2008-03-07Fix both sides of provide/conflict handling to work the same way
Dan Winship <danw@gnome.org> [Fri, 07 Mar 2008 13:45:37 -0500] rev 156
Fix both sides of provide/conflict handling to work the same way

thus fixing conflicts-on-properties-rather-than-names, thus fixing the
last remaining unpassed test

2008-03-07kill warning
Dan Winship <danw@gnome.org> [Fri, 07 Mar 2008 13:44:22 -0500] rev 155
kill warning

2008-03-07another checkpoint...
Dan Winship <danw@gnome.org> [Fri, 07 Mar 2008 08:52:09 -0500] rev 154
another checkpoint...

there are a bunch of problems with doing things this way, and file
provide removals are currently unhandled

2008-03-07fix version parsing to not sometimes use octal :)
Dan Winship <danw@gnome.org> [Fri, 07 Mar 2008 08:24:09 -0500] rev 153
fix version parsing to not sometimes use octal :)

2008-03-06Use the razor transaction logic when installing rpms.
Kristian H?gsberg <krh@redhat.com> [Thu, 06 Mar 2008 01:41:50 -0500] rev 152
Use the razor transaction logic when installing rpms.

2008-03-06Split install root creation out into command_init().
Kristian H?gsberg <krh@redhat.com> [Thu, 06 Mar 2008 01:33:08 -0500] rev 151
Split install root creation out into command_init().

2008-03-06Initalize empty package sets with a proper empty file list.
Kristian H?gsberg <krh@redhat.com> [Thu, 06 Mar 2008 01:33:08 -0500] rev 150
Initalize empty package sets with a proper empty file list.

2008-03-06Fix bug in razor_create_dir() path component loop.
Kristian H?gsberg <krh@redhat.com> [Thu, 06 Mar 2008 00:55:51 -0500] rev 149
Fix bug in razor_create_dir() path component loop.

2008-03-05Fix epoch in rpm and rpmdb importer: epoch is an integer tag.
Kristian H?gsberg <krh@redhat.com> [Wed, 05 Mar 2008 23:49:05 -0500] rev 148
Fix epoch in rpm and rpmdb importer: epoch is an integer tag.

2008-03-05keep track of both the install and the remove on an update
Dan Winship <danw@gnome.org> [Wed, 05 Mar 2008 19:02:29 -0500] rev 147
keep track of both the install and the remove on an update

razor_transaction_satisfy_removes() is now quite broken, as it thinks
it should remove a package if a package it depends on was updated.