[prev in list] [next in list] [prev in thread] [next in thread] 

List:       ktexteditor-devel
Subject:    Re: Access to the currently displayed document range
From:       Hamish Rodda <rodda () kde ! org>
Date:       2008-03-27 14:02:26
Message-ID: 200803280102.28754.rodda () kde ! org
[Download RAW message or body]

[Attachment #2 (multipart/signed)]


Hi,

On Thu, 27 Mar 2008 07:57:20 am Michel Ludwig wrote:
> Hi,
>
> On Wed 26 Mar 2008, Dominik Haumann wrote:
> > 1. We just discussed this on IRC and came to the conclusions:
> > - rename DisplayInterface to ViewportInterface
> > - rename getDisplayedRange to displayRange
> > - rename displayedRangeChanged to displayRangeChanged
> > - change Q_SIGNALS to public, as the class does not inherit QObject
> >   (see variableinterface.h for an example)
> > - add documentation that also states that the columns are of importance,
> >   think of dynamically wrapped lines
> > - the range returned by displayRange() is a object and no pointer, which
> >   was a typo in the first place
> >
> > If I missed anything, please add it :)
>
> I've attached a new version.
>
> > 3. With the above change, we also thought about the already existing
> > signals KTE::View::horizontalScrollPositionChanged
> >   KTE::View::verticalScrollPositionChanged
> > Question is whether they are still needed (or needed at all):
> > - Are there use cases where displayedRangeChanged is not enough with
> >   regard to verticalScrollPositionChanged?
> >   If not, verticalScrollPositionChanged should be marked as \deprecated
> > for KDE5 and the ViewportInterface should be merged into KTE::View.
>
> What if a really long line is shown that is dynamically wrapped and is
> longer than what can be shown in the viewport? Then one could change the
> vertical scroll position but the displayed range would not change...
>
> (I know it's an adventurous use case :-))
>
> > - Is horizontalScrollPositionChanged needed at all? displayedRangeChanged
> >   does NOT tell you about this. But I wonder whether anyone uses this
> >   signal. If not, it should be marked as \deprecated, too.
>
> My experience as a developer that uses KTextEditor interfaces tells me that
> there are never enough signals! ;-)

The interface looks ok (I would probably add the new range to the signal, to 
avoid the receiver having to look it up) however I would suggest also that we 
try to pin down why the dynamic highlighting interface is slow - I designed 
it to scale well, so there must be a bottleneck (hopefully that can be 
fixed).  I may have some time over the next few months after my exams are 
finished to take a look at it.

Cheers,

Hamish.

["signature.asc" (application/pgp-signature)]

_______________________________________________
KTextEditor-Devel mailing list
KTextEditor-Devel@kde.org
https://mail.kde.org/mailman/listinfo/ktexteditor-devel


[prev in list] [next in list] [prev in thread] [next in thread]