Attributes not showing for the "target-text" in a parallel corpus

[starkadur]

I have created a parallel corpus of normalized and original texts. I made an alignment for both norm -> orig and orig -> norm. When I search for a word in the normalized text I can see its structural and positional attributes appear on the right side. But if I click on a word in the original text (which appears below the normalized text) the boxes for the attributes are empty. The same happens if I search for a word in the original text - then the boxes are empty if I click on a word in the normalized text. Is this normal behaviour?

[MartinHammarstedt]

It is possible to see the attributes for not only the primary text but also the linked text. Here is an example in our Korp, where we have attributes for both the Swedish and the Dutch texts.

It's been ages since we worked on the parallel corpora, but the way it's implemented in Korp right now, I think you need to have the same attributes for both the primary and the linked text (with empty values for the attributes that are not applicable to the language in question). This is because of how the attributes are parsed from the CWB output. We could probably improve this, but it would require some work.