Use a metaclass to cache Atom instances transparently. This should improve
authorZac Medico <zmedico@gentoo.org>
Fri, 20 Jun 2008 16:18:46 +0000 (16:18 -0000)
committerZac Medico <zmedico@gentoo.org>
Fri, 20 Jun 2008 16:18:46 +0000 (16:18 -0000)
commit89a06fc9d0ab02ace3f8847da2357c5c6f9794a1
tree5a5caebb79319d983f14a63eac719bbe604a868b
parent36a91804330dca702f9d924bfa2ca98c00deb7c6
Use a metaclass to cache Atom instances transparently. This should improve
performance and conserve memory in cases when the same atom is more than
once.

svn path=/main/trunk/; revision=10741
pym/portage/dep.py