PSeitz

As reported by a user: ... Since that showed promised, I tried increasing split_num_docs_target to 20 million. However, my index appears to be corrupt and is returning the following error...

The way incorrect types on typed columns are handled in columnar is a footgun. We should ignore it or return an error. ```rust #[test] fn test_date_range_query() { let mut schema_builder...

The `precision_threshold` parameter is not supported in the cardinality aggregation currently. It defines until which count the aggregation is exact and switches to an approximation afterwards. Currently the aggregation always...

As reported in discord: ```rust use tantivy::collector::FacetCollector; use tantivy::query::{BooleanQuery, Occur, Query, TermQuery}; use tantivy::schema::*; use tantivy::{doc, Index, IndexWriter}; fn main() -> tantivy::Result { let mut schema_builder = Schema::builder(); let id...

RegexPhraseQuery supports phrase queries with regex. It supports regex and wildcards. E.g. a query with wildcards: "b* b* wolf" matches "big bad wolf" Slop is supported as well: "b* wolf"~2...

Potential pruning optimization for phrase queries: The score of a phrase query is defined by how often a phrase occurs in a document (`term_freq` is equal to `phrase_count`). Once we...

temp unbump to 0.22 for easier release with `cargo release`

use new expect lint in clippy, which works like allow, but errors if the lint is not required anymore convert clippy allow to clippy expect and remove unused cleanup

``` failures: actors::indexing_service::tests::test_indexing_service_shutdown_merge_pipeline_when_no_indexing_pipeline test result: FAILED. 0 passed; 1 failed; 0 ignored; 0 measured; 159 filtered out; finished in 1.03s --- TRY 2 STDERR: quickwit-indexing actors::indexing_service::tests::test_indexing_service_shutdown_merge_pipeline_when_no_indexing_pipeline --- thread 'actors::indexing_service::tests::test_indexing_service_shutdown_merge_pipeline_when_no_indexing_pipeline' panicked...