Can't really blame BeautifulSoup for this, but our crawler hit a page
("http://clagnut.com/privacy/";) with an out of range character escape:
𔃷
in this text:
If you provide a name, email address and/or website and choose ‘Remember
me𔃷, these details will be stored as a cookie on your computer.
The author clearly meant "’", which is a single close quote.
The traceback as BeautifulSoup aborts:
SGMLParser.feed(self, markup or "")
File "/usr/local/lib/python2.5/sgmllib.py", line 99, in feed
self.goahead(0)
File "/usr/local/lib/python2.5/sgmllib.py", line 181, in goahead
self.handle_charref(name)
File "/var/www/vhosts/sitetruth.com/cgi-bin/sitetruth/BeautifulSoup.py", line
1250, in handle_charref
data = unichr(int(ref))
ValueError: unichr() arg not in range(0x10000) (narrow Python build)
Another item in our ongoing saga of "What happens when you parse real-world
HTML".
A try-block in handle_charref would be appropriate.
John Nagle
SiteTruth
--
http://mail.python.org/mailman/listinfo/python-list
