[prev in list] [next in list] [prev in thread] [next in thread]
List: kde-kimageshop
Subject: Re: Tool overlay and cursor shapes
From: Cyrille Berger <cberger () cberger ! net>
Date: 2009-09-05 21:48:13
Message-ID: 200909052348.14177.cberger () cberger ! net
[Download RAW message or body]
[Attachment #2 (multipart/alternative)]
On Saturday 05 September 2009, LukasT.dev@gmail.com wrote:
> On Friday 04 September 2009 10:52:20 Cyrille Berger wrote:
> > On Friday 04 September 2009, LukasT.dev@gmail.com wrote:
> > > I'm waiting whether somebody approve my commit which uses QImage as
> > > buffer in canvas although we have many buffers in the render chain...
> >
> > That is still a concern.
>
> Any chance to test this patch through beta stages and revert if it shows
> performance problems?
I leave that decision to Boudewijn :)
> > > At least my
> > > patch would help me to fix some issues (I could use various compositing
> > > modes, e.g. XOR, without fear that some platform or X11 would not work
> > > correctly)
> >
> > What's wrong with the black&white lines solution ?
>
> I really think that some composite operation is better because it can help
> you identify e.g. the borders of some area you paint. You can see on the
> outline where is different colour. Black and white just covers part of the
> image.
>
> Second argument is pure subjective -- it is ugly :)
since xor is ugly as well...
> > > I would prefer some intelligent XOR instead of black&white versions.
> > > When the under-laying pixel has brightness less then 50 use black
> > > pixel, white otherwise...
> >
> > Yes but we don't have that in Qt, do we ?
>
> Yep, we don't. It is possible to have own code for that outside Qt? E.g.
> some method that would raster line on QImage according bresensham algorithm
> and would take pixels in QImage into account? Then just use value from
> QColor as the limit for black&white? Would be that usable?
we would have to rewrite part of arthur in krita... maybe trolltech would
accept such a patch... but that would be for nex year.
--
Cyrille Berger
[Attachment #5 (text/html)]
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0//EN" \
"http://www.w3.org/TR/REC-html40/strict.dtd"><html><head><meta name="qrichtext" \
content="1" /><style type="text/css">p, li { white-space: pre-wrap; \
}</style></head><body style=" font-family:'DejaVu Sans'; font-size:9pt; \
font-weight:400; font-style:normal;">On Saturday 05 September 2009, \
LukasT.dev@gmail.com wrote:<br> > On Friday 04 September 2009 10:52:20 Cyrille \
Berger wrote:<br> > > On Friday 04 September 2009, LukasT.dev@gmail.com \
wrote:<br> > > > I'm waiting whether somebody approve my commit which uses \
QImage as<br> > > > buffer in canvas although we have many buffers in the \
render chain...<br> > ><br>
> > That is still a concern.<br>
><br>
> Any chance to test this patch through beta stages and revert if it shows<br>
> performance problems?<br>
I leave that decision to Boudewijn :)<br>
<p style="-qt-paragraph-type:empty; margin-top:0px; margin-bottom:0px; \
margin-left:0px; margin-right:0px; -qt-block-indent:0; text-indent:0px; \
-qt-user-state:0;"><br></p>> > > At least my<br> > > > patch would \
help me to fix some issues (I could use various compositing<br> > > > modes, \
e.g. XOR, without fear that some platform or X11 would not work<br> > > > \
correctly)<br> > ><br>
> > What's wrong with the black&white lines solution ?<br>
><br>
> I really think that some composite operation is better because it can help<br>
> you identify e.g. the borders of some area you paint. You can see on the<br>
> outline where is different colour. Black and white just covers part of the<br>
> image.<br>
><br>
> Second argument is pure subjective -- it is ugly :)<br>
since xor is ugly as well...<br>
<p style="-qt-paragraph-type:empty; margin-top:0px; margin-bottom:0px; \
margin-left:0px; margin-right:0px; -qt-block-indent:0; text-indent:0px; \
-qt-user-state:0;"><br></p>> > > I would prefer some intelligent XOR instead \
of black&white versions.<br> > > > When the under-laying pixel has \
brightness less then 50 use black<br> > > > pixel, white otherwise...<br>
> ><br>
> > Yes but we don't have that in Qt, do we ?<br>
><br>
> Yep, we don't. It is possible to have own code for that outside Qt? E.g.<br>
> some method that would raster line on QImage according bresensham algorithm<br>
> and would take pixels in QImage into account? Then just use value from<br>
> QColor as the limit for black&white? Would be that usable?<br>
we would have to rewrite part of arthur in krita... maybe trolltech would accept such \
a patch... but that would be for nex year.<br> <p style="-qt-paragraph-type:empty; \
margin-top:0px; margin-bottom:0px; margin-left:0px; margin-right:0px; \
-qt-block-indent:0; text-indent:0px; -qt-user-state:0;"><br></p>-- <br> Cyrille \
Berger</p></body></html>
_______________________________________________
kimageshop mailing list
kimageshop@kde.org
https://mail.kde.org/mailman/listinfo/kimageshop
[prev in list] [next in list] [prev in thread] [next in thread]
Configure |
About |
News |
Add a list |
Sponsored by KoreLogic