• 13

A PHP Error was encountered

Severity: Notice

Message: Undefined index: userid

Filename: views/question.php

Line Number: 191


File: /home/prodcxja/public_html/questions/application/views/question.php
Line: 191
Function: _error_handler

File: /home/prodcxja/public_html/questions/application/controllers/Questions.php
Line: 433
Function: view

File: /home/prodcxja/public_html/questions/index.php
Line: 315
Function: require_once

name Punditsdkoslkdosdkoskdo

Python Dictionary to CSV

I have written code to read a CSV into a python dictionary, which works fine. I'm trying to get the dictionary back to a CSV. I have written the following:

import csv


listReader = csv.reader(open('/Users/broberts/Desktop/Sum_CSP1.csv','rU'), delimiter = ',', quotechar='|')

for row in listReader:
    fID = row[0]
    fClassRange = row[1]
    fArea = row[2]

    if itemDict.has_key(fID):
        itemDict[fID] = {'5.0 to 5.25':'','5.25 to 5.5':'','5.5 to 5.75':'','5.75 to 6.0':'','6.0 to 6.25':'','6.25 to 6.5':'','6.5 to 6.75':'','6.75 to 7.0':'','7.0 to 7.25':'','7.25 to 7.5':'','7.5 to 7.75':'','7.75 to 8.0':'','8.0 to 8.25':'',}

listWriter = csv.writer(open('/Users/broberts/Desktop/Sum_CSP1_output.csv', 'wb'), delimiter=',', quotechar='|', quoting=csv.QUOTE_MINIMAL)

for a in itemDict:
    print a

In the last block, listWriter will not write anything to the CSV though it will print a. I believe this has something to do with a dictionary being unordered. I really need to write out the fID and each of the keys associated with each fID (fClassRange ex. "5.0 to 5.25") and then the value fArea associated with each fClassRange to the CSV, but I haven't even gotten that far in my code since I can't figure out how to write out even the fID.

I looked into using DictWriter, but I can't figure out how to tell it what the required fieldnames are.

    • Have to recommend Kenneth Reitz's tablib. It does more than what you were looking for here, so this is not a direct answer, just want to recommend this library to others. It has a great, easy to use API and you can serialize to csv, tsv, json, yaml, and xlsx effortlessly.

The default writer expects a list, which is why it won't work for you. To use the dictwriter, just change your listwriter = line to this:

with open('/Users/broberts/Desktop/Sum_CSP1_output.csv', 'wb') as outfile:
    listWriter = csv.DictWriter(

Or, you can just set fieldnames to be fieldnames=['arbitrary','list','of','keys'] if you know what the fields are supposed to be.

  • 21
Reply Report
    • @bojo Whoops! You're making a dict of dicts. So to get the correct fieldnames you need the internal dictionary fieldnames. Also, each dictionary has to have the same fieldnames. Or else you need to use the DictWriter options for incorrect fieldnames. I've updated the line to use the correct dictionary for getting the fieldnames.
      • 1
    • @bojo Hmm, on your itemDict[fid] = line, you set the fieldnames of your new dictionary to some default values. Could you use those in a list as your fields? Don't forget the fClassRange field you add afterwards. Or, change your print a statement to print a.keys() and just use that.
      • 1
    • I really, really appreciate your help. I tried changing "print a" to "print a.keys()" just to see what happened, and I got: AttributeError: 'str' object has no attribute 'keys' -- maybe this indicates a problem? I also tried using the DictWriter statement where "fieldnames = myFieldnames" and "myFieldnames" was equal to the list of default class ranges as you suggest, but I still get the same ValueError.
    • Wow, thanks! But now I'm getting this message: ValueError: dict contains fields not in fieldnames: 4, 6, 3, 3, 9

Sample data:

mydict = [{"col1": 1000, "col2": 2000}, {"col1": 3000, "col2": 4000}]

One-liner for converting a list of dicts to CSV, using pandas:

import pandas as pd

pd.DataFrame(mydict).to_csv('out.csv', index=False)


  • 13
Reply Report

This is what i use, its simple and works fine for me. when you have only one dictionary, use this

my_dict = {"tester": 1, "testers": 2}
with open('mycsvfile.csv', 'wb') as f:  
    w = csv.DictWriter(f, my_dict.keys())
    w.writerow(dict((fn,fn) for fn in my_dict.keys()))

$ cat mycsvfile.csv

When you have a list of dictionaries, like what you get from SQL queries, you do like this.

my_dict = ({"tester": 1, "testers": 2},{"tester": 14, "testers": 28})
with open('mycsvfile.csv', 'wb') as f:  
    w = csv.DictWriter(f, my_dict[0].keys())
    w.writerow(dict((fn,fn) for fn in my_dict[0].keys()))

cat mycsvfile.csv
  • 3
Reply Report

For posterity:

You should use iteritems() to iterate over a dictionary, so the last part becomes

for name, values in itemDict.iteritems():
    print values
  • 1
Reply Report

Easiest Way

You can convert the dictionary into Dataframe and write it to csv Eg

import pandas as pd
my_dict = {"tester": 1, "testers": 2}
df.to_csv("path and name of your csv.csv")


   tester  testers
0       1        2
  • -2
Reply Report

Trending Tags