Hi Eirik,
I start to understand your point of view. You view edge from programming
point of view?
Like if you have a chess board, an edge cell is one where one cannot
take one cell step to all directions?
If I would have to point an edge of a chess board, I would point to the
physical edge.
If you draw a chess board to a grid and ask for an edge, I would point
the line intervals at the edges.
Is there a possibility to add a picture to Qt documentation where you
have a rectangle grid and shade the
edge cells with some color to identify them as edges? If I'm the only
one confused of the term edge, not edge cell, in integer
rectangle, then this is not needed.
Best Regards,
Tomi Pannila
On 21.4.2022 11.39, Eirik Aavitsland wrote:
On 4/20/22 17:50, Tomi Pannila wrote:
Hi Lars and Eirik,
thank you for the explanations. I guess Eirik kind of guessed right
that I view QRect from $\mathbb{R}^2$ point of view.
I'm a mathematician and this is the most natural point of view for
me. For me, rectangles have edges=boundary which are measure zero.
But QRect is not a mathematical (cartesian) rectangle, instead it
describes a chess board. If one asks anybody which squares are on the
edge of the board, they will correctly point out row 1 and 8, and
column A and H. That is exactly the meaning of "edge" in the doc of
QRect::contains() - it is the everyday, intuitive and obvious meaning.
So I don't see the problem with the current api, except for the
anonymosity of the bool parameter, as Lars pointed out. Also, there is
an overload of QRect::contains() without the bool parameter, so no
edge to worry about at all, so I don't really understand what the
problem is.
(And if one wants to do operate on rects with borders, one can simply
use QMargins together with QRect, e.g. QRect::marginsAdded()).
Cheers,
- Eirik Aa.
_______________________________________________
Development mailing list
Development@qt-project.org
https://lists.qt-project.org/listinfo/development
