Creating Date Intervals in Python

Question

I want to use a for loop to print every date between 2 dates. Actually going to merge this with a MySQL query to pass the date into the query itself and into the filename of the output.

So, how can I change this:

sum = 0
for i in range(1,11):
 print sum
 sum += i

To this?

InputDate = '2009-01-01'
for i in range('2009-01-01','2009-07-01'):
 print InputDate 
 InputDate += i

I realize there is something in rrule that does this exact function:

a = date(2009, 1, 1)
b = date(2009, 7, 1)
for dt in rrule(DAILY, dtstart=a, until=b):
 print dt.strftime("%Y-%m-%d")

But, I am restricted to older version of python.

This is the shell script version of what I am trying to do, if this helps clarify:

while [InputDate <= EndDate]
do
    sql="SELECT Date,SUM(CostUsd) FROM DailyStats WHERE Date = '$InputDate' GROUP BY Date"
    name=$(mysql -h -sN -u -p -e "$sql" > DateLoop-$InputDate.txt db)
    echo "$name"
    InputDate=$(( InputDate + 1 ))
done

So how can I do this in Python?

Adding follow up question here for readability. Unfortunately I can not use standard MySQL library as we have a proprietary setup with numerous instances running in parallel. The only way to run this type of query is to connect to one instance at a time, on the command line.

while day <= b: print "Running extract for :" day

sql="SELECT Date,SUM(CostUsd) FROM Stats d WHERE d.Date = " + day + " GROUP BY Date"

os.system('mysql -h -sN -u -p -e " + sql + " > DateLoop-" + day + ".txt db')

day += one_day

Answer 1

This will work:

import datetime

a = datetime.date(2009, 1, 1)
b = datetime.date(2009, 7, 1)
one_day = datetime.timedelta(1)

day = a

while day <= b:
    # do important stuff
    day += one_day

Answer 2

Try this:

import datetime
dt1 = datetime.date(2009, 1, 1)
dt2 = datetime.date(2009, 7, 1)
dt = dt1
while dt <= dt2:
    print dt.strftime("%Y-%m-%d")
    dt += datetime.timedelta(days=1)

You say you are restricted to an older version of Python. If you don't have the datetime module (Python < 2.3), then you can also do:

import time
dt1 = time.mktime(time.strptime('2009-01-01', '%Y-%m-%d'))
dt2 = time.mktime(time.strptime('2009-07-01', '%Y-%m-%d'))
ONE_DAY = 86400
dt = dt1
while dt <= dt2:
    print time.strftime('%Y-%m-%d', time.gmtime(dt))
    dt += ONE_DAY

Answer 3

In [1]: from dateutil.relativedelta import *

In [2]: from datetime import *

In [3]: aday = datetime.today()

In [4]: nextweek = aday+relativedelta(weeks=1)

In [5]: while aday<nextweek:
   ...:     print datetime.strftime(aday,format='%Y-%b-%d')
   ...:     aday+=relativedelta(days=1)
   ...:     
   ...:     
    #Output
2009-Aug-03
2009-Aug-04
2009-Aug-05
2009-Aug-06
2009-Aug-07
2009-Aug-08
2009-Aug-09

While you can do, most of the stuff about datetime using the stdlib, (IMO) dateutil lets you do it faster and better.