TypeError: 'dict_keys' object is not subscriptable
Question:
I have this code that errors out in python3:
self.instance_id = get_instance_metadata(data='meta-data/instance-id').keys()[0]
TypeError: 'dict_keys' object is not subscriptable
I changed my code and I get different error (I guess I need more experience):
self.instance_id = get_instance_metadata(list(data='meta-data/instance-id').keys())[0]
TypeError: list() takes no keyword arguments
Answers:
.keys() is a set-like view, not a sequence, and you can only index sequences.
If you just want the first key, you can manually create an iterator for the dict (with iter) and advance it once (with next):
self.instance_id = next(iter(get_instance_metadata(data='meta-data/instance-id')))
Your second attempt was foiled by mistakes in where you performed the conversion to list, and should have been:
self.instance_id = list(get_instance_metadata(data='meta-data/instance-id').keys())[0] # The .keys() is unnecessary, but mostly harmless
but it would be less efficient than the solution I suggest, as your solution would have to make a shallow copy of the entire set of keys as a list just to get the first element (big-O O(n) work), where next(iter(thedict)) is O(1).
I have this code that errors out in python3:
self.instance_id = get_instance_metadata(data='meta-data/instance-id').keys()[0]
TypeError: 'dict_keys' object is not subscriptable
I changed my code and I get different error (I guess I need more experience):
self.instance_id = get_instance_metadata(list(data='meta-data/instance-id').keys())[0]
TypeError: list() takes no keyword arguments
.keys() is a set-like view, not a sequence, and you can only index sequences.
If you just want the first key, you can manually create an iterator for the dict (with iter) and advance it once (with next):
self.instance_id = next(iter(get_instance_metadata(data='meta-data/instance-id')))
Your second attempt was foiled by mistakes in where you performed the conversion to list, and should have been:
self.instance_id = list(get_instance_metadata(data='meta-data/instance-id').keys())[0] # The .keys() is unnecessary, but mostly harmless
but it would be less efficient than the solution I suggest, as your solution would have to make a shallow copy of the entire set of keys as a list just to get the first element (big-O O(n) work), where next(iter(thedict)) is O(1).