NAME multi_key_sort
MODULE multi_key_sort
DESCRIPTION Module: multi_key_sort Contains functions for multiple-key, multiple-direction sort of iterable data
Author: HAP Proctor
Email: 48.41.50@gmail.com
GitHub: https://github.com/48-41-50/python-multi-key-sort
FUNCTIONS asorted(iterable, *keys) asorted(iterable, *keys) -> new sorted list This sort will perform a Pythonic stable sort on the iterable using operator.attrgetter and based on the keys.
Each key is a keyspec which is:
directional-prefix: ''
directional-prefix: '-'
directional-prefix: '+'
directional-suffix: ''
directional-suffix: ' desc'
directional-suffix: ' asc'
key-string: any valid key string
keyspec: directional-prefix key-string directional-suffix
So if you had data in a list of classes like this:
class SomeObj(object):
def __init__(self, name, age, sex):
self.name = name
self.age = age
self.sex = sex
data = [
SomeObj('Jim', 35, 'M'),
SomeObj('Sarah', 20, 'F'),
SomeObj("Alex", 35, 'M'),
SomeObj('Bill', 40, 'M'),
SomeObj('Wallace', 20, 'M'),
SomeObj('Linda', 25, F')
]
and you wanted to sort descending by age and ascending by name,
valid keyspecs would be:
'-age' or 'age desc'
'name' or '+name' or 'name asc'
Ascending sort is default so the '+' and ' asc' directionals are optional.
Given all of the above, the call to asorted would be like this:
asorted(data, 'age desc', 'name')
and return a new list of the dictionaries sorted by the keys and directionals.
complex_sorted(iterable, getter, *keys)
complex_sorted(iterable, getter, *keys) -> new sorted list
This sort will perform a Pythonic stable sort on the iterable and
the getter and based on the keys.
The getter must be either operator.itemgetter or operator.attrgetter
Each key is a keyspec which is:
directional-prefix: ''
directional-prefix: '-'
directional-prefix: '+'
directional-suffix: ''
directional-suffix: ' desc'
directional-suffix: ' asc'
key-string: any valid key string
keyspec: directional-prefix key-string directional-suffix
So if you had data in a list of dictionaries like this:
data = [
{'name': 'Jim', 'age': 35, 'sex': 'M'},
{'name': 'Sarah', 'age': 20, 'sex': 'F'},
{'name': "Alex", 'age': 35, 'sex': 'M'},
{'name': 'Bill', 'age': 40, 'sex': 'M'},
{'name': 'Wallace', 'age': 20, 'sex': 'M'},
{'name': 'Linda', 'age': 25, 'sex': 'F'}
]
and you wanted to sort descending by age and ascending by name,
valid keyspecs would be:
'-age' or 'age desc'
'name' or '+name' or 'name asc'
Ascending sort is default so the '+' and ' asc' directionals are optional.
Given all of the above, the call to complex_sorted would be like this:
...
from operator import itemgetter
...
complex_sorted(data, itemgetter, 'age desc', 'name')
and return a new list of the dictionaries sorted by the keys and directionals.
flatten_all(obj, res=None)
This will flatten a list of lists or a mix of elements and lists
into a single list. It will flatten ALL levels.
Ex:
flatten_all([1, [2, [3, 4]], 5]) -> [1, 2, 3, 4, 5]
isorted(iterable, *keys)
isorted(iterable, *keys) -> new sorted list
This sort will perform a Pythonic stable sort on the iterable using
operator.itemgetter and based on the keys.
Each key is a keyspec which is:
directional-prefix: ''
directional-prefix: '-'
directional-prefix: '+'
directional-suffix: ''
directional-suffix: ' desc'
directional-suffix: ' asc'
key-string: any valid key string
keyspec: directional-prefix key-string directional-suffix
So if you had data in a list of dictionaries like this:
data = [
{'name': 'Jim', 'age': 35, 'sex': 'M'},
{'name': 'Sarah', 'age': 20, 'sex': 'F'},
{'name': "Alex", 'age': 35, 'sex': 'M'},
{'name': 'Bill', 'age': 40, 'sex': 'M'},
{'name': 'Wallace', 'age': 20, 'sex': 'M'},
{'name': 'Linda', 'age': 25, 'sex': 'F'}
]
and you wanted to sort descending by age and ascending by name,
valid keyspecs would be:
'-age' or 'age desc'
'name' or '+name' or 'name asc'
Ascending sort is default so the '+' and ' asc' directionals are optional.
Given all of the above, the call to isorted would be like this:
isorted(data, 'age desc', 'name')
and return a new list of the dictionaries sorted by the keys and directionals.
msort(self, *keys)
If you subclass from a iterable base class, you can add this method
to an instance or to the class.
If you subclass from list, it will act like list.sort() except it will return
a reference to the instance.
If you do not subclass from list, it will return a new sorted list.
You will have to set a class or instance attribute named 'getter' to either
operator.itemgetter or operator.attrgetter depending on the data being sorted.
This sort will perform a Pythonic stable sort on the iterable using
the getter set by the instance or class attribute 'getter' and based on the keys.
Each key is a keyspec which is:
directional-prefix: ''
directional-prefix: '-'
directional-prefix: '+'
directional-suffix: ''
directional-suffix: ' desc'
directional-suffix: ' asc'
key-string: any valid key string
keyspec: directional-prefix key-string directional-suffix
So if you had data in a list of classes like this:
class SomeObj(object):
def __init__(self, name, age, sex):
self.name = name
self.age = age
self.sex = sex
data = [
SomeObj('Jim', 35, 'M'),
SomeObj('Sarah', 20, 'F'),
SomeObj("Alex", 35, 'M'),
SomeObj('Bill', 40, 'M'),
SomeObj('Wallace', 20, 'M'),
SomeObj('Linda', 25, F')
]
and you wanted to sort descending by age and ascending by name,
valid keyspecs would be:
'-age' or 'age desc'
'name' or '+name' or 'name asc'
Ascending sort is default so the '+' and ' asc' directionals are optional.
Given all of the above, the call to the method (assuming it was bound as 'msort')
would be like this:
self.msort('age desc', 'name')
and return a new list of the dictionaries sorted by the keys and directionals.
