Noel Yap wrote:

        [...]

> Taking another stance, "cvs add dir" will still act on files within dir.
> The
> difference with my proposition is that "dir/CVS" will be created whether
> or not
> there're files within the hierarchy.  And, "dir" will get created upon
> "cvs ci"
> whether or not there're files within the hierarchy.  There's absolutely no
> reason to have additional complicated code to do this sort of checking.
> 
        [...]

        Well, I suppose I've said it before...

        There is slightly more than "absolutely no reason".  If 
        you consider the differing behavior of "cvs co" on branches 
        vs. "cvs co" on the trunk with regard to empty
        directories a bug, and you consider the proper fix to 
        have the trunk behave as branches currently do...(that is,  
        "-P" is always in force.), as I do, rather than
        the other way around, as I gather you do...

        But, I think we simply disagree on this point...

        Anyway, my opinion: I have no problem with "cvs 
        add emptydir" creating  emptydir/CVS etc., but 
        I'd prefer emptydir not get created in the repository
        at "cvs ci" time unless there are files in it...
        however, even if that happens it would be alright,
        if "cvs update" were fixed so -P is the default.

        And if "cvs update" isn't fixed, well, the new 
        "cvs add" would still be an improvement over
        the status quo, so how can I complain?
        (yet I still do :-)

        -- steve

Reply via email to