0

the job of the layout is to place vertexes at given locations. if the layout is iterative, then the layout's job is to iterate through an algo, moving the vertexes with each step, until the final layout configuration is achieved.

I have a multi-level graph - say 100 objects of type A; each A object has 10 objects as children; call the children type B objects.

I would like the layout location placement algos to operate on objects of type A only (let's say) - and ignore the B objects.

The cleanest way to achieve this objective might be to define a transform to expose those elements that should participate in the 'algo' placement operation via the step method.

Currently, the step methods, assuming they respect the lock flag at all, do their calculations including the locked vertexes first - so lock/unlock won't work in this case.

Is it possible to do this somehow without resorting to multiple graph objects?

chris fabri
  • 1
  • 1
  • 1

1 Answers1

0

If you want to ignore the B objects entirely, then the simplest option is to create a graph consisting only of the A objects, lay it out, and use the locations from that layout.

That said, it's not clear how you intend to assign locations to the B objects. And if the A objects aren't connected to each other at all, then this approach won't make much sense. (OTOH, if they aren't connected to each other then you're really just laying out a bunch of trees.)

Joshua O'Madadhain
  • 2,704
  • 1
  • 14
  • 18