[
https://issues.apache.org/jira/browse/PIG-1505?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12891347#action_12891347
]
Richard Ding commented on PIG-1505:
-----------------------------------
Thank you for the update. A few more comments:
* According to Pig Latin manual, user can also register additional files (to
use with user's Pig script) via the command line using the
-Dpig.additional.jars option (in addition to the REGISTER statement inside a
Pig script). I suggest you call FileLocalizer.fetchFile from the shared method
PigServer.registerJar so both cases will be covered.
* Can you change the method signature to
{code}
public static FetchFileRet fetchFile(Properties properties, String filePath)
throws IOException
{code}
The reason is that we have deprecated all other public methods on FileLocalizer
which has DataStorage as a parameter (so we can deprecate DataStorage in the
future). I think this is safe since the condition in the method
{code}
((fileUri.getScheme() == null) && (dfs == null))
{code}
is not used in the patch.
* You need to add a unit test in the patch (by first copying a Pig script to
the mini-cluster).
* Finally, since this is a new feature, can you add a release note (On jira,
there is a Release Note field) so that it will be incorporated in the next Pig
release notes.
> support jars and scripts in dfs
> -------------------------------
>
> Key: PIG-1505
> URL: https://issues.apache.org/jira/browse/PIG-1505
> Project: Pig
> Issue Type: Improvement
> Reporter: Andrew Hitchcock
> Assignee: Andrew Hitchcock
> Attachments: pig-jars-and-scripts-from-dfs-3.patch,
> pig-jars-and-scripts-from-dfs-trunk-1.patch,
> pig-jars-and-scripts-from-dfs-trunk-2.patch,
> pig-jars-and-scripts-from-dfs-trunk.patch
>
>
> Pig can't operate on files stored in Amazon S3.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
