Hi, On Mon, 23 Nov 2015 19:33:29 +0100 jan <d_...@ymail.com> wrote: > Am 23.11.2015 um 09:44 schrieb Thomas Friedrichsmeier: > > 1c. The warning symbol could be repeated next to / inside the > > "Submit"-button. > If we can only show the warning icon in case there is an error, it > makes sense to me.
yes, that's the idea (and that's what makes it non-trivial to implement). > > 1d. On pressing the "Submit"-button, there could be a warning > > dialog: "Do you really want to do this?", with a "Don't ask again" > > option. > It would be really good, if the text in not generic, but offers some > help to decide what to do. Instead of "Do you really want…" we could > have "The Variable [name] for [varslot] has the type [type] but the > function only takes [right type]. The calculation might not give a > result." I don’t know how hard it is to create, but we should at least > have all the information we need. Yes, should be doable. At least the part specifying where the problem is found, and what property exactly does not look right. Rather difficult, again, for specifying what exactly might go wrong, in case the user choses to proceed. In current master, I've changed the behavior to soft checks for now. The per-item warning now says "may lead to failures or unexpected results". I don't think the dialog would be any more specific in this respect, either. > > 2. If allowing for hard checks, and in particular if we allow for > > hard checks _in addition to_ soft checks, these should work by > > refusing the object in question right away. I am still not sure, > > how to solve this, UI wise, but perhaps it's ok'ish to simply pop > > up a dialog "No, you can't do this, because...", when the user > > clicks the "Add"-button with an invalid object selected. That will > > be somewhat annoying, but OTOH it is not something users would try > > to do all the time. > Seems like a good idea. Though we should try to save the user the > click on "o.k.". (There might still be cases in which – e.g. desperate > students – just try try try… :-) ) Could we show this like a system > notification that sort of pulls down or the like? > (https://en.wikipedia.org/wiki/Pop-up_notification) Hm, that should be possible. But if we want to avoid the user having to click "ok"/"close", how / when will the message go away? > Again, a text quoting variable name and type and target type would > help the user greatly to know what goes on here. Yes. Regards Thomas
pgpDiIfyi7X1E.pgp
Description: OpenPGP digital signature
_______________________________________________ rkward-devel mailing list rkward-devel@kde.org https://mail.kde.org/mailman/listinfo/rkward-devel