summaryrefslogtreecommitdiffstats
path: root/tests/dialects/test_spark.py
blob: 7c8ca1bbe2fd54046c15d4d01a7703d7388d8d27 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
from tests.dialects.test_dialect import Validator


class TestSpark(Validator):
    dialect = "spark"

    def test_ddl(self):
        self.validate_identity("CREATE TABLE foo (col VARCHAR(50))")
        self.validate_identity("CREATE TABLE foo (col STRUCT<struct_col_a: VARCHAR((50))>)")

        self.validate_all(
            "CREATE TABLE db.example_table (col_a struct<struct_col_a:int, struct_col_b:string>)",
            write={
                "duckdb": "CREATE TABLE db.example_table (col_a STRUCT(struct_col_a INT, struct_col_b TEXT))",
                "presto": "CREATE TABLE db.example_table (col_a ROW(struct_col_a INTEGER, struct_col_b VARCHAR))",
                "hive": "CREATE TABLE db.example_table (col_a STRUCT<struct_col_a INT, struct_col_b STRING>)",
                "spark": "CREATE TABLE db.example_table (col_a STRUCT<struct_col_a: INT, struct_col_b: STRING>)",
            },
        )
        self.validate_all(
            "CREATE TABLE db.example_table (col_a struct<struct_col_a:int, struct_col_b:struct<nested_col_a:string, nested_col_b:string>>)",
            write={
                "bigquery": "CREATE TABLE db.example_table (col_a STRUCT<struct_col_a INT64, struct_col_b STRUCT<nested_col_a STRING, nested_col_b STRING>>)",
                "duckdb": "CREATE TABLE db.example_table (col_a STRUCT(struct_col_a INT, struct_col_b STRUCT(nested_col_a TEXT, nested_col_b TEXT)))",
                "presto": "CREATE TABLE db.example_table (col_a ROW(struct_col_a INTEGER, struct_col_b ROW(nested_col_a VARCHAR, nested_col_b VARCHAR)))",
                "hive": "CREATE TABLE db.example_table (col_a STRUCT<struct_col_a INT, struct_col_b STRUCT<nested_col_a STRING, nested_col_b STRING>>)",
                "spark": "CREATE TABLE db.example_table (col_a STRUCT<struct_col_a: INT, struct_col_b: STRUCT<nested_col_a: STRING, nested_col_b: STRING>>)",
            },
        )
        self.validate_all(
            "CREATE TABLE db.example_table (col_a array<int>, col_b array<array<int>>)",
            write={
                "bigquery": "CREATE TABLE db.example_table (col_a ARRAY<INT64>, col_b ARRAY<ARRAY<INT64>>)",
                "duckdb": "CREATE TABLE db.example_table (col_a INT[], col_b INT[][])",
                "presto": "CREATE TABLE db.example_table (col_a ARRAY(INTEGER), col_b ARRAY(ARRAY(INTEGER)))",
                "hive": "CREATE TABLE db.example_table (col_a ARRAY<INT>, col_b ARRAY<ARRAY<INT>>)",
                "spark": "CREATE TABLE db.example_table (col_a ARRAY<INT>, col_b ARRAY<ARRAY<INT>>)",
                "snowflake": "CREATE TABLE db.example_table (col_a ARRAY, col_b ARRAY)",
            },
        )
        self.validate_all(
            "CREATE TABLE x USING ICEBERG PARTITIONED BY (MONTHS(y)) LOCATION 's3://z'",
            write={
                "duckdb": "CREATE TABLE x",
                "presto": "CREATE TABLE x WITH (FORMAT='ICEBERG', PARTITIONED_BY=ARRAY['MONTHS'])",
                "hive": "CREATE TABLE x STORED AS ICEBERG PARTITIONED BY (MONTHS(y)) LOCATION 's3://z'",
                "spark": "CREATE TABLE x USING ICEBERG PARTITIONED BY (MONTHS(y)) LOCATION 's3://z'",
            },
        )
        self.validate_all(
            "CREATE TABLE test STORED AS PARQUET AS SELECT 1",
            write={
                "duckdb": "CREATE TABLE test AS SELECT 1",
                "presto": "CREATE TABLE test WITH (FORMAT='PARQUET') AS SELECT 1",
                "hive": "CREATE TABLE test STORED AS PARQUET AS SELECT 1",
                "spark": "CREATE TABLE test USING PARQUET AS SELECT 1",
            },
        )
        self.validate_all(
            """CREATE TABLE blah (col_a INT) COMMENT "Test comment: blah" PARTITIONED BY (date STRING) STORED AS ICEBERG TBLPROPERTIES('x' = '1')""",
            write={
                "duckdb": """CREATE TABLE blah (
  col_a INT
)""",  # Partition columns should exist in table
                "presto": """CREATE TABLE blah (
  col_a INTEGER,
  date VARCHAR
)
COMMENT='Test comment: blah'
WITH (
  PARTITIONED_BY=ARRAY['date'],
  FORMAT='ICEBERG',
  x='1'
)""",
                "hive": """CREATE TABLE blah (
  col_a INT
)
COMMENT 'Test comment: blah'
PARTITIONED BY (
  date STRING
)
STORED AS ICEBERG
TBLPROPERTIES (
  'x'='1'
)""",
                "spark": """CREATE TABLE blah (
  col_a INT
)
COMMENT 'Test comment: blah'
PARTITIONED BY (
  date STRING
)
USING ICEBERG
TBLPROPERTIES (
  'x'='1'
)""",
            },
            pretty=True,
        )

        self.validate_all(
            "CACHE TABLE testCache OPTIONS ('storageLevel' 'DISK_ONLY') SELECT * FROM testData",
            write={
                "spark": "CACHE TABLE testCache OPTIONS('storageLevel' = 'DISK_ONLY') AS SELECT * FROM testData"
            },
        )
        self.validate_all(
            "ALTER TABLE StudentInfo ADD COLUMNS (LastName STRING, DOB TIMESTAMP)",
            write={
                "spark": "ALTER TABLE StudentInfo ADD COLUMNS (LastName STRING, DOB TIMESTAMP)",
            },
        )
        self.validate_all(
            "ALTER TABLE StudentInfo DROP COLUMNS (LastName, DOB)",
            write={
                "spark": "ALTER TABLE StudentInfo DROP COLUMNS (LastName, DOB)",
            },
        )
        self.validate_all(
            "CREATE TABLE x USING ICEBERG PARTITIONED BY (MONTHS(y)) LOCATION 's3://z'",
            identify=True,
            write={
                "spark": "CREATE TABLE `x` USING ICEBERG PARTITIONED BY (MONTHS(`y`)) LOCATION 's3://z'",
            },
        )

    def test_to_date(self):
        self.validate_all(
            "TO_DATE(x, 'yyyy-MM-dd')",
            write={
                "duckdb": "CAST(x AS DATE)",
                "hive": "TO_DATE(x)",
                "presto": "CAST(SUBSTR(CAST(x AS VARCHAR), 1, 10) AS DATE)",
                "spark": "TO_DATE(x)",
            },
        )
        self.validate_all(
            "TO_DATE(x, 'yyyy')",
            write={
                "duckdb": "CAST(STRPTIME(x, '%Y') AS DATE)",
                "hive": "TO_DATE(x, 'yyyy')",
                "presto": "CAST(DATE_PARSE(x, '%Y') AS DATE)",
                "spark": "TO_DATE(x, 'yyyy')",
            },
        )

    def test_hint(self):
        self.validate_all(
            "SELECT /*+ COALESCE(3) */ * FROM x",
            write={
                "spark": "SELECT /*+ COALESCE(3) */ * FROM x",
                "bigquery": "SELECT * FROM x",
            },
        )
        self.validate_all(
            "SELECT /*+ COALESCE(3), REPARTITION(1) */ * FROM x",
            write={
                "spark": "SELECT /*+ COALESCE(3), REPARTITION(1) */ * FROM x",
                "bigquery": "SELECT * FROM x",
            },
        )
        self.validate_all(
            "SELECT /*+ BROADCAST(table) */ cola FROM table",
            write={
                "spark": "SELECT /*+ BROADCAST(table) */ cola FROM table",
                "bigquery": "SELECT cola FROM table",
            },
        )
        self.validate_all(
            "SELECT /*+ BROADCASTJOIN(table) */ cola FROM table",
            write={
                "spark": "SELECT /*+ BROADCASTJOIN(table) */ cola FROM table",
                "bigquery": "SELECT cola FROM table",
            },
        )
        self.validate_all(
            "SELECT /*+ MAPJOIN(table) */ cola FROM table",
            write={
                "spark": "SELECT /*+ MAPJOIN(table) */ cola FROM table",
                "bigquery": "SELECT cola FROM table",
            },
        )
        self.validate_all(
            "SELECT /*+ MERGE(table) */ cola FROM table",
            write={
                "spark": "SELECT /*+ MERGE(table) */ cola FROM table",
                "bigquery": "SELECT cola FROM table",
            },
        )
        self.validate_all(
            "SELECT /*+ SHUFFLEMERGE(table) */ cola FROM table",
            write={
                "spark": "SELECT /*+ SHUFFLEMERGE(table) */ cola FROM table",
                "bigquery": "SELECT cola FROM table",
            },
        )
        self.validate_all(
            "SELECT /*+ MERGEJOIN(table) */ cola FROM table",
            write={
                "spark": "SELECT /*+ MERGEJOIN(table) */ cola FROM table",
                "bigquery": "SELECT cola FROM table",
            },
        )
        self.validate_all(
            "SELECT /*+ SHUFFLE_HASH(table) */ cola FROM table",
            write={
                "spark": "SELECT /*+ SHUFFLE_HASH(table) */ cola FROM table",
                "bigquery": "SELECT cola FROM table",
            },
        )
        self.validate_all(
            "SELECT /*+ SHUFFLE_REPLICATE_NL(table) */ cola FROM table",
            write={
                "spark": "SELECT /*+ SHUFFLE_REPLICATE_NL(table) */ cola FROM table",
                "bigquery": "SELECT cola FROM table",
            },
        )

    def test_spark(self):
        self.validate_identity("INTERVAL -86 days")
        self.validate_identity("SELECT UNIX_TIMESTAMP()")
        self.validate_identity("TRIM('    SparkSQL   ')")
        self.validate_identity("TRIM(BOTH 'SL' FROM 'SSparkSQLS')")
        self.validate_identity("TRIM(LEADING 'SL' FROM 'SSparkSQLS')")
        self.validate_identity("TRIM(TRAILING 'SL' FROM 'SSparkSQLS')")
        self.validate_identity("SPLIT(str, pattern, lim)")

        self.validate_all(
            "SELECT * FROM ((VALUES 1))", write={"spark": "SELECT * FROM (VALUES (1))"}
        )
        self.validate_all(
            "SELECT CAST(STRUCT('fooo') AS STRUCT<a: VARCHAR(2)>)",
            write={"spark": "SELECT CAST(STRUCT('fooo') AS STRUCT<a: STRING>)"},
        )
        self.validate_all(
            "SELECT CAST(123456 AS VARCHAR(3))",
            write={
                "": "SELECT TRY_CAST(123456 AS TEXT)",
                "spark": "SELECT CAST(123456 AS STRING)",
            },
        )
        self.validate_all(
            "SELECT piv.Q1 FROM (SELECT * FROM produce PIVOT(SUM(sales) FOR quarter IN ('Q1', 'Q2'))) AS piv",
            read={
                "snowflake": "SELECT piv.Q1 FROM produce PIVOT(SUM(sales) FOR quarter IN ('Q1', 'Q2')) piv",
            },
        )
        self.validate_all(
            "SELECT piv.Q1 FROM (SELECT * FROM (SELECT * FROM produce) PIVOT(SUM(sales) FOR quarter IN ('Q1', 'Q2'))) AS piv",
            read={
                "snowflake": "SELECT piv.Q1 FROM (SELECT * FROM produce) PIVOT(SUM(sales) FOR quarter IN ('Q1', 'Q2')) piv",
            },
        )
        self.validate_all(
            "SELECT * FROM produce PIVOT(SUM(produce.sales) FOR quarter IN ('Q1', 'Q2'))",
            read={
                "snowflake": "SELECT * FROM produce PIVOT (SUM(produce.sales) FOR produce.quarter IN ('Q1', 'Q2'))",
            },
        )
        self.validate_all(
            "SELECT * FROM produce AS p PIVOT(SUM(p.sales) AS sales FOR quarter IN ('Q1' AS Q1, 'Q2' AS Q1))",
            read={
                "bigquery": "SELECT * FROM produce AS p PIVOT(SUM(p.sales) AS sales FOR p.quarter IN ('Q1' AS Q1, 'Q2' AS Q1))",
            },
        )
        self.validate_all(
            "SELECT DATEDIFF(MONTH, '2020-01-01', '2020-03-05')",
            write={
                "databricks": "SELECT DATEDIFF(MONTH, TO_DATE('2020-01-01'), TO_DATE('2020-03-05'))",
                "hive": "SELECT MONTHS_BETWEEN(TO_DATE('2020-03-05'), TO_DATE('2020-01-01'))",
                "presto": "SELECT DATE_DIFF('MONTH', CAST(SUBSTR(CAST('2020-01-01' AS VARCHAR), 1, 10) AS DATE), CAST(SUBSTR(CAST('2020-03-05' AS VARCHAR), 1, 10) AS DATE))",
                "spark": "SELECT DATEDIFF(MONTH, TO_DATE('2020-01-01'), TO_DATE('2020-03-05'))",
                "spark2": "SELECT MONTHS_BETWEEN(TO_DATE('2020-03-05'), TO_DATE('2020-01-01'))",
                "trino": "SELECT DATE_DIFF('MONTH', CAST(SUBSTR(CAST('2020-01-01' AS VARCHAR), 1, 10) AS DATE), CAST(SUBSTR(CAST('2020-03-05' AS VARCHAR), 1, 10) AS DATE))",
            },
        )

        for data_type in ("BOOLEAN", "DATE", "DOUBLE", "FLOAT", "INT", "TIMESTAMP"):
            self.validate_all(
                f"{data_type}(x)",
                write={
                    "": f"CAST(x AS {data_type})",
                    "spark": f"CAST(x AS {data_type})",
                },
            )
        self.validate_all(
            "STRING(x)",
            write={
                "": "CAST(x AS TEXT)",
                "spark": "CAST(x AS STRING)",
            },
        )

        self.validate_all(
            "CAST(x AS TIMESTAMP)", read={"trino": "CAST(x AS TIMESTAMP(6) WITH TIME ZONE)"}
        )
        self.validate_all(
            "SELECT DATE_ADD(my_date_column, 1)",
            write={
                "spark": "SELECT DATE_ADD(my_date_column, 1)",
                "bigquery": "SELECT DATE_ADD(my_date_column, INTERVAL 1 DAY)",
            },
        )
        self.validate_all(
            "AGGREGATE(my_arr, 0, (acc, x) -> acc + x, s -> s * 2)",
            write={
                "trino": "REDUCE(my_arr, 0, (acc, x) -> acc + x, s -> s * 2)",
                "duckdb": "REDUCE(my_arr, 0, (acc, x) -> acc + x, s -> s * 2)",
                "hive": "REDUCE(my_arr, 0, (acc, x) -> acc + x, s -> s * 2)",
                "presto": "REDUCE(my_arr, 0, (acc, x) -> acc + x, s -> s * 2)",
                "spark": "AGGREGATE(my_arr, 0, (acc, x) -> acc + x, s -> s * 2)",
            },
        )
        self.validate_all(
            "TRIM('SL', 'SSparkSQLS')", write={"spark": "TRIM('SL' FROM 'SSparkSQLS')"}
        )
        self.validate_all(
            "ARRAY_SORT(x, (left, right) -> -1)",
            write={
                "duckdb": "ARRAY_SORT(x)",
                "presto": "ARRAY_SORT(x, (left, right) -> -1)",
                "hive": "SORT_ARRAY(x)",
                "spark": "ARRAY_SORT(x, (left, right) -> -1)",
            },
        )
        self.validate_all(
            "ARRAY(0, 1, 2)",
            write={
                "bigquery": "[0, 1, 2]",
                "duckdb": "LIST_VALUE(0, 1, 2)",
                "presto": "ARRAY[0, 1, 2]",
                "hive": "ARRAY(0, 1, 2)",
                "spark": "ARRAY(0, 1, 2)",
            },
        )

        self.validate_all(
            "SELECT fname, lname, age FROM person ORDER BY age DESC NULLS FIRST, fname ASC NULLS LAST, lname",
            write={
                "clickhouse": "SELECT fname, lname, age FROM person ORDER BY age DESC NULLS FIRST, fname, lname NULLS FIRST",
                "duckdb": "SELECT fname, lname, age FROM person ORDER BY age DESC NULLS FIRST, fname, lname NULLS FIRST",
                "postgres": "SELECT fname, lname, age FROM person ORDER BY age DESC, fname, lname NULLS FIRST",
                "presto": "SELECT fname, lname, age FROM person ORDER BY age DESC NULLS FIRST, fname, lname NULLS FIRST",
                "hive": "SELECT fname, lname, age FROM person ORDER BY age DESC NULLS FIRST, fname NULLS LAST, lname",
                "spark": "SELECT fname, lname, age FROM person ORDER BY age DESC NULLS FIRST, fname NULLS LAST, lname",
                "snowflake": "SELECT fname, lname, age FROM person ORDER BY age DESC, fname, lname NULLS FIRST",
            },
        )
        self.validate_all(
            "SELECT APPROX_COUNT_DISTINCT(a) FROM foo",
            write={
                "duckdb": "SELECT APPROX_COUNT_DISTINCT(a) FROM foo",
                "presto": "SELECT APPROX_DISTINCT(a) FROM foo",
                "hive": "SELECT APPROX_COUNT_DISTINCT(a) FROM foo",
                "spark": "SELECT APPROX_COUNT_DISTINCT(a) FROM foo",
            },
        )
        self.validate_all(
            "MONTH('2021-03-01')",
            write={
                "duckdb": "MONTH(CAST('2021-03-01' AS DATE))",
                "presto": "MONTH(CAST(SUBSTR(CAST('2021-03-01' AS VARCHAR), 1, 10) AS DATE))",
                "hive": "MONTH(TO_DATE('2021-03-01'))",
                "spark": "MONTH(TO_DATE('2021-03-01'))",
            },
        )
        self.validate_all(
            "YEAR('2021-03-01')",
            write={
                "duckdb": "YEAR(CAST('2021-03-01' AS DATE))",
                "presto": "YEAR(CAST(SUBSTR(CAST('2021-03-01' AS VARCHAR), 1, 10) AS DATE))",
                "hive": "YEAR(TO_DATE('2021-03-01'))",
                "spark": "YEAR(TO_DATE('2021-03-01'))",
            },
        )
        self.validate_all(
            "'\u6bdb'",
            write={
                "duckdb": "'毛'",
                "presto": "'毛'",
                "hive": "'毛'",
                "spark": "'毛'",
            },
        )
        self.validate_all(
            "SELECT LEFT(x, 2), RIGHT(x, 2)",
            write={
                "duckdb": "SELECT LEFT(x, 2), RIGHT(x, 2)",
                "presto": "SELECT SUBSTRING(x, 1, 2), SUBSTRING(x, LENGTH(x) - (2 - 1))",
                "hive": "SELECT SUBSTRING(x, 1, 2), SUBSTRING(x, LENGTH(x) - (2 - 1))",
                "spark": "SELECT LEFT(x, 2), RIGHT(x, 2)",
            },
        )
        self.validate_all(
            "MAP_FROM_ARRAYS(ARRAY(1), c)",
            write={
                "duckdb": "MAP(LIST_VALUE(1), c)",
                "presto": "MAP(ARRAY[1], c)",
                "hive": "MAP(ARRAY(1), c)",
                "spark": "MAP_FROM_ARRAYS(ARRAY(1), c)",
                "snowflake": "OBJECT_CONSTRUCT([1], c)",
            },
        )
        self.validate_all(
            "SELECT ARRAY_SORT(x)",
            write={
                "duckdb": "SELECT ARRAY_SORT(x)",
                "presto": "SELECT ARRAY_SORT(x)",
                "hive": "SELECT SORT_ARRAY(x)",
                "spark": "SELECT ARRAY_SORT(x)",
            },
        )

    def test_iif(self):
        self.validate_all(
            "SELECT IIF(cond, 'True', 'False')",
            write={"spark": "SELECT IF(cond, 'True', 'False')"},
        )

    def test_bool_or(self):
        self.validate_all(
            "SELECT a, LOGICAL_OR(b) FROM table GROUP BY a",
            write={"spark": "SELECT a, BOOL_OR(b) FROM table GROUP BY a"},
        )

    def test_current_user(self):
        self.validate_all(
            "CURRENT_USER",
            write={"spark": "CURRENT_USER()"},
        )
        self.validate_all(
            "CURRENT_USER()",
            write={"spark": "CURRENT_USER()"},
        )