15 Jan
2015
15 Jan
'15
11:17 a.m.
On my side I don't think we really need a rule on this. It's just good
to know about new stuff, thanks.
Also I'm not a big fan of adding a dependency just for that (but it's
just a -0).
On Thu, Jan 15, 2015 at 10:54 AM, vincent(a)massol.net <vincent(a)massol.net> wrote:
Thanks a lot for your replies.
My opinion:
* I didn’t know the “precision” issue pointed out by Denis so thanks about that and
indeed it can be a problem in some cases
* I made this proposal only because ExpectedException is part of JUnit and I thought it
was the new best practice
* I don’t like too much introducing a third-party library because it makes our tests
non-standard (i.e. if you only know JUnit you’ll also need to understand that other
library) and I don’t think the complexity introduced by a new library outweights the gains
in this case. But it everyone feels strongly about introducing a new lib, I’m ok too.
* I’m perfectly fine to continue using our try/catch idiom.
Thanks
-Vincent
On 15 Jan 2015 at 10:48:19, Eduard Moraru
(enygma2002@gmail.com(mailto:enygma2002@gmail.com)) wrote:
--
Thomas Mortagne
A nice review on the various ways of testing for
exceptions:
http://blog.codeleak.pl/2013/07/3-ways-of-handling-exceptions-in-junit.html
(also talks about and links to a getting started post on catch-exception).
Thanks,
Eduard
On Thu, Jan 15, 2015 at 11:42 AM, Eduard Moraru
wrote:
_______________________________________________
devs mailing list
devs(a)xwiki.org
http://lists.xwiki.org/mailman/listinfo/devs +0, leaning to -0, mainly due to the lack of
precision that could cause
some not very obvious to debug test failures.
What about this: https://code.google.com/p/catch-exception/ ?
It seems to have a very flexible syntax and you don`t have (AFAICS) the
drawback/issues pointed out by Denis.
Thanks,
Eduard
On Wed, Jan 14, 2015 at 1:05 PM, Marius Dumitru Florea <
mariusdumitru.florea(a)xwiki.com> wrote:
_______________________________________________
devs mailing list
devs(a)xwiki.org
http://lists.xwiki.org/mailman/listinfo/devs +1, although defining expectations is more like
JMock than Mockito
with respect to code style. I guess there's no way to verify the
exception after invoking the component under test.
Thanks,
Marius
On Sat, Jan 10, 2015 at 1:51 PM, vincent(a)massol.net
wrote:
> Hi devs,
>
> I’d like to propose that we use the strategy described at
https://github.com/junit-team/junit/wiki/Exception-testing (thanks to
Lyes for pointing out this page to me).
>
> For example:
>
> @Test
> public void sendSynchronousWithErrors() throws Exception
> {
> ...
> this.thrown.expect(MessagingException.class);
> this.thrown.expectMessage("Some messages have failed to be sent
for the following reasons: "
> +
"[[[errorsummary1],[errordescription1]][[errorsummary2],[errordescription2]]]");
>
>
this.mocker.getComponentUnderTest().send(Arrays.asList(message), session);
> }
>
> I feel it’s slightly better than our current idiom based on try/catch.
One advantage is that we cannot forget to put the fail().
>
> WDYT?
>
> Thanks
> -Vincent
>
>
> _______________________________________________
> devs mailing list
> devs(a)xwiki.org
> http://lists.xwiki.org/mailman/listinfo/devs
_______________________________________________
devs mailing list
devs(a)xwiki.org
http://lists.xwiki.org/mailman/listinfo/devs