On Thu, Jul 24, 2008 at 04:09:43PM -0700, Brad Hall wrote:
> Link to new CR: http://cr.opensolaris.org/~bhall/bug-2627-1/
We use getopt in the rest of the code, rather than optparse. Unless
there's something that optparse really buys you, please use getopt.
I think -p is more consistent with the rest of solaris than -m (p for
parseable). Particularly your use of "mach" as a short form throughout, it
strongly means "machine type" to me, as in the command "mach".
Is it worth assuming that the baseline for a test is "pass", and keeping
the baseline files shorter?
baseline.py:
- line 49: perhaps just make the generate default to False, and pass in
True instead of 1? Or you could do bool(generate) here. Similarly on
line 74, I think.
- line 63: return self.results.get(name, None)
- line 87: "if lst"
- line 105, 124: I don't think this work. This will catch exceptions of
type IOError or type e.
- line 106: "storing"
- line 130: no need for parens on the lhs
Danek
_______________________________________________
pkg-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/pkg-discuss
