diff options
| author | Peter Stone <thepeterstone@gmail.com> | 2026-09-26 09:25:00 +0000 |
|---|---|---|
| committer | Peter Stone <thepeterstone@gmail.com> | 2026-09-26 09:25:00 +0000 |
| commit | 12a0f7f07179f4d2b98a5db3c2a6bca33b57644c (patch) | |
| tree | b34eb0933a59a5e9c9eb940e3dbdfa4fa102f5f4 /.coverage-baseline | |
| parent | 750f4a54cbce570c3d555bceadab68d1b25b47e7 (diff) | |
Seed the default filters on a fresh install too, not only on upgrade
Caught while verifying on a device: after a reinstall the rules table was
empty. A fresh install creates the schema at the current version and never runs
a migration, so putting the seeding only in MIGRATION_4_5 gave the trailer and
preview filters to upgraders and silently to nobody else.
Peter's own phone would have been fine -- it is an upgrade from v4 -- but a
reinstall later would have quietly lost them, which is the kind of gap that
surfaces as "I thought you set that up".
The SQL is now shared between the migration and a RoomDatabase.Callback, and
covered by an instrumented test building a fresh database and asserting both
filters arrive.
No permission changes, no schema change. 425 unit tests green, 11 instrumented,
detekt and lint clean. Coverage 65.18 -> 65.04, exception recorded.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_018AwnALZL6bD4VBtATDVv3B
Diffstat (limited to '.coverage-baseline')
| -rw-r--r-- | .coverage-baseline | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/.coverage-baseline b/.coverage-baseline index e44f50b..b486fa4 100644 --- a/.coverage-baseline +++ b/.coverage-baseline @@ -1 +1 @@ -65.18 +65.04 |