On Tue, 21 Sep 1999, Alastair Reid wrote:
>
> Metanote: I'm not sure how Debian package management are supposed to
> work but I'd have thought Antti-Juhani Kaijanaho should be receiving
> bug reports about the packaging and things which might be related to
> the packaging. Both of Havoc's reports are very clearly general
> problems in Hugs.
>
The idea is that package maintainers know how to report bugs for the
individual package, and they maintain contact with the upstream
distributors, so you report to Debian and it's forwarded by the Debian
maintainer if appropriate.
If you think about it, someone who's installed Debian's 2000 packages
hasn't been to the Hugs web site and isn't familiar with the Hugs bug
reporting procedure; especially for some more obscure packages without
official bug report mechanisms, it's useful to have the maintainer as
intermediary to be sure things are routed properly. Then Debian users only
have to learn one bug reporting system.
I can mail hugs-bugs directly in the future of course, if that's the
preferred thing.
> Or, to put it another way... before we can implement the spec, we
> first have to figure out what the spec means. Ideally, this means
> that for each operation we specify exactly how errno values (or
> whatever) map onto IOError values (for each operation?) for every
> platform we care about. (I suppose this should be added to the
> "improve IO libraries" entry in the wishlist.)
>
What's the appropriate forum for this? I can make up a suggested mapping
if it's helpful. But I know little about Haskell. Does ghc have some
mapping already in place?
Taking my cheesy little program as a case study, there are two clear
impracticalities in the Hugs implementation: this error handling
issue, and the fact that strings are written with a write() for every
character. Both of those are show-stoppers for "practical" applications.
The error thing is the most important since presumably "practical" apps
are compiled for production use instead of using Hugs, but the errno has
to be fixed so you can test using the interpreter, then compile later, and
get the same semantics in both cases.
Thanks for the prompt replies from Antti-Juhani and the Hugs people!
Havoc
