get_code_from_markdown

Extract code from within triple-quotes (```) in markdown files (code blocks).

Usage

Getting code blocks from markdown files

The language parameter defaults to python. This will find codeblocks that start with ```python

from get_code_from_markdown import *

blocks = get_code_from_markdown(["# get_code_from_markdown", "```python", "print('hello world')", "```"])
# or
blocks = get_code_from_markdown_filename("README.md", language="python")
# or
with open("README.md", "r") as f:
    blocks = get_code_from_markdown_file(f, language="python")

Running code blocks from markdown files

You can use a subprocess or just exec the code to see if it works.

from get_code_from_markdown import run_code_from_markdown_blocks
blocks = ["print('hello world')"]

# run function uses exec() by default
run_code_from_markdown_blocks(blocks)

# you can choose to use subprocess output instead
# this will use python3 by default
run_code_from_markdown_blocks(blocks, method=RunMethods.SUBPROCESS)

# give arguments to specify a different command
run_code_from_markdown_blocks(blocks, method=RunMethods.SUBPROCESS(["python3", "-c"]))

Format of outputted code blocks

The format is a list of \n-delimited strings, each representing a block of code from the parsed markdown file. You can split the code blocks further into individual lines of code with string.split("\n")

Development

• Use black (code linter)
• Run pytest (unit tests)
• Build a package for PyPI: python -m build --sdist --wheel

Legal

This is a derivative work of Codeblocks which provides a simpler version with less dependencies. It is covered by the same Apache license (see LICENSE file in this git repo). Use Codeblocks for its CLI and automatic formatting features.