Sorting date objects in python lists incorrect order

Question

So I have my python script which lists a report about each ec2 instance. I have a function which sorts the networkattatchtime of an instance but everytime i use stftime to format the output of the date , it seems to mess up the sorting and lists instances in a random order and not by the most oldest CreationDateTime. The output i get is as follows :

i-09dc54328002240ff,Aug 05 2021,asg-workxxx
i-048e92c5a4741d2b1,Mar 09 2017,False
i-0d649cebdf54bd2f4,Mar 12 2020,asg-dyyyyy
i-0ff596f1dc01b61d8,Mar 17 2021,asg-base-test
i-06db4eb158ad0b071,May 12 2021,False
i-0f285277529543462,May 18 2018,False
i-0f67cf99fb9f96c3f,Oct 14 2020,asg-elk-test
i-01734dfef0159c5c8,Oct 20 2020,asg-lb-test
i-0539c8dfc839cbfda,Oct 26 2020,asg-stand-base-test

You can see the CreationDateTime are not in a sorted order.

My code is as follows :

       response = ec2_client.describe_instances(
       
        MaxResults=10
    )
    
    # return json data from describe instances and filter what is needed
    instances = (len(response['Reservations']))
    header_row = 'InstanceID, CreationDateTime, AutoScalingGroupName' + '\n'
  

    for x in range(instances):

        # Get InstanceId
        instance_id = (response['Reservations'][x]
                       ['Instances'][0]['InstanceId'])
        # Get NetworkInterfacws AttatchTime
        network_interface_id = (
            response['Reservations'][x]['Instances'][0]['NetworkInterfaces'][0]['NetworkInterfaceId'])
        
        network_interface_details = ec2_client.describe_network_interfaces(
            NetworkInterfaceIds=[network_interface_id])
        networkinterface_id_attachedtime = network_interface_details[
            'NetworkInterfaces'][0]['Attachment']['AttachTime']
    
        time_between_insertion = datetime.now(
            timezone.utc) - networkinterface_id_attachedtime

        # Get Autoscaling GroupName
        tags = (response['Reservations'][x]['Instances'][0]['Tags'])
        autoscaling_group_name = get_tag(tags, 'aws:autoscaling:groupName')

        # print results
        if time_between_insertion.days > max_age:
            line = '{},{},{}'.format(
                instance_id, formatted_date_networkinterface_id, autoscaling_group_name)
            instances_list.append(line)

    sorted_list= sorted(instances_list, key=lambda v: v.split(',')[1])

    for instance in sorted_list:
        print(instance) ```

You are sorting a list of strings by sub-string sorted(instances_list, key=lambda v: v.split(',')[1]) so there is no "date" per se. Note in your output the "A" month dates appear first... — JonSG
– JonSG, Commented Sep 15, 2021 at 12:46

Stef · Accepted Answer · 2021-09-15 13:07:57Z

1

To parse dates, you can use datetime.datetime.strptime:

import datetime  # strptime
import operator  # itemgetter

data_unparsed = ['i-09dc54328002240ff,Aug 05 2021,asg-workxxx',
'i-048e92c5a4741d2b1,Mar 09 2017,False',
'i-0d649cebdf54bd2f4,Mar 12 2020,asg-dyyyyy',
'i-0ff596f1dc01b61d8,Mar 17 2021,asg-base-test',
'i-06db4eb158ad0b071,May 12 2021,False',
'i-0f285277529543462,May 18 2018,False',
'i-0f67cf99fb9f96c3f,Oct 14 2020,asg-elk-test',
'i-01734dfef0159c5c8,Oct 20 2020,asg-lb-test',
'i-0539c8dfc839cbfda,Oct 26 2020,asg-stand-base-test']

data = [((row := s.split(','))[0], datetime.datetime.strptime(row[1], '%b %d %Y'), row[2]) for s in data_unparsed]

data_sorted = sorted(data, key=operator.itemgetter(1))
print(data_sorted)
# [('i-048e92c5a4741d2b1', datetime.datetime(2017, 3, 9, 0, 0), 'False'),
#  ('i-0f285277529543462', datetime.datetime(2018, 5, 18, 0, 0), 'False'),
#  ('i-0d649cebdf54bd2f4', datetime.datetime(2020, 3, 12, 0, 0), 'asg-dyyyyy'),
#  ('i-0f67cf99fb9f96c3f', datetime.datetime(2020, 10, 14, 0, 0), 'asg-elk-test'),
#  ('i-01734dfef0159c5c8', datetime.datetime(2020, 10, 20, 0, 0), 'asg-lb-test'),
#  ('i-0539c8dfc839cbfda', datetime.datetime(2020, 10, 26, 0, 0), 'asg-stand-base-test'),
#  ('i-0ff596f1dc01b61d8', datetime.datetime(2021, 3, 17, 0, 0), 'asg-base-test'),
#  ('i-06db4eb158ad0b071', datetime.datetime(2021, 5, 12, 0, 0), 'False'),
#  ('i-09dc54328002240ff', datetime.datetime(2021, 8, 5, 0, 0), 'asg-workxxx')]

Alternatively, as a one-liner:

data_sorted = sorted(data_unparsed, key=lambda s: datetime.datetime.strptime(s.split(',')[1], '%b %d %Y'))

print(data_sorted)
# ['i-048e92c5a4741d2b1,Mar 09 2017,False',
#  'i-0f285277529543462,May 18 2018,False',
#  'i-0d649cebdf54bd2f4,Mar 12 2020,asg-dyyyyy',
#  'i-0f67cf99fb9f96c3f,Oct 14 2020,asg-elk-test',
#  'i-01734dfef0159c5c8,Oct 20 2020,asg-lb-test',
#  'i-0539c8dfc839cbfda,Oct 26 2020,asg-stand-base-test',
#  'i-0ff596f1dc01b61d8,Mar 17 2021,asg-base-test',
#  'i-06db4eb158ad0b071,May 12 2021,False',
#  'i-09dc54328002240ff,Aug 05 2021,asg-workxxx']

Relevant documentation:

answered Sep 15, 2021 at 13:07

Stef

15.7k2 gold badges22 silver badges39 bronze badges

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

3 Comments

devdude Over a year ago

I have just tried that and i get this error : raise ValueError("time data %r does not match format %r" % ValueError: time data '2020-03-12 10:07:07+00:00' does not match format '%b %d %Y' @Stef

Stef Over a year ago

@devdude, aye, the format I specified matches strings like "Aug 05 2021". If the strings are different, such as "2020-03-12 10:07:07+00:00", you need a different format. The documentation I linked explains how to write the format (probably "%Y-%m-%d %H:%M:%S%z" in the case of '2020-03-12 10:07:07+00:00'). Or you could try not specifying a format at all, and letting datetime figure it out. The danger in this case is that you have to make sure that datetime is not confusing day and month (ie 07-12 is 12 of july and not 7 of december).

Stef Over a year ago

If you have only those two formats, you could use a try / except block to try applying a format and if it fails, try applying the second format.

Collectives™ on Stack Overflow

Sorting date objects in python lists incorrect order

1 Answer 1

3 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

3 Comments

Your Answer

Sign up or log in

Post as a guest

Related