Re: [Libvirt-cim] [PATCH] [TEST] Fixing 41_cs_to_settingdefinestate.py tc of CS to work with the latest libvirt-cim changes
Deepti B. Kalakeri wrote:
# HG changeset patch
# User Deepti B. Kalakeri <deeptik(a)linux.vnet.ibm.com>
# Date 1229602962 28800
# Node ID ff966617a775d47376459a69aca191f5367874dd
# Parent adfa38df8c3b9600ff08796631c0eae7d3249ffe
[TEST] Fixing 41_cs_to_settingdefinestate.py tc of CS to work with the latest libvirt-cim
changes.
Libvirt-cim now supports GraphicsRASD for revision >=725 and InputRASD >= 746,
hence
updating/fixing the tc to accomodate these changes.
Tested with KVM on F9 , Xen/KVM with current sources.
Signed-off-by: Deepti B. Kalakeri <deeptik(a)linux.vnet.ibm.com>
diff -r adfa38df8c3b -r ff966617a775
suites/libvirt-cim/cimtest/ComputerSystem/41_cs_to_settingdefinestate.py
- # Verifying that the in_vssdc_list contains 4 entries one each for mem rasd,
- # network rasd, processor rasd and disk rasd.
- exp_len = 4
+ # Verifying that the in_vssdc_list contains
+ # 4 entries one each for mem rasd, network rasd, processor rasd and
+ # disk rasd for libvirt-cim revision < 725
+ # 5 entries one each for mem rasd, network rasd, processor rasd and
+ # disk rasd, graphics rasd for libvirt-cim revision >= 725 and < 746
+ # 6 entries one each for mem rasd, network rasd, processor rasd and
+ # disk rasd, graphics rasd, input rasd for
+ # libvirt-cim revision >= 746
+ curr_cim_rev, changeset = get_provider_version(virt, server)
+ if curr_cim_rev < libvirt_graphics_changes:
+ exp_len = 4
+ elif curr_cim_rev >= libvirt_graphics_changes and \
+ curr_cim_rev < libvirt_input_dev_changes:
+ exp_len = 5
+ elif curr_cim_rev >= libvirt_input_dev_changes:
+ exp_len = 6
if check_len(an, in_vssdc_list, qcn, exp_len) != PASS:
vsxml.destroy(server)
return FAIL
I think this little bit of logic is confusing. I recently updated
SettingsDefine/01_forward.py and SettingsDefine/02_reverse.py with an
improved way of getting the RASD and device instances.
I would suggest looking at that test, and using the enum_dev() and
possibly the enum_rasds() functions to generate you instance lists.
I think this would reduce the amount of code and possibly reduce the
number of times you need to branch the test.
This test is pretty old, so I think it's worth updating it some. Thoughts?
--
Kaitlin Rupert
IBM Linux Technology Center
kaitlin(a)linux.vnet.ibm.com