mirror of
https://github.com/qdrant/qdrant.git
synced 2026-10-29 09:04:50 +08:00
9b85d7b5b0
* [BM25] BM25 text search in edge-shard-query Add a `text` subcommand to `edge-shard-query`, running BM25 over the text index of a field through the query path, a `bm25` payload index type to `edge-tool create` to build a shard to run it on, and `edge-tool load-beir` to fill it with a BEIR corpus. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * load-beir: say a smaller re-run leaves points, reject an id overflow - a re-run loading fewer documents leaves the earlier ones past its end: documented in the args and README - a --start-id that overflows u64 is an error instead of a panic or wrap Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> * Open the ambient trace format in the uio trace visualizer #11041 changed the trace records: spans as span_start/span_end pairs, started/ended and timestamp in epoch nanoseconds, and one columnar cpu record. The visualizer read only the old fields, so new traces showed nothing. It now reads both, rebasing epoch times on the earliest one. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com> Co-authored-by: generall <andrey@vasnetsov.com>