Re: DO NOT REPLY [Bug 19868] - Need the InterruptedIOException when timeout

[Oleg Kalnichevski <o....@dplanet.ch>]

I agree it does seem to be an overkill in this particular case. But I
personally find the flexibility offered by chained exceptions worth
paying the price. This might wait until 2.1, though.

You are absolutely right about the overall concept of exception
handling  getting a bit vague. There used to be a pretty clear
convention before with HttpException representing protocol errors and
IOException representing transport errors. It got lost somehow. 

I think we do not have to invent much. We just have to be a bit more
stringent about respecting the convention that has already been
established. 

Cheers

Oleg

On Wed, 2003-05-14 at 21:48, Jandalf wrote:
> The SomeOtherGawdAwefulException was just an example of continuing the 
> idea to have another Exception type possibly contained in the 
> ChainedException class.  We are not going through all this work just 
> because there will only ever be one chained exception, IIOE, right?
> 
> I'm not against exception chaining, but we are taliking about changing 
> the semantics of the exception handling for the sake of exposing another 
> exception type while preserving the interface.  It just does not feel 
> good over here; it does not seem fully motivated.
> 
> Thinking about this a bit more, we really do not have any "Exception 
> rules" for exceptions thrown out to user code.  We just have some casual 
> conventions that determine which exceptions are thrown out, which are 
> converted to HttpClient types and which are swallowed.
> 
> 
> 
> bugzilla@apache.org wrote:
> 
> >DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG 
> >RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
> ><http://nagoya.apache.org/bugzilla/show_bug.cgi?id=19868>.
> >ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND 
> >INSERTED IN THE BUG DATABASE.
> >
> >http://nagoya.apache.org/bugzilla/show_bug.cgi?id=19868
> >
> >Need the InterruptedIOException when timeout
> >
> >
> >
> >
> >
> >------- Additional Comments From olegk@apache.org  2003-05-14 19:37 -------
> >Actually I do not. The problem is that HttpClient#executeMethod may not throw
> >SomeOtherGawdAwefulException (unless it's a runtime exception or a sub class of
> >either IOException or HttpException). The only way to get hold of the original
> >exception along with all its properties is to chain it. The alternative would be
> >to implement all the possible exception types as sub classes of HttpException,
> >which is in my opinion is as ugly. Exception chaining is the most flexible way
> >of dealing with heterogeneous exceptions, and flexibility usually carries a
> >price tag. 
> >
> >Oleg
> >
> >---------------------------------------------------------------------
> >To unsubscribe, e-mail: commons-httpclient-dev-unsubscribe@jakarta.apache.org
> >For additional commands, e-mail: commons-httpclient-dev-help@jakarta.apache.org
> >
> >
> >  
> >
> 
> 
> 
> 
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: commons-httpclient-dev-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: commons-httpclient-dev-help@jakarta.apache.org
>