Incorrect language mapping for character 'ë'

[sandernugterenedia]

For example, with the text: "Deze patiënt is al een tijd ziek" the detected language is sq (Albanian) while with the text "Deze patient is al een tijd ziek" (the same text without the "ë" character) the language is correctly identified as nl (Dutch). This also seems to happen for longer texts (if the text contains one or more "ë" characters it is classified as sq, when these characters are replaced with "e" the text is correctly detected as nl).

[pemistahl]

When I implemented this back then, I thought that the letter ë would not be part of the main Dutch orthography. But now, after reading this Wikipedia article you seem to be right. I consider this a bug then.

Thank you for letting me know about this. I will correct it soon.