Test the positive case too, true + invalid formula allowed, test that the default is true, test that flipping it from false to true with an invalid formula in place, resulting in an invalid workbook, is somehow handled so users don't get odd errors later. Maybe validate all formulas then? That would be expensive. If we just allow an invalid state then, the javadoc needs to clearly note that for the boolean setter and getter, and the formula setter and getter.
It will make it easy for users to get things all mixed up but appear valid, so the docs need to be vocal about it. On Tue, Apr 25, 2017, 08:10 <bugzi...@apache.org> wrote: > https://bz.apache.org/bugzilla/show_bug.cgi?id=61033 > > --- Comment #2 from Travis Burtrum <admin.apa...@moparisthebest.com> --- > It's basically just a setter/getter for a boolean, and then not calling > validation logic if false, I'm not sure of a great way to unit test that, > especially since it only applies to XSSF too? > > I guess maybe I could set it false and send in an invalid formula and > assert I > didn't get an exception? Anything else? > > -- > You are receiving this mail because: > You are the assignee for the bug. > --------------------------------------------------------------------- > To unsubscribe, e-mail: dev-unsubscr...@poi.apache.org > For additional commands, e-mail: dev-h...@poi.apache.org > >
