serhiy-bzhezytskyy commented on PR #4812: URL: https://github.com/apache/solr/pull/4812#issuecomment-5601724739
In this PR, as its own commit -- so no stacked PR needed. It only makes sense on top of the `SpellCheckToken` record this PR adds, and splitting it out would mean reviewing the same five call sites twice. One divergence from your sketch: `QueryWordsTokenStream` stays. `convert()` has to keep returning a `TokenStream` -- a converter may return any stream, and `SimpleQueryConverter` returns one straight from the analyzer -- so the drain belongs in `SpellCheckComponent`, not in the converter. The class stays, and the extension point with it. Analyses per request go from 1 + N (+1 on shard requests) to 1. The peek is `!tokens.isEmpty()`, `AttributeReader` is private now that only `drain` uses it, and the per-word `catch` is back with a test that fails without it. 24 classes / 90 tests green; spotless, forbidden-apis, ecj-lint and javadoc clean. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
