snowykami/meilisearch
1
0
Fork 0
mirror of https://github.com/meilisearch/meilisearch.git synced 2024-11-22 10:07:40 +08:00
Code Issues Actions 9 Packages Projects Releases Wiki Activity

Rename and use the try_arc_for_each_try_init method

Browse Source
This commit is contained in:
Clément Renault 2024-09-29 17:42:26 +02:00
parent d83c9a4074
commit 00e045b249
No known key found for this signature in database
GPG Key ID: F250A4C4E3AE5F5F
7 changed files with 54 additions and 51 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

13
milli/src/update/new/extract/faceted/extract_facets.rs
View File

@ -212,7 +212,6 @@ impl DocidsExtractor for FacetedDocidsExtractor {
let context_pool = ItemsPool::new(|| {
Ok((
index.read_txn().map_err(Error::from).map_err(Arc::new)?,
fields_ids_map.clone(),
Vec::new(),
CboCachedSorter::new(
@ -234,12 +233,12 @@ impl DocidsExtractor for FacetedDocidsExtractor {
let span =
tracing::trace_span!(target: "indexing::documents::extract", "docids_extraction");
let _entered = span.enter();
document_changes.into_par_iter().try_for_each_try_init(
|| Ok(()),
|_, document_change| {
context_pool.with(|(rtxn, fields_ids_map, buffer, cached_sorter)| {
document_changes.into_par_iter().try_arc_for_each_try_init(
|| index.read_txn().map_err(Error::from),
|rtxn, document_change| {
context_pool.with(|(fields_ids_map, buffer, cached_sorter)| {
Self::extract_document_change(
&*rtxn,
rtxn,
index,
buffer,
fields_ids_map,
@ -261,7 +260,7 @@ impl DocidsExtractor for FacetedDocidsExtractor {
let readers: Vec<_> = context_pool
.into_items()
.par_bridge()
.map(|(_rtxn, _tokenizer, _fields_ids_map, cached_sorter)| {
.map(|(_tokenizer, _fields_ids_map, cached_sorter)| {
let sorter = cached_sorter.into_sorter()?;
sorter.into_reader_cursors()
})

19
milli/src/update/new/extract/searchable/extract_word_docids.rs
View File

@ -5,7 +5,7 @@ use std::sync::Arc;
use grenad::{Merger, MergerBuilder};
use heed::RoTxn;
use rayon::iter::{IntoParallelIterator, ParallelIterator};
use rayon::iter::IntoParallelIterator;
use super::tokenize_document::{tokenizer_builder, DocumentTokenizer};
use crate::update::new::extract::cache::CboCachedSorter;
@ -341,7 +341,6 @@ impl WordDocidsExtractors {
let context_pool = ItemsPool::new(|| {
Ok((
index.read_txn().map_err(Error::from).map_err(Arc::new)?,
&document_tokenizer,
fields_ids_map.clone(),
WordDocidsCachedSorters::new(
@ -357,13 +356,12 @@ impl WordDocidsExtractors {
let span =
tracing::trace_span!(target: "indexing::documents::extract", "docids_extraction");
let _entered = span.enter();
document_changes.into_par_iter().try_for_each_try_init(
|| Ok(()),
|_, document_change| {
context_pool.with(
|(rtxn, document_tokenizer, fields_ids_map, cached_sorter)| {
document_changes.into_par_iter().try_arc_for_each_try_init(
|| index.read_txn().map_err(Error::from),
|rtxn, document_change| {
context_pool.with(|(document_tokenizer, fields_ids_map, cached_sorter)| {
Self::extract_document_change(
&*rtxn,
rtxn,
index,
document_tokenizer,
fields_ids_map,
@ -371,8 +369,7 @@ impl WordDocidsExtractors {
document_change?,
)
.map_err(Arc::new)
},
)
})
},
)?;
}
@ -382,7 +379,7 @@ impl WordDocidsExtractors {
tracing::trace_span!(target: "indexing::documents::extract", "merger_building");
let _entered = span.enter();
let mut builder = WordDocidsMergerBuilders::new();
for (_rtxn, _tokenizer, _fields_ids_map, cache) in context_pool.into_items() {
for (_tokenizer, _fields_ids_map, cache) in context_pool.into_items() {
builder.add_sorters(cache)?;
}

17
milli/src/update/new/extract/searchable/mod.rs
View File

@ -60,7 +60,6 @@ pub trait SearchableExtractor {
let context_pool = ItemsPool::new(|| {
Ok((
index.read_txn().map_err(Error::from).map_err(Arc::new)?,
&document_tokenizer,
fields_ids_map.clone(),
CboCachedSorter::new(
@ -82,13 +81,12 @@ pub trait SearchableExtractor {
let span =
tracing::trace_span!(target: "indexing::documents::extract", "docids_extraction");
let _entered = span.enter();
document_changes.into_par_iter().try_for_each_try_init(
|| Ok(()),
|_, document_change| {
context_pool.with(
|(rtxn, document_tokenizer, fields_ids_map, cached_sorter)| {
document_changes.into_par_iter().try_arc_for_each_try_init(
|| index.read_txn().map_err(Error::from),
|rtxn, document_change| {
context_pool.with(|(document_tokenizer, fields_ids_map, cached_sorter)| {
Self::extract_document_change(
&*rtxn,
rtxn,
index,
document_tokenizer,
fields_ids_map,
@ -96,8 +94,7 @@ pub trait SearchableExtractor {
document_change?,
)
.map_err(Arc::new)
},
)
})
},
)?;
}
@ -110,7 +107,7 @@ pub trait SearchableExtractor {
let readers: Vec<_> = context_pool
.into_items()
.par_bridge()
.map(|(_rtxn, _tokenizer, _fields_ids_map, cached_sorter)| {
.map(|(_tokenizer, _fields_ids_map, cached_sorter)| {
let sorter = cached_sorter.into_sorter()?;
sorter.into_reader_cursors()
})

3
milli/src/update/new/indexer/document_operation.rs
View File

@ -5,11 +5,10 @@ use std::sync::Arc;
use heed::types::Bytes;
use heed::RoTxn;
use memmap2::Mmap;
use rayon::iter::{IndexedParallelIterator, IntoParallelIterator, ParallelIterator};
use rayon::iter::{IndexedParallelIterator, IntoParallelIterator};
use IndexDocumentsMethod as Idm;
use super::super::document_change::DocumentChange;
use super::super::items_pool::ItemsPool;
use super::super::{CowStr, TopLevelMap};
use super::DocumentChanges;
use crate::documents::{DocumentIdExtractionError, PrimaryKey};

5
milli/src/update/new/indexer/mod.rs
View File

@ -81,7 +81,8 @@ where
// document but we need to create a function that collects and compresses documents.
let document_sender = extractor_sender.document_sender();
document_changes.clone().into_par_iter().try_for_each_try_init(|| Ok(()) as Result<_>, |_, result| {
document_changes.clone().into_par_iter().try_arc_for_each::<_, Error>(
|result| {
match result? {
DocumentChange::Deletion(deletion) => {
let docid = deletion.docid();
@ -99,7 +100,7 @@ where
// extracted_dictionary_sender.send(self, dictionary: &[u8]);
}
}
Ok(()) as std::result::Result<_, Arc<_>>
Ok(())
})?;
document_sender.finish().unwrap();

2
milli/src/update/new/indexer/partial_dump.rs
View File

@ -1,6 +1,6 @@
use std::sync::Arc;
use rayon::iter::{IndexedParallelIterator, ParallelBridge, ParallelIterator};
use rayon::iter::IndexedParallelIterator;
use super::DocumentChanges;
use crate::documents::{DocumentIdExtractionError, PrimaryKey};

14
milli/src/update/new/items_pool.rs
View File

@ -1,4 +1,3 @@
use std::convert::identity;
use std::sync::Arc;
use crossbeam_channel::{Receiver, Sender, TryRecvError};
@ -38,7 +37,7 @@ pub trait ParallelIteratorExt: ParallelIterator {
/// A method to run a closure of all the items and return an owned error.
///
/// The init function is ran only as necessary which is basically once by thread.
fn try_for_each_try_init<F, INIT, T, E>(self, init: INIT, op: F) -> Result<(), E>
fn try_arc_for_each_try_init<F, INIT, T, E>(self, init: INIT, op: F) -> Result<(), E>
where
E: Send + Sync,
F: Fn(&mut T, Self::Item) -> Result<(), Arc<E>> + Sync + Send + Clone,
@ -60,6 +59,17 @@ pub trait ParallelIteratorExt: ParallelIterator {
Err(err) => Err(Arc::into_inner(err).expect("the error must be only owned by us")),
}
}
fn try_arc_for_each<F, E>(self, op: F) -> Result<(), E>
where
E: Send + Sync,
F: Fn(Self::Item) -> Result<(), Arc<E>> + Sync + Send + Clone,
{
match self.try_for_each(op) {
Ok(()) => Ok(()),
Err(err) => Err(Arc::into_inner(err).expect("the error must be only owned by us")),
}
}
}
impl<T: ParallelIterator> ParallelIteratorExt for T {}