[darcs-devel] darcs patch: Enable better progress reporting when ge... (and 1 more)

[David Roundy]

On Thu, Mar 27, 2008 at 04:01:57PM +0100, Nicolas Pouillard wrote:
> Excerpts from David Roundy's message of Thu Mar 27 15:43:50 +0100 2008:
> > I've written a patch to handle this.  That cabal repository is
> > certainly a tedious one to grab!
> 
> Thanks.
> 
> However  perhaps you could try this patch too, since yesterday that I use it I
> feel that it's not too verbose.

The patch you wrote restores earlier behavior that I removed precisely
because it was confusing.  It wasn't that it was too verbose, the problem
is that these low-level progress messages end up interfering with the more
useful high-level messages.

> > True, but I'd rather fix those bugs than hide them.
> 
> Why not add an awful debug message (or a failure) when "s < t" since this should not happen.

Not too bad an idea, except that it's not something I'm particularly
concerned about.

> > > Concerning progressList, I found it quite nice.  I've tried with
> > > {begin,end}Tedious + finishedOneIO at first but failed to see anything.
> > 
> > Indeed, progressList is much easier to use.  The problem is that it just
> > doesn't handle the counting properly (unless you only use one identifier
> > per list), and that's a major drawback.
> 
> Perhaps progressList could have one of these types:
> 
>   -- a title, a list of names, a list
>   progressList :: String -> [String] -> [a] -> [a]
> 
>   -- a title, a list of names and values
>   progressList :: String -> [(String, a)] -> [a]
> 
>   -- a title, a function that compute a name given an element, and a list
>   progressList :: String -> (a -> String) -> [a] -> [a]

I'm not sure how any of these would help us.  Would they, or would they not
show a progress counter? If they do, then they'd be just as buggy as the
existing progressList.  If they don't, then they will be less useful than
the existing progressList.  Unless you've also got a redesign of the
progress-reporting framework in mind.
-- 
David Roundy
Department of Physics
Oregon State University
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: Digital signature
Url : http://lists.osuosl.org/pipermail/darcs-devel/attachments/20080327/d41346c9/attachment.pgp