I'm updating a Shiny app and am getting a very different visualization output than I have in the past. (Apologies for the lengthy data generation - auto-generated sets don't replicate the problem.) The following code running R 4.1.2 with visNetwork_2.1.0:
The same data with visNetwork_2.0.7 (with a few other parameters for color but substantively the same code; I didn't record the R version) generated this output:
None of the additional visHierarchicalLayout parameters seem to change anything with the exception of lowering levelSeparation from the default of 150 - that brings the two main levels closer together, but the sub-levels within them sit on top of each other and it's still a mess.
I have a different app with different data running visNetwork_2.0.9 that is behaving just fine but also shows the same problem with v2.1.0, so I suspect the issue cropped up between v2.0.9 and v2.1.0. I'm not doing anything differently and I can't find any documentation stating that the parameters need a different kind of treatment, so this feels like a bug.
(Also posted to StackOverflow 2021/10/19)
I'm updating a Shiny app and am getting a very different visualization output than I have in the past. (Apologies for the lengthy data generation - auto-generated sets don't replicate the problem.) The following code running R 4.1.2 with
visNetwork_2.1.0
:generates the following output:
The same data with
visNetwork_2.0.7
(with a few other parameters for color but substantively the same code; I didn't record theR
version) generated this output:None of the additional
visHierarchicalLayout
parameters seem to change anything with the exception of loweringlevelSeparation
from the default of 150 - that brings the two main levels closer together, but the sub-levels within them sit on top of each other and it's still a mess.I have a different app with different data running
visNetwork_2.0.9
that is behaving just fine but also shows the same problem with v2.1.0, so I suspect the issue cropped up between v2.0.9 and v2.1.0. I'm not doing anything differently and I can't find any documentation stating that the parameters need a different kind of treatment, so this feels like a bug.