Re: errors and exceptions in xproc

James Fuller <james.fuller.2007@gmail.com> writes:
> A few thoughts on error processing in xproc:
>
> * has it been considered that try/catch throws exceptions and
> everything else errors, I like c:exception better versus c:error

Is there a semantic difference between error and exception, or would
you just like the spec better if we did a s/error/exception?

> * should a p:error have a "error" output port

What would it mean? A p:error always aborts execution, possibly captured
by an ancesor p:try.

If p:try does catch it, I expect the input to p:error to appear on the
p:catch "error" port, though perhaps that's not clear.

> * I was thinking of how nice it would be to output all error messages
> as well formed documents ... perhaps this could be achieved using a
> top level try/catch but I guess there is scope for this type of
> functionality as an extension attribute on p:pipeline and leave it to
> the implementator to provide

Given that errors appear on an error port, I exect them to be WF.

                                        Be seeing you,
                                          norm

-- 
Norman Walsh <ndw@nwalsh.com> | Two starving men cannot be twice as
http://nwalsh.com/            | hungry as one; but two rascals can be
                              | ten times as vicious as one.--George
                              | Bernard Shaw