searchengines-web/MeiliSearch
searchengines-web/MeiliSearch
1
0
Fork 0
mirror of https://github.com/meilisearch/MeiliSearch synced 2024-11-30 08:44:27 +01:00
Code Issues Releases Wiki Activity

Remove the useless option from the document changes

Browse Source
This commit is contained in:
Clément Renault 2024-09-02 15:21:00 +02:00
parent 521775f788
commit ab01679a8f
No known key found for this signature in database
GPG Key ID: F250A4C4E3AE5F5F
5 changed files with 98 additions and 96 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
milli/src/update/new/indexer/document_deletion.rs
View File

@ -28,7 +28,7 @@ impl<'p> DocumentChanges<'p> for DocumentDeletion {
fn document_changes(
self,
param: Self::Parameter,
) -> Result<impl ParallelIterator<Item = Result<Option<DocumentChange>>> + 'p> {
) -> Result<impl ParallelIterator<Item = Result<DocumentChange>> + 'p> {
let (index, fields, primary_key) = param;
let items = Arc::new(ItemsPool::new(|| index.read_txn().map_err(crate::Error::from)));
Ok(self.to_delete.into_iter().par_bridge().map_with(items, |items, docid| {
@ -42,11 +42,11 @@ impl<'p> DocumentChanges<'p> for DocumentDeletion {
.into()),
}?;
Ok(Some(DocumentChange::Deletion(Deletion::create(
Ok(DocumentChange::Deletion(Deletion::create(
docid,
external_docid,
current.boxed(),
))))
)))
})
}))
}

119
milli/src/update/new/indexer/document_operation.rs
View File

@ -76,7 +76,7 @@ impl<'p> DocumentChanges<'p> for DocumentOperation {
fn document_changes(
self,
param: Self::Parameter,
) -> Result<impl ParallelIterator<Item = Result<Option<DocumentChange>>> + 'p> {
) -> Result<impl ParallelIterator<Item = Result<DocumentChange>> + 'p> {
let (index, rtxn, fields_ids_map, primary_key) = param;
let documents_ids = index.documents_ids(rtxn)?;
@ -170,7 +170,9 @@ impl<'p> DocumentChanges<'p> for DocumentOperation {
}
}
Ok(docids_version_offsets.into_par_iter().map_with(
Ok(docids_version_offsets
.into_par_iter()
.map_with(
Arc::new(ItemsPool::new(|| index.read_txn().map_err(crate::Error::from))),
move |context_pool, (external_docid, (internal_docid, operations))| {
context_pool.with(|rtxn| {
@ -190,7 +192,63 @@ impl<'p> DocumentChanges<'p> for DocumentOperation {
)
})
},
))
)
.filter_map(Result::transpose))
}
}
/// Returns only the most recent version of a document based on the updates from the payloads.
///
/// This function is only meant to be used when doing a replacement and not an update.
fn merge_document_for_replacements(
rtxn: &RoTxn,
index: &Index,
fields_ids_map: &FieldsIdsMap,
docid: DocumentId,
external_docid: String,
operations: &[InnerDocOp],
) -> Result<Option<DocumentChange>> {
let current = index.documents.remap_data_type::<Bytes>().get(rtxn, &docid)?;
let current: Option<&KvReaderFieldId> = current.map(Into::into);
match operations.last() {
Some(InnerDocOp::Addition(DocumentOffset { content, offset })) => {
let reader = DocumentsBatchReader::from_reader(Cursor::new(content.as_ref()))?;
let (mut cursor, batch_index) = reader.into_cursor_and_fields_index();
let update = cursor.get(*offset)?.expect("must exists");
let mut document_entries = Vec::new();
update.into_iter().for_each(|(k, v)| {
let field_name = batch_index.name(k).unwrap();
let id = fields_ids_map.id(field_name).unwrap();
document_entries.push((id, v));
});
document_entries.sort_unstable_by_key(|(id, _)| *id);
let mut writer = KvWriterFieldId::memory();
document_entries.into_iter().for_each(|(id, value)| writer.insert(id, value).unwrap());
let new = writer.into_boxed();
match current {
Some(current) => {
let update = Update::create(docid, external_docid, current.boxed(), new);
Ok(Some(DocumentChange::Update(update)))
}
None => {
let insertion = Insertion::create(docid, external_docid, new);
Ok(Some(DocumentChange::Insertion(insertion)))
}
}
}
Some(InnerDocOp::Deletion) => match current {
Some(current) => {
let deletion = Deletion::create(docid, external_docid, current.boxed());
Ok(Some(DocumentChange::Deletion(deletion)))
}
None => Ok(None),
},
None => Ok(None), // but it's strange
}
}
@ -271,58 +329,3 @@ fn merge_document_for_updates(
}
}
}
/// Returns only the most recent version of a document based on the updates from the payloads.
///
/// This function is only meant to be used when doing a replacement and not an update.
fn merge_document_for_replacements(
rtxn: &RoTxn,
index: &Index,
fields_ids_map: &FieldsIdsMap,
docid: DocumentId,
external_docid: String,
operations: &[InnerDocOp],
) -> Result<Option<DocumentChange>> {
let current = index.documents.remap_data_type::<Bytes>().get(rtxn, &docid)?;
let current: Option<&KvReaderFieldId> = current.map(Into::into);
match operations.last() {
Some(InnerDocOp::Addition(DocumentOffset { content, offset })) => {
let reader = DocumentsBatchReader::from_reader(Cursor::new(content.as_ref()))?;
let (mut cursor, batch_index) = reader.into_cursor_and_fields_index();
let update = cursor.get(*offset)?.expect("must exists");
let mut document_entries = Vec::new();
update.into_iter().for_each(|(k, v)| {
let field_name = batch_index.name(k).unwrap();
let id = fields_ids_map.id(field_name).unwrap();
document_entries.push((id, v));
});
document_entries.sort_unstable_by_key(|(id, _)| *id);
let mut writer = KvWriterFieldId::memory();
document_entries.into_iter().for_each(|(id, value)| writer.insert(id, value).unwrap());
let new = writer.into_boxed();
match current {
Some(current) => {
let update = Update::create(docid, external_docid, current.boxed(), new);
Ok(Some(DocumentChange::Update(update)))
}
None => {
let insertion = Insertion::create(docid, external_docid, new);
Ok(Some(DocumentChange::Insertion(insertion)))
}
}
}
Some(InnerDocOp::Deletion) => match current {
Some(current) => {
let deletion = Deletion::create(docid, external_docid, current.boxed());
Ok(Some(DocumentChange::Deletion(deletion)))
}
None => Ok(None),
},
None => Ok(None), // but it's strange
}
}

11
milli/src/update/new/indexer/mod.rs
View File

@ -1,4 +1,4 @@
use std::thread;
use std::thread::{self, Builder};
use big_s::S;
pub use document_deletion::DocumentDeletion;
@ -28,7 +28,7 @@ pub trait DocumentChanges<'p> {
fn document_changes(
self,
param: Self::Parameter,
) -> Result<impl ParallelIterator<Item = Result<Option<DocumentChange>>> + 'p>;
) -> Result<impl ParallelIterator<Item = Result<DocumentChange>> + 'p>;
}
/// This is the main function of this crate.
@ -43,7 +43,7 @@ pub fn index<PI>(
document_changes: PI,
) -> Result<()>
where
PI: IntoParallelIterator<Item = Result<Option<DocumentChange>>> + Send,
PI: IntoParallelIterator<Item = Result<DocumentChange>> + Send,
PI::Iter: Clone,
{
let (merger_sender, writer_receiver) = merger_writer_channel(100);
@ -52,8 +52,7 @@ where
thread::scope(|s| {
// TODO manage the errors correctly
let handle =
thread::Builder::new().name(S("indexer-extractors")).spawn_scoped(s, || {
let handle = Builder::new().name(S("indexer-extractors")).spawn_scoped(s, || {
pool.in_place_scope(|_s| {
// word docids
// document_changes.into_par_iter().try_for_each(|_dc| Ok(()) as Result<_>)
@ -65,7 +64,7 @@ where
})?;
// TODO manage the errors correctly
let handle2 = thread::Builder::new().name(S("indexer-merger")).spawn_scoped(s, || {
let handle2 = Builder::new().name(S("indexer-merger")).spawn_scoped(s, || {
let rtxn = index.read_txn().unwrap();
merge_grenad_entries(merger_receiver, merger_sender, &rtxn, index)
})?;

4
milli/src/update/new/indexer/partial_dump.rs
View File

@ -31,7 +31,7 @@ where
fn document_changes(
self,
param: Self::Parameter,
) -> Result<impl ParallelIterator<Item = Result<Option<DocumentChange>>> + 'p> {
) -> Result<impl ParallelIterator<Item = Result<DocumentChange>> + 'p> {
let (fields_ids_map, concurrent_available_ids, primary_key) = param;
Ok(self.iter.into_iter().par_bridge().map(|object| {
@ -68,7 +68,7 @@ where
}?;
let insertion = Insertion::create(docid, external_docid, document);
Ok(Some(DocumentChange::Insertion(insertion)))
Ok(DocumentChange::Insertion(insertion))
}))
}
}

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

@ -12,7 +12,7 @@ impl<'p> DocumentChanges<'p> for UpdateByFunction {
fn document_changes(
self,
_param: Self::Parameter,
) -> Result<impl ParallelIterator<Item = Result<Option<DocumentChange>>> + 'p> {
) -> Result<impl ParallelIterator<Item = Result<DocumentChange>> + 'p> {
todo!();
Ok(vec![].into_par_iter())
}