Group.add_nodes() function should be able to get a list of 'pk' or 'uuid' as the argument.

[yakutovicha]

To export a significant amount of nodes it would be good to NOT load them directly, but rather to provide a list of pk or uuid numbers to the add_nodes() function.

I think rather then doing that:

n = QueryBuilder()
n.append(Node)
flat_list = [item for sublist in n.all() for item in sublist]
some_group.add_nodes(flat_list)

it would be much nicer to do:

n = QueryBuilder()
n.append(Node, project=['id'])
flat_list = [item for sublist in n.all() for item in sublist]
some_group.add_nodes(flat_list)

Calling some_group.add_nodes() for every loaded node is very slow, so I would prefer to avoid this.

[ltalirz]

I don't quite get the need for the sublist here (the desired list of nodes could simply come out of the query by applying the appropriate filters), but I completely agree that it should not be necessary to load a node into memory in order to add it to a group.