I'll add a review comment of my own: the use of osnet.publish in
usr/src/pkg/transforms/publish is leftover from when this was
expressed differently.  I should probably get rid of it, and just
abort directly from the architecture match.

> My caveat: I'm really quite ill and haven't managed a terribly
> coherent or thorough review.  But, I'd rather see this go back than
> wait on me since we still have a full workspace code review to go,
> and I'll also be in there again soon so may have more comments then.

Get better.

> - Yes, the consolidation dependency should be in the transforms.  I
>   just hadn't gotten around to doing it, so thanks.

Thanks for the confirmation.

> - Comments in the makefile are good, again thanks.

You're welcome, and thanks for the confirmation.

> - I'm not at all convinced that we want any incorporation at all for
>   the 'extra' packages.  But, I can't articulate well why right now.

Should I leave it or take it out?  I don't really like the thought
that a package is arbitrarily unconstrained as a function of a legal
or political boundary, rather than a technical one.

> - For the depend-on-incorporation, I was thinking pkgmogrify would
>   get a directive that said "if you see this pattern, *add* this
>   new action" rather than transform the existing action.  (Multiple
>   matches would just add multiple of the same new action.)

This would be a significant RFE to pkgmogrify: it would violate the
1:1 input to output line constraint.  Currently, a transform may
change only the current action, and no other.

> - Extensive shell syntax in the makefile does make me nervous.  What
>   about users who use csh and its derivatives?

Like me?  :)  I have tested this with tcsh as a user shell, and can do
others if folks want.  ISTR John running into an issue with something
like that a while back, with one of the (thankfully) temporary hacks.
John, what shell do you use?

> That's it so far.

Thanks.

--Mark
_______________________________________________
pkg-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/pkg-discuss

Reply via email to