Fix the errors when using the try_map_try_init method

milli/src/update/new/extract/searchable/extract_word_docids.rs

@@ -1,6 +1,7 @@
 use std::collections::HashMap;
 use std::fs::File;
 use std::num::NonZero;
+use std::sync::Arc;
 
 use grenad::{Merger, MergerBuilder};
 use heed::RoTxn;
@@ -12,7 +13,7 @@ use crate::update::new::extract::perm_json_p::contained_in;
 use crate::update::new::{DocumentChange, ItemsPool};
 use crate::update::{create_sorter, GrenadParameters, MergeDeladdCboRoaringBitmaps};
 use crate::{
-    bucketed_position, DocumentId, FieldId, GlobalFieldsIdsMap, Index, Result,
+    bucketed_position, DocumentId, Error, FieldId, GlobalFieldsIdsMap, Index, Result,
     MAX_POSITION_PER_ATTRIBUTE,
 };
 
@@ -303,7 +304,9 @@ impl WordDocidsExtractors {
         index: &Index,
         fields_ids_map: &GlobalFieldsIdsMap,
         indexer: GrenadParameters,
-        document_changes: impl IntoParallelIterator<Item = Result<DocumentChange>>,
+        document_changes: impl IntoParallelIterator<
+            Item = std::result::Result<DocumentChange, Arc<Error>>,
+        >,
     ) -> Result<WordDocidsMergers> {
         let max_memory = indexer.max_memory_by_thread();
 

milli/src/update/new/indexer/partial_dump.rs

@@ -1,8 +1,11 @@
-use rayon::iter::IndexedParallelIterator;
+use std::sync::Arc;
+
+use rayon::iter::{IndexedParallelIterator, ParallelBridge, ParallelIterator};
 
 use super::DocumentChanges;
 use crate::documents::{DocumentIdExtractionError, PrimaryKey};
 use crate::update::concurrent_available_ids::ConcurrentAvailableIds;
+use crate::update::new::items_pool::ParallelIteratorExt;
 use crate::update::new::{DocumentChange, Insertion, KvWriterFieldId};
 use crate::{all_obkv_to_json, Error, FieldsIdsMap, Object, Result, UserError};
 
@@ -37,7 +40,9 @@ where
     > {
         let (fields_ids_map, concurrent_available_ids, primary_key) = param;
 
-        Ok(self.iter.map(|object| {
+        Ok(self.iter.try_map_try_init(
+            || Ok(()),
+            |_, object| {
                 let docid = match concurrent_available_ids.next() {
                     Some(id) => id,
                     None => return Err(Error::UserError(UserError::DocumentLimitReached)),
@@ -55,7 +60,9 @@ where
                 let document = writer.into_boxed();
                 let external_docid = match primary_key.document_id(&document, fields_ids_map)? {
                     Ok(document_id) => Ok(document_id),
-                Err(DocumentIdExtractionError::InvalidDocumentId(user_error)) => Err(user_error),
+                    Err(DocumentIdExtractionError::InvalidDocumentId(user_error)) => {
+                        Err(user_error)
+                    }
                     Err(DocumentIdExtractionError::MissingDocumentId) => {
                         Err(UserError::MissingDocumentId {
                             primary_key: primary_key.name().to_string(),
@@ -72,6 +79,7 @@ where
 
                 let insertion = Insertion::create(docid, document);
                 Ok(DocumentChange::Insertion(insertion))
-        }))
+            },
+        ))
     }
 }

milli/src/update/new/items_pool.rs

@@ -4,7 +4,9 @@ use crossbeam_channel::{Receiver, Sender, TryRecvError};
 use rayon::iter::{MapInit, ParallelIterator};
 
 pub trait ParallelIteratorExt: ParallelIterator {
-    /// A method on a parallel iterator to map
+    /// Maps items based on the init function.
+    ///
+    /// The init function is ran only as necessary which is basically once by thread.
     fn try_map_try_init<F, INIT, T, E, R>(
         self,
         init: INIT,