Hi Telofy,

Thanks, you're right,

I've fixed it at one place but haven't fixed at another place.
Compare with line 59 at the same file:
https://github.com/Telofy/python-readability/blob/a355c6ea72961a431cbc3a8fac35557188543e5c/readability/htmls.py#L59

I don't remember yet why it's there and why I don't use your way...

Anyway, I'll fix it over the weekend.

BTW: I finally started to use it myself (in another project), now can't decide where to write a long post on how to apply the module correctly to get not 95% of article texts correctly but rather 99.5% of them.

Oh, 99.5% sounds tremendous. Please do link me that blog post when it’s done!

Any news on this PR ?

in latest readability from PyPI, I got this :

Traceback (most recent call last):
  File "test.py", line 77, in <module>
    test()
  File "test.py", line 62, in test
    extractor = ftr.process(url)
  File "/home/olive/sources/python-ftr/ftr/process.py", line 148, in ftr_process
    if extractor.process(html=content):
  File "/home/olive/sources/python-ftr/ftr/extractor.py", line 620, in process
    self._auto_extract_if_failed()
  File "/home/olive/sources/python-ftr/ftr/extractor.py", line 512, in _auto_extract_if_failed
    title = readabilitized.title().strip()
  File "/home/olive/.virtualenvs/1flow/lib/python2.7/site-packages/readability/readability.py", line 136, in title
    return get_title(self._html(True))
  File "/home/olive/.virtualenvs/1flow/lib/python2.7/site-packages/readability/htmls.py", line 46, in get_title
    if title is None or len(title.text) == 0:
TypeError: object of type 'NoneType' has no len()

This happens on http://wheelyric.com/lyrics/121#2. For information, I use readability as a fallback / helper in python-ftr.

PS: I know why you test explicitely is not None at the other place: that's probably because you got FutureWarning: The behavior of this method will change in future versions. Use specific 'len(elem)' or 'elem is not None' test instead. at some point while coding.

Could you use a github version or do you need a pypi release?

Released both ( https://pypi.python.org/pypi/readability-lxml and https://github.com/buriy/python-readability/releases/tag/0.3.0.6 )
I do "wget http://wheelyric.com/lyrics/121#2" and get
Length: 0 [text/html]

P.S. I'm going to finish soon v0.5 with py3 support, improved logging and on a file-based news sites downloader.
Sorry for delay.

No problem for the delay, we all have busy lives ;-)
Thanks for the release. Nice to have pushed it on pypi !
I must leave now but will update/close the issue given my result when I come back.

regards,

Just to let you know, version 0.3.0.6 works perfectly. Thanks !