summaryrefslogtreecommitdiff
path: root/cloudinit/user_data.py
blob: df069ff846f91cf4bda0c827e54e2aabbb3a5b59 (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
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
# vi: ts=4 expandtab
#
#    Copyright (C) 2012 Canonical Ltd.
#    Copyright (C) 2012 Hewlett-Packard Development Company, L.P.
#    Copyright (C) 2012 Yahoo! Inc.
#
#    Author: Scott Moser <scott.moser@canonical.com>
#    Author: Juerg Haefliger <juerg.haefliger@hp.com>
#    Author: Joshua Harlow <harlowja@yahoo-inc.com>
#
#    This program is free software: you can redistribute it and/or modify
#    it under the terms of the GNU General Public License version 3, as
#    published by the Free Software Foundation.
#
#    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/>.

import os

import email
from email.mime.base import MIMEBase
from email.mime.multipart import MIMEMultipart
from email.mime.text import MIMEText

from cloudinit import handlers
from cloudinit import log as logging
from cloudinit import util

LOG = logging.getLogger(__name__)

# Constants copied in from the handler module
NOT_MULTIPART_TYPE = handlers.NOT_MULTIPART_TYPE
PART_FN_TPL = handlers.PART_FN_TPL
OCTET_TYPE = handlers.OCTET_TYPE

# Saves typing errors
CONTENT_TYPE = 'Content-Type'

# Various special content types that cause special actions
TYPE_NEEDED = ["text/plain", "text/x-not-multipart"]
INCLUDE_TYPES = ['text/x-include-url', 'text/x-include-once-url']
ARCHIVE_TYPES = ["text/cloud-config-archive"]
UNDEF_TYPE = "text/plain"
ARCHIVE_UNDEF_TYPE = "text/cloud-config"

# Msg header used to track attachments
ATTACHMENT_FIELD = 'Number-Attachments'

# Only the following content types can have there launch index examined
# in there payload, evey other content type can still provide a header
EXAMINE_FOR_LAUNCH_INDEX = ["text/cloud-config"]


class UserDataProcessor(object):
    def __init__(self, paths):
        self.paths = paths
        self.ssl_details = util.fetch_ssl_details(paths)

    def process(self, blob):
        accumulating_msg = MIMEMultipart()
        self._process_msg(convert_string(blob), accumulating_msg)
        return accumulating_msg

    def _process_msg(self, base_msg, append_msg):
        for part in base_msg.walk():
            if is_skippable(part):
                continue

            ctype = None
            ctype_orig = part.get_content_type()
            payload = part.get_payload(decode=True)

            if not ctype_orig:
                ctype_orig = UNDEF_TYPE

            if ctype_orig in TYPE_NEEDED:
                ctype = handlers.type_from_starts_with(payload)

            if ctype is None:
                ctype = ctype_orig

            if ctype != ctype_orig:
                if CONTENT_TYPE in part:
                    part.replace_header(CONTENT_TYPE, ctype)
                else:
                    part[CONTENT_TYPE] = ctype

            if ctype in INCLUDE_TYPES:
                self._do_include(payload, append_msg)
                continue

            if ctype in ARCHIVE_TYPES:
                self._explode_archive(payload, append_msg)
                continue

            # Should this be happening, shouldn't
            # the part header be modified and not the base?
            if CONTENT_TYPE in base_msg:
                base_msg.replace_header(CONTENT_TYPE, ctype)
            else:
                base_msg[CONTENT_TYPE] = ctype

            self._attach_part(append_msg, part)

    def _attach_launch_index(self, msg):
        header_idx = msg.get('Launch-Index', None)
        payload_idx = None
        if msg.get_content_type() in EXAMINE_FOR_LAUNCH_INDEX:
            try:
                # See if it has a launch-index field
                # that might affect the final header
                payload = util.load_yaml(msg.get_payload(decode=True))
                if payload:
                    payload_idx = payload.get('launch-index')
            except:
                pass
        # Header overrides contents, for now (?) or the other way around?
        if header_idx is not None:
            payload_idx = header_idx
        # Nothing found in payload, use header (if anything there)
        if payload_idx is None:
            payload_idx = header_idx
        if payload_idx is not None:
            try:
                msg.add_header('Launch-Index', str(int(payload_idx)))
            except (ValueError, TypeError):
                pass

    def _get_include_once_filename(self, entry):
        entry_fn = util.hash_blob(entry, 'md5', 64)
        return os.path.join(self.paths.get_ipath_cur('data'),
                            'urlcache', entry_fn)

    def _process_before_attach(self, msg, attached_id):
        if not msg.get_filename():
            msg.add_header('Content-Disposition',
                           'attachment', filename=PART_FN_TPL % (attached_id))
        self._attach_launch_index(msg)

    def _do_include(self, content, append_msg):
        # Include a list of urls, one per line
        # also support '#include <url here>'
        # or #include-once '<url here>'
        include_once_on = False
        for line in content.splitlines():
            lc_line = line.lower()
            if lc_line.startswith("#include-once"):
                line = line[len("#include-once"):].lstrip()
                # Every following include will now
                # not be refetched.... but will be
                # re-read from a local urlcache (if it worked)
                include_once_on = True
            elif lc_line.startswith("#include"):
                line = line[len("#include"):].lstrip()
                # Disable the include once if it was on
                # if it wasn't, then this has no effect.
                include_once_on = False
            if line.startswith("#"):
                continue
            include_url = line.strip()
            if not include_url:
                continue

            include_once_fn = None
            content = None
            if include_once_on:
                include_once_fn = self._get_include_once_filename(include_url)
            if include_once_on and os.path.isfile(include_once_fn):
                content = util.load_file(include_once_fn)
            else:
                resp = util.read_file_or_url(include_url,
                                             ssl_details=self.ssl_details)
                if include_once_on and resp.ok():
                    util.write_file(include_once_fn, str(resp), mode=0600)
                if resp.ok():
                    content = str(resp)
                else:
                    LOG.warn(("Fetching from %s resulted in"
                              " a invalid http code of %s"),
                             include_url, resp.code)

            if content is not None:
                new_msg = convert_string(content)
                self._process_msg(new_msg, append_msg)

    def _explode_archive(self, archive, append_msg):
        entries = util.load_yaml(archive, default=[], allowed=(list, set))
        for ent in entries:
            # ent can be one of:
            #  dict { 'filename' : 'value', 'content' :
            #       'value', 'type' : 'value' }
            #    filename and type not be present
            # or
            #  scalar(payload)
            if isinstance(ent, (str, basestring)):
                ent = {'content': ent}
            if not isinstance(ent, (dict)):
                # TODO(harlowja) raise?
                continue

            content = ent.get('content', '')
            mtype = ent.get('type')
            if not mtype:
                mtype = handlers.type_from_starts_with(content,
                                                       ARCHIVE_UNDEF_TYPE)

            maintype, subtype = mtype.split('/', 1)
            if maintype == "text":
                msg = MIMEText(content, _subtype=subtype)
            else:
                msg = MIMEBase(maintype, subtype)
                msg.set_payload(content)

            if 'filename' in ent:
                msg.add_header('Content-Disposition',
                               'attachment', filename=ent['filename'])
            if 'launch-index' in ent:
                msg.add_header('Launch-Index', str(ent['launch-index']))

            for header in list(ent.keys()):
                if header in ('content', 'filename', 'type', 'launch-index'):
                    continue
                msg.add_header(header, ent[header])

            self._attach_part(append_msg, msg)

    def _multi_part_count(self, outer_msg, new_count=None):
        """
        Return the number of attachments to this MIMEMultipart by looking
        at its 'Number-Attachments' header.
        """
        if ATTACHMENT_FIELD not in outer_msg:
            outer_msg[ATTACHMENT_FIELD] = '0'

        if new_count is not None:
            outer_msg.replace_header(ATTACHMENT_FIELD, str(new_count))

        fetched_count = 0
        try:
            fetched_count = int(outer_msg.get(ATTACHMENT_FIELD))
        except (ValueError, TypeError):
            outer_msg.replace_header(ATTACHMENT_FIELD, str(fetched_count))
        return fetched_count

    def _attach_part(self, outer_msg, part):
        """
        Attach a message to an outer message. outermsg must be a MIMEMultipart.
        Modifies a header in the outer message to keep track of number of attachments.
        """
        part_count = self._multi_part_count(outer_msg)
        self._process_before_attach(part, part_count + 1)
        outer_msg.attach(part)
        self._multi_part_count(outer_msg, part_count + 1)


def is_skippable(part):
    # multipart/* are just containers
    part_maintype = part.get_content_maintype() or ''
    if part_maintype.lower() == 'multipart':
        return True
    return False


# Coverts a raw string into a mime message
def convert_string(raw_data, headers=None):
    if not raw_data:
        raw_data = ''
    if not headers:
        headers = {}
    data = util.decomp_gzip(raw_data)
    if "mime-version:" in data[0:4096].lower():
        msg = email.message_from_string(data)
        for (key, val) in headers.iteritems():
            if key in msg:
                msg.replace_header(key, val)
            else:
                msg[key] = val
    else:
        mtype = headers.get(CONTENT_TYPE, NOT_MULTIPART_TYPE)
        maintype, subtype = mtype.split("/", 1)
        msg = MIMEBase(maintype, subtype, *headers)
        msg.set_payload(data)
    return msg