Python Analyzer for lucene

[mustafa0x]

I'm using classical solr/lucene. I'd like to write a python analyzer (part-of-speech NLP library). I couldn't find a good way to do this using solr/lucene. I came across this file which makes me think lupyne might be a solution:

https://github.com/coady/lupyne/blob/main/lupyne/engine/analyzers.py

Or should I instead process the text before inserting into lucene? Eg replace <word> with <pos>|<word>, then tell lucene to split at |?

[coady]

Lucene supports custom analyzers, and lupyne makes writing one on easier. I'm not as familiar with Solr, but it appears to supports custom analyzers through plugins.

[mustafa0x]

Thanks for the reply!

Yes, solr does support custom analyzers, but the problem is that they have to be in Java.

https://solr.apache.org/guide/solr/latest/indexing-guide/analyzers.html

The simplest way to configure an analyzer is with a single element whose class attribute is a fully qualified Java class name.

However the library I plan to use is in Python, hence my questions.

Thanks again!