I'm developing a calendaring application in Django.
The relevant model structure is as follows:
class Lesson(models.Model):
RECURRENCE_CHOICES = (
(0, 'None'),
(1, 'Daily'),
(7, 'Weekly'),
(14, 'Biweekly')
)
frequency = models.IntegerField(choices=RECURRENCE_CHOICES)
lessonTime = models.TimeField('Lesson Time')
startDate = models.DateField('Start Date')
endDate = models.DateField('End Date')
student = models.ForeignKey(Student)
class CancelledLesson(models.Model):
lesson = models.ForeignKey(Lesson)
student = models.ForeignKey(Student)
cancelledLessonDate = models.DateField() # Actual date lesson has been cancelled, this is startDate + Frequency
class PaidLesson(models.Model):
lesson = models.ForeignKey(Lesson)
student = models.ForeignKey(Student)
actualDate = models.DateField() # Actual date lesson took place
paidAmt = models.DecimalField('Amount Paid', max_digits=5, decimal_places=2)
paidDate = models.DateField('date paid')
class CompositeLesson(models.Model):
# only used to aggregate lessons for individual lesson management
lesson = models.ForeignKey(Lesson)
student = models.ForeignKey(Student)
actualDate = models.DateTimeField()
isCancelled = models.BooleanField()
canLesson = models.ForeignKey(CancelledLesson, blank=True, null=True)
payLesson = models.ForeignKey(PaidLesson, blank=True, null=True)
Apparently this is all causing issues with displaying the lessons that belong to a particular student. What I am attempting to do is display a table that shows the Student name plus all instances of scheduled lessons. I am calculating the recurrence dynamically to avoid blowing up my database. Exceptions to the recurrences (i.e. lesson cancellations) are stored in their own tables. Recurrences are checked against the cancelled lesson table when the recurrences are generated.
See my code to generate recurrences (as well as a small catalog of what issues this is causing) here: Can't get key to display in Django template
I'm relatively inexperienced with Python, and am using this project as a way to get my head around a lot of the concepts, so if I'm missing something that's inherently "Pythonic", I apologize.
The key part of your problem is that you're using a handful of models to track just one concept, so you're introducing a lot of duplication and complexity. Each of the additional models is a "type" of Lesson, so you should be using inheritance here. Additionally, most of the additional models are merely tracking a particular characteristic of a Lesson, and as a result should not actually be models themselves. This is how I would have set it up:
class Lesson(models.Model):
RECURRENCE_CHOICES = (
(0, 'None'),
(1, 'Daily'),
(7, 'Weekly'),
(14, 'Biweekly')
)
student = models.ForeignKey(Student)
frequency = models.IntegerField(choices=RECURRENCE_CHOICES)
lessonTime = models.TimeField('Lesson Time')
startDate = models.DateField('Start Date')
endDate = models.DateField('End Date')
cancelledDate = models.DateField('Cancelled Date', blank=True, null=True)
paidAmt = models.DecimalField('Amount Paid', max_digits=5, decimal_places=2, blank=True, null=True)
paidDate = models.DateField('Date Paid', blank=True, null=True)
class CancelledLessonManager(models.Manager):
def get_query_set(self):
return self.filter(cancelledDate__isnull=False)
class CancelledLesson(Lesson):
class Meta:
proxy = True
objects = CancelledLessonManager()
class PaidLessonManager(models.Manager):
def get_query_set(self):
return self.filter(paidDate__isnull=False)
class PaidLesson(Lesson):
class Meta:
proxy = True
objects = PaidLessonManager()
You'll notice that I moved all the attributes onto Lesson. This is the way it should be. For example, Lesson has a cancelledDate field. If that field is NULL then it's not cancelled. If it's an actual date, then it is cancelled. There's no need for another model.
However, I have left both CancelledLesson and PaidLesson for instructive purposes. These are now what's called in Django "proxy models". They don't get their own database table (so no nasty data duplication). They're purely for convenience. Each has a custom manager to return the appropriate matching Lessons, so you can do CancelledLesson.objects.all() and get only those Lessons that are cancelled, for example. You can also use proxy models to create unique views in the admin. If you wanted to have an administration area only for CancelledLessons you can, while all the data still goes into the one table for Lesson.
CompositeLesson is gone, and good riddance. This was a product of trying to compose these three other models into one cohesive thing. That's no longer necessary, and your queries will be dramatically easier as a result.
EDIT
I neglected to mention that you can and should add utility methods to the Lesson model. For example, while tracking cancelled/not by whether the field is NULL or not makes sense from a database perspective, from programming perspective it's not as intuitive as it could be. As a result, you might want to do things like:
#property
def is_cancelled(self):
return self.cancelledDate is not None
...
if lesson.is_cancelled:
print 'This lesson is cancelled'
Or:
import datetime
...
def cancel(self, date=None, commit=True):
self.cancelledDate = date or datetime.date.today()
if commit:
self.save()
Then, you can cancel a lesson simply by calling lesson.cancel(), and it will default to cancelling it today. If you want to future cancel it, you can pass a date: lesson.cancel(date=tommorrow) (where tomorrow is a datetime). If you want to do other processing before saving, pass commit=False, and it won't actually save the object to the database yet. Then, call lesson.save() when you're ready.
This is what I came up with. I feel that lessons_in_range() may not be as "Pythonic" as I could get it, but this does what I need it to do.
class Lesson(models.Model):
RECURRENCE_CHOICES = (
(0, 'None'),
(1, 'Daily'),
(7, 'Weekly'),
(14, 'Biweekly')
)
relatedLesson = models.ForeignKey('self', null=True, blank=True)
student = models.ForeignKey(Student)
frequency = models.IntegerField(choices=RECURRENCE_CHOICES, null=True, blank=True)
lessonTime = models.TimeField('Lesson Time', null=True, blank=True)
startDate = models.DateField('Start Date')
endDate = models.DateField('End Date', null=True, blank=True)
isCancelled = models.BooleanField(default = False)
amtBilled = models.DecimalField(max_digits=5, decimal_places=2, null=True, blank=True)
amtPaid = models.DecimalField(max_digits=5, decimal_places=2, null=True, blank=True)
def get_exceptions(self):
return Lesson.objects.filter(relatedLesson = self.id)
def cancel(self, date=None):
if date:
x = Lesson()
x = self
x.pk = None
x.relatedLesson = self.id
x.isCancelled = True
x.startDate = date
x.endDate = date
x.save()
else:
self.endDate = datetime.date.today()
self.save()
return
def pay_lesson(self, date, amount):
x = Lesson()
x = self
x.pk = None
x.relatedLesson = self.id
x.amtPaid = amount
x.startDate = date
x.endDate = date
x.save()
return
def lessons_in_range(self, startDate, endDate):
if (self.startDate > endDate) or (self.endDate < startDate):
return None
if self.endDate < endDate:
endDate = self.endDate
ex = self.get_exceptions()
if self.frequency == 0:
if ex:
return ex
else:
return self
sd = next_date(self.startDate, self.frequency, startDate)
lessonList = []
while (sd <= endDate):
exf = ex.filter(startDate = sd)
if exf:
# lesson already exists in database, add it
lessonList.append(exf)
elif sd == self.startDate:
# lesson is the original lesson, add that
lessonList.append(self)
else:
# lesson does not exist, create it in the database then add it to the list
x = Lesson()
x.student = self.student
x.frequency = 0
x.lessonTime = self.lessonTime
x.relatedLesson = self
x.startDate = sd
x.endDate = sd
x.isCancelled = False
x.amtBilled = self.amtBilled
x.amtPaid = None
x.save()
lessonList.append(x)
sd += timedelta(self.frequency)
return lessonList
Related
I've got a Stock table and a StockArchive table.
My Stock table consists of roughly that 10000 stocks that I update daily. The reason I have a StockArchive table is because I still wanna some historic data and not just update existing records. My question is, is this a proper way of doing it?
First, my models:
class Stock(models.Model):
objects = BulkUpdateOrCreateQuerySet.as_manager()
stock = models.CharField(max_length=200)
ticker = models.CharField(max_length=200)
exchange = models.ForeignKey(Exchange, on_delete=models.DO_NOTHING)
eod_price = models.DecimalField(max_digits=12, decimal_places=4)
currency = models.CharField(max_length=20, blank=True, null=True)
last_modified = models.DateTimeField(blank=True, null=True)
class Meta:
db_table = "stock"
class StockArchive(models.Model):
objects = BulkUpdateOrCreateQuerySet.as_manager()
stock = models.ForeignKey(Stock, on_delete=models.DO_NOTHING)
eod_price = models.DecimalField(max_digits=12, decimal_places=4)
archive_date = models.DateField()
class Meta:
db_table = "stock_archive"
I proceed on doing the following:
#transaction.atomic
def my_func():
archive_stocks = []
batch_size = 100
old_stocks = Stock.objects.all()
for stock in old_stocks:
archive_stocks.append(
StockArchive(
stock=stock.stock,
eod_price = stock.eod_price,
archive_date = date.today(),
)
)
# insert into stock archive table
StockArchive.objects.bulk_create(archive_stocks, batch_size)
# delete stock table
Stock.objects.all().delete()
# proceed to bulk_insert new stocks
I also wrapped the function with a #transaction.atomic to make sure that everything is committed and not just one of the transactions.
Is my thought process correct, or should I do something differently? Perhaps more efficient?
I searched for a similar case on SO and Google with no luck.
SHORT EXPLANATION
I have transactions that belong to an account, and an account belongs to an account aggrupation.
I want to get a list of accounts aggrupations, with their accounts, and I want to know the total balance of each account (an account balance is calculated by adding all its transactions amount).
LONG EXPLANATION
I have the following models (I include mixins for the sake of completeness):
class UniqueNameMixin(models.Model):
class Meta:
abstract = True
name = models.CharField(verbose_name=_('name'), max_length=100, unique=True)
def __str__(self):
return self.name
class PercentageMixin(UniqueNameMixin):
class Meta:
abstract = True
_validators = [MinValueValidator(0), MaxValueValidator(100)]
current_percentage = models.DecimalField(max_digits=5,
decimal_places=2,
validators=_validators,
null=True,
blank=True)
ideal_percentage = models.DecimalField(max_digits=5,
decimal_places=2,
validators=_validators,
null=True,
blank=True)
class AccountsAggrupation(PercentageMixin):
pass
class Account(PercentageMixin):
aggrupation = models.ForeignKey(AccountsAggrupation, models.PROTECT)
class Transaction(models.Model):
date = models.DateField()
concept = models.ForeignKey(Concept, models.PROTECT, blank=True, null=True)
amount = models.DecimalField(max_digits=10, decimal_places=2)
account = models.ForeignKey(Account, models.PROTECT)
detail = models.CharField(max_length=100, blank=True, null=True)
def __str__(self):
return '{} - {} - {} - {}'.format(self.date, self.concept, self.amount, self.account)
I want to be able to do this in Django ORM:
select ca.*, ca2.*, sum(ct.amount)
from core_accountsaggrupation ca
join core_account ca2 on ca2.aggrupation_id = ca.id
join core_transaction ct on ct.account_id = ca2.id
group by ca2.name
order by ca.name;
It would appear that nesting navigation through sets is not possible:
Wrong: AccountsAggrupation.objects.prefetch_related('account_set__transaction_set')
(or any similar approach). The way to work with this is the way around: go from transaction to account and then to account_aggroupation.
But, as I needed to have a dict with account_aggroupation, pointing each key to its set of accounts (and the balance for each), I ended up doing this:
def get_accounts_aggrupations_data(self):
accounts_aggrupations_data = {}
accounts_balances = Account.objects.annotate(balance=Sum('transaction__amount'))
for aggrupation in self.queryset:
aggrupations_accounts = accounts_balances.filter(aggrupation__id=aggrupation.id)
aggrupation.balance = aggrupations_accounts.aggregate(Sum('balance'))['balance__sum']
accounts_aggrupations_data[aggrupation] = aggrupations_accounts
current_month = datetime.today().replace(day=1).date()
date = current_month.strftime('%B %Y')
total_balance = Transaction.objects.aggregate(Sum('amount'))['amount__sum']
return {'balances': accounts_aggrupations_data, 'date': date, 'total_balance': total_balance}
Note that since I'm iterating through the accounts_aggrupations, that query (self.queryset, which leads to AccountsAggrupation.objects.all()) is executed to the DB.
The rest of the queries I do, do not execute yet because I'm not iterating through them (until consuming the info at the template).
Also note that the dictionary accounts_aggrupations_data has an accounts_aggrupation object as key.
I am new to programming. I am trying to create a project management site. I created a model as follows:
class Boqmodel(models.Model):
code = models.IntegerField(unique=True)
building = models.ForeignKey(building, on_delete=models.SET_NULL, null=True)
level = models.ForeignKey(level, on_delete=models.SET_NULL, null=True)
activity = models.ForeignKey(activity, on_delete=models.SET_NULL, null=True)
subactivity = models.ForeignKey(sub_activity, on_delete=models.SET_NULL, null=True)
duration = models.IntegerField()
linkactivity = models.IntegerFeild(null=True) #contains code (same as code field) which this specific code is linked to
linktype = models.CharField(max_length=300, null=True)# only two choices start or finish
linkduration = models.IntegerField(null=True)
plannedstart = models.DateField()
plannedfinish = models.DateField()
The problem is I need my planned start as a computed column. The planned column should be as follows:
if linkactivity is null, then it should take a default value 01-01-2019
if else then it should look the linkactivity in code field and then
if linktype is start, then it should specify the start date of code activity +duration
or if linktype is finish plannedfinish + duration
Example
First entry:
code=1
building=A-1
Level=L-1
Activity=Activity-1
Subactivity-Subactivity-1
duration=2
linkactivity=null
linktype=null
linkduration=null
planned start=01-01-2019(as linkactivity=null)
plannedfinish=03-01-2019(planned start+duration)
Second entry:
code=2
building=A-1
Level=L-1
Activity=Activity-2
Subactivity-Subactivity-2
duration=3
linkactivity=1
linktype=start
linkduration=1
planned start=02-01-2019(as linkactivity=1,it searches code1 ,as linktype=start,it searches startdate of code 1 it is 01-01-2019 ; finally 01-01-2019+link duration(1)=02-01-2019)
plannedfinish=05-01-2019(planned start+duration)
Any help would be really appreciated
When you call Model.objects.create() method, the django ORM implicitly calls the Model.save() method inside your model. Since, you are inheriting BoqModel from models.Model and there is no explicit BoqModel.save() method in your BoqModel, Model.objects.create() method calls the save() method from the parent class. Inside Model.save() method you can customize your model save behavior. In order to achieve your intended action, override the save() method as follows -
class Boqmodel(models.Model):
code = models.IntegerField()
building = models.ForeignKey(building, on_delete=models.SET_NULL, null=True)
level = models.ForeignKey(level, on_delete=models.SET_NULL, null=True)
activity = models.ForeignKey(activity, on_delete=models.SET_NULL, null=True)
subactivity = models.ForeignKey(sub_activity, on_delete=models.SET_NULL, null=True)
duration = models.IntegerField()
linkactivity = models.IntegerFeild(null=True) #contains code (same as code field) which this specific code is linked to
linktype = models.CharField(max_length=300, null=True)# only two choices start or finish
linkduration = models.IntegerField(null=True)
plannedstart = models.DateField()
plannedfinish = models.DateField()
def save(self, *args, **kwargs):
if self.linkactivity is None:
self.plannedstart = datetime.date(year=2019, month=1, day=1)
else:
boq = BoqModel.objects.get(code=self.linkactivity)
if self.linktype is not None:
if self.linktype == "start":
self.plannedstart = boq.plannedstart + datetime.timedelta(days=self.linkduration)
elif self.linktype == "finish":
self.plannedstart = boq.plannedfinish + datetime.timdelta(days=self.linkduration)
else:
self.plannedstart = datetime.date(year=2019, month=1, day=1)
else:
self.plannedstart = datetime.date(year=2019, month=1, day=1)
self.plannedfinish = self.plannedstart + datetime.tiemdelta(days=self.linkduration)
super(BoqModel, self).save(*args, **kwargs)
The above is not optimized. But you get the general idea of how to do it. Also learn more about overriding predefined model methdos here.
absolute n00b here, just fiddling around. Trying to make a very simple app to track my personal expenses. I have a class for entering the expenses, a class for the categories and a class for my account balance. Plan is to create en entry in the account balance everytime I create an expense.
How do I update my account balance? I'll have to get fields from the latest entry in expenses to do the math with in my balance class, right?
This is what I have. Any help appreciated.
from django.db import models
from django.utils import timezone
class Category(models.Model):
category = models.CharField(max_length=200,blank=False)
def __str__(self):
return self.category
class Balance(models.Model):
date = models.DateTimeField(default=timezone.now)
previous_balance = ????
transaction = ????
current_balance = previous_balance - transaction
class Expense(models.Model):
date = models.DateTimeField(default=timezone.now)
spender = models.ForeignKey('auth.User')
description = models.CharField(max_length=200)
category = models.ForeignKey(Category,default=1)
ABN = 'ABN'
ING = 'ING'
ACCOUNT_CHOICES = (
(ABN, 'ABN'),
(ING, 'ING'),
)
account = models.CharField(
max_length=30,
choices=ACCOUNT_CHOICES,
default=ABN,
)
amount = models.DecimalField(max_digits=10, decimal_places=2)
def commit(self):
self.commit_date = timezone.now()
self.save()
def __str__(self):
return u"%s. Kosten: %s" % (self.description, self.amount)
If I'm understanding your question correctly, you want to be able to get your current balance after creating Expenses. If so, you can use Django's aggregation:
from django.db.models import Sum
class Balance(models.Model):
date = models.DateTimeField(default=timezone.now)
# Keep the amount you start with
starting_balance = models.IntegerField()
# Get the Sum of all expenses and do some simple subtraction
def get_current_balance(self):
total_expenses = Expense.objects.all().aggregate(Sum('amount'))
return self.starting_balance - total_expenses['amount__sum']
Then in your views, you can do something like:
current_balance = some_balance_instance.get_current_balance()
considered balance change will be trigger by expense record change, you can overwrite save on Expense model. then balance table can be maintain in auto.
import datetime
class Expense(models.Model):
date = models.DateTimeField(default=timezone.now)
spender = models.ForeignKey('auth.User')
description = models.CharField(max_length=200)
category = models.ForeignKey(Category,default=1)
ABN = 'ABN'
ING = 'ING'
ACCOUNT_CHOICES = (
(ABN, 'ABN'),
(ING, 'ING'),
)
account = models.CharField(
max_length=30,
choices=ACCOUNT_CHOICES,
default=ABN,
)
amount = models.DecimalField(max_digits=10, decimal_places=2)
def save(self, *args, **kwargs):
super(Expense, self).save(*args, **kwargs)
last_bal = Balance.objects.order_by('id').last()
Balance.objects.create(date=datetime.datetime.now(), previouse_balance=last_bal.current_balance,
transaction=self, current_balance=last_bal.current_balance + self.amount)
I have a model called "Comments" which uses a generic foreign key:
class CommentManager(models.Manager):
def for_model(self, model):
"""
QuerySet for all comments for a particular model (either an instance or
a class).
"""
ct = ContentType.objects.get_for_model(model)
qs = self.get_query_set().filter(content_type=ct)
if isinstance(model, models.Model):
qs = qs.filter(object_pk=force_text(model._get_pk_val()))
return qs
class Comment(models.Model):
"""
A user comment about some object.
"""
status = models.CharField(max_length=12, blank=True, null=True)
sub_status = models.CharField(max_length=500, blank=True, null=True)
comment = models.TextField()
content_type = models.ForeignKey(
ContentType,
verbose_name=_('content type'),
related_name="content_type_set_for_%(class)s")
object_pk = models.TextField(_('object ID'))
content_object = generic.GenericForeignKey(ct_field="content_type",
fk_field="object_pk")
One of the things you can put comments on are Tickets:
class Ticket(CommonModel):
type = models.ForeignKey(TicketType)
priority = models.PositiveSmallIntegerField()
status = models.ForeignKey(TicketStatus)
access_serial_number = models.PositiveSmallIntegerField(null=True)
parent = models.ForeignKey('self', null=True, related_name='child')
submitted = models.DateTimeField()
I do a lot of filtering of Tickets - in the past all I did with Comments on Tickets is that when I displayed a Ticket, I used Comments.objects.for_model(ticket) to find all the Comments for it. But now what I want to do is find Tickets that have a specific text in the comment. There is no comment_set or equivalent with GenericForeignKey.
This is what I've come up with, but it's pretty horrible:
comment_ticket_ids = [int(c.object_pk) for c in Comment.objects.for_model(Ticket).filter(comment__icontains='error')]
tickets = Ticket.filter(status=open_status, id__in=comment_ticket_ids)
There must be a better way.
Perhaps you could add something using .extra() to help. Instead of
comment_ticket_ids = [int(c.object_pk) for c in Comment.objects.for_model(Ticket).filter(comment__icontains='error')]
tickets = Ticket.filter(status=open_status, id__in=comment_ticket_ids)
You could attach the field id_wanted:
extra_select = {
'id_wanted': 'CASE WHEN (SELECT COUNT(*) FROM tickets WHERE comments something) > 0 THEN TRUE ELSE FALSE END'
}
Then filter for tickets with the extra select:
tickets = Tickets.objects.filter(ticket__status=open_status, id_wanted__isnull=False).extra(extra_select)
It's not fully clear to me why you can't have a FKey relationship between these two models.