Hi Danek & Sean,

Thanks for the code reviews, I've updated the webrev at the same url.

On Wed, 2010-01-27 at 17:19 -0800, Danek Duvall wrote:
> Tim Foster wrote:
> 
> > http://cr.opensolaris.org/~timf/14038,14039,14133-pkgsend/
> 
> publish.py:
> 
>   - line 209, et al: shouldn't need str()

Yep.

>   - line 389: while it makes some sense to abandon a "publish" and return
>     an error, here in "include", I think it makes more sense to just keep
>     going after warning.  There's no way to undo just what succeeded in the
>     incomplete "include", and there's no way to resume where you left off,
>     so you're kinda stuck.  If you really want the command to fail, allow
>     it to be ignorable, by returning 3, which is the "partial success" code
>     we've been using.

I agree - I've done the last suggestion, warning (and not including the
"unknown" action in the transaction) and eventually returning 3.

>   - line 434, 455: no need for the parens around action and err.

Ok.

>   - line 436: in keeping with "publish", shouldn't you be abandoning the
>     transaction and returning an error?

Yes, I've done that.

> t_pkgsend.py:
> 
>   - why use a dictionary with ".1", ".2" strings, rather than creating an
>     empty list before line 560, and appending each path to the list as you
>     create it?  What's there seems convoluted, if correct.

Yep, that looked weird to me too now that I look at it in the cold light
of day, sorry :-/

        cheers,
                        tim

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

Reply via email to