Question

I have a list of objects, each of the same kind.

Each object has its own list of objects (usually just 5-10 items)

What I used to do was:

for o in main_object_list:
    obj_list = o.get_this_object_list()
    for i in obj_list:
        if i in main_object_list:
            //do something with i

While this approach works, when main_object_list has, say, 100.000 elements,
it goes horribly slow.

My workaround has been this:

for o in main_object_list:
    o.flag = True 

for o in main_object_list:
    obj_list = o.get_this_object_list()
    for i in obj_list:
        if i.flag:
            //do something with i

It goes several orders of magnitude faster (from 22 minutes to as little as 17 secs)
but I suspect there may be a different, and better, approach. Moreover, this example works just becaue each object has a flag property, and by the way it is not so
elegant to use a flag that may well have been setted/unsetted in other functions
(if this function is called in the body of a parent function which uses the same
flag mechanisms, this would mess everything up, setting every objects flag)

Is there a more correct pythonesque way to quickly check if an object is in
main_object_list?

Answer 1

If you want to use your own flag, you could do:

for o in main_object_list:
    o.my_special_flag = True 

for o in main_object_list:
    obj_list = o.get_this_object_list()
    for i in obj_list:
        if hasattr(i, 'my_special_flag'):

Otherwise set.intersection is as fast as it gets:

main_object_set = set(main_object_list)

for o in main_object_list:
    obj_list = o.get_this_object_list()
    objs_in_main_list = main_object_set.intersection(obj_list)
    for i in objs_in_main_list:
        //do something with i

Or:

main_object_set = set(main_object_list)

objs_in_main_list = set().update(
                           *(o.get_this_object_list() for o in main_object_list))
objs_in_main_list.intersection_update(main_object_set)