Merge branch 'develop' into e2e_backups

1 files changed, 83 insertions, 0 deletions

diff --git a/synapse/storage/events_worker.py b/synapse/storage/events_worker.py
index 9b4cfeb899..59822178ff 100644
--- a/synapse/storage/events_worker.py
+++ b/synapse/storage/events_worker.py
@@ -12,6 +12,7 @@
 # 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 itertools
 import logging
 from collections import namedtuple
 
@@ -442,3 +443,85 @@ class EventsWorkerStore(SQLBaseStore):
             self._get_event_cache.prefill((original_ev.event_id,), cache_entry)
 
         defer.returnValue(cache_entry)
+
+    @defer.inlineCallbacks
+    def have_events_in_timeline(self, event_ids):
+        """Given a list of event ids, check if we have already processed and
+        stored them as non outliers.
+        """
+        rows = yield self._simple_select_many_batch(
+            table="events",
+            retcols=("event_id",),
+            column="event_id",
+            iterable=list(event_ids),
+            keyvalues={"outlier": False},
+            desc="have_events_in_timeline",
+        )
+
+        defer.returnValue(set(r["event_id"] for r in rows))
+
+    @defer.inlineCallbacks
+    def have_seen_events(self, event_ids):
+        """Given a list of event ids, check if we have already processed them.
+
+        Args:
+            event_ids (iterable[str]):
+
+        Returns:
+            Deferred[set[str]]: The events we have already seen.
+        """
+        results = set()
+
+        def have_seen_events_txn(txn, chunk):
+            sql = (
+                "SELECT event_id FROM events as e WHERE e.event_id IN (%s)"
+                % (",".join("?" * len(chunk)), )
+            )
+            txn.execute(sql, chunk)
+            for (event_id, ) in txn:
+                results.add(event_id)
+
+        # break the input up into chunks of 100
+        input_iterator = iter(event_ids)
+        for chunk in iter(lambda: list(itertools.islice(input_iterator, 100)),
+                          []):
+            yield self.runInteraction(
+                "have_seen_events",
+                have_seen_events_txn,
+                chunk,
+            )
+        defer.returnValue(results)
+
+    def get_seen_events_with_rejections(self, event_ids):
+        """Given a list of event ids, check if we rejected them.
+
+        Args:
+            event_ids (list[str])
+
+        Returns:
+            Deferred[dict[str, str|None):
+                Has an entry for each event id we already have seen. Maps to
+                the rejected reason string if we rejected the event, else maps
+                to None.
+        """
+        if not event_ids:
+            return defer.succeed({})
+
+        def f(txn):
+            sql = (
+                "SELECT e.event_id, reason FROM events as e "
+                "LEFT JOIN rejections as r ON e.event_id = r.event_id "
+                "WHERE e.event_id = ?"
+            )
+
+            res = {}
+            for event_id in event_ids:
+                txn.execute(sql, (event_id,))
+                row = txn.fetchone()
+                if row:
+                    _, rejected = row
+                    res[event_id] = rejected
+
+            return res
+
+        return self.runInteraction("get_rejection_reasons", f)