Please note: this project is inactive since early 2006

9.1 Example: using FixedPoint for a price attribute

Sometimes you need more than this. Let's take the test package AuthorBooks, and suppose we want to use FixedPoint^9.1.

change the model so that price is a string/VARCHAR (was: a float/NUMERIC(10,2)): we will store the FixedPoint object as a string, not as a float, because of the inherent imprecision which goes any (binary) representation of float^9.2

add _setPrice() and _getPrice() to AuthorBooks.Book:

   PRECISION=2
   def _setPrice(self, value):
     if value is None: 
       self._price=None
     else:
       self._price  = FixedPoint(value, PRECISION)
     
   def _getPrice(self):
     if self._price:
       return None
     else:
       return str(self._price)

Now let's test this: (remember to change the DB schema so that table BOOK's attribute price is a VARCHAR)

>>> from fixedpoint import FixedPoint
>>> from AuthorBooks.Book import Book
>>> from Modeling.EditingContext import EditingContext
>>> ec=EditingContext()
>>> book=Book()
>>> book.setTitle('Test FixedPoint')
>>> book.setPrice(FixedPoint("3.341"))
>>> book.getTitle(), book.getPrice()
('Test FixedPoint', FixedPoint('3.34', 2)) # precision=2
>>> ec.insert(book)
>>> ec.saveChanges()
>>> book.getTitle(), book.getPrice()
('Test FixedPoint', FixedPoint('3.34', 2))

Here you can check in you db that it was stored as a varchar, as expected. Start a new python and test the fetch:

>>> from fixedpoint import FixedPoint
>>> from Modeling.EditingContext import EditingContext
>>> ec=EditingContext()
>>> books=ec.fetch('Book')
>>> books[0].getTitle(), books[0].getPrice()
('Test FixedPoint', FixedPoint('3.34', 2))

As you can see, FixedPoint is now correctly and automatically handled by the framework.

This technique can be used for any custom type you want to use. The next section gives some details on how this works.

Footnotes

...FixedPoint ^9.1: FixedPoint package can be found on sourceforge
... float ^9.2: try to type '0.7' in a python interpreter:
>>> 0.7
0.69999999999999996

Comments are welcome: Sebastien Bigaret / Modeling Home Page
Hosted by: