Taras Bobrovytsky has posted comments on this change.

Change subject: IMPALA-2904: Support INSERT and LOAD DATA on S3 and between 
filesystems
......................................................................


Patch Set 5:

(2 comments)

http://gerrit.cloudera.org:8080/#/c/2574/5/tests/common/impala_test_suite.py
File tests/common/impala_test_suite.py:

Line 120: generic_client
> We need hdfs_client as some pure HDFS tests use that. Do you think we shoul
Yes, I think it would be much cleaner if we only had filesystem_client and got 
rid of hdfs_client. The test should not worry what the underlying filesystem is 
(hdfs or S3). (also imagine if we add support for many other filesystems, 
things will get complicated). Also may be a good idea to get a second opinion 
about this.


http://gerrit.cloudera.org:8080/#/c/2574/5/tests/util/hdfs_util.py
File tests/util/hdfs_util.py:

Line 119:   def exists(self, path):
> This function needs to be there because PyWebHdfsClient implements exists()
I'm a bit confused. Where is super(PyWebHdfsClientWithChmod, self).exists(path) 
defined anyways? I can't find it in the module:
https://github.com/pywebhdfs/pywebhdfs/blob/master/pywebhdfs/webhdfs.py

The fact that it's complaining that you didn't override the abstract method 
exists makes me a little worried, which code will get executed when someone 
calls this method? Does this even work?


-- 
To view, visit http://gerrit.cloudera.org:8080/2574
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-MessageType: comment
Gerrit-Change-Id: I94e15ad67752dce21c9b7c1dced6e114905a942d
Gerrit-PatchSet: 5
Gerrit-Project: Impala
Gerrit-Branch: cdh5-trunk
Gerrit-Owner: Sailesh Mukil <[email protected]>
Gerrit-Reviewer: Henry Robinson <[email protected]>
Gerrit-Reviewer: Michael Brown <[email protected]>
Gerrit-Reviewer: Sailesh Mukil <[email protected]>
Gerrit-Reviewer: Taras Bobrovytsky <[email protected]>
Gerrit-HasComments: Yes

Reply via email to