Hello,
I’m using Hops inside clusters.
It works well, but sometimes one of the clusters is missing the Hops component and therefore doesn’t work.
Inside, it looks like this:
And on the same script I have many instances of the same cluster that includes the Hops component and works fine:
Any idea what could cause such an issue?
Many thanks,
Y
Are you opening this file on the same computer and sometimes is shows up and other times it doesn’t? Or are you trying to open the file on a different machine which doesn’t have the Hops plugin installed? Could you include a simplified example file for me to review?
Hi Andy,
In my case the Hops script is saved on a local network (not my own local drive), and I’m reaching it through another script that is saved on the same network. I do have Hops installed.
I’m adding a simple example of the cluster I’m using that has the Hops component inside.
In my larger script I have 20+ instances of this cluster, and as said it works most of the times, but some times one of the clusters is failing. And when I go inside I see that the Hops component is missing.
I’ve noticed that if I go back and open a different instance that is not failing, change something and save. It also corrects the instance that failed before. I hope this is clear…
Hop Test Cluster.gh (11.0 KB)
Hops Test.gh (15.0 KB)
Hi Andy,
I’ve noticed that if I enter a working cluster and go out with the saving option it corrects the missing components in some of the clusters. I’ve also noticed it usually happens only in 1 cluster. Is there a workaround fix that can update all the clusters to reload or something similar?
Many thanks
Does the missing Hops component ever occur if it’s not inside a cluster?
Hi Andy, I believe it will work outside the cluster. One of the reasons it is inside the cluster is that the cluster is password protected and therefore it is a workaround to make it difficult to use the script outside of our office network.
Hello Andy,
Another issue that came up is that when one of the clusters is missing the Hops component, the grasshopper UI also works very slow (just loading components and connecting them). I’ve realized that after resaving the cluster, the UI also came back to a normal speed.
All the best and appreciate your help,
Y
I get the same issue, except that I have the paths to the Hops components be built dynamically relative to the directory of the Grasshopper definition. See my thread about relative paths. After renaming the directory, Hops inside a cluster go missing:
Zoomed into where one of the Hops was (Path is correct!):
How it looked before renaming the directory:
The issue does not happen outside of clusters.
Here comes a test case.
Grasshopper definition: 2024-04-21_definition.gh (20.6 KB)
Hops function: 2024-04-21_double.gh (2.9 KB)
The Hops function doubles a number:
Place the above files into C:\test, and things are fine:
The Python scripts in the definition and the clusters generate a relative path, see my thread about relative paths. Things turn bad when you rename C:\test to C:\test2. The Hops components inside the clusters disappear:
Sometimes, it even works. Looks like a some sort of race condition.
Is it possible to report this as a bug somewhere?
I have created a YT issue for this topic to investigate this when I can carve out some time.
It certainly happens when the Hops functions go missing after a rename of the directory. If the paths are updated by Python scripts, as in my example above, that update will happen too late. It would be nice to have the ability to specify Hops with relative paths, to keep them fixed within a project.
(But maybe I’m the only one who goes crazy with these Hops. It’s like discovering test-driven procedural programming all over again.)
