Skip to content

[SPARK-25659][PYTHON][TEST] Test type inference specification for createDataFrame in PySpark #22653

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
wants to merge 1 commit into from
Closed

[SPARK-25659][PYTHON][TEST] Test type inference specification for createDataFrame in PySpark #22653

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
69 changes: 69 additions & 0 deletions python/pyspark/sql/tests.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1149,6 +1149,75 @@ def test_infer_schema(self):
result = self.spark.sql("SELECT l[0].a from test2 where d['key'].d = '2'")
self.assertEqual(1, result.head()[0])

def test_infer_schema_specification(self):
from decimal import Decimal

class A(object):
def __init__(self):
self.a = 1

data = [
True,
1,
"a",
u"a",
Copy link
Member

@viirya viirya Oct 9, 2018

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

nit: since this is for unicode string, how about using a non-ascii string?

Copy link
Member Author

@HyukjinKwon HyukjinKwon Oct 9, 2018

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Fair point. Will change when I fix some codes around here.

datetime.date(1970, 1, 1),
datetime.datetime(1970, 1, 1, 0, 0),
1.0,
array.array("d", [1]),
[1],
(1, ),
{"a": 1},
bytearray(1),
Decimal(1),
Row(a=1),
Row("a")(1),
A(),
Copy link
Member

@BryanCutler BryanCutler Oct 8, 2018

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I didn't this was possible - does it just look at the variable attributes in the object to get the fields?

Copy link
Member Author

@HyukjinKwon HyukjinKwon Oct 8, 2018

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Yea, it uses __dict__ attribute.. looks that's not possible in UDFs .. possibly an issue.

]

df = self.spark.createDataFrame([data])
actual = list(map(lambda x: x.dataType.simpleString(), df.schema))
expected = [
'boolean',
'bigint',
'string',
'string',
'date',
'timestamp',
'double',
'array<double>',
'array<bigint>',
'struct<_1:bigint>',
'map<string,bigint>',
'binary',
'decimal(38,18)',
'struct<a:bigint>',
'struct<a:bigint>',
'struct<a:bigint>',
]
self.assertEqual(actual, expected)

actual = list(df.first())
expected = [
True,
1,
'a',
u"a",
datetime.date(1970, 1, 1),
datetime.datetime(1970, 1, 1, 0, 0),
1.0,
[1.0],
[1],
Row(_1=1),
{"a": 1},
bytearray(b'\x00'),
Decimal('1.000000000000000000'),
Row(a=1),
Row(a=1),
Row(a=1),
]
self.assertEqual(actual, expected)

def test_infer_schema_not_enough_names(self):
df = self.spark.createDataFrame([["a", "b"]], ["col1"])
self.assertEqual(df.columns, ['col1', '_2'])
Expand Down