© 2004
Canonical Ltd.
•
Terms of use
•
Data privacy
•
Contact Launchpad Support
•
Blog
•
Careers
•
System status
•
73416ba
(Get the code!)
(In reply to comment #5)
> Adding "set" and "not_set" comparisons sounds sensible to make things clearer,
> but I don't think the "all" behavior is a bug.
the behavior of qual="all" itself isn't a problem. I just think either of cases should be checked with qual="all" compare="not_eq" even if the testcase contains any invalid thing, as you indicated in the last example. that really makes sense for what you want to. but relying on "All zero values match" isn't a good idea, because it's the side-effect because of no-error-handling.
> Please don't change the behavior of all.
> It is this behavior that makes all,not_eq the complement of any,eq.
If there are too much cases relying on this behavior, we should show a warning at least if something failed to parse, because it's difficult to see whether something might looks like a typo is desired or not. let's see how much things using this trick and we can avoid an regression as much as possible.