Hi Asko,
Asko Soukka wrote:
Log message:
abstracting WheelView, quick Sundew implementation -> almost an
illusion of virtual space :)
Good start indeed. Unforch, still a lot to do :-)
One thing is we need to figure out how to fix interpolation. One
possibility is my earlier proposal for a special connection-aware
matcher. However, I'm still lost for an algorithm to actually fulfill
the constraints set by addConnection() (see my earlier mail) -- any ideas?
An easier point is that for the 'virtual space' perception, we must be
able to see what we move to if we take *two* steps right or left. I.e.,
let's say we have four nodes ::
A -- B -- C
\
D
and A is the focus. In a view like the above, when we move two steps
right, we need to be on C. Currently, after moving one step right, we
often get this picture:
C
/
A -- B -- D
The reason is that the Cursor places D opposite to A on the wheel around
B, but when the focus is A, the WheelView doesn't take this into account.
Also, we need billowing. The farther-away nodes must be smaller. This is
important because currently, I sometimes take the wrong node to be the
focus-- it's actually a selection, and that confuses me to no end (when
I'm not thinking ;-) ).
Sorry to give even more problems, but these are all really important for
Loom to *work*...
What's the problem with the rotated stubs, btw? It should just be
calculating the lines' ends a little differently?
Thanks for your work,
- Benja
_______________________________________________
Gzz-commits mailing list
address@hidden
http://mail.nongnu.org/mailman/listinfo/gzz-commits