Bugs: UnicodeDecodeError: invalid continuation byte

[Chura]

Beta3 still have it

Traceback (most recent call last):
File "cherrypy\_cprequest.pyo", line 670, in respond
File "cherrypy\lib\encoding.pyo", line 217, in __call__
File "cherrypy\_cpdispatch.pyo", line 61, in __call__
File "sabnzbd\interface.pyo", line 423, in tapi
File "sabnzbd\api.pyo", line 102, in api_handler
File "sabnzbd\api.pyo", line 509, in _api_history
File "sabnzbd\api.pyo", line 970, in report
File "json\__init__.pyo", line 243, in dumps
File "json\encoder.pyo", line 207, in encode
File "json\encoder.pyo", line 270, in iterencode
UnicodeDecodeError: 'utf8' codec can't decode byte 0xf9 in position 6: invalid start byte

[shypike]

The next Beta is 4 and not out yet.

[Chura]

Beta4 installed, so far so good
I'll give it more time and update here.

Thanks!!!!!

[shypike]

Please make sure that logging is set to at least Info level (Status page).
Then, after a day, download the log file and use a text editor to search for this text in the log:

Code: Select all

JSON encoder sees Latin-1

Then post here the full logging line.
That will help us to find the actual cause.

[Chura]

Code: Select all

{ ~ }-> grep "JSON encoder sees Latin-1" Downloads/sabnzbd.log 
2015-12-25 22:48:52,214::INFO::[json:78] JSON encoder sees Latin-1: 22:49 éåí å 25 ãöî
2015-12-25 22:48:52,214::INFO::[json:78] JSON encoder sees Latin-1: 22:49 éåí å 25 ãöî
2015-12-25 22:48:54,214::INFO::[json:78] JSON encoder sees Latin-1: 22:49 éåí å 25 ãöî
2015-12-25 22:48:54,214::INFO::[json:78] JSON encoder sees Latin-1: 22:49 éåí å 25 ãöî
2015-12-25 22:48:54,825::INFO::[json:78] JSON encoder sees Latin-1: 22:49 éåí å 25 ãöî
2015-12-25 22:48:54,825::INFO::[json:78] JSON encoder sees Latin-1: 22:49 éåí å 25 ãöî
2015-12-25 22:48:54,934::INFO::[json:78] JSON encoder sees Latin-1: 22:49 éåí å 25 ãöî
2015-12-25 22:48:55,825::INFO::[json:78] JSON encoder sees Latin-1: 22:49 éåí å 25 ãöî
2015-12-25 22:48:55,825::INFO::[json:78] JSON encoder sees Latin-1: 22:49 éåí å 25 ãöî
2015-12-25 22:48:55,887::INFO::[json:78] JSON encoder sees Latin-1: 22:49 éåí å 25 ãöî

[shypike]

Which language do you use within SABnzbd?

[Chura]

language = en

Its running on Windows 10, i have hebrew keyboard over there but OS is english.

[shypike]

It looks like a date and time string.
Is it possible that OS is set to display time and dates (especially months) in Hebrew?

[Chura]

Yes
my system is full English, but my region is Israel and I can see now that date appear in Hebrew.

[shypike]

OK, we found the spot where it goes wrong.
Thanks for you patience and assistance.

[Chura]

awesome waiting for next release...

[firesale]

Awesome you found the bug! Just FYI, this is what the log says for me. Swedish localization.

Code: Select all

2016-01-03 22:43:09,111::INFO::[json:78] JSON encoder sees Latin-1: 22:43 sön 03 jan

[shypike]

Any language which has accent marks or diacriticals in month or day names will give this issue.
We overlooked the date/time conversion.

[Chura]

I guess it will be fixed permanently now ?

[safihre]

I guess it will be fixed permanently now ?

Yes, fixed in the correct way now without need of special fallback