nutship nutship - 1 month ago 8
Python Question

Getting the closest date to a given date

Given this base date:

base_date = "10/29 06:58 AM"


I want to find a tuple within the list that contains the closest date to the
base_date
, but it must not be an earlier date.

list_date = [('10/30 02:18 PM', '-103', '-107'), ('10/30 02:17 PM', '+100', '-110'), \
('10/29 02:15 AM', '-101', '-109')


so here the output should be
('10/30 02:17 PM', '+100', '-110')
(it can't be the 3rd tuple because the date there happened earlier than the base date)

My question is, does it exist any module for such date comparison? I tried to first change the data all to
AM
format and then compare but my code gets ugly with lots of slicing.

@edit:

Big list to test:

[('10/30 02:18 PM', '+13 -103', '-13 -107'), ('10/30 02:17 PM', '+13 +100', '-13 -110'), ('10/30 02:15 PM', '+13 -101', '-13 -109'), ('10/30 02:14 PM', '+13 -103', '-13 -107'), ('10/30 01:59 PM', '+13 -105', '-13 -105'), ('10/30 01:46 PM', '+13 -106', '-13 -104'), ('10/30 01:37 PM', '+13 -105', '-13 -105'), ('10/30 01:24 PM', '+13 -107', '-13 -103'), ('10/30 01:23 PM', '+13 -106', '-13 -104'), ('10/30 01:05 PM', '+13 -103', '-13 -107'), ('10/30 01:02 PM', '+13 -104', '-13 -106'), ('10/30 12:55 PM', '+13 -103', '-13 -107'), ('10/30 12:51 PM', '+13.5 -110', '-13.5 +100'), ('10/30 12:44 PM', '+13.5 -108', '-13.5 -102'), ('10/30 12:38 PM', '+13.5 -107', '-13.5 -103'), ('10/30 12:35 PM', '+13 -102', '-13 -108'), ('10/30 12:34 PM', '+13 -103', '-13 -107'), ('10/30 12:06 PM', '+13.5 -110', '-13.5 +100'), ('10/30 11:57 AM', '+13.5 -108', '-13.5 -102'), ('10/30 11:36 AM', '+13.5 -107', '-13.5 -103'), ('10/30 09:01 AM', '+13.5 -110', '-13.5 +100'), ('10/30 08:59 AM', '+13.5 -108', '-13.5 -102'), ('10/30 08:13 AM', '+13.5 -105', '-13.5 -105'), ('10/30 06:11 AM', '+13.5 +100', '-13.5 -110'), ('10/30 06:09 AM', '+13.5 -105', '-13.5 -105'), ('10/30 06:04 AM', '+13.5 -110', '-13.5 +100'), ('10/30 05:32 AM', '+13.5 -105', '-13.5 -105'), ('10/30 04:48 AM', '+13.5 -107', '-13.5 -103'), ('10/30 12:51 AM', '+13.5 -110', '-13.5 +100'), ('10/29 01:31 PM', '+13.5 -105', '-13.5 -105'), ('10/29 01:31 PM', '+13 +103', '-13 -113'), ('10/29 01:28 PM', '+13 -102', '-13 -108'), ('10/29 07:59 AM', '+13 -105', '-13 -105'), ('10/29 07:20 AM', '+13 -103', '-13 -107'), ('10/29 07:14 AM', '+13 -105', '-13 -105'), ('10/29 04:47 AM', '+13 +100', '-13 -110'), ('10/29 04:14 AM', '+13 -105', '-13 -105'), ('10/28 08:17 PM', '+12.5 +100', '-12.5 -110'), ('10/28 12:52 PM', '+12.5 -105', '-12.5 -105')]


Big list to test2:

[('10/30 04:30 PM', '+1.5 -111', '-1.5 +101'), ('10/30 04:24 PM', '+1.5 -110', '-1.5 +100'), ('10/30 04:21 PM', '+1.5 -111', '-1.5 +101'), ('10/30 04:15 PM', '+1.5 -112', '-1.5 +102'), ('10/30 04:14 PM', '+1.5 -110', '-1.5 +100'), ('10/30 03:57 PM', '+1.5 -111', '-1.5 +101'), ('10/30 03:40 PM', '+1.5 -110', '-1.5 +100'), ('10/30 03:31 PM', '+1.5 -111', '-1.5 +101'), ('10/30 03:30 PM', '+1.5 -109', '-1.5 -101'), ('10/30 03:25 PM', '+1.5 -107', '-1.5 -103'), ('10/30 03:24 PM', '+1.5 -110', '-1.5 +100'), ('10/30 03:23 PM', '+1.5 -108', '-1.5 -102'), ('10/30 03:22 PM', '+1.5 -106', '-1.5 -104'), ('10/30 02:14 PM', '+1.5 -104', '-1.5 -106'), ('10/30 01:41 PM', '+1.5 -105', '-1.5 -105'), ('10/30 01:37 PM', '+1.5 -107', '-1.5 -103'), ('10/30 01:36 PM', '+1.5 -105', '-1.5 -105'), ('10/30 01:06 PM', '+1.5 -103', '-1.5 -107'), ('10/30 12:56 PM', '+2 -111', '-2 +101'), ('10/30 12:53 PM', '+2 -110', '-2 +100'), ('10/30 12:50 PM', '+2 -113', '-2 +103'), ('10/30 12:49 PM', '+2 -112', '-2 +102'), ('10/30 12:46 PM', '+2 -113', '-2 +103'), ('10/30 12:45 PM', '+2 -110', '-2 +100'), ('10/30 12:43 PM', '+2 -108', '-2 -102'), ('10/30 12:38 PM', '+2.5 -116', '-2.5 +106'), ('10/30 12:38 PM', '+2.5 -113', '-2.5 +103'), ('10/30 12:37 PM', '+2.5 -110', '-2.5 +100'), ('10/30 10:30 AM', '+2.5 -105', '-2.5 -105'), ('10/30 10:07 AM', '+3 -113', '-3 +103'), ('10/30 09:55 AM', '+3 -112', '-3 +102'), ('10/30 09:51 AM', '+3 -110', '-3 +100'), ('10/30 09:32 AM', '+3 -109', '-3 -101'), ('10/30 06:04 AM', '+3 -110', '-3 +100'), ('10/30 03:16 AM', '+3 -107', '-3 -103'), ('10/30 03:14 AM', '+3.5 -116', '-3.5 +106'), ('10/30 01:03 AM', '+3.5 -115', '-3.5 +105'), ('10/30 12:17 AM', '+3.5 -110', '-3.5 +100'), ('10/29 08:52 PM', '+3.5 -108', '-3.5 -102'), ('10/29 01:31 PM', '+3.5 -105', '-3.5 -105'), ('10/29 06:48 AM', '+3.5 -110', '-3.5 +100'), ('10/29 06:47 AM', '+3.5 -109', '-3.5 -101'), ('10/29 05:39 AM', '+3.5 -113', '-3.5 +103'), ('10/29 03:34 AM', '+3.5 -108', '-3.5 -102'), ('10/29 12:44 AM', '+3.5 -110', '-3.5 +100'), ('10/29 12:41 AM', '+3.5 -107', '-3.5 -103'), ('10/29 12:40 AM', '+3.5 -105', '-3.5 -105'), ('10/28 12:52 PM', '+4 -105', '-4 -105')]

Answer
>>> from datetime import timedelta, datetime
>>> base_date = "10/29 06:58 AM"
>>> b_d = datetime.strptime(base_date, "%m/%d %I:%M %p")
def func(x):
    d =  datetime.strptime(x[0], "%m/%d %I:%M %p")
    delta =  d - b_d if d > b_d else timedelta.max
    return delta
... 
>>> min(list_date, key = func)
('10/30 02:17 PM', '+100', '-110')

datetime.strptime converts the date to a datetime object, so b_d now looks something like this :

>>> b_d
datetime.datetime(1900, 10, 29, 6, 58)

Now we can write a function that can be passed to key parameter of min:

delta =  d - b_d if d > b_d else timedelta.max

if d > b_d i.e if the date passed to min is greater than base_date then assign their difference to delta else assign timedelta.max to it.

>>> timedelta.max
datetime.timedelta(999999999, 86399, 999999)

Update:

>>> from datetime import timedelta, datetime
>>> base_date = '10/29 06:59 AM'
>>> b_d = datetime.strptime(base_date, "%m/%d %I:%M %p")
>>> def func(x):
...         d =  datetime.strptime(x[0], "%m/%d %I:%M %p")
...         delta =  d - b_d if d > b_d else timedelta.max
...         return delta
... 
>>> lis2 = [('10/30 04:30 PM', '+1.5 -111', '-1.5 +101'), ('10/30 04:24 PM', '+1.5 -110', '-1.5     +100'), ('10/30 04:21 PM', '+1.5 -111', '-1.5 +101'), ('10/30 04:15 PM', '+1.5 -112', '-1.5 +102'), ('10/30 04:14 PM', '+1.5 -110', '-1.5 +100'), ('10/30 03:57 PM', '+1.5 -111', '-1.5 +101'), ('10/30 03:40 PM', '+1.5 -110', '-1.5 +100'), ('10/30 03:31 PM', '+1.5 -111', '-1.5 +101'), ('10/30 03:30 PM', '+1.5 -109', '-1.5 -101'), ('10/30 03:25 PM', '+1.5 -107', '-1.5 -103'), ('10/30 03:24 PM', '+1.5 -110', '-1.5 +100'), ('10/30 03:23 PM', '+1.5 -108', '-1.5 -102'), ('10/30 03:22 PM', '+1.5 -106', '-1.5 -104'), ('10/30 02:14 PM', '+1.5 -104', '-1.5 -106'), ('10/30 01:41 PM', '+1.5 -105', '-1.5 -105'), ('10/30 01:37 PM', '+1.5 -107', '-1.5 -103'), ('10/30 01:36 PM', '+1.5 -105', '-1.5 -105'), ('10/30 01:06 PM', '+1.5 -103', '-1.5 -107'), ('10/30 12:56 PM', '+2 -111', '-2 +101'), ('10/30 12:53 PM', '+2 -110', '-2 +100'), ('10/30 12:50 PM', '+2 -113', '-2 +103'), ('10/30 12:49 PM', '+2 -112', '-2 +102'), ('10/30 12:46 PM', '+2 -113', '-2 +103'), ('10/30 12:45 PM', '+2 -110', '-2 +100'), ('10/30 12:43 PM', '+2 -108', '-2 -102'), ('10/30 12:38 PM', '+2.5 -116', '-2.5 +106'), ('10/30 12:38 PM', '+2.5 -113', '-2.5 +103'), ('10/30 12:37 PM', '+2.5 -110', '-2.5 +100'), ('10/30 10:30 AM', '+2.5 -105', '-2.5 -105'), ('10/30 10:07 AM', '+3 -113', '-3 +103'), ('10/30 09:55 AM', '+3 -112', '-3 +102'), ('10/30 09:51 AM', '+3 -110', '-3 +100'), ('10/30 09:32 AM', '+3 -109', '-3 -101'), ('10/30 06:04 AM', '+3 -110', '-3 +100'), ('10/30 03:16 AM', '+3 -107', '-3 -103'), ('10/30 03:14 AM', '+3.5 -116', '-3.5 +106'), ('10/30 01:03 AM', '+3.5 -115', '-3.5 +105'), ('10/30 12:17 AM', '+3.5 -110', '-3.5 +100'), ('10/29 08:52 PM', '+3.5 -108', '-3.5 -102'), ('10/29 01:31 PM', '+3.5 -105', '-3.5 -105'), ('10/29 06:48 AM', '+3.5 -110', '-3.5 +100'), ('10/29 06:47 AM', '+3.5 -109', '-3.5 -101'), ('10/29 05:39 AM', '+3.5 -113', '-3.5 +103'), ('10/29 03:34 AM', '+3.5 -108', '-3.5 -102'), ('10/29 12:44 AM', '+3.5 -110', '-3.5 +100'), ('10/29 12:41 AM', '+3.5 -107', '-3.5 -103'), ('10/29 12:40 AM', '+3.5 -105', '-3.5 -105'), ('10/28 12:52 PM', '+4 -105', '-4 -105')]
>>> min(lis2, key = func)
('10/29 01:31 PM', '+3.5 -105', '-3.5 -105')

Timing comparisons:

Script:

from datetime import datetime, timedelta
import sys
import time
list_date = [('10/30 04:30 PM', '+1.5 -111', '-1.5 +101'), ('10/30 04:24 PM', '+1.5 -110', '-1.5     +100'), ('10/30 04:21 PM', '+1.5 -111', '-1.5 +101'), ('10/30 04:15 PM', '+1.5 -112', '-1.5 +102'), ('10/30 04:14 PM', '+1.5 -110', '-1.5 +100'), ('10/30 03:57 PM', '+1.5 -111', '-1.5 +101'), ('10/30 03:40 PM', '+1.5 -110', '-1.5 +100'), ('10/30 03:31 PM', '+1.5 -111', '-1.5 +101'), ('10/30 03:30 PM', '+1.5 -109', '-1.5 -101'), ('10/30 03:25 PM', '+1.5 -107', '-1.5 -103'), ('10/30 03:24 PM', '+1.5 -110', '-1.5 +100'), ('10/30 03:23 PM', '+1.5 -108', '-1.5 -102'), ('10/30 03:22 PM', '+1.5 -106', '-1.5 -104'), ('10/30 02:14 PM', '+1.5 -104', '-1.5 -106'), ('10/30 01:41 PM', '+1.5 -105', '-1.5 -105'), ('10/30 01:37 PM', '+1.5 -107', '-1.5 -103'), ('10/30 01:36 PM', '+1.5 -105', '-1.5 -105'), ('10/30 01:06 PM', '+1.5 -103', '-1.5 -107'), ('10/30 12:56 PM', '+2 -111', '-2 +101'), ('10/30 12:53 PM', '+2 -110', '-2 +100'), ('10/30 12:50 PM', '+2 -113', '-2 +103'), ('10/30 12:49 PM', '+2 -112', '-2 +102'), ('10/30 12:46 PM', '+2 -113', '-2 +103'), ('10/30 12:45 PM', '+2 -110', '-2 +100'), ('10/30 12:43 PM', '+2 -108', '-2 -102'), ('10/30 12:38 PM', '+2.5 -116', '-2.5 +106'), ('10/30 12:38 PM', '+2.5 -113', '-2.5 +103'), ('10/30 12:37 PM', '+2.5 -110', '-2.5 +100'), ('10/30 10:30 AM', '+2.5 -105', '-2.5 -105'), ('10/30 10:07 AM', '+3 -113', '-3 +103'), ('10/30 09:55 AM', '+3 -112', '-3 +102'), ('10/30 09:51 AM', '+3 -110', '-3 +100'), ('10/30 09:32 AM', '+3 -109', '-3 -101'), ('10/30 06:04 AM', '+3 -110', '-3 +100'), ('10/30 03:16 AM', '+3 -107', '-3 -103'), ('10/30 03:14 AM', '+3.5 -116', '-3.5 +106'), ('10/30 01:03 AM', '+3.5 -115', '-3.5 +105'), ('10/30 12:17 AM', '+3.5 -110', '-3.5 +100'), ('10/29 08:52 PM', '+3.5 -108', '-3.5 -102'), ('10/29 01:31 PM', '+3.5 -105', '-3.5 -105'), ('10/29 06:48 AM', '+3.5 -110', '-3.5 +100'), ('10/29 06:47 AM', '+3.5 -109', '-3.5 -101'), ('10/29 05:39 AM', '+3.5 -113', '-3.5 +103'), ('10/29 03:34 AM', '+3.5 -108', '-3.5 -102'), ('10/29 12:44 AM', '+3.5 -110', '-3.5 +100'), ('10/29 12:41 AM', '+3.5 -107', '-3.5 -103'), ('10/29 12:40 AM', '+3.5 -105', '-3.5 -105'), ('10/28 12:52 PM', '+4 -105', '-4 -105')]

base_date = "10/29 06:58 AM"

def func1(list_date):
    #http://stackoverflow.com/a/17249420/846892
    get_datetime = lambda s: datetime.strptime(s, "%m/%d %I:%M %p")
    base = get_datetime(base_date)
    later = filter(lambda d: get_datetime(d[0]) > base, list_date)
    return min(later, key = lambda d: get_datetime(d[0]))

def func2(list_date):
    #http://stackoverflow.com/a/17249470/846892
    b_d = datetime.strptime(base_date, "%m/%d %I:%M %p")
    def func(x):
       d =  datetime.strptime(x[0], "%m/%d %I:%M %p")
       delta =  d - b_d if d > b_d else timedelta.max
       return delta
    return min(list_date, key = func)

def func3(list_date):
    #http://stackoverflow.com/a/17249529/846892
    fmt = '%m/%d %I:%M %p'
    d = datetime.strptime(base_date, fmt)
    def foo(x):
        return (datetime.strptime(x[0],fmt)-d).total_seconds() > 0
    return sorted(list_date, key=foo)[-1]

def func4(list_date):
    #http://stackoverflow.com/a/17249441/846892
    fmt = '%m/%d %I:%M %p'
    base_d = datetime.strptime(base_date, fmt)
    candidates = ((datetime.strptime(d, fmt), d, x, y) for d, x, y in list_date)
    candidates = min((dt, d, x, y) for dt, d, x, y in candidates if dt > base_d)
    return  candidates[1:]

Results:

>>> from so import *

#check output irst
>>> func1(list_date)
('10/29 01:31 PM', '+3.5 -105', '-3.5 -105')
>>> func2(list_date)
('10/29 01:31 PM', '+3.5 -105', '-3.5 -105')
>>> func3(list_date)
('10/29 01:31 PM', '+3.5 -105', '-3.5 -105')
>>> func4(list_date)
('10/29 01:31 PM', '+3.5 -105', '-3.5 -105')

>>> %timeit func1(list_date)
100 loops, best of 3: 3.07 ms per loop
>>> %timeit func2(list_date)
100 loops, best of 3: 1.59 ms per loop      #winner
>>> %timeit func3(list_date)
100 loops, best of 3: 1.91 ms per loop
>>> %timeit func4(list_date)
1000 loops, best of 3: 2.02 ms per loop

#increase the input size
>>> list_date = list_date *10**3
>>> len(list_date)
48000
>>> %timeit func1(list_date)
1 loops, best of 3: 3.6 s per loop
>>> %timeit func2(list_date)            #winner
1 loops, best of 3: 1.99 s per loop      
>>> %timeit func3(list_date)
1 loops, best of 3: 2.09 s per loop
>>> %timeit func4(list_date)
1 loops, best of 3: 2.02 s per loop


#increase the input size again

>>> list_date = list_date *10
>>> len(list_date)
480000
>>> %timeit func1(list_date)
1 loops, best of 3: 36.4 s per loop
>>> %timeit func2(list_date)                  #winner
1 loops, best of 3: 20.2 s per loop           
>>> %timeit func3(list_date)
1 loops, best of 3: 22.8 s per loop
>>> %timeit func4(list_date)
1 loops, best of 3: 22.7 s per loop
Comments