From 02df6cdb000c8dbf739abda2af321a4f90d1b059 Mon Sep 17 00:00:00 2001 From: Daniel Baumann Date: Sun, 10 Dec 2023 11:45:55 +0100 Subject: Adding upstream version 20.1.0. Signed-off-by: Daniel Baumann --- tests/dialects/test_duckdb.py | 124 ++++++++++++++++++++++++++++++++---------- 1 file changed, 96 insertions(+), 28 deletions(-) (limited to 'tests/dialects/test_duckdb.py') diff --git a/tests/dialects/test_duckdb.py b/tests/dialects/test_duckdb.py index f9de953..687a807 100644 --- a/tests/dialects/test_duckdb.py +++ b/tests/dialects/test_duckdb.py @@ -1,4 +1,5 @@ from sqlglot import ErrorLevel, UnsupportedError, exp, parse_one, transpile +from sqlglot.helper import logger as helper_logger from tests.dialects.test_dialect import Validator @@ -71,7 +72,7 @@ class TestDuckDB(Validator): "SELECT UNNEST(ARRAY[1, 2, 3]), UNNEST(ARRAY[4, 5]), UNNEST(ARRAY[6])", write={ "bigquery": "SELECT IF(pos = pos_2, col, NULL) AS col, IF(pos = pos_3, col_2, NULL) AS col_2, IF(pos = pos_4, col_3, NULL) AS col_3 FROM UNNEST(GENERATE_ARRAY(0, GREATEST(ARRAY_LENGTH([1, 2, 3]), ARRAY_LENGTH([4, 5]), ARRAY_LENGTH([6])) - 1)) AS pos CROSS JOIN UNNEST([1, 2, 3]) AS col WITH OFFSET AS pos_2 CROSS JOIN UNNEST([4, 5]) AS col_2 WITH OFFSET AS pos_3 CROSS JOIN UNNEST([6]) AS col_3 WITH OFFSET AS pos_4 WHERE ((pos = pos_2 OR (pos > (ARRAY_LENGTH([1, 2, 3]) - 1) AND pos_2 = (ARRAY_LENGTH([1, 2, 3]) - 1))) AND (pos = pos_3 OR (pos > (ARRAY_LENGTH([4, 5]) - 1) AND pos_3 = (ARRAY_LENGTH([4, 5]) - 1)))) AND (pos = pos_4 OR (pos > (ARRAY_LENGTH([6]) - 1) AND pos_4 = (ARRAY_LENGTH([6]) - 1)))", - "presto": "SELECT IF(pos = pos_2, col) AS col, IF(pos = pos_3, col_2) AS col_2, IF(pos = pos_4, col_3) AS col_3 FROM UNNEST(SEQUENCE(1, GREATEST(CARDINALITY(ARRAY[1, 2, 3]), CARDINALITY(ARRAY[4, 5]), CARDINALITY(ARRAY[6])))) AS _u(pos) CROSS JOIN UNNEST(ARRAY[1, 2, 3]) WITH ORDINALITY AS _u_2(col, pos_2) CROSS JOIN UNNEST(ARRAY[4, 5]) WITH ORDINALITY AS _u_3(col_2, pos_3) CROSS JOIN UNNEST(ARRAY[6]) WITH ORDINALITY AS _u_4(col_3, pos_4) WHERE ((pos = pos_2 OR (pos > CARDINALITY(ARRAY[1, 2, 3]) AND pos_2 = CARDINALITY(ARRAY[1, 2, 3]))) AND (pos = pos_3 OR (pos > CARDINALITY(ARRAY[4, 5]) AND pos_3 = CARDINALITY(ARRAY[4, 5])))) AND (pos = pos_4 OR (pos > CARDINALITY(ARRAY[6]) AND pos_4 = CARDINALITY(ARRAY[6])))", + "presto": "SELECT IF(_u.pos = _u_2.pos_2, _u_2.col) AS col, IF(_u.pos = _u_3.pos_3, _u_3.col_2) AS col_2, IF(_u.pos = _u_4.pos_4, _u_4.col_3) AS col_3 FROM UNNEST(SEQUENCE(1, GREATEST(CARDINALITY(ARRAY[1, 2, 3]), CARDINALITY(ARRAY[4, 5]), CARDINALITY(ARRAY[6])))) AS _u(pos) CROSS JOIN UNNEST(ARRAY[1, 2, 3]) WITH ORDINALITY AS _u_2(col, pos_2) CROSS JOIN UNNEST(ARRAY[4, 5]) WITH ORDINALITY AS _u_3(col_2, pos_3) CROSS JOIN UNNEST(ARRAY[6]) WITH ORDINALITY AS _u_4(col_3, pos_4) WHERE ((_u.pos = _u_2.pos_2 OR (_u.pos > CARDINALITY(ARRAY[1, 2, 3]) AND _u_2.pos_2 = CARDINALITY(ARRAY[1, 2, 3]))) AND (_u.pos = _u_3.pos_3 OR (_u.pos > CARDINALITY(ARRAY[4, 5]) AND _u_3.pos_3 = CARDINALITY(ARRAY[4, 5])))) AND (_u.pos = _u_4.pos_4 OR (_u.pos > CARDINALITY(ARRAY[6]) AND _u_4.pos_4 = CARDINALITY(ARRAY[6])))", }, ) @@ -79,7 +80,7 @@ class TestDuckDB(Validator): "SELECT UNNEST(ARRAY[1, 2, 3]), UNNEST(ARRAY[4, 5]), UNNEST(ARRAY[6]) FROM x", write={ "bigquery": "SELECT IF(pos = pos_2, col, NULL) AS col, IF(pos = pos_3, col_2, NULL) AS col_2, IF(pos = pos_4, col_3, NULL) AS col_3 FROM x, UNNEST(GENERATE_ARRAY(0, GREATEST(ARRAY_LENGTH([1, 2, 3]), ARRAY_LENGTH([4, 5]), ARRAY_LENGTH([6])) - 1)) AS pos CROSS JOIN UNNEST([1, 2, 3]) AS col WITH OFFSET AS pos_2 CROSS JOIN UNNEST([4, 5]) AS col_2 WITH OFFSET AS pos_3 CROSS JOIN UNNEST([6]) AS col_3 WITH OFFSET AS pos_4 WHERE ((pos = pos_2 OR (pos > (ARRAY_LENGTH([1, 2, 3]) - 1) AND pos_2 = (ARRAY_LENGTH([1, 2, 3]) - 1))) AND (pos = pos_3 OR (pos > (ARRAY_LENGTH([4, 5]) - 1) AND pos_3 = (ARRAY_LENGTH([4, 5]) - 1)))) AND (pos = pos_4 OR (pos > (ARRAY_LENGTH([6]) - 1) AND pos_4 = (ARRAY_LENGTH([6]) - 1)))", - "presto": "SELECT IF(pos = pos_2, col) AS col, IF(pos = pos_3, col_2) AS col_2, IF(pos = pos_4, col_3) AS col_3 FROM x, UNNEST(SEQUENCE(1, GREATEST(CARDINALITY(ARRAY[1, 2, 3]), CARDINALITY(ARRAY[4, 5]), CARDINALITY(ARRAY[6])))) AS _u(pos) CROSS JOIN UNNEST(ARRAY[1, 2, 3]) WITH ORDINALITY AS _u_2(col, pos_2) CROSS JOIN UNNEST(ARRAY[4, 5]) WITH ORDINALITY AS _u_3(col_2, pos_3) CROSS JOIN UNNEST(ARRAY[6]) WITH ORDINALITY AS _u_4(col_3, pos_4) WHERE ((pos = pos_2 OR (pos > CARDINALITY(ARRAY[1, 2, 3]) AND pos_2 = CARDINALITY(ARRAY[1, 2, 3]))) AND (pos = pos_3 OR (pos > CARDINALITY(ARRAY[4, 5]) AND pos_3 = CARDINALITY(ARRAY[4, 5])))) AND (pos = pos_4 OR (pos > CARDINALITY(ARRAY[6]) AND pos_4 = CARDINALITY(ARRAY[6])))", + "presto": "SELECT IF(_u.pos = _u_2.pos_2, _u_2.col) AS col, IF(_u.pos = _u_3.pos_3, _u_3.col_2) AS col_2, IF(_u.pos = _u_4.pos_4, _u_4.col_3) AS col_3 FROM x, UNNEST(SEQUENCE(1, GREATEST(CARDINALITY(ARRAY[1, 2, 3]), CARDINALITY(ARRAY[4, 5]), CARDINALITY(ARRAY[6])))) AS _u(pos) CROSS JOIN UNNEST(ARRAY[1, 2, 3]) WITH ORDINALITY AS _u_2(col, pos_2) CROSS JOIN UNNEST(ARRAY[4, 5]) WITH ORDINALITY AS _u_3(col_2, pos_3) CROSS JOIN UNNEST(ARRAY[6]) WITH ORDINALITY AS _u_4(col_3, pos_4) WHERE ((_u.pos = _u_2.pos_2 OR (_u.pos > CARDINALITY(ARRAY[1, 2, 3]) AND _u_2.pos_2 = CARDINALITY(ARRAY[1, 2, 3]))) AND (_u.pos = _u_3.pos_3 OR (_u.pos > CARDINALITY(ARRAY[4, 5]) AND _u_3.pos_3 = CARDINALITY(ARRAY[4, 5])))) AND (_u.pos = _u_4.pos_4 OR (_u.pos > CARDINALITY(ARRAY[6]) AND _u_4.pos_4 = CARDINALITY(ARRAY[6])))", }, ) self.validate_all( @@ -96,7 +97,6 @@ class TestDuckDB(Validator): ) self.validate_identity("SELECT i FROM RANGE(5) AS _(i) ORDER BY i ASC") - self.validate_identity("[x.STRING_SPLIT(' ')[1] FOR x IN ['1', '2', 3] IF x.CONTAINS('1')]") self.validate_identity("INSERT INTO x BY NAME SELECT 1 AS y") self.validate_identity("SELECT 1 AS x UNION ALL BY NAME SELECT 2 AS x") self.validate_identity("SELECT SUM(x) FILTER (x = 1)", "SELECT SUM(x) FILTER(WHERE x = 1)") @@ -109,6 +109,10 @@ class TestDuckDB(Validator): parse_one("a // b", read="duckdb").assert_is(exp.IntDiv).sql(dialect="duckdb"), "a // b" ) + self.validate_identity("SELECT EPOCH_MS(10) AS t") + self.validate_identity("SELECT MAKE_TIMESTAMP(10) AS t") + self.validate_identity("SELECT TO_TIMESTAMP(10) AS t") + self.validate_identity("SELECT UNNEST(column, recursive := TRUE) FROM table") self.validate_identity("VAR_POP(a)") self.validate_identity("SELECT * FROM foo ASOF LEFT JOIN bar ON a = b") self.validate_identity("PIVOT Cities ON Year USING SUM(Population)") @@ -151,11 +155,18 @@ class TestDuckDB(Validator): self.validate_all("0x1010", write={"": "0 AS x1010"}) self.validate_all("x ~ y", write={"duckdb": "REGEXP_MATCHES(x, y)"}) self.validate_all("SELECT * FROM 'x.y'", write={"duckdb": 'SELECT * FROM "x.y"'}) + self.validate_all( + "SELECT * FROM produce PIVOT(SUM(sales) FOR quarter IN ('Q1', 'Q2'))", + read={ + "duckdb": "SELECT * FROM produce PIVOT(SUM(sales) FOR quarter IN ('Q1', 'Q2'))", + "snowflake": "SELECT * FROM produce PIVOT(SUM(produce.sales) FOR produce.quarter IN ('Q1', 'Q2'))", + }, + ) self.validate_all( "SELECT UNNEST([1, 2, 3])", write={ "duckdb": "SELECT UNNEST([1, 2, 3])", - "snowflake": "SELECT IFF(pos = pos_2, col, NULL) AS col FROM (SELECT value FROM TABLE(FLATTEN(INPUT => ARRAY_GENERATE_RANGE(0, (GREATEST(ARRAY_SIZE([1, 2, 3])) - 1) + 1)))) AS _u(pos) CROSS JOIN (SELECT value, index FROM TABLE(FLATTEN(INPUT => [1, 2, 3]))) AS _u_2(col, pos_2) WHERE pos = pos_2 OR (pos > (ARRAY_SIZE([1, 2, 3]) - 1) AND pos_2 = (ARRAY_SIZE([1, 2, 3]) - 1))", + "snowflake": "SELECT IFF(_u.pos = _u_2.pos_2, _u_2.col, NULL) AS col FROM TABLE(FLATTEN(INPUT => ARRAY_GENERATE_RANGE(0, (GREATEST(ARRAY_SIZE([1, 2, 3])) - 1) + 1))) AS _u(seq, key, path, index, pos, this) CROSS JOIN TABLE(FLATTEN(INPUT => [1, 2, 3])) AS _u_2(seq, key, path, pos_2, col, this) WHERE _u.pos = _u_2.pos_2 OR (_u.pos > (ARRAY_SIZE([1, 2, 3]) - 1) AND _u_2.pos_2 = (ARRAY_SIZE([1, 2, 3]) - 1))", }, ) self.validate_all( @@ -355,14 +366,14 @@ class TestDuckDB(Validator): "STRUCT_PACK(x := 1, y := '2')", write={ "duckdb": "{'x': 1, 'y': '2'}", - "spark": "STRUCT(x = 1, y = '2')", + "spark": "STRUCT(1 AS x, '2' AS y)", }, ) self.validate_all( "STRUCT_PACK(key1 := 'value1', key2 := 42)", write={ "duckdb": "{'key1': 'value1', 'key2': 42}", - "spark": "STRUCT(key1 = 'value1', key2 = 42)", + "spark": "STRUCT('value1' AS key1, 42 AS key2)", }, ) self.validate_all( @@ -440,6 +451,16 @@ class TestDuckDB(Validator): "hive": "SELECT DATE_ADD(TO_DATE(x), 1)", }, ) + self.validate_all( + "SELECT CAST('2018-01-01 00:00:00' AS DATE) + INTERVAL 3 DAY", + read={ + "hive": "SELECT DATE_ADD('2018-01-01 00:00:00', 3)", + }, + write={ + "duckdb": "SELECT CAST('2018-01-01 00:00:00' AS DATE) + INTERVAL '3' DAY", + "hive": "SELECT CAST('2018-01-01 00:00:00' AS DATE) + INTERVAL '3' DAY", + }, + ) self.validate_all( "SELECT CAST('2020-05-06' AS DATE) - INTERVAL 5 DAY", read={"bigquery": "SELECT DATE_SUB(CAST('2020-05-06' AS DATE), INTERVAL 5 DAY)"}, @@ -483,6 +504,35 @@ class TestDuckDB(Validator): self.validate_identity("SELECT ISNAN(x)") + def test_array_index(self): + with self.assertLogs(helper_logger) as cm: + self.validate_all( + "SELECT some_arr[1] AS first FROM blah", + read={ + "bigquery": "SELECT some_arr[0] AS first FROM blah", + }, + write={ + "bigquery": "SELECT some_arr[0] AS first FROM blah", + "duckdb": "SELECT some_arr[1] AS first FROM blah", + "presto": "SELECT some_arr[1] AS first FROM blah", + }, + ) + self.validate_identity( + "[x.STRING_SPLIT(' ')[1] FOR x IN ['1', '2', 3] IF x.CONTAINS('1')]" + ) + + self.assertEqual( + cm.output, + [ + "WARNING:sqlglot:Applying array index offset (-1)", + "WARNING:sqlglot:Applying array index offset (1)", + "WARNING:sqlglot:Applying array index offset (1)", + "WARNING:sqlglot:Applying array index offset (1)", + "WARNING:sqlglot:Applying array index offset (-1)", + "WARNING:sqlglot:Applying array index offset (1)", + ], + ) + def test_time(self): self.validate_identity("SELECT CURRENT_DATE") self.validate_identity("SELECT CURRENT_TIMESTAMP") @@ -533,16 +583,16 @@ class TestDuckDB(Validator): self.validate_all( "EPOCH_MS(x)", write={ - "bigquery": "UNIX_TO_TIME(x / 1000)", - "duckdb": "TO_TIMESTAMP(x / 1000)", - "presto": "FROM_UNIXTIME(x / 1000)", - "spark": "CAST(FROM_UNIXTIME(x / 1000) AS TIMESTAMP)", + "bigquery": "TIMESTAMP_MILLIS(x)", + "duckdb": "EPOCH_MS(x)", + "presto": "FROM_UNIXTIME(CAST(x AS DOUBLE) / 1000)", + "spark": "TIMESTAMP_MILLIS(x)", }, ) self.validate_all( "STRFTIME(x, '%y-%-m-%S')", write={ - "bigquery": "TIME_TO_STR(x, '%y-%-m-%S')", + "bigquery": "FORMAT_DATE('%y-%-m-%S', x)", "duckdb": "STRFTIME(x, '%y-%-m-%S')", "postgres": "TO_CHAR(x, 'YY-FMMM-SS')", "presto": "DATE_FORMAT(x, '%y-%c-%s')", @@ -552,6 +602,7 @@ class TestDuckDB(Validator): self.validate_all( "STRFTIME(x, '%Y-%m-%d %H:%M:%S')", write={ + "bigquery": "FORMAT_DATE('%Y-%m-%d %H:%M:%S', x)", "duckdb": "STRFTIME(x, '%Y-%m-%d %H:%M:%S')", "presto": "DATE_FORMAT(x, '%Y-%m-%d %T')", "hive": "DATE_FORMAT(x, 'yyyy-MM-dd HH:mm:ss')", @@ -570,7 +621,7 @@ class TestDuckDB(Validator): self.validate_all( "TO_TIMESTAMP(x)", write={ - "bigquery": "UNIX_TO_TIME(x)", + "bigquery": "TIMESTAMP_SECONDS(x)", "duckdb": "TO_TIMESTAMP(x)", "presto": "FROM_UNIXTIME(x)", "hive": "FROM_UNIXTIME(x)", @@ -651,22 +702,25 @@ class TestDuckDB(Validator): "CAST(ROW(1, ROW(1)) AS STRUCT(number BIGINT, row STRUCT(number BIGINT)))" ) - self.validate_all("CAST(x AS NUMERIC(1, 2))", write={"duckdb": "CAST(x AS DECIMAL(1, 2))"}) - self.validate_all("CAST(x AS HUGEINT)", write={"duckdb": "CAST(x AS INT128)"}) - self.validate_all("CAST(x AS CHAR)", write={"duckdb": "CAST(x AS TEXT)"}) - self.validate_all("CAST(x AS BPCHAR)", write={"duckdb": "CAST(x AS TEXT)"}) - self.validate_all("CAST(x AS STRING)", write={"duckdb": "CAST(x AS TEXT)"}) - self.validate_all("CAST(x AS INT1)", write={"duckdb": "CAST(x AS TINYINT)"}) - self.validate_all("CAST(x AS FLOAT4)", write={"duckdb": "CAST(x AS REAL)"}) - self.validate_all("CAST(x AS FLOAT)", write={"duckdb": "CAST(x AS REAL)"}) - self.validate_all("CAST(x AS INT4)", write={"duckdb": "CAST(x AS INT)"}) - self.validate_all("CAST(x AS INTEGER)", write={"duckdb": "CAST(x AS INT)"}) - self.validate_all("CAST(x AS SIGNED)", write={"duckdb": "CAST(x AS INT)"}) - self.validate_all("CAST(x AS BLOB)", write={"duckdb": "CAST(x AS BLOB)"}) - self.validate_all("CAST(x AS BYTEA)", write={"duckdb": "CAST(x AS BLOB)"}) - self.validate_all("CAST(x AS BINARY)", write={"duckdb": "CAST(x AS BLOB)"}) - self.validate_all("CAST(x AS VARBINARY)", write={"duckdb": "CAST(x AS BLOB)"}) - self.validate_all("CAST(x AS LOGICAL)", write={"duckdb": "CAST(x AS BOOLEAN)"}) + self.validate_identity("CAST(x AS INT64)", "CAST(x AS BIGINT)") + self.validate_identity("CAST(x AS INT32)", "CAST(x AS INT)") + self.validate_identity("CAST(x AS INT16)", "CAST(x AS SMALLINT)") + self.validate_identity("CAST(x AS NUMERIC(1, 2))", "CAST(x AS DECIMAL(1, 2))") + self.validate_identity("CAST(x AS HUGEINT)", "CAST(x AS INT128)") + self.validate_identity("CAST(x AS CHAR)", "CAST(x AS TEXT)") + self.validate_identity("CAST(x AS BPCHAR)", "CAST(x AS TEXT)") + self.validate_identity("CAST(x AS STRING)", "CAST(x AS TEXT)") + self.validate_identity("CAST(x AS INT1)", "CAST(x AS TINYINT)") + self.validate_identity("CAST(x AS FLOAT4)", "CAST(x AS REAL)") + self.validate_identity("CAST(x AS FLOAT)", "CAST(x AS REAL)") + self.validate_identity("CAST(x AS INT4)", "CAST(x AS INT)") + self.validate_identity("CAST(x AS INTEGER)", "CAST(x AS INT)") + self.validate_identity("CAST(x AS SIGNED)", "CAST(x AS INT)") + self.validate_identity("CAST(x AS BLOB)", "CAST(x AS BLOB)") + self.validate_identity("CAST(x AS BYTEA)", "CAST(x AS BLOB)") + self.validate_identity("CAST(x AS BINARY)", "CAST(x AS BLOB)") + self.validate_identity("CAST(x AS VARBINARY)", "CAST(x AS BLOB)") + self.validate_identity("CAST(x AS LOGICAL)", "CAST(x AS BOOLEAN)") self.validate_all( "CAST(x AS NUMERIC)", write={ @@ -799,3 +853,17 @@ class TestDuckDB(Validator): "duckdb": "SELECT CAST(w AS TIMESTAMP_S), CAST(x AS TIMESTAMP_MS), CAST(y AS TIMESTAMP), CAST(z AS TIMESTAMP_NS)", }, ) + + def test_isnan(self): + self.validate_all( + "ISNAN(x)", + read={"bigquery": "IS_NAN(x)"}, + write={"bigquery": "IS_NAN(x)", "duckdb": "ISNAN(x)"}, + ) + + def test_isinf(self): + self.validate_all( + "ISINF(x)", + read={"bigquery": "IS_INF(x)"}, + write={"bigquery": "IS_INF(x)", "duckdb": "ISINF(x)"}, + ) -- cgit v1.2.3