initial WIP of a tentative preview_url endpoint - incomplete, untested, experimental, etc. just putting it here for safekeeping for now

This commit is contained in:
Matthew Hodgson 2016-01-24 18:47:27 -05:00
parent f92fe15897
commit 7dd0c1730a
5 changed files with 327 additions and 1 deletions

74
docs/url_previews.rst Normal file
View File

@ -0,0 +1,74 @@
URL Previews
============
Design notes on a URL previewing service for Matrix:
Options are:
1. Have an AS which listens for URLs, downloads them, and inserts an event that describes their metadata.
* Pros:
* Decouples the implementation entirely from Synapse.
* Uses existing Matrix events & content repo to store the metadata.
* Cons:
* Which AS should provide this service for a room, and why should you trust it?
* Doesn't work well with E2E; you'd have to cut the AS into every room
* the AS would end up subscribing to every room anyway.
2. Have a generic preview API (nothing to do with Matrix) that provides a previewing service:
* Pros:
* Simple and flexible; can be used by any clients at any point
* Cons:
* If each HS provides one of these independently, all the HSes in a room may needlessly DoS the target URI
* We need somewhere to store the URL metadata rather than just using Matrix itself
* We can't piggyback on matrix to distribute the metadata between HSes.
3. Make the synapse of the sending user responsible for spidering the URL and inserting an event asynchronously which describes the metadata.
* Pros:
* Works transparently for all clients
* Piggy-backs nicely on using Matrix for distributing the metadata.
* No confusion as to which AS
* Cons:
* Doesn't work with E2E
* We might want to decouple the implementation of the spider from the HS, given spider behaviour can be quite complicated and evolve much more rapidly than the HS. It's more like a bot than a core part of the server.
4. Make the sending client use the preview API and insert the event itself when successful.
* Pros:
* Works well with E2E
* No custom server functionality
* Lets the client customise the preview that they send (like on FB)
* Cons:
* Entirely specific to the sending client, whereas it'd be nice if /any/ URL was correctly previewed if clients support it.
5. Have the option of specifying a shared (centralised) previewing service used by a room, to avoid all the different HSes in the room DoSing the target.
Best solution is probably a combination of both 2 and 4.
* Sending clients do their best to create and send a preview at the point of sending the message, perhaps delaying the message until the preview is computed? (This also lets the user validate the preview before sending)
* Receiving clients have the option of going and creating their own preview if one doesn't arrive soon enough (or if the original sender didn't create one)
This is a bit magical though in that the preview could come from two entirely different sources - the sending HS or your local one. However, this can always be exposed to users: "Generate your own URL previews if none are available?"
This is tantamount also to senders calculating their own thumbnails for sending in advance of the main content - we are trusting the sender not to lie about the content in the thumbnail. Whereas currently thumbnails are calculated by the receiving homeserver to avoid this attack.
However, this kind of phishing attack does exist whether we let senders pick their thumbnails or not, in that a malicious sender can send normal text messages around the attachment claiming it to be legitimate. We could rely on (future) reputation/abuse management to punish users who phish (be it with bogus metadata or bogus descriptions). Bogus metadata is particularly bad though, especially if it's avoidable.
As a first cut, let's do #2 and have the receiver hit the API to calculate its own previews (as it does currently for image thumbnails). We can then extend/optimise this to option 4 as a special extra if needed.
API
---
GET /_matrix/media/r0/previewUrl?url=http://wherever.com
200 OK
{
"og:type" : "article"
"og:url" : "https://twitter.com/matrixdotorg/status/684074366691356672"
"og:title" : "Matrix on Twitter"
"og:image" : "https://pbs.twimg.com/profile_images/500400952029888512/yI0qtFi7_400x400.png"
"og:description" : "“Synapse 0.12 is out! Lots of polishing, performance & bugfixes: /sync API, /r0 prefix, fulltext search, 3PID invites https://t.co/5alhXLLEGP”"
"og:site_name" : "Twitter"
}
* Downloads the URL
* If HTML, just stores it in RAM and parses it for OG meta tags
* Download any media OG meta tags to the media repo, and refer to them in the OG via mxc:// URIs.
* If a media filetype we know we can thumbnail: store it on disk, and hand it to the thumbnailer. Generate OG meta tags from the thumbnailer contents.
* Otherwise, don't bother downloading further.

View File

@ -53,6 +53,7 @@ class ContentRepositoryConfig(Config):
def read_config(self, config):
self.max_upload_size = self.parse_size(config["max_upload_size"])
self.max_image_pixels = self.parse_size(config["max_image_pixels"])
self.max_spider_size = self.parse_size(config["max_spider_size"])
self.media_store_path = self.ensure_directory(config["media_store_path"])
self.uploads_path = self.ensure_directory(config["uploads_path"])
self.dynamic_thumbnails = config["dynamic_thumbnails"]
@ -73,6 +74,9 @@ class ContentRepositoryConfig(Config):
# The largest allowed upload size in bytes
max_upload_size: "10M"
# The largest allowed URL preview spidering size in bytes
max_spider_size: "10M"
# Maximum number of pixels that will be thumbnailed
max_image_pixels: "32M"
@ -80,7 +84,7 @@ class ContentRepositoryConfig(Config):
# the resolution requested by the client. If true then whenever
# a new resolution is requested by the client the server will
# generate a new thumbnail. If false the server will pick a thumbnail
# from a precalcualted list.
# from a precalculated list.
dynamic_thumbnails: false
# List of thumbnail to precalculate when an image is uploaded.

View File

@ -238,6 +238,87 @@ class SimpleHttpClient(object):
else:
raise CodeMessageException(response.code, body)
# XXX: FIXME: This is horribly copy-pasted from matrixfederationclient.
# The two should be factored out.
@defer.inlineCallbacks
def get_file(self, url, output_stream, args={}, max_size=None):
"""GETs a file from a given URL
Args:
url (str): The URL to GET
output_stream (file): File to write the response body to.
Returns:
A (int,dict) tuple of the file length and a dict of the response
headers.
"""
def body_callback(method, url_bytes, headers_dict):
self.sign_request(destination, method, url_bytes, headers_dict)
return None
response = yield self.request(
"GET",
url.encode("ascii"),
headers=Headers({
b"User-Agent": [self.user_agent],
})
)
headers = dict(response.headers.getAllRawHeaders())
if headers['Content-Length'] > max_size:
logger.warn("Requested URL is too large > %r bytes" % (self.max_size,))
# XXX: do we want to explicitly drop the connection here somehow? if so, how?
raise # what should we be raising here?
# TODO: if our Content-Type is HTML or something, just read the first
# N bytes into RAM rather than saving it all to disk only to read it
# straight back in again
try:
length = yield preserve_context_over_fn(
_readBodyToFile,
response, output_stream, max_size
)
except:
logger.exception("Failed to download body")
raise
defer.returnValue((length, headers))
# XXX: FIXME: This is horribly copy-pasted from matrixfederationclient.
# The two should be factored out.
class _ReadBodyToFileProtocol(protocol.Protocol):
def __init__(self, stream, deferred, max_size):
self.stream = stream
self.deferred = deferred
self.length = 0
self.max_size = max_size
def dataReceived(self, data):
self.stream.write(data)
self.length += len(data)
if self.max_size is not None and self.length >= self.max_size:
logger.warn("Requested URL is too large > %r bytes" % (self.max_size,))
self.deferred = defer.Deferred()
self.transport.loseConnection()
def connectionLost(self, reason):
if reason.check(ResponseDone):
self.deferred.callback(self.length)
else:
self.deferred.errback(reason)
# XXX: FIXME: This is horribly copy-pasted from matrixfederationclient.
# The two should be factored out.
def _readBodyToFile(response, stream, max_size):
d = defer.Deferred()
response.deliverBody(_ReadBodyToFileProtocol(stream, d, max_size))
return d
class CaptchaServerHttpClient(SimpleHttpClient):
"""

View File

@ -17,6 +17,7 @@ from .upload_resource import UploadResource
from .download_resource import DownloadResource
from .thumbnail_resource import ThumbnailResource
from .identicon_resource import IdenticonResource
from .preview_url_resource import PreviewUrlResource
from .filepath import MediaFilePaths
from twisted.web.resource import Resource
@ -78,3 +79,5 @@ class MediaRepositoryResource(Resource):
self.putChild("download", DownloadResource(hs, filepaths))
self.putChild("thumbnail", ThumbnailResource(hs, filepaths))
self.putChild("identicon", IdenticonResource())
self.putChild("preview_url", PreviewUrlResource(hs, filepaths))

View File

@ -0,0 +1,164 @@
# Copyright 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.
from twisted.web.resource import Resource
from lxml import html
from synapse.http.client import SimpleHttpClient
from synapse.http.server import respond_with_json_bytes
from simplejson import json
import logging
logger = logging.getLogger(__name__)
class PreviewUrlResource(Resource):
isLeaf = True
def __init__(self, hs, filepaths):
Resource.__init__(self)
self.client = SimpleHttpClient(hs)
self.filepaths = filepaths
self.max_spider_size = hs.config.max_spider_size
self.server_name = hs.hostname
self.clock = hs.get_clock()
def render_GET(self, request):
self._async_render_GET(request)
return NOT_DONE_YET
@request_handler
@defer.inlineCallbacks
def _async_render_GET(self, request):
url = request.args.get("url")
try:
# TODO: keep track of whether there's an ongoing request for this preview
# and block and return their details if there is one.
media_info = self._download_url(url)
except:
os.remove(fname)
raise
if self._is_media(media_type):
dims = yield self._generate_local_thumbnails(
media_info.filesystem_id, media_info
)
og = {
"og:description" : media_info.download_name,
"og:image" : "mxc://%s/%s" % (self.server_name, media_info.filesystem_id),
"og:image:type" : media_info.media_type,
"og:image:width" : dims.width,
"og:image:height" : dims.height,
}
# define our OG response for this media
elif self._is_html(media_type):
tree = html.parse(media_info.filename)
# suck it up into lxml and define our OG response.
# if we see any URLs in the OG response, then spider them
# (although the client could choose to do this by asking for previews of those URLs to avoid DoSing the server)
# "og:type" : "article"
# "og:url" : "https://twitter.com/matrixdotorg/status/684074366691356672"
# "og:title" : "Matrix on Twitter"
# "og:image" : "https://pbs.twimg.com/profile_images/500400952029888512/yI0qtFi7_400x400.png"
# "og:description" : "“Synapse 0.12 is out! Lots of polishing, performance & bugfixes: /sync API, /r0 prefix, fulltext search, 3PID invites https://t.co/5alhXLLEGP”"
# "og:site_name" : "Twitter"
og = {}
for tag in tree.xpath("//*/meta[starts-with(@property, 'og:')]"):
og[tag.attrib['property']] = tag.attrib['content']
# TODO: store our OG details in a cache (and expire them when stale)
# TODO: delete the content to stop diskfilling, as we only ever cared about its OG
respond_with_json_bytes(request, 200, json.dumps(og), send_cors=True)
def _download_url(url):
requester = yield self.auth.get_user_by_req(request)
# XXX: horrible duplication with base_resource's _download_remote_file()
file_id = random_string(24)
fname = self.filepaths.local_media_filepath(file_id)
self._makedirs(fname)
try:
with open(fname, "wb") as f:
length, headers = yield self.client.get_file(
url, output_stream=f, max_size=self.max_spider_size,
)
media_type = headers["Content-Type"][0]
time_now_ms = self.clock.time_msec()
content_disposition = headers.get("Content-Disposition", None)
if content_disposition:
_, params = cgi.parse_header(content_disposition[0],)
download_name = None
# First check if there is a valid UTF-8 filename
download_name_utf8 = params.get("filename*", None)
if download_name_utf8:
if download_name_utf8.lower().startswith("utf-8''"):
download_name = download_name_utf8[7:]
# If there isn't check for an ascii name.
if not download_name:
download_name_ascii = params.get("filename", None)
if download_name_ascii and is_ascii(download_name_ascii):
download_name = download_name_ascii
if download_name:
download_name = urlparse.unquote(download_name)
try:
download_name = download_name.decode("utf-8")
except UnicodeDecodeError:
download_name = None
else:
download_name = None
yield self.store.store_local_media(
media_id=fname,
media_type=media_type,
time_now_ms=self.clock.time_msec(),
upload_name=download_name,
media_length=length,
user_id=requester.user,
)
except:
os.remove(fname)
raise
return {
"media_type": media_type,
"media_length": length,
"download_name": download_name,
"created_ts": time_now_ms,
"filesystem_id": file_id,
"filename": fname,
}
def _is_media(content_type):
if content_type.lower().startswith("image/"):
return True
def _is_html(content_type):
content_type = content_type.lower()
if content_type == "text/html" or
content_type.startswith("application/xhtml"):
return True