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
|
from sqlglot import UnsupportedError
from tests.dialects.test_dialect import Validator
class TestSnowflake(Validator):
dialect = "snowflake"
def test_snowflake(self):
self.validate_all(
'x:a:"b c"',
write={
"duckdb": "x['a']['b c']",
"hive": "x['a']['b c']",
"presto": "x['a']['b c']",
"snowflake": "x['a']['b c']",
"spark": "x['a']['b c']",
},
)
self.validate_all(
"SELECT a FROM test WHERE a = 1 GROUP BY a HAVING a = 2 QUALIFY z ORDER BY a LIMIT 10",
write={
"bigquery": "SELECT a FROM test WHERE a = 1 GROUP BY a HAVING a = 2 QUALIFY z ORDER BY a NULLS LAST LIMIT 10",
"snowflake": "SELECT a FROM test WHERE a = 1 GROUP BY a HAVING a = 2 QUALIFY z ORDER BY a LIMIT 10",
},
)
self.validate_all(
"SELECT a FROM test AS t QUALIFY ROW_NUMBER() OVER (PARTITION BY a ORDER BY Z) = 1",
write={
"bigquery": "SELECT a FROM test AS t QUALIFY ROW_NUMBER() OVER (PARTITION BY a ORDER BY Z NULLS LAST) = 1",
"snowflake": "SELECT a FROM test AS t QUALIFY ROW_NUMBER() OVER (PARTITION BY a ORDER BY Z) = 1",
},
)
self.validate_all(
"SELECT TO_TIMESTAMP(1659981729)",
write={
"bigquery": "SELECT UNIX_TO_TIME(1659981729)",
"snowflake": "SELECT TO_TIMESTAMP(1659981729)",
"spark": "SELECT FROM_UNIXTIME(1659981729)",
},
)
self.validate_all(
"SELECT TO_TIMESTAMP(1659981729000, 3)",
write={
"bigquery": "SELECT UNIX_TO_TIME(1659981729000, 'millis')",
"snowflake": "SELECT TO_TIMESTAMP(1659981729000, 3)",
"spark": "SELECT TIMESTAMP_MILLIS(1659981729000)",
},
)
self.validate_all(
"SELECT TO_TIMESTAMP('1659981729')",
write={
"bigquery": "SELECT UNIX_TO_TIME('1659981729')",
"snowflake": "SELECT TO_TIMESTAMP('1659981729')",
"spark": "SELECT FROM_UNIXTIME('1659981729')",
},
)
self.validate_all(
"SELECT TO_TIMESTAMP(1659981729000000000, 9)",
write={
"bigquery": "SELECT UNIX_TO_TIME(1659981729000000000, 'micros')",
"snowflake": "SELECT TO_TIMESTAMP(1659981729000000000, 9)",
"spark": "SELECT TIMESTAMP_MICROS(1659981729000000000)",
},
)
self.validate_all(
"SELECT TO_TIMESTAMP('2013-04-05 01:02:03')",
write={
"bigquery": "SELECT PARSE_TIMESTAMP('%Y-%m-%d %H:%M:%S', '2013-04-05 01:02:03')",
"snowflake": "SELECT TO_TIMESTAMP('2013-04-05 01:02:03', 'yyyy-mm-dd hh24:mi:ss')",
"spark": "SELECT TO_TIMESTAMP('2013-04-05 01:02:03', 'yyyy-MM-dd HH:mm:ss')",
},
)
self.validate_all(
"SELECT TO_TIMESTAMP('04/05/2013 01:02:03', 'mm/dd/yyyy hh24:mi:ss')",
read={
"bigquery": "SELECT PARSE_TIMESTAMP('%m/%d/%Y %H:%M:%S', '04/05/2013 01:02:03')",
"duckdb": "SELECT STRPTIME('04/05/2013 01:02:03', '%m/%d/%Y %H:%M:%S')",
"snowflake": "SELECT TO_TIMESTAMP('04/05/2013 01:02:03', 'mm/dd/yyyy hh24:mi:ss')",
},
write={
"bigquery": "SELECT PARSE_TIMESTAMP('%m/%d/%Y %H:%M:%S', '04/05/2013 01:02:03')",
"snowflake": "SELECT TO_TIMESTAMP('04/05/2013 01:02:03', 'mm/dd/yyyy hh24:mi:ss')",
"spark": "SELECT TO_TIMESTAMP('04/05/2013 01:02:03', 'MM/dd/yyyy HH:mm:ss')",
},
)
self.validate_all(
"SELECT IFF(TRUE, 'true', 'false')",
write={
"snowflake": "SELECT IFF(TRUE, 'true', 'false')",
},
)
self.validate_all(
"SELECT fname, lname, age FROM person ORDER BY age DESC NULLS FIRST, fname ASC NULLS LAST, lname",
write={
"duckdb": "SELECT fname, lname, age FROM person ORDER BY age DESC NULLS FIRST, fname NULLS LAST, lname NULLS LAST",
"postgres": "SELECT fname, lname, age FROM person ORDER BY age DESC, fname, lname",
"presto": "SELECT fname, lname, age FROM person ORDER BY age DESC NULLS FIRST, fname, lname",
"hive": "SELECT fname, lname, age FROM person ORDER BY age DESC NULLS FIRST, fname NULLS LAST, lname NULLS LAST",
"spark": "SELECT fname, lname, age FROM person ORDER BY age DESC NULLS FIRST, fname NULLS LAST, lname NULLS LAST",
"snowflake": "SELECT fname, lname, age FROM person ORDER BY age DESC, fname, lname",
},
)
self.validate_all(
"SELECT ARRAY_AGG(DISTINCT a)",
write={
"spark": "SELECT COLLECT_LIST(DISTINCT a)",
"snowflake": "SELECT ARRAY_AGG(DISTINCT a)",
},
)
self.validate_all(
"SELECT * FROM a INTERSECT ALL SELECT * FROM b",
write={
"snowflake": UnsupportedError,
},
)
self.validate_all(
"SELECT * FROM a EXCEPT ALL SELECT * FROM b",
write={
"snowflake": UnsupportedError,
},
)
self.validate_all(
"SELECT ARRAY_UNION_AGG(a)",
write={
"snowflake": "SELECT ARRAY_UNION_AGG(a)",
},
)
self.validate_all(
"SELECT NVL2(a, b, c)",
write={
"snowflake": "SELECT NVL2(a, b, c)",
},
)
self.validate_all(
"SELECT $$a$$",
write={
"snowflake": "SELECT 'a'",
},
)
self.validate_all(
r"SELECT $$a ' \ \t \x21 z $ $$",
write={
"snowflake": r"SELECT 'a \' \\ \\t \\x21 z $ '",
},
)
self.validate_identity("SELECT REGEXP_LIKE(a, b, c)")
self.validate_all(
"SELECT RLIKE(a, b)",
write={
"snowflake": "SELECT REGEXP_LIKE(a, b)",
},
)
self.validate_all(
"SELECT a FROM test SAMPLE BLOCK (0.5) SEED (42)",
write={
"snowflake": "SELECT a FROM test TABLESAMPLE BLOCK (0.5) SEED (42)",
},
)
self.validate_all(
"SELECT a FROM test pivot",
write={
"snowflake": "SELECT a FROM test AS pivot",
},
)
self.validate_all(
"SELECT a FROM test unpivot",
write={
"snowflake": "SELECT a FROM test AS unpivot",
},
)
def test_null_treatment(self):
self.validate_all(
r"SELECT FIRST_VALUE(TABLE1.COLUMN1) OVER (PARTITION BY RANDOM_COLUMN1, RANDOM_COLUMN2 ROWS BETWEEN UNBOUNDED PRECEDING AND UNBOUNDED FOLLOWING) AS MY_ALIAS FROM TABLE1",
write={
"snowflake": r"SELECT FIRST_VALUE(TABLE1.COLUMN1) OVER (PARTITION BY RANDOM_COLUMN1, RANDOM_COLUMN2 ROWS BETWEEN UNBOUNDED PRECEDING AND UNBOUNDED FOLLOWING) AS MY_ALIAS FROM TABLE1"
},
)
self.validate_all(
r"SELECT FIRST_VALUE(TABLE1.COLUMN1 RESPECT NULLS) OVER (PARTITION BY RANDOM_COLUMN1, RANDOM_COLUMN2 ROWS BETWEEN UNBOUNDED PRECEDING AND UNBOUNDED FOLLOWING) AS MY_ALIAS FROM TABLE1",
write={
"snowflake": r"SELECT FIRST_VALUE(TABLE1.COLUMN1) OVER (PARTITION BY RANDOM_COLUMN1, RANDOM_COLUMN2 ROWS BETWEEN UNBOUNDED PRECEDING AND UNBOUNDED FOLLOWING) AS MY_ALIAS FROM TABLE1"
},
)
self.validate_all(
r"SELECT FIRST_VALUE(TABLE1.COLUMN1) RESPECT NULLS OVER (PARTITION BY RANDOM_COLUMN1, RANDOM_COLUMN2 ROWS BETWEEN UNBOUNDED PRECEDING AND UNBOUNDED FOLLOWING) AS MY_ALIAS FROM TABLE1",
write={
"snowflake": r"SELECT FIRST_VALUE(TABLE1.COLUMN1) RESPECT NULLS OVER (PARTITION BY RANDOM_COLUMN1, RANDOM_COLUMN2 ROWS BETWEEN UNBOUNDED PRECEDING AND UNBOUNDED FOLLOWING) AS MY_ALIAS FROM TABLE1"
},
)
self.validate_all(
r"SELECT FIRST_VALUE(TABLE1.COLUMN1 IGNORE NULLS) OVER (PARTITION BY RANDOM_COLUMN1, RANDOM_COLUMN2 ROWS BETWEEN UNBOUNDED PRECEDING AND UNBOUNDED FOLLOWING) AS MY_ALIAS FROM TABLE1",
write={
"snowflake": r"SELECT FIRST_VALUE(TABLE1.COLUMN1 IGNORE NULLS) OVER (PARTITION BY RANDOM_COLUMN1, RANDOM_COLUMN2 ROWS BETWEEN UNBOUNDED PRECEDING AND UNBOUNDED FOLLOWING) AS MY_ALIAS FROM TABLE1"
},
)
self.validate_all(
r"SELECT FIRST_VALUE(TABLE1.COLUMN1) IGNORE NULLS OVER (PARTITION BY RANDOM_COLUMN1, RANDOM_COLUMN2 ROWS BETWEEN UNBOUNDED PRECEDING AND UNBOUNDED FOLLOWING) AS MY_ALIAS FROM TABLE1",
write={
"snowflake": r"SELECT FIRST_VALUE(TABLE1.COLUMN1) IGNORE NULLS OVER (PARTITION BY RANDOM_COLUMN1, RANDOM_COLUMN2 ROWS BETWEEN UNBOUNDED PRECEDING AND UNBOUNDED FOLLOWING) AS MY_ALIAS FROM TABLE1"
},
)
def test_timestamps(self):
self.validate_all(
"SELECT CAST(a AS TIMESTAMP)",
write={
"snowflake": "SELECT CAST(a AS TIMESTAMPNTZ)",
},
)
self.validate_all(
"SELECT a::TIMESTAMP_LTZ(9)",
write={
"snowflake": "SELECT CAST(a AS TIMESTAMPLTZ(9))",
},
)
self.validate_all(
"SELECT a::TIMESTAMPLTZ",
write={
"snowflake": "SELECT CAST(a AS TIMESTAMPLTZ)",
},
)
self.validate_all(
"SELECT a::TIMESTAMP WITH LOCAL TIME ZONE",
write={
"snowflake": "SELECT CAST(a AS TIMESTAMPLTZ)",
},
)
self.validate_identity("SELECT EXTRACT(month FROM a)")
self.validate_all(
"SELECT EXTRACT('month', a)",
write={
"snowflake": "SELECT EXTRACT('month' FROM a)",
},
)
self.validate_all(
"SELECT DATE_PART('month', a)",
write={
"snowflake": "SELECT EXTRACT('month' FROM a)",
},
)
self.validate_all(
"SELECT DATE_PART(month, a::DATETIME)",
write={
"snowflake": "SELECT EXTRACT(month FROM CAST(a AS DATETIME))",
},
)
self.validate_all(
"SELECT DATE_PART(epoch_second, foo) as ddate from table_name",
write={
"snowflake": "SELECT EXTRACT(epoch_second FROM CAST(foo AS TIMESTAMPNTZ)) AS ddate FROM table_name",
"presto": "SELECT TO_UNIXTIME(CAST(foo AS TIMESTAMP)) AS ddate FROM table_name",
},
)
self.validate_all(
"SELECT DATE_PART(epoch_milliseconds, foo) as ddate from table_name",
write={
"snowflake": "SELECT EXTRACT(epoch_second FROM CAST(foo AS TIMESTAMPNTZ)) * 1000 AS ddate FROM table_name",
"presto": "SELECT TO_UNIXTIME(CAST(foo AS TIMESTAMP)) * 1000 AS ddate FROM table_name",
},
)
def test_semi_structured_types(self):
self.validate_identity("SELECT CAST(a AS VARIANT)")
self.validate_all(
"SELECT a::VARIANT",
write={
"snowflake": "SELECT CAST(a AS VARIANT)",
"tsql": "SELECT CAST(a AS SQL_VARIANT)",
},
)
self.validate_identity("SELECT CAST(a AS ARRAY)")
self.validate_all(
"ARRAY_CONSTRUCT(0, 1, 2)",
write={
"snowflake": "[0, 1, 2]",
"bigquery": "[0, 1, 2]",
"duckdb": "LIST_VALUE(0, 1, 2)",
"presto": "ARRAY[0, 1, 2]",
"spark": "ARRAY(0, 1, 2)",
},
)
self.validate_all(
"SELECT a::OBJECT",
write={
"snowflake": "SELECT CAST(a AS OBJECT)",
},
)
def test_ddl(self):
self.validate_identity(
"CREATE TABLE a (x DATE, y BIGINT) WITH (PARTITION BY (x), integration='q', auto_refresh=TRUE, file_format=(type = parquet))"
)
self.validate_identity("CREATE MATERIALIZED VIEW a COMMENT='...' AS SELECT 1 FROM x")
def test_user_defined_functions(self):
self.validate_all(
"CREATE FUNCTION a(x DATE, y BIGINT) RETURNS ARRAY LANGUAGE JAVASCRIPT AS $$ SELECT 1 $$",
write={
"snowflake": "CREATE FUNCTION a(x DATE, y BIGINT) RETURNS ARRAY LANGUAGE JAVASCRIPT AS ' SELECT 1 '",
},
)
self.validate_all(
"CREATE FUNCTION a() RETURNS TABLE (b INT) AS 'SELECT 1'",
write={
"snowflake": "CREATE FUNCTION a() RETURNS TABLE (b INT) AS 'SELECT 1'",
"bigquery": "CREATE TABLE FUNCTION a() RETURNS TABLE <b INT64> AS SELECT 1",
},
)
self.validate_all(
"CREATE FUNCTION a() RETURNS INT IMMUTABLE AS 'SELECT 1'",
write={
"snowflake": "CREATE FUNCTION a() RETURNS INT IMMUTABLE AS 'SELECT 1'",
},
)
def test_stored_procedures(self):
self.validate_identity("CALL a.b.c(x, y)")
self.validate_identity(
"CREATE PROCEDURE a.b.c(x INT, y VARIANT) RETURNS OBJECT EXECUTE AS CALLER AS 'BEGIN SELECT 1; END;'"
)
def test_table_literal(self):
# All examples from https://docs.snowflake.com/en/sql-reference/literals-table.html
self.validate_all(
r"""SELECT * FROM TABLE('MYTABLE')""", write={"snowflake": r"""SELECT * FROM TABLE('MYTABLE')"""}
)
self.validate_all(
r"""SELECT * FROM TABLE('MYDB."MYSCHEMA"."MYTABLE"')""",
write={"snowflake": r"""SELECT * FROM TABLE('MYDB."MYSCHEMA"."MYTABLE"')"""},
)
# Per Snowflake documentation at https://docs.snowflake.com/en/sql-reference/literals-table.html
# one can use either a " ' " or " $$ " to enclose the object identifier.
# Capturing the single tokens seems like lot of work. Hence adjusting tests to use these interchangeably,
self.validate_all(
r"""SELECT * FROM TABLE($$MYDB. "MYSCHEMA"."MYTABLE"$$)""",
write={"snowflake": r"""SELECT * FROM TABLE('MYDB. "MYSCHEMA"."MYTABLE"')"""},
)
self.validate_all(r"""SELECT * FROM TABLE($MYVAR)""", write={"snowflake": r"""SELECT * FROM TABLE($MYVAR)"""})
self.validate_all(r"""SELECT * FROM TABLE(?)""", write={"snowflake": r"""SELECT * FROM TABLE(?)"""})
self.validate_all(
r"""SELECT * FROM TABLE(:BINDING)""", write={"snowflake": r"""SELECT * FROM TABLE(:BINDING)"""}
)
self.validate_all(
r"""SELECT * FROM TABLE($MYVAR) WHERE COL1 = 10""",
write={"snowflake": r"""SELECT * FROM TABLE($MYVAR) WHERE COL1 = 10"""},
)
|
