Shawn Walker wrote:
Bart Smaalders wrote:
http://cr.opensolaris.org/~barts/12817/
12817 Traceback installing a package which depends(?) on an installed
package from a disabled publisher
13195 pkg(5) SAT solver can spend too much time trying to converge
fixes various problems w/ updating project gates while switching
repos, and incidentally prints better diagnostics when install -v
fails.
tests/cli/t_pkg_install.py
--------------------------
line 2938: so this tests disabled publisher, but not really a deleted
(removed) (unset-)publisher; could you change this to check both cases?
line 2944-2945: leftover debugging?
Output Nit
----------
From the test above, when "pkg install [email protected]" fails, the user see
this message:
Planning for install failed:
Use -v option for more details
pkg: No version of baz can be installed:
pkg://test1/[email protected],5.11-0:20091217T172947Z: Suitable required
dependency pkg:/[email protected],5.11 cannot be found
However, if you actually provide -v, you don't see any more detail:
Planning for install failed:
pkg: No version of baz can be installed:
pkg://test1/[email protected],5.11-0:20091217T172923Z: Suitable required
dependency pkg:/[email protected],5.11 cannot be found
Is the -v bit just a general message or is there smoething more missing?
Otherwise, LGTM.
I'll improve the messaging... when we catch the problem very early
before we go to the solver, no additional output is available
(or should be needed).
- Bart
--
Bart Smaalders Solaris Kernel Performance
[email protected] http://blogs.sun.com/barts
"You will contribute more with mercurial than with thunderbird."
_______________________________________________
pkg-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/pkg-discuss
