tokejepsen Posted May 25, 2017 Report Share Posted May 25, 2017 Hey, I'm looking to add import functionality with the Vectorfield node. Everything is in place but ftrack-connect-nuke discover code for getting the scripts assets seem a bit hardcoded; https://bitbucket.org/ftrack/ftrack-connect-nuke/src/74ea5d8ba0589256b035a375152641ed9c6c9376/source/ftrack_connect_nuke/connector/nukecon.py?at=master&fileviewer=file-view-default#nukecon.py-32 I'm up for doing a PR, but wanted to discuss this a bit further first. Was the original thought behind limiting the search of the nodes, for optimization? Maybe there has been complains about the performance when search all nodes for the "componentId" knob? Link to comment Share on other sites More sharing options...
tokejepsen Posted May 27, 2017 Author Report Share Posted May 27, 2017 Actually we are scraping this feature, in favour of using gizmos for LUTs instead. Would still be nice to hear the reason behind this. I know in the future we'll be implementing 3d nodes, so the same problem will arise. Link to comment Share on other sites More sharing options...
Mattias Lagergren Posted May 30, 2017 Report Share Posted May 30, 2017 I think it is partly a performance thing but the code that you refer to is quite old and should be re-evaluated. Once we are done with refreshing the publishing part of Connect integrations we will look at import and later the asset manager. The goal would be to come up with something that is more flexible and easy to extend. Link to comment Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.