From ac51bd581aa98b8972d785a898d6233def9b636a Mon Sep 17 00:00:00 2001
From: Patrick Cloke <clokep@users.noreply.github.com>
Date: Tue, 16 Jun 2020 10:43:29 -0400
Subject: Include a user agent in federation requests. (#7677)

---
 synapse/http/federation/matrix_federation_agent.py | 10 +++++++++-
 synapse/http/federation/well_known_resolver.py     | 17 +++++++++++++++--
 synapse/http/matrixfederationclient.py             |  9 ++++++++-
 3 files changed, 32 insertions(+), 4 deletions(-)

(limited to 'synapse/http')

diff --git a/synapse/http/federation/matrix_federation_agent.py b/synapse/http/federation/matrix_federation_agent.py
index f5f917f5ae..c5fc746f2f 100644
--- a/synapse/http/federation/matrix_federation_agent.py
+++ b/synapse/http/federation/matrix_federation_agent.py
@@ -48,6 +48,9 @@ class MatrixFederationAgent(object):
         tls_client_options_factory (FederationPolicyForHTTPS|None):
             factory to use for fetching client tls options, or none to disable TLS.
 
+        user_agent (bytes):
+            The user agent header to use for federation requests.
+
         _srv_resolver (SrvResolver|None):
             SRVResolver impl to use for looking up SRV records. None to use a default
             implementation.
@@ -61,6 +64,7 @@ class MatrixFederationAgent(object):
         self,
         reactor,
         tls_client_options_factory,
+        user_agent,
         _srv_resolver=None,
         _well_known_resolver=None,
     ):
@@ -78,6 +82,7 @@ class MatrixFederationAgent(object):
             ),
             pool=self._pool,
         )
+        self.user_agent = user_agent
 
         if _well_known_resolver is None:
             _well_known_resolver = WellKnownResolver(
@@ -87,6 +92,7 @@ class MatrixFederationAgent(object):
                     pool=self._pool,
                     contextFactory=tls_client_options_factory,
                 ),
+                user_agent=self.user_agent,
             )
 
         self._well_known_resolver = _well_known_resolver
@@ -149,7 +155,7 @@ class MatrixFederationAgent(object):
             parsed_uri = urllib.parse.urlparse(uri)
 
         # We need to make sure the host header is set to the netloc of the
-        # server.
+        # server and that a user-agent is provided.
         if headers is None:
             headers = Headers()
         else:
@@ -157,6 +163,8 @@ class MatrixFederationAgent(object):
 
         if not headers.hasHeader(b"host"):
             headers.addRawHeader(b"host", parsed_uri.netloc)
+        if not headers.hasHeader(b"user-agent"):
+            headers.addRawHeader(b"user-agent", self.user_agent)
 
         res = yield make_deferred_yieldable(
             self._agent.request(method, uri, headers, bodyProducer)
diff --git a/synapse/http/federation/well_known_resolver.py b/synapse/http/federation/well_known_resolver.py
index 7ddfad286d..89a3b041ce 100644
--- a/synapse/http/federation/well_known_resolver.py
+++ b/synapse/http/federation/well_known_resolver.py
@@ -23,6 +23,7 @@ import attr
 from twisted.internet import defer
 from twisted.web.client import RedirectAgent, readBody
 from twisted.web.http import stringToDatetime
+from twisted.web.http_headers import Headers
 
 from synapse.logging.context import make_deferred_yieldable
 from synapse.util import Clock
@@ -78,7 +79,12 @@ class WellKnownResolver(object):
     """
 
     def __init__(
-        self, reactor, agent, well_known_cache=None, had_well_known_cache=None
+        self,
+        reactor,
+        agent,
+        user_agent,
+        well_known_cache=None,
+        had_well_known_cache=None,
     ):
         self._reactor = reactor
         self._clock = Clock(reactor)
@@ -92,6 +98,7 @@ class WellKnownResolver(object):
         self._well_known_cache = well_known_cache
         self._had_valid_well_known_cache = had_well_known_cache
         self._well_known_agent = RedirectAgent(agent)
+        self.user_agent = user_agent
 
     @defer.inlineCallbacks
     def get_well_known(self, server_name):
@@ -227,6 +234,10 @@ class WellKnownResolver(object):
         uri = b"https://%s/.well-known/matrix/server" % (server_name,)
         uri_str = uri.decode("ascii")
 
+        headers = {
+            b"User-Agent": [self.user_agent],
+        }
+
         i = 0
         while True:
             i += 1
@@ -234,7 +245,9 @@ class WellKnownResolver(object):
             logger.info("Fetching %s", uri_str)
             try:
                 response = yield make_deferred_yieldable(
-                    self._well_known_agent.request(b"GET", uri)
+                    self._well_known_agent.request(
+                        b"GET", uri, headers=Headers(headers)
+                    )
                 )
                 body = yield make_deferred_yieldable(readBody(response))
 
diff --git a/synapse/http/matrixfederationclient.py b/synapse/http/matrixfederationclient.py
index 7b33b9f10a..18f6a8fd29 100644
--- a/synapse/http/matrixfederationclient.py
+++ b/synapse/http/matrixfederationclient.py
@@ -197,7 +197,14 @@ class MatrixFederationHttpClient(object):
 
         self.reactor = Reactor()
 
-        self.agent = MatrixFederationAgent(self.reactor, tls_client_options_factory)
+        user_agent = hs.version_string
+        if hs.config.user_agent_suffix:
+            user_agent = "%s %s" % (user_agent, hs.config.user_agent_suffix)
+        user_agent = user_agent.encode("ascii")
+
+        self.agent = MatrixFederationAgent(
+            self.reactor, tls_client_options_factory, user_agent
+        )
 
         # Use a BlacklistingAgentWrapper to prevent circumventing the IP
         # blacklist via IP literals in server names
-- 
cgit 1.4.1