Write Python OrderedDict to CSV

Question

I have an ordered dictionary that, as one would expect, contains a number of key-value pairs. I need to parse the contents of the ordered dictionary into CSV format, with the keys in the top row and their respective values in the second row. I've got the first row down, but I'm at a loss as to how to write the second row. The values in the OrderedDict are what you'd expect (what it looks like printed out in the terminal): ([(a, 1), (b, 2), (c, 3)]). Here's what I have:

import csv
with open("frequencies.csv", "w") as outfile:
    csvwriter = csv.writer(outfile)
    csvwriter.writerow(myDict)

I tried iterating over the values and appending them to a list, but the list didn't maintain the original order of the ordered dictionary, which, as far as I could tell, would have made it pretty difficult to print the values from the list to the CSV so that they'd match up with the correct keys.

Thanks in advance for your help!

Solution 1

Alright, I'm going to answer my own question here. A couple of people were kind enough to offer suggestions in the comments. As suggested, I was working on accomplishing this with Pandas. As I was doing so, however, it occurred to me that I could do this without having to learn the ins and outs of the Pandas module. Here's what I came up with:

import csv

keys, values = [], []

for key, value in myOrderedDict.items():
    keys.append(key)
    values.append(value)       

with open("frequencies.csv", "w") as outfile:
    csvwriter = csv.writer(outfile)
    csvwriter.writerow(keys)
    csvwriter.writerow(values)

So here's what's going on here:

1. Create two empty lists corresponding to the keys and values in my ordered dictionary

2. Iterate over the key/value pairs in my ordered dictionary, appending each pair to its respective list. Because lists in Python retain their order, this ensures that items of corresponding indices in either list belong together

3. Write the keys to the first row of my CSV, and the values to the second

I'm sure there are more elegant ways to do this, but this is is sufficient for my purposes.

Solution 2

As of Python 3.7 dictionaries retain order so you can just use dict() to turn an ordered dictionary into a usable dictionary.

with open("frequencies.csv", "w") as outfile:
    csvwriter = csv.writer(outfile)
    csvwriter.writerow(dict(myDict))
    csvwriter.writerow(dict(myDict).values())

Solution 3

Here is another, more general solution assuming you don't have a list of rows (maybe they don't fit in memory) or a copy of the headers (maybe the write_csv function is generic):

def gen_rows():
    yield OrderedDict(a=1, b=2)

def write_csv():
    it = genrows()
    first_row = it.next()  # __next__ in py3
    with open("frequencies.csv", "w") as outfile:
        wr = csv.DictWriter(outfile, fieldnames=list(first_row))
        wr.writeheader()
        wr.writerow(first_row)
        wr.writerows(it)

Comments

[Marat]

did you try csv.DictWriter?

[colidyre]

OP does not want an unordered dict.

[Frederick Chute]

OP wants to "maintain the original order of the ordered dictionary" in his csv which this code does.

[colidyre]

How is the ordering kept by using dict()-function?

[Frederick Chute]

Its a language feature in python 3.7

[colidyre]

Thank you for clarifying it! It would be nice to add the information about the version into answer's text. The question was asked on december 2016. I think, Python 3.5 was the latest version here without any inserted order on dicts which come into play on 3.6 (without reliability), I think (not 100% sure about this) and is considered as language feature by Guido on 3.7 - - that's right. :)

[Nagabhushan S N]

Please reformat your code. import statement has got excluded from code