[
https://issues.apache.org/jira/browse/HADOOP-2664?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Chris Douglas updated HADOOP-2664:
----------------------------------
Status: Open (was: Patch Available)
-1
I'm pulling this back. The writes from the cstr and its (related) silent
incompatibility with SequenceFile are sufficient to prevent it from being
checked in. It reads and writes lzop-compatible files, but it is inadequate as
a general compression codec. SequenceFile explicitly checks for a non-native
version of GzipCodec, but surely there's a better way to effect this.
That said, it should be noted that one can still write ".lzo" files from
LzoCodec that aren't. The incompatible change in this patch- that asserts
precedence for the .lzo extension and changes the former to .lzo_deflate -
should be considered for 0.17 regardless of what happens with this patch.
> lzop-compatible CompresionCodec
> -------------------------------
>
> Key: HADOOP-2664
> URL: https://issues.apache.org/jira/browse/HADOOP-2664
> Project: Hadoop Core
> Issue Type: New Feature
> Components: io
> Reporter: Chris Douglas
> Assignee: Chris Douglas
> Fix For: 0.17.0
>
> Attachments: 2664-0.patch, 2664-1.patch
>
>
> The current lzo codec is not compatible with the standard .lzo file format
> used by lzop.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
