Extracting the entire documentation to a MarkDown file is useful if you want to use AI LLMs (Large Language Models) such as or to help you write new documentation pages, as the generated MarkDown file can be used to provide the AI with context.
To extract documentation from our , run this Python script:
import os
import git
import glob
import shutil # Import shutil for removing directories
# Function to get the directory of the current script
def get_script_dir():
return os.path.dirname(os.path.realpath(__file__))
# Clone the repository
repo_url = 'https://gitlab.com/bloohq/documentation'
script_dir = get_script_dir() # Get the directory of the script
repo_dir = os.path.join(script_dir, 'documentation') # Directory to clone the repo
# Check if the script_dir is writable
if not os.access(script_dir, os.W_OK):
print(f"The directory {script_dir} is not writable. Please check your permissions.")
else:
if not os.path.exists(repo_dir):
git.Repo.clone_from(repo_url, repo_dir)
# Navigate through the cloned repository and find all .md files
md_files = glob.glob(repo_dir + '/**/*.md', recursive=True)
# Concatenate the contents of all .md files into one
combined_md_content = ''
for md_file in md_files:
with open(md_file, 'r', encoding='utf-8') as file:
combined_md_content += file.read() + '\n\n'
# Save the combined content into one .md file
combined_md_filename = os.path.join(script_dir, 'blue_docs.md')
with open(combined_md_filename, 'w', encoding='utf-8') as combined_file:
combined_file.write(combined_md_content)
print(f'All Markdown files have been combined into {combined_md_filename}')
# Delete the cloned directory and its contents
shutil.rmtree(repo_dir)
print(f'The directory {repo_dir} and its contents have been deleted.')
