Skip to content

fix(sync): verify stored messages before trusting history_complete #184

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
rnavarro wants to merge 4 commits into
base: main
Choose a base branch
from
Open

fix(sync): verify stored messages before trusting history_complete #184

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
0647ec0
fix(sync): don't trust history_complete when no messages are stored
rnavarro Aug 20, 2026
365cb32
fix(sync): always crawl full history when verifying a channel
rnavarro Aug 20, 2026
f123ed4
fix(sync): keep the verification marker restore off the crawl context
rnavarro Aug 20, 2026
93ea122
fix(sync): retire verified_empty once a channel holds messages
rnavarro Aug 20, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
10 changes: 10 additions & 0 deletions internal/store/query.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -59,6 +59,16 @@ func (s *Store) GetSyncState(ctx context.Context, scope string) (string, error)
return cursor, nil
}

// ChannelHasMessages reports whether any message rows are stored for the
// channel. It is an O(1) existence probe against idx_messages_channel_id and is
// safe to call once per channel per sync.
func (s *Store) ChannelHasMessages(ctx context.Context, channelID string) (bool, error) {
if channelID == "" {
return false, nil
}
return s.q.ChannelHasMessages(ctx, channelID)
}

func (s *Store) ChannelMessageBounds(ctx context.Context, channelID string) (string, string, error) {
row, err := s.q.ChannelMessageBounds(ctx, channelID)
if err != nil {
Expand Down
3 changes: 3 additions & 0 deletions internal/store/sqlc/queries.sql
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -14,6 +14,9 @@ on conflict(scope) do update set
delete from sync_state
where scope = ?;

-- name: ChannelHasMessages :one
select exists(select 1 from messages where channel_id = ? limit 1) as has_messages;

-- name: ChannelMessageBounds :one
select cast(coalesce(min(id), '') as text) as oldest_id,
cast(coalesce(max(id), '') as text) as newest_id
Expand Down
11 changes: 11 additions & 0 deletions internal/store/storedb/queries.sql.go
View file
Open in desktop

Some generated files are not rendered by default. Learn more about how customized files appear on GitHub.

14 changes: 14 additions & 0 deletions internal/syncer/channel_catalog_test.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -508,6 +508,13 @@ func TestSyncSkipsUnchangedThreadsWhenHistoryComplete(t *testing.T) {
}))
require.NoError(t, s.SetSyncState(ctx, channelLatestScope("t1"), "200"))
require.NoError(t, s.SetSyncState(ctx, channelHistoryCompleteScope("t1"), "1"))
// A genuinely complete channel has the messages to show for it. Without a
// stored row the channel is re-crawled to recover the missing history.
require.NoError(t, s.UpsertMessage(ctx, store.MessageRecord{
ID: "200", GuildID: "g1", ChannelID: "t1", ChannelName: "bug-report",
AuthorID: "u1", AuthorName: "User", CreatedAt: time.Now().UTC().Format(time.RFC3339Nano),
Content: "hello", NormalizedContent: "hello", RawJSON: `{}`,
}))

client := &fakeClient{
guilds: []*discordgo.UserGuild{{ID: "g1", Name: "Guild"}},
Expand Down Expand Up @@ -556,6 +563,13 @@ func TestSyncSkipsUnchangedTextChannelsWhenHistoryComplete(t *testing.T) {
}))
require.NoError(t, s.SetSyncState(ctx, channelLatestScope("c1"), "200"))
require.NoError(t, s.SetSyncState(ctx, channelHistoryCompleteScope("c1"), "1"))
// A genuinely complete channel has the messages to show for it. Without a
// stored row the channel is re-crawled to recover the missing history.
require.NoError(t, s.UpsertMessage(ctx, store.MessageRecord{
ID: "200", GuildID: "g1", ChannelID: "c1", ChannelName: "general",
AuthorID: "u1", AuthorName: "User", CreatedAt: time.Now().UTC().Format(time.RFC3339Nano),
Content: "hello", NormalizedContent: "hello", RawJSON: `{}`,
}))

client := &fakeClient{
guilds: []*discordgo.UserGuild{{ID: "g1", Name: "Guild"}},
Expand Down
Loading