From 2ec8ca5e6046b207eabe008101631b978758ac6d Mon Sep 17 00:00:00 2001 From: Richard van der Hoff <1389908+richvdh@users.noreply.github.com> Date: Tue, 12 Jan 2021 12:34:16 +0000 Subject: Remove SynapseRequest.get_user_agent (#9069) SynapseRequest is in danger of becoming a bit of a dumping-ground for "useful stuff relating to Requests", which isn't really its intention (its purpose is to override render, finished and connectionLost to set up the LoggingContext and write the right entries to the request log). Putting utility functions inside SynapseRequest means that lots of our code ends up requiring a SynapseRequest when there is nothing synapse-specific about the Request at all, and any old twisted.web.iweb.IRequest will do. This increases code coupling and makes testing more difficult. In short: move get_user_agent out to a utility function. --- synapse/api/auth.py | 3 ++- synapse/handlers/auth.py | 6 +++--- synapse/handlers/sso.py | 5 +++-- synapse/http/__init__.py | 15 +++++++++++++++ synapse/http/site.py | 18 ++---------------- 5 files changed, 25 insertions(+), 22 deletions(-) (limited to 'synapse') diff --git a/synapse/api/auth.py b/synapse/api/auth.py index 48c4d7b0be..6d6703250b 100644 --- a/synapse/api/auth.py +++ b/synapse/api/auth.py @@ -33,6 +33,7 @@ from synapse.api.errors import ( from synapse.api.room_versions import KNOWN_ROOM_VERSIONS from synapse.appservice import ApplicationService from synapse.events import EventBase +from synapse.http import get_request_user_agent from synapse.http.site import SynapseRequest from synapse.logging import opentracing as opentracing from synapse.storage.databases.main.registration import TokenLookupResult @@ -187,7 +188,7 @@ class Auth: """ try: ip_addr = self.hs.get_ip_from_request(request) - user_agent = request.get_user_agent("") + user_agent = get_request_user_agent(request) access_token = self.get_access_token_from_request(request) diff --git a/synapse/handlers/auth.py b/synapse/handlers/auth.py index f4434673dc..5b86ee85c7 100644 --- a/synapse/handlers/auth.py +++ b/synapse/handlers/auth.py @@ -49,8 +49,10 @@ from synapse.api.errors import ( UserDeactivatedError, ) from synapse.api.ratelimiting import Ratelimiter +from synapse.handlers._base import BaseHandler from synapse.handlers.ui_auth import INTERACTIVE_AUTH_CHECKERS from synapse.handlers.ui_auth.checkers import UserInteractiveAuthChecker +from synapse.http import get_request_user_agent from synapse.http.server import finish_request, respond_with_html from synapse.http.site import SynapseRequest from synapse.logging.context import defer_to_thread @@ -62,8 +64,6 @@ from synapse.util.async_helpers import maybe_awaitable from synapse.util.msisdn import phone_number_to_msisdn from synapse.util.threepids import canonicalise_email -from ._base import BaseHandler - if TYPE_CHECKING: from synapse.app.homeserver import HomeServer @@ -539,7 +539,7 @@ class AuthHandler(BaseHandler): # authentication flow. await self.store.set_ui_auth_clientdict(sid, clientdict) - user_agent = request.get_user_agent("") + user_agent = get_request_user_agent(request) await self.store.add_user_agent_ip_to_ui_auth_session( session.session_id, user_agent, clientip diff --git a/synapse/handlers/sso.py b/synapse/handlers/sso.py index 2da1ea2223..740df7e4a0 100644 --- a/synapse/handlers/sso.py +++ b/synapse/handlers/sso.py @@ -23,6 +23,7 @@ from typing_extensions import NoReturn, Protocol from twisted.web.http import Request from synapse.api.errors import Codes, RedirectException, SynapseError +from synapse.http import get_request_user_agent from synapse.http.server import respond_with_html from synapse.http.site import SynapseRequest from synapse.types import JsonDict, UserID, contains_invalid_mxid_characters @@ -362,7 +363,7 @@ class SsoHandler: attributes, auth_provider_id, remote_user_id, - request.get_user_agent(""), + get_request_user_agent(request), request.getClientIP(), ) @@ -628,7 +629,7 @@ class SsoHandler: attributes, session.auth_provider_id, session.remote_user_id, - request.get_user_agent(""), + get_request_user_agent(request), request.getClientIP(), ) diff --git a/synapse/http/__init__.py b/synapse/http/__init__.py index 59b01b812c..4bc3cb53f0 100644 --- a/synapse/http/__init__.py +++ b/synapse/http/__init__.py @@ -17,6 +17,7 @@ import re from twisted.internet import task from twisted.web.client import FileBodyProducer +from twisted.web.iweb import IRequest from synapse.api.errors import SynapseError @@ -50,3 +51,17 @@ class QuieterFileBodyProducer(FileBodyProducer): FileBodyProducer.stopProducing(self) except task.TaskStopped: pass + + +def get_request_user_agent(request: IRequest, default: str = "") -> str: + """Return the last User-Agent header, or the given default. + """ + # There could be raw utf-8 bytes in the User-Agent header. + + # N.B. if you don't do this, the logger explodes cryptically + # with maximum recursion trying to log errors about + # the charset problem. + # c.f. https://github.com/matrix-org/synapse/issues/3471 + + h = request.getHeader(b"User-Agent") + return h.decode("ascii", "replace") if h else default diff --git a/synapse/http/site.py b/synapse/http/site.py index 5a5790831b..12ec3f851f 100644 --- a/synapse/http/site.py +++ b/synapse/http/site.py @@ -20,7 +20,7 @@ from twisted.python.failure import Failure from twisted.web.server import Request, Site from synapse.config.server import ListenerConfig -from synapse.http import redact_uri +from synapse.http import get_request_user_agent, redact_uri from synapse.http.request_metrics import RequestMetrics, requests_counter from synapse.logging.context import LoggingContext, PreserveLoggingContext from synapse.types import Requester @@ -113,15 +113,6 @@ class SynapseRequest(Request): method = self.method.decode("ascii") return method - def get_user_agent(self, default: str) -> str: - """Return the last User-Agent header, or the given default. - """ - user_agent = self.requestHeaders.getRawHeaders(b"User-Agent", [None])[-1] - if user_agent is None: - return default - - return user_agent.decode("ascii", "replace") - def render(self, resrc): # this is called once a Resource has been found to serve the request; in our # case the Resource in question will normally be a JsonResource. @@ -292,12 +283,7 @@ class SynapseRequest(Request): # and can see that we're doing something wrong. authenticated_entity = repr(self.requester) # type: ignore[unreachable] - # ...or could be raw utf-8 bytes in the User-Agent header. - # N.B. if you don't do this, the logger explodes cryptically - # with maximum recursion trying to log errors about - # the charset problem. - # c.f. https://github.com/matrix-org/synapse/issues/3471 - user_agent = self.get_user_agent("-") + user_agent = get_request_user_agent(self, "-") code = str(self.code) if not self.finished: -- cgit 1.4.1