On 08/30/11 14:11, Danek Duvall wrote:
Brock Pytlik wrote:
https://cr.opensolaris.org/action/browse/pkg/bpytlik/18872-v1
publisher.py:
- line 2277: this implies that if we could move the CRL into place, we'd
use the CRL from disk. But we never remove entries from __tmp_crls, so
that's not actually true (within one instance of this object). I'd
correct the comment, since it seems to me that returning the CRL from
memory is better than going to disk.
Actually, the comment is correct. The code from 2254-2272 handles the
case where we previously moved the crl into place successfully.
I agree that not rereading the crl file from disk during the same
operation is probably a good thing, I'll rework this a bit more so that
we're always populating __tmp_crls.
Brock
Danek
_______________________________________________
pkg-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/pkg-discuss
