..and also, is there an example that uses nodes as lanes?

That's interesting, thanks Stoyan.

Do the nodes have to be locked for this to work? The reason I ask is that I shortly need to add code to 'roll up' some of the lanes.

What I have to achieve is to effectively make the lanes work like containers. The lanes in our system represent hierarchical data, e.g.

Lane 1
-Lane 1.1
--Lane 1.1.1
-Lane 1.2
Lane 2

The client wants to represent the data as lanes, but wants to be able to minimise the data below a certain point in the hierarchy, e.g. to rollup all lanes below Lane 1 (1.1 and 1.1.1 in the example above). There is a fairly reasonable aim for this, which is to allow 'summary' views of the data to be printed with only the top-level processes (which are represented by the lanes) to be displayed.

However, they still want to see the links that are connected to nodes in Lane 1 that join with nodes in the (now rolled up) lanes 1.1. and 1.1.1.

My plan for this was to replace lanes 1.1 and 1.1.1 with a single lane that contains all of the nodes that were in 1.1. and 1.1.1 and retains their horizontal positions, and then to draw one (probably very wide) node across the whole new lane on top of the 'real' nodes. The hope is that all the links will therefore keep working.

How this relates to the nodes as lanes question is that I was considering using nodes for lanes so that I could create a custom lane node that had a minimize/maximise button, but if this node has to be locked then I may be better just placing a node on top of the grid's row headers.

Any general thoughts you or your team have about the above (i.e. is it a really stupid idea) would be very gratefully received!!

Thanks

Tom

Thanks Stoyan, that's really good info.

I've already had a good look at the containers and they very nearly did what I needed. The issue I hit there was that when a container is folded the nodes inside the container all seemed to move to a single point within the container, or at least that was what seemed to be happening from where the links were going when they entered/exited the container.

This seemed to happen even if I forced the container to retain a minimum width equal to the width of the area containing the nodes in their unfolded state.

What I am looking to do is to retain the horizontal positions of the nodes so that if say a link is coming vertically down from a node in the lane above the container it continues to point vertically down even when the container is folded up.

Any idea if such a thing is possible?

Thanks Stoyan.

I'll go for the long node approach for the time being, it's probably simpler than trying to work with a mix of containers and lanes.

Quote:

my type of solution!