Best way to store date/time in mongodb

Question

I've seen using strings, integer timestamps and mongo datetime objects.

Answer 1

The best way is to store native JavaScript Date objects, which map onto BSON native Date objects.

> db.test.insert({date: ISODate()})
> db.test.insert({date: new Date()})
> db.test.find()
{ "_id" : ObjectId("..."), "date" : ISODate("2014-02-10T10:50:42.389Z") }
{ "_id" : ObjectId("..."), "date" : ISODate("2014-02-10T10:50:57.240Z") }

The native type supports a whole range of useful methods out of the box, which you can use in your map-reduce jobs, for example.

If you need to, you can easily convert Date objects to and from Unix timestamps¹⁾, using the getTime() method and Date(milliseconds) constructor, respectively.

¹⁾ Strictly speaking, the Unix timestamp is measured in seconds. The JavaScript Date object measures in milliseconds since the Unix epoch.

Answer 2

One datestamp is already in the _id object, representing insert time

So if the insert time is what you need, it's already there:

Login to mongodb shell

ubuntu@ip-10-0-1-223:~$ mongo 10.0.1.223
MongoDB shell version: 2.4.9
connecting to: 10.0.1.223/test

Create your database by inserting items

> db.penguins.insert({"penguin": "skipper"})
> db.penguins.insert({"penguin": "kowalski"})
> 

Lets make that database the one we are on now

> use penguins
switched to db penguins

Get the rows back:

> db.penguins.find()
{ "_id" : ObjectId("5498da1bf83a61f58ef6c6d5"), "penguin" : "skipper" }
{ "_id" : ObjectId("5498da28f83a61f58ef6c6d6"), "penguin" : "kowalski" }

Get each row in yyyy-MM-dd HH:mm:ss format:

> db.penguins.find().forEach(function (doc){ d = doc._id.getTimestamp(); print(d.getFullYear()+"-"+(d.getMonth()+1)+"-"+d.getDate() + " " + d.getHours() + ":" + d.getMinutes() + ":" + d.getSeconds()) })
2014-12-23 3:4:41
2014-12-23 3:4:53

If that last one-liner confuses you I have a walkthrough on how that works here: https://stackoverflow.com/a/27613766/445131

I'm not really sold on this nosql mongodb is webscale stuff. Maybe it's faster, in some cases, for very large json objects in some contexts, but the syntax and usage is strange and annoying.

Answer 3

I figured when you use pymongo, MongoDB will store the native Python datetime object as a Date field. This Date field in MongoDB could facilitate date-related queries later (e.g. querying intervals). Therefore, a code like this would work in Python

from datetime import datetime

datetime_now = datetime.utcnow()
new_doc = db.content.insert_one({"updated": datetime_now})

After this, I can see in my database a field like the following (I am using Mongo Compass to view my db). Note how it is not stored as a string (no quotation) and it shows Date as the field type.

Regarding javascript usage, this should also work there. As long as you have the +00:00 (UTC in my case) or Z at the end of your date, Javascript should be able to read the date properly with timezone information.

Answer 4

Use the code below to create a datetime variable that can be assigned in a document (Note that I'm creating a datetime object, not a date object):

from datetime import date
from datetime import datetime
import random

def random(date):
    my_year=random.randint(2020,2022)
    my_month=random.randint(1,12)
    my_day=random.randint(1,28)

    selected=datetime(year = my_year, month = my_month, day = my_day, hour = 0, minute = 0, second = 0)


def insert_objects(collection):

      collection.insert_one( { "mydate": random_date() })

Answer 5

BSON Date is a 64-bit integer that represents the number of milliseconds since the Unix epoch (Jan 1, 1970). This results in a representable date range of about 290 million years into the past and future.

The official BSON specification refers to the BSON Date type as the UTC datetime.

BSON Date type is signed. Negative values represent dates before 1970.