I only have superficial meta-level comments.  I'm assuming my job now is
to wait for you to submit another version of this patch (bureaucrat mode)

On Wed, May 13, 2009 at 22:03:28 +0200, Reinier Lamers wrote:
> > Have you looked at
> > the HUnit framework?  You may be reinventing the wheel.
> 
> At least there are no test frameworks that allow running tests in parallel, 
> which is my I started writing my own.

It would be really good is if we could make this a test framework
that everybody wants to use :-)

I'm not sure what our policy should be on external modules that get
spun out.  I guess it's up to their maintainers to set their own
policies on what patches they apply.  But then what should darcs
do with versions of those packages?  I guess we'd have to do the
usual case-by-case thinking.  For example, we probably want to be
very liberal about which version of the tuxtester package we use.
 
> > I haven't yet seen another function in your patch that uses underscores, so
> > even this would be preferrable to me, initialRunnerState. This is a minor
> > nitpick and I wouldn't want to halt the acceptance of these patches on this
> > item.
> 
> There is a page on the wiki that tells me to use underscores in local 
> functions and camelCase in exported ones.

Note that since you are planning on making this a separate package
in the future, it would make sense to just use your own convention
here, most likely the de facto Haskell camelCase standard
 
> Or to be precise, http://wiki.darcs.net/index.html/CodingStyle says: "favour 
> camelCase for top level functions (instead of the current camel/underscore 
> hodgepodge)
> 
>     * yes, underscores can be useful in certain situations, don't be 
> overzealous
>     * In the existing code, camelCase is typically used for names that are 
> exported from modules and underscores for names that are internal to a 
> module. 
> "

I am wiling to see this change to all camelCase if this is what people
what, but I won't push for it.  The motivation for such a switch is that
we then have a really simple policy (camelCase with rare exceptions
where the underscores are useful).

-- 
Eric Kow <http://www.nltg.brighton.ac.uk/home/Eric.Kow>
PGP Key ID: 08AC04F9

Attachment: pgp52cng80V6J.pgp
Description: PGP signature

_______________________________________________
darcs-users mailing list
[email protected]
http://lists.osuosl.org/mailman/listinfo/darcs-users

Reply via email to