snowykami/meilisearch
1
0
Fork 0
mirror of https://github.com/meilisearch/meilisearch.git synced 2025-06-04 19:25:32 +00:00
Code Issues Packages Projects Releases Wiki Activity

kickstart new settings indexer

Browse Source
This commit is contained in:
Louis Dureuil 2025-04-10 09:56:03 +02:00
parent 066fc2a3c9
commit c05c83a99d
No known key found for this signature in database
1 changed files with 72 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

72
crates/milli/src/update/settings.rs
View File

@ -25,6 +25,7 @@ use crate::index::{
DEFAULT_MIN_WORD_LEN_TWO_TYPOS,
};
use crate::order_by_map::OrderByMap;
use crate::progress::Progress;
use crate::prompt::default_max_bytes;
use crate::proximity::ProximityPrecision;
use crate::update::index_documents::IndexDocumentsMethod;
@ -1271,6 +1272,77 @@ impl<'a, 't, 'i> Settings<'a, 't, 'i> {
Ok(())
}
pub fn new_execute<'indexer, MSP>(
mut self,
must_stop_processing: &'indexer MSP,
progress: &'indexer Progress,
) -> Result<()>
where
MSP: Fn() -> bool + Sync,
{
// force the old indexer if the environment says so
if std::env::var_os("MEILI_EXPERIMENTAL_NO_EDITION_2024_FOR_SETTINGS").is_some() {
return self.execute(
|indexing_step| tracing::debug!("update: {:?}", indexing_step),
must_stop_processing,
);
}
// only use the new indexer when only the embedder possibly changed
if let Self {
searchable_fields: Setting::NotSet,
displayed_fields: Setting::NotSet,
filterable_fields: Setting::NotSet,
sortable_fields: Setting::NotSet,
criteria: Setting::NotSet,
stop_words: Setting::NotSet,
non_separator_tokens: Setting::NotSet,
separator_tokens: Setting::NotSet,
dictionary: Setting::NotSet,
distinct_field: Setting::NotSet,
synonyms: Setting::NotSet,
primary_key: Setting::NotSet,
authorize_typos: Setting::NotSet,
min_word_len_two_typos: Setting::NotSet,
min_word_len_one_typo: Setting::NotSet,
exact_words: Setting::NotSet,
exact_attributes: Setting::NotSet,
max_values_per_facet: Setting::NotSet,
sort_facet_values_by: Setting::NotSet,
pagination_max_total_hits: Setting::NotSet,
proximity_precision: Setting::NotSet,
embedder_settings: _,
search_cutoff: Setting::NotSet,
localized_attributes_rules: Setting::NotSet,
prefix_search: Setting::NotSet,
facet_search: Setting::NotSet,
wtxn: _,
index: _,
indexer_config: _,
} = &self
{
todo!()
// 1. First we want to update the database and compute the settings diff, we might reuse a bunch of existing functions here
// 2. Pick which pipelines we need to run.
// 3. Execute extraction pipelines
// 4. Wait for DB writes to complete
// 5. Perform post-processing operations that are necessary (FST, facets, arroy)
// need SettingsChange object + settings::Extractor trait with fn process(document: DocumentFromDb) + fn init_data function
// need settings::extract function that accepts an extractor and performs all support functions (cancelation, progress)
// as well as iterates on document calling the Extractor::process function
// we might encapsulate the `init_data` -> `process` -> `merge_data` loop even further
} else {
self.execute(
|indexing_step| tracing::debug!("update: {:?}", indexing_step),
must_stop_processing,
)
}
// create rtxn, populate FieldIdMapWithMetadata (old + new)
}
}
pub struct InnerIndexSettingsDiff {