order raws based on name and number of header python-CodePudding

I have a csv file with raws of different sizes that contain values and a header for each raw.

00_A1_s1
00_A1_s2
00_A1_s3
04_A1_s1
04_A1_s2
04_A1_s3
08_A1_s1
08_A1_s2
08_A1_s3

and I want to group the raws by the name and then order them by the number in front of the coordinates

00_A1_s1
04_A1_s1
08_A1_s1
00_A1_s2
04_A1_s2
08_A1_s2
00_A1_s3
04_A1_s3
08_A1_s3

Thanks!

CodePudding user response：

You could use sort built-in function with the key parameter which accepts a function where you can infer the order condition on the elements. It returns a new list.

# file names
l = """00_A1_s1
00_A1_s2
00_A1_s3
04_A1_s1
04_A1_s2
04_A1_s3
08_A1_s1
08_A1_s2
08_A1_s3"""

l = l.split('\n') # file names as list

# sort wrt the last group of characters
print(sorted(l, key=lambda p: p.split('_')[-1]))

Output

04_A1_s1
08_A1_s1
00_A1_s2
04_A1_s2
08_A1_s2
00_A1_s3
04_A1_s3
08_A1_s3

CodePudding user response：

You can use sorted and use tuple for each item that is most important in sorting order like below:

txt = """00_A1_s1
00_A1_s2
00_A1_s3
04_A1_s1
04_A1_s2
04_A1_s3
08_A1_s1
08_A1_s2
08_A1_s3"""

lst = txt.split('\n') 

def orderSort(item):
    splt_itm = item.split('_')
    # first sorting on `A*` then on `s*` and at end consider number
    return (splt_itm[1] , splt_itm[2] , splt_itm[0])
    

result = sorted(lst, key=lambda x : orderSort(x))
print('\n'.join(result))

Output:

00_A1_s1
04_A1_s1
08_A1_s1
00_A1_s2
04_A1_s2
08_A1_s2
00_A1_s3
04_A1_s3
08_A1_s3