Pattern Matcher: Options to exclude phrases

[ik4Rus]

Hi,
first of all thank you very much for the amazing package and the fantastic documentation. :)
I didn't find a solution for the following challenge and I'm not sure if it would be a new feature or I didn't find it in the documentation.

Minimal example

Using rule-based matching I would like to find all patterns, where the lemma 'house' occurs, except the lemma 'in' occurs in any token before it:
Expected:
I love the house. - match
I am in the house. - no match
We were playing in their great house. - no match

(Not working) approach

Patterns like [{'LEMMA': {'NOT_IN': ['in']}}, {'OP': '*'}, {'LEMMA': 'house'}] do not work, because then e.g. 'the' matches {'LEMMA': {'NOT_IN': ['in']}}}.

Related documentation

I did not find anything in the documentation (https://spacy.io/usage/rule-based-matching#_title) and forums.

Thank you :)

[svlandeg]

I could be mistaken, but I think you're right that the current matcher does not offer this functionality. There are some related feature requests, such as #2262 which is a little less complex than what you're describing.

In general I think we'd be happy to accept contributions to enhance the functionality in the Matcher but we may not have time to look into this ourselves in the near future.

[delzac]

@ik4Rus Perhaps you can do something like this: [{'LEMMA': {'NOT_IN': ['in']}, "OP": "*"}, {"LEMMA": "house"}]. This way it seems like anything before house that has an in will not be matched.