._continuation_ws, splitchars)\r
        return zip(chunks, [charset]*len(chunks))\r
\r
    def _encode_chunks(self, newchunks, maxlinelen):\r
        # MIME-encode a header with many different charsets and/or encodings.\r
        #\r
        # Given a list of pairs (string, charset), return a MIME-encoded\r
        # string suitable for use in a header field.  Each pair may have\r
        # different charsets and/or encodings, and the resulting header will\r
        # accurately reflect each setting.\r
        #\r
        # Each encoding can be email.utils.QP (quoted-printable, for\r
        # ASCII-like character sets like iso-8859-1), email.utils.BASE64\r
        # (Base64, for non-ASCII like character sets like KOI8-R and\r
        # iso-2022-jp), or None (no encoding).\r
        #\r
        # Each pair will be represented on a separate line; the resulting\r
        # string will be in the format:\r
        #\r
        # =?charset1?q?Mar=EDa_Gonz=E1lez_Alonso?=\n\r
        #  =?charset2?b?SvxyZ2VuIEL2aW5n?="\r
        chunks = []\r
        for header, charset in newchunks:\r
            if not header:\r
                continue\r
            if charset is None or charset.header_encoding is None:\r
                s = header\r
            else:\r
                s = charset.header_encode(header)\r
            # Don't add more folding whitespace than necessary\r
            if chunks and chunks[-1].endswith(' '):\r
                extra = ''\r
            else:\r
                extra = ' '\r
            _max_append(chunks, s, maxlinelen, extra)\r
        joiner = NL + self._continuation_ws\r
        return joiner.join(chunks)\r
\r
    def encode(self, splitchars=';, '):\r
        """Encode a message header into an RFC-compliant format.\r
\r
        There are many issues involved in converting a given string for use in\r
        an email header.  Only certain character sets are readable in most\r
        email clients, and as header strings can only contain a subset of\r
        7-bit ASCII, care must be taken to properly convert and encode (with\r
        Base64 or quoted-printable) header strings.  In addition, there is a\r
        75-character length limit on any given encoded header field, so\r
        line-wrapping must be performed, even with double-byte character sets.\r
\r
        This method will do its best to convert the string to the correct\r
        character set used in email, and encode and line wrap it safely with\r
        the appropriate scheme for that character set.\r
\r
        If the given charset is not known or an error occurs during\r
        conversion, this function will return the header untouched.\r
\r
        Optional splitchars is a string containing characters to split long\r
        ASCII lines on, in rough support of RFC 2822's `highest level\r
        syntactic breaks'.  This doesn't affect RFC 2047 encoded lines.\r
        """\r
        newchunks = []\r
        maxlinelen = self._firstlinelen\r
        lastlen = 0\r
        for s, charset in self._chunks:\r
            # The first bit of the next chunk should be just long enough to\r
            # fill the next line.  Don't forget the space separating the\r
            # encoded words.\r
            targetlen = maxlinelen - lastlen - 1\r
            if targetlen < charset.encoded_header_len(''):\r
                # Stick it on the next line\r
                targetlen = maxlinelen\r
            newchunks += self._split(s, charset, targetlen, splitchars)\r
            lastchunk, lastcharset = newchunks[-1]\r
            lastlen = lastcharset.encoded_header_len(lastchunk)\r
        value = self._encode_chunks(newchunks, maxlinelen)\r
        if _embeded_header.search(value):\r
            raise HeaderParseError("header value appears to contain "\r
                "an embedded header: {!r}".format(value))\r
        return value\r
\r
\r
\f\r
def _split_ascii(s, firstlen, restlen, continuation_ws, splitchars):\r
    lines = []\r
    maxlen = firstlen\r
    for line in s.splitlines():\r
        # Ignore any leading whitespace (i.e. continuation whitespace) already\r
        # on the line, since we'll be adding our own.\r
        line = line.lstrip()\r
        if len(line) < maxlen:\r
            lines.append(line)\r
            maxlen = restlen\r
            continue\r
        # Attempt to split the line at the highest-level syntactic break\r
        # possible.  Note that we don't have a lot of smarts about field\r
        # syntax; we just try to break on semi-colons, then commas, then\r
        # whitespace.\r
        for ch in splitchars:\r
            if ch in line:\r
                break\r
        else:\r
            # There's nothing useful to split the line on, not even spaces, so\r
            # just append this line unchanged\r
            lines.append(line)\r
            maxlen = restlen\r
            continue\r
        # Now split the line on the character plus trailing whitespace\r
        cre = re.compile(r'%s\s*' % ch)\r
        if ch in ';,':\r
            eol = ch\r
        else:\r
            eol = ''\r
        joiner = eol + ' '\r
        joinlen = len(joiner)\r
        wslen = len(continuation_ws.replace('\t', SPACE8))\r
        this = []\r
        linelen = 0\r
        for part in cre.split(line):\r
            curlen = linelen + max(0, len(this)-1) * joinlen\r
            partlen = len(part)\r
            onfirstline = not lines\r
            # We don't want to split after the field name, if we're on the\r
            # first line and the field name is present in the header string.\r
            if ch == ' ' and onfirstline and \\r
                   len(this) == 1 and fcre.match(this[0]):\r
                this.append(part)\r
                linelen += partlen\r
            elif curlen + partlen > maxlen:\r
                if this:\r
                    lines.append(joiner.join(this) + eol)\r
                # If this part is longer than maxlen and we aren't already\r
                # splitting on whitespace, try to recursively split this line\r
                # on whitespace.\r
                if partlen > maxlen and ch != ' ':\r
                    subl = _split_ascii(part, maxlen, restlen,\r
                                        continuation_ws, ' ')\r
                    lines.extend(subl[:-1])\r
                    this = [subl[-1]]\r
                else:\r
                    this = [part]\r
                linelen = wslen + len(this[-1])\r
                maxlen = restlen\r
            else:\r
                this.append(part)\r
                linelen += partlen\r
        # Put any left over parts on a line by themselves\r
        if this:\r
            lines.append(joiner.join(this))\r
    return lines\r
\r
\r
\f\r
def _binsplit(splittable, charset, maxlinelen):\r
    i = 0\r
    j = len(splittable)\r
    while i < j:\r
        # Invariants:\r
        # 1. splittable[:k] fits for all k <= i (note that we *assume*,\r
        #    at the start, that splittable[:0] fits).\r
        # 2. splittable[:k] does not fit for any k > j (at the start,\r
        #    this means we shouldn't look at any k > len(splittable)).\r
        # 3. We don't know about splittable[:k] for k in i+1..j.\r
        # 4. We want to set i to the largest k that fits, with i <= k <= j.\r
        #\r
        m = (i+j+1) >> 1  # ceiling((i+j)/2); i < m <= j\r
        chunk = charset.from_splittable(splittable[:m], True)\r
        chunklen = charset.encoded_header_len(chunk)\r
        if chunklen <= maxlinelen:\r
            # m is acceptable, so is a new lower bound.\r
            i = m\r
        else:\r
            # m is not acceptable, so final i must be < m.\r
            j = m - 1\r
    # i == j.  Invariant #1 implies that splittable[:i] fits, and\r
    # invariant #2 implies that splittable[:i+1] does not fit, so i\r
    # is what we're looking for.\r
    first = charset.from_splittable(splittable[:i], False)\r
    last  = charset.from_splittable(splittable[i:], False)\r
    return first, last\r