From 7f0565e029791e336933806522d0e87bce870f64 Mon Sep 17 00:00:00 2001
From: Erik Johnston <erik@matrix.org>
Date: Fri, 10 Sep 2021 10:16:52 +0100
Subject: [PATCH] Don't needlessly batch in `add_event_to_cache` (#10784)

We've already batched up the events previously, and assume in other
places in the events.py file that we have. Removing this makes it easier
to adjust the batch sizes in one place.
---
 changelog.d/10784.misc                   |  1 +
 synapse/storage/databases/main/events.py | 49 +++++++++++-------------
 2 files changed, 24 insertions(+), 26 deletions(-)
 create mode 100644 changelog.d/10784.misc

diff --git a/changelog.d/10784.misc b/changelog.d/10784.misc
new file mode 100644
index 000000000..3b7acff03
--- /dev/null
+++ b/changelog.d/10784.misc
@@ -0,0 +1 @@
+Minor speed ups when joining large rooms over federation.
diff --git a/synapse/storage/databases/main/events.py b/synapse/storage/databases/main/events.py
index 14ada8a8b..8e691678e 100644
--- a/synapse/storage/databases/main/events.py
+++ b/synapse/storage/databases/main/events.py
@@ -1547,35 +1547,32 @@ class PersistEventsStore:
         to_prefill = []
 
         rows = []
-        N = 200
-        for i in range(0, len(events_and_contexts), N):
-            ev_map = {e[0].event_id: e[0] for e in events_and_contexts[i : i + N]}
-            if not ev_map:
-                break
 
-            sql = (
-                "SELECT "
-                " e.event_id as event_id, "
-                " r.redacts as redacts,"
-                " rej.event_id as rejects "
-                " FROM events as e"
-                " LEFT JOIN rejections as rej USING (event_id)"
-                " LEFT JOIN redactions as r ON e.event_id = r.redacts"
-                " WHERE "
-            )
+        ev_map = {e.event_id: e for e, _ in events_and_contexts}
+        if not ev_map:
+            return
 
-            clause, args = make_in_list_sql_clause(
-                self.database_engine, "e.event_id", list(ev_map)
-            )
+        sql = (
+            "SELECT "
+            " e.event_id as event_id, "
+            " r.redacts as redacts,"
+            " rej.event_id as rejects "
+            " FROM events as e"
+            " LEFT JOIN rejections as rej USING (event_id)"
+            " LEFT JOIN redactions as r ON e.event_id = r.redacts"
+            " WHERE "
+        )
 
-            txn.execute(sql + clause, args)
-            rows = self.db_pool.cursor_to_dict(txn)
-            for row in rows:
-                event = ev_map[row["event_id"]]
-                if not row["rejects"] and not row["redacts"]:
-                    to_prefill.append(
-                        _EventCacheEntry(event=event, redacted_event=None)
-                    )
+        clause, args = make_in_list_sql_clause(
+            self.database_engine, "e.event_id", list(ev_map)
+        )
+
+        txn.execute(sql + clause, args)
+        rows = self.db_pool.cursor_to_dict(txn)
+        for row in rows:
+            event = ev_map[row["event_id"]]
+            if not row["rejects"] and not row["redacts"]:
+                to_prefill.append(_EventCacheEntry(event=event, redacted_event=None))
 
         def prefill():
             for cache_entry in to_prefill: