I bet with careful regex you can just process the httpd log without needing
a convertor. Or simply awk it to a CSV format



On Mon, Mar 15, 2010 at 11:43 PM, Robin Anil <[email protected]> wrote:

> Anything can be a feature (entity) (String, Integer, Object).This can be
> done programmatically and you can write your own input transaction iterator.
>
> But the main entry point works only on String features, which is created by
> splitting the transaction file. So you just have to specify the right
> splititng pattern for your preprocessed/unprocessed log file.
>

Reply via email to