[solved] Danish language?

Comments

• sebastiaan

  May 2014 edited 2:56AM

  Hi Mona,

  Generally speaking, a UnicodeDecodeError occurs when you are working with text that contains special characters (like Danish, which uses accents like 'ø'), but don't explicitly specify your text encoding. OpenSesame should be unicode-safe, but if you write your own inline_script items, you have to take care not to trigger these errors. It can be tricky stuff.

  I would first take a look at the full error message, which you can find in the debug window. This should tell you whether it's triggered by an inline_script (I suspect so), or by OpenSesame itself. If the latter is the case, it's a bug. But either way, please post the full error message and (if any) the inline_script that caused it.

  Is it not possible to use the Danish letters?

  

  Cheers,
  Sebastiaan

  
  
  Check out SigmundAI.eu for our OpenSesame AI assistant!
  
• Mona

  May 2014 edited May 2014

  Hi Sebastiaan,

  thank you for your nice reply

  It indeed has to do with the inline script. It works up until I change the name of the questions into Danish.

  In a normal text display it works fine, indeed.

  Mona

  the 'stripped' code that gives the error after including Danish letters:

  for the questions:

  Question_number1 = "Jeg tænkte på andre"
  

  to build the questionnaire:

  from libopensesame import widgets
  form = widgets.form(self.experiment, cols=[2,1], rows=[1,1,1,1,1,1,1,1,1,1,1,1])
  question01 = widgets.label(form, text=Question_number1, center=False)
  form.set_widget(question01, (0,rows[0]))
  form._exec()
  

  to log name of question:

  self.log('Question1 = "Jeg tænkte på andre"')
  

  Python traceback:

  inline_error: Error: Inline script error
  In: inline_10 (run phase)
  File "libopensesame/item.pyc", line 989, in log
  UnicodeDecodeError: 'ascii' codec can't decode byte 0xc3 in position 18: ordinal not in range(128)
  
• sebastiaan

  May 2014 edited 2:56AM

  Hi Mona,

  The problem is the logging part, here:

  self.log('Question1 = "Jeg tænkte på andre"')
  

  You're running into a pretty tricky aspect of Python, which causes a lot of trouble for a lot of people. Basically, if you define a str (e.g., s = 'my str'), Python doesn't know what the character encoding is, and--unless you explicitly tell Python otherwise--assumes that it is ascii, which doesn't contain any special characters. This is weird, because you can see the special characters in the editor, but nevertheless Python doesn't know how to deal with them.

  To work around this in a more-or-less reliable way you should use the unicode type, like so:

  self.log(u'Question1 = "Jeg tænkte på andre"')
  

  The difference in notation is the little u in front of the text, but this makes all the difference. You now declare the text in a non-ambiguous way, i.e. in a such a way that Python knows how to deal with special characters.

  This is a really tricky subject, which is explained reasonably well here:

  • https://docs.python.org/2/howto/unicode.html

  But for now, I would simply recommend always using unicode text in inline_script items. As a general (but non-fool-proof) rule this works best.

  Cheers!
  Sebastiaan

  
  
  Check out SigmundAI.eu for our OpenSesame AI assistant!
  
• Mona

  June 2014 edited 2:56AM

  Perfect! All that was between me and the solution was a little "u"

  Thank you so much for the great help!

  Now we can send the questionnaire (and open sesame) to Denmark!

  Mona