[Libvirt-cim] [PATCH 04/10] cimtest: VSSDC - 02-reverse.py - adjust iteration of association
John Ferlan
jferlan at redhat.com
Fri Apr 4 16:12:53 UTC 2014
Upcoming changes to libvirt-cim allow for multiple types of objects
per class name for the Input class and a new upcoming Controller class.
Adjust the logic in the code to change the iteration to handle the
situation where there could be more than one type of device in the class.
An innocent bystander of this change was the extra check in the code:
if status == FAIL and virt == "LXC":
return XFAIL_RC(bug_libvirt)
Since I wasn't sure what this was testing and perhaps it's actually
been fixed - I removed the check. If it's still a bug, then it needs
to be fixed and there's only one way to find out!
Signed-off-by: John Ferlan <jferlan at redhat.com>
---
.../02_reverse.py | 102 ++++++++++++---------
1 file changed, 61 insertions(+), 41 deletions(-)
diff --git a/suites/libvirt-cim/cimtest/VirtualSystemSettingDataComponent/02_reverse.py b/suites/libvirt-cim/cimtest/VirtualSystemSettingDataComponent/02_reverse.py
index 3a13b2a..1a2218b 100644
--- a/suites/libvirt-cim/cimtest/VirtualSystemSettingDataComponent/02_reverse.py
+++ b/suites/libvirt-cim/cimtest/VirtualSystemSettingDataComponent/02_reverse.py
@@ -63,18 +63,6 @@ test_vcpus = 1
test_mem = 128
test_mac = "00:11:22:33:44:aa"
-def check_rasd_values(id, exp_id):
- try:
- if id != exp_id:
- logger.error("Returned %s instead of %s", id, exp_id)
- return FAIL
-
- except Exception, detail :
- logger.error("Exception evaluating InstanceID: %s", detail)
- return FAIL
-
- return PASS
-
def assoc_values(ip, assoc_info, virt="Xen"):
"""
The association info of
@@ -99,8 +87,11 @@ def assoc_values(ip, assoc_info, virt="Xen"):
}
+ expect_rasds = len(rasd_list)
+
try:
- if len(assoc_info) <= 0:
+ assoc_count = len(assoc_info)
+ if assoc_count <= 0:
logger.error("No RASD instances returned")
return FAIL
@@ -111,34 +102,63 @@ def assoc_values(ip, assoc_info, virt="Xen"):
input_cn = get_typed_class(virt, 'InputResourceAllocationSettingData')
grap_cn = get_typed_class(virt, 'GraphicsResourceAllocationSettingData')
- for inst in assoc_info:
- if inst.classname == proc_cn:
- status = check_rasd_values(inst['InstanceID'],
- rasd_list['proc_rasd'])
- elif inst.classname == net_cn:
- status = check_rasd_values(inst['InstanceID'],
- rasd_list['net_rasd'])
- elif inst.classname == disk_cn:
- status = check_rasd_values(inst['InstanceID'],
- rasd_list['disk_rasd'])
- elif inst.classname == mem_cn:
- status = check_rasd_values(inst['InstanceID'],
- rasd_list['mem_rasd'])
- elif inst.classname == input_cn:
- status = check_rasd_values(inst['InstanceID'],
- rasd_list['input_rasd'])
- if status == FAIL and virt == "LXC":
- return XFAIL_RC(bug_libvirt)
- elif inst.classname == grap_cn:
- status = check_rasd_values(inst['InstanceID'],
- rasd_list['grap_rasd'])
- else:
- logger.error("Unexpected RASD instance type" )
- status = FAIL
-
- if status == FAIL:
- logger.error("Mistmatching association value" )
- break
+ rasd_cns = [proc_cn, net_cn, disk_cn, mem_cn, input_cn, grap_cn]
+
+ # Iterate over the rasds, looking for the expected InstanceID
+ # listed in the rasd_list dictionary for the same classname in
+ # the returned assoc_info list
+ try:
+ found_rasds = 0
+ # Keep track of what worked
+ found_list = {}
+ assoc_list = {}
+ for cn in rasd_cns:
+ for rasd_key, rasd_value in rasd_list.iteritems():
+ for j, inst in enumerate(assoc_info):
+ if inst.classname == cn and \
+ inst['InstanceID'] == rasd_value:
+ #found_list.append((rasd_key, rasd_value))
+ found_list.update({rasd_key: rasd_value})
+ assoc_list.update({rasd_value: cn})
+ found_rasds += 1
+ except Exception, detail:
+ logger.error("Exception evaluating InstanceID: %s", detail)
+ for (i,j) in found_list:
+ logger.error("Found cn=%s exp_id=%s", i, j)
+ return FAIL
+
+ # Check for errors
+ if expect_rasds != found_rasds:
+ logger.error("RASD instances don't match expect=%d found=%d.",
+ expect_rasds, found_rasds)
+ status = FAIL
+ # What did we expect to find from the rasd_list, but did not
+ # find in the found_list (key'd by rasd name)?
+ # This means we're missing some device or perhaps the
+ # InstanceID format changed...
+ for k, v in rasd_list.iteritems():
+ if k not in found_list:
+ logger.error("rasd_list ('%s','%s') not in found_list",
+ k , v)
+ # Thankfully the alternative is not possible - after all how
+ # could there be something in the found list that isn't in the
+ # rasd_list to start with...
+
+ if assoc_count != found_rasds:
+ status = FAIL
+ logger.error("Assoc instances don't match expect=%d found=%d.",
+ assoc_count, found_rasds)
+ # What's in the assoc_info that's not in found assoc_list (key'd
+ # by InstanceID)
+ # Meaning there's a new device type that we haven't accounted for
+ for j, inst in enumerate(assoc_info):
+ if inst['InstanceID'] not in assoc_list:
+ logger.error("Did not find association id=%s in assoc_list",
+ inst['InstanceID'])
+ # Thankfully the alternative is not possible - after all how
+ # could there be something in the found list that isn't in the
+ # assoc_info list to start with...
+
except Exception, detail :
logger.error("Exception in assoc_values function: %s", detail)
--
1.8.5.3
More information about the Libvirt-cim
mailing list