On Dec 7, 2007 5:13 PM, Dan McGee <dpmcgee@gmail.com> wrote:
On Dec 7, 2007 3:56 PM, Nathan Jones <nathanj@insightbb.com> wrote:
I found this useful when testing the shortest_delta_path code. Recently, I ran valgrind on the test which helped to discover the memory leak.
Not sure if you guys will want to apply this, but here it is.
Signed-off-by: Nathan Jones <nathanj@insightbb.com>
I really think we do need some test programs directly on libalpm- things that pactest can't fully set on fire.
I'd definitely like to write a unit test harness for alpm_list_t, so this may provide some inspiration for that.
I agree. Nathan, this is great. I started a makepkg unit test setup somewhere, but that went by the wayside. A pacman test setup is also great. If we're testing JUST the libalpm functions we may not need all the complexity pactest has with creating DBs and the like, so we could go with the precedent here