Appending a nested dictionary for values with same key

Question

I've created a nested dictionary from a file as so:

import csv

d = {}
with open("file.txt", 'r') as f:
    data = csv.DictReader(f, delimiter="\t")
    for row in data:
        item = d.get(row["value"], dict())
        item[row["value"]] = row["need"]
        d[row["item"]] = item

print d

However, there should be multiple entries for every item. For example, here is what the output from this code looks like:

d = {'key1': {'153908482': 'ctaggaacca'}, 'key2': {'115057116': 'gtaattctga'}, 'key3': {'133381371': 'ctgaaaagat'}

The dictionary should look something like this:

d = {'key1': {'153908482': 'ctaggaacca', '143543': 'atccgatcgg'}, 'key2': {'115057116': 'gtaattctga', '12321333': 'accccgta'}, 'key3': {'133381371': 'ctgaaaagat'}

I'm losing some information in the creation of my dictionary. Any help would be greatly appreciated.

Answer 1

The problem is that you are assigning item to a different key than you get from the dict.

item = d.get(row["value"], dict()) <-- you never assign to row["value"] so
                                       this always returns an empty dict
item[row["value"]] = row["need"]
d[row["item"]] = item              <-- since you didn't get the existing
                                       row["item"], you keep overwriting
                                       them

I reworked your script a bit so that it is runnable with test data but the only real change is grabbing data from correct key

import csv
from cStringIO import StringIO

f = StringIO("""item    value   need
key1    153908482   ctaggaacca
key2    115057116   gtaattctga
key3    133381371   ctgaaaagat
key1    143543  atccgatcgg
key2    212321333   accccgta
""")

d = {}
#with open("file.txt", 'r') as f:

data = csv.DictReader(f, delimiter="\t")
for row in data:
#    item = d.get(row["value"], dict())
    item = d.get(row["item"], dict())
    item[row["value"]] = row["need"]
    d[row["item"]] = item

print d

And here's an alternate implementation using defaultdict which creates the inner dict for you

import collections
d = collections.defaultdict(dict)
data = csv.DictReader(f, delimiter="\t")
for row in data:
    d[row["item"]][row["value"]] = row["need"]
print d