Convert a csv.DictReader object to a list of dictionaries?

Question

A csv file names.csv has content:

first_name last_name Baked Beans Lovely Spam Wonderful Spam 

I would like to read it into a list of dictionaries, with the first row containing the keys:

>>> import csv >>> with open('names.csv') as csvfile: ...     reader = csv.DictReader(csvfile) ...     for row in reader: ...         print(row['first_name'], row['last_name']) ... Baked Beans Lovely Spam Wonderful Spam 

But is the type of reader csv.DictReader? How can I convert reader into a list of dictionaries? Thanks.

Answer 1

import csv with open("in.csv") as csvfile:     reader = csv.DictReader(csvfile,delimiter=" ")     print(list(reader)) [{'first_name': 'Baked', 'last_name': 'Beans'}, {'first_name': 'Lovely', 'last_name': 'Spam'}, {'first_name': 'Wonderful', 'last_name': 'Spam'}] 

If the delimiter is not actually a , you need to specify " " or whatever it is.

Just to clear any confusion, the code works fine for python3.6 also, the only difference is that using DictReader gives Orderdicts by default:

In [1]: import csv    ...: with open("in.csv") as csvfile:    ...:     reader = csv.DictReader(csvfile, delimiter=" ")    ...:     print(list(reader))    ...:      [OrderedDict([('first_name', 'Baked'), ('last_name', 'Beans')]), OrderedDict([('first_name', 'Lovely'), ('last_name', 'Spam')]), OrderedDict([('first_name', 'Wonderful'), ('last_name', 'Spam')])] 

You can access keys exactly the same, an OrderedDict just keeps key insertion order:

In [2]: import csv    ...: with open("in.csv") as csvfile:    ...:     reader = csv.DictReader(csvfile, delimiter=" ")    ...:     for dct in reader:    ...:         print(f"{dct['first_name']} {dct['last_name']}")    ...:             ...:      Baked Beans Lovely Spam Wonderful Spam 

Which py3.6 actually does too, so if for some reason you really want a dict:

In [5]: import csv    ...: with open("in.csv") as csvfile:    ...:     reader = csv.DictReader(csvfile, delimiter=" ")    ...:     for dct in map(dict, reader):    ...:         print(dct)    ...:         print(f"{dct['first_name']} {dct['last_name']}")    ...:             ...:      {'first_name': 'Baked', 'last_name': 'Beans'} Baked Beans {'first_name': 'Lovely', 'last_name': 'Spam'} Lovely Spam {'first_name': 'Wonderful', 'last_name': 'Spam'} Wonderful Spam 

The ordering retention on insertion in py3.6 is an implementation detail and may change, but if enough of us use it, it may just have to stay :)