On Wednesday, 15 May 2013 at 18:28:53 UTC, Idan Arye wrote:
I'm making an idioms
library(http://forum.dlang.org/thread/[email protected])
to add to Phobos, and one of the idioms is the singleton design
pattern. Since I'm gonna send it to Phobos, it needs to have a
unit test. Here is my unit test:
static class Foo
{
mixin LowLockSingleton;
this()
{
Thread.sleep(dur!"msecs"(500));
}
}
Foo[10] foos;
foreach(i; parallel(iota(foos.length)))
{
foos[i] = Foo.instance;
}
foreach(i; 1 .. foos.length)
{
assert(foos[0] == foos[i]);
}
This unit test works - it doesn't fail, but if I remove the
`synchronized` from my singleton implementation it does fail.
Now, this is my concern: I'm doing here a 500 millisecond sleep
in the constructor, and this sleep is required to guarantee a
race condition. But I'm not sure about two things:
- Is it enough? If a slow or busy computer runs this test, the
500ms sleep of the first iteration might be over before the
second iteration even starts!
- Is it too much? Phobos has many unit tests, and they need to
be run many times by many machines - is it really OK to add a
500ms delay for a single item's implementation?
Your opinion?
Also note that unit tests are supposed to run fast (definitely <
50 ms), because running 500-5000 of them should be relatively
cheap. Apart from that, races aren't something to test for in a
unit test. It may be more a kind of integration test, though the
sporadic nature of races makes them candidates for "One of those
occasionally failing bad guys hamstringing development... Let's
just ignore him."-type thinking.
I would create a seperate Test suite for race tests and execute
them multiple times if need be... But there are a million better
sources on the internet than me on that topic.
tldr; don't test for races in unit tests.
