#!/usr/bin/env python3
"""
cgit about-filter: renders README.md as HTML with Pygments syntax highlighting.

cgit calls this script with the readme filename as argv[1] and file content on stdin.
Output is HTML written to stdout.
"""

import sys
import re
import markdown
from pygments import highlight
from pygments.lexers import get_lexer_by_name, TextLexer
from pygments.formatters import HtmlFormatter
from pygments.util import ClassNotFound


def highlight_code_blocks(html):
    def replacer(match):
        lang = match.group(1) or ""
        code = match.group(2)
        # Unescape HTML entities from markdown output
        code = code.replace("&amp;", "&").replace("&lt;", "<").replace("&gt;", ">").replace("&quot;", '"')
        try:
            lexer = get_lexer_by_name(lang.strip(), stripnl=False) if lang.strip() else TextLexer(stripnl=False)
        except ClassNotFound:
            lexer = TextLexer(stripnl=False)
        formatter = HtmlFormatter(nowrap=True, cssclass="highlight", noclasses=False)
        highlighted = highlight(code, lexer, formatter)
        return f'<pre><code class="highlight">{highlighted}</code></pre>'

    # Match <pre><code class="language-LANG"> or <pre><code>
    pattern = re.compile(
        r'<pre><code(?:\s+class="language-([^"]*)")?>(.*?)</code></pre>',
        re.DOTALL
    )
    return pattern.sub(replacer, html)


content = sys.stdin.read()
html = markdown.markdown(content, extensions=["fenced_code", "tables", "toc"])
print(highlight_code_blocks(html))