Introduce a parameter to disable the engine to autogenerate docids

src/update/index_documents/mod.rs

@@ -202,6 +202,7 @@ pub struct IndexDocuments<'t, 'u, 'i> {
     indexing_jobs: Option<usize>,
     update_method: IndexDocumentsMethod,
     update_format: UpdateFormat,
+    autogenerate_docids: bool,
 }
 
 impl<'t, 'u, 'i> IndexDocuments<'t, 'u, 'i> {
@@ -219,6 +220,7 @@ impl<'t, 'u, 'i> IndexDocuments<'t, 'u, 'i> {
             indexing_jobs: None,
             update_method: IndexDocumentsMethod::ReplaceDocuments,
             update_format: UpdateFormat::Json,
+            autogenerate_docids: true,
         }
     }
 
@@ -272,6 +274,16 @@ impl<'t, 'u, 'i> IndexDocuments<'t, 'u, 'i> {
         self
     }
 
+    pub fn enable_autogenerate_docids(&mut self) -> &mut Self {
+        self.autogenerate_docids = true;
+        self
+    }
+
+    pub fn disable_autogenerate_docids(&mut self) -> &mut Self {
+        self.autogenerate_docids = false;
+        self
+    }
+
     pub fn execute<R, F>(self, reader: R, progress_callback: F) -> anyhow::Result<()>
     where
         R: io::Read,
@@ -288,6 +300,7 @@ impl<'t, 'u, 'i> IndexDocuments<'t, 'u, 'i> {
             max_nb_chunks: self.max_nb_chunks,
             max_memory: self.max_memory,
             index_documents_method: self.update_method,
+            autogenerate_docids: self.autogenerate_docids,
         };
 
         let output = match self.update_format {
@@ -636,6 +649,56 @@ mod tests {
         drop(rtxn);
     }
 
+    #[test]
+    fn not_auto_generated_csv_documents_ids() {
+        let path = tempfile::tempdir().unwrap();
+        let mut options = EnvOpenOptions::new();
+        options.map_size(10 * 1024 * 1024); // 10 MB
+        let index = Index::new(options, &path).unwrap();
+
+        // First we send 3 documents with ids from 1 to 3.
+        let mut wtxn = index.write_txn().unwrap();
+        let content = &b"name\nkevin\nkevina\nbenoit\n"[..];
+        let mut builder = IndexDocuments::new(&mut wtxn, &index);
+        builder.disable_autogenerate_docids();
+        builder.update_format(UpdateFormat::Csv);
+        assert!(builder.execute(content, |_, _| ()).is_err());
+        wtxn.commit().unwrap();
+
+        // Check that there is no document.
+        let rtxn = index.read_txn().unwrap();
+        let count = index.number_of_documents(&rtxn).unwrap();
+        assert_eq!(count, 0);
+        drop(rtxn);
+    }
+
+    #[test]
+    fn not_auto_generated_json_documents_ids() {
+        let path = tempfile::tempdir().unwrap();
+        let mut options = EnvOpenOptions::new();
+        options.map_size(10 * 1024 * 1024); // 10 MB
+        let index = Index::new(options, &path).unwrap();
+
+        // First we send 3 documents and 2 without ids.
+        let mut wtxn = index.write_txn().unwrap();
+        let content = &br#"[
+            { "name": "kevina", "id": 21 },
+            { "name": "kevin" },
+            { "name": "benoit" }
+        ]"#[..];
+        let mut builder = IndexDocuments::new(&mut wtxn, &index);
+        builder.disable_autogenerate_docids();
+        builder.update_format(UpdateFormat::Json);
+        assert!(builder.execute(content, |_, _| ()).is_err());
+        wtxn.commit().unwrap();
+
+        // Check that there is no document.
+        let rtxn = index.read_txn().unwrap();
+        let count = index.number_of_documents(&rtxn).unwrap();
+        assert_eq!(count, 0);
+        drop(rtxn);
+    }
+
     #[test]
     fn simple_auto_generated_documents_ids() {
         let path = tempfile::tempdir().unwrap();

src/update/index_documents/transform.rs

@@ -33,6 +33,7 @@ pub struct Transform<'t, 'i> {
     pub max_nb_chunks: Option<usize>,
     pub max_memory: Option<usize>,
     pub index_documents_method: IndexDocumentsMethod,
+    pub autogenerate_docids: bool,
 }
 
 impl Transform<'_, '_> {
@@ -57,7 +58,14 @@ impl Transform<'_, '_> {
             None => {
                 match documents.get(0).and_then(|doc| doc.keys().find(|k| k.contains("id"))) {
                     Some(key) => fields_ids_map.insert(&key).context("field id limit reached")?,
-                    None => fields_ids_map.insert("id").context("field id limit reached")?,
+                    None => {
+                        if !self.autogenerate_docids {
+                            // If there is no primary key in the current document batch, we must
+                            // return an error and not automatically generate any document id.
+                            return Err(anyhow!("missing primary key"))
+                        }
+                        fields_ids_map.insert("id").context("field id limit reached")?
+                    },
                 }
             },
         };
@@ -130,6 +138,9 @@ impl Transform<'_, '_> {
                     _ => return Err(anyhow!("documents ids must be either strings or numbers")),
                 },
                 None => {
+                    if !self.autogenerate_docids {
+                        return Err(anyhow!("missing primary key"));
+                    }
                     let uuid = uuid::Uuid::new_v4().to_hyphenated().encode_lower(&mut uuid_buffer);
                     Cow::Borrowed(uuid)
                 },
@@ -180,11 +191,16 @@ impl Transform<'_, '_> {
         let primary_key_field_id = match user_id_pos {
             Some(pos) => fields_ids_map.id(&headers[pos]).expect("found the primary key"),
             None => {
-                let id = fields_ids_map.insert("id").context("field id limit reached")?;
+                if !self.autogenerate_docids {
+                    // If there is no primary key in the current document batch, we must
+                    // return an error and not automatically generate any document id.
+                    return Err(anyhow!("missing primary key"))
+                }
+                let field_id = fields_ids_map.insert("id").context("field id limit reached")?;
                 // We make sure to add the primary key field id to the fields ids,
                 // this way it is added to the obks.
-                fields_ids.push((id, usize::max_value()));
-                id
+                fields_ids.push((field_id, usize::max_value()));
+                field_id
             },
         };