From 3a25137ee42d1f6d98db6f9e569baae40cb1949f Mon Sep 17 00:00:00 2001 From: Kerollmops Date: Wed, 17 Mar 2021 13:55:24 +0100 Subject: [PATCH] Expose and use the WordsLevelPositions update --- milli/src/update/index_documents/mod.rs | 17 +++++++++++++++++ milli/src/update/update_builder.rs | 20 +++++++++++++++++++- 2 files changed, 36 insertions(+), 1 deletion(-) diff --git a/milli/src/update/index_documents/mod.rs b/milli/src/update/index_documents/mod.rs index 8fc35b654..e7143bde0 100644 --- a/milli/src/update/index_documents/mod.rs +++ b/milli/src/update/index_documents/mod.rs @@ -263,6 +263,8 @@ pub struct IndexDocuments<'t, 'u, 'i, 'a> { facet_min_level_size: Option, words_prefix_threshold: Option, max_prefix_length: Option, + words_positions_level_group_size: Option, + words_positions_min_level_size: Option, update_method: IndexDocumentsMethod, update_format: UpdateFormat, autogenerate_docids: bool, @@ -290,6 +292,8 @@ impl<'t, 'u, 'i, 'a> IndexDocuments<'t, 'u, 'i, 'a> { facet_min_level_size: None, words_prefix_threshold: None, max_prefix_length: None, + words_positions_level_group_size: None, + words_positions_min_level_size: None, update_method: IndexDocumentsMethod::ReplaceDocuments, update_format: UpdateFormat::Json, autogenerate_docids: true, @@ -740,6 +744,19 @@ impl<'t, 'u, 'i, 'a> IndexDocuments<'t, 'u, 'i, 'a> { } builder.execute()?; + // Run the words level positions update operation. + let mut builder = WordsLevelPositions::new(self.wtxn, self.index, self.update_id); + builder.chunk_compression_type = self.chunk_compression_type; + builder.chunk_compression_level = self.chunk_compression_level; + builder.chunk_fusing_shrink_size = self.chunk_fusing_shrink_size; + if let Some(value) = self.words_positions_level_group_size { + builder.level_group_size(value); + } + if let Some(value) = self.words_positions_min_level_size { + builder.min_level_size(value); + } + builder.execute()?; + debug_assert_eq!(database_count, total_databases); info!("Transform output indexed in {:.02?}", before_indexing.elapsed()); diff --git a/milli/src/update/update_builder.rs b/milli/src/update/update_builder.rs index c966f72d2..9a4fb850e 100644 --- a/milli/src/update/update_builder.rs +++ b/milli/src/update/update_builder.rs @@ -2,7 +2,10 @@ use grenad::CompressionType; use rayon::ThreadPool; use crate::Index; -use super::{ClearDocuments, DeleteDocuments, IndexDocuments, Settings, Facets, WordsPrefixes}; +use super::{ + ClearDocuments, DeleteDocuments, IndexDocuments, Settings, + Facets, WordsPrefixes, WordsLevelPositions, +}; pub struct UpdateBuilder<'a> { pub(crate) log_every_n: Option, @@ -150,4 +153,19 @@ impl<'a> UpdateBuilder<'a> { builder } + + pub fn words_level_positions<'t, 'u, 'i>( + self, + wtxn: &'t mut heed::RwTxn<'i, 'u>, + index: &'i Index, + ) -> WordsLevelPositions<'t, 'u, 'i> + { + let mut builder = WordsLevelPositions::new(wtxn, index, self.update_id); + + builder.chunk_compression_type = self.chunk_compression_type; + builder.chunk_compression_level = self.chunk_compression_level; + builder.chunk_fusing_shrink_size = self.chunk_fusing_shrink_size; + + builder + } }