2016-04-19 06:31:43 -04:00
|
|
|
# -*- coding: utf-8 -*-
|
|
|
|
# Copyright 2014-2016 OpenMarket Ltd
|
|
|
|
#
|
|
|
|
# Licensed under the Apache License, Version 2.0 (the "License");
|
|
|
|
# you may not use this file except in compliance with the License.
|
|
|
|
# You may obtain a copy of the License at
|
|
|
|
#
|
|
|
|
# http://www.apache.org/licenses/LICENSE-2.0
|
|
|
|
#
|
|
|
|
# Unless required by applicable law or agreed to in writing, software
|
|
|
|
# distributed under the License is distributed on an "AS IS" BASIS,
|
|
|
|
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
|
|
# See the License for the specific language governing permissions and
|
|
|
|
# limitations under the License.
|
|
|
|
|
2018-07-09 02:09:20 -04:00
|
|
|
import logging
|
|
|
|
import os
|
|
|
|
|
2018-11-15 16:55:58 -05:00
|
|
|
from six import PY3
|
2018-09-12 06:41:31 -04:00
|
|
|
from six.moves import urllib
|
2016-04-19 06:31:43 -04:00
|
|
|
|
|
|
|
from twisted.internet import defer
|
|
|
|
from twisted.protocols.basic import FileSender
|
|
|
|
|
2018-07-09 02:09:20 -04:00
|
|
|
from synapse.api.errors import Codes, SynapseError, cs_error
|
|
|
|
from synapse.http.server import finish_request, respond_with_json
|
|
|
|
from synapse.util import logcontext
|
2016-04-19 06:31:43 -04:00
|
|
|
from synapse.util.stringutils import is_ascii
|
|
|
|
|
|
|
|
logger = logging.getLogger(__name__)
|
|
|
|
|
|
|
|
|
|
|
|
def parse_media_id(request):
|
|
|
|
try:
|
|
|
|
# This allows users to append e.g. /test.png to the URL. Useful for
|
|
|
|
# clients that parse the URL to see content type.
|
|
|
|
server_name, media_id = request.postpath[:2]
|
2018-09-12 06:41:31 -04:00
|
|
|
|
|
|
|
if isinstance(server_name, bytes):
|
|
|
|
server_name = server_name.decode('utf-8')
|
|
|
|
media_id = media_id.decode('utf8')
|
|
|
|
|
2016-04-19 06:31:43 -04:00
|
|
|
file_name = None
|
|
|
|
if len(request.postpath) > 2:
|
|
|
|
try:
|
2018-09-12 06:41:31 -04:00
|
|
|
file_name = urllib.parse.unquote(request.postpath[-1].decode("utf-8"))
|
2016-04-19 06:31:43 -04:00
|
|
|
except UnicodeDecodeError:
|
|
|
|
pass
|
|
|
|
return server_name, media_id, file_name
|
2017-10-23 10:52:32 -04:00
|
|
|
except Exception:
|
2016-04-19 06:31:43 -04:00
|
|
|
raise SynapseError(
|
2018-11-15 16:55:58 -05:00
|
|
|
404, "Invalid media id token %r" % (request.postpath,), Codes.UNKNOWN
|
2016-04-19 06:31:43 -04:00
|
|
|
)
|
|
|
|
|
|
|
|
|
|
|
|
def respond_404(request):
|
|
|
|
respond_with_json(
|
2018-11-15 16:55:58 -05:00
|
|
|
request,
|
|
|
|
404,
|
|
|
|
cs_error("Not found %r" % (request.postpath,), code=Codes.NOT_FOUND),
|
|
|
|
send_cors=True,
|
2016-04-19 06:31:43 -04:00
|
|
|
)
|
|
|
|
|
|
|
|
|
|
|
|
@defer.inlineCallbacks
|
2018-11-15 16:55:58 -05:00
|
|
|
def respond_with_file(request, media_type, file_path, file_size=None, upload_name=None):
|
2016-04-19 06:31:43 -04:00
|
|
|
logger.debug("Responding with %r", file_path)
|
|
|
|
|
|
|
|
if os.path.isfile(file_path):
|
|
|
|
if file_size is None:
|
|
|
|
stat = os.stat(file_path)
|
|
|
|
file_size = stat.st_size
|
|
|
|
|
2018-01-05 10:09:43 -05:00
|
|
|
add_file_headers(request, media_type, file_size, upload_name)
|
2016-04-19 06:31:43 -04:00
|
|
|
|
|
|
|
with open(file_path, "rb") as f:
|
2017-10-23 09:34:27 -04:00
|
|
|
yield logcontext.make_deferred_yieldable(
|
|
|
|
FileSender().beginFileTransfer(f, request)
|
|
|
|
)
|
2016-04-19 06:31:43 -04:00
|
|
|
|
|
|
|
finish_request(request)
|
|
|
|
else:
|
|
|
|
respond_404(request)
|
2018-01-05 10:09:43 -05:00
|
|
|
|
|
|
|
|
|
|
|
def add_file_headers(request, media_type, file_size, upload_name):
|
|
|
|
"""Adds the correct response headers in preparation for responding with the
|
|
|
|
media.
|
|
|
|
|
|
|
|
Args:
|
|
|
|
request (twisted.web.http.Request)
|
|
|
|
media_type (str): The media/content type.
|
|
|
|
file_size (int): Size in bytes of the media, if known.
|
|
|
|
upload_name (str): The name of the requested file, if any.
|
|
|
|
"""
|
2018-11-15 16:55:58 -05:00
|
|
|
|
2018-09-12 06:41:31 -04:00
|
|
|
def _quote(x):
|
|
|
|
return urllib.parse.quote(x.encode("utf-8"))
|
|
|
|
|
2018-01-05 10:09:43 -05:00
|
|
|
request.setHeader(b"Content-Type", media_type.encode("UTF-8"))
|
|
|
|
if upload_name:
|
|
|
|
if is_ascii(upload_name):
|
2018-11-15 16:55:58 -05:00
|
|
|
disposition = "inline; filename=%s" % (_quote(upload_name),)
|
2018-01-05 10:09:43 -05:00
|
|
|
else:
|
2018-11-15 16:55:58 -05:00
|
|
|
disposition = "inline; filename*=utf-8''%s" % (_quote(upload_name),)
|
2018-09-12 06:41:31 -04:00
|
|
|
|
2018-11-15 16:55:58 -05:00
|
|
|
request.setHeader(b"Content-Disposition", disposition.encode('ascii'))
|
2018-01-05 10:09:43 -05:00
|
|
|
|
|
|
|
# cache for at least a day.
|
|
|
|
# XXX: we might want to turn this off for data we don't want to
|
|
|
|
# recommend caching as it's sensitive or private - or at least
|
|
|
|
# select private. don't bother setting Expires as all our
|
|
|
|
# clients are smart enough to be happy with Cache-Control
|
2018-11-15 16:55:58 -05:00
|
|
|
request.setHeader(b"Cache-Control", b"public,max-age=86400,s-maxage=86400")
|
|
|
|
request.setHeader(b"Content-Length", b"%d" % (file_size,))
|
2018-01-08 11:58:09 -05:00
|
|
|
|
|
|
|
|
|
|
|
@defer.inlineCallbacks
|
|
|
|
def respond_with_responder(request, responder, media_type, file_size, upload_name=None):
|
|
|
|
"""Responds to the request with given responder. If responder is None then
|
|
|
|
returns 404.
|
|
|
|
|
|
|
|
Args:
|
|
|
|
request (twisted.web.http.Request)
|
2018-01-12 06:15:31 -05:00
|
|
|
responder (Responder|None)
|
2018-01-08 11:58:09 -05:00
|
|
|
media_type (str): The media/content type.
|
2018-01-12 10:02:46 -05:00
|
|
|
file_size (int|None): Size in bytes of the media. If not known it should be None
|
|
|
|
upload_name (str|None): The name of the requested file, if any.
|
2018-01-08 11:58:09 -05:00
|
|
|
"""
|
|
|
|
if not responder:
|
|
|
|
respond_404(request)
|
|
|
|
return
|
|
|
|
|
2018-05-02 08:29:16 -04:00
|
|
|
logger.debug("Responding to media request with responder %s")
|
2018-01-08 11:58:09 -05:00
|
|
|
add_file_headers(request, media_type, file_size, upload_name)
|
2019-02-25 14:17:22 -05:00
|
|
|
try:
|
|
|
|
with responder:
|
|
|
|
yield responder.write_to_consumer(request)
|
|
|
|
except Exception as e:
|
|
|
|
# The majority of the time this will be due to the client having gone
|
|
|
|
# away. Unfortunately, Twisted simply throws a generic exception at us
|
|
|
|
# in that case.
|
|
|
|
logger.warning("Failed to write to consumer: %s %s", type(e), e)
|
|
|
|
|
2018-01-08 11:58:09 -05:00
|
|
|
finish_request(request)
|
|
|
|
|
|
|
|
|
|
|
|
class Responder(object):
|
|
|
|
"""Represents a response that can be streamed to the requester.
|
|
|
|
|
2018-01-12 08:32:03 -05:00
|
|
|
Responder is a context manager which *must* be used, so that any resources
|
|
|
|
held can be cleaned up.
|
2018-01-08 11:58:09 -05:00
|
|
|
"""
|
2018-11-15 16:55:58 -05:00
|
|
|
|
2018-01-08 11:58:09 -05:00
|
|
|
def write_to_consumer(self, consumer):
|
|
|
|
"""Stream response into consumer
|
|
|
|
|
|
|
|
Args:
|
|
|
|
consumer (IConsumer)
|
|
|
|
|
|
|
|
Returns:
|
|
|
|
Deferred: Resolves once the response has finished being written
|
|
|
|
"""
|
|
|
|
pass
|
|
|
|
|
2018-01-12 08:32:03 -05:00
|
|
|
def __enter__(self):
|
|
|
|
pass
|
|
|
|
|
|
|
|
def __exit__(self, exc_type, exc_val, exc_tb):
|
2018-01-08 11:58:09 -05:00
|
|
|
pass
|
|
|
|
|
|
|
|
|
|
|
|
class FileInfo(object):
|
|
|
|
"""Details about a requested/uploaded file.
|
|
|
|
|
|
|
|
Attributes:
|
|
|
|
server_name (str): The server name where the media originated from,
|
|
|
|
or None if local.
|
|
|
|
file_id (str): The local ID of the file. For local files this is the
|
|
|
|
same as the media_id
|
|
|
|
url_cache (bool): If the file is for the url preview cache
|
|
|
|
thumbnail (bool): Whether the file is a thumbnail or not.
|
|
|
|
thumbnail_width (int)
|
|
|
|
thumbnail_height (int)
|
2018-01-12 06:15:31 -05:00
|
|
|
thumbnail_method (str)
|
|
|
|
thumbnail_type (str): Content type of thumbnail, e.g. image/png
|
2018-01-08 11:58:09 -05:00
|
|
|
"""
|
2018-11-15 16:55:58 -05:00
|
|
|
|
|
|
|
def __init__(
|
|
|
|
self,
|
|
|
|
server_name,
|
|
|
|
file_id,
|
|
|
|
url_cache=False,
|
|
|
|
thumbnail=False,
|
|
|
|
thumbnail_width=None,
|
|
|
|
thumbnail_height=None,
|
|
|
|
thumbnail_method=None,
|
|
|
|
thumbnail_type=None,
|
|
|
|
):
|
2018-01-08 11:58:09 -05:00
|
|
|
self.server_name = server_name
|
|
|
|
self.file_id = file_id
|
|
|
|
self.url_cache = url_cache
|
|
|
|
self.thumbnail = thumbnail
|
|
|
|
self.thumbnail_width = thumbnail_width
|
|
|
|
self.thumbnail_height = thumbnail_height
|
|
|
|
self.thumbnail_method = thumbnail_method
|
|
|
|
self.thumbnail_type = thumbnail_type
|
2018-11-15 16:55:58 -05:00
|
|
|
|
|
|
|
|
|
|
|
def get_filename_from_headers(headers):
|
|
|
|
"""
|
|
|
|
Get the filename of the downloaded file by inspecting the
|
|
|
|
Content-Disposition HTTP header.
|
|
|
|
|
|
|
|
Args:
|
|
|
|
headers (twisted.web.http_headers.Headers): The HTTP
|
|
|
|
request headers.
|
|
|
|
|
|
|
|
Returns:
|
|
|
|
A Unicode string of the filename, or None.
|
|
|
|
"""
|
|
|
|
content_disposition = headers.get(b"Content-Disposition", [b''])
|
|
|
|
|
|
|
|
# No header, bail out.
|
|
|
|
if not content_disposition[0]:
|
|
|
|
return
|
|
|
|
|
|
|
|
# dict of unicode: bytes, corresponding to the key value sections of the
|
|
|
|
# Content-Disposition header.
|
|
|
|
params = {}
|
|
|
|
parts = content_disposition[0].split(b";")
|
|
|
|
for i in parts:
|
|
|
|
# Split into key-value pairs, if able
|
|
|
|
# We don't care about things like `inline`, so throw it out
|
|
|
|
if b"=" not in i:
|
|
|
|
continue
|
|
|
|
|
|
|
|
key, value = i.strip().split(b"=")
|
|
|
|
params[key.decode('ascii')] = value
|
|
|
|
|
|
|
|
upload_name = None
|
|
|
|
|
|
|
|
# First check if there is a valid UTF-8 filename
|
|
|
|
upload_name_utf8 = params.get("filename*", None)
|
|
|
|
if upload_name_utf8:
|
|
|
|
if upload_name_utf8.lower().startswith(b"utf-8''"):
|
|
|
|
upload_name_utf8 = upload_name_utf8[7:]
|
|
|
|
# We have a filename*= section. This MUST be ASCII, and any UTF-8
|
|
|
|
# bytes are %-quoted.
|
|
|
|
if PY3:
|
|
|
|
try:
|
|
|
|
# Once it is decoded, we can then unquote the %-encoded
|
|
|
|
# parts strictly into a unicode string.
|
|
|
|
upload_name = urllib.parse.unquote(
|
|
|
|
upload_name_utf8.decode('ascii'), errors="strict"
|
|
|
|
)
|
|
|
|
except UnicodeDecodeError:
|
|
|
|
# Incorrect UTF-8.
|
|
|
|
pass
|
|
|
|
else:
|
|
|
|
# On Python 2, we first unquote the %-encoded parts and then
|
|
|
|
# decode it strictly using UTF-8.
|
|
|
|
try:
|
|
|
|
upload_name = urllib.parse.unquote(upload_name_utf8).decode('utf8')
|
|
|
|
except UnicodeDecodeError:
|
|
|
|
pass
|
|
|
|
|
|
|
|
# If there isn't check for an ascii name.
|
|
|
|
if not upload_name:
|
|
|
|
upload_name_ascii = params.get("filename", None)
|
|
|
|
if upload_name_ascii and is_ascii(upload_name_ascii):
|
|
|
|
# Make sure there's no %-quoted bytes. If there is, reject it as
|
|
|
|
# non-valid ASCII.
|
|
|
|
if b"%" not in upload_name_ascii:
|
|
|
|
upload_name = upload_name_ascii.decode('ascii')
|
|
|
|
|
|
|
|
# This may be None here, indicating we did not find a matching name.
|
|
|
|
return upload_name
|