On 05/08/12 03:40, Takeshi Asano wrote:
Hi Shawn,
Thank you for the fix.
The _varcet.c fix LGTM.
Regarding the test script, from the first half of
added part (#182-196), it seems that there's assumption
that nl_ZA is evaluated first.
Yes, this test relies on the fact that the way Python currently builds
the attributes dictionary for the action always results in the nl_ZA
facet being evaluated first.
That is the only way to trigger this bug.
I have added a comment indicating that.
But in that case, the latter part (from #198) seems to
pass regardless of the _varcet.c fix.
Yes, because there's no way for me to control the evaluation order.
This test is just there to ensure that if the evaluation order changes
that everything still works.
The issue is only seen when the first facet evaluated matches a wildcard
facet set to False. The order facets are set in the image doesn't matter.
-Shawn
_______________________________________________
pkg-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/pkg-discuss
