[Haskell] Re: Better Exception Handling

[Peter Simons]

Bayley, Alistair writes:

 > data SqliteException = SqliteException Int String
 >   deriving (Typeable)

 > catchSqlite :: IO a -> (SqliteException -> IO a) -> IO a
 > catchSqlite = catchDyn

 > throwSqlite :: SqliteException -> a
 > throwSqlite = throwDyn

I, too, think that's a good way to do it. And in the cases
where a custom data type seems to be overhead, I simply use
(fail "Foo.Bar: buffer overflow") and catch it with:

  isBufferOverflow :: IOError -> Bool
  isBufferOverflow e =
    (isUserError e) && (ioeGetErrorString e == "FooBar: buffer overflow")

Once you have attributes for individual exceptions, defining
them for arbitrary groups of exceptions is trivial. And the
fact that I _use_ the message internally encourages me to
use clear and consistent error messages which also make
sense in case someone just 'show's them.


 >> Python can work that way, but also adds another feature:
 >>
 >> try:
 >>   blah
 >>   moreblah
 >> finally:
 >>   foo

 > And in Haskell we have catch(Dyn), bracket, and finally.
 > Are these not enough?

We also have Control.Exception.try. :-)

Peter