[
https://issues.apache.org/jira/browse/SPARK-13174?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15316789#comment-15316789
]
Felix Cheung commented on SPARK-13174:
--------------------------------------
This hasn't said to be R specific - but in the context of PR 11457 above, we
discussed R implementation with the name `read.csv` would masked the R base
package implementation (ie. making it not callable).
As a follow up, I did look into this and found the R base `read.csv` to be too
generic
{code}
read.csv(file, header = TRUE, sep = ",", quote = "\"",
dec = ".", fill = TRUE, comment.char = "", ...)
{code}
I was not sure if there was a way to implement this in SparkR without masking
the base version.
Perhaps if we would starting prefixing SparkR methods with `spark.` or
`sparkR.` (as being done in some specific cases)
> Add API and options for csv data sources
> ----------------------------------------
>
> Key: SPARK-13174
> URL: https://issues.apache.org/jira/browse/SPARK-13174
> Project: Spark
> Issue Type: New Feature
> Components: Input/Output
> Affects Versions: 2.0.0
> Reporter: Davies Liu
>
> We should have a API to load csv data source (with some options as
> arguments), similar to json() and jdbc()
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
