Inconsistent parameter name `top_cluster` / `top_clusters`

[amjjbonvin]

There is an inconsitent use of top_cluster and top_clusters in the files of the seletopclusts module, which causes this parameter not to be used. As a consequence be default all clusters are passed to the next stage.

Module/Workflow/Library affected

seletopclusters

Expected behavior

When defining top_cluster the number of clusters passed to the next stage should be limited to the value set.

Actual behavior

default.yaml and __init__.py use top_cluster as variable, while seletopclusts.py uses top_clusters

As a consequence the selection of topX clusters is not working and all clusters are selected, which can lead to a lot of models being passed to the next stage.

Suggestions on how to fix it

change top_cluster to top_clusters in default.yaml and __init__.py

Version

036a68a12a9937edb168405727e1791dd3fd14f7

Additional context

This might have affected work where cluster selection was applied after rigidbody.

[amjjbonvin]

Note that the variable name should then also be changes in two test files. Also it makes more sense to use top_clusters rather than top_cluster as one would expect to select typically more than one cluster

[mgiulini]

@amjjbonvin can you give me a practical example where you encountered the problem? top_cluster is passed as an argument to the function select_top_clusts_models (here https://github.com/haddocking/haddock3/blob/036a68a12a9937edb168405727e1791dd3fd14f7/src/haddock/modules/analysis/seletopclusts/seletopclusts.py#L10), so the variable's name doesn't matter. I ran several examples and the selection is working fine.

[VGPReys]

I agree that top_clusters with an s is better than top_cluster. But, same as marco, I also do not see the error in the current version

[amjjbonvin]

None of our current examples test that actually... But there must have been a glitch somewhere as I currently can't reproduce the issue anymore...

[rvhonorato]

@amjjbonvin i think you forgot to add the steps to replicate - could you please update the issue description?

[amjjbonvin]

Actually I could not replicate the problem. Must have been some glitch. So in principle, unless we want to change the syntax for consistency, this one can be closed.