One liner to restructure/nest a dictionary-CodePudding

I'm wondering if in Python3 there is a one liner that can turn:

in_dict = {
    ("ID1", "TEXT_A"): "TIMESTAMP_X",
    ("ID1", "TEXT_B"): "TIMESTAMP_Y",
    ("ID2", "TEXT_C"): "TIMESTAMP_X",
    ("ID3", "TEXT_E"): "TIMESTAMP_Y",
}

into

out_dict = {
    "ID1": [
        {"TEXT_A": "TIMESTAMP_X"}, 
        {"TEXT_B": "TIMESTAMP_Y"}
    ],
    "ID2": [{"TEXT_C": "TIMESTAMP_X"}],
    "ID3": [{"TEXT_E": "TIMESTAMP_Y"}],
}

I know I can easily do this in a for loop but I was wondering if there is a super fancy way.

CodePudding user response：

One way of doing this is using groupby if keys of dictionary are in sorted manner as in your case

from itertools import groupby
in_dict = {
    ("ID1", "TEXT_A"): "TIMESTAMP_X",
    ("ID1", "TEXT_B"): "TIMESTAMP_Y",
    ("ID2", "TEXT_C"): "TIMESTAMP_X",
    ("ID3", "TEXT_E"): "TIMESTAMP_Y",
}
print({k : [{y[1]: in_dict[y]} for y in v]  for k,v in groupby(in_dict, key=lambda x: x[0])})
 # output
 {'ID1': [{'TEXT_A': 'TIMESTAMP_X'}, {'TEXT_B': 'TIMESTAMP_Y'}],
 'ID2': [{'TEXT_C': 'TIMESTAMP_X'}],
 'ID3': [{'TEXT_E': 'TIMESTAMP_Y'}]}

CodePudding user response：

This way,

{ uniq_k : [ {k[1] : v } for k, v in in_dict.items() if uniq_k in k ] for uniq_k in {k[0] for k in in_dict} }