UnicodeEncodeError in Task name

[Jana Mizikova]

Hello,

recently I've been working on the custom monthly tasks report via Python API (1.7.0). And it fails everytime when tasks are created by managers with special characters in a task name, otherwise it works fine on its own. I don't want to manually rename the task in the web UI everytime it happens. I wanted to fix it in my script with exception handling block for UnicodeEncodeError. But for some reason I can't even query the task from the session to replace the wrong character in my script. I get this error:

>>> session.query('Task where parent.name is "Test Subjects" and type is "R&D"').one()

Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "P:\.sys\python\Python27\lib\site-packages\ftrack_api\query.py", line 164, in one
    results, metadata = self._session._query(expression)
  File "P:\.sys\python\Python27\lib\site-packages\ftrack_api\session.py", line 821, in _query
    results = self._call(batch)
  File "P:\.sys\python\Python27\lib\site-packages\ftrack_api\session.py", line 1637, in _call
    raise ftrack_api.exception.ServerError(error_message)
ftrack_api.exception.ServerError: Server reported error: AttributeError('unicode' object has no attribute '_sa_instance_state')

I can query the task if conditions result in getting multiple tasks from the query, but when it gets to passing the name of the task for further processing, I get this:

UnicodeEncodeError: 'ascii' codec can't encode character u'\u0161' in position 28: ordinal not in range(128)

Am I doing something wrong here? I don't understand how and why it is even possible to create a task with a special character in the task name if it's using unicode as a variable type and then it cannot be queried. Is there another way how to query the task from session? Can I change encoding in the process of reading a task name?

 

Thanks in advance for the reply!

- Jane

[tomas]

Hi Jana!

if you want select task by type, you must use "type.name" in query because "type" is an object and you want use name of object in filter. So you can change your query to:

session.query('Task where parent.name is "Test Subjects" and type.name is "R&D"').all()

And Unicode error you get mean the name of your task contain some utf8 character ('š' I guess) and API return you not "string" object but "unicode".

print type(your_task["name"])

>> <type 'unicode'>

If you try cast unicode with utf character to string, you will get unicode error (https://stackoverflow.com/questions/18034272/python-str-vs-unicode-types)...So you can try following: 

print type(your_task["name"].encode("utf-8"))

>> <type 'str'>

and you get string

answer from stackoverflow:

https://stackoverflow.com/questions/9942594/unicodeencodeerror-ascii-codec-cant-encode-character-u-xa0-in-position-20

[Jana Mizikova]

Thanks for the reply ! 

I thought those two errors are connected and caused by one thing but apparently I did two separate mistakes there : ) The query works fine now. I fixed also the part with encoding to utf-8 and works smoothly. Thank you very much !

- Jana