"""
Utilidades para DevDocs MCP
Conversión HTML a Markdown y otras funciones auxiliares
"""
import re
def html_to_markdown(html_content: str) -> str:
"""
Convierte HTML a Markdown de forma simple pero efectiva.
Optimizado para el formato de documentación de DevDocs.
"""
if not html_content:
return ""
# Remover scripts y estilos
html_content = re.sub(r'<script[^>]*>.*?</script>', '', html_content, flags=re.DOTALL)
html_content = re.sub(r'<style[^>]*>.*?</style>', '', html_content, flags=re.DOTALL)
# Convertir headings
html_content = re.sub(r'<h1[^>]*>(.*?)</h1>', r'\n# \1\n', html_content, flags=re.DOTALL)
html_content = re.sub(r'<h2[^>]*>(.*?)</h2>', r'\n## \1\n', html_content, flags=re.DOTALL)
html_content = re.sub(r'<h3[^>]*>(.*?)</h3>', r'\n### \1\n', html_content, flags=re.DOTALL)
html_content = re.sub(r'<h4[^>]*>(.*?)</h4>', r'\n#### \1\n', html_content, flags=re.DOTALL)
html_content = re.sub(r'<h5[^>]*>(.*?)</h5>', r'\n##### \1\n', html_content, flags=re.DOTALL)
html_content = re.sub(r'<h6[^>]*>(.*?)</h6>', r'\n###### \1\n', html_content, flags=re.DOTALL)
# Convertir bloques de código (pre antes que code para evitar conflictos)
html_content = re.sub(
r'<pre[^>]*><code[^>]*class="([^"]*)"[^>]*>(.*?)</code></pre>',
lambda m: f'\n```{extract_language(m.group(1))}\n{m.group(2)}\n```\n',
html_content,
flags=re.DOTALL
)
html_content = re.sub(r'<pre[^>]*>(.*?)</pre>', r'\n```\n\1\n```\n', html_content, flags=re.DOTALL)
# Convertir código inline
html_content = re.sub(r'<code[^>]*>(.*?)</code>', r'`\1`', html_content, flags=re.DOTALL)
# Convertir listas
html_content = re.sub(r'<ul[^>]*>', '\n', html_content)
html_content = re.sub(r'</ul>', '\n', html_content)
html_content = re.sub(r'<ol[^>]*>', '\n', html_content)
html_content = re.sub(r'</ol>', '\n', html_content)
html_content = re.sub(r'<li[^>]*>(.*?)</li>', r'- \1\n', html_content, flags=re.DOTALL)
# Convertir párrafos
html_content = re.sub(r'<p[^>]*>(.*?)</p>', r'\1\n\n', html_content, flags=re.DOTALL)
# Convertir enlaces
html_content = re.sub(r'<a[^>]*href="([^"]*)"[^>]*>(.*?)</a>', r'[\2](\1)', html_content, flags=re.DOTALL)
# Convertir negrita y cursiva
html_content = re.sub(r'<strong[^>]*>(.*?)</strong>', r'**\1**', html_content, flags=re.DOTALL)
html_content = re.sub(r'<b[^>]*>(.*?)</b>', r'**\1**', html_content, flags=re.DOTALL)
html_content = re.sub(r'<em[^>]*>(.*?)</em>', r'*\1*', html_content, flags=re.DOTALL)
html_content = re.sub(r'<i[^>]*>(.*?)</i>', r'*\1*', html_content, flags=re.DOTALL)
# Convertir blockquotes
html_content = re.sub(r'<blockquote[^>]*>(.*?)</blockquote>', r'> \1\n', html_content, flags=re.DOTALL)
# Convertir líneas horizontales
html_content = re.sub(r'<hr[^>]*/?>', '\n---\n', html_content)
# Convertir saltos de línea
html_content = re.sub(r'<br[^>]*/?>', '\n', html_content)
# Convertir tablas (básico)
html_content = re.sub(r'<table[^>]*>', '\n', html_content)
html_content = re.sub(r'</table>', '\n', html_content)
html_content = re.sub(r'<tr[^>]*>', '', html_content)
html_content = re.sub(r'</tr>', ' |\n', html_content)
html_content = re.sub(r'<th[^>]*>(.*?)</th>', r'| **\1** ', html_content, flags=re.DOTALL)
html_content = re.sub(r'<td[^>]*>(.*?)</td>', r'| \1 ', html_content, flags=re.DOTALL)
# Remover divs y spans
html_content = re.sub(r'<div[^>]*>', '\n', html_content)
html_content = re.sub(r'</div>', '\n', html_content)
html_content = re.sub(r'<span[^>]*>(.*?)</span>', r'\1', html_content, flags=re.DOTALL)
# Remover todas las demás etiquetas HTML
html_content = re.sub(r'<[^>]+>', '', html_content)
# Decodificar entidades HTML comunes
html_content = html_content.replace('<', '<')
html_content = html_content.replace('>', '>')
html_content = html_content.replace('&', '&')
html_content = html_content.replace('"', '"')
html_content = html_content.replace(''', "'")
html_content = html_content.replace(' ', ' ')
html_content = html_content.replace('–', '–')
html_content = html_content.replace('—', '—')
html_content = html_content.replace('©', '©')
html_content = html_content.replace('®', '®')
# Limpiar múltiples líneas vacías
html_content = re.sub(r'\n{3,}', '\n\n', html_content)
# Limpiar espacios al inicio de líneas
html_content = re.sub(r'\n[ \t]+', '\n', html_content)
return html_content.strip()
def extract_language(class_name: str) -> str:
"""Extrae el lenguaje de programación de una clase CSS"""
# Patrones comunes: language-python, lang-js, highlight-python
patterns = [
r'language-(\w+)',
r'lang-(\w+)',
r'highlight-(\w+)',
]
for pattern in patterns:
match = re.search(pattern, class_name)
if match:
return match.group(1)
return ""
def truncate_text(text: str, max_length: int = 10000) -> str:
"""Trunca texto largo manteniendo palabras completas"""
if len(text) <= max_length:
return text
truncated = text[:max_length]
# Cortar en el último espacio para no cortar palabras
last_space = truncated.rfind(' ')
if last_space > max_length * 0.8:
truncated = truncated[:last_space]
return truncated + "\n\n... [contenido truncado]"
