Neil Dewar created SPARK-16467:
----------------------------------
Summary: After importing R data.frame, although DataFrame columns
show . replaced by _, the describe() function gives warnings on . in the name
Key: SPARK-16467
URL: https://issues.apache.org/jira/browse/SPARK-16467
Project: Spark
Issue Type: Bug
Components: SparkR
Affects Versions: 1.6.1
Reporter: Neil Dewar
Priority: Minor
When importing an R data.frame into a Spark DataFrame using createDataFrame(),
if the R data.frame column names contain period characters "." they are
converted to underscores "_". Most SparkR functions then display the column
names with the underscores.
If the describe() function is used, with the option for a specific function
name, warning messages are displayed about the use of period characters.
Example:
sdfIris <- createDataFrame(sqlContext, iris)
str(sdfIris) # works fine
collect(describe(sdfIris, Species))
the last function above throws a string of warnings such as:
1: In FUN(X[[i]], ...) : Use Sepal_Length instead of Sepal.Length as column
name
Note, the describe() function appears to work ok if no column name is specified
but fails if a column name is specified, even if the specified name did not
contain a period.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
