[racket-dev] [plt] Push #22821: master branch updated

From: Robby Findler (robby at eecs.northwestern.edu)
Date: Sat Jun 18 03:56:58 EDT 2011

FWIW, I think there's two other little complexities that come if we
did want to change it (essentially these being reasons not to bother):
sometimes the focus is on a button (if you click "run" on some
platforms); the insertion point changes to the location of the error
if the error is in the definitions. And I'm probably forgetting
something, too.

There more I think about it, the more I think there's a lot of stuff
happening there that has slowly been shaped for a long time.

(And at this point I'm really wishing that there was a test suite that
we could rely on to actually know what is happening, but I never made
one. Shame on me.)

Robby

On Sat, Jun 18, 2011 at 2:51 AM, Eli Barzilay <eli at barzilay.org> wrote:
> Yes, I know that it's been doing that for a long time -- the reason I
> said that I'd hope for a change is because I disliked it for a long
> time...  I think that making the focus change without explicit
> intention to do so is usually a bad idea.
>
> But since it's been this way for a while I knew that it's unlikely to
> change.  (Also, since it's been this way your fix makes more sense as
> things currently are.)
>
>
> Four minutes ago, Robby Findler wrote:
>> There isn't really a "where it is". The focus goes to the
>> interaction window unless there's an error with a source in the
>> definitions, in which case it goes to the definitions window. Things
>> have been doing that for a _long_ time so I'm reluctant to mess with
>> that.
>>
>> Robby
>>
>> On Fri, Jun 17, 2011 at 8:55 PM, Eli Barzilay <eli at barzilay.org> wrote:
>> > I'd hope for this to go the other way -- for the focus to remain
>> > where it is.  (But I'd say the same on the behavior when both
>> > windows are visible too, and the focus is in the interactions --
>> > I'd like it to show the highlighted error in that case, but keep
>> > the focus in the interactions.)
>
> --
>          ((lambda (x) (x x)) (lambda (x) (x x)))          Eli Barzilay:
>                    http://barzilay.org/                   Maze is Life!
>



Posted on the dev mailing list.