document_fromstring

[commissar]

I use lxml to parse the contents of the url(http://www.dtzww.cn/files/article/fulltext/23/23208.html),the lxml is been blocking,and don't rasie exception. The CPU utilization rate is 100%.

My environment is lxml-2.2.2. ubutnu-8.04-amd64-server python-2.5.2My code is fellow:

import lxml.html as htmltool
import urlib

url = "http://www.dtzww.cn/files/article/fulltext/23/23208.html"
f = urllib.urlopen(url)
data = f.read()

doc = htmltool.document_fromstring(data) ## <--- Block this

and use the following code if it is feasible.

====sample 1:
import lxml.html as htmltool
import urlib

url = "http://www.dtzww.cn/files/article/fulltext/23/23208.html"
doc = htmltool.parse(url) ## <--- ok

====My question:
why the document_fromstring can not work ?and the lxml.html.parse and lxml.html.document_fromstring are not the same used in a way?

Thank you

[scoder (scoder)]

Reproducible with libxml2 2.7.3. This seems to be a problem in libxml2 rather than lxml. I'll ask over there.

[commissar (commissarster)]

OK,thanks

[scoder (scoder)]

This has been fixed in libxml2, likely to be released in 2.7.4.

http://bugzilla.gnome.org/show_bug.cgi?id=592430

[scoder (scoder)]

Marking this as "invalid", as it can't be solved in lxml itself.