Welcome to yEd Q&A!
Here you can ask questions and receive answers from other members of the community and yEd developers. And you can tell us your most wanted feature requests.

Context View Limitations for High Edge Density Nodes

+13 votes

I'm not sure if this is a limitation for yEd or what, but certain nodes in my graph never display in a context view, the neighborhood or successors windows remain empty. These only appear to happen for nodes with high edge densities.

 

Take a look here

 

thanks so much!

 

 

Jaime

 

asked Nov 22, 2011 in Finished Features by jxramos (390 points)
recategorized Dec 22, 2014 by thomas.behr

2 Answers

0 votes
Yes, this is a limitiation in yEd. Currently, context views will only display graphs with at most 50 nodes and 100 edges.
answered Nov 23, 2011 by thomas.behr [yWorks] (121,330 points)
Is there a way to overcone this limitation ?
Yes Please.

Or at least give us the option to extend it at our own risk.
Our applications data model and process model have about 30% of the nodes with more then 100 edges.  These are also the nodes I most need to look at and track down a connection.
Indeed; if for deep groups it showed "..." at the end of branches after a certain depth, and for broad groups if it showed a "# more" node to indicate not all siblings for a branch were shown.. Or something along these lines. Often I'm working with deep trees and am only interested in the immediate relatives of the selected node, so showing only the 50 'nearest relatives' would be significantly better than showing nothing at all.
Fully agree with guys above - this limitation should be up to the end user who could decide to take a risk or get significantly slower performance if that's the issue.
In my opinion this should be:
1. the limit should be dynamically adjustable (slider, text box input) by user
2. everything down to a certain depth defined by limit set should be shown plus "..."/"#" indications if there's more beyond
3. a feature proposed - should be option to automatically select all preceding or following siblings in the main model. This option should also allow to "add to current selection". This would be elegant and quick way to isolate a subset of a graph and continue work on it.

Re:

3. a feature proposed - should be option to automatically select all preceding or following siblings in the main model. This option should also allow to "add to current selection". This would be elegant and quick way to isolate a subset of a graph and continue work on it.

This feature is already available.
Use "Tools" -> "Select Elements". Make sure that options "Use These Criteria" is selected only in tab "Nodes". For "Criteria" on tab "Nodes" choose "Successors of Selected Nodes" or "Predecessors of Selected Nodes". For "Maximal Path Distance" use either "1" if you want only direct successors/predecessors or the number of nodes in your graph if you want all successors/predecessors. Finally, if option "Clear Selection First" on tab "General" is not selected, then you get the desired "add to current selection" behavior.

Another vote here for extending this number. My neighborhood views too often disappear as I add new objects to the selection. If this limit were settable in a preference file that would be sufficient.
Adding one more vote as this feature is very important as we use this feature to track dependencies.  This item was first raised in 2011 - what are the perspectives to make it actionable?  is there a threshold in terms of votes which would trigger implementation?
0 votes
As of yEd 3.14, it is possible to adjust the maximum number of nodes and edges to be displayed in context views. To do so, go to "File" -> "Preferences", tab "Display" and adjust settings "Maximum Number of Nodes in Context Views" and "Maximum Number of Edges in Context Views".
answered Dec 22, 2014 by thomas.behr [yWorks] (121,330 points)
Imprint | Privacy Policy