On Dec 2, 2007, at 11:50 AM, Toby Thain wrote:
On 29-Nov-07, at 3:06 PM, Paul J. Lucas wrote:
On Nov 28, 2007, at 11:31 PM, Sumit Lohia wrote:
Also I can code against interfaces that know nothing about
Restlets so don't
have to add 'throws RestletException' even though the
implementation might
throw a derived exception like NotFoundException.
RestletException could conceivably be derived from IOException.
An interface that didn't anticipate that an implementation might
need to do some kind of I/O as part of said implementation wasn't
designed with sufficient foresight, IMHO.
Maybe I misunderstood... But surely you can't mean an exception
with semantics such as 'http resource not found' (or forbidden,
etc) - because those are not i/o related but resource related.
Funny, because java.io.FileNotFoundException is derived from
IOException. So clearly there's precedent.
What kind of thing would a RestletException subclassed from
IOException be used for?
The same things as if it weren't derived from IOException that have
already been mentioned. However, if it is derived from IOException,
it will integrate better with existing APIs.
- Paul
