bettersql

A better SQL engine for DataFrames, Lists and Dictionaries!

Motivation

Pandasql is a great project but it lacked the ability to call Python UDF's or use SQL to update a DataFrame. Bettersql allows both of these features plus there is no need to define a lambda to pass globals() because I wrote the function to automatically get globals() from the caller level.

Installation

To install bettersql, simply:

$ pip install bettersql

Usage

A simple example:

from pandas import DataFrame
from bettersql import sqldf

def reverse(x):
    # sample function with one parameter
    if x is not None:
        return x[::-1]

# DataFrame source
names = DataFrame({'id':[1, 2, 3], 'name':['Alpha', 'Beta', 'Gamma'], 'category':[1, 2, 2]})    

# Dict of lists source
categories = {'id':[1, 2, 3], 'name':['One', 'Two', 'Three']}

sql = '''
SELECT n.id, n.name, n.category, c.name as categoryname, reverse(n.name) as reverse
FROM names AS n
LEFT JOIN categories AS c on n.category= c.id
'''

# The first reverse is the alias for the function in SQL and the second reverse is a reference to the Python UDF
r = sqldf(sql, reverse = reverse) 
print(r)