How can I create Dictionary in Python from file, where values are a list

Question

I have a txt file and I want to read values into a dictionary. Different from common dictionary, the value of each key is a value pair, for example:

tiger eat meat
tiger eat people
rabbit eat carrot
people can walk
trees has root
people has hand

I want to get a dictionary that,

tiger, {eat, meat}, {eat, people}
rabbit, {eat, carrot}
trees, {has, root}
people, {can, walk}, {has, hand}

Should I just read lines, split(\n) into 3 items and store the first one as the key and the rest two ones as the values? Or there is a better way to store the two values?

My objective is that, when I query what does a tiger eat, I want to get the answer meat and people.

Can you show the actual dictionary structure you want, rather than this strange non-Python pseudocode? — Daniel Roseman, May 17 '16 at 08:09
You need to upload what you've done so far. More so, your data structure does not have any dictionary. Is this what you intend: `{'tiger': [{'eat': 'meat'}, {'eat': 'people'}]}`? — Moses Koledoye, May 17 '16 at 08:12

riteshtch · Answer 1 · 2016-05-17T08:47:01.167

import collections

lines=[]
with open('data1', 'r') as f:
    lines=list(map(lambda line:line.strip(), f.readlines()))

d, flag=collections.defaultdict(list), False
for line in lines:
    temp=list(map(lambda x:x.strip(), line.split()))
    d[temp[0]].append(temp[1:])
print(d)

Here is the output:

$ cat data1
tiger eat meat
tiger eat people
rabbit eat carrot
people can walk
trees has root
people has hand
$ python3 a.py 
defaultdict(<class 'list'>, {'rabbit': [['eat', 'carrot']], 'trees': [['has', 'root']], 'tiger': [['eat', 'meat'], ['eat', 'people']], 'people': [['can', 'walk'], ['has', 'hand']]})

And if you want this structure:

$ python3 a.py 
defaultdict(<class 'list'>, {'people': [{'can': 'walk'}, {'has': 'hand'}], 'tiger': [{'eat': 'meat'}, {'eat': 'people'}], 'trees': [{'has': 'root'}], 'rabbit': [{'eat': 'carrot'}]})

replace the 2nd last line in the script to:

d[temp[0]].append({temp[1]:temp[2]})

score 1 · Answer 2 · answered May 17 '16 at 08:17

First, you can accumulate the data, based on the subjects and the verbs, like this

data = {}
with open("Input.txt") as fin:
    for line in fin:
        subject, verb, obj = line.strip().split()
        data.setdefault(subject, {}).setdefault(verb, []).append(obj)

Now, data will look like this

{'people': {'can': ['walk'], 'has': ['hand']},
 'rabbit': {'eat': ['carrot']},
 'tiger': {'eat': ['meat', 'people']},
 'trees': {'has': ['root']}}

we basically have created nested dictionaries with the values as lists.

Now, its just a simple matter of iterating and printing the result, in the manner you like

for subject in data:
    print subject,
    for verb in data[subject]:
        for obj in data[subject][verb]:
            print "{{{}, {}}}".format(verb, obj),
    print

Output

tiger {eat, meat} {eat, people}
trees {has, root}
rabbit {eat, carrot}
people {has, hand} {can, walk}

Note: If the original order of the data is important, then instead of using normal dictionaries, you can use collections.OrderedDict, like this

from collections import OrderedDict


data = OrderedDict()
with open("Input.txt") as fin:
    for line in fin:
        subject, verb, obj = line.strip().split()
        data.setdefault(subject, OrderedDict()).setdefault(verb, []).append(obj)

Moses Koledoye · Answer 3 · 2016-05-17T15:37:51.083

Create a dictionary whose keys are the subjects and whose values is a list containing dictionaries with verbs as keys and objects as values (see results).

animal_attr = {} #Don't mind the name :)
with open (filename,"r") as f:
    for line in f:
        items = line.split()
        if items[0] not in animal_attr.keys():
            animal_attr[items[0]] = []            
        animal_attr[items[0]].append({items[1]: items[2]})

print(animal_attr)
#{'tiger': [{'eat': 'meat'}, {'eat': 'people'}], 'trees': [{'has': 'root'}],
# 'rabbit': [{'eat': 'carrot'}], 'people': [{'can': 'walk'}, {'has': 'hand'}]}

score 0 · Answer 4 · edited May 23 '17 at 12:16

Once, you have read the lines from the file, you can create a nested defaultdict for this purpose:

d = defaultdict(lambda: defaultdict(list))

for line in lines:
    words = line.split()
    d[words[0]][words[1]].append(words[2])

If you print(d) you will get following:

defaultdict(<function <lambda> at 0x7fa858444320>, {'tiger': defaultdict(<type 'list'>, {'eat': ['meat', 'people'], 'eats': []}), 'trees': defaultdict(<type 'list'>, {'has': ['root']}), 'rabbit': defaultdict(<type 'list'>, {'eat': ['carrot']}), 'people': defaultdict(<type 'list'>, {'has': ['hand'], 'can': ['walk']})})

And, you can access what tiger eats as following:

>>> d['tiger']['eat']
['meat', 'people']

If, you want to see what all can a people do:

>>> d['people']['can']
['walk']

score 0 · Answer 5 · answered May 17 '16 at 08:37

import collections

d=collections.defaultdict(list)
with open('text.txt', 'r') as lines:
    for line in lines:
        temp=line.split()
        d[temp[0]].append({temp[1]: temp[2]})
print(d)

Output:

defaultdict(<type 'list'>, {'tiger': [{'eat': 'meat'}, {'eat': 'people'}], 'trees': [{'has': 'root'}], 'rabbit': [{'eat': 'carrot'}], 'people': [{'can': 'walk'}, {'has': 'hand'}]})

How can I create Dictionary in Python from file, where values are a list

5 Answers5