On Dec 3, 2007, at 0:32, dormando wrote:
I haven't tested the code, but I gave it a brief review and believe
it's
good as a starting point.
I'm still using my python test suite for general protocol
compliance. It occasionally fails on my ``reserved delete'' test:
def testReservedDelete(self):
"""Test a delete with a reservation timestamp."""
self.mc.set("x", 5, 19, "somevalue")
self.assertEquals((19, "somevalue"), self.mc.get("x"))
self.mc.delete("x", 1)
self.assertNotExists("x")
try:
self.mc.add("x", 5, 19, "ex2")
self.fail("Expected failure to add during timed delete")
except MemcachedError, e:
self.assertEquals(memcacheConstants.ERR_EXISTS, e.status)
time.sleep(1.1)
self.mc.add("x", 5, 19, "ex2")
On the last line, it's occasionally reporting that the key still
exists. Other than that, things still look fine.
I assume it'd be good to have this test suite included in the tree so
it's a bit easier for others to get to. It's pretty much plain
python, which makes it not quite fit with the rest of the test suite,
but I don't have much of a desire to rewrite it in perl. Any opinions?
--
Dustin Sallings
