diff options
| -rw-r--r-- | docs/url_previews.rst | 74 | ||||
| -rw-r--r-- | synapse/config/repository.py | 6 | ||||
| -rw-r--r-- | synapse/http/client.py | 122 | ||||
| -rw-r--r-- | synapse/python_dependencies.py | 1 | ||||
| -rw-r--r-- | synapse/rest/media/v1/base_resource.py | 1 | ||||
| -rw-r--r-- | synapse/rest/media/v1/media_repository.py | 2 | ||||
| -rw-r--r-- | synapse/rest/media/v1/preview_url_resource.py | 404 | ||||
| -rw-r--r-- | synapse/rest/media/v1/thumbnail_resource.py | 22 | ||||
| -rw-r--r-- | synapse/storage/media_repository.py | 55 | ||||
| -rw-r--r-- | synapse/storage/schema/delta/30/local_media_repository_url_cache.sql | 27 |
10 files changed, 709 insertions, 5 deletions
diff --git a/docs/url_previews.rst b/docs/url_previews.rst new file mode 100644 index 0000000000..634d9d907f --- /dev/null +++ b/docs/url_previews.rst @@ -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/preview_url?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 &amp; 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. diff --git a/synapse/config/repository.py b/synapse/config/repository.py index 2e96c09013..f4ab705701 100644 --- a/synapse/config/repository.py +++ b/synapse/config/repository.py @@ -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. diff --git a/synapse/http/client.py b/synapse/http/client.py index cbd45b2bbe..442b4bb73d 100644 --- a/synapse/http/client.py +++ b/synapse/http/client.py @@ -15,17 +15,22 @@ from OpenSSL import SSL from OpenSSL.SSL import VERIFY_NONE -from synapse.api.errors import CodeMessageException +from synapse.api.errors import ( + CodeMessageException, SynapseError, Codes, +) from synapse.util.logcontext import preserve_context_over_fn import synapse.metrics from canonicaljson import encode_canonical_json -from twisted.internet import defer, reactor, ssl +from twisted.internet import defer, reactor, ssl, protocol from twisted.web.client import ( - Agent, readBody, FileBodyProducer, PartialDownloadError, + BrowserLikeRedirectAgent, ContentDecoderAgent, GzipDecoder, Agent, + readBody, FileBodyProducer, PartialDownloadError, ) +from twisted.web.http import PotentialDataLoss from twisted.web.http_headers import Headers +from twisted.web._newclient import ResponseDone from StringIO import StringIO @@ -238,6 +243,96 @@ 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, 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,string,int) tuple of the file length, dict of the response + headers, absolute URI of the response and HTTP response code. + """ + + response = yield self.request( + "GET", + url.encode("ascii"), + headers=Headers({ + b"User-Agent": [self.user_agent], + }) + ) + + headers = dict(response.headers.getAllRawHeaders()) + + if 'Content-Length' in headers and 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? + + if response.code > 299: + logger.warn("Got %d when downloading %s" % (response.code, url)) + raise + + # 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, response.request.absoluteURI, response.code)) + + +# 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: + self.deferred.errback(SynapseError( + 502, + "Requested file is too large > %r bytes" % (self.max_size,), + Codes.TOO_LARGE, + )) + self.deferred = defer.Deferred() + self.transport.loseConnection() + + def connectionLost(self, reason): + if reason.check(ResponseDone): + self.deferred.callback(self.length) + elif reason.check(PotentialDataLoss): + # stolen from https://github.com/twisted/treq/pull/49/files + # http://twistedmatrix.com/trac/ticket/4840 + 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): """ @@ -269,6 +364,27 @@ class CaptchaServerHttpClient(SimpleHttpClient): defer.returnValue(e.response) +class SpiderHttpClient(SimpleHttpClient): + """ + Separate HTTP client for spidering arbitrary URLs. + Special in that it follows retries and has a UA that looks + like a browser. + + used by the preview_url endpoint in the content repo. + """ + def __init__(self, hs): + SimpleHttpClient.__init__(self, hs) + # clobber the base class's agent and UA: + self.agent = ContentDecoderAgent(BrowserLikeRedirectAgent(Agent( + reactor, + connectTimeout=15, + contextFactory=hs.get_http_client_context_factory() + )), [('gzip', GzipDecoder)]) + # We could look like Chrome: + # self.user_agent = ("Mozilla/5.0 (%s) (KHTML, like Gecko) + # Chrome Safari" % hs.version_string) + + def encode_urlencode_args(args): return {k: encode_urlencode_arg(v) for k, v in args.items()} diff --git a/synapse/python_dependencies.py b/synapse/python_dependencies.py index cf1414b4db..86b8331760 100644 --- a/synapse/python_dependencies.py +++ b/synapse/python_dependencies.py @@ -36,6 +36,7 @@ REQUIREMENTS = { "blist": ["blist"], "pysaml2>=3.0.0,<4.0.0": ["saml2>=3.0.0,<4.0.0"], "pymacaroons-pynacl": ["pymacaroons"], + "lxml>=3.6.0": ["lxml"], "pyjwt": ["jwt"], } CONDITIONAL_REQUIREMENTS = { diff --git a/synapse/rest/media/v1/base_resource.py b/synapse/rest/media/v1/base_resource.py index 58ef91c0b8..2b1938dc8e 100644 --- a/synapse/rest/media/v1/base_resource.py +++ b/synapse/rest/media/v1/base_resource.py @@ -72,6 +72,7 @@ class BaseMediaResource(Resource): self.store = hs.get_datastore() self.max_upload_size = hs.config.max_upload_size self.max_image_pixels = hs.config.max_image_pixels + self.max_spider_size = hs.config.max_spider_size self.filepaths = filepaths self.version_string = hs.version_string self.downloads = {} diff --git a/synapse/rest/media/v1/media_repository.py b/synapse/rest/media/v1/media_repository.py index 7dfb027dd1..11f672aeab 100644 --- a/synapse/rest/media/v1/media_repository.py +++ b/synapse/rest/media/v1/media_repository.py @@ -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 +79,4 @@ 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)) diff --git a/synapse/rest/media/v1/preview_url_resource.py b/synapse/rest/media/v1/preview_url_resource.py new file mode 100644 index 0000000000..f5ec32d8f2 --- /dev/null +++ b/synapse/rest/media/v1/preview_url_resource.py @@ -0,0 +1,404 @@ +# -*- coding: utf-8 -*- +# 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 .base_resource import BaseMediaResource + +from twisted.web.server import NOT_DONE_YET +from twisted.internet import defer +from lxml import html +from urlparse import urlparse, urlunparse + +from synapse.api.errors import Codes +from synapse.util.stringutils import random_string +from synapse.util.caches.expiringcache import ExpiringCache +from synapse.http.client import SpiderHttpClient +from synapse.http.server import ( + request_handler, respond_with_json, respond_with_json_bytes +) +from synapse.util.async import ObservableDeferred +from synapse.util.stringutils import is_ascii + +import os +import re +import cgi +import ujson as json + +import logging +logger = logging.getLogger(__name__) + + +class PreviewUrlResource(BaseMediaResource): + isLeaf = True + + def __init__(self, hs, filepaths): + BaseMediaResource.__init__(self, hs, filepaths) + self.client = SpiderHttpClient(hs) + + # simple memory cache mapping urls to OG metadata + self.cache = ExpiringCache( + cache_name="url_previews", + clock=self.clock, + # don't spider URLs more often than once an hour + expiry_ms=60 * 60 * 1000, + ) + self.cache.start() + + self.downloads = {} + + def render_GET(self, request): + self._async_render_GET(request) + return NOT_DONE_YET + + @request_handler + @defer.inlineCallbacks + def _async_render_GET(self, request): + + try: + # XXX: if get_user_by_req fails, what should we do in an async render? + requester = yield self.auth.get_user_by_req(request) + url = request.args.get("url")[0] + if "ts" in request.args: + ts = int(request.args.get("ts")[0]) + else: + ts = self.clock.time_msec() + + # first check the memory cache - good to handle all the clients on this + # HS thundering away to preview the same URL at the same time. + try: + og = self.cache[url] + respond_with_json_bytes(request, 200, json.dumps(og), send_cors=True) + return + except: + pass + + # then check the URL cache in the DB (which will also provide us with + # historical previews, if we have any) + cache_result = yield self.store.get_url_cache(url, ts) + if ( + cache_result and + cache_result["download_ts"] + cache_result["expires"] > ts and + cache_result["response_code"] / 100 == 2 + ): + respond_with_json_bytes( + request, 200, cache_result["og"].encode('utf-8'), + send_cors=True + ) + return + + # Ensure only one download for a given URL is active at a time + download = self.downloads.get(url) + if download is None: + download = self._download_url(url, requester.user) + download = ObservableDeferred( + download, + consumeErrors=True + ) + self.downloads[url] = download + + @download.addBoth + def callback(media_info): + del self.downloads[url] + return media_info + media_info = yield download.observe() + + # FIXME: we should probably update our cache now anyway, so that + # even if the OG calculation raises, we don't keep hammering on the + # remote server. For now, leave it uncached to aid debugging OG + # calculation problems + + logger.debug("got media_info of '%s'" % media_info) + + if self._is_media(media_info['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'], + "matrix:image:size": media_info['media_length'], + } + + if dims: + og["og:image:width"] = dims['width'] + og["og:image:height"] = dims['height'] + else: + logger.warn("Couldn't get dims for %s" % url) + + # define our OG response for this media + elif self._is_html(media_info['media_type']): + # TODO: somehow stop a big HTML tree from exploding synapse's RAM + + try: + tree = html.parse(media_info['filename']) + og = yield self._calc_og(tree, media_info, requester) + except UnicodeDecodeError: + # XXX: evil evil bodge + file = open(media_info['filename']) + body = file.read() + file.close() + tree = html.fromstring(body.decode('utf-8', 'ignore')) + og = yield self._calc_og(tree, media_info, requester) + + else: + logger.warn("Failed to find any OG data in %s", url) + og = {} + + logger.debug("Calculated OG for %s as %s" % (url, og)) + + # store OG in ephemeral in-memory cache + self.cache[url] = og + + # store OG in history-aware DB cache + yield self.store.store_url_cache( + url, + media_info["response_code"], + media_info["etag"], + media_info["expires"], + json.dumps(og), + media_info["filesystem_id"], + media_info["created_ts"], + ) + + respond_with_json_bytes(request, 200, json.dumps(og), send_cors=True) + except: + # XXX: if we don't explicitly respond here, the request never returns. + # isn't this what server.py's wrapper is meant to be doing for us? + respond_with_json( + request, + 500, + { + "error": "Internal server error", + "errcode": Codes.UNKNOWN, + }, + send_cors=True + ) + raise + + @defer.inlineCallbacks + def _calc_og(self, tree, media_info, requester): + # suck our tree into lxml and define our OG response. + + # if we see any image 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" : "video", + # "og:url" : "https://www.youtube.com/watch?v=LXDBoHyjmtw", + # "og:site_name" : "YouTube", + # "og:video:type" : "application/x-shockwave-flash", + # "og:description" : "Fun stuff happening here", + # "og:title" : "RemoteJam - Matrix team hack for Disrupt Europe Hackathon", + # "og:image" : "https://i.ytimg.com/vi/LXDBoHyjmtw/maxresdefault.jpg", + # "og:video:url" : "http://www.youtube.com/v/LXDBoHyjmtw?version=3&autohide=1", + # "og:video:width" : "1280" + # "og:video:height" : "720", + # "og:video:secure_url": "https://www.youtube.com/v/LXDBoHyjmtw?version=3", + + og = {} + for tag in tree.xpath("//*/meta[starts-with(@property, 'og:')]"): + og[tag.attrib['property']] = tag.attrib['content'] + + # TODO: grab article: meta tags too, e.g.: + + # "article:publisher" : "https://www.facebook.com/thethudonline" /> + # "article:author" content="https://www.facebook.com/thethudonline" /> + # "article:tag" content="baby" /> + # "article:section" content="Breaking News" /> + # "article:published_time" content="2016-03-31T19:58:24+00:00" /> + # "article:modified_time" content="2016-04-01T18:31:53+00:00" /> + + if 'og:title' not in og: + # do some basic spidering of the HTML + title = tree.xpath("(//title)[1] | (//h1)[1] | (//h2)[1] | (//h3)[1]") + og['og:title'] = title[0].text.strip() if title else None + + if 'og:image' not in og: + # TODO: extract a favicon failing all else + meta_image = tree.xpath( + "//*/meta[translate(@itemprop, 'IMAGE', 'image')='image']/@content" + ) + if meta_image: + og['og:image'] = self._rebase_url(meta_image[0], media_info['uri']) + else: + # TODO: consider inlined CSS styles as well as width & height attribs + images = tree.xpath("//img[@src][number(@width)>10][number(@height)>10]") + images = sorted(images, key=lambda i: ( + -1 * int(i.attrib['width']) * int(i.attrib['height']) + )) + if not images: + images = tree.xpath("//img[@src]") + if images: + og['og:image'] = images[0].attrib['src'] + + # pre-cache the image for posterity + # FIXME: it might be cleaner to use the same flow as the main /preview_url request + # itself and benefit from the same caching etc. But for now we just rely on the + # caching on the master request to speed things up. + if 'og:image' in og and og['og:image']: + image_info = yield self._download_url( + self._rebase_url(og['og:image'], media_info['uri']), requester.user + ) + + if self._is_media(image_info['media_type']): + # TODO: make sure we don't choke on white-on-transparent images + dims = yield self._generate_local_thumbnails( + image_info['filesystem_id'], image_info + ) + if dims: + og["og:image:width"] = dims['width'] + og["og:image:height"] = dims['height'] + else: + logger.warn("Couldn't get dims for %s" % og["og:image"]) + + og["og:image"] = "mxc://%s/%s" % ( + self.server_name, image_info['filesystem_id'] + ) + og["og:image:type"] = image_info['media_type'] + og["matrix:image:size"] = image_info['media_length'] + else: + del og["og:image"] + + if 'og:description' not in og: + meta_description = tree.xpath( + "//*/meta" + "[translate(@name, 'DESCRIPTION', 'description')='description']" + "/@content") + if meta_description: + og['og:description'] = meta_description[0] + else: + # text_nodes = tree.xpath("//h1/text() | //h2/text() | //h3/text() | " + # "//p/text() | //div/text() | //span/text() | //a/text()") + text_nodes = tree.xpath("//text()[not(ancestor::header | ancestor::nav | " + "ancestor::aside | ancestor::footer | " + "ancestor::script | ancestor::style)]" + + "[ancestor::body]") + text = '' + for text_node in text_nodes: + if len(text) < 500: + text += text_node + ' ' + else: + break + text = re.sub(r'[\t ]+', ' ', text) + text = re.sub(r'[\t \r\n]*[\r\n]+', '\n', text) + text = text.strip()[:500] + og['og:description'] = text if text else None + + # TODO: delete the url downloads to stop diskfilling, + # as we only ever cared about its OG + defer.returnValue(og) + + def _rebase_url(self, url, base): + base = list(urlparse(base)) + url = list(urlparse(url)) + if not url[0]: # fix up schema + url[0] = base[0] or "http" + if not url[1]: # fix up hostname + url[1] = base[1] + if not url[2].startswith('/'): + url[2] = re.sub(r'/[^/]+$', '/', base[2]) + url[2] + return urlunparse(url) + + @defer.inlineCallbacks + def _download_url(self, url, user): + # TODO: we should probably honour robots.txt... except in practice + # we're most likely being explicitly triggered by a human rather than a + # bot, so are we really a robot? + + # 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: + logger.debug("Trying to get url '%s'" % url) + length, headers, uri, code = yield self.client.get_file( + url, output_stream=f, max_size=self.max_spider_size, + ) + # FIXME: pass through 404s and other error messages nicely + + 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=file_id, + media_type=media_type, + time_now_ms=self.clock.time_msec(), + upload_name=download_name, + media_length=length, + user_id=user, + ) + + except: + os.remove(fname) + raise + + defer.returnValue({ + "media_type": media_type, + "media_length": length, + "download_name": download_name, + "created_ts": time_now_ms, + "filesystem_id": file_id, + "filename": fname, + "uri": uri, + "response_code": code, + # FIXME: we should calculate a proper expiration based on the + # Cache-Control and Expire headers. But for now, assume 1 hour. + "expires": 60 * 60 * 1000, + "etag": headers["ETag"][0] if "ETag" in headers else None, + }) + + def _is_media(self, content_type): + if content_type.lower().startswith("image/"): + return True + + def _is_html(self, content_type): + content_type = content_type.lower() + if ( + content_type.startswith("text/html") or + content_type.startswith("application/xhtml") + ): + return True diff --git a/synapse/rest/media/v1/thumbnail_resource.py b/synapse/rest/media/v1/thumbnail_resource.py index ab52499785..513b445688 100644 --- a/synapse/rest/media/v1/thumbnail_resource.py +++ b/synapse/rest/media/v1/thumbnail_resource.py @@ -72,6 +72,11 @@ class ThumbnailResource(BaseMediaResource): self._respond_404(request) return + if media_info["media_type"] == "image/svg+xml": + file_path = self.filepaths.local_media_filepath(media_id) + yield self._respond_with_file(request, media_info["media_type"], file_path) + return + thumbnail_infos = yield self.store.get_local_media_thumbnails(media_id) if thumbnail_infos: @@ -103,6 +108,11 @@ class ThumbnailResource(BaseMediaResource): self._respond_404(request) return + if media_info["media_type"] == "image/svg+xml": + file_path = self.filepaths.local_media_filepath(media_id) + yield self._respond_with_file(request, media_info["media_type"], file_path) + return + thumbnail_infos = yield self.store.get_local_media_thumbnails(media_id) for info in thumbnail_infos: t_w = info["thumbnail_width"] == desired_width @@ -138,6 +148,11 @@ class ThumbnailResource(BaseMediaResource): desired_method, desired_type): media_info = yield self._get_remote_media(server_name, media_id) + if media_info["media_type"] == "image/svg+xml": + file_path = self.filepaths.remote_media_filepath(server_name, media_id) + yield self._respond_with_file(request, media_info["media_type"], file_path) + return + thumbnail_infos = yield self.store.get_remote_media_thumbnails( server_name, media_id, ) @@ -181,6 +196,11 @@ class ThumbnailResource(BaseMediaResource): # We should proxy the thumbnail from the remote server instead. media_info = yield self._get_remote_media(server_name, media_id) + if media_info["media_type"] == "image/svg+xml": + file_path = self.filepaths.remote_media_filepath(server_name, media_id) + yield self._respond_with_file(request, media_info["media_type"], file_path) + return + thumbnail_infos = yield self.store.get_remote_media_thumbnails( server_name, media_id, ) @@ -208,6 +228,8 @@ class ThumbnailResource(BaseMediaResource): @defer.inlineCallbacks def _respond_default_thumbnail(self, request, media_info, width, height, method, m_type): + # XXX: how is this meant to work? store.get_default_thumbnails + # appears to always return [] so won't this always 404? media_type = media_info["media_type"] top_level_type = media_type.split("/")[0] sub_type = media_type.split("/")[-1].split(";")[0] diff --git a/synapse/storage/media_repository.py b/synapse/storage/media_repository.py index 9d3ba32478..c9dd20eed8 100644 --- a/synapse/storage/media_repository.py +++ b/synapse/storage/media_repository.py @@ -25,7 +25,7 @@ class MediaRepositoryStore(SQLBaseStore): def get_local_media(self, media_id): """Get the metadata for a local piece of media Returns: - None if the meia_id doesn't exist. + None if the media_id doesn't exist. """ return self._simple_select_one( "local_media_repository", @@ -50,6 +50,59 @@ class MediaRepositoryStore(SQLBaseStore): desc="store_local_media", ) + def get_url_cache(self, url, ts): + """Get the media_id and ts for a cached URL as of the given timestamp + Returns: + None if the URL isn't cached. + """ + def get_url_cache_txn(txn): + # get the most recently cached result (relative to the given ts) + sql = ( + "SELECT response_code, etag, expires, og, media_id, max(download_ts)" + " FROM local_media_repository_url_cache" + " WHERE url = ? AND download_ts <= ?" + ) + txn.execute(sql, (url, ts)) + row = txn.fetchone() + + if not row[3]: + # ...or if we've requested a timestamp older than the oldest + # copy in the cache, return the oldest copy (if any) + sql = ( + "SELECT response_code, etag, expires, og, media_id, min(download_ts)" + " FROM local_media_repository_url_cache" + " WHERE url = ? AND download_ts > ?" + ) + txn.execute(sql, (url, ts)) + row = txn.fetchone() + + if not row[3]: + return None + + return dict(zip(( + 'response_code', 'etag', 'expires', 'og', 'media_id', 'download_ts' + ), row)) + + return self.runInteraction( + "get_url_cache", get_url_cache_txn + ) + + def store_url_cache(self, url, response_code, etag, expires, og, media_id, + download_ts): + return self._simple_insert( + "local_media_repository_url_cache", + { + "url": url, + "response_code": response_code, + "etag": etag, + "expires": expires, + "og": og, + "media_id": media_id, + "download_ts": download_ts, + }, + desc="store_url_cache", + ) + def get_local_media_thumbnails(self, media_id): return self._simple_select_list( "local_media_repository_thumbnails", diff --git a/synapse/storage/schema/delta/30/local_media_repository_url_cache.sql b/synapse/storage/schema/delta/30/local_media_repository_url_cache.sql new file mode 100644 index 0000000000..9efb4280eb --- /dev/null +++ b/synapse/storage/schema/delta/30/local_media_repository_url_cache.sql @@ -0,0 +1,27 @@ +/* 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. + */ + +CREATE TABLE local_media_repository_url_cache( + url TEXT, -- the URL being cached + response_code INTEGER, -- the HTTP response code of this download attempt + etag TEXT, -- the etag header of this response + expires INTEGER, -- the number of ms this response was valid for + og TEXT, -- cache of the OG metadata of this URL as JSON + media_id TEXT, -- the media_id, if any, of the URL's content in the repo + download_ts BIGINT -- the timestamp of this download attempt +); + +CREATE INDEX local_media_repository_url_cache_by_url_download_ts + ON local_media_repository_url_cache(url, download_ts); |