dbTable

dbTable is a innovative library built on the Python SQLite Library that allows you to manipulate views on tables in Databases, select data from them, and modify them in many ways. This library abstracts out all SQL queries with native Python syntax, and has built-in lazy computation as well. Furthermore, this library provides an interface to load tables from SQLite into in-memory table structures like Pandas DataFrames or Numpy Array for further analysis.

Check out the full documentation at DOCUMENTATION

Quickstart Guide:

1. Installation

Make sure you have numpy and tabulate installed

pip install numpy
pip install tabulate

Install dbTable through PyPI (This’ll automatically install numpy and tabulate)

pip install dbTable

You can also install the built distribution in the dist/ folder, or for portable cases, just download the dbTable.py file, and place it in your project folder

2. Load dbTable into your interpreter/project

import dbTable

3. Make a connection to a database

db = Database('test.db') # Replace with your database or use ":memory:" to load an in-memory database

4. Select a table:

table1 = db['table1']

5. Exporting / Importing

To convert a dbTable to any other form, use:

new_df = db['table1'].to_df() # Returns a Pandas DataFrame Object
new_table = db['table1'].to_table() # Returns a UCB Datascience Table
new_ndarray = db['table1'].to_numpy_array() # Returns a Numpy Array

To save a DataFrame, UCB Table, or Numpy Array, use:

db.store_table(new_df,'newtablename')

Here’s a sampling of modifications and filters you can do with dbTable

table2 = table1.select(['col1','col2']) # Selecting columns from database
table3 = table2.where(table2['col1']> 20) # Filters
table4 = table3.sort('col2',descending=False) # Sort data
grouped_table = table4.group(table4['col1'])
table4['col2'] = table4['col1']*10 # Modify columns
table4['col3'] = table4['col1']+table4['col2']
table4['col3'] *= 2

Have fun!

Check out the full documentation at DOCUMENTATION