I'm trying to understand the preferred way a plug-in can create and find devices it owns. My plug-in will have about 10 devices.
1) Should it create one root device with 10 child devices or 10 root devices? With 1 root device, the plug-in has found all of them once it finds the root device.
2) How should the plug-in determine if any devices exist that were previously created by the plug-in? It looks like this is based on the name of the plug-in, so how can I ensure my plug-in name is unique? If I pick a name like 'weather', it seems the system will not be able to run another plug-in someone calls 'weather'. I see where the plug-in should not use the address or code fields to attempt to create a unique device name. I envision ending up with some plug-in's that I don't publish (not useful to anybody but me) and some plug-in's from the community. Thus, HST can not be the gatekeeper to ensure unique names.
Thanks for any insight.
1) Should it create one root device with 10 child devices or 10 root devices? With 1 root device, the plug-in has found all of them once it finds the root device.
2) How should the plug-in determine if any devices exist that were previously created by the plug-in? It looks like this is based on the name of the plug-in, so how can I ensure my plug-in name is unique? If I pick a name like 'weather', it seems the system will not be able to run another plug-in someone calls 'weather'. I see where the plug-in should not use the address or code fields to attempt to create a unique device name. I envision ending up with some plug-in's that I don't publish (not useful to anybody but me) and some plug-in's from the community. Thus, HST can not be the gatekeeper to ensure unique names.
Thanks for any insight.