On Thu, Sep 27, 2012 at 12:53 PM, Iustin Pop <[email protected]> wrote:
> On Thu, Sep 27, 2012 at 11:28:42AM +0200, Agata Murawska wrote:
>> On Wed, Sep 26, 2012 at 9:23 AM, Iustin Pop <[email protected]> wrote:
>> > On Tue, Sep 25, 2012 at 06:43:44PM +0200, Agata Murawska wrote:
>> >> This was repeated a lot, so we should abstract this into a type.
>> >
>> > LGTM; the ERPCError is a bit too sneaky though.
>> Agreed, but I can't think of a nicer name - RpcErrorOrA? ;)
>
> Me neither… what about forgetting the error part? 'Either RpcError' is a
> container for either an error or an actual result, so RpcResult? (Guess
> you already have that) RpcContainer?
Ha! Not any more ;) But this is not accurate, because we sometimes put
just a string (http response string to be exact) there for Right type,
or even an http request.

>
> Anyway, leave it as such for now, we can clear it later.
I did leave it, but now I'm doing all the realignments and such,
including turning this thing, whatever the name, to a monad ;) So it
sort of jumped at me ;) But I can leave it for now, just thought you
might have an idea..

>
> iustin

Reply via email to