Sorting in python with a dict

Question

I have an array of dicts. This works

tickets.sort(key=itemgetter('start')), however if the starts equal each other, I need a second criteria to sort by (If there's an easy way to do this, that would solve everything), so I'd like to write my own sort. I've done it before, in the form

tickets.sort(sort_function)

but it doesn't work with this, I get

unhashable type: 'dict' error.

Is there a way to do this?

EDIT

I forgot something important. For the second criteria, it's not a simple greater/less than case. It's along the lines of if start == "specific", return -1

you should add more info about the sort criteria you need. The solution should be something like key=lambda x: ... — Facundo Casco
– Facundo Casco, Commented Jul 18, 2011 at 22:14

g.d.d.c · Accepted Answer · 2011-07-18 20:59:46Z

5

You can simply pass a second parameter to itemgetter and it will create the necessary function for you:

tickets.sort(key = itemgetter('start', 'end'))

answered Jul 18, 2011 at 20:59

g.d.d.c

48.3k12 gold badges105 silver badges116 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

Community · Accepted Answer · 2017-05-23 11:47:50Z

…however if the starts equal each other, I need a second criteria to sort by (If there's an easy way to do this, that would solve everything)

There is an easy way to do this: use a tuple for each item's key. Because tuple comparison treats earlier items as more significant, later items are only considered when earlier ones match.

For example, given a list of items with two numbers, where sorting should be ascending on the first (more significant) number and descending on the second:

>>> data = [
...   (1, 0),
...   (1, 1),
...   (1, 2),
...   (0, 0),
...   (2, 0),
... ]
>>> data.sort(key=lambda (a, b): (a, -b))
>>> data
[(0, 0), (1, 2), (1, 1), (1, 0), (2, 0)]

I forgot something important. For the second criteria, it's not a simple greater/less than case. It's along the lines of if start == "specific", return -1

Use your special values in later items of the tuple in a way that turns them into a simple <, ==, or > comparison:

def ticket_key(ticket):
  start = ticket["start"]
  if start == "specific":
    second = -1
  else:
    second = ticket["blah"]
  return (start, second)

tickets.sort(key=ticket_key)

Now for items where ticket["start"] is "specific", -1 is used instead of ticket["blah"]:

>>> data = [
...   {"start": "specific"}, # "blah" doesn't even exist here
...   {"start": "specific", "blah": 2},
...   {"start": "42", "blah": 1},
...   {"start": "42", "blah": 0},
... ]
>>> data.sort(key=ticket_key)
>>> data
[{'start': '42', 'blah': 0},
 {'start': '42', 'blah': 1},
 {'start': 'specific'},
 {'start': 'specific', 'blah': 2}]

fletom · Accepted Answer · 2011-07-18 21:00:45Z

0

I think your problem is that the function you wrote somehow tries to use a dictionary as a key for another dictionary. Since dictionaries are unhashable, you get that error.

answered Jul 18, 2011 at 21:00

fletom

2,03814 silver badges18 bronze badges

Collectives™ on Stack Overflow

Sorting in python with a dict

3 Answers 3

Comments

Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

3 Answers 3

Comments

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related