--analyze option for create-index, insert, update commands, closes #379, closes #365

Author: simonw

docs/cli.rst

@@ -761,6 +761,8 @@ You can delete all the existing rows in the table before inserting the new recor
 
     $ sqlite-utils insert dogs.db dogs dogs.json --truncate
 
+You can add the ``--analyze`` option to run ``ANALYZE`` against the table after the rows have been inserted.
+
 .. _cli_inserting_data_binary:
 
 Inserting binary data
@@ -1697,6 +1699,8 @@ This will create an index on that table on ``(col1, col2 desc, col3)``.
 
 If your column names are already prefixed with a hyphen you'll need to manually execute a ``CREATE INDEX`` SQL statement to add indexes to them rather than using this tool.
 
+Add the ``--analyze`` option to run ``ANALYZE`` against the index after it has been created.
+
 .. _cli_fts:
 
 Configuring full-text search
@@ -1817,6 +1821,8 @@ You can run it against specific tables, or against specific named indexes, by pa
 
     $ sqlite-utils analyze mydb.db mytable idx_mytable_name
 
+You can also run ``ANALYZE`` as part of another command using the ``--analyze`` option. This is supported by the ``create-index``, ``insert`` and ``upsert`` commands.
+
 .. _cli_vacuum:
 
 Vacuum

sqlite_utils/cli.py

@@ -490,8 +490,15 @@ def index_foreign_keys(path, load_extension):
     default=False,
     is_flag=True,
 )
+@click.option(
+    "--analyze",
+    help="Run ANALYZE after creating the index",
+    is_flag=True,
+)
 @load_extension_option
-def create_index(path, table, column, name, unique, if_not_exists, load_extension):
+def create_index(
+    path, table, column, name, unique, if_not_exists, analyze, load_extension
+):
     """
     Add an index to the specified table covering the specified columns.
     Use "sqlite-utils create-index mydb -- -column" to specify descending
@@ -506,7 +513,11 @@ def create_index(path, table, column, name, unique, if_not_exists, load_extensio
             col = DescIndex(col[1:])
         columns.append(col)
     db[table].create_index(
-        columns, index_name=name, unique=unique, if_not_exists=if_not_exists
+        columns,
+        index_name=name,
+        unique=unique,
+        if_not_exists=if_not_exists,
+        analyze=analyze,
     )
 
 
@@ -726,6 +737,11 @@ def insert_upsert_options(fn):
                 envvar="SQLITE_UTILS_DETECT_TYPES",
                 help="Detect types for columns in CSV/TSV data",
             ),
+            click.option(
+                "--analyze",
+                is_flag=True,
+                help="Run ANALYZE at the end of this operation",
+            ),
             load_extension_option,
             click.option("--silent", is_flag=True, help="Do not show progress bar"),
         )
@@ -761,6 +777,7 @@ def insert_upsert_implementation(
     default=None,
     encoding=None,
     detect_types=None,
+    analyze=False,
     load_extension=None,
     silent=False,
 ):
@@ -853,7 +870,12 @@ def insert_upsert_implementation(
         else:
             docs = (fn(doc) or doc for doc in docs)
 
-    extra_kwargs = {"ignore": ignore, "replace": replace, "truncate": truncate}
+    extra_kwargs = {
+        "ignore": ignore,
+        "replace": replace,
+        "truncate": truncate,
+        "analyze": analyze,
+    }
     if not_null:
         extra_kwargs["not_null"] = set(not_null)
     if default:
@@ -950,6 +972,7 @@ def insert(
     alter,
     encoding,
     detect_types,
+    analyze,
     load_extension,
     silent,
     ignore,
@@ -1005,6 +1028,7 @@ def insert(
             truncate=truncate,
             encoding=encoding,
             detect_types=detect_types,
+            analyze=analyze,
             load_extension=load_extension,
             silent=silent,
             not_null=not_null,
@@ -1039,6 +1063,7 @@ def upsert(
     default,
     encoding,
     detect_types,
+    analyze,
     load_extension,
     silent,
 ):
@@ -1072,6 +1097,7 @@ def upsert(
             default=default,
             encoding=encoding,
             detect_types=detect_types,
+            analyze=analyze,
             load_extension=load_extension,
             silent=silent,
         )

tests/test_cli.py

@@ -224,6 +224,17 @@ def test_create_index(db_path):
     )
 
 
+def test_create_index_analyze(db_path):
+    db = Database(db_path)
+    assert "sqlite_stat1" not in db.table_names()
+    assert [] == db["Gosh"].indexes
+    result = CliRunner().invoke(
+        cli.cli, ["create-index", db_path, "Gosh", "c1", "--analyze"]
+    )
+    assert result.exit_code == 0
+    assert "sqlite_stat1" in db.table_names()
+
+
 def test_create_index_desc(db_path):
     db = Database(db_path)
     assert [] == db["Gosh"].indexes
@@ -889,6 +900,20 @@ def test_upsert(db_path, tmpdir):
     ]
 
 
+def test_upsert_analyze(db_path, tmpdir):
+    db = Database(db_path)
+    db["rows"].insert({"id": 1, "foo": "x", "n": 3}, pk="id")
+    db["rows"].create_index(["n"])
+    assert "sqlite_stat1" not in db.table_names()
+    result = CliRunner().invoke(
+        cli.cli,
+        ["upsert", db_path, "rows", "-", "--nl", "--analyze", "--pk", "id"],
+        input='{"id": 2, "foo": "bar", "n": 1}',
+    )
+    assert 0 == result.exit_code, result.output
+    assert "sqlite_stat1" in db.table_names()
+
+
 def test_upsert_flatten(tmpdir):
     db_path = str(tmpdir / "flat.db")
     db = Database(db_path)

tests/test_cli_insert.py

@@ -327,6 +327,20 @@ def test_insert_alter(db_path, tmpdir):
     ] == list(db.query("select foo, n, baz from from_json_nl"))
 
 
+def test_insert_analyze(db_path):
+    db = Database(db_path)
+    db["rows"].insert({"foo": "x", "n": 3})
+    db["rows"].create_index(["n"])
+    assert "sqlite_stat1" not in db.table_names()
+    result = CliRunner().invoke(
+        cli.cli,
+        ["insert", db_path, "rows", "-", "--nl", "--analyze"],
+        input='{"foo": "bar", "n": 1}\n{"foo": "baz", "n": 2}',
+    )
+    assert 0 == result.exit_code, result.output
+    assert "sqlite_stat1" in db.table_names()
+
+
 def test_insert_lines(db_path):
     result = CliRunner().invoke(
         cli.cli,