[
https://issues.apache.org/jira/browse/SPARK-14663?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Sergey updated SPARK-14663:
---------------------------
Description:
I am trying to specify
spark.hadoop.textinputformat.record.delimiter in spark-defaults.conf, namely,
to set it to "\n" (the #10 character). I know how to do it in
sc.newAPIHadoopFile, but I'd like to set it in configuration, so I can keep
using sc.textFile (because it also works with zipped files).
However, I can't find a way to accomplish it.
I have tried
spark.hadoop.textinputformat.record.delimiter \n
spark.hadoop.textinputformat.record.delimiter '\n'
spark.hadoop.textinputformat.record.delimiter "\n"
spark.hadoop.textinputformat.record.delimiter \\n (that's two slashes and the
letter n)
spark.hadoop.textinputformat.record.delimiter (just pressing enter)
None of them works. I check in sc._conf.getAll(), and none of them gives me the
right result.
was:
I am trying to specify
spark.hadoop.textinputformat.record.delimiter in spark-defaults.conf, namely,
to set it to "\n" (the #10 character). I know how to do it in
sc.newAPIHadoopFile, but I'd like to set it in configuration, so I can keep
using sc.textFile (because it also works with zipped files).
However, I can't find a way to accomplish it.
I have tried
spark.hadoop.textinputformat.record.delimiter \n
spark.hadoop.textinputformat.record.delimiter '\n'
spark.hadoop.textinputformat.record.delimiter "\n"
spark.hadoop.textinputformat.record.delimiter \\n
spark.hadoop.textinputformat.record.delimiter (just pressing enter)
None of them works. I check in sc._conf.getAll(), and none of them gives me the
right result.
> Parse escape sequences in spark-defaults.conf
> ---------------------------------------------
>
> Key: SPARK-14663
> URL: https://issues.apache.org/jira/browse/SPARK-14663
> Project: Spark
> Issue Type: Bug
> Components: Spark Core
> Affects Versions: 1.6.1
> Reporter: Sergey
>
> I am trying to specify
> spark.hadoop.textinputformat.record.delimiter in spark-defaults.conf, namely,
> to set it to "\n" (the #10 character). I know how to do it in
> sc.newAPIHadoopFile, but I'd like to set it in configuration, so I can keep
> using sc.textFile (because it also works with zipped files).
> However, I can't find a way to accomplish it.
> I have tried
> spark.hadoop.textinputformat.record.delimiter \n
> spark.hadoop.textinputformat.record.delimiter '\n'
> spark.hadoop.textinputformat.record.delimiter "\n"
> spark.hadoop.textinputformat.record.delimiter \\n (that's two slashes and
> the letter n)
> spark.hadoop.textinputformat.record.delimiter (just pressing enter)
> None of them works. I check in sc._conf.getAll(), and none of them gives me
> the right result.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
