[
https://issues.apache.org/jira/browse/HBASE-20884?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16547922#comment-16547922
]
Mike Drob commented on HBASE-20884:
-----------------------------------
bq. Out of interest, any differences between our old encoding and this new one?
The old one would add new lines after 76 characters to be MIME-encoding
compatible (with comments in the implementation that this wasn't strictly
base64 encoding compatible), the new one doesn't do this. We were stripping out
the line breaks before anyway.
> Replace usage of our Base64 implementation with java.util.Base64
> ----------------------------------------------------------------
>
> Key: HBASE-20884
> URL: https://issues.apache.org/jira/browse/HBASE-20884
> Project: HBase
> Issue Type: Task
> Reporter: Mike Drob
> Assignee: Mike Drob
> Priority: Major
> Fix For: 3.0.0, 1.5.0, 1.2.7, 1.3.3, 1.4.6, 2.0.2, 2.1.1
>
> Attachments: HBASE-20884.branch-1.001.patch,
> HBASE-20884.branch-1.002.patch, HBASE-20884.master.001.patch
>
>
> We have a public domain implementation of Base64 that is copied into our code
> base and infrequently receives updates. We should replace usage of that with
> the new Java 8 java.util.Base64 where possible.
> For the migration, I propose a phased approach.
> * Deprecate on 1.x and 2.x to signal to users that this is going away.
> * Replace usages on branch-2 and master with j.u.Base64
> * Delete our implementation of Base64 on master.
> Does this seem in line with our API compatibility requirements?
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
