Python nested dictionary spread (**kwargs)

Question

Python has the ** operator, which spreads a dictionary within another dictionary or function arguments. This works well in the example:

default = {'a':5}
a = {'a':15}
{**default, **a}
// => {'a': 15}

However, if one of the value is a dictionary itself, it is overwritten completetly.

default = {'a': {'aa' : 10, 'ab' : 15}}
a = {'a': {'aa': 15}}
{**default, **a}
// => {'a': {'aa': 15}}
// expected => {'a': {'aa' : 15, 'ab' : 15}}

It is overwritten completely. For now I'll make a function that unpacks them recursively, but i was wondering if there is an actual way to do this within python (not necessarily a syntax feature, a standard library function would be OK as well).

If there is no such feature, that is also an acceptable answer.

What did you expect - what output were you looking for? The value is overwritten completely _whether or not_ it's a dictionary, the `5` in your first example is replaced. — jonrsharpe, Jun 16 '21 at 14:45
The spread operator is always shallow, that is it only access the top level items of the thing it is trying to spread. To handle a special shape of item or a special deep spread you would need to implement your own logic. — zr0gravity7, Jun 16 '21 at 14:47
@jonrsharpe i already made my own, I was just wondering if theres functionality for it by default. Thanks for the help — DownloadPizza, Jun 16 '21 at 15:05
@jonrsharpe i did not see that question, but feel free to mark as dupe — DownloadPizza, Jun 16 '21 at 15:07

score 0 · Accepted Answer · answered Jun 16 '21 at 15:52

This issue is less about the spread operator (which just expands a dictionary), and more about how the new dictionary is being constructed.

From the dict docs:

If a key occurs more than once, the last value for that key becomes the corresponding value in the new dictionary.

If a key being added is already present, the value from the keyword argument replaces the value from the positional argument.

This is apparent from the first example you gave, where the value of a:5 is overwritten with a:15, rather than being summed to a:20.

The same thing is happening with the dictionaries: the previous values are being overwritten rather than added togather, so if you want the previous values to be concatenated you will need to define that operation manually.

Python nested dictionary spread (**kwargs)

1 Answers1