|
| 1 | +import re |
| 2 | +from pathlib import Path |
| 3 | + |
| 4 | +# Mapping section titles to their corresponding filenames |
| 5 | +SECTION_MAPPING = { |
| 6 | + "": "home_page.md", |
| 7 | + "About the project": "about.md", |
| 8 | + "Installation": "installation.md", |
| 9 | + "Documentation": "docs.md", |
| 10 | + "License": "license.md", |
| 11 | + "Credits and acknowledgements": "credits.md", |
| 12 | + "Contact and feedback": "contact.md", |
| 13 | +} |
| 14 | + |
| 15 | + |
| 16 | +def extract_section(readme, section_title): |
| 17 | + """Extracts content between current section and next ## heading""" |
| 18 | + pattern = rf"## {re.escape(section_title)}(.*?)(?=\n## |\Z)" |
| 19 | + match = re.search(pattern, readme, flags=re.DOTALL) |
| 20 | + return match.group(1).strip() if match else "" |
| 21 | + |
| 22 | + |
| 23 | +def extract_links_from_readme(readme): |
| 24 | + """Extract link references from README.md into a dictionary""" |
| 25 | + link_pattern = r"\[([^\]]+)\]: (\S+)" |
| 26 | + links = {} |
| 27 | + |
| 28 | + matches = re.findall(link_pattern, readme) |
| 29 | + for ref, url in matches: |
| 30 | + links[ref] = url |
| 31 | + |
| 32 | + return links |
| 33 | + |
| 34 | + |
| 35 | +def convert_gfm_to_sphinx(content, links): |
| 36 | + """Convert GitHub Flavored Markdown to Sphinx-style syntax.""" |
| 37 | + # Convert GFM admonitions (like > [!IMPORTANT] and > [!NOTE]) |
| 38 | + content = re.sub( |
| 39 | + r"(^|\n)> \[!(\w+)\]([^\n]*)((?:\n> [^\n]*)*)", |
| 40 | + lambda m: f"\n:::{{{m.group(2)}}}\n" # Note the curly braces here |
| 41 | + + re.sub(r"^> ", "", m.group(4), flags=re.MULTILINE).strip() |
| 42 | + + "\n:::\n", |
| 43 | + content, |
| 44 | + ) |
| 45 | + |
| 46 | + # Replace link references dynamically using the links dictionary |
| 47 | + for ref, url in links.items(): |
| 48 | + content = re.sub(rf"\[{re.escape(ref)}\]", f"({url})", content) |
| 49 | + |
| 50 | + return content |
| 51 | + |
| 52 | + |
| 53 | +def decrease_header_levels(content): |
| 54 | + """Decrease each Markdown header by one level.""" |
| 55 | + lines = content.splitlines() |
| 56 | + new_lines = [] |
| 57 | + for line in lines: |
| 58 | + if re.match(r"^(#{2,6})\s", line): |
| 59 | + num_hashes = len(line.split()[0]) |
| 60 | + new_line = "#" * (num_hashes - 1) + line[num_hashes:] |
| 61 | + new_lines.append(new_line) |
| 62 | + else: |
| 63 | + new_lines.append(line) |
| 64 | + return "\n".join(new_lines) |
| 65 | + |
| 66 | + |
| 67 | +def clean_trailing_links(content): |
| 68 | + """Remove trailing links and clean up extra empty lines.""" |
| 69 | + # Remove [label]: link style |
| 70 | + content = re.sub(r"^\[.+?\]:\s+\S+$", "", content, flags=re.MULTILINE) |
| 71 | + # Remove (url): url style |
| 72 | + content = re.sub( |
| 73 | + r"^\(https?://[^\s)]+\):\s*https?://[^\s)]+$", "", content, flags=re.MULTILINE |
| 74 | + ) |
| 75 | + content = re.sub( |
| 76 | + r"^\(mailto:[^\s)]+\):\s*mailto:[^\s)]+$", "", content, flags=re.MULTILINE |
| 77 | + ) |
| 78 | + # Remove empty lines |
| 79 | + content = re.sub(r"\n{2,}", "\n\n", content).strip() |
| 80 | + return content |
| 81 | + |
| 82 | + |
| 83 | +def process_readme(readme_path, output_dir): |
| 84 | + readme = Path(readme_path).read_text(encoding="utf-8") |
| 85 | + |
| 86 | + # Extract links from README |
| 87 | + links = extract_links_from_readme(readme) |
| 88 | + |
| 89 | + # Create output directory |
| 90 | + output_dir.mkdir(exist_ok=True, parents=True) |
| 91 | + |
| 92 | + for section_title, filename in SECTION_MAPPING.items(): |
| 93 | + content = extract_section(readme, section_title) |
| 94 | + if content: |
| 95 | + myst_content = ( |
| 96 | + f"## {section_title}\n\n{convert_gfm_to_sphinx(content, links)}" |
| 97 | + ) |
| 98 | + if filename.lower() == "contact.md": |
| 99 | + myst_content = clean_trailing_links(myst_content) |
| 100 | + myst_content = decrease_header_levels(myst_content) |
| 101 | + (output_dir / filename).write_text(myst_content) |
| 102 | + print(f"Generated {filename}") |
| 103 | + else: |
| 104 | + raise ValueError(f"Section '{section_title}' not found in README") |
| 105 | + |
| 106 | + # Include CONTRIBUTING.md with its own link references |
| 107 | + contrib_path = readme_path.parent / "CONTRIBUTING.md" |
| 108 | + try: |
| 109 | + raw_contrib = contrib_path.read_text() |
| 110 | + contrib_links = extract_links_from_readme(raw_contrib) |
| 111 | + |
| 112 | + # Convert content |
| 113 | + contrib_converted = convert_gfm_to_sphinx(raw_contrib, contrib_links) |
| 114 | + |
| 115 | + # Remove trailing link definitions |
| 116 | + contrib_converted = clean_trailing_links(contrib_converted) |
| 117 | + |
| 118 | + # Write output |
| 119 | + (output_dir / "contributing.md").write_text(contrib_converted) |
| 120 | + print("Generated contributing.md") |
| 121 | + except FileNotFoundError as e: |
| 122 | + raise FileNotFoundError(f"CONTRIBUTING.md not found at {contrib_path}") from e |
| 123 | + |
| 124 | + |
| 125 | +if __name__ == "__main__": |
| 126 | + default_readme = Path(__file__).resolve().parent.parent / "README.md" |
| 127 | + output_sections_readme = Path("./sections_readme") |
| 128 | + process_readme(default_readme, output_sections_readme) |
0 commit comments