jlab-method-pull 1.1.0

Pull class methods into Jupyter cells and inject edited versions back — with optional persistence to source files

jlab_method_pull

Pull class methods into editable JupyterLab cells and inject modified versions back — optionally rewriting the source file for persistence.

Install

pip install jlab_method_pull

To make pullMethodCode and injectMethod available automatically in every JupyterLab kernel, run once:

from jlab_method_pull import install
install()

Usage

Pull a method into a new cell

from jlab_method_pull import pullMethodCode, injectMethod
from mymodule import MyClass

pullMethodCode(MyClass.some_method)

A new cell appears below, pre-filled with the method's source, all imports from the source file, and a ready-to-run injectMethod call:

import numpy as np
from mymodule import MyClass, OtherClass

def some_method(self, x):
    ...

injectMethod(some_method, MyClass, persistent=False)

Inject a method back

# In-memory only (survives the session, not reimports):
injectMethod(some_method, MyClass, persistent=False)

# Persistent (also rewrites mymodule.py on disk):
injectMethod(some_method, MyClass, persistent=True)

persistent=True uses AST to locate the exact lines of the old method in the source file and replaces them. It works correctly even if the method was previously monkey-patched in memory.

How it compares

Tool	Cell injection	Runtime patch	Rewrites source file
IPython %load	file-level only	—	—
gorilla	—	yes	—
jlab_method_pull	method-level	yes	yes

Requirements

• Python ≥ 3.9
• IPython (only required for the Jupyter cell injection and install())