Path: csiph.com!usenet.pasdenom.info!weretis.net!feeder4.news.weretis.net!feeds.phibee-telecom.net!newsfeed.xs4all.nl!newsfeed3.news.xs4all.nl!xs4all!post.news.xs4all.nl!not-for-mail
To: python-list@python.org
From: Peter Otten <__peter__@web.de>
Subject: Re: Group by interval time
Date: Thu, 12 Feb 2015 14:29:59 +0100
Organization: None
References: <6f6f567d-7e25-4a74-96ef-d20cb44a9646@googlegroups.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="ISO-8859-1"
Content-Transfer-Encoding: 7Bit
User-Agent: KNode/4.13.3
Precedence: list
Newsgroups: comp.lang.python
Message-ID: <mailman.18690.1423747817.18130.python-list@python.org>
Lines: 74
NNTP-Posting-Host: 2001:888:2000:d::a6
Xref: csiph.com comp.lang.python:85582

charles.sartori@gmail.com wrote:

> Hello there!
> 
> I`m trying to group by a list of Row() objects in 12days interval and
> sum(). values. Here is an example of the list
> 
> [Row(time=datetime.datetime(2013, 1, 1, 0, 0), sum=4676557380615),
> [Row(time=datetime.datetime(2013, 1, 2, 0, 0), sum=6549630855895),
> [Row(time=datetime.datetime(2013, 1, 3, 0, 0), sum=6549630855895), ...]
> 
> Row() objects has two vars: row.time and row.sum
> 
> 
> The result that I`m looking for is:
> [[datetime.datetime(2013, 1, 1, 0, 0), value],
> [datetime.datetime(2013, 1, 12, 0, 0), value],
> [datetime.datetime(2013, 1, 24, 0, 0), value]
> ]
> Where value is the sum() of all row.sum in that interval.
> 
> I`m trying to use itertools.groupby by I could not get it to work yet.

If the data is sorted by time then you can use groupby, otherwise consider 
putting it into a (default)dict. Here is an example for both methods:

import datetime
import random
from itertools import groupby
from collections import defaultdict, namedtuple

BASEDATE = datetime.datetime(2015, 1, 1)
INTERVAL = 12 # time interval in days
DAYS = datetime.timedelta(days=INTERVAL)

def make_sample_rows():
    random.seed(42)
    Row = namedtuple("Row", "time sum")
    return [
        Row(BASEDATE + datetime.timedelta(days=random.randrange(-20, 80)),
            random.randrange(300))
        for i in range(30)]

def get_key(row):
    offset = (row.time - BASEDATE).days // INTERVAL
    return BASEDATE + datetime.timedelta(days=offset)

def format_time(time):
    return time.strftime("%Y-%m-%d")

if __name__ == "__main__":
    rows = make_sample_rows()

    # with groupby()
    for key, group in groupby(sorted(rows), key=get_key):
        print("{} - {}".format(format_time(key), format_time(key+DAYS)))
        print("-" * 23)

        group = list(group)
        for row in group:
            print("{} {:4}".format(format_time(row.time), row.sum))
        print("{:>15}".format("----"))
        print("{:15}".format(sum(row.sum for row in group)))
        print("")

    # with defaultdict
    d = defaultdict(int)
    for row in rows:
        d[get_key(row)] += row.sum

    for time, sum in sorted(d.items()):
        print("{} {:4}".format(format_time(time), sum))