Problem with getting data from csv using pythonand writing them to a dictionary

Question

I have a CSV in a below format:

Expected: Where I want to convert this to dictionary like this:

{'Masterfolder': ['Training'], 'Childfolder': ['Training videos', 'Training documents', 'Training workouts', 'Training practicals']}

So far I have done the following code,

import csv

with open('features.csv', mode='r') as f:
reader = csv.reader(f)
checker = lambda i: bool(i and i.strip()) 
mydict = {rows[0]: list(filter(checker, rows[0:])) for rows in reader}

print(mydict)

And my output is something like this:

{'Master folder': ['Child - 1', 'Child - 2', 'Child - 3', 'Child - 4'], 'Training': ['Training videos', 'Training documents', 'Training workouts', 'Training practicals']}

How can I improve this code to get in order to get the result as I am expecting?

correct column separator used? (I mean like ';' instead of ',' - check the file by opening it in a text edior, like notepad++) — FObersteiner, Aug 05 '19 at 13:58
I think you need to separate the header. Also is the checker to make sure you only take cells 1 forward and only non empty data? — Buckeye14Guy, Aug 05 '19 at 14:10
@Buckeye14Guy, If it is there a way then how can I name all the column values from B2 till X2 as "Childfolder" ? and only A2 as "Masterfolder" ? thats my issue here — ilexcel, Aug 05 '19 at 14:13
Can you have multiple rows in your csv and if so what would the results look like? — DaveStSomeWhere, Aug 05 '19 at 14:42
@DaveStSomeWhere I wont have multiple rows anyways, I will have to use just these two rows — ilexcel, Aug 05 '19 at 14:44

score 1 · Answer 1 · answered Aug 05 '19 at 14:20

You can use Pandas

Read your CSV file with pd.read_csv and change the index

import pandas as pd 

df = pd.read_csv('Classeur1.csv', sep=';', index_col='Master folder')

output :

                   Child - 1          Child - 2          Child - 3           Child - 4
Master folder               
Training           Training videos    Training document Training workouts   Training praticals

Then make your dict

mydict = {'Master folder' : list(df.index),
          'Childfolder' : list(df.iloc[0])}

output :

{'Master folder': ['Training'],
 'Childfolder': ['Training videos','Training document','Training workouts','Training praticals']}

Buckeye14Guy · Answer 2 · 2019-08-05T15:01:45.800

0

This is what I mean by separating the header. The first element is your header so using next to get that out is best. Then you can work on the other rows.

with open(r'./data/temp.csv', 'r') as f:
    checker = lambda i: bool(i and i.strip())
    reader = csv.reader(f)
    header = next(reader)
    row = next(reader)
    mydict = {
        header[0]: [row[0]],
        'Childfolder': list(filter(checker, row[1:]))
    }

Or this way if you don't want to store the row at all

with open(r'./data/temp.csv', 'r') as f:
    checker = lambda i: bool(i and i.strip())
    reader = csv.reader(f)
    header = next(reader)
    mydict = next({
                      header[0]: [row[0]],
                      'Childfolder': list(filter(checker, row[1:]))
                  } for row in reader
                  )

edited Aug 05 '19 at 15:01

answered Aug 05 '19 at 14:18

Buckeye14Guy

831
6
12

So, how to separate for Masterfolder which is there in "A2" – ilexcel Aug 05 '19 at 14:20
I don't understand what you mean by separate Masterfolder. This gives the output you wanted though. A2 has 'Training; not Masterfolder. The only reason why I am returning a list of dictionaries is because I am assuming you have a csv file with many rows. – Buckeye14Guy Aug 05 '19 at 14:21
I mean when i add a variable like this: masterfolder = mydict['Master folder'] am getting error: TypeError: list indices must be integers or slices, not str – ilexcel Aug 05 '19 at 14:26
Where I should get the list of items from that keyword right? – ilexcel Aug 05 '19 at 14:29
Yea you should take the first element from the list. I mean your code is on a file with a single row or many rows? I added to the answer. `masterfolder = mydicts[0].get('Master folder')` please use the latest code. I changed `mydict` to `mydicts`. If you had more than 1 row and you wanted the master folder for the second row you could just do `masterfolder_row2 = mydicts[1].get('Master folder')` etc. – Buckeye14Guy Aug 05 '19 at 14:31
ya, but isn't there a way to assign all the items to a variable using keyword? – ilexcel Aug 05 '19 at 14:33
I'm basically trying this: masterfolder = mydicts['Master folder'] TypeError: list indices must be integers or slices, not str – ilexcel Aug 05 '19 at 14:36
and I am saying `mydicts` is not a dictionary. it is a list :). I mean think about it. If you have more than 1 row, the dictionary you have will not work – Buckeye14Guy Aug 05 '19 at 14:36
Let us [continue this discussion in chat](https://chat.stackoverflow.com/rooms/197502/discussion-between-ilexcel-and-buckeye14guy). – ilexcel Aug 05 '19 at 14:37
To simply use one dict comprehension as opposed to a list comprehension you can use this answer https://stackoverflow.com/questions/23983908/multiple-key-value-pairs-in-dict-comprehension – Buckeye14Guy Aug 05 '19 at 14:49
yes the new update should work. I am just not using a dictionary comprehension – Buckeye14Guy Aug 05 '19 at 15:00

Problem with getting data from csv using pythonand writing them to a dictionary

2 Answers2