[
https://issues.apache.org/jira/browse/SPARK-16874?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Abou Haydar Elias updated SPARK-16874:
--------------------------------------
Description:
I was porting some code I had from spark 1.6.2 to 2.0.0 and while reading a CSV
file, I've stumbled into the following error :
{code} org.apache.spark.sql.AnalysisException: Unable to resolve row.names
given [row.names, sbp, tobacco, ldl, adiposity, famhist, typea, obesity,
alcohol, age, chd];
{code}
There is is of course a work around using a predefined schema but it's not very
practical if there is lots of columns
I've provided in the following link code snippets that can generate the error
for spark 2.0 and another that show how it works perfectly with 1.6.2 using the
spark-csv package :
https://gist.github.com/eliasah/5dfc6bc8ddcbe920311049e37a58855b
was:
I was porting some code I had from spark 1.6.2 to 2.0.0 and while reading a CSV
file, I've stumbled into the following error :
{code} org.apache.spark.sql.AnalysisException: Unable to resolve row.names
given [row.names, sbp, tobacco, ldl, adiposity, famhist, typea, obesity,
alcohol, age, chd];
{code}
There is is of course a work around using a predefined schema but it's not very
practical if there is lots of columns
I've provided in the foolowing link code snippets that can generate the error
for spark 2.0 and another that show how it works perfectly with 1.6.2 using the
spark-csv package :
https://gist.github.com/eliasah/5dfc6bc8ddcbe920311049e37a58855b
> CSV Reader : Can't resolve column name with a point
> ---------------------------------------------------
>
> Key: SPARK-16874
> URL: https://issues.apache.org/jira/browse/SPARK-16874
> Project: Spark
> Issue Type: Bug
> Components: Input/Output
> Affects Versions: 2.0.0
> Reporter: Abou Haydar Elias
> Priority: Minor
>
> I was porting some code I had from spark 1.6.2 to 2.0.0 and while reading a
> CSV file, I've stumbled into the following error :
> {code} org.apache.spark.sql.AnalysisException: Unable to resolve row.names
> given [row.names, sbp, tobacco, ldl, adiposity, famhist, typea, obesity,
> alcohol, age, chd];
> {code}
> There is is of course a work around using a predefined schema but it's not
> very practical if there is lots of columns
> I've provided in the following link code snippets that can generate the error
> for spark 2.0 and another that show how it works perfectly with 1.6.2 using
> the spark-csv package :
> https://gist.github.com/eliasah/5dfc6bc8ddcbe920311049e37a58855b
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
