docs(search): document English stemming for the built-in full-text engine - #2032
Merged
Conversation
…gine spiceai/spiceai#12220 switches the built-in Tantivy full-text index from the default tokenizer to `en_stem` (English Snowball stemming, with positions retained). Indexed and query terms are both stemmed, so `running` now matches `run`/`runs`. vNext only — the change is not in v2.1.2.
Contributor
✅ Pull with Spice PassedPassing checks:
|
Contributor
|
🚀 deployed to https://96ad88d9.spiceai-org-website.pages.dev |
Collaborator
Author
|
@copilot review |
lukekim
approved these changes
Aug 1, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
spiceai/spiceai#12220 changes how the built-in full-text engine analyzes text: the Tantivy index now uses the
en_stemtokenizer (English Snowball stemming, lowercasing,WithFreqsAndPositionsso phrase queries keep working) instead of Tantivy's defaultTEXToptions. Tantivy'sQueryParseris constructed with the index's tokenizer manager, so query terms are stemmed the same way —runningnow matches documents containingrunandruns.This is user-visible matching behavior with no configuration knob, and the Full-text Search page said nothing about analysis at all. Adds a short Text Analysis section under Engines covering:
on_zero_results: use_sourcefallback) use Elasticsearch's own analyzer.Source PRs
Doc pages changed (1)
website/docs/features/search/full-text.md— vNext onlyScoped to vNext:
en_stemis not present atv2.1.2(git grep en_stem v2.1.2 -- crates/search/→ no hits), so no versioned snapshot needs it.Test plan
cd website && npm run buildpasses (the added#warm-tierlink resolves to the#### Warm tierheading on the same page)