binwiederhier 58dfef9aa8 Create both replay indexes in a 16 -> 17 migration ...

Instead of on every start: the schema is defined in one place again. Postgres gets
idx_message_topic_published_id (topic, published, id), SQLite idx_topic_published_time (topic,
published, time), both with IF NOT EXISTS (ntfy.sh got the Postgres one by hand). A downgrade to
v2.31.0 or earlier now needs the schema version set back by hand, as described in the release notes;
from v2.31.1 on, a newer schema only logs a warning (#2059).

2026-10-16 22:46:18 +00:00

..

cache_postgres_internal_test.go

Add the (topic, published, id) index, and split the replay by time

2026-10-16 21:08:31 +00:00

cache_postgres_schema.go

Create both replay indexes in a 16 -> 17 migration

2026-10-16 22:46:18 +00:00

cache_postgres_test.go

Create both replay indexes in a 16 -> 17 migration

2026-10-16 22:46:18 +00:00

cache_postgres.go

Create both replay indexes in a 16 -> 17 migration

2026-10-16 22:46:18 +00:00

cache_sqlite_internal_test.go

Check generic plans, and split the tail query the same way

2026-10-16 19:03:54 +00:00

cache_sqlite_schema.go

Create both replay indexes in a 16 -> 17 migration

2026-10-16 22:46:18 +00:00

cache_sqlite_test.go

Create both replay indexes in a 16 -> 17 migration

2026-10-16 22:46:18 +00:00

cache_sqlite.go

Create both replay indexes in a 16 -> 17 migration

2026-10-16 22:46:18 +00:00

cache_test.go

Handle since=<id> only through ResolveMarker and MessagesAfterMarker

2026-10-14 21:03:02 +00:00

cache.go

Explain the since=<id> replay query on MessagesAfterMarker

2026-10-15 09:21:28 +00:00

tail_test.go

Let the tail carry its own stop channel: Done and Close

2026-10-14 13:56:48 +00:00

tail.go

Order tail.go in read order: Next and its helpers first, Close last

2026-10-14 20:48:45 +00:00