qdrant / qdrant/fastembed

[Feature]: Support pluggable stemmer for BM25 (enable Polish and other non‑Snowball languages)

Open
#654 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
3.2k
Forks
248
Avg merge
4d 8h
Merged PRs (30d)
4

Description

What feature would you like to request?

Motivation: SnowballStemmer from py_rust_stemmers has no Polish algorithm, so Bm25(language="polish") is impossible even though Polish is a common target language for Qdrant users. Same applies to Czech, Ukrainian, Slovak, Bulgarian, Vietnamese…

Proposal: allow Bm25(stemmer=<callable/obj with .stem_word>, stopwords=[...], language="polish"). Skip supported_languages check when a custom stemmer is provided. Skip stopwords file download when stopwords are provided inline.

Alternative: ship a PolishStemmer wrapping pystempel behind an extra: pip install fastembed[polish].

Is there any additional information you would like to provide?

No response

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start at the Bm25 entry point and trace how SnowballStemmer, supported_languages, and stopword file downloads are handled. Check the proposed callable or .stem_word interface and inline stopwords behavior. Done means custom stemmers work for languages outside the supported list and inline stopwords avoid a download, with coverage for these cases.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
search
Issue type
Feature
Difficulty
4/5
Estimated time
3-5 days
Activity status
Quiet
Clarity
Mostly clear
Newbie friendliness
48/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.