sanitize_contrib_titles.py

import bleach
import re
import HTMLParser

h = HTMLParser.HTMLParser()

queries = [
    Contribution.query.filter(Contribution.title.like('%</%>%') | Contribution.title.like('%&%;%')),
    Session.query.filter(Session.title.like('%</%>%') | Session.title.like('%&%;%')),
    SessionBlock.query.filter(SessionBlock.title.like('%</%>%') | SessionBlock.title.like('%&%;%')),
    Break.query.filter(Break.title.like('%</%>%') | Break.title.like('%&%;%'))
]

for query in queries:
    for c in query:
        bleached = bleach.clean(c.title, strip=True, tags=['br', 'sub', 'sup'])
        # unescape HTML entities
        unescaped = h.unescape(re.sub(r'&rightarrow', '&rarr', bleached, re.IGNORECASE))
        # convert <br> to ' - '
        text = re.sub(r'<\w*br\w*/?>', ' - ', re.sub(r'<br/?>\w*$', '', unescaped, re.IGNORECASE)).strip()
        c.title = text
        print c.id, text