Box.update is slow

Is Box.update supposed to be much slower than dict.update? I expected the performance to be comparable, and ran into trouble. The code below shows the timing for dict.update and Box. .update

import random
import box
import string
import time

def random_string(N):
    return ''.join(random.choices(string.ascii_uppercase + string.digits, k=N))


a = dict((random_string(64), 0) for i in range(10**3))
for i in a:
    a[i.upper()] = 0

b = box.Box(dict((random_string(64), 0) for i in range(10)))
c = dict((random_string(64), 0) for i in range(10))

st = time.time()
c.update(a)
et = time.time()
print(et - st)

st = time.time()
b.update(a)
et = time.time()
print(et - st)

Output:

2.09808349609375e-05
4.840667724609375

With

Python 3.7.5 (default, Nov  1 2019, 02:16:32) 
[Clang 11.0.0 (clang-1100.0.33.8)] on darwin

[jkylling]

The above was tested with Box 3.4.6, while in 4.0.4 Box.update is closer to dict.update and the performance is much better. Sorry for the confusion. The 3.4.6 version of Box.update lives on in Box.merge_update. For Box.merge_update the performance is still very slow, the same as for the old Box.update in 3.4.6. See the code below. This performance issue is a consequence of __setattr__ being slow because of _conversion_checks iterating over all keys in the dictionary on every call of __setattr__. This also seems to be the case for __delitem__, and __getattr__. In practice this means that the basic operations (creates, lookup, updates, deletions) of Box have the same performance as a list. The below code shows that merge_update is considerably slower than dict.update in 4.0.4.

By using some extra internal bookkeeping data in Box it is possible to get performance comparable to dict.

import random
import box
import string
import time

def random_string(N):
    return ''.join(random.choices(string.ascii_uppercase + string.digits, k=N))


a = dict((random_string(64), 0) for i in range(10**2))
for i in a:
    a[i.upper()] = 0

b = box.Box(dict((random_string(64), 0) for i in range(10**3)))
c = dict((random_string(64), 0) for i in range(10**3))

st = time.time()
c.update(a)
et = time.time()
print(et - st)

st = time.time()
b.merge_update(a)
et = time.time()
print(et - st)
timing.py (END)

[cdgriffith]

4.2.0 is now released, so going to close this issue hoping it is fixed, but if there are any issues let me know!

Thanks for all the help!