summaryrefslogtreecommitdiff
path: root/src/leap/mail/imap/messageparts.py
blob: a47ea1d2af9320d64792155fabcb51dfecba749b (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
# -*- coding: utf-8 -*-
# messageparts.py
# Copyright (C) 2014 LEAP
#
# This program is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program.  If not, see <http://www.gnu.org/licenses/>.
"""
MessagePart implementation. Used from LeapMessage.
"""
import logging
import re
import StringIO

from enum import Enum
from zope.interface import implements
from twisted.mail import imap4

from leap.common.decorators import memoized_method
from leap.common.mail import get_email_charset
from leap.mail.imap.fields import fields
from leap.mail.utils import first

MessagePartType = Enum("hdoc", "fdoc", "cdoc")


logger = logging.getLogger(__name__)


CHARSET_PATTERN = r"""charset=([\w-]+)"""
CHARSET_RE = re.compile(CHARSET_PATTERN, re.IGNORECASE)


class MessagePart(object):
    """
    IMessagePart implementor.
    It takes a subpart message and is able to find
    the inner parts.

    Excusatio non petita: see the interface documentation.
    """

    implements(imap4.IMessagePart)

    def __init__(self, soledad, part_map):
        """
        Initializes the MessagePart.

        :param part_map: a dictionary containing the parts map for this
                         message
        :type part_map: dict
        """
        # TODO
        # It would be good to pass the uid/mailbox also
        # for references while debugging.

        # We have a problem on bulk moves, and is
        # that when the fetch on the new mailbox is done
        # the parts maybe are not complete.
        # So we should be able to fail with empty
        # docs until we solve that. The ideal would be
        # to gather the results of the deferred operations
        # to signal the operation is complete.
        #leap_assert(part_map, "part map dict cannot be null")
        self._soledad = soledad
        self._pmap = part_map

    def getSize(self):
        """
        Return the total size, in octets, of this message part.

        :return: size of the message, in octets
        :rtype: int
        """
        if not self._pmap:
            return 0
        size = self._pmap.get('size', None)
        if not size:
            logger.error("Message part cannot find size in the partmap")
        return size

    def getBodyFile(self):
        """
        Retrieve a file object containing only the body of this message.

        :return: file-like object opened for reading
        :rtype: StringIO
        """
        fd = StringIO.StringIO()
        if self._pmap:
            multi = self._pmap.get('multi')
            if not multi:
                phash = self._pmap.get("phash", None)
            else:
                pmap = self._pmap.get('part_map')
                first_part = pmap.get('1', None)
                if first_part:
                    phash = first_part['phash']

            if not phash:
                logger.warning("Could not find phash for this subpart!")
                payload = str("")
            else:
                payload = self._get_payload_from_document(phash)

        else:
            logger.warning("Message with no part_map!")
            payload = str("")

        if payload:
            content_type = self._get_ctype_from_document(phash)
            charset = first(CHARSET_RE.findall(content_type))
            logger.debug("Got charset from header: %s" % (charset,))
            if not charset:
                charset = self._get_charset(payload)
            try:
                payload = payload.encode(charset)
            except UnicodeError as exc:
                logger.error("Unicode error {0}".format(exc))
                payload = payload.encode(charset, 'replace')

        fd.write(payload)
        fd.seek(0)
        return fd

    # TODO cache the phash retrieval
    def _get_payload_from_document(self, phash):
        """
        Gets the message payload from the content document.

        :param phash: the payload hash to retrieve by.
        :type phash: basestring
        """
        cdocs = self._soledad.get_from_index(
            fields.TYPE_P_HASH_IDX,
            fields.TYPE_CONTENT_VAL, str(phash))

        cdoc = first(cdocs)
        if not cdoc:
            logger.warning(
                "Could not find the content doc "
                "for phash %s" % (phash,))
        payload = cdoc.content.get(fields.RAW_KEY, "")
        return payload

    # TODO cache the pahash retrieval
    def _get_ctype_from_document(self, phash):
        """
        Gets the content-type from the content document.

        :param phash: the payload hash to retrieve by.
        :type phash: basestring
        """
        cdocs = self._soledad.get_from_index(
            fields.TYPE_P_HASH_IDX,
            fields.TYPE_CONTENT_VAL, str(phash))

        cdoc = first(cdocs)
        if not cdoc:
            logger.warning(
                "Could not find the content doc "
                "for phash %s" % (phash,))
        ctype = cdoc.content.get('ctype', "")
        return ctype

    @memoized_method
    def _get_charset(self, stuff):
        # TODO put in a common class with LeapMessage
        """
        Gets (guesses?) the charset of a payload.

        :param stuff: the stuff to guess about.
        :type stuff: basestring
        :returns: charset
        """
        # XXX existential doubt 2. shouldn't we make the scope
        # of the decorator somewhat more persistent?
        # ah! yes! and put memory bounds.
        return get_email_charset(unicode(stuff))

    def getHeaders(self, negate, *names):
        """
        Retrieve a group of message headers.

        :param names: The names of the headers to retrieve or omit.
        :type names: tuple of str

        :param negate: If True, indicates that the headers listed in names
                       should be omitted from the return value, rather
                       than included.
        :type negate: bool

        :return: A mapping of header field names to header field values
        :rtype: dict
        """
        if not self._pmap:
            logger.warning("No pmap in Subpart!")
            return {}
        headers = dict(self._pmap.get("headers", []))

        # twisted imap server expects *some* headers to be lowercase
        # We could use a CaseInsensitiveDict here...
        headers = dict(
            (str(key), str(value)) if key.lower() != "content-type"
            else (str(key.lower()), str(value))
            for (key, value) in headers.items())

        names = map(lambda s: s.upper(), names)
        if negate:
            cond = lambda key: key.upper() not in names
        else:
            cond = lambda key: key.upper() in names

        # unpack and filter original dict by negate-condition
        filter_by_cond = [
            map(str, (key, val)) for
            key, val in headers.items()
            if cond(key)]
        filtered = dict(filter_by_cond)
        return filtered

    def isMultipart(self):
        """
        Return True if this message is multipart.
        """
        if not self._pmap:
            logger.warning("Could not get part map!")
            return False
        multi = self._pmap.get("multi", False)
        return multi

    def getSubPart(self, part):
        """
        Retrieve a MIME submessage

        :type part: C{int}
        :param part: The number of the part to retrieve, indexed from 0.
        :raise IndexError: Raised if the specified part does not exist.
        :raise TypeError: Raised if this message is not multipart.
        :rtype: Any object implementing C{IMessagePart}.
        :return: The specified sub-part.
        """
        if not self.isMultipart():
            raise TypeError
        sub_pmap = self._pmap.get("part_map", {})
        try:
            part_map = sub_pmap[str(part + 1)]
        except KeyError:
            logger.debug("getSubpart for %s: KeyError" % (part,))
            raise IndexError

        # XXX check for validity
        return MessagePart(self._soledad, part_map)