This is an automated email from the ASF dual-hosted git repository.
gurwls223 pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/spark.git
The following commit(s) were added to refs/heads/master by this push:
new af3b747a09f [SPARK-41149][PYTHON] Fix `SparkSession.builder.config` to
support bool
af3b747a09f is described below
commit af3b747a09f0df85856896c6241f5734b0b1dc54
Author: itholic <haejoon....@databricks.com>
AuthorDate: Thu Nov 17 15:01:33 2022 +0900
[SPARK-41149][PYTHON] Fix `SparkSession.builder.config` to support bool
### What changes were proposed in this pull request?
This PR proposes to support `bool` type for `SparkSession.builder.config`
when building the new `SparkSession`.
### Why are the changes needed?
Currently, Python `bool` type `True` or `False` is not working correctly,
since JVM side only treat the String type of "true" or "false" as a valid
option value.
So, this PR basically fix the current behavior as below:
**Before**
```python
>>> s1 = spark.builder.config("spark.sql.pyspark.jvmStacktrace.enabled",
True).getOrCreate()
>>> s1.conf.get("spark.sql.pyspark.jvmStacktrace.enabled")
'True' # invalid value, no effect for options.
```
**After**
```python
>>> s1 = spark.builder.config("spark.sql.pyspark.jvmStacktrace.enabled",
True).getOrCreate()
>>> s1.conf.get("spark.sql.pyspark.jvmStacktrace.enabled")
'true' # the lower case "true" only valid for options.
```
### Does this PR introduce _any_ user-facing change?
No.
### How was this patch tested?
Unittest added.
Closes #38673 from itholic/SPARK-41149.
Authored-by: itholic <haejoon....@databricks.com>
Signed-off-by: Hyukjin Kwon <gurwls...@apache.org>
---
python/pyspark/sql/session.py | 6 +++---
python/pyspark/sql/tests/test_session.py | 21 +++++++++++++++++++++
2 files changed, 24 insertions(+), 3 deletions(-)
diff --git a/python/pyspark/sql/session.py b/python/pyspark/sql/session.py
index ee04c94cbd5..eec3246cac3 100644
--- a/python/pyspark/sql/session.py
+++ b/python/pyspark/sql/session.py
@@ -60,7 +60,7 @@ from pyspark.sql.types import (
_parse_datatype_string,
_from_numpy_type,
)
-from pyspark.sql.utils import install_exception_handler,
is_timestamp_ntz_preferred
+from pyspark.sql.utils import install_exception_handler,
is_timestamp_ntz_preferred, to_str
if TYPE_CHECKING:
from pyspark.sql._typing import AtomicValue, RowLike, OptionalPrimitiveType
@@ -256,9 +256,9 @@ class SparkSession(SparkConversionMixin):
self._options[k] = v
elif map is not None:
for k, v in map.items(): # type: ignore[assignment]
- self._options[k] = str(v)
+ self._options[k] = to_str(v)
else:
- self._options[cast(str, key)] = str(value)
+ self._options[cast(str, key)] = to_str(value)
return self
def master(self, master: str) -> "SparkSession.Builder":
diff --git a/python/pyspark/sql/tests/test_session.py
b/python/pyspark/sql/tests/test_session.py
index 11544bf3131..80c05e1a3cb 100644
--- a/python/pyspark/sql/tests/test_session.py
+++ b/python/pyspark/sql/tests/test_session.py
@@ -327,6 +327,27 @@ class SparkSessionBuilderTests(unittest.TestCase):
if sc is not None:
sc.stop()
+ def test_create_spark_context_with_initial_session_options_bool(self):
+ session = None
+ # Test if `True` is set as "true".
+ try:
+ session = SparkSession.builder.config(
+ "spark.sql.pyspark.jvmStacktrace.enabled", True
+ ).getOrCreate()
+
self.assertEqual(session.conf.get("spark.sql.pyspark.jvmStacktrace.enabled"),
"true")
+ finally:
+ if session is not None:
+ session.stop()
+ # Test if `False` is set as "false".
+ try:
+ session = SparkSession.builder.config(
+ "spark.sql.pyspark.jvmStacktrace.enabled", False
+ ).getOrCreate()
+
self.assertEqual(session.conf.get("spark.sql.pyspark.jvmStacktrace.enabled"),
"false")
+ finally:
+ if session is not None:
+ session.stop()
+
class SparkExtensionsTest(unittest.TestCase):
# These tests are separate because it uses 'spark.sql.extensions' which is
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscr...@spark.apache.org
For additional commands, e-mail: commits-h...@spark.apache.org
