sqlite-utils optimize command, .optimize() and .detect_fts() table methods

sqlite_utils/cli.py

@@ -37,3 +37,21 @@ def table_names(path, fts4, fts5):
 def vacuum(path):
     """Run VACUUM against the database"""
     sqlite_utils.Database(path).vacuum()
+
+
+@cli.command()
+@click.argument(
+    "path",
+    type=click.Path(exists=True, file_okay=True, dir_okay=False, allow_dash=False),
+    required=True,
+)
+@click.option("--no-vacuum", help="Don't run VACUUM", default=False, is_flag=True)
+def optimize(path, no_vacuum):
+    """Optimize all FTS tables and then run VACUUM - should shrink the database file"""
+    db = sqlite_utils.Database(path)
+    tables = db.table_names(fts4=True) + db.table_names(fts5=True)
+    with db.conn:
+        for table in tables:
+            db[table].optimize()
+    if not no_vacuum:
+        db.vacuum()

sqlite_utils/db.py

@@ -243,6 +243,40 @@ class Table:
         self.db.conn.executescript(sql)
         return self
 
+    def detect_fts(self):
+        "Detect if table has a corresponding FTS virtual table and return it"
+        rows = self.db.conn.execute(
+            """
+            SELECT name FROM sqlite_master
+                WHERE rootpage = 0
+                AND (
+                    sql LIKE '%VIRTUAL TABLE%USING FTS%content="{table}"%'
+                    OR (
+                        tbl_name = "{table}"
+                        AND sql LIKE '%VIRTUAL TABLE%USING FTS%'
+                    )
+                )
+        """.format(
+                table=self.name
+            )
+        ).fetchall()
+        if len(rows) == 0:
+            return None
+        else:
+            return rows[0][0]
+
+    def optimize(self):
+        fts_table = self.detect_fts()
+        if fts_table is not None:
+            self.db.conn.execute(
+                """
+                INSERT INTO [{table}] ([{table}]) VALUES ("optimize");
+            """.format(
+                    table=fts_table
+                )
+            )
+        return self
+
     def detect_column_types(self, records):
         all_column_types = {}
         for record in records: