Sorry Fedor, but again, I have to disagree :-)

>declaring implementation-specific exception in the API thus binding API
>to it's particular implementation

It's not an implementation-specific exception.  All the contrary, it's an
API-interface-specific exception.  And it does not bind the API to it's
implementation.  All the contrary, it shields the API from any changes to
its implementation.

-- Mathieu

-----Original Message-----
From: Fedor Karpelevitch [mailto:[EMAIL PROTECTED]]
Sent: February 21, 2002 12:21 PM
To: 'Turbine Developers List'
Subject: RE: [vote] Standardizing OM for throwing TorqueException


> Mathieu, you have my +1 to modify the Torque Java source and Velocity
> templates to throw TorqueException instead of Exception.  This is a
> fully backwards compatible change, and we can debate at length whether
> TorqueException should be a checked or unchecked exception _after_ the
> change is made.
>
>                              Thanks, Dan

Just to clarify: I agree that this is good (note - i did not vote against
it). My point is that this solves only part of the problem - that is,
throwing too generic exception, but does not solve the other problem that
is, declaring implementation-specific exception in the API thus binding API
to it's particular implementation. I am afraid there would have to be a
second round of changes and I do not see why same job needs to be done
twice.

fedor.

--
To unsubscribe, e-mail:
<mailto:[EMAIL PROTECTED]>
For additional commands, e-mail:
<mailto:[EMAIL PROTECTED]>


--
To unsubscribe, e-mail:   <mailto:[EMAIL PROTECTED]>
For additional commands, e-mail: <mailto:[EMAIL PROTECTED]>

Reply via email to