idea_content_link.py 文件源码

def _infer_text_fragment_inner(self, title, body, post_id):
        # dead code? If not needs to be refactored with langstrings
        # and moved within text_fragment, maybe?
        body = sanitize_html(body, [])
        quote = self.quote.replace("\r", "")
        try:
            # for historical reasons
            quote = quopri.decodestring(quote)
        except:
            pass
        quote = sanitize_html(quote, [])
        if quote != self.body:
            self.body = quote
        quote = quote.replace("\n", "")
        start = body.find(quote)
        lookin = 'message-body'
        if start < 0:
            xpath = "//div[@id='%s']/div[class='post_title']" % (post_id)
            start = title.find(quote)
            if start < 0:
                return None
            lookin = 'message-subject'
        xpath = "//div[@id='message-%s']//div[@class='%s']" % (
            Post.uri_generic(post_id), lookin)
        tfi = self.db.query(TextFragmentIdentifier).filter_by(
            extract=self).first()
        if not tfi:
            tfi = TextFragmentIdentifier(extract=self)
        tfi.xpath_start = tfi.xpath_end = xpath
        tfi.offset_start = start
        tfi.offset_end = start+len(quote)
        return tfi