[ 
https://issues.apache.org/jira/browse/AVRO-386?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12828889#action_12828889
 ] 

Philip Zeyliger commented on AVRO-386:
--------------------------------------

Jeff, when JIRA was acting up, suggested the following feedback:

-----------

Some comments:

* Remove the TODO's that you implemented
* In line 143, you set {{uncompressed_data = self.buffer_writer.getvalue()}}; 
use this variable in line 149 instead of calling 
{{self.buffer_writer.getvalue()}} again?
* Extra newline added after {{self._meta = {}}} in DataFileWriter#__init__: 
remove
* Will "assert" raise a DataFileException? I haven't tried it.
* Default "codec" to "'null'" instead of "None" in DFW#__init__?
* What are you up to in line 213 with {{self.codec = self.codec}}?
* Line 263 is wider than 80 characters
* Line 281: if you're skipping a long, why not use "skip_long()"? Not a big 
deal.
* Line 285: comment on what the argument of "-15" means to zlib.decompress

Thanks for tackling this one, Philip!

> Python implementation of compression
> ------------------------------------
>
>                 Key: AVRO-386
>                 URL: https://issues.apache.org/jira/browse/AVRO-386
>             Project: Avro
>          Issue Type: Improvement
>          Components: python
>    Affects Versions: 1.3.0
>            Reporter: Philip Zeyliger
>            Assignee: Philip Zeyliger
>         Attachments: AVRO-386.patch.txt
>
>
> AVRO-135 introduced compression.  AVRO-368 changed some metadata keys.  The 
> patch to follow adds this functionality to the python implementation.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

Reply via email to