gistfile1.py

# -*- coding: utf-8 -*-
from lxml.html import parse
import re

arquivo = open('Playboyentrevistas2001.html')
soup = parse(arquivo).getroot()

nome = re.compile(r'([A-Z]*)')
perguntas = soup.xpath("//p[starts-with(text(),'PLAYBOY')]")

raw = []
for p in perguntas:
	pergunta = p.text_content()
	resposta = p.getnext().getnext().text_content()
	autor = re.search(nome, resposta).group()
	raw.append('"'+pergunta+'";"'+resposta+'";"'+autor+'"')

zing = open('2001.csv', 'w')
zing.write('\n'.join(raw).encode('utf-8'))
zing.close()