Reduce unnecessary whitespace in JSON. (#7372)

15 files changed, 56 insertions, 53 deletions

diff --git a/changelog.d/7372.misc b/changelog.d/7372.misc
new file mode 100644
index 0000000000..67a39f0471
--- /dev/null
+++ b/changelog.d/7372.misc
@@ -0,0 +1 @@
+Reduce the amount of whitespace in JSON stored and sent in responses. Contributed by David Vo.
diff --git a/synapse/http/server.py b/synapse/http/server.py
index 94ab29974a..ffe6cfa09e 100644
--- a/synapse/http/server.py
+++ b/synapse/http/server.py
@@ -25,7 +25,7 @@ from io import BytesIO
 from typing import Any, Callable, Dict, Tuple, Union
 
 import jinja2
-from canonicaljson import encode_canonical_json, encode_pretty_printed_json, json
+from canonicaljson import encode_canonical_json, encode_pretty_printed_json
 
 from twisted.internet import defer
 from twisted.python import failure
@@ -46,6 +46,7 @@ from synapse.api.errors import (
 from synapse.http.site import SynapseRequest
 from synapse.logging.context import preserve_fn
 from synapse.logging.opentracing import trace_servlet
+from synapse.util import json_encoder
 from synapse.util.caches import intern_dict
 
 logger = logging.getLogger(__name__)
@@ -538,7 +539,7 @@ def respond_with_json(
             # canonicaljson already encodes to bytes
             json_bytes = encode_canonical_json(json_object)
         else:
-            json_bytes = json.dumps(json_object).encode("utf-8")
+            json_bytes = json_encoder.encode(json_object).encode("utf-8")
 
     return respond_with_json_bytes(request, code, json_bytes, send_cors=send_cors)
 
diff --git a/synapse/replication/tcp/commands.py b/synapse/replication/tcp/commands.py
index f33801f883..d853e4447e 100644
--- a/synapse/replication/tcp/commands.py
+++ b/synapse/replication/tcp/commands.py
@@ -18,11 +18,12 @@ The VALID_SERVER_COMMANDS and VALID_CLIENT_COMMANDS define which commands are
 allowed to be sent by which side.
 """
 import abc
-import json
 import logging
 from typing import Tuple, Type
 
-_json_encoder = json.JSONEncoder()
+from canonicaljson import json
+
+from synapse.util import json_encoder as _json_encoder
 
 logger = logging.getLogger(__name__)
 
diff --git a/synapse/rest/media/v1/preview_url_resource.py b/synapse/rest/media/v1/preview_url_resource.py
index f4768a9e8b..4bb454c36f 100644
--- a/synapse/rest/media/v1/preview_url_resource.py
+++ b/synapse/rest/media/v1/preview_url_resource.py
@@ -27,7 +27,6 @@ from typing import Dict, Optional
 from urllib import parse as urlparse
 
 import attr
-from canonicaljson import json
 
 from twisted.internet import defer
 from twisted.internet.error import DNSLookupError
@@ -43,6 +42,7 @@ from synapse.http.servlet import parse_integer, parse_string
 from synapse.logging.context import make_deferred_yieldable, run_in_background
 from synapse.metrics.background_process_metrics import run_as_background_process
 from synapse.rest.media.v1._base import get_filename_from_headers
+from synapse.util import json_encoder
 from synapse.util.async_helpers import ObservableDeferred
 from synapse.util.caches.expiringcache import ExpiringCache
 from synapse.util.stringutils import random_string
@@ -355,7 +355,7 @@ class PreviewUrlResource(DirectServeJsonResource):
 
         logger.debug("Calculated OG for %s as %s", url, og)
 
-        jsonog = json.dumps(og)
+        jsonog = json_encoder.encode(og)
 
         # store OG in history-aware DB cache
         await self.store.store_url_cache(
diff --git a/synapse/storage/databases/main/account_data.py b/synapse/storage/databases/main/account_data.py
index 2193d8fdc5..cf039e7f7d 100644
--- a/synapse/storage/databases/main/account_data.py
+++ b/synapse/storage/databases/main/account_data.py
@@ -18,13 +18,12 @@ import abc
 import logging
 from typing import List, Tuple
 
-from canonicaljson import json
-
 from twisted.internet import defer
 
 from synapse.storage._base import SQLBaseStore, db_to_json
 from synapse.storage.database import DatabasePool
 from synapse.storage.util.id_generators import StreamIdGenerator
+from synapse.util import json_encoder
 from synapse.util.caches.descriptors import cached, cachedInlineCallbacks
 from synapse.util.caches.stream_change_cache import StreamChangeCache
 
@@ -327,7 +326,7 @@ class AccountDataStore(AccountDataWorkerStore):
         Returns:
             A deferred that completes once the account_data has been added.
         """
-        content_json = json.dumps(content)
+        content_json = json_encoder.encode(content)
 
         with self._account_data_id_gen.get_next() as next_id:
             # no need to lock here as room_account_data has a unique constraint
@@ -373,7 +372,7 @@ class AccountDataStore(AccountDataWorkerStore):
         Returns:
             A deferred that completes once the account_data has been added.
         """
-        content_json = json.dumps(content)
+        content_json = json_encoder.encode(content)
 
         with self._account_data_id_gen.get_next() as next_id:
             # no need to lock here as account_data has a unique constraint on
diff --git a/synapse/storage/databases/main/deviceinbox.py b/synapse/storage/databases/main/deviceinbox.py
index 874ecdf8d2..76ec954f44 100644
--- a/synapse/storage/databases/main/deviceinbox.py
+++ b/synapse/storage/databases/main/deviceinbox.py
@@ -16,13 +16,12 @@
 import logging
 from typing import List, Tuple
 
-from canonicaljson import json
-
 from twisted.internet import defer
 
 from synapse.logging.opentracing import log_kv, set_tag, trace
 from synapse.storage._base import SQLBaseStore, db_to_json, make_in_list_sql_clause
 from synapse.storage.database import DatabasePool
+from synapse.util import json_encoder
 from synapse.util.caches.expiringcache import ExpiringCache
 
 logger = logging.getLogger(__name__)
@@ -354,7 +353,7 @@ class DeviceInboxStore(DeviceInboxWorkerStore, DeviceInboxBackgroundUpdateStore)
             )
             rows = []
             for destination, edu in remote_messages_by_destination.items():
-                edu_json = json.dumps(edu)
+                edu_json = json_encoder.encode(edu)
                 rows.append((destination, stream_id, now_ms, edu_json))
             txn.executemany(sql, rows)
 
@@ -432,7 +431,7 @@ class DeviceInboxStore(DeviceInboxWorkerStore, DeviceInboxBackgroundUpdateStore)
                 # Handle wildcard device_ids.
                 sql = "SELECT device_id FROM devices WHERE user_id = ?"
                 txn.execute(sql, (user_id,))
-                message_json = json.dumps(messages_by_device["*"])
+                message_json = json_encoder.encode(messages_by_device["*"])
                 for row in txn:
                     # Add the message for all devices for this user on this
                     # server.
@@ -454,7 +453,7 @@ class DeviceInboxStore(DeviceInboxWorkerStore, DeviceInboxBackgroundUpdateStore)
                     # Only insert into the local inbox if the device exists on
                     # this server
                     device = row[0]
-                    message_json = json.dumps(messages_by_device[device])
+                    message_json = json_encoder.encode(messages_by_device[device])
                     messages_json_for_user[device] = message_json
 
             if messages_json_for_user:
diff --git a/synapse/storage/databases/main/devices.py b/synapse/storage/databases/main/devices.py
index 88a7aadfc6..81e64de126 100644
--- a/synapse/storage/databases/main/devices.py
+++ b/synapse/storage/databases/main/devices.py
@@ -17,8 +17,6 @@
 import logging
 from typing import List, Optional, Set, Tuple
 
-from canonicaljson import json
-
 from twisted.internet import defer
 
 from synapse.api.errors import Codes, StoreError
@@ -36,6 +34,7 @@ from synapse.storage.database import (
     make_tuple_comparison_clause,
 )
 from synapse.types import Collection, get_verify_key_from_cross_signing_key
+from synapse.util import json_encoder
 from synapse.util.caches.descriptors import (
     Cache,
     cached,
@@ -397,7 +396,7 @@ class DeviceWorkerStore(SQLBaseStore):
             values={
                 "stream_id": stream_id,
                 "from_user_id": from_user_id,
-                "user_ids": json.dumps(user_ids),
+                "user_ids": json_encoder.encode(user_ids),
             },
         )
 
@@ -1032,7 +1031,7 @@ class DeviceStore(DeviceWorkerStore, DeviceBackgroundUpdateStore):
                 txn,
                 table="device_lists_remote_cache",
                 keyvalues={"user_id": user_id, "device_id": device_id},
-                values={"content": json.dumps(content)},
+                values={"content": json_encoder.encode(content)},
                 # we don't need to lock, because we assume we are the only thread
                 # updating this user's devices.
                 lock=False,
@@ -1088,7 +1087,7 @@ class DeviceStore(DeviceWorkerStore, DeviceBackgroundUpdateStore):
                 {
                     "user_id": user_id,
                     "device_id": content["device_id"],
-                    "content": json.dumps(content),
+                    "content": json_encoder.encode(content),
                 }
                 for content in devices
             ],
@@ -1209,7 +1208,7 @@ class DeviceStore(DeviceWorkerStore, DeviceBackgroundUpdateStore):
                     "device_id": device_id,
                     "sent": False,
                     "ts": now,
-                    "opentracing_context": json.dumps(context)
+                    "opentracing_context": json_encoder.encode(context)
                     if whitelisted_homeserver(destination)
                     else "{}",
                 }
diff --git a/synapse/storage/databases/main/e2e_room_keys.py b/synapse/storage/databases/main/e2e_room_keys.py
index 90152edc3c..c4aaec3993 100644
--- a/synapse/storage/databases/main/e2e_room_keys.py
+++ b/synapse/storage/databases/main/e2e_room_keys.py
@@ -14,13 +14,12 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-from canonicaljson import json
-
 from twisted.internet import defer
 
 from synapse.api.errors import StoreError
 from synapse.logging.opentracing import log_kv, trace
 from synapse.storage._base import SQLBaseStore, db_to_json
+from synapse.util import json_encoder
 
 
 class EndToEndRoomKeyStore(SQLBaseStore):
@@ -50,7 +49,7 @@ class EndToEndRoomKeyStore(SQLBaseStore):
                 "first_message_index": room_key["first_message_index"],
                 "forwarded_count": room_key["forwarded_count"],
                 "is_verified": room_key["is_verified"],
-                "session_data": json.dumps(room_key["session_data"]),
+                "session_data": json_encoder.encode(room_key["session_data"]),
             },
             desc="update_e2e_room_key",
         )
@@ -77,7 +76,7 @@ class EndToEndRoomKeyStore(SQLBaseStore):
                     "first_message_index": room_key["first_message_index"],
                     "forwarded_count": room_key["forwarded_count"],
                     "is_verified": room_key["is_verified"],
-                    "session_data": json.dumps(room_key["session_data"]),
+                    "session_data": json_encoder.encode(room_key["session_data"]),
                 }
             )
             log_kv(
@@ -360,7 +359,7 @@ class EndToEndRoomKeyStore(SQLBaseStore):
                     "user_id": user_id,
                     "version": new_version,
                     "algorithm": info["algorithm"],
-                    "auth_data": json.dumps(info["auth_data"]),
+                    "auth_data": json_encoder.encode(info["auth_data"]),
                 },
             )
 
@@ -387,7 +386,7 @@ class EndToEndRoomKeyStore(SQLBaseStore):
         updatevalues = {}
 
         if info is not None and "auth_data" in info:
-            updatevalues["auth_data"] = json.dumps(info["auth_data"])
+            updatevalues["auth_data"] = json_encoder.encode(info["auth_data"])
         if version_etag is not None:
             updatevalues["etag"] = version_etag
 
diff --git a/synapse/storage/databases/main/end_to_end_keys.py b/synapse/storage/databases/main/end_to_end_keys.py
index 40354b8304..6126376a6f 100644
--- a/synapse/storage/databases/main/end_to_end_keys.py
+++ b/synapse/storage/databases/main/end_to_end_keys.py
@@ -16,7 +16,7 @@
 # limitations under the License.
 from typing import Dict, List, Tuple
 
-from canonicaljson import encode_canonical_json, json
+from canonicaljson import encode_canonical_json
 
 from twisted.enterprise.adbapi import Connection
 from twisted.internet import defer
@@ -24,6 +24,7 @@ from twisted.internet import defer
 from synapse.logging.opentracing import log_kv, set_tag, trace
 from synapse.storage._base import SQLBaseStore, db_to_json
 from synapse.storage.database import make_in_list_sql_clause
+from synapse.util import json_encoder
 from synapse.util.caches.descriptors import cached, cachedList
 from synapse.util.iterutils import batch_iter
 
@@ -700,7 +701,7 @@ class EndToEndKeyStore(EndToEndKeyWorkerStore, SQLBaseStore):
                 values={
                     "user_id": user_id,
                     "keytype": key_type,
-                    "keydata": json.dumps(key),
+                    "keydata": json_encoder.encode(key),
                     "stream_id": stream_id,
                 },
             )
diff --git a/synapse/storage/databases/main/event_push_actions.py b/synapse/storage/databases/main/event_push_actions.py
index b8cefb4d5e..7c246d3e4c 100644
--- a/synapse/storage/databases/main/event_push_actions.py
+++ b/synapse/storage/databases/main/event_push_actions.py
@@ -17,11 +17,10 @@
 import logging
 from typing import List
 
-from canonicaljson import json
-
 from synapse.metrics.background_process_metrics import run_as_background_process
 from synapse.storage._base import LoggingTransaction, SQLBaseStore, db_to_json
 from synapse.storage.database import DatabasePool
+from synapse.util import json_encoder
 from synapse.util.caches.descriptors import cachedInlineCallbacks
 
 logger = logging.getLogger(__name__)
@@ -50,7 +49,7 @@ def _serialize_action(actions, is_highlight):
     else:
         if actions == DEFAULT_NOTIF_ACTION:
             return ""
-    return json.dumps(actions)
+    return json_encoder.encode(actions)
 
 
 def _deserialize_action(actions, is_highlight):
diff --git a/synapse/storage/databases/main/group_server.py b/synapse/storage/databases/main/group_server.py
index a98181f445..75ea6d4b2f 100644
--- a/synapse/storage/databases/main/group_server.py
+++ b/synapse/storage/databases/main/group_server.py
@@ -16,12 +16,11 @@
 
 from typing import List, Tuple
 
-from canonicaljson import json
-
 from twisted.internet import defer
 
 from synapse.api.errors import SynapseError
 from synapse.storage._base import SQLBaseStore, db_to_json
+from synapse.util import json_encoder
 
 # The category ID for the "default" category. We don't store as null in the
 # database to avoid the fun of null != null
@@ -752,7 +751,7 @@ class GroupServerStore(GroupServerWorkerStore):
         if profile is None:
             insertion_values["profile"] = "{}"
         else:
-            update_values["profile"] = json.dumps(profile)
+            update_values["profile"] = json_encoder.encode(profile)
 
         if is_public is None:
             insertion_values["is_public"] = True
@@ -783,7 +782,7 @@ class GroupServerStore(GroupServerWorkerStore):
         if profile is None:
             insertion_values["profile"] = "{}"
         else:
-            update_values["profile"] = json.dumps(profile)
+            update_values["profile"] = json_encoder.encode(profile)
 
         if is_public is None:
             insertion_values["is_public"] = True
@@ -1007,7 +1006,7 @@ class GroupServerStore(GroupServerWorkerStore):
                         "group_id": group_id,
                         "user_id": user_id,
                         "valid_until_ms": remote_attestation["valid_until_ms"],
-                        "attestation_json": json.dumps(remote_attestation),
+                        "attestation_json": json_encoder.encode(remote_attestation),
                     },
                 )
 
@@ -1131,7 +1130,7 @@ class GroupServerStore(GroupServerWorkerStore):
                     "is_admin": is_admin,
                     "membership": membership,
                     "is_publicised": is_publicised,
-                    "content": json.dumps(content),
+                    "content": json_encoder.encode(content),
                 },
             )
 
@@ -1143,7 +1142,7 @@ class GroupServerStore(GroupServerWorkerStore):
                     "group_id": group_id,
                     "user_id": user_id,
                     "type": "membership",
-                    "content": json.dumps(
+                    "content": json_encoder.encode(
                         {"membership": membership, "content": content}
                     ),
                 },
@@ -1171,7 +1170,7 @@ class GroupServerStore(GroupServerWorkerStore):
                             "group_id": group_id,
                             "user_id": user_id,
                             "valid_until_ms": remote_attestation["valid_until_ms"],
-                            "attestation_json": json.dumps(remote_attestation),
+                            "attestation_json": json_encoder.encode(remote_attestation),
                         },
                     )
             else:
@@ -1240,7 +1239,7 @@ class GroupServerStore(GroupServerWorkerStore):
             keyvalues={"group_id": group_id, "user_id": user_id},
             updatevalues={
                 "valid_until_ms": attestation["valid_until_ms"],
-                "attestation_json": json.dumps(attestation),
+                "attestation_json": json_encoder.encode(attestation),
             },
             desc="update_remote_attestion",
         )
diff --git a/synapse/storage/databases/main/push_rule.py b/synapse/storage/databases/main/push_rule.py
index 97cc12931d..264521635f 100644
--- a/synapse/storage/databases/main/push_rule.py
+++ b/synapse/storage/databases/main/push_rule.py
@@ -18,8 +18,6 @@ import abc
 import logging
 from typing import List, Tuple, Union
 
-from canonicaljson import json
-
 from twisted.internet import defer
 
 from synapse.push.baserules import list_with_base_rules
@@ -33,6 +31,7 @@ from synapse.storage.databases.main.receipts import ReceiptsWorkerStore
 from synapse.storage.databases.main.roommember import RoomMemberWorkerStore
 from synapse.storage.push_rule import InconsistentRuleException, RuleNotFoundException
 from synapse.storage.util.id_generators import ChainedIdGenerator
+from synapse.util import json_encoder
 from synapse.util.caches.descriptors import cachedInlineCallbacks, cachedList
 from synapse.util.caches.stream_change_cache import StreamChangeCache
 
@@ -411,8 +410,8 @@ class PushRuleStore(PushRulesWorkerStore):
         before=None,
         after=None,
     ):
-        conditions_json = json.dumps(conditions)
-        actions_json = json.dumps(actions)
+        conditions_json = json_encoder.encode(conditions)
+        actions_json = json_encoder.encode(actions)
         with self._push_rules_stream_id_gen.get_next() as ids:
             stream_id, event_stream_ordering = ids
             if before or after:
@@ -681,7 +680,7 @@ class PushRuleStore(PushRulesWorkerStore):
 
     @defer.inlineCallbacks
     def set_push_rule_actions(self, user_id, rule_id, actions, is_default_rule):
-        actions_json = json.dumps(actions)
+        actions_json = json_encoder.encode(actions)
 
         def set_push_rule_actions_txn(txn, stream_id, event_stream_ordering):
             if is_default_rule:
diff --git a/synapse/storage/databases/main/receipts.py b/synapse/storage/databases/main/receipts.py
index 6255977c92..1920a8a152 100644
--- a/synapse/storage/databases/main/receipts.py
+++ b/synapse/storage/databases/main/receipts.py
@@ -18,13 +18,12 @@ import abc
 import logging
 from typing import List, Tuple
 
-from canonicaljson import json
-
 from twisted.internet import defer
 
 from synapse.storage._base import SQLBaseStore, db_to_json, make_in_list_sql_clause
 from synapse.storage.database import DatabasePool
 from synapse.storage.util.id_generators import StreamIdGenerator
+from synapse.util import json_encoder
 from synapse.util.async_helpers import ObservableDeferred
 from synapse.util.caches.descriptors import cached, cachedInlineCallbacks, cachedList
 from synapse.util.caches.stream_change_cache import StreamChangeCache
@@ -459,7 +458,7 @@ class ReceiptsStore(ReceiptsWorkerStore):
             values={
                 "stream_id": stream_id,
                 "event_id": event_id,
-                "data": json.dumps(data),
+                "data": json_encoder.encode(data),
             },
             # receipts_linearized has a unique constraint on
             # (user_id, room_id, receipt_type), so no need to lock
@@ -585,7 +584,7 @@ class ReceiptsStore(ReceiptsWorkerStore):
                 "room_id": room_id,
                 "receipt_type": receipt_type,
                 "user_id": user_id,
-                "event_ids": json.dumps(event_ids),
-                "data": json.dumps(data),
+                "event_ids": json_encoder.encode(event_ids),
+                "data": json_encoder.encode(data),
             },
         )
diff --git a/synapse/util/__init__.py b/synapse/util/__init__.py
index c63256d3bd..b3f76428b6 100644
--- a/synapse/util/__init__.py
+++ b/synapse/util/__init__.py
@@ -17,6 +17,7 @@ import logging
 import re
 
 import attr
+from canonicaljson import json
 
 from twisted.internet import defer, task
 
@@ -24,6 +25,9 @@ from synapse.logging import context
 
 logger = logging.getLogger(__name__)
 
+# Create a custom encoder to reduce the whitespace produced by JSON encoding.
+json_encoder = json.JSONEncoder(separators=(",", ":"))
+
 
 def unwrapFirstError(failure):
     # defer.gatherResults and DeferredLists wrap failures.
diff --git a/synapse/util/frozenutils.py b/synapse/util/frozenutils.py
index eab78dd256..0e445e01d7 100644
--- a/synapse/util/frozenutils.py
+++ b/synapse/util/frozenutils.py
@@ -63,5 +63,8 @@ def _handle_frozendict(obj):
     )
 
 
-# A JSONEncoder which is capable of encoding frozendicts without barfing
-frozendict_json_encoder = json.JSONEncoder(default=_handle_frozendict)
+# A JSONEncoder which is capable of encoding frozendicts without barfing.
+# Additionally reduce the whitespace produced by JSON encoding.
+frozendict_json_encoder = json.JSONEncoder(
+    default=_handle_frozendict, separators=(",", ":"),
+)