CODEWITHSUNDEEP
pythonlistpython-2.7
walko1234
walko1234

Reputation: 151

Sort list of dictonaries by dict value

Halo, want to sort lists of dicts by value:

a = [{'cavity' : '11A', 'ltg_nr' : 'Ltgnr11A'},
        {'cavity' : '20', 'ltg_nr' : 'Ltgnr20'},
        {'cavity' : '10', 'ltg_nr' : 'Ltgnr10'},
        {'cavity' : '17B', 'ltg_nr' : 'Ltgnr17B'},
        {'cavity' : '1K', 'ltg_nr' : 'Ltgnr1K'},
        {'cavity' : '11A', 'ltg_nr' : 'Ltgnr11A'},
        {'cavity' : '11A', 'ltg_nr' : 'Ltgnr11A'},
        {'cavity' : '22', 'ltg_nr' : 'Ltgnr22'},
        {'cavity' : '21A', 'ltg_nr' : 'Ltgnr21A'},
        {'cavity' : '21C', 'ltg_nr' : 'Ltgnr21C'},
        {'cavity' : '21B', 'ltg_nr' : 'Ltgnr170'},
        {'cavity' : '101', 'ltg_nr' : 'Ltgnr170'},
        {'cavity' : '22AA', 'ltg_nr' : 'Ltgnr170'},
        {'cavity' : '100A', 'ltg_nr' : 'Ltgnr170'}]
newList= sorted(a, key=lambda k: (len(k['cavity']),k['cavity']) )
print "-"*20 + ' sorted ' + "-"*20
for k in newList:
    print k['cavity']

and the result is not what I excepted:

C:\Python27>python.exe test_dict.py
-------------------- sorted --------------------
10
1K
20
22
101
11A
11A
11A
17B
21A
21B
21C
100A
22AA

I will first sort by numeric and then by alpha characters. How can I do that :

1K
10
11A
11A
11A
17B
20
21A
21B
21C
22
22AA
100A
101

All sugestions are welcome :) Thanx Piotr

Upvotes: 2

Views: 131

Answers (2)

jfs
jfs

Reputation: 414179

Using natural_key() function:

import re

def natural_key(astr):
    """See http://www.codinghorror.com/blog/archives/001018.html"""
    return [int(s) if re.match(r'\d+$', s) else s
            for s in re.split(r'(\d+)', astr)]

sorted_list = sorted(a, key=lambda d: natural_key(d['cavity']))

Result

[{'cavity': '1K', 'ltg_nr': 'Ltgnr1K'},
 {'cavity': '10', 'ltg_nr': 'Ltgnr10'},
 {'cavity': '11A', 'ltg_nr': 'Ltgnr11A'},
 {'cavity': '11A', 'ltg_nr': 'Ltgnr11A'},
 {'cavity': '11A', 'ltg_nr': 'Ltgnr11A'},
 {'cavity': '17B', 'ltg_nr': 'Ltgnr17B'},
 {'cavity': '20', 'ltg_nr': 'Ltgnr20'},
 {'cavity': '21A', 'ltg_nr': 'Ltgnr21A'},
 {'cavity': '21B', 'ltg_nr': 'Ltgnr170'},
 {'cavity': '21C', 'ltg_nr': 'Ltgnr21C'},
 {'cavity': '22', 'ltg_nr': 'Ltgnr22'},
 {'cavity': '22AA', 'ltg_nr': 'Ltgnr170'},
 {'cavity': '100A', 'ltg_nr': 'Ltgnr170'},
 {'cavity': '101', 'ltg_nr': 'Ltgnr170'}]

Upvotes: 3

jdi
jdi

Reputation: 92569

You are on the right track, but the primary key should probably not be len(k['cavity']). That will only consider the length of the string.

I think you need to extract the numeric component and specific have it sort on the number:

import re

sorted(a, key=lambda k: (int(re.sub(r'[A-Z]+$', '', k['cavity'])), k['cavity']) )

[{'cavity': '1K', 'ltg_nr': 'Ltgnr1K'},
 {'cavity': '10', 'ltg_nr': 'Ltgnr10'},
 {'cavity': '11A', 'ltg_nr': 'Ltgnr11A'},
 {'cavity': '11A', 'ltg_nr': 'Ltgnr11A'},
 {'cavity': '11A', 'ltg_nr': 'Ltgnr11A'},
 {'cavity': '17B', 'ltg_nr': 'Ltgnr17B'},
 {'cavity': '20', 'ltg_nr': 'Ltgnr20'},
 {'cavity': '21A', 'ltg_nr': 'Ltgnr21A'},
 {'cavity': '21B', 'ltg_nr': 'Ltgnr170'},
 {'cavity': '21C', 'ltg_nr': 'Ltgnr21C'},
 {'cavity': '22', 'ltg_nr': 'Ltgnr22'},
 {'cavity': '22AA', 'ltg_nr': 'Ltgnr170'},
 {'cavity': '100A', 'ltg_nr': 'Ltgnr170'},
 {'cavity': '101', 'ltg_nr': 'Ltgnr170'}]

The regular expression I am using is a bit specific. It expects that capital A-Z letters can be stripped away from the end of each cavity. I am basing that on your data set. It may need to be altered if cavity is not <numbers><optional letters>

Upvotes: 1

Related Questions