Hi,

On Monday 19 March 2012, meik michalke wrote:
> > 2) For all functions starting with a dot, add some marker like '#TODO
> > REVIEW EXPORT' on the '@export'-line to remind us that we will have to
> > check whether it is really necessary to export these.
> 
> i've tried to separate supposed internal functions already, by placing them
> in "internal*.R" files. so we can just check all functions in these files
> later, and if they need to be exported, rather move them to some other
> file.

good point. So essentially, any '@export'-tag inside an 'internal*.R' file 
signifies a TODO. Yes, I think I agree that should be good enough.
 
> > 4) Over time, we will take a look at the functions marked for review, one
> > by one, and check whether / how they can be hidden from the export. (Is
> > there a corresponding tag like '@internal', '@private', '@noexport' or
> > something in roxygen? Otherwise, perhaps we should make up one
> > ourselves, to avoid ambiguities).
> 
> the mechanism is that (only) functions/methods/classes marked "@export"
> will be exported (i.e. available to the user), and everything else remains
> internal without the need for further tagging. so once you remove the
> "@export" tag, a function will already be hidden.

Yes, I was aware of that. I was looking for a tag saying "somebody has 
actually looked at this, and concluded it can really remain hidden". But then 
I guess this may not be necessary after all. Once the NAMESPACE file is active, 
forgetting to add an '@export' to a newly added function should result in an 
obvious failure, anyway.

> that is, basically all there is left to do is to add the line "#' @export"
> before each function definition that is not tagged yet, and call
> roxygenize().

Yes. Would you do that?

Regards
Thomas

Attachment: signature.asc
Description: This is a digitally signed message part.

------------------------------------------------------------------------------
This SF email is sponsosred by:
Try Windows Azure free for 90 days Click Here 
http://p.sf.net/sfu/sfd2d-msazure
_______________________________________________
RKWard-devel mailing list
RKWard-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/rkward-devel

Reply via email to