Specify output file names when using hadoop streaming

[Usman Waheed <us...@opera.com>]

Hi,

I am a newbie to hadoop. I have a simple MAP only job but would like to 
output the results to different file names instead of the files 
part-0000x(s).
I would like to name these output files based on the log file that were 
processed.

As an example if I processed log1 and log2 i would like the map keys 
output from log1 to go to part-log1 and map keys output from log2 to 
part-log2 and so forth.
My mapper is in Perl and am using hadoop streaming.

 From the documentation I have read so far looks like i can do this in 
Java but was wondering if this is doable via hadoop streaming also.

Any clues would help.

Thanks,
Usman