Nodes reference

The nodes have consistent behaviors as much as possible, so here are a few notes written once here instead of being repeated all over the place:

• When a node creates Blender objects, it will be in the current scene, and currently active collection.
• When a node offers the option to delete what it creates, that means creating will not delete or overwrite the previous items. Also Blender does not allow deleting datablocks that have several users (eg. a mesh that is used by at least two objects), so those will be silently ignored.
• When a node has a mandatory input socket which is not connected, it will display an error message near the top, so you always know what the minimum required inputs are.
• When a node asks for a name prefix for what it will create (objects, meshes datablocks etc…), try to make that prefix unique to that node, because if the node has a socket output of what it creates, probably the items will be output based on their name prefix, even if they haven't been created by the node. Moreover if the node has a button to delete what it creates, it will delete the items based on their name prefix.
• If a button on a node says “Create / update” it means that the item will be created if it doesn’t exist yet, or updated otherwise. When the data is updated, it means all the data inside (for example all verts and faces for a mesh, or all the pixels color for an image) will be overwritten, and any manual change will be lost.
• If a node has a random seed, that means that, even though the operation it performs is based on randomness, it will always give the same result again and again as long as you do not change any setting. Changing any setting will give a totally new random result. If you simply want the same kind of result with the same settings, but a different random one, then change the seed. The exact value of a seed doesn’t matter, but it can be useful to note it down if you find a particular value that gives a result you like, while you’re experimenting.