How to parse specific columns in a csv with python

Question

I would like to turn this data into a dictionary however I want the 'name' column to be the key and only the age and height as values to be associated with the name and everything else is irrelevant.

The CSV file is arranged like so name | age | color | height | weight steven | 15 | red | 70 | 150 many other names with characteristics are below this row as well

I am new to parsing csv files and can't seem to find the framework for what I am trying to do. Is there a place I can maybe put the indexes of the columns?

with open("search_rez.csv","w+", encoding="utf-8") as out:
    out.write(final_data)
    out.close
with open('search_rez.csv', 'r') as fh:
    rd = csv.DictReader(fh, delimiter=',') 
    for row in rd:
    print(row)

A dictionary of key's(names) and values(many characteristics) that I can access and is stored into a variable. Eventually I would like to be able to search the dictionary with a given 'name' and return the values associated with it.

The question is a bit confusing. Can you add (a) example of your csv file and (b) what exactly you are trying to achieve from that csv file data. That question does not provide any information on that. — Gravity Mass
– Gravity Mass, Commented Jul 24, 2019 at 17:51

Tim Mironov · Accepted Answer · 2019-07-24 18:14:41Z

2

I've created a dummy csv file:

import pandas as pd
import csv

test_data = {'name': ['sansa', 'bran', 'aria'], 'sex': ['F','M','F'], 'age': [15,10,12]}
test_data = pd.DataFrame(test_data)
test_data.to_csv("search_rez.csv", index=False)

This file looks like this:

data = pd.read_csv("search_rez.csv")
print(data)

    name sex  age
0  sansa   F   15
1   bran   M   10
2   aria   F   12

I've modified your code to implement the desired output (as I understood it):

my_dict = {}
with open('search_rez.csv', 'r') as fh:
    rd = csv.DictReader(fh, delimiter=',') 
    for row in rd:
        new_key = row['name']
        new_values = (row['sex'], row['age'])
        my_dict[new_key] = new_values

Which gives:

print(my_dict)

{'sansa': ('F', '15'), 'bran': ('M', '10'), 'aria': ('F', '12')}

edited Jul 24, 2019 at 18:14

answered Jul 24, 2019 at 18:01

Tim Mironov

8599 silver badges22 bronze badges

Sign up to request clarification or add additional context in comments.

4 Comments

ubersuperuser Over a year ago

Yes! This will definetely help, where you use "A", "B", etc should I use the column headers or an index?

Tim Mironov Over a year ago

@ubersuperuser, These are column headers, just play with it and look how the resulting csv looks like. I just saw your edit on original question, so in your case it would be the name|age| etc.

ubersuperuser Over a year ago

Thanks! Do i need to use the panda module at all? I saw you installed it.

Tim Mironov Over a year ago

@ubersuperuser, I just used it to create an example csv, since I didn't have your file.

Ram K · Accepted Answer · 2019-07-24 18:02:37Z

1

The question is quite vague but I think you are looking for something like this :

df = pd.read_csv("your file here")

your_map = dict(
        df.set_index('name')
        .groupby(level=0)
        .apply(lambda x: x.to_dict(orient='records'))
)

answered Jul 24, 2019 at 18:02

Ram K

1,7952 gold badges16 silver badges25 bronze badges

Comments

Nikos Oikou · Accepted Answer · 2019-07-24 18:23:54Z

1

You can try this:

import csv

with open('search_rez.csv') as csvfile:
    reader = csv.reader(csvfile)
    next(reader)  # Skip header 
    my_dict = {rows[0]: rows[1:] for rows in reader}

Example input:

name,D,Q,L
nikos,1,,5
maria,7,'xa',4

And the output, stored in my_dict:

{'nikos': ['1', '', '5'], 'maria': ['7', "'xa'", '4']}

edited Jul 24, 2019 at 18:23

answered Jul 24, 2019 at 18:03

Nikos Oikou

2411 silver badge5 bronze badges

Comments

score 0 · Accepted Answer · 2019-07-24 17:59:16Z

0

Something like this?

>>> import csv
>>> with open("search_rez.csv", newline='') as csvfile:
...     reader = csv.DictReader(csvfile)
...     for row in reader:
...         if row['name'] == 'some_name':
...             print(row['columnD'], row['columnQ'])
...

edited Jul 24, 2019 at 17:59

answered Jul 24, 2019 at 17:47

user11553043

1 Comment

ubersuperuser Over a year ago

Not exactly, but is there a way to pre-associate the 'name' with specific column values?

Collectives™ on Stack Overflow

How to parse specific columns in a csv with python

4 Answers 4

4 Comments

Comments

Comments

1 Comment

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

4 Answers 4

4 Comments

Comments

Comments

1 Comment

Your Answer

Sign up or log in

Post as a guest

Related