Brock Pytlik wrote:
Dan Price wrote:
On Fri 22 Jan 2010 at 02:16PM, Brock Pytlik wrote:
Brock Pytlik wrote:
Webrev:
http://cr.opensolaris.org/~bpytlik/ips-14020-v1/
Bugs:
14020 test for alt python versions shouldn't hardcode paths
This makes the tests for 13059 and python combinations find the
paths that should appear in the results dynamically since it
varies from system to system. It's been tested on the system that
originally demonstrated the problem and seems to fix the issue.
Thanks,
Brock
_______________________________________________
pkg-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/pkg-discuss
Now with comments:
http://cr.opensolaris.org/~bpytlik/ips-14020-v2/
Sorry-- one added annoying thought-- could this be commonized
to python2.6 as well? That is to say, does this combine
cleanly with make_res_payload_1()?
-dp
In short, yes, but I'm not sure it wouldn't make things more
confusing. The biggest difference is that, because the 2.6 analysis
takes place without a fork, it happens in the environment of the test
suite (which means that the proto area of the workspace is in
sys.path) while the fork for peforming analysis of 2.4 and 2.5 clears
that out. It would make get_ver_paths more complicated, but might
improve the ability to seamlessly move forward to the next python
version.
I'll work up what things would look like if we went that direction and
see what I think.
Brock
_______________________________________________
pkg-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/pkg-discuss
Turns out this isn't too bad.
New webrev:
http://cr.opensolaris.org/~bpytlik/ips-14020-v3/
Thanks,
Brock
_______________________________________________
pkg-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/pkg-discuss
