count objects created in django application in past X days, for each day

Question

I have following unsorted dict (dates are keys):

{"23-09-2014": 0, "11-10-2014": 0, "30-09-2014": 0, "26-09-2014": 0, "03-10-2014": 0, "19-10-2014": 0, "15-10-2014": 0, "22-09-2014": 0, "17-10-2014": 0, "29-09-2014": 0, "13-10-2014": 0, "16-10-2014": 0, "12-10-2014": 0, "25-09-2014": 0, "14-10-2014": 0, "08-10-2014": 0, "02-10-2014": 0, "09-10-2014": 0, "18-10-2014": 0, "24-09-2014": 0, "28-09-2014": 0, "10-10-2014": 0, "21-10-2014": 0, "20-10-2014": 0, "06-10-2014": 0, "04-10-2014": 0, "27-09-2014": 0, "05-10-2014": 0, "01-10-2014": 0, "07-10-2014": 0}

I am trying to sort it from oldest to newest. I've tried code:

mydict = OrderedDict(sorted(mydict .items(), key=lambda t: t[0], reverse=True))

to sort it, and it almost worked. It produced sorted dict, but it has ignored months:

{"01-10-2014": 0, "02-10-2014": 0, "03-10-2014": 0, "04-10-2014": 0, "05-10-2014": 0, "06-10-2014": 0, "07-10-2014": 0, "08-10-2014": 0, "09-10-2014": 0, "10-10-2014": 0, "11-10-2014": 0, "12-10-2014": 0, "13-10-2014": 0, "14-10-2014": 0, "15-10-2014": 0, "16-10-2014": 0, "17-10-2014": 0, "18-10-2014": 0, "19-10-2014": 0, "20-10-2014": 0, "21-10-2014": 0, "22-09-2014": 0, "23-09-2014": 0, "24-09-2014": 0, "25-09-2014": 0, "26-09-2014": 0, "27-09-2014": 0, "28-09-2014": 0, "29-09-2014": 0, "30-09-2014": 0}

How can I fix this?

EDIT: I need this to count objects created in django application in past X days, for each day.

event_chart = {}
date_list = [datetime.datetime.today() - datetime.timedelta(days=x) for x in range(0, 30)]
for date in date_list:
    event_chart[formats.date_format(date, "SHORT_DATE_FORMAT")] = Event.objects.filter(project=project_name, created=date).count()

event_chart = OrderedDict(sorted(event_chart.items(), key=lambda t: t[0]))
return HttpResponse(json.dumps(event_chart))

You need to convert those strings to datetime objects to make them sort correctly - have a look at the docs. Alternatively, if you can change format to 'yyyy-mm-dd' they will sort correctly as strings. — jonrsharpe
– jonrsharpe, Commented Oct 21, 2014 at 14:12
The problem is, I am already converting these dates from datetime objects. And each of them should localised. Please check my edit — dease
– dease, Commented Oct 21, 2014 at 14:20

ch3ka · Accepted Answer · 2014-10-21 14:18:09Z

0

You can use the datetime module to parse the strings into actual dates:

>>> from datetime import datetime
>>> sorted(mydict .items(), key=lambda t:datetime.strptime(t[0], '%d-%m-%Y'), reverse=True)

answered Oct 21, 2014 at 14:18

ch3ka

12.2k5 gold badges33 silver badges28 bronze badges

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

2 Comments

dease Over a year ago

I am using django, and each date should be localised. Please check my edit

ch3ka Over a year ago

Well, even better - there you already have date items - in order, even. Why do you destroy the order, only to re-sort in that order afterwards?

jfs · Accepted Answer · 2014-10-22 02:50:34Z

0

If you want to create a json response in the format: {"22-09-2014": 0, 23-09-2014": 0, "localized date": count_for_that_date} so that oldest dates will appear earlier in the output then you could make event_chart an OrderedDict:

event_chart = OrderedDict()
today = DT.date.today() # use DT.datetime.combine(date, DT.time()) if needed
for day in range(29, -1, -1): # last 30 days
    date = today - DT.timedelta(days=day)
    localized_date = formats.date_format(date, "SHORT_DATE_FORMAT")
    day_count = Event.objects.filter(project=name, created=date).count()
    event_chart[localized_date] = day_count
return HttpResponse(json.dumps(event_chart))

answered Oct 22, 2014 at 2:50

jfs

417k210 gold badges1k silver badges1.7k bronze badges

Collectives™ on Stack Overflow

count objects created in django application in past X days, for each day

2 Answers 2

2 Comments

Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

2 Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related