9
9
@pytest .mark .repeat (1 )
10
10
@pytest .mark .parametrize ("benchmark_time" , [150 ])
11
11
def test_s3_select (benchmark_time ):
12
-
13
12
path = "s3://ursa-labs-taxi-data/2018/1*.parquet"
14
13
with ExecutionTimer ("elapsed time of wr.s3.select_query()" ) as timer :
15
14
wr .s3 .select_query (
@@ -23,23 +22,21 @@ def test_s3_select(benchmark_time):
23
22
assert timer .elapsed_time < benchmark_time
24
23
25
24
26
- @pytest .mark .parametrize ("benchmark_time" , [120 ])
25
+ @pytest .mark .parametrize ("benchmark_time" , [90 ])
27
26
def test_s3_read_parquet_simple (benchmark_time ):
28
-
29
27
path = "s3://ursa-labs-taxi-data/2018/"
30
28
with ExecutionTimer ("elapsed time of wr.s3.read_parquet() simple" ) as timer :
31
29
wr .s3 .read_parquet (path = path , parallelism = 10000 )
32
30
33
31
assert timer .elapsed_time < benchmark_time
34
32
35
33
36
- @pytest .mark .parametrize ("benchmark_time" , [120 ])
34
+ @pytest .mark .parametrize ("benchmark_time" , [90 ])
37
35
def test_s3_read_parquet_partition_filter (benchmark_time ):
38
-
39
36
path = "s3://amazon-reviews-pds/parquet/"
40
37
with ExecutionTimer ("elapsed time of wr.s3.read_parquet() partition filter" ) as timer :
41
38
filter = lambda x : True if x ["product_category" ].startswith ("Wireless" ) else False
42
- wr .s3 .read_parquet (path = path , parallelism = 10000 , partition_filter = filter , dataset = True )
39
+ wr .s3 .read_parquet (path = path , parallelism = 10000 , dataset = True , partition_filter = filter )
43
40
44
41
assert timer .elapsed_time < benchmark_time
45
42
0 commit comments