Re: [cdt-dev] codan test does not fail if StackOverflowError is thrown

[Alena Laskavaia <elaskavaia.cdt@xxxxxxxxx>]

In reality you can propagate exception only so much. You don't want cdt to pop up exceptions in user's face because of failed checker, the only reasonable thing

is to log it - and maybe not each one because it will spam the log too. Maybe reasonable thing to do would be to have a special test mode where exceptions 

are propagated and only enable it during unit testing? Otherwise normal checkers test are just for errors it finds or not.

On Fri, Mar 28, 2014 at 8:05 AM, Andrew Eidsness <eclipse@xxxxxxxxxx> wrote:

On 14-03-28 01:53 AM, Nathan Ridge wrote:

> It seems strange to me that all we do is log it; shouldn't we record
> somewhere that running that checker failed, and then have the test harness
> check that and have the test case fail that happens?

I don't have that code here right now, but I thought there is at least some of the Codan tests that looks for problem
markers.  If that exists in real life and not just my memory ;-) then it sounds like what you've described here (which
sounds like a good idea to me too).

-Andrew

_______________________________________________
cdt-dev mailing list
cdt-dev@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/cdt-dev