[
https://issues.apache.org/jira/browse/SPARK-15031?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Dongjoon Hyun updated SPARK-15031:
----------------------------------
Description:
This PR aims to update Scala/Python examples by replacing SQLContext with newly
added SparkSession. Also, this fixes the following examples.
**sql.py**
{code}
- people = sqlContext.jsonFile(path)
+ people = sqlContext.read.json(path)
- people.registerAsTable("people")
+ people.registerTempTable("people")
{code}
**dataframe_example.py**
{code}
- features = df.select("features").map(lambda r: r.features)
+ features = df.select("features").rdd.map(lambda r: r.features)
{code}
Note that the following examples are untouched in this PR since it fails some
unknown issue.
- `simple_params_example.py`
- `aft_survival_regression.py`
was:
Currently, Python SQL example, `sql.py`, fails.
{code}
bin/spark-submit examples/src/main/python/sql.py
Traceback (most recent call last):
File
"/Users/dongjoon/spark-release/spark-2.0/examples/src/main/python/sql.py", line
60, in <module>
people = sqlContext.jsonFile(path)
AttributeError: 'SQLContext' object has no attribute 'jsonFile'
{code}
{code}
Traceback (most recent call last):
File
"/Users/dongjoon/spark-release/spark-2.0/examples/src/main/python/sql.py", line
72, in <module>
people.registerAsTable("people")
File "/Users/dongjoon/spark/python/lib/pyspark.zip/pyspark/sql/dataframe.py",
line 795, in __getattr__
AttributeError: 'DataFrame' object has no attribute 'registerAsTable'
{code}
This issue fixes them by the following fix.
{code}
- people = sqlContext.jsonFile(path)
+ people = sqlContext.read.json(path)
...
- people.registerAsTable("people")
+ people.registerTempTable("people")
{code}
> Use SparkSession in Scala/Python example.
> -----------------------------------------
>
> Key: SPARK-15031
> URL: https://issues.apache.org/jira/browse/SPARK-15031
> Project: Spark
> Issue Type: Improvement
> Components: Examples
> Reporter: Dongjoon Hyun
> Priority: Trivial
>
> This PR aims to update Scala/Python examples by replacing SQLContext with
> newly added SparkSession. Also, this fixes the following examples.
> **sql.py**
> {code}
> - people = sqlContext.jsonFile(path)
> + people = sqlContext.read.json(path)
> - people.registerAsTable("people")
> + people.registerTempTable("people")
> {code}
> **dataframe_example.py**
> {code}
> - features = df.select("features").map(lambda r: r.features)
> + features = df.select("features").rdd.map(lambda r: r.features)
> {code}
> Note that the following examples are untouched in this PR since it fails some
> unknown issue.
> - `simple_params_example.py`
> - `aft_survival_regression.py`
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
