On Mon 02 Feb 2009 at 09:28AM, Danek Duvall wrote:
> On Mon, Feb 02, 2009 at 03:42:17AM -0800, Dan Price wrote:
>
> > http://cr.opensolaris.org/~dp/deploy-fixes
>
> setup.py:
>
> - line 392: will the use of -R/usr/sfw/lib put this in the openssl
> modules even on systems where it's not required? Or are you simply
> depending on the fact that we're now using the pyopenssl delivered in
> Nevada as of build 106?
Yes, it will. That's harmless. -R just adds something to the linker's
runtime path. Before 106, we don't have pyopenssl, and before (I think)
b104, we don't have openssl in /usr/lib at all, it's in /usr/sfw.
So all we're doing here is adding the -R for those systems on which
we still have to build pyopenssl for ourselves.
> - line 490: this seems a bit sketchy. Why is it okay to do this here?
> ... Never mind, I see what you're doing.
I pulled the syntax check out into a separate routine, and added
some comments. Eventually I would like to run the syntax check
over publish.py, etc., too.
> - line 491: why not catch a PyCompileError?
Fixed, thanks for noticing this.
> - line 494: you don't need the str() call here.
Fixed.
> svc-pkg-depot:
>
> - line 119: you don't need to fork off grep to do these checks. This and
> the two ssl-key-file ones can simply be done with
>
> [[ $bool_ops == *'--readonly'* ]]
>
> and I'm pretty sure you can do the ssl-dialog check with
>
> [[ $option_ops == *--ssl-dialog=@(smf|exec):* ]]
>
> which should work with ksh88, too. Maybe a "+" instead of the "@".
Ok, fixed.
-dp
--
Daniel Price, Solaris Kernel Engineering http://blogs.sun.com/dp
_______________________________________________
pkg-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/pkg-discuss
