Pylog - Pythonic Prolog Features

1.1 · active · verified Mon Apr 13

Pylog is a lightweight Python library that implements core features of Prolog, enabling logic programming paradigms within Python. It allows defining facts and rules to perform simple logical deductions and goal-oriented queries. The current stable version is 1.1, with an infrequent release cadence focusing on stability and minor enhancements.

Warnings

breaking The project underwent significant restructuring around versions `1.0`/`1.1` (including the 'test1' release), which changed the import paths for core functions.
Fix: Update all import statements. Instead of `import pylog` or `from pylog import symbol`, use `from pylog.pylog import symbol` for functions like `facts`, `rules`, `goal`, and `run`.
gotcha Pylog heavily relies on global state for its `facts` and `rules` knowledge base. Calls to `facts()` and `rules()` modify a shared global state.
Fix: Be aware that the knowledge base is global. For applications requiring isolated logical contexts, manually manage state or consider external mechanisms, as `pylog` does not provide explicit APIs for context management or resetting the knowledge base directly.
gotcha Pylog implements a subset of Prolog's features. It may not support advanced Prolog constructs like complex data structures, explicit cut/fail predicates, or certain types of recursion with optimal performance.
Fix: Consult the library's GitHub repository for current capabilities and examples. Do not expect full Prolog compatibility; plan your logic programming tasks accordingly within the library's demonstrated scope.

Install

pip install pylog Install Pylog

Imports

facts, rules, goal, run
```
from pylog.pylog import facts, rules, goal, run
```
The project structure was reorganized, requiring imports from the `pylog.pylog` submodule rather than the top-level package.

Quickstart

This quickstart demonstrates how to define facts, create a simple rule using other predicates, set a goal, and retrieve solutions using the `run()` function. It showcases basic logical inference capabilities.

from pylog.pylog import facts, rules, goal, run

# Define facts
facts("human(socrates)")
facts("human(plato)")

# Define a rule: All humans are mortal
# The body is a list of predicates that must be true for the head to be true.
rules("mortal(X)", ["human(X)"])

# Define a goal to find all mortal beings
goal("mortal(X)")

# Run the inference engine to find solutions
solutions = run()

print("Solutions for mortal(X):")
for s in solutions:
    print(s)

# Example for a specific query
goal("mortal(socrates)")
solutions_socrates = run()
print("\nSolutions for mortal(socrates):")
for s in solutions_socrates:
    print(s)

view raw JSON →