置換処理の置換後の文字列の代わりに関数を指定する

今後も使えそうなのでメモ。

import re
from django.core.exceptions import ObjectDoesNotExist
from sample.web.models import WebPage
PATTERN = r'<a href="%(dummy_url)s" >%(real_url)s</a>'
HTML_TEMPLATE = re.compile(r"((http|https)://[-_.!~*\'()a-zA-Z0-9;\/?:\@&=+\$,%#]+)", re.I)
def substitute_url(m):
url = m.group(0)
webpage = None
try:
webpage = WebPage.objects.get(url=url)
except ObjectDoesNotExist:
webpage = WebPage()
webpage.url = url
webpage.save()
return PATTERN % {"real_url": url, "dummy_url": "/bookmark/%s/" % (webpage.id,),}
print HTML_TEMPLATE.sub(substitute_url, "http://b.hatena.ne.jp/ が面白かったので紹介します。")