You are viewing a plain text version of this content. The canonical link for it is here.
Posted to mapreduce-user@hadoop.apache.org by Charles Earl <ch...@gmail.com> on 2012/02/14 17:49:11 UTC

Extending Pipes, context.getInputValue() to support read of binary file data

Hi,
I'm trying to extend the pipes interface as defined in Pipes.hh to support the read of binary input data. 
I believe that would mean extending the getInputValue() method of context to return char *, which would be buffer containing binary data.
I'm guessing the best way  to do this would be to use a custom InputFormat on the java side that would have BytesWritable values.
Does this sound reasonable, and are there examples around?
C