> Oops. I didn't find that. Thanks for the pointer. (Thinks:
> does it work with HUGS?)
It works with the latest hugs, November 2002 (use "hugs +N" to get the
hierarchical libraries).
> A significant amount of my work went into the test cases and
> matching the
> parsing code against the (revised work-in-progress) URI spec,
> so maybe
> there's some residual value there...?
When I wrote Network.URI I took the grammar from the RFC, and also a
bunch of test cases, which are in GHC's testsuite
(fptools/testsuite/tests/ghc-regress/lib/net/uri001.hs). It might not
be up to date with the latest rev of the spec, however. Any extra
testing you can do and/or comments you have on the API would be
appreciated.
One thing I know is questionable is that parseURI returns a Maybe URI,
when in fact there are no strings for which it can return Nothing
(*everything* is a valid URI, it seems!).
Cheers,
Simon
_______________________________________________
Haskell mailing list
[EMAIL PROTECTED]
http://www.haskell.org/mailman/listinfo/haskell
