Recursive query: comparing one field to another of the same class - django

I have a model similar to this one:
class Trip(models.Model):
departure = models.DateTimeField()
arrival = models.DateTimeField()
And I want to make a query that returns objects where the arrival is at least 2 hours later than the departure.
Trip.objects.filter(arrival__gt = "departure" + timedelta(hours=2))
Is this even possible? thanks

You are looking for filters that reference fields ont the model
But what if you want to compare the value of a model field with
another field on the same model?
See this sample:
>>> from datetime import timedelta
>>> Entry.objects.filter(mod_date__gt=F('pub_date') + timedelta(days=3))
For your case:
from django.db.models import F
Trip.objects.filter(arrival__gt = F('departure') + timedelta(hours=2))

Related

Django / PostGreSQL: Create queryset grouped by 'date' when each row has a different timezone

Let's say I have two models:
from django.db import model
class Company(model.Model):
name = models.TextField()
timezone = models.TextField()
class Sale(models.Model):
amount = models.IntegerField()
company = models.ForeignKey('Company')
time = models.DateTimeField()
I want to create a queryset grouped by date and company, where date refers to the calendar date of the sale at the timezone specified on the Company object.
This query:
result = Sale.objects.values(
'company', 'time__date'
).aggregate(
models.Sum('amount')
)
This returns the data in a format that works for me. However, the sales are grouped by UTC day. I want them grouped by the timezone on the Company objects.
What is the cleanest, quickest way to do this?
I know I could dump the entire set of values into Python, like this:
result = Sale.objects.values(
'amount', 'company__timezone', 'time'
).order_by(
'company_timezone'
)
for r in result:
r.date = r.time.astimezone(pytz.timezone(r.company_timezone)).date()
and then groupby, but is there a better way?
The solution is to use the TruncDate function, and pass the timezone string as an argument.
from django.db.models.functions import TruncDate
from django.db.models import F
...
local_time_daily_sales = Sale.objects.annotate(
date=TruncDate(tzinfo=F('company__timezone'))
).values(
date
).annotate(Sum('amount'))

Django: Combine a date and time field and filter

I have a django model that has a date field and a separate time field. I am trying to use a filter to find a value on the latest record by date/time that is less than the current record's date time.
How do I use annotate/aggregate to combine the date and time fields into one and then do a filter on it?
models.py
class Note(models.model):
note_date = models.DateField(null=True)
note_time = models.TimeField(null=True)
note_value = models.PositiveIntegerField(null=True)
def get_last(n):
"""
n: Note
return: Return the note_value of the most recent Note prior to given Note.
"""
latest = Note.objects.filter(
note_date__lte=n.note_date
).order_by(
'-note_date', '-note_time'
).first()
return latest.note_value if latest else return 0
This will return any notes from a previous date, but if I have a two notes on the same date, one at 3pm and one at 1pm, and I send the 3pm note to the function, I want to get the value of the 1pm note. Is there a way to annotate the two fields into one for comparison, or do I have to perform a raw SQL query? Is there a way to convert the date and time component into one, similar to how you could use Concat for strings?
Note.objects.annotate(
my_dt=Concat('note_date', 'note_time')
).filter(
my_dt__lt=Concat(models.F('note_date'), models.F('note_time')
).first()
I am too late but here is what I did
from django.db.models import DateTimeField, ExpressionWrapper, F
notes = Note.objects.annotate(my_dt=ExpressionWrapper(F('note_date') + F('note_time'), output_field=DateTimeField()))
Now we have added a new field my_dt of datetime type and can add a filter further to do operations
Found an answer using models.Q here: filter combined date and time in django
Note.objects.filter(
models.Q(note_date__lt=n.note_date) | models.Q(
note_date=n.note_date,
note_time__lt=n.note_time
)
).first()
I guess I just wasn't searching by the right criteria.
Here is another Approach which is more authentic
from django.db.models import Value, DateTimeField
from django.db.models.functions import Cast, Concat
notes = Note.objects.annotate(my_dt=Cast(
Concat('note_date', Value(" "), 'note_time', output_field=DateTimeField()),
output_field=DateTimeField()
).filter(my_dt__lte=datetime.now())
Here is another solution following others.
def get_queryset(self):
from django.db import models
datetime_wrapper = models.ExpressionWrapper(models.F('note_date') + models.F('note_time'), output_field=models.DateTimeField())
return Note.objects.annotate(
note_datetime=datetime_wrapper
).filter(note_datetime__gt=timezone.now()).order_by('note_datetime')

Django ORM: Conditional filtering

In a Django app, I keep daily scores of users in such a model:
class Score(models.Model):
user = models.ForeignKey(User)
score = models.PositiveIntegerField(default=0)
date = models.DateField(auto_now_add=True)
I want to find out the days when a user's score has changed drastically compared to a consequent day. That is, if for example, the user scores 10 times higher than the previous day.
How can I include such a condition in a query filter using Django ORM? Is it possible with a single query using conditional expressions as described here: https://docs.djangoproject.com/en/1.9/ref/models/conditional-expressions/
Thanks.
If you change your Score class slightly to include the previous day's score (which is admittedly pretty wasteful), you can pack the query into one line using F expressions.
Your new class:
class Score(models.Model):
user = models.ForeignKey(User)
score = models.PositiveIntegerField(default=0)
lastscore = models.PositiveIntegerField(default=0)
date = models.DateField(auto_now_add=True)
Then the filter becomes:
from django.db.models import F
daily_chg = 10
big_changes = Score.objects.filter(score__gt=daily_chg*F('lastscore'))
Instead of using timedeltas to search for and set the previous day's score field, I'd look into establishing an ordering via a Meta class and calling latest() when saving the current day's score.
Using timedelta we can test for the last week's days for a given user as such:
from my_app.models import Score
import datetime
def some_view(request):
days_with_score_boost = []
today = datetime.date.today()
for each_day in xrange(0,7):
day_to_test, day_before_to_test = today - datetime.timedelta(days=each_day), today - datetime.timedelta(days=each_day + 1)
day_before_score = Score.objects.get(user=request.user,date=today - datetime.timedelta(days=each_day)).score # will need to catch the exception raised if .get return None and process appropriately
if Score.objects.filter(user=request.user,score__gte=days_before_score * 10,date=day_before_to_test).count() > 0:
days_with_score_boost.append(day_to_test)
days_with_score_boost will be a list of datetime.date objects where the score increased by 10 or more from the day before.
In reponse to your comment, I'd make a measure that checks at save time whether the score boost has occured. However, I would get rid of auto_now_add in favor of writing in the save method.
from django.utils import timezone
from django.core.exceptions import ObjectDoesNotExist
class Score(models.Model):
user = models.ForeignKey(User)
score = models.PositiveIntegerField(default=0)
date = models.DateField(null=True,blank=True)
increased_ten_over_previous_day = models.BooleanField(null=True,blank=True)
def save(self, *args, **kwargs):
self.date = timezone.now().today()
try:
yesterday_score = Score.objects.get(date=self.date-timezone.timedelta(days=1)).score
self.increased_ten_over_previous_day = (yesterday_score * 10) <= self.score
except ObjectDoesNotExist: # called if Score.object.get returns no object; requires you only have one score per user per date
self.increased_ten_over_previous_day = False
super(self, Score).save(*args, **kwargs)
Then you could filter objects for a date_range where increased_ten_over_previous_day is True.

Subtract django.db.models.DateField from python's datetime.date to get age

In django, I want to get the age (in days) of an instance of a class. I tried doing that by subtracting its creation date field from today, but it does not seem to work properly. date.today() works fine, but DateField is giving me trouble. I looked at its source code and the django docs online for my version but I'm not sure how to manipulate it to perform the subtraction.
import datetime.date
from django.db import models
class MyItem(models.Model):
item_name = models.CharField(max_length = 30)
creation_date = models.DateField()
def age(self):
return date.today() - creation_date
my_first_item = MyItem(item_name = 'First', creation_date = '2005-11-01')
print my_first_item.age.days
Any input would be greatly appreciated!
Your problem is that you are trying to use a field instance outside of a model to represent a value.
models.DateField is a class which represents a database field with a type of "date". I suspect that you are looking to do one of the following:
Just do straight date math
Work with a value returned by a model
In the case of 1, you don't want to use Django's models at all. All you need and want is python's date and time handling classes. For your specific example all you need to use is a pair of date objects and you will end up with a timedelta object.
To do what you were trying to do in your example with the python standard classes, see the example below:
from datetime import date
birthday = date(year=2005, month=11, day=1)
today = date.today()
age = today - birthday
print age.days()
Here we instantiate a date with the birthdate values, we get a date with today's values, subtract them to get a timedelta, and finally print the number of days between the two dates.
In the case of 2, let's look at an example model:
class Person(models.Model):
name = models.CharField(max_length=100)
birthday = models.DateField()
Here we have a model where we've used models.CharField and models.DateField to describe a table in the database which contains a "varchar" column and a "date" column. When we fetch instances of this model using the ORM, Django handles converting whatever value the database returns to a native datatype. Now let's look at some code that figures out the age of an instance of a person:
from datetime import date
from myapp.models import Person
person = Person.objects.get(id=1)
age = date.today() - person.birthday
print age.days
Here you can see that we fetch an instance of the person model from the database and then we subtract their birthday from today. We're able to do this here, because when we access "person.birthday" Django is transforming whatever value the database returned into a python date object. This is the same type as the date object returned by "date.today()" so the "-" operator makes sense. The result of the subtraction operation is a timedelta object.

Custom model method in Django

I'm developing a web app in Django that manages chores on a reoccurring weekly basis. These are the models I've come up with so far. Chores need to be able to be assigned multiple weekdays and times. So the chore of laundry could be Sunday # 8:00 am and Wednesday # 5:30 pm. I first want to confirm the models below are the best way to represent this. Secondly, I'm a little confused about model relationships and custom model methods. Since these chores are on a reoccurring basis, I need to be able to check if there has been a CompletedEvent in this week. Since this is row level functionality, that would be a model method correct? Based on the models below, how would I check for this? It has me scratching my head.
models.py:
from django.db import models
from datetime import date
class ChoreManager(models.Manager):
def by_day(self, day_name):
return self.filter(scheduledday__day_name = day_name)
def today(self):
todays_day_name = date.today().strftime('%A')
return self.filter(scheduledday__day_name = todays_day_name)
class Chore(models.Model):
objects = ChoreManager()
name = models.CharField(max_length=50)
notes = models.TextField()
class Meta:
ordering = ['scheduledday__time']
class ScheduledDay(models.Model):
day_name = models.CharField(max_length=8)
time = models.TimeField()
chore = models.ForeignKey('Chore')
class CompletedEvent(models.Model):
date_completed = DateTimeField(auto_now_add=True)
chore = models.ForeignKey('Chore')
Then all you need to do is:
monday_of_week = some_date - datetime.timedetla(days=some_date.weekday())
end_of_week = date + datetime.timedelta(days=7)
chore = Chore.objects.get(name='The chore your looking for')
ScheduledDay.objects.filter(completed_date__gte=monday_of_week,
completed_date__lt=end_of_week,
chore=chore)
A neater (and faster) option is to use Bitmasks!
Think of the days of the week you want a chore to be repeated on as a binary number—a bit for each day. For example, if you wanted a chore repeated every Tuesday, Friday and Sunday then you would get the binary number 1010010 (or 82 in decimal):
S S F T W T M
1 0 1 0 0 1 0 = 1010010
Days are reversed for sake of illustration
And to check if a chore should be done today, simply get the number of that day and do an &:
from datetime import datetime as dt
if dt.today().weekday() & 0b1010100:
print("Do chores!")
Models
Your models.py would look a bit like this:
from django.contrib.auth.models import User
from django.db import models
from django.utils.functional import cached_property
class Chore(models.Model):
name = models.CharField(max_length=128)
notes = models.TextField()
class ChoreUser(models.Model):
chore_detail = models.ForeignKey('ChoreDetail')
user = models.ForeignKey('ChoreDetail')
completed_time = models.DateTimeField(null=True, blank=True)
class ChoreDetail(models.Model):
chore = models.ForeignKey('Chore')
chore_users = models.ManyToManyField('User', through=ChoreUser)
time = models.DateTimeField()
date_begin = models.DateField()
date_end = models.DateField()
schedule = models.IntegerField(help_text="Bitmask of Weekdays")
#cached_property
def happens_today(self):
return bool(dt.today().weekday() & self.weekly_schedule)
This schema has a M2M relationship between a User and a Chore's Schedule. So you can extend your idea, like record the duration of the chore (if you want to), or even have many users participating in the same chore.
And to answer your question, if you'd like to get the list of completed events this week, you could could put this in a Model Manager for ChoreUser:
from datetime import datetime as dt, timedelta
week_start = dt.today() - timedelta(days=dt.weekday())
week_end = week_start + timedelta(days=6)
chore_users = ChoreUser.objects.filter(completed_time__range=(week_start, week_end))
Now you have all the information you need in a single DB call:
user = chore_users[0].user
time = chore_users[0].chore_detail.time
name = chore_users[0].chore_detail.chore.name
happens_today = chore_users[0].chore_detail.happens_today
You could also get all the completed chores for a user easily:
some_user.choreuser_set.filter(completed_time__range=(week_start, week_end))