Thread: [Metadata Source Plugin] Kobo Books
View Single Post
Old 10-01-2020, 01:07 AM   #77
davidfor
Grand Sorcerer
davidfor ought to be getting tired of karma fortunes by now.davidfor ought to be getting tired of karma fortunes by now.davidfor ought to be getting tired of karma fortunes by now.davidfor ought to be getting tired of karma fortunes by now.davidfor ought to be getting tired of karma fortunes by now.davidfor ought to be getting tired of karma fortunes by now.davidfor ought to be getting tired of karma fortunes by now.davidfor ought to be getting tired of karma fortunes by now.davidfor ought to be getting tired of karma fortunes by now.davidfor ought to be getting tired of karma fortunes by now.davidfor ought to be getting tired of karma fortunes by now.
 
Posts: 24,905
Karma: 47303824
Join Date: Jul 2011
Location: Sydney, Australia
Device: Kobo:Touch,Glo, AuraH2O, GloHD,AuraONE, ClaraHD, Libra H2O; tolinoepos
Quote:
Originally Posted by DNSB View Post
Not sure if this is a bug in the Kobo Books plugin (1.8.0), calibre version 5.0.1, a Python 3 issue or a thread safe issue with calibre 5.0.1 but I get an erratic error when downloading metadata with this plugin. The lovely part is that it is erratic which is what leans me to the thread safe issue.

Code:
Traceback (most recent call last):
  File "calibre_plugins.kobobooks.__init__", line 167, in identify
  File "site-packages\mechanize\_mechanize.py", line 241, in open_novisit
  File "site-packages\mechanize\_mechanize.py", line 277, in _mech_open
  File "site-packages\mechanize\_opener.py", line 158, in _request
  File "site-packages\mechanize\_request.py", line 74, in __init__
  File "site-packages\mechanize\_urllib2_fork.py", line 192, in __init__
  File "site-packages\mechanize\_urllib2_fork.py", line 178, in normalize_url
  File "site-packages\mechanize\_urllib2_fork.py", line 175, in safe_encode
LookupError: unknown encoding: idna

idna? International Domain Names for Applications?
It could be the thread-safe problem. The error in the mechanize library is where it is parsing the URL before sending it. I might not have encoded it properly. Are there any unusual characters in the author or title? And can you post the full log to see what the URL and input looks like.

I've seen a couple of other reports like this. They did seem to be for books being fetched that would need more than basic ASCII. Again, I don't know if it is mishandling the encoding at the plugin end, the thread safe issue, or maybe something missing in mechanize.
davidfor is offline   Reply With Quote