Github user marmbrus commented on a diff in the pull request:
https://github.com/apache/spark/pull/4446#discussion_r24432619
--- Diff: python/pyspark/sql_tests.py ---
@@ -285,6 +285,38 @@ def test_aggregator(self):
self.assertTrue(95 <
g.agg(Dsl.approxCountDistinct(df.key)).first()[0])
self.assertEqual(100,
g.agg(Dsl.countDistinct(df.value)).first()[0])
+ def test_save_and_load(self):
+ df = self.df
+ tmpPath = tempfile.mkdtemp()
+ shutil.rmtree(tmpPath)
+ df.save(tmpPath, "org.apache.spark.sql.json", "error")
+ actual = self.sqlCtx.load(tmpPath, "org.apache.spark.sql.json")
+ self.assertTrue(sorted(df.collect()) == sorted(actual.collect()))
+
+ from pyspark.sql import StructType, StructField, StringType
+ schema = StructType([StructField("value", StringType(), True)])
+ actual = self.sqlCtx.load(tmpPath, "org.apache.spark.sql.json",
schema)
+ self.assertTrue(sorted(df.select("value").collect()) ==
sorted(actual.collect()))
+
+ df.save(tmpPath, "org.apache.spark.sql.json", "overwrite")
+ actual = self.sqlCtx.load(tmpPath, "org.apache.spark.sql.json")
+ self.assertTrue(sorted(df.collect()) == sorted(actual.collect()))
+
+ df.save(dataSourceName="org.apache.spark.sql.json",
mode="overwrite", path=tmpPath,
+ noUse="this options will not be used in save.")
+ actual =
self.sqlCtx.load(dataSourceName="org.apache.spark.sql.json", path=tmpPath,
+ noUse="this options will not be used in
load.")
+ self.assertTrue(sorted(df.collect()) == sorted(actual.collect()))
+
+ defaultDataSourceName =
self.sqlCtx._ssql_ctx.getConf("spark.sql.sources.default",
+
"org.apache.spark.sql.parquet")
+ self.sqlCtx.sql("SET
spark.sql.sources.default=org.apache.spark.sql.json")
--- End diff --
I think we should just try org.apache.spark.sql.* whenever what they give
us fails to resolve on its own.
On Feb 10, 2015 9:43 AM, "Yin Huai" <[email protected]> wrote:
> In python/pyspark/sql_tests.py
> <https://github.com/apache/spark/pull/4446#discussion_r24432467>:
>
> > + actual = self.sqlCtx.load(tmpPath,
"org.apache.spark.sql.json", schema)
> > + self.assertTrue(sorted(df.select("value").collect()) ==
sorted(actual.collect()))
> > +
> > + df.save(tmpPath, "org.apache.spark.sql.json", "overwrite")
> > + actual = self.sqlCtx.load(tmpPath, "org.apache.spark.sql.json")
> > + self.assertTrue(sorted(df.collect()) ==
sorted(actual.collect()))
> > +
> > + df.save(dataSourceName="org.apache.spark.sql.json",
mode="overwrite", path=tmpPath,
> > + noUse="this options will not be used in save.")
> > + actual =
self.sqlCtx.load(dataSourceName="org.apache.spark.sql.json", path=tmpPath,
> > + noUse="this options will not be used
in load.")
> > + self.assertTrue(sorted(df.collect()) ==
sorted(actual.collect()))
> > +
> > + defaultDataSourceName =
self.sqlCtx._ssql_ctx.getConf("spark.sql.sources.default",
> > +
"org.apache.spark.sql.parquet")
> > + self.sqlCtx.sql("SET
spark.sql.sources.default=org.apache.spark.sql.json")
>
> @davies <https://github.com/davies> You meant a shorter name for the conf
> key "spark.sql.sources.default"?
>
> â
> Reply to this email directly or view it on GitHub
> <https://github.com/apache/spark/pull/4446/files#r24432467>.
>
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
