[Qt-creator] Virtual functions resolving

[Orgad Shaneh]

On Thu, Nov 22, 2012 at 1:52 PM, Sven Putze <
mailinglist.Qt-creator at hardcodes.de> wrote:

> >
> >> 1. Conceptually, when users call Find Usages, they want to know which
> calls can reach this piece of code. For virtual functions, calls to 'lower'
> (i.e. more concrete) types which implement the function, will not reach the
> function in 'higher' types.
> >> For example, Find Usages on Base::v (usages only, we'll leave
> declarations for later) should only list entries 4 and 5 and 6. 7 is not
> calling Base::v since it is reimplemented in MostDerived.
> >
> > I think this is a cool idea. A bit of background: when we implemented
> this feature, we did have some discussion on what to  show, mostly
> regarding overloads/overrides, but wanted to make sure that a user would
> not get confused too much (otherwise they would always choose the "gimme
> all stuff" option anyway).
>
> I'd like to represent the user here :-)
> From a user perspective it would be great to have just one entry in the
> context menu "find usages", but a subdivided output, sorted by the usecases
> -> perhaps with a header what every section stands for.
>
>
> Sven Putze
> mailinglist.Qt-creator at hardcodes.de
>
>
I agree (though a new action which initially opens find usages with this
'smart' mode can come in handy). That's what I introduced in section 0 :)

@Erik, I didn't mean replacing current behavior, only extending it. I wrote
"I suggest adding a new action" on the follow symbol part :)

I'm not familiar with the cplusplus code. I'll try to start this project,
will probably poke you on IRC occasionally ;)

Anyone else volunteers to step up for this feature?

- Orgad
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.qt-project.org/pipermail/qt-creator/attachments/20121122/adff2585/attachment.html>