fix(core,fuse): compact the search indexes and stop pruning after partial scans

FTS5 keeps deleted entries until their segments are merged, which with
these indexes is close to never. On one machine local_fts held 317 MiB for
40,000 paths, against 20 MiB rebuilt, and search ran four times slower. The
indexes are now compacted once a day after a local scan. A scan that hit the
entry cap or lost a batch no longer prunes the paths it never reached, which
the next scan used to add back (B217).

crates/pdfs-core/src/db/tests.rs

@@ -1429,6 +1429,77 @@ fn rescanning_an_unchanged_file_keeps_exactly_one_index_entry() {
     }
 }
 
+/// A scan that stopped early or lost a batch did not see every file, so it
+/// must not prune the ones it missed (`docs/BUGS.md` B217).
+#[test]
+fn a_partial_local_scan_prunes_nothing() {
+    let db = Db::open_in_memory().unwrap();
+    let g = db.local_begin_scan().unwrap();
+    db.local_upsert_batch(
+        g,
+        &[
+            local("/home/u/a.txt", "a.txt", false),
+            local("/home/u/b.txt", "b.txt", false),
+        ],
+    )
+    .unwrap();
+    assert_eq!(db.local_finish_scan(g, 1).unwrap(), 2);
+
+    let g = db.local_begin_scan().unwrap();
+    db.local_upsert_batch(g, &[local("/home/u/a.txt", "a.txt", false)])
+        .unwrap();
+    assert_eq!(db.local_finish_partial_scan(2).unwrap(), 2);
+    assert_eq!(db.search_local("b.txt", 10).unwrap().len(), 1);
+    assert_eq!(db.local_indexed_at().unwrap(), Some(2));
+}
+
+/// Deleted entries pile up in an FTS5 index until it is compacted. The daily
+/// compaction must leave both indexes whole and searchable, and must not run
+/// again the same day (`docs/BUGS.md` B217).
+#[test]
+fn compacting_the_search_indexes_keeps_them_whole_and_runs_once_a_day() {
+    let db = Db::open_in_memory().unwrap();
+    db.upsert_node(&folder("root", None, "My Files")).unwrap();
+    db.upsert_node(&file("f1", "root", "quarterly-report.pdf", 1))
+        .unwrap();
+    for round in 0..20 {
+        let g = db.local_begin_scan().unwrap();
+        let name = format!("churn-{round}.txt");
+        db.local_upsert_batch(
+            g,
+            &[
+                local("/home/u/keep.txt", "keep.txt", false),
+                local(&format!("/home/u/{name}"), &name, false),
+            ],
+        )
+        .unwrap();
+        // Far in the past, so no scan compacts on its own.
+        db.local_finish_scan(g, 0).unwrap();
+    }
+
+    let day = 24 * 60 * 60;
+    db.compact_search_indexes_if_due(day).unwrap();
+    {
+        let conn = db.conn.lock();
+        for table in ["local_fts", "nodes_fts"] {
+            conn.execute(
+                &format!("INSERT INTO {table}({table}) VALUES('integrity-check')"),
+                [],
+            )
+            .unwrap();
+        }
+    }
+    assert_eq!(db.search_local("keep", 10).unwrap().len(), 1);
+    assert_eq!(db.search_local("churn-19", 10).unwrap().len(), 1);
+    assert!(db.search_local("churn-3.txt", 10).unwrap().is_empty());
+    assert_eq!(db.search("quarterly", 10).unwrap().len(), 1);
+    let stamp = |db: &Db| db.state_str("fts_compacted_at").unwrap();
+    assert_eq!(stamp(&db).as_deref(), Some(day.to_string().as_str()));
+
+    db.compact_search_indexes_if_due(day + 60).unwrap();
+    assert_eq!(stamp(&db).as_deref(), Some(day.to_string().as_str()));
+}
+
 /// Queries below the trigram minimum still match, via the `LIKE` fallback.
 #[test]
 fn local_search_short_query_like_fallback() {