markdown-parser

What is it?

This is an auxiliary Python script for parsing and checking text files. The main problem it solves is verifying Markdown links between text files. For example, the Docusaurus, a static site generator that we use for documentation, recognizes the link to a non-existing file like [a link to a file](non-existing-path-to-file). But it cannot recognize a link to a non-existing heading inside a file: [link to a paragraph](existing-file#non-existing-heading). So, when we change a heading we might break a link to it in other files. This script is intended to catch such situations.

Configuration

The script is configurable. You can define configuration parameters either in a config.ini file or pass them as arguments when running the script.

Config file

The configuration must have a section [MAIN] and the following parameters:

root is the root directory where the files are scanned. Note that the script works with all files of the given extension in the root recursively. The root can be an absolute path or a relative path.
file_extension is the extension of files that are being parsed. Accepts only one value with a leading dot, for example .md. You can use any text extension that can be opened with the default Python open function.
action is the intended action of the script. Accepts one of the following values:
- check_links parses all files with a given extension in a given root and checks all Markdown links. If there are broken links, it outputs them in a stream specified in output parameter.

search asks for a non-empty string to search, parses all files with the given extension in the given root, and searches the string. Then outputs the result in a stream specified in output parameter.

If config.ini misses [MAIN] section, the default configuration is used. If config.ini misses one or more obligatory parameters in [MAIN] section, the default values are used. If a path specified in root parameter is not found or not a directory or empty string, the default value is used. If action has a value that is not accepted, the default value is used. All other sections and parameters are ignored.

Default configuration:

[MAIN]
root: .
file_extension: .mdx
action: check_links

Arguments

Note if you pass configuration arguments, the config file is ignored.

All arguments are optional, if you don't pass an argument the default value will be used. You can use the following configuration arguments in command line:

-r or --root to define the root directory
-f or --file to define the file extension

-a or --action to define the action. The argument accepts check_links or search values only

Run python3 main.py -h to see the help in console.

Example of using arguments:

python3 main.py -r ../../docs -f .md -a print_links

Script files and directories

TBD.

Using as Git action

You can run the script as a Github action, for example, when someone creates a new PR.

Create a yaml file in .github/workflows directory.
Copy the following snippet into yaml file:

name: Check Markdown links

on: 
  pull_request:    
    branches: [ "main", "test", "develop" ]

  workflow_dispatch:

jobs:
  links-checker:
    name: links-checker
    runs-on: ubuntu-latest
    steps:
    - name: Normal checkout
      uses: actions/checkout@v3
      
    - name: Check-out parser repository
      uses: actions/checkout@v2
      with:
        repository: paveltovchigrechko/markdown-parser
        path: "markdown-parser"
        
    - name: Setup Python
      uses: actions/setup-python@v4
      with:
        python-version: 3.11

    - name: Run parser
      run: python3 markdown-parser/main.py