[jira] [Commented] (ARROW-13141) [C++][Python] HadoopFileSystem: automatically set CLASSPATH based on HADOOP_HOME env variable?

["Antoine Pitrou (Jira)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/ARROW-13141?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17367182#comment-17367182 ] 

Antoine Pitrou commented on ARROW-13141:
----------------------------------------

We could do that in C++, but I have two concerns. 1) It adds some complication and also runtime cost (spawning an external subprocess). 2) It sets an environment variable from a library.

Another thing we can do is simply document how to set the CLASSPATH from a command line, which is actually simple. For example on a Unix shell:
{code}
export CLASSPATH=$($HADOOP_HOME/bin/hadoop classpath --glob)
{code}

On a Windows shell this might be (untested):
{code}
%HADOOP_HOME%/bin/hadoop classpath --glob > %CLASSPATH%
{code}




> [C++][Python] HadoopFileSystem: automatically set CLASSPATH based on HADOOP_HOME env variable?
> ----------------------------------------------------------------------------------------------
>
>                 Key: ARROW-13141
>                 URL: https://issues.apache.org/jira/browse/ARROW-13141
>             Project: Apache Arrow
>          Issue Type: Improvement
>          Components: C++, Python
>            Reporter: Joris Van den Bossche
>            Priority: Major
>              Labels: filesystem, hdfs
>
> In the "legacy" python-specific HadoopFileSystem implementation, we have a {{_maybe_set_hadoop_classpath}} function which has some logic to set the {{CLASSPATH}} environment variable based on {{HADOOP_HOME}} or the hadoop executable: https://github.com/apache/arrow/blob/c43fab3d621bedef15470a1be43570be2026af20/python/pyarrow/hdfs.py#L134-L149
> This is also mentioned in the documentation of the new HadoopFileSystem (https://arrow.apache.org/docs/python/filesystems.html#hadoop-file-system-hdfs):
> > If CLASSPATH is not set, then it will be set automatically if the hadoop executable is in your system path, or if HADOOP_HOME is set.
> However, this sentence was probably simply copied over from the docs about the legacy filesystem. And for the new HadoopFileSystem implementation, we don't have this logic to automatically set up {{CLASSPATH}}. 
> Do we want to add this logic to the new implementation as well? (in cython, or actually in C++?) Or if not, we should update the docs to clarify that {{CLASSPATH}} is actually required.
> cc [~apitrou]



--
This message was sent by Atlassian Jira
(v8.3.4#803005)