Le 07/02/12 16:06, David Rajchenbach-Teller a écrit :
> Perhaps we could compare the results with those of the previous 
> indexing?
> If, say, at least 5% of the indices are the same, then we probably have 
> succeeded.
>

This looks dangerous: what about the first indexing? What if you
completely failed your first indexing? What if your code went through a
heavy refactoring? (then, I assume, you could force the update
nevertheless).

And then: why 5%? (note, it might indeed be a good solution, just trying
to play devil's advocate here).

Lionel

PS: hey David, I think I've seen you from a long distance at FOSDEM but
I was not able to say hello, too far ;-)
_______________________________________________
dev-static-analysis mailing list
[email protected]
https://lists.mozilla.org/listinfo/dev-static-analysis

Reply via email to