Merge pull request #193 from meilisearch/get-documents-id

Add a method to get an iterator over all documents ids

meilidb-data/src/database/index/documents_index.rs

@@ -55,6 +55,11 @@ impl DocumentsIndex {
         Ok(DocumentFieldsIter(iter))
     }
 
+    pub fn documents_ids(&self) -> RocksDbResult<DocumentsIdsIter> {
+        let iter = DocumentsKeysIter(self.0.iter()?);
+        Ok(DocumentsIdsIter { inner: iter, last: None })
+    }
+
     pub fn documents_fields_repartition(&self, schema: Schema) -> RocksDbResult<HashMap<String, u64>> {
         let iter = self.0.iter()?;
         let mut repartition_attributes_id = HashMap::new();
@@ -120,3 +125,22 @@ impl Iterator for DocumentsKeysIter<'_> {
         }
     }
 }
+
+pub struct DocumentsIdsIter<'a> {
+    inner: DocumentsKeysIter<'a>,
+    last: Option<DocumentId>,
+}
+
+impl Iterator for DocumentsIdsIter<'_> {
+    type Item = DocumentId;
+
+    fn next(&mut self) -> Option<Self::Item> {
+        for DocumentAttrKey { document_id, .. } in &mut self.inner {
+            if self.last != Some(document_id) {
+                self.last = Some(document_id);
+                return Some(document_id)
+            }
+        }
+        None
+    }
+}

meilidb-data/src/database/index/mod.rs

@@ -19,6 +19,7 @@ use crate::serde::{Deserializer, DeserializerError};
 
 pub use self::custom_settings_index::{CustomSettingsIndex, RankingOrdering, StopWords, RankingOrder, DistinctField, RankingRules};
 pub use self::common_index::CommonIndex;
+pub use self::documents_index::DocumentsIdsIter;
 use self::docs_words_index::DocsWordsIndex;
 use self::documents_index::DocumentsIndex;
 use self::main_index::MainIndex;
@@ -374,6 +375,10 @@ impl Index {
         Ok(self.update_status(update_id)?.unwrap())
     }
 
+    pub fn documents_ids(&self) -> Result<DocumentsIdsIter, Error> {
+        Ok(self.documents_index.documents_ids()?)
+    }
+
     pub fn document<T>(
         &self,
         fields: Option<&HashSet<&str>>,

meilidb-data/tests/updates.rs

@@ -189,3 +189,27 @@ fn custom_settings() {
     assert_eq!(ret_distinct_field, distinct_field);
     assert_eq!(ret_ranking_rules, ranking_rules);
 }
+
+#[test]
+fn documents_ids() {
+    let tmp_dir = tempfile::tempdir().unwrap();
+    let database = Database::open(&tmp_dir).unwrap();
+
+    let schema = simple_schema();
+    let index = database.create_index("hello", schema).unwrap();
+
+    let doc1 = json!({ "objectId": 123, "title": "hello" });
+    let doc2 = json!({ "objectId": 456, "title": "world" });
+    let doc3 = json!({ "objectId": 789 });
+
+    let mut addition = index.documents_addition();
+    addition.update_document(&doc1);
+    addition.update_document(&doc2);
+    addition.update_document(&doc3);
+    let update_id = addition.finalize().unwrap();
+    let status = index.update_status_blocking(update_id).unwrap();
+    assert!(status.result.is_ok());
+
+    let documents_ids_count = index.documents_ids().unwrap().count();
+    assert_eq!(documents_ids_count, 3);
+}