adriangb commented on code in PR #21714: URL: https://github.com/apache/datafusion/pull/21714#discussion_r4420905011
########## datafusion/sqllogictest/test_files/sql_extension_types.slt: ########## @@ -0,0 +1,136 @@ +# SPDX-License-Identifier: Apache-2.0 +# SPDX-FileCopyrightText: See the NOTICE file distributed with this work for additional information regarding copyright ownership +# SPDX-FileContributor: Licensed to the Apache Software Foundation (ASF) under one or more contributor license agreements + +# Tests for built-in mapping of SQL types to Arrow canonical extension types. +# See: https://arrow.apache.org/docs/format/CanonicalExtensions.html + +####### +# UUID Extension Type +####### + +# CAST to UUID preserves extension metadata +query ?T +SELECT CAST(X'00010203040506070809000102030506' AS UUID), + arrow_metadata(CAST(X'00010203040506070809000102030506' AS UUID), 'ARROW:extension:name'); +---- +00010203040506070809000102030506 arrow.uuid + +# Test table creation and various insertion methods (Null, Naked Binary, and Cast) +statement ok +CREATE TABLE test_uuid (value UUID); + +query TTT +DESCRIBE test_uuid; +---- +value FixedSizeBinary(16) YES + +# 1. NULL insertion +statement ok +INSERT INTO test_uuid VALUES (NULL); + +# 2. Exercises Implicit Promotion to UUID from binary +statement ok +INSERT INTO test_uuid VALUES (X'00010203040506070809000102030506'); + +# 3. Explicit CAST +statement ok +INSERT INTO test_uuid VALUES (CAST(X'00010203040506070809000102030507' AS UUID)); + +query T +SELECT arrow_metadata(value, 'ARROW:extension:name') FROM test_uuid; +---- +arrow.uuid +arrow.uuid +arrow.uuid + +# Batch of multiple values (some NULL) +statement ok +INSERT INTO test_uuid VALUES + (CAST(X'00010203040506070809000102030506' AS UUID)), + (NULL), + (CAST(X'00010203040506070809000102030507' AS UUID)); + +query I +SELECT COUNT(*) FROM test_uuid; +---- +6 + +statement ok +DROP TABLE test_uuid; + +####### +# JSON Extension Type +####### + +# CAST to JSON preserves extension metadata +query TT +SELECT CAST('{"a": 1}' AS JSON), + arrow_metadata(CAST('{"a": 1}' AS JSON), 'ARROW:extension:name'); +---- +{"a": 1} arrow.json + +statement ok +CREATE TABLE test_json (data JSON); + +query TTT +DESCRIBE test_json; +---- +data Utf8 YES + +# 1. NULL insertion +statement ok +INSERT INTO test_json VALUES (NULL); + +# 2. Bare string literal (Implicit Promotion) +statement ok +INSERT INTO test_json VALUES ('{"bare": "literal"}'); + +# 3. Explicit CAST +statement ok +INSERT INTO test_json VALUES (CAST('{"hello": "world"}' AS JSON)); + +query T +SELECT arrow_metadata(data, 'ARROW:extension:name') FROM test_json; +---- +arrow.json +arrow.json +arrow.json + +# Verify Type Decay: Concatenation results in a bare Utf8 string +query T +SELECT arrow_metadata(data || '{"extra": 1}', 'ARROW:extension:name') FROM test_json; +---- +NULL +NULL +NULL + +statement ok +DROP TABLE test_json; + +####### +# Error Handling +####### + +# Error: Cannot cast an integer to UUID +statement error Unsupported CAST from Int64 to FixedSizeBinary\(16\) +SELECT CAST(12345 AS UUID); + +# Error: Cannot cast a short string to UUID +statement error Unsupported CAST from Utf8 to FixedSizeBinary\(16\) +SELECT CAST('too-short' AS UUID); + +# This should trigger cast plan_err: "Cannot cast Utf8 to extension type at column 0" Review Comment: ```suggestion # This should trigger cast plan_err: "Types don't match and no valid cast exists" ``` -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
