Disambiguate queries on state_key (#11497)

We're going to add a `state_key` column to the `events` table, so we need to
add some disambiguation to queries which use it.

Author: richvdh

changelog.d/11497.misc

@@ -0,0 +1 @@
+Preparation for database schema simplifications: disambiguate queries on `state_key`.

synapse/storage/databases/main/event_federation.py

@@ -1552,9 +1552,9 @@ def delete_event_auth(txn):
                 DELETE FROM event_auth
                 WHERE event_id IN (
                     SELECT event_id FROM events
-                    LEFT JOIN state_events USING (room_id, event_id)
+                    LEFT JOIN state_events AS se USING (room_id, event_id)
                     WHERE ? <= stream_ordering AND stream_ordering < ?
-                        AND state_key IS null
+                        AND se.state_key IS null
                 )
             """
 

synapse/storage/databases/main/events.py

@@ -575,9 +575,9 @@ def _add_chain_cover_index(
         # fetch their auth event info.
         while missing_auth_chains:
             sql = """
-                SELECT event_id, events.type, state_key, chain_id, sequence_number
+                SELECT event_id, events.type, se.state_key, chain_id, sequence_number
                 FROM events
-                INNER JOIN state_events USING (event_id)
+                INNER JOIN state_events AS se USING (event_id)
                 LEFT JOIN event_auth_chains USING (event_id)
                 WHERE
             """

synapse/storage/databases/main/events_worker.py

@@ -1408,10 +1408,10 @@ def get_all_new_forward_event_rows(
         ) -> List[Tuple[int, str, str, str, str, str, str, str, str]]:
             sql = (
                 "SELECT e.stream_ordering, e.event_id, e.room_id, e.type,"
-                " state_key, redacts, relates_to_id, membership, rejections.reason IS NOT NULL"
+                " se.state_key, redacts, relates_to_id, membership, rejections.reason IS NOT NULL"
                 " FROM events AS e"
                 " LEFT JOIN redactions USING (event_id)"
-                " LEFT JOIN state_events USING (event_id)"
+                " LEFT JOIN state_events AS se USING (event_id)"
                 " LEFT JOIN event_relations USING (event_id)"
                 " LEFT JOIN room_memberships USING (event_id)"
                 " LEFT JOIN rejections USING (event_id)"
@@ -1449,11 +1449,11 @@ def get_ex_outlier_stream_rows_txn(
         ) -> List[Tuple[int, str, str, str, str, str, str, str, str]]:
             sql = (
                 "SELECT event_stream_ordering, e.event_id, e.room_id, e.type,"
-                " state_key, redacts, relates_to_id, membership, rejections.reason IS NOT NULL"
+                " se.state_key, redacts, relates_to_id, membership, rejections.reason IS NOT NULL"
                 " FROM events AS e"
                 " INNER JOIN ex_outlier_stream AS out USING (event_id)"
                 " LEFT JOIN redactions USING (event_id)"
-                " LEFT JOIN state_events USING (event_id)"
+                " LEFT JOIN state_events AS se USING (event_id)"
                 " LEFT JOIN event_relations USING (event_id)"
                 " LEFT JOIN room_memberships USING (event_id)"
                 " LEFT JOIN rejections USING (event_id)"
@@ -1507,10 +1507,10 @@ def get_all_new_backfill_event_rows(
         ) -> Tuple[List[Tuple[int, Tuple[str, str, str, str, str, str]]], int, bool]:
             sql = (
                 "SELECT -e.stream_ordering, e.event_id, e.room_id, e.type,"
-                " state_key, redacts, relates_to_id"
+                " se.state_key, redacts, relates_to_id"
                 " FROM events AS e"
                 " LEFT JOIN redactions USING (event_id)"
-                " LEFT JOIN state_events USING (event_id)"
+                " LEFT JOIN state_events AS se USING (event_id)"
                 " LEFT JOIN event_relations USING (event_id)"
                 " WHERE ? > stream_ordering AND stream_ordering >= ?"
                 "  AND instance_name = ?"
@@ -1537,11 +1537,11 @@ def get_all_new_backfill_event_rows(
 
             sql = (
                 "SELECT -event_stream_ordering, e.event_id, e.room_id, e.type,"
-                " state_key, redacts, relates_to_id"
+                " se.state_key, redacts, relates_to_id"
                 " FROM events AS e"
                 " INNER JOIN ex_outlier_stream AS out USING (event_id)"
                 " LEFT JOIN redactions USING (event_id)"
-                " LEFT JOIN state_events USING (event_id)"
+                " LEFT JOIN state_events AS se USING (event_id)"
                 " LEFT JOIN event_relations USING (event_id)"
                 " WHERE ? > event_stream_ordering"
                 " AND event_stream_ordering >= ?"

synapse/storage/databases/main/purge_events.py

@@ -118,7 +118,7 @@ def _purge_history_txn(
 
         logger.info("[purge] looking for events to delete")
 
-        should_delete_expr = "state_key IS NULL"
+        should_delete_expr = "state_events.state_key IS NULL"
         should_delete_params: Tuple[Any, ...] = ()
         if not delete_local_events:
             should_delete_expr += " AND event_id NOT LIKE ?"

synapse/storage/databases/main/roommember.py

@@ -476,7 +476,7 @@ def _get_rooms_for_user_with_stream_ordering_txn(
                 INNER JOIN events AS e USING (room_id, event_id)
                 WHERE
                     c.type = 'm.room.member'
-                    AND state_key = ?
+                    AND c.state_key = ?
                     AND c.membership = ?
             """
         else:
@@ -487,7 +487,7 @@ def _get_rooms_for_user_with_stream_ordering_txn(
                 INNER JOIN events AS e USING (room_id, event_id)
                 WHERE
                     c.type = 'm.room.member'
-                    AND state_key = ?
+                    AND c.state_key = ?
                     AND m.membership = ?
             """
 

synapse/storage/schema/__init__.py

@@ -12,7 +12,7 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-SCHEMA_VERSION = 65  # remember to update the list below when updating
+SCHEMA_VERSION = 66  # remember to update the list below when updating
 """Represents the expectations made by the codebase about the database schema
 
 This should be incremented whenever the codebase changes its requirements on the
@@ -46,6 +46,10 @@
     - MSC2716: Remove unique event_id constraint from insertion_event_edges
       because an insertion event can have multiple edges.
     - Remove unused tables `user_stats_historical` and `room_stats_historical`.
+
+Changes in SCHEMA_VERSION = 66:
+    - Queries on state_key columns are now disambiguated (ie, the codebase can handle
+      the `events` table having a `state_key` column).
 """