Matrix-Mirrors/anonymousland-synapse
1
0
Fork 0
mirror of https://git.anonymousland.org/anonymousland/synapse.git synced 2025-05-04 17:24:59 -04:00
Code Issues Projects Releases Packages Wiki Activity

Support filtering by relations per MSC3440 (#11236)

Browse source 
Adds experimental support for `relation_types` and `relation_senders`
fields for filters.
This commit is contained in:
Patrick Cloke 2021-11-09 08:10:58 -05:00 • committed by GitHub
parent 4b3e30c276
commit a19d01c3d9
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
15 changed files with 680 additions and 110 deletions
Download patch file Download diff file Expand all files Collapse all files

86
synapse/storage/databases/main/stream.py
View file

@ -272,31 +272,37 @@ def filter_to_clause(event_filter: Optional[Filter]) -> Tuple[str, List[str]]:
args = []
if event_filter.types:
clauses.append("(%s)" % " OR ".join("type = ?" for _ in event_filter.types))
clauses.append(
"(%s)" % " OR ".join("event.type = ?" for _ in event_filter.types)
)
args.extend(event_filter.types)
for typ in event_filter.not_types:
clauses.append("type != ?")
clauses.append("event.type != ?")
args.append(typ)
if event_filter.senders:
clauses.append("(%s)" % " OR ".join("sender = ?" for _ in event_filter.senders))
clauses.append(
"(%s)" % " OR ".join("event.sender = ?" for _ in event_filter.senders)
)
args.extend(event_filter.senders)
for sender in event_filter.not_senders:
clauses.append("sender != ?")
clauses.append("event.sender != ?")
args.append(sender)
if event_filter.rooms:
clauses.append("(%s)" % " OR ".join("room_id = ?" for _ in event_filter.rooms))
clauses.append(
"(%s)" % " OR ".join("event.room_id = ?" for _ in event_filter.rooms)
)
args.extend(event_filter.rooms)
for room_id in event_filter.not_rooms:
clauses.append("room_id != ?")
clauses.append("event.room_id != ?")
args.append(room_id)
if event_filter.contains_url:
clauses.append("contains_url = ?")
clauses.append("event.contains_url = ?")
args.append(event_filter.contains_url)
# We're only applying the "labels" filter on the database query, because applying the
@ -307,6 +313,23 @@ def filter_to_clause(event_filter: Optional[Filter]) -> Tuple[str, List[str]]:
clauses.append("(%s)" % " OR ".join("label = ?" for _ in event_filter.labels))
args.extend(event_filter.labels)
# Filter on relation_senders / relation types from the joined tables.
if event_filter.relation_senders:
clauses.append(
"(%s)"
% " OR ".join(
"related_event.sender = ?" for _ in event_filter.relation_senders
)
)
args.extend(event_filter.relation_senders)
if event_filter.relation_types:
clauses.append(
"(%s)"
% " OR ".join("relation_type = ?" for _ in event_filter.relation_types)
)
args.extend(event_filter.relation_types)
return " AND ".join(clauses), args
@ -1116,7 +1139,7 @@ class StreamWorkerStore(EventsWorkerStore, SQLBaseStore, metaclass=abc.ABCMeta):
bounds = generate_pagination_where_clause(
direction=direction,
column_names=("topological_ordering", "stream_ordering"),
column_names=("event.topological_ordering", "event.stream_ordering"),
from_token=from_bound,
to_token=to_bound,
engine=self.database_engine,
@ -1133,32 +1156,51 @@ class StreamWorkerStore(EventsWorkerStore, SQLBaseStore, metaclass=abc.ABCMeta):
select_keywords = "SELECT"
join_clause = ""
# Using DISTINCT in this SELECT query is quite expensive, because it
# requires the engine to sort on the entire (not limited) result set,
# i.e. the entire events table. Only use it in scenarios that could result
# in the same event ID occurring multiple times in the results.
needs_distinct = False
if event_filter and event_filter.labels:
# If we're not filtering on a label, then joining on event_labels will
# return as many row for a single event as the number of labels it has. To
# avoid this, only join if we're filtering on at least one label.
join_clause = """
join_clause += """
LEFT JOIN event_labels
USING (event_id, room_id, topological_ordering)
"""
if len(event_filter.labels) > 1:
# Using DISTINCT in this SELECT query is quite expensive, because it
# requires the engine to sort on the entire (not limited) result set,
# i.e. the entire events table. We only need to use it when we're
# filtering on more than two labels, because that's the only scenario
# in which we can possibly to get multiple times the same event ID in
# the results.
select_keywords += "DISTINCT"
# Multiple labels could cause the same event to appear multiple times.
needs_distinct = True
# If there is a filter on relation_senders and relation_types join to the
# relations table.
if event_filter and (
event_filter.relation_senders or event_filter.relation_types
):
# Filtering by relations could cause the same event to appear multiple
# times (since there's no limit on the number of relations to an event).
needs_distinct = True
join_clause += """
LEFT JOIN event_relations AS relation ON (event.event_id = relation.relates_to_id)
"""
if event_filter.relation_senders:
join_clause += """
LEFT JOIN events AS related_event ON (relation.event_id = related_event.event_id)
"""
if needs_distinct:
select_keywords += " DISTINCT"
sql = """
%(select_keywords)s
event_id, instance_name,
topological_ordering, stream_ordering
FROM events
event.event_id, event.instance_name,
event.topological_ordering, event.stream_ordering
FROM events AS event
%(join_clause)s
WHERE outlier = ? AND room_id = ? AND %(bounds)s
ORDER BY topological_ordering %(order)s,
stream_ordering %(order)s LIMIT ?
WHERE event.outlier = ? AND event.room_id = ? AND %(bounds)s
ORDER BY event.topological_ordering %(order)s,
event.stream_ordering %(order)s LIMIT ?
""" % {
"select_keywords": select_keywords,
"join_clause": join_clause,