Sort nested list: Exclude first item from sorting

Question

SET

I have a csv file which includes the current balance of my refreshments (it is comma separated, but in this example commas , are removed for improved readability):

NAME         AMOUNT         PRICE
Coca-Cola    8              1.25
Fanta        6              1.29
Dr. Pepper   2              2.20
Sprite       10             1.35
Guarana      6              1.80
Pepsi        4              1.25

I read all the data into the memory (list) and do all the necessary editing using:

import csv
# format list for refreshments
items_list = []

with open("my_refresments.csv", newline='') as fileOpener:
    open_csv = csv.reader(fileOpener)
    for rows in open_csv:
        items_list.append(rows)

The list 'items_list' now prints out as:

items_list = [["NAME","AMOUNT","PRICE"],
              ["Coca-Cola","8","1.25"],
              ["Fanta","6","1.29"],
              ["Dr. Pepper","2","2.20"],
              ["Sprite","10","1.35"],
              ["Guarana","6","1.80"],
              ["Pepsi","4","1.25"]]

QUESTION

What gives me this kind of sorting:

# Notice that this is sorted, but items_list[0] is at it's place

items_list = [["NAME","AMOUNT","PRICE"],
              ["Coca-Cola","8","1.25"],
              ["Fanta","6","1.29"],
              ["Dr. Pepper","2","2.20"],
              ["Sprite","10","1.35"],
              ["Guarana","6","1.80"],
              ["pepsi","4","1.25"]]

Everything except first row items_list[0] should stay unsorted ergo untouched where it stands.

Sorting the list

Generally a.sort() is a good option with nested lists because it will not sort all the individual items inside the lists, but instead it sorts the lists compared one to another. In this case I can't use items_list.sort(), because it will include also the headers in csv file, which I want to stay just where it is already located: items_list[0]. Here is what happens when using items_list.sort():

items_list.sort()
[['Coca-Cola', '8', '1.25'],
 ['Fanta', '6', '1.29'], 
 ['Dr. Pepper', '2', '2.20'], 
 ['Sprite', '10', '1.35'], 
 ['Guarana', '6', '1.80'], 
 ['NAME', 'AMOUNT', 'PRICE'], 
 ['Pepsi', '4', '1.25']]

Strangely, if I write all the refreshments with lower case, it will work, because TEXT > text, but I don't want to do that. I would like to use slice to exclude the first row (headers), but it doesn't seem to have any effect what so ever (or maybe I'm doing it wrong):

# This doesn't do anything
items_list[1:].sort()

One possible solution

What I could do is to

first copy headers into another list in one way or another temp_list = [items_list[0]]
delete items_list[0] using the del items_list[0] statement
sort the list using items_list.sort(), and finally
insert headers into the sorted list's index(0) with items_list.insert(0, temp_list)

like this:

def sort_csv_list_exclude_headers(file_name):

    items_list = []

    with open(file_name, newline='') as fileOpener:
        open_csv = csv.reader(fileOpener)
        for rows in open_csv:
            items_list.append(rows)

    temp_list = [items_list[0]]
    del items_list[0]
    items_list.sort()
    items_list.insert(0, temp_list[0])

    with open(file_name, "w") as fileWriter:
        write_csv = csv.writer(fileWriter, lineterminator='\n')
        write_csv.writerows(items_list)

sort_csv_list_exclude_headers("SODA_BALANCE.csv")

Actually this is pretty good and simple in general when using csv with > 1 000 000 rows of "refreshments".

Subquestion

Is there any other (more simple) method?

the `.sort` method works in place, so doing `items_list[1:].sort()` creates a new list, `items_list[1:]`, then you sort that new list, but the method returns `None`, and you never capture the result and thus the list is immediately discarded. Use `sorted`, which *returns a new list*. — juanpa.arrivillaga, Sep 25 '17 at 05:39

score 4 · Accepted Answer · answered Sep 25 '17 at 05:40

4

A clean one liner will be:

items_list[1:] = sorted(items_list[1:])

Refer: https://stackoverflow.com/a/5827649/937153

answered Sep 25 '17 at 05:40

Unni

5,348
6
36
55

Jiun Bae · Answer 2 · 2017-09-25T05:49:30.300

0

It is simple possible solution,

[items_list[0]] + sorted(items_list[1:])

or Use pandas If there's no reason not to use

pandas sort

thanks to juanpa.arrivillaga

To explain why use sorted() not .sort()

sorted() return sorted list, but .sort() is change list it self. (not return)

so If you want to use .sort() then code will be

values = items_list[1:].sort()
item_list = [item_list[0]] + values

using sorted is more short!

edited Sep 25 '17 at 05:49

answered Sep 25 '17 at 05:35

Jiun Bae

550
3
14

1

You should explain *why* `.sort` doesn't work – juanpa.arrivillaga Sep 25 '17 at 05:38
`.sort` is work well, but `sorted()` return sorted list, but `.sort` doesn't – Jiun Bae Sep 25 '17 at 05:42
@juanpa.arrivillaga I think this is good comment! – Jiun Bae Sep 25 '17 at 05:43
yes, add your explanation to the question by clicking [edit](https://stackoverflow.com/posts/46398224/edit) – juanpa.arrivillaga Sep 25 '17 at 05:44
What are the advatages of `pandas` in my application? – Kasperi Koski Sep 25 '17 at 05:50
Pandas offers many different features, and its very fast. when I handle large csv files or many columns, values, `pandas` is faster than just `csv` – Jiun Bae Sep 25 '17 at 05:54
@Maybe: Switching to pandas just to get this working is an absolute overkill. In general they are useful for handling large amount of data. – Unni Sep 25 '17 at 07:08
It may be, cuz I do not know the whole program – Jiun Bae Sep 25 '17 at 07:20

score 0 · Answer 3 · answered Sep 25 '17 at 05:38

The easiest way I can think of is to use pandas.

Read the csv using pandas:

df = pd.read_csv("my_refresments.csv")
df.sort_values("PRICE")

which sorts the data according to price, and the header is stored in df.columns, so it is not sorted with the actual data.

Sort nested list: Exclude first item from sorting

SET

QUESTION

Sorting the list

One possible solution

Subquestion

3 Answers3

Linked