forked from databricks/databricks-sql-python
-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathcore_tests.py
More file actions
221 lines (209 loc) · 7.96 KB
/
core_tests.py
File metadata and controls
221 lines (209 loc) · 7.96 KB
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
import decimal
import datetime
from collections import namedtuple
TypeFailure = namedtuple(
"TypeFailure",
"query,columnType,resultType,resultValue,"
"actualValue,actualType,description,conf",
)
ResultFailure = namedtuple(
"ResultFailure",
"query,columnType,resultType,resultValue,"
"actualValue,actualType,description,conf",
)
ExecFailure = namedtuple(
"ExecFailure",
"query,columnType,resultType,resultValue,"
"actualValue,actualType,description,conf,error",
)
class SmokeTestMixin:
def test_smoke_test(self):
with self.cursor() as cursor:
cursor.execute("select 0")
rows = cursor.fetchall()
assert len(rows) == 1
assert rows[0][0] == 0
class CoreTestMixin:
"""
This mixin expects to be mixed with a CursorTest-like class with the following extra attributes:
validate_row_value_type: bool
validate_result: bool
"""
# A list of (subquery, column_type, python_type, expected_result)
# To be executed as "SELECT {} FROM RANGE(...)" and "SELECT {}"
range_queries = [
("TRUE", "boolean", bool, True),
("cast(1 AS TINYINT)", "byte", int, 1),
("cast(1000 AS SMALLINT)", "short", int, 1000),
("cast(100000 AS INTEGER)", "integer", int, 100000),
("cast(10000000000000 AS BIGINT)", "long", int, 10000000000000),
("cast(100.001 AS DECIMAL(6, 3))", "decimal", decimal.Decimal, 100.001),
("date '2020-02-20'", "date", datetime.date, datetime.date(2020, 2, 20)),
("unhex('f000')", "binary", bytes, b"\xf0\x00"), # pyodbc internal mismatch
("'foo'", "string", str, "foo"),
# SPARK-32130: 6.x: "4 weeks 2 days" vs 7.x: "30 days"
# ("interval 30 days", str, str, "interval 4 weeks 2 days"),
# ("interval 3 days", str, str, "interval 3 days"),
("CAST(NULL AS DOUBLE)", "double", type(None), None),
]
# Full queries, only the first column of the first row is checked
queries = [("NULL UNION (SELECT 1) order by 1", "integer", type(None), None)]
def run_tests_on_queries(self, default_conf):
failures = []
for query, columnType, rowValueType, answer in self.range_queries:
with self.cursor(default_conf) as cursor:
failures.extend(
self.run_query(
cursor, query, columnType, rowValueType, answer, default_conf
)
)
failures.extend(
self.run_range_query(
cursor, query, columnType, rowValueType, answer, default_conf
)
)
for query, columnType, rowValueType, answer in self.queries:
with self.cursor(default_conf) as cursor:
failures.extend(
self.run_query(
cursor, query, columnType, rowValueType, answer, default_conf
)
)
if failures:
self.fail(
"Failed testing result set with Arrow. "
"Failed queries: {}".format("\n\n".join([str(f) for f in failures]))
)
def run_query(self, cursor, query, columnType, rowValueType, answer, conf):
full_query = "SELECT {}".format(query)
expected_column_types = self.expected_column_types(columnType)
try:
cursor.execute(full_query)
(result,) = cursor.fetchone()
if not all(
cursor.description[0][1] == type for type in expected_column_types
):
return [
TypeFailure(
full_query,
expected_column_types,
rowValueType,
answer,
result,
type(result),
cursor.description,
conf,
)
]
if self.validate_row_value_type and type(result) is not rowValueType:
return [
TypeFailure(
full_query,
expected_column_types,
rowValueType,
answer,
result,
type(result),
cursor.description,
conf,
)
]
if self.validate_result and str(answer) != str(result):
return [
ResultFailure(
full_query,
query,
expected_column_types,
rowValueType,
answer,
result,
type(result),
cursor.description,
conf,
)
]
return []
except Exception as e:
return [
ExecFailure(
full_query,
columnType,
rowValueType,
None,
None,
None,
cursor.description,
conf,
e,
)
]
def run_range_query(self, cursor, query, columnType, rowValueType, expected, conf):
full_query = "SELECT {}, id FROM RANGE({})".format(query, 5000)
expected_column_types = self.expected_column_types(columnType)
try:
cursor.execute(full_query)
while True:
rows = cursor.fetchmany(1000)
if len(rows) <= 0:
break
for index, (result, id) in enumerate(rows):
if not all(
cursor.description[0][1] == type
for type in expected_column_types
):
return [
TypeFailure(
full_query,
expected_column_types,
rowValueType,
expected,
result,
type(result),
cursor.description,
conf,
)
]
if (
self.validate_row_value_type
and type(result) is not rowValueType
):
return [
TypeFailure(
full_query,
expected_column_types,
rowValueType,
expected,
result,
type(result),
cursor.description,
conf,
)
]
if self.validate_result and str(expected) != str(result):
return [
ResultFailure(
full_query,
expected_column_types,
rowValueType,
expected,
result,
type(result),
cursor.description,
conf,
)
]
return []
except Exception as e:
return [
ExecFailure(
full_query,
columnType,
rowValueType,
None,
None,
None,
cursor.description,
conf,
e,
)
]