f11r

The same problem seems to appear when parsing lists like the following (from wikipedia), which makes this problem much worse since it isn't just about texts in the wrong language....

Yes, this "fixes" it by segmenting/tokenizing the document again. However, I need to keep the tokenization the way it is provided to corenlp and only use the parser (I'm aware...

This is an old issue but for anyone that comes across this looking for a solution you can use the following: ``` .where(pypika.terms.Term.wrap_constant("python") == ARRAY_ANY(proj_t.languages)) ```