Merge branch 'develop' of github.com:matrix-org/synapse into erikj/refactor_repl_servlet

3 files changed, 66 insertions, 54 deletions

diff --git a/synapse/http/client.py b/synapse/http/client.py
index 25b6307884..3771e0b3f6 100644
--- a/synapse/http/client.py
+++ b/synapse/http/client.py
@@ -39,12 +39,7 @@ from twisted.web.client import (
 from twisted.web.http import PotentialDataLoss
 from twisted.web.http_headers import Headers
 
-from synapse.api.errors import (
-    CodeMessageException,
-    Codes,
-    MatrixCodeMessageException,
-    SynapseError,
-)
+from synapse.api.errors import Codes, HttpResponseException, SynapseError
 from synapse.http import cancelled_to_request_timed_out_error, redact_uri
 from synapse.http.endpoint import SpiderEndpoint
 from synapse.util.async import add_timeout_to_deferred
@@ -132,6 +127,11 @@ class SimpleHttpClient(object):
 
         Returns:
             Deferred[object]: parsed json
+
+        Raises:
+            HttpResponseException: On a non-2xx HTTP response.
+
+            ValueError: if the response was not JSON
         """
 
         # TODO: Do we ever want to log message contents?
@@ -155,7 +155,10 @@ class SimpleHttpClient(object):
 
         body = yield make_deferred_yieldable(readBody(response))
 
-        defer.returnValue(json.loads(body))
+        if 200 <= response.code < 300:
+            defer.returnValue(json.loads(body))
+        else:
+            raise HttpResponseException(response.code, response.phrase, body)
 
     @defer.inlineCallbacks
     def post_json_get_json(self, uri, post_json, headers=None):
@@ -169,6 +172,11 @@ class SimpleHttpClient(object):
 
         Returns:
             Deferred[object]: parsed json
+
+        Raises:
+            HttpResponseException: On a non-2xx HTTP response.
+
+            ValueError: if the response was not JSON
         """
         json_str = encode_canonical_json(post_json)
 
@@ -193,9 +201,7 @@ class SimpleHttpClient(object):
         if 200 <= response.code < 300:
             defer.returnValue(json.loads(body))
         else:
-            raise self._exceptionFromFailedRequest(response, body)
-
-        defer.returnValue(json.loads(body))
+            raise HttpResponseException(response.code, response.phrase, body)
 
     @defer.inlineCallbacks
     def get_json(self, uri, args={}, headers=None):
@@ -213,14 +219,12 @@ class SimpleHttpClient(object):
             Deferred: Succeeds when we get *any* 2xx HTTP response, with the
             HTTP body as JSON.
         Raises:
-            On a non-2xx HTTP response. The response body will be used as the
-            error message.
+            HttpResponseException On a non-2xx HTTP response.
+
+            ValueError: if the response was not JSON
         """
-        try:
-            body = yield self.get_raw(uri, args, headers=headers)
-            defer.returnValue(json.loads(body))
-        except CodeMessageException as e:
-            raise self._exceptionFromFailedRequest(e.code, e.msg)
+        body = yield self.get_raw(uri, args, headers=headers)
+        defer.returnValue(json.loads(body))
 
     @defer.inlineCallbacks
     def put_json(self, uri, json_body, args={}, headers=None):
@@ -239,7 +243,9 @@ class SimpleHttpClient(object):
             Deferred: Succeeds when we get *any* 2xx HTTP response, with the
             HTTP body as JSON.
         Raises:
-            On a non-2xx HTTP response.
+            HttpResponseException On a non-2xx HTTP response.
+
+            ValueError: if the response was not JSON
         """
         if len(args):
             query_bytes = urllib.urlencode(args, True)
@@ -266,10 +272,7 @@ class SimpleHttpClient(object):
         if 200 <= response.code < 300:
             defer.returnValue(json.loads(body))
         else:
-            # NB: This is explicitly not json.loads(body)'d because the contract
-            # of CodeMessageException is a *string* message. Callers can always
-            # load it into JSON if they want.
-            raise CodeMessageException(response.code, body)
+            raise HttpResponseException(response.code, response.phrase, body)
 
     @defer.inlineCallbacks
     def get_raw(self, uri, args={}, headers=None):
@@ -287,8 +290,7 @@ class SimpleHttpClient(object):
             Deferred: Succeeds when we get *any* 2xx HTTP response, with the
             HTTP body at text.
         Raises:
-            On a non-2xx HTTP response. The response body will be used as the
-            error message.
+            HttpResponseException on a non-2xx HTTP response.
         """
         if len(args):
             query_bytes = urllib.urlencode(args, True)
@@ -311,16 +313,7 @@ class SimpleHttpClient(object):
         if 200 <= response.code < 300:
             defer.returnValue(body)
         else:
-            raise CodeMessageException(response.code, body)
-
-    def _exceptionFromFailedRequest(self, response, body):
-        try:
-            jsonBody = json.loads(body)
-            errcode = jsonBody['errcode']
-            error = jsonBody['error']
-            return MatrixCodeMessageException(response.code, error, errcode)
-        except (ValueError, KeyError):
-            return CodeMessageException(response.code, body)
+            raise HttpResponseException(response.code, response.phrase, body)
 
     # XXX: FIXME: This is horribly copy-pasted from matrixfederationclient.
     # The two should be factored out.
diff --git a/synapse/http/server.py b/synapse/http/server.py
index c70fdbdfd2..6dacb31037 100644
--- a/synapse/http/server.py
+++ b/synapse/http/server.py
@@ -13,12 +13,13 @@
 # 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.
+
 import cgi
 import collections
 import logging
-import urllib
 
-from six.moves import http_client
+from six import PY3
+from six.moves import http_client, urllib
 
 from canonicaljson import encode_canonical_json, encode_pretty_printed_json, json
 
@@ -35,7 +36,6 @@ from synapse.api.errors import (
     Codes,
     SynapseError,
     UnrecognizedRequestError,
-    cs_exception,
 )
 from synapse.http.request_metrics import requests_counter
 from synapse.util.caches import intern_dict
@@ -76,16 +76,13 @@ def wrap_json_request_handler(h):
     def wrapped_request_handler(self, request):
         try:
             yield h(self, request)
-        except CodeMessageException as e:
+        except SynapseError as e:
             code = e.code
-            if isinstance(e, SynapseError):
-                logger.info(
-                    "%s SynapseError: %s - %s", request, code, e.msg
-                )
-            else:
-                logger.exception(e)
+            logger.info(
+                "%s SynapseError: %s - %s", request, code, e.msg
+            )
             respond_with_json(
-                request, code, cs_exception(e), send_cors=True,
+                request, code, e.error_dict(), send_cors=True,
                 pretty_print=_request_user_agent_is_curl(request),
             )
 
@@ -264,6 +261,7 @@ class JsonResource(HttpServer, resource.Resource):
         self.hs = hs
 
     def register_paths(self, method, path_patterns, callback):
+        method = method.encode("utf-8")  # method is bytes on py3
         for path_pattern in path_patterns:
             logger.debug("Registering for %s %s", method, path_pattern.pattern)
             self.path_regexs.setdefault(method, []).append(
@@ -296,8 +294,19 @@ class JsonResource(HttpServer, resource.Resource):
         # here. If it throws an exception, that is handled by the wrapper
         # installed by @request_handler.
 
+        def _unquote(s):
+            if PY3:
+                # On Python 3, unquote is unicode -> unicode
+                return urllib.parse.unquote(s)
+            else:
+                # On Python 2, unquote is bytes -> bytes We need to encode the
+                # URL again (as it was decoded by _get_handler_for request), as
+                # ASCII because it's a URL, and then decode it to get the UTF-8
+                # characters that were quoted.
+                return urllib.parse.unquote(s.encode('ascii')).decode('utf8')
+
         kwargs = intern_dict({
-            name: urllib.unquote(value).decode("UTF-8") if value else value
+            name: _unquote(value) if value else value
             for name, value in group_dict.items()
         })
 
@@ -313,9 +322,9 @@ class JsonResource(HttpServer, resource.Resource):
             request (twisted.web.http.Request):
 
         Returns:
-            Tuple[Callable, dict[str, str]]: callback method, and the dict
-                mapping keys to path components as specified in the handler's
-                path match regexp.
+            Tuple[Callable, dict[unicode, unicode]]: callback method, and the
+                dict mapping keys to path components as specified in the
+                handler's path match regexp.
 
                 The callback will normally be a method registered via
                 register_paths, so will return (possibly via Deferred) either
@@ -327,7 +336,7 @@ class JsonResource(HttpServer, resource.Resource):
         # Loop through all the registered callbacks to check if the method
         # and path regex match
         for path_entry in self.path_regexs.get(request.method, []):
-            m = path_entry.pattern.match(request.path)
+            m = path_entry.pattern.match(request.path.decode('ascii'))
             if m:
                 # We found a match!
                 return path_entry.callback, m.groupdict()
@@ -383,7 +392,7 @@ class RootRedirect(resource.Resource):
         self.url = path
 
     def render_GET(self, request):
-        return redirectTo(self.url, request)
+        return redirectTo(self.url.encode('ascii'), request)
 
     def getChild(self, name, request):
         if len(name) == 0:
@@ -404,12 +413,14 @@ def respond_with_json(request, code, json_object, send_cors=False,
         return
 
     if pretty_print:
-        json_bytes = encode_pretty_printed_json(json_object) + "\n"
+        json_bytes = (encode_pretty_printed_json(json_object) + "\n"
+                      ).encode("utf-8")
     else:
         if canonical_json or synapse.events.USE_FROZEN_DICTS:
+            # canonicaljson already encodes to bytes
             json_bytes = encode_canonical_json(json_object)
         else:
-            json_bytes = json.dumps(json_object)
+            json_bytes = json.dumps(json_object).encode("utf-8")
 
     return respond_with_json_bytes(
         request, code, json_bytes,
diff --git a/synapse/http/servlet.py b/synapse/http/servlet.py
index 882816dc8f..69f7085291 100644
--- a/synapse/http/servlet.py
+++ b/synapse/http/servlet.py
@@ -171,8 +171,16 @@ def parse_json_value_from_request(request, allow_empty_body=False):
     if not content_bytes and allow_empty_body:
         return None
 
+    # Decode to Unicode so that simplejson will return Unicode strings on
+    # Python 2
     try:
-        content = json.loads(content_bytes)
+        content_unicode = content_bytes.decode('utf8')
+    except UnicodeDecodeError:
+        logger.warn("Unable to decode UTF-8")
+        raise SynapseError(400, "Content not JSON.", errcode=Codes.NOT_JSON)
+
+    try:
+        content = json.loads(content_unicode)
     except Exception as e:
         logger.warn("Unable to parse JSON: %s", e)
         raise SynapseError(400, "Content not JSON.", errcode=Codes.NOT_JSON)