Run In Notebook

Prerequisite

To get started with Logica quickly, ensure that Logica and Jupyter are installed on your system, or use Google Colab with Logica pre-installed. Besides, you should install the following Python packages to ensure proper functionality (Yes, this enable you to use DuckDB!!!):

pip install duckdb pandas graphviz

CAUTION

Without graphviz, you can still run Logica in a notebook, but the visualization of execution process will be limited to an ASCII version instead of the full graphical representation.

Package Import

from logica import colab_logica

Cell Magic

Logica uses Cell Magic in Notebooks to support its Datalog-like syntax. To enable this feature, start the cell with %%logica. The first line of the cell should specify the predicates to be printed as output.

For example, to write a "Hello, World!" program in a Notebook, use:

%%logica Greeting
@Engine("sqlite");
Greeting("Hello World!");

WARNING

After %%logica, specify the predicate to output, similar to Clingo's #show. Also, don't forget to specify the engine to use.

Result Interpretation

With this script, you will get the following:

From this, we know that the compiled query is stored in the Greeting_sql variable. The output of the Logica query is stored in a table variable named Greeting, which contains one row (0) and one column (col0) with the value Hello World!.

With a foundational understanding of using Logica in both the terminal and a notebook, let's explore the basics in greater detail.