[
https://issues.apache.org/jira/browse/FLINK-10218?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16593285#comment-16593285
]
ASF GitHub Bot commented on FLINK-10218:
----------------------------------------
zentol commented on a change in pull request #6616: [FLINK-10218] Allow writing
DataSet without explicit path parameter
URL: https://github.com/apache/flink/pull/6616#discussion_r212897718
##########
File path: flink-java/src/main/java/org/apache/flink/api/java/DataSet.java
##########
@@ -1727,6 +1727,21 @@ public void printToErr() throws Exception {
return output(new PrintingOutputFormat<T>(sinkIdentifier,
true));
}
+ /**
+ * Writes a DataSet using a {@link FileOutputFormat} to a specified
location.
+ * This method adds a data sink to the program.
+ *
+ * @param outputFormat The FileOutputFormat to write the DataSet.
+ * @return The DataSink that writes the DataSet.
+ *
+ * @see FileOutputFormat
+ */
+ public DataSink<T> write(FileOutputFormat<T> outputFormat) {
+ Preconditions.checkNotNull(outputFormat, "Output format must
not be null.");
+ Preconditions.checkNotNull(outputFormat.getOutputFilePath(),
"File path must not be null.");
+ return output(outputFormat);
Review comment:
this right here is already a viable alternative for users, hence I would
reject this PR.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
> Allow writing DataSet without explicit path parameter
> -----------------------------------------------------
>
> Key: FLINK-10218
> URL: https://issues.apache.org/jira/browse/FLINK-10218
> Project: Flink
> Issue Type: Improvement
> Components: DataSet API
> Affects Versions: 1.6.0
> Reporter: Paul Lin
> Priority: Minor
> Labels: pull-request-available
>
> Currently, DataSet API has two overloaded `write` methods for using
> FileOutputFormat as output format, and both require a path parameter, but the
> output path could already be set in the FileOutputFormat object. What's more,
> the subclasses of FileOutputFormat mostly don't have default constructors and
> required a path parameter too, so users have to set output path twice in the
> code, like:
> {code:java}
> String output = "hdfs:///tmp/";
> dataset.write(new TextOutputFormat<>(new Path(output)), output);
> {code}
> So I propose to add another write helper method that requires no path
> parameter. May someone assign this issue to me?
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
