-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 roger peppe wrote: > I was wondering if it was possible to implement synchronous channels > within STM. In particular, I'd like to have CSP-like send and recv primitives > on a channel that each block until the other side arrives to complete > the transaction. > > I think I've convinced myself that it's not possible, but > anyone care to differ?
This sounds similar to what I once tried for a FRP implementation. I decided that what I was trying to do was impossible (at least without recalculating things all the time whenever something retries), and then constructed an IVar implementation which was working very elegantly (before I decided to change the core ideas behind my FRP implementation). As luck would have it, Luke Palmer recently added his own IVar implementation to Hackage, which you may find a use for. - - Jake -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkjtMQwACgkQye5hVyvIUKk39QCcDdsFY7z0k2xwkJ/zBK3tVsbT n28An2D9LzUfWiWTpKIU7J7tt/VGs4EB =kQGz -----END PGP SIGNATURE----- _______________________________________________ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe