on an x86 machine glsa-check -f 200509-11 failes because net-libs/gecko-sdk-1.7.12 has been masked in favor of net-libs/gecko-sdk-1.7.12-r1 Reproducible: Always Steps to Reproduce: See Details Actual Results: !!! All ebuilds that could satisfy "=net-libs/gecko-sdk-1.7.12" have been masked. !!! One of the following masked packages is required to complete your request: - net-libs/gecko-sdk-1.7.12 (masked by: ~x86 keyword) Expected Results: If you replace line 196 of glsa-check with: exitcode = os.system("emerge '>="+pkg+"'") It will emerge '>=package' as opposed to '=package' and in this case geckp-sdk-1.7.12-r1 would have been installed instead
# $Header: /var/cvsroot/gentoolkit/src/glsa-check/glsa-check,v 1.6 2004/12/08 10:52:43
emerge sync, net-libs/gecko-sdk-1.7.12-r1 is marked stable on all ARCHs.
I did emerge sync! Apparently I was not clear enough glsa-check -f for fixing the bugs: uses the form "emerge '=packagename' " the problem with this is when a package is later masked/version bumped, then glsa-check trys to masked the exact package (which has now been masked) as opposed to allowing emerge to see that the revision will do just fine EX: YESTURDAY: package-version stable TODAY : package-version unstable, package-version-r1 stable I feel that having glsa-check use the form: emerge '>=packagename' would make sense and fix this problem. Or perhaps the form: emerge '>=packagename-version' that sytax might be incorrect but the idea is to allow for a newer revision to be emerged as opposed to attempting an exact match.
Uhm, bleh... I can't read, indeed. Which gentoolkit version are you using?
