Python Inheritance with Datatables

---

Review of concepts

To review key concepts up until this point...

• we setup our database and added calcution results to it.
• we learned about python classes, and in particular, the importance of the Structure class.
• we learned how to explore the documentation and use new classes.

Now, we want to bring these ideas together in order to expolore our database.

---

An example table in python

Let's start simple... All datatables are represented by a class, where the general format looks like this:

from simmate.database.base_data_types import DatabaseTable, table_column

class MyExampleTable(DatabaseTable):
   column_01 = table_columns.CharField()  # CharField --> means we store text
   column_02 = table_columns.BoolField()  # BoolField --> means we store True/False
   column_03 = table_columns.FloatField()  # FloatField --> means we store a number/decimal

And the corresponding table (with random data added) would look like...

column_01	column_02	column_03
jack	True	3.1456
lauren	False	299792458
siona	True	1.6180
scott	False	1.602e-19
...	...	...

That's how all tables are made! We just make a class, say it is a DatabaseTable and then list off our desired columns.

---

Building tables with inheritance

However, this could get really repetitive if we have a bunch of tables that contain similar information. For example, we may want to store structures in many different tables -- each one with columns like density, number of sites, number of elements, etc.. To save time, we use what is known as python "inheritance". Here's how it works:

First, we define a table with common information (let's say a Person).

from simmate.database.base_data_types import DatabaseTable, table_column

class Person(DatabaseTable):
   name = table_columns.CharField()
   age = table_columns.IntField()
   height = table_columns.FloatField()

Next, we want a separate table to contain this type of information and more:

class Student(Person):  # <--- note we have Person here instead of DatabaseTable
   year = table_columns.IntField()  # e.g. class of 2020
   gpa = table_columns.FloatField()

The Student datatable now looks like this:

name	age	height	year	gpa
jack	15	6.1	2020	3.6
lauren	16	5.8	2019	4.0
siona	15	5.6	2020	3.7
scott	14	6.2	2021	3.2
...	...	...	...	...

Simmate uses this idea with common materials science data -- such as structures, thermodynamic data, site forces, and more. You'll find our fundamental building blocks for tables in the simmate.database.base_data_types module (covered here).

All of our datatables start from these classes and build up. Up next, we'll look at an actual database table and learn how to use it to view data.

---