tests: Add run-tests.py -i to control execution from stdin

The new -i command line argument can be used to control test case
execution from stdin to run-tests.py and vm-run.sh. This can be used,
e.g., to run multiple repeated test sequences in a virtual machine
without havign to restart the VM between each iteration.

Signed-off-by: Jouni Malinen <j@w1.fi>
This commit is contained in:
Jouni Malinen 2014-11-19 02:01:27 +02:00
parent cb8cb33e55
commit 0d7456e9fd

View file

@ -204,6 +204,8 @@ def main():
type=str, choices=[[]] + test_modules, nargs='+')
parser.add_argument('-l', metavar='<modules file>', dest='mfile',
help='test modules file name')
parser.add_argument('-i', action='store_true', dest='stdin_ctrl',
help='stdin-controlled test case execution')
parser.add_argument('tests', metavar='<test>', nargs='*', type=str,
help='tests to run (only valid without -f)',
choices=[[]] + test_names)
@ -339,7 +341,34 @@ def main():
shuffle(tests_to_run)
count = 0
for t in tests_to_run:
if args.stdin_ctrl:
print "READY"
sys.stdout.flush()
else:
remaining_tests = tests_to_run
while True:
if args.stdin_ctrl:
test = sys.stdin.readline()
if not test:
break
test = test.splitlines()[0]
if test == '':
break
t = None
for tt in tests:
name = tt.__name__.replace('test_', '', 1)
if name == test:
t = tt
break
if not t:
print "NOT-FOUND"
sys.stdout.flush()
continue
else:
if len(remaining_tests) == 0:
break
t = remaining_tests.pop(0)
name = t.__name__.replace('test_', '', 1)
if log_handler:
log_handler.stream.close()