From 7af27f2932b4b66477dd0b86c4c8eb56293f7331 Mon Sep 17 00:00:00 2001
From: awais qureshi <awais.qureshi@arbisoft.com>
Date: Thu, 1 Jan 2026 18:19:39 +0500
Subject: [PATCH 1/8] feat: Adds an optional skipCreation. When set to true,
 documents that don't exist in the index are silently ignored rather than
 created.

---
 meilisearch/index.py | 77 ++++++++++++++++++++++++++++++++++++--------
 1 file changed, 64 insertions(+), 13 deletions(-)

diff --git a/meilisearch/index.py b/meilisearch/index.py
index cfbff045..388ff6b7 100644
--- a/meilisearch/index.py
+++ b/meilisearch/index.py
@@ -454,6 +454,7 @@ def add_documents(
         primary_key: Optional[str] = None,
         *,
         serializer: Optional[Type[JSONEncoder]] = None,
+        skip_creation: Optional[bool] = None,
     ) -> TaskInfo:
         """Add documents to the index.
 
@@ -466,6 +467,9 @@ def add_documents(
         serializer (optional):
             A custom JSONEncode to handle serializing fields that the build in json.dumps
             cannot handle, for example UUID and datetime.
+        skip_creation (optional):
+            If True, documents that don't exist in the index are silently ignored rather
+            than created. Default is False, preserving existing behavior.
 
         Returns
         -------
@@ -478,7 +482,7 @@ def add_documents(
         MeilisearchApiError
             An error containing details about why Meilisearch can't process your request. Meilisearch error codes are described here: https://www.meilisearch.com/docs/reference/errors/error_codes#meilisearch-errors
         """
-        url = self._build_url(primary_key)
+        url = self._build_url(primary_key, skip_creation=skip_creation)
         add_document_task = self.http.post(url, documents, serializer=serializer)
         return TaskInfo(**add_document_task)
 
@@ -489,6 +493,7 @@ def add_documents_in_batches(
         primary_key: Optional[str] = None,
         *,
         serializer: Optional[Type[JSONEncoder]] = None,
+        skip_creation: Optional[bool] = None,
     ) -> List[TaskInfo]:
         """Add documents to the index in batches.
 
@@ -503,6 +508,9 @@ def add_documents_in_batches(
         serializer (optional):
             A custom JSONEncode to handle serializing fields that the build in json.dumps
             cannot handle, for example UUID and datetime.
+        skip_creation (optional):
+            If True, documents that don't exist in the index are silently ignored rather
+            than created. Default is False, preserving existing behavior.
 
         Returns
         -------
@@ -520,7 +528,7 @@ def add_documents_in_batches(
         tasks: List[TaskInfo] = []
 
         for document_batch in self._batch(documents, batch_size):
-            task = self.add_documents(document_batch, primary_key, serializer=serializer)
+            task = self.add_documents(document_batch, primary_key, serializer=serializer, skip_creation=skip_creation)
             tasks.append(task)
 
         return tasks
@@ -531,6 +539,7 @@ def add_documents_json(
         primary_key: Optional[str] = None,
         *,
         serializer: Optional[Type[JSONEncoder]] = None,
+        skip_creation: Optional[bool] = None,
     ) -> TaskInfo:
         """Add documents to the index from a byte-encoded JSON string.
 
@@ -543,6 +552,9 @@ def add_documents_json(
         serializer (optional):
             A custom JSONEncode to handle serializing fields that the build in json.dumps
             cannot handle, for example UUID and datetime.
+        skip_creation (optional):
+            If True, documents that don't exist in the index are silently ignored rather
+            than created. Default is False, preserving existing behavior.
 
         Returns
         -------
@@ -556,7 +568,7 @@ def add_documents_json(
             An error containing details about why Meilisearch can't process your request. Meilisearch error codes are described here: https://www.meilisearch.com/docs/reference/errors/error_codes#meilisearch-errors
         """
         return self.add_documents_raw(
-            str_documents, primary_key, "application/json", serializer=serializer
+            str_documents, primary_key, "application/json", serializer=serializer, skip_creation=skip_creation
         )
 
     def add_documents_csv(
@@ -564,6 +576,7 @@ def add_documents_csv(
         str_documents: bytes,
         primary_key: Optional[str] = None,
         csv_delimiter: Optional[str] = None,
+        skip_creation: Optional[bool] = None,
     ) -> TaskInfo:
         """Add documents to the index from a byte-encoded CSV string.
 
@@ -575,6 +588,9 @@ def add_documents_csv(
             The primary-key used in index. Ignored if already set up.
         csv_delimiter:
             One ASCII character used to customize the delimiter for CSV. Comma used by default.
+        skip_creation (optional):
+            If True, documents that don't exist in the index are silently ignored rather
+            than created. Default is False, preserving existing behavior.
 
         Returns
         -------
@@ -587,12 +603,13 @@ def add_documents_csv(
         MeilisearchApiError
             An error containing details about why Meilisearch can't process your request. Meilisearch error codes are described here: https://www.meilisearch.com/docs/reference/errors/error_codes#meilisearch-errors
         """
-        return self.add_documents_raw(str_documents, primary_key, "text/csv", csv_delimiter)
+        return self.add_documents_raw(str_documents, primary_key, "text/csv", csv_delimiter, skip_creation=skip_creation)
 
     def add_documents_ndjson(
         self,
         str_documents: bytes,
         primary_key: Optional[str] = None,
+        skip_creation: Optional[bool] = None,
     ) -> TaskInfo:
         """Add documents to the index from a byte-encoded NDJSON string.
 
@@ -602,6 +619,9 @@ def add_documents_ndjson(
             Byte-encoded NDJSON string.
         primary_key (optional):
             The primary-key used in index. Ignored if already set up.
+        skip_creation (optional):
+            If True, documents that don't exist in the index are silently ignored rather
+            than created. Default is False, preserving existing behavior.
 
         Returns
         -------
@@ -614,7 +634,7 @@ def add_documents_ndjson(
         MeilisearchApiError
             An error containing details about why Meilisearch can't process your request. Meilisearch error codes are described here: https://www.meilisearch.com/docs/reference/errors/error_codes#meilisearch-errors
         """
-        return self.add_documents_raw(str_documents, primary_key, "application/x-ndjson")
+        return self.add_documents_raw(str_documents, primary_key, "application/x-ndjson", skip_creation=skip_creation)
 
     def add_documents_raw(
         self,
@@ -624,6 +644,7 @@ def add_documents_raw(
         csv_delimiter: Optional[str] = None,
         *,
         serializer: Optional[Type[JSONEncoder]] = None,
+        skip_creation: Optional[bool] = None,
     ) -> TaskInfo:
         """Add documents to the index from a byte-encoded string.
 
@@ -641,6 +662,9 @@ def add_documents_raw(
         serializer (optional):
             A custom JSONEncode to handle serializing fields that the build in json.dumps
             cannot handle, for example UUID and datetime.
+        skip_creation (optional):
+            If True, documents that don't exist in the index are silently ignored rather
+            than created. Default is False, preserving existing behavior.
 
         Returns
         -------
@@ -653,7 +677,7 @@ def add_documents_raw(
         MeilisearchApiError
             An error containing details about why Meilisearch can't process your request. Meilisearch error codes are described here: https://www.meilisearch.com/docs/reference/errors/error_codes#meilisearch-errors
         """
-        url = self._build_url(primary_key=primary_key, csv_delimiter=csv_delimiter)
+        url = self._build_url(primary_key=primary_key, csv_delimiter=csv_delimiter, skip_creation=skip_creation)
         response = self.http.post(url, str_documents, content_type, serializer=serializer)
         return TaskInfo(**response)
 
@@ -663,6 +687,7 @@ def update_documents(
         primary_key: Optional[str] = None,
         *,
         serializer: Optional[Type[JSONEncoder]] = None,
+        skip_creation: Optional[bool] = None,
     ) -> TaskInfo:
         """Update documents in the index.
 
@@ -675,6 +700,9 @@ def update_documents(
         serializer (optional):
             A custom JSONEncode to handle serializing fields that the build in json.dumps
             cannot handle, for example UUID and datetime.
+        skip_creation (optional):
+            If True, documents that don't exist in the index are silently ignored rather
+            than created. Default is False, preserving existing behavior.
 
         Returns
         -------
@@ -687,7 +715,7 @@ def update_documents(
         MeilisearchApiError
             An error containing details about why Meilisearch can't process your request. Meilisearch error codes are described here: https://www.meilisearch.com/docs/reference/errors/error_codes#meilisearch-errors
         """
-        url = self._build_url(primary_key)
+        url = self._build_url(primary_key, skip_creation=skip_creation)
         response = self.http.put(url, documents, serializer=serializer)
         return TaskInfo(**response)
 
@@ -695,6 +723,7 @@ def update_documents_ndjson(
         self,
         str_documents: str,
         primary_key: Optional[str] = None,
+        skip_creation: Optional[bool] = None,
     ) -> TaskInfo:
         """Update documents as a ndjson string in the index.
 
@@ -704,6 +733,9 @@ def update_documents_ndjson(
             String of document from a NDJSON file.
         primary_key (optional):
             The primary-key used in index. Ignored if already set up
+        skip_creation (optional):
+            If True, documents that don't exist in the index are silently ignored rather
+            than created. Default is False, preserving existing behavior.
 
         Returns
         -------
@@ -716,7 +748,7 @@ def update_documents_ndjson(
         MeilisearchApiError
             An error containing details about why Meilisearch can't process your request. Meilisearch error codes are described here: https://www.meilisearch.com/docs/reference/errors/error_codes#meilisearch-errors
         """
-        return self.update_documents_raw(str_documents, primary_key, "application/x-ndjson")
+        return self.update_documents_raw(str_documents, primary_key, "application/x-ndjson", skip_creation=skip_creation)
 
     def update_documents_json(
         self,
@@ -724,6 +756,7 @@ def update_documents_json(
         primary_key: Optional[str] = None,
         *,
         serializer: Optional[Type[JSONEncoder]] = None,
+        skip_creation: Optional[bool] = None,
     ) -> TaskInfo:
         """Update documents as a json string in the index.
 
@@ -736,6 +769,9 @@ def update_documents_json(
         serializer (optional):
             A custom JSONEncode to handle serializing fields that the build in json.dumps
             cannot handle, for example UUID and datetime.
+        skip_creation (optional):
+            If True, documents that don't exist in the index are silently ignored rather
+            than created. Default is False, preserving existing behavior.
 
         Returns
         -------
@@ -749,7 +785,7 @@ def update_documents_json(
             An error containing details about why Meilisearch can't process your request. Meilisearch error codes are described here: https://www.meilisearch.com/docs/reference/errors/error_codes#meilisearch-errors
         """
         return self.update_documents_raw(
-            str_documents, primary_key, "application/json", serializer=serializer
+            str_documents, primary_key, "application/json", serializer=serializer, skip_creation=skip_creation
         )
 
     def update_documents_csv(
@@ -757,6 +793,7 @@ def update_documents_csv(
         str_documents: str,
         primary_key: Optional[str] = None,
         csv_delimiter: Optional[str] = None,
+        skip_creation: Optional[bool] = None,
     ) -> TaskInfo:
         """Update documents as a csv string in the index.
 
@@ -768,6 +805,9 @@ def update_documents_csv(
             The primary-key used in index. Ignored if already set up.
         csv_delimiter:
             One ASCII character used to customize the delimiter for CSV. Comma used by default.
+        skip_creation (optional):
+            If True, documents that don't exist in the index are silently ignored rather
+            than created. Default is False, preserving existing behavior.
 
         Returns
         -------
@@ -780,7 +820,7 @@ def update_documents_csv(
         MeilisearchApiError
             An error containing details about why Meilisearch can't process your request. Meilisearch error codes are described here: https://www.meilisearch.com/docs/reference/errors/error_codes#meilisearch-errors
         """
-        return self.update_documents_raw(str_documents, primary_key, "text/csv", csv_delimiter)
+        return self.update_documents_raw(str_documents, primary_key, "text/csv", csv_delimiter, skip_creation=skip_creation)
 
     def update_documents_raw(
         self,
@@ -790,6 +830,7 @@ def update_documents_raw(
         csv_delimiter: Optional[str] = None,
         *,
         serializer: Optional[Type[JSONEncoder]] = None,
+        skip_creation: Optional[bool] = None,
     ) -> TaskInfo:
         """Update documents as a string in the index.
 
@@ -807,6 +848,9 @@ def update_documents_raw(
         serializer (optional):
             A custom JSONEncode to handle serializing fields that the build in json.dumps
             cannot handle, for example UUID and datetime.
+        skip_creation (optional):
+            If True, documents that don't exist in the index are silently ignored rather
+            than created. Default is False, preserving existing behavior.
 
         Returns
         -------
@@ -819,7 +863,7 @@ def update_documents_raw(
         MeilisearchApiError
             An error containing details about why Meilisearch can't process your request. Meilisearch error codes are described here: https://www.meilisearch.com/docs/reference/errors/error_codes#meilisearch-errors
         """
-        url = self._build_url(primary_key=primary_key, csv_delimiter=csv_delimiter)
+        url = self._build_url(primary_key=primary_key, csv_delimiter=csv_delimiter, skip_creation=skip_creation)
         response = self.http.put(url, str_documents, content_type, serializer=serializer)
         return TaskInfo(**response)
 
@@ -829,6 +873,7 @@ def update_documents_in_batches(
         batch_size: int = 1000,
         primary_key: Optional[str] = None,
         serializer: Optional[Type[JSONEncoder]] = None,
+        skip_creation: Optional[bool] = None,
     ) -> List[TaskInfo]:
         """Update documents to the index in batches.
 
@@ -843,6 +888,9 @@ def update_documents_in_batches(
         serializer (optional):
             A custom JSONEncode to handle serializing fields that the build in json.dumps
             cannot handle, for example UUID and datetime.
+        skip_creation (optional):
+            If True, documents that don't exist in the index are silently ignored rather
+            than created. Default is False, preserving existing behavior.
 
         Returns
         -------
@@ -860,7 +908,7 @@ def update_documents_in_batches(
         tasks = []
 
         for document_batch in self._batch(documents, batch_size):
-            update_task = self.update_documents(document_batch, primary_key, serializer=serializer)
+            update_task = self.update_documents(document_batch, primary_key, serializer=serializer, skip_creation=skip_creation)
             tasks.append(update_task)
 
         return tasks
@@ -2331,13 +2379,16 @@ def _build_url(
         self,
         primary_key: Optional[str] = None,
         csv_delimiter: Optional[str] = None,
+        skip_creation: Optional[bool] = None,
     ) -> str:
         parameters = {}
         if primary_key:
             parameters["primaryKey"] = primary_key
         if csv_delimiter:
             parameters["csvDelimiter"] = csv_delimiter
-        if primary_key is None and csv_delimiter is None:
+        if skip_creation is True:
+            parameters["skipCreation"] = "true"
+        if not parameters:
             return f"{self.config.paths.index}/{self.uid}/{self.config.paths.document}"
         return f"{self.config.paths.index}/{self.uid}/{self.config.paths.document}?{parse.urlencode(parameters)}"
 

From 917838045e8703540ccb72f0b4ac094c40c8b8dd Mon Sep 17 00:00:00 2001
From: awais qureshi <awais.qureshi@arbisoft.com>
Date: Thu, 1 Jan 2026 18:27:22 +0500
Subject: [PATCH 2/8] feat: Adds an optional skipCreation. When set to true,
 documents that don't exist in the index are silently ignored rather than
 created.

---
 meilisearch/index.py | 44 ++++++++++++++++++++++++++++++++++----------
 1 file changed, 34 insertions(+), 10 deletions(-)

diff --git a/meilisearch/index.py b/meilisearch/index.py
index 388ff6b7..5325a125 100644
--- a/meilisearch/index.py
+++ b/meilisearch/index.py
@@ -528,7 +528,9 @@ def add_documents_in_batches(
         tasks: List[TaskInfo] = []
 
         for document_batch in self._batch(documents, batch_size):
-            task = self.add_documents(document_batch, primary_key, serializer=serializer, skip_creation=skip_creation)
+            task = self.add_documents(
+                document_batch, primary_key, serializer=serializer, skip_creation=skip_creation
+            )
             tasks.append(task)
 
         return tasks
@@ -568,7 +570,11 @@ def add_documents_json(
             An error containing details about why Meilisearch can't process your request. Meilisearch error codes are described here: https://www.meilisearch.com/docs/reference/errors/error_codes#meilisearch-errors
         """
         return self.add_documents_raw(
-            str_documents, primary_key, "application/json", serializer=serializer, skip_creation=skip_creation
+            str_documents,
+            primary_key,
+            "application/json",
+            serializer=serializer,
+            skip_creation=skip_creation,
         )
 
     def add_documents_csv(
@@ -603,7 +609,9 @@ def add_documents_csv(
         MeilisearchApiError
             An error containing details about why Meilisearch can't process your request. Meilisearch error codes are described here: https://www.meilisearch.com/docs/reference/errors/error_codes#meilisearch-errors
         """
-        return self.add_documents_raw(str_documents, primary_key, "text/csv", csv_delimiter, skip_creation=skip_creation)
+        return self.add_documents_raw(
+            str_documents, primary_key, "text/csv", csv_delimiter, skip_creation=skip_creation
+        )
 
     def add_documents_ndjson(
         self,
@@ -634,7 +642,9 @@ def add_documents_ndjson(
         MeilisearchApiError
             An error containing details about why Meilisearch can't process your request. Meilisearch error codes are described here: https://www.meilisearch.com/docs/reference/errors/error_codes#meilisearch-errors
         """
-        return self.add_documents_raw(str_documents, primary_key, "application/x-ndjson", skip_creation=skip_creation)
+        return self.add_documents_raw(
+            str_documents, primary_key, "application/x-ndjson", skip_creation=skip_creation
+        )
 
     def add_documents_raw(
         self,
@@ -677,7 +687,9 @@ def add_documents_raw(
         MeilisearchApiError
             An error containing details about why Meilisearch can't process your request. Meilisearch error codes are described here: https://www.meilisearch.com/docs/reference/errors/error_codes#meilisearch-errors
         """
-        url = self._build_url(primary_key=primary_key, csv_delimiter=csv_delimiter, skip_creation=skip_creation)
+        url = self._build_url(
+            primary_key=primary_key, csv_delimiter=csv_delimiter, skip_creation=skip_creation
+        )
         response = self.http.post(url, str_documents, content_type, serializer=serializer)
         return TaskInfo(**response)
 
@@ -748,7 +760,9 @@ def update_documents_ndjson(
         MeilisearchApiError
             An error containing details about why Meilisearch can't process your request. Meilisearch error codes are described here: https://www.meilisearch.com/docs/reference/errors/error_codes#meilisearch-errors
         """
-        return self.update_documents_raw(str_documents, primary_key, "application/x-ndjson", skip_creation=skip_creation)
+        return self.update_documents_raw(
+            str_documents, primary_key, "application/x-ndjson", skip_creation=skip_creation
+        )
 
     def update_documents_json(
         self,
@@ -785,7 +799,11 @@ def update_documents_json(
             An error containing details about why Meilisearch can't process your request. Meilisearch error codes are described here: https://www.meilisearch.com/docs/reference/errors/error_codes#meilisearch-errors
         """
         return self.update_documents_raw(
-            str_documents, primary_key, "application/json", serializer=serializer, skip_creation=skip_creation
+            str_documents,
+            primary_key,
+            "application/json",
+            serializer=serializer,
+            skip_creation=skip_creation,
         )
 
     def update_documents_csv(
@@ -820,7 +838,9 @@ def update_documents_csv(
         MeilisearchApiError
             An error containing details about why Meilisearch can't process your request. Meilisearch error codes are described here: https://www.meilisearch.com/docs/reference/errors/error_codes#meilisearch-errors
         """
-        return self.update_documents_raw(str_documents, primary_key, "text/csv", csv_delimiter, skip_creation=skip_creation)
+        return self.update_documents_raw(
+            str_documents, primary_key, "text/csv", csv_delimiter, skip_creation=skip_creation
+        )
 
     def update_documents_raw(
         self,
@@ -863,7 +883,9 @@ def update_documents_raw(
         MeilisearchApiError
             An error containing details about why Meilisearch can't process your request. Meilisearch error codes are described here: https://www.meilisearch.com/docs/reference/errors/error_codes#meilisearch-errors
         """
-        url = self._build_url(primary_key=primary_key, csv_delimiter=csv_delimiter, skip_creation=skip_creation)
+        url = self._build_url(
+            primary_key=primary_key, csv_delimiter=csv_delimiter, skip_creation=skip_creation
+        )
         response = self.http.put(url, str_documents, content_type, serializer=serializer)
         return TaskInfo(**response)
 
@@ -908,7 +930,9 @@ def update_documents_in_batches(
         tasks = []
 
         for document_batch in self._batch(documents, batch_size):
-            update_task = self.update_documents(document_batch, primary_key, serializer=serializer, skip_creation=skip_creation)
+            update_task = self.update_documents(
+                document_batch, primary_key, serializer=serializer, skip_creation=skip_creation
+            )
             tasks.append(update_task)
 
         return tasks

From f0333d4727563d6ddbb1356861e2b2c98fe0f7e7 Mon Sep 17 00:00:00 2001
From: awais qureshi <awais.qureshi@arbisoft.com>
Date: Thu, 1 Jan 2026 18:41:27 +0500
Subject: [PATCH 3/8] feat: Adds an optional skipCreation. When set to true,
 documents that don't exist in the index are silently ignored rather than
 created.

---
 .../index/test_index_document_meilisearch.py  | 267 ++++++++++++++++++
 1 file changed, 267 insertions(+)

diff --git a/tests/index/test_index_document_meilisearch.py b/tests/index/test_index_document_meilisearch.py
index 35bc5f33..256d0e1d 100644
--- a/tests/index/test_index_document_meilisearch.py
+++ b/tests/index/test_index_document_meilisearch.py
@@ -555,3 +555,270 @@ def test_update_documents_ndjson(index_with_documents, songs_ndjson):
     task = index.wait_for_task(response.task_uid)
     assert task.status == "succeeded"
     assert index.get_primary_key() == "id"
+
+
+# Tests for skip_creation parameter
+def test_add_documents_with_skip_creation_true(empty_index):
+    """Tests that skip_creation=True prevents creation of new documents."""
+    index = empty_index()
+    documents = [
+        {"id": "1", "title": "Existing Document"},
+    ]
+
+    # First add a document normally
+    task = index.add_documents(documents)
+    index.wait_for_task(task.task_uid)
+
+    # Verify document exists
+    doc = index.get_document("1")
+    assert doc.title == "Existing Document"
+
+    # Now try to add a new document with skip_creation=True - should be ignored
+    new_documents = [
+        {"id": "2", "title": "New Document"},
+    ]
+    task = index.add_documents(new_documents, skip_creation=True)
+    index.wait_for_task(task.task_uid)
+
+    # New document should not exist
+    with pytest.raises(Exception):
+        index.get_document("2")
+
+    # Existing document should still be there
+    doc = index.get_document("1")
+    assert doc.title == "Existing Document"
+
+
+def test_add_documents_with_skip_creation_false(empty_index):
+    """Tests that skip_creation=False allows creation of new documents (default behavior)."""
+    index = empty_index()
+    documents = [
+        {"id": "1", "title": "New Document"},
+    ]
+
+    # Add document with skip_creation=False (should work same as default)
+    task = index.add_documents(documents, skip_creation=False)
+    index.wait_for_task(task.task_uid)
+
+    # Document should exist
+    doc = index.get_document("1")
+    assert doc.title == "New Document"
+
+
+def test_add_documents_skip_creation_updates_existing(empty_index):
+    """Tests that skip_creation=True still allows updating existing documents."""
+    index = empty_index()
+    documents = [
+        {"id": "1", "title": "Original Title"},
+    ]
+
+    # Add document initially
+    task = index.add_documents(documents)
+    index.wait_for_task(task.task_uid)
+
+    # Update with skip_creation=True - should update existing document
+    updated_documents = [
+        {"id": "1", "title": "Updated Title"},
+    ]
+    task = index.add_documents(updated_documents, skip_creation=True)
+    index.wait_for_task(task.task_uid)
+
+    # Document should be updated
+    doc = index.get_document("1")
+    assert doc.title == "Updated Title"
+
+
+def test_update_documents_with_skip_creation_true(empty_index):
+    """Tests that update_documents with skip_creation=True prevents creation of new documents."""
+    index = empty_index()
+    documents = [
+        {"id": "1", "title": "Existing Document"},
+    ]
+
+    # First add a document
+    task = index.add_documents(documents)
+    index.wait_for_task(task.task_uid)
+
+    # Now try to update with a new document - should be ignored
+    new_documents = [
+        {"id": "2", "title": "New Document"},
+    ]
+    task = index.update_documents(new_documents, skip_creation=True)
+    index.wait_for_task(task.task_uid)
+
+    # New document should not exist
+    with pytest.raises(Exception):
+        index.get_document("2")
+
+    # Existing document should still be there and unchanged
+    doc = index.get_document("1")
+    assert doc.title == "Existing Document"
+
+
+def test_update_documents_skip_creation_updates_existing(empty_index):
+    """Tests that update_documents with skip_creation=True still updates existing documents."""
+    index = empty_index()
+    documents = [
+        {"id": "1", "title": "Original Title"},
+    ]
+
+    # Add document initially
+    task = index.add_documents(documents)
+    index.wait_for_task(task.task_uid)
+
+    # Update with skip_creation=True - should update existing document
+    updated_documents = [
+        {"id": "1", "title": "Updated Title"},
+    ]
+    task = index.update_documents(updated_documents, skip_creation=True)
+    index.wait_for_task(task.task_uid)
+
+    # Document should be updated
+    doc = index.get_document("1")
+    assert doc.title == "Updated Title"
+
+
+def test_add_documents_in_batches_with_skip_creation(empty_index, small_movies):
+    """Tests that skip_creation parameter works with add_documents_in_batches."""
+    index = empty_index()
+
+    # Add some documents first
+    initial_docs = small_movies[:5]
+    task = index.add_documents(initial_docs)
+    index.wait_for_task(task.task_uid)
+
+    # Try to add more documents with skip_creation=True
+    new_docs = small_movies[5:10]
+    task = index.add_documents_in_batches(new_docs, batch_size=2, skip_creation=True)
+    assert isinstance(task, list)
+    for t in task:
+        index.wait_for_task(t.task_uid)
+
+    # Only original documents should exist
+    all_docs = index.get_documents().results
+    existing_ids = {doc.id for doc in all_docs}
+    original_ids = {doc["id"] for doc in initial_docs}
+    assert existing_ids == original_ids
+
+
+def test_update_documents_in_batches_with_skip_creation(empty_index, small_movies):
+    """Tests that skip_creation parameter works with update_documents_in_batches."""
+    index = empty_index()
+
+    # Add some documents first
+    initial_docs = small_movies[:5]
+    task = index.add_documents(initial_docs)
+    index.wait_for_task(task.task_uid)
+
+    # Try to update with new documents with skip_creation=True
+    new_docs = small_movies[5:10]
+    task = index.update_documents_in_batches(new_docs, batch_size=2, skip_creation=True)
+    assert isinstance(task, list)
+    for t in task:
+        index.wait_for_task(t.task_uid)
+
+    # Only original documents should exist
+    all_docs = index.get_documents().results
+    existing_ids = {doc.id for doc in all_docs}
+    original_ids = {doc["id"] for doc in initial_docs}
+    assert existing_ids == original_ids
+
+
+def test_add_documents_json_with_skip_creation(empty_index, small_movies_json_file):
+    """Tests that skip_creation parameter works with add_documents_json."""
+    import json
+
+    index = empty_index()
+    documents = json.loads(small_movies_json_file.decode("utf-8"))
+
+    # Add first document
+    first_doc = json.dumps([documents[0]]).encode("utf-8")
+    task = index.add_documents_json(first_doc)
+    index.wait_for_task(task.task_uid)
+
+    # Try to add new document with skip_creation=True
+    new_doc = json.dumps([documents[1]]).encode("utf-8")
+    task = index.add_documents_json(new_doc, skip_creation=True)
+    index.wait_for_task(task.task_uid)
+
+    # Only first document should exist
+    all_docs = index.get_documents().results
+    assert len(all_docs) == 1
+    assert all_docs[0].id == documents[0]["id"]
+
+
+def test_update_documents_json_with_skip_creation(empty_index, small_movies_json_file):
+    """Tests that skip_creation parameter works with update_documents_json."""
+    import json
+
+    index = empty_index()
+    documents = json.loads(small_movies_json_file.decode("utf-8"))
+
+    # Add first document
+    first_doc = json.dumps([documents[0]]).encode("utf-8")
+    task = index.add_documents(first_doc.decode("utf-8"))
+    index.wait_for_task(task.task_uid)
+
+    # Try to update with new document with skip_creation=True
+    new_doc = json.dumps([documents[1]]).encode("utf-8")
+    task = index.update_documents_json(new_doc.decode("utf-8"), skip_creation=True)
+    index.wait_for_task(task.task_uid)
+
+    # Only first document should exist
+    all_docs = index.get_documents().results
+    assert len(all_docs) == 1
+    assert all_docs[0].id == documents[0]["id"]
+
+
+def test_add_documents_csv_with_skip_creation(empty_index, songs_csv):
+    """Tests that skip_creation parameter works with add_documents_csv."""
+    index = empty_index()
+
+    # Add first few lines manually
+    first_line = songs_csv.split(b"\n")[0] + b"\n"
+    task = index.add_documents_csv(first_line)
+    index.wait_for_task(task.task_uid)
+
+    # Try to add more with skip_creation=True
+    task = index.add_documents_csv(songs_csv, skip_creation=True)
+    index.wait_for_task(task.task_uid)
+
+    # Only first document should exist
+    all_docs = index.get_documents().results
+    assert len(all_docs) == 1
+
+
+def test_update_documents_csv_with_skip_creation(empty_index, songs_csv):
+    """Tests that skip_creation parameter works with update_documents_csv."""
+    index = empty_index()
+
+    # Add first few lines manually
+    first_line = songs_csv.split(b"\n")[0] + b"\n"
+    task = index.add_documents_csv(first_line)
+    index.wait_for_task(task.task_uid)
+
+    # Try to update with more with skip_creation=True
+    task = index.update_documents_csv(songs_csv.decode("utf-8"), skip_creation=True)
+    index.wait_for_task(task.task_uid)
+
+    # Only first document should exist
+    all_docs = index.get_documents().results
+    assert len(all_docs) == 1
+
+
+def test_add_documents_ndjson_with_skip_creation(empty_index, songs_ndjson):
+    """Tests that skip_creation parameter works with add_documents_ndjson."""
+    index = empty_index()
+
+    # Add first line
+    first_line = songs_ndjson.split(b"\n")[0] + b"\n"
+    task = index.add_documents_ndjson(first_line)
+    index.wait_for_task(task.task_uid)
+
+    # Try to add more with skip_creation=True
+    task = index.add_documents_ndjson(songs_ndjson, skip_creation=True)
+    index.wait_for_task(task.task_uid)
+
+    # Only first document should exist
+    all_docs = index.get_documents().results
+    assert len(all_docs) == 1

From 0c3ee2a4fec68526c67e924b8aaac527c774db0f Mon Sep 17 00:00:00 2001
From: awais qureshi <awais.qureshi@arbisoft.com>
Date: Thu, 1 Jan 2026 18:49:12 +0500
Subject: [PATCH 4/8] feat: Adds an optional skipCreation. When set to true,
 documents that don't exist in the index are silently ignored rather than
 created.

---
 .../index/test_index_document_meilisearch.py  | 42 ++++++++++++-------
 1 file changed, 27 insertions(+), 15 deletions(-)

diff --git a/tests/index/test_index_document_meilisearch.py b/tests/index/test_index_document_meilisearch.py
index 256d0e1d..318d3144 100644
--- a/tests/index/test_index_document_meilisearch.py
+++ b/tests/index/test_index_document_meilisearch.py
@@ -754,14 +754,14 @@ def test_update_documents_json_with_skip_creation(empty_index, small_movies_json
     index = empty_index()
     documents = json.loads(small_movies_json_file.decode("utf-8"))
 
-    # Add first document
-    first_doc = json.dumps([documents[0]]).encode("utf-8")
-    task = index.add_documents(first_doc.decode("utf-8"))
+    # Add first document using add_documents (expects list of dicts)
+    task = index.add_documents([documents[0]])
     index.wait_for_task(task.task_uid)
 
     # Try to update with new document with skip_creation=True
-    new_doc = json.dumps([documents[1]]).encode("utf-8")
-    task = index.update_documents_json(new_doc.decode("utf-8"), skip_creation=True)
+    # update_documents_json expects a string (JSON encoded)
+    new_doc = json.dumps([documents[1]])
+    task = index.update_documents_json(new_doc, skip_creation=True)
     index.wait_for_task(task.task_uid)
 
     # Only first document should exist
@@ -774,36 +774,48 @@ def test_add_documents_csv_with_skip_creation(empty_index, songs_csv):
     """Tests that skip_creation parameter works with add_documents_csv."""
     index = empty_index()
 
-    # Add first few lines manually
-    first_line = songs_csv.split(b"\n")[0] + b"\n"
-    task = index.add_documents_csv(first_line)
+    # Add first two lines (header + first data row) manually
+    lines = songs_csv.split(b"\n")
+    first_two_lines = b"\n".join(lines[:2]) + b"\n"
+    task = index.add_documents_csv(first_two_lines)
     index.wait_for_task(task.task_uid)
 
+    # Verify first document exists
+    all_docs = index.get_documents().results
+    initial_count = len(all_docs)
+    assert initial_count == 1
+
     # Try to add more with skip_creation=True
     task = index.add_documents_csv(songs_csv, skip_creation=True)
     index.wait_for_task(task.task_uid)
 
-    # Only first document should exist
+    # Only first document should exist (no new ones created)
     all_docs = index.get_documents().results
-    assert len(all_docs) == 1
+    assert len(all_docs) == initial_count
 
 
 def test_update_documents_csv_with_skip_creation(empty_index, songs_csv):
     """Tests that skip_creation parameter works with update_documents_csv."""
     index = empty_index()
 
-    # Add first few lines manually
-    first_line = songs_csv.split(b"\n")[0] + b"\n"
-    task = index.add_documents_csv(first_line)
+    # Add first two lines (header + first data row) manually
+    lines = songs_csv.split(b"\n")
+    first_two_lines = b"\n".join(lines[:2]) + b"\n"
+    task = index.add_documents_csv(first_two_lines)
     index.wait_for_task(task.task_uid)
 
+    # Verify first document exists
+    all_docs = index.get_documents().results
+    initial_count = len(all_docs)
+    assert initial_count == 1
+
     # Try to update with more with skip_creation=True
     task = index.update_documents_csv(songs_csv.decode("utf-8"), skip_creation=True)
     index.wait_for_task(task.task_uid)
 
-    # Only first document should exist
+    # Only first document should exist (no new ones created)
     all_docs = index.get_documents().results
-    assert len(all_docs) == 1
+    assert len(all_docs) == initial_count
 
 
 def test_add_documents_ndjson_with_skip_creation(empty_index, songs_ndjson):

From f1778a9c068cc4797b8a3f80e2d00802696b39e2 Mon Sep 17 00:00:00 2001
From: awais qureshi <awais.qureshi@arbisoft.com>
Date: Thu, 1 Jan 2026 18:53:54 +0500
Subject: [PATCH 5/8] feat: Adds an optional skipCreation. When set to true,
 documents that don't exist in the index are silently ignored rather than
 created.

---
 tests/index/test_index_document_meilisearch.py | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/tests/index/test_index_document_meilisearch.py b/tests/index/test_index_document_meilisearch.py
index 318d3144..20c9142c 100644
--- a/tests/index/test_index_document_meilisearch.py
+++ b/tests/index/test_index_document_meilisearch.py
@@ -759,8 +759,9 @@ def test_update_documents_json_with_skip_creation(empty_index, small_movies_json
     index.wait_for_task(task.task_uid)
 
     # Try to update with new document with skip_creation=True
-    # update_documents_json expects a string (JSON encoded)
-    new_doc = json.dumps([documents[1]])
+    # update_documents_json accepts bytes (like the fixture) or list of dicts
+    # Create bytes like the fixture does
+    new_doc = json.dumps([documents[1]]).encode("utf-8")
     task = index.update_documents_json(new_doc, skip_creation=True)
     index.wait_for_task(task.task_uid)
 

From 805d0c326459d512bbd74a7dec6aca3270851500 Mon Sep 17 00:00:00 2001
From: awais qureshi <awais.qureshi@arbisoft.com>
Date: Thu, 1 Jan 2026 18:55:53 +0500
Subject: [PATCH 6/8] feat: Adds an optional skipCreation. When set to true,
 documents that don't exist in the index are silently ignored rather than
 created.

---
 .code-samples.meilisearch.yaml                 | 4 ++--
 tests/index/test_index_document_meilisearch.py | 5 +----
 2 files changed, 3 insertions(+), 6 deletions(-)

diff --git a/.code-samples.meilisearch.yaml b/.code-samples.meilisearch.yaml
index bf7062f3..d683624c 100644
--- a/.code-samples.meilisearch.yaml
+++ b/.code-samples.meilisearch.yaml
@@ -42,13 +42,13 @@ add_or_replace_documents_1: |-
     'poster': 'https://image.tmdb.org/t/p/w1280/xnopI5Xtky18MPhK40cZAGAOVeV.jpg',
     'overview': 'A boy is given the ability to become an adult superhero in times of need with a single magic word.',
     'release_date': '2019-03-23'
-  }])
+  }], skip_creation=True)
 add_or_update_documents_1: |-
   client.index('movies').update_documents([{
       'id': 287947,
       'title': 'Shazam ⚡️',
       'genres': 'comedy'
-  }])
+  }], skip_creation=True)
 delete_all_documents_1: |-
   client.index('movies').delete_all_documents()
 delete_one_document_1: |-
diff --git a/tests/index/test_index_document_meilisearch.py b/tests/index/test_index_document_meilisearch.py
index 20c9142c..92914f0d 100644
--- a/tests/index/test_index_document_meilisearch.py
+++ b/tests/index/test_index_document_meilisearch.py
@@ -1,5 +1,6 @@
 # pylint: disable=invalid-name
 
+import json
 from datetime import datetime
 from json import JSONEncoder
 from math import ceil
@@ -726,8 +727,6 @@ def test_update_documents_in_batches_with_skip_creation(empty_index, small_movie
 
 def test_add_documents_json_with_skip_creation(empty_index, small_movies_json_file):
     """Tests that skip_creation parameter works with add_documents_json."""
-    import json
-
     index = empty_index()
     documents = json.loads(small_movies_json_file.decode("utf-8"))
 
@@ -749,8 +748,6 @@ def test_add_documents_json_with_skip_creation(empty_index, small_movies_json_fi
 
 def test_update_documents_json_with_skip_creation(empty_index, small_movies_json_file):
     """Tests that skip_creation parameter works with update_documents_json."""
-    import json
-
     index = empty_index()
     documents = json.loads(small_movies_json_file.decode("utf-8"))
 

From 80001ac7cffd1ab2185c5e28b48513962f15f8ed Mon Sep 17 00:00:00 2001
From: awais qureshi <awais.qureshi@arbisoft.com>
Date: Thu, 1 Jan 2026 19:13:05 +0500
Subject: [PATCH 7/8] feat: Adds an optional skipCreation. When set to true,
 documents that don't exist in the index are silently ignored rather than
 created.

---
 meilisearch/index.py | 24 ++++++++++++------------
 1 file changed, 12 insertions(+), 12 deletions(-)

diff --git a/meilisearch/index.py b/meilisearch/index.py
index 5325a125..0b4a5a4d 100644
--- a/meilisearch/index.py
+++ b/meilisearch/index.py
@@ -469,7 +469,7 @@ def add_documents(
             cannot handle, for example UUID and datetime.
         skip_creation (optional):
             If True, documents that don't exist in the index are silently ignored rather
-            than created. Default is False, preserving existing behavior.
+            than created. If False or None (default), existing behavior is preserved.
 
         Returns
         -------
@@ -510,7 +510,7 @@ def add_documents_in_batches(
             cannot handle, for example UUID and datetime.
         skip_creation (optional):
             If True, documents that don't exist in the index are silently ignored rather
-            than created. Default is False, preserving existing behavior.
+            than created. If False or None (default), existing behavior is preserved.
 
         Returns
         -------
@@ -556,7 +556,7 @@ def add_documents_json(
             cannot handle, for example UUID and datetime.
         skip_creation (optional):
             If True, documents that don't exist in the index are silently ignored rather
-            than created. Default is False, preserving existing behavior.
+            than created. If False or None (default), existing behavior is preserved.
 
         Returns
         -------
@@ -596,7 +596,7 @@ def add_documents_csv(
             One ASCII character used to customize the delimiter for CSV. Comma used by default.
         skip_creation (optional):
             If True, documents that don't exist in the index are silently ignored rather
-            than created. Default is False, preserving existing behavior.
+            than created. If False or None (default), existing behavior is preserved.
 
         Returns
         -------
@@ -629,7 +629,7 @@ def add_documents_ndjson(
             The primary-key used in index. Ignored if already set up.
         skip_creation (optional):
             If True, documents that don't exist in the index are silently ignored rather
-            than created. Default is False, preserving existing behavior.
+            than created. If False or None (default), existing behavior is preserved.
 
         Returns
         -------
@@ -674,7 +674,7 @@ def add_documents_raw(
             cannot handle, for example UUID and datetime.
         skip_creation (optional):
             If True, documents that don't exist in the index are silently ignored rather
-            than created. Default is False, preserving existing behavior.
+            than created. If False or None (default), existing behavior is preserved.
 
         Returns
         -------
@@ -714,7 +714,7 @@ def update_documents(
             cannot handle, for example UUID and datetime.
         skip_creation (optional):
             If True, documents that don't exist in the index are silently ignored rather
-            than created. Default is False, preserving existing behavior.
+            than created. If False or None (default), existing behavior is preserved.
 
         Returns
         -------
@@ -747,7 +747,7 @@ def update_documents_ndjson(
             The primary-key used in index. Ignored if already set up
         skip_creation (optional):
             If True, documents that don't exist in the index are silently ignored rather
-            than created. Default is False, preserving existing behavior.
+            than created. If False or None (default), existing behavior is preserved.
 
         Returns
         -------
@@ -785,7 +785,7 @@ def update_documents_json(
             cannot handle, for example UUID and datetime.
         skip_creation (optional):
             If True, documents that don't exist in the index are silently ignored rather
-            than created. Default is False, preserving existing behavior.
+            than created. If False or None (default), existing behavior is preserved.
 
         Returns
         -------
@@ -825,7 +825,7 @@ def update_documents_csv(
             One ASCII character used to customize the delimiter for CSV. Comma used by default.
         skip_creation (optional):
             If True, documents that don't exist in the index are silently ignored rather
-            than created. Default is False, preserving existing behavior.
+            than created. If False or None (default), existing behavior is preserved.
 
         Returns
         -------
@@ -870,7 +870,7 @@ def update_documents_raw(
             cannot handle, for example UUID and datetime.
         skip_creation (optional):
             If True, documents that don't exist in the index are silently ignored rather
-            than created. Default is False, preserving existing behavior.
+            than created. If False or None (default), existing behavior is preserved.
 
         Returns
         -------
@@ -912,7 +912,7 @@ def update_documents_in_batches(
             cannot handle, for example UUID and datetime.
         skip_creation (optional):
             If True, documents that don't exist in the index are silently ignored rather
-            than created. Default is False, preserving existing behavior.
+            than created. If False or None (default), existing behavior is preserved.
 
         Returns
         -------

From 3c22c85dd6062bc8beb1ffe4c870b6d956c4c968 Mon Sep 17 00:00:00 2001
From: awais qureshi <awais.qureshi@arbisoft.com>
Date: Thu, 1 Jan 2026 19:18:07 +0500
Subject: [PATCH 8/8] feat: Adds an optional skipCreation. When set to true,
 documents that don't exist in the index are silently ignored rather than
 created.

---
 tests/index/test_index_document_meilisearch.py | 15 ++++++++-------
 1 file changed, 8 insertions(+), 7 deletions(-)

diff --git a/tests/index/test_index_document_meilisearch.py b/tests/index/test_index_document_meilisearch.py
index 92914f0d..485d1978 100644
--- a/tests/index/test_index_document_meilisearch.py
+++ b/tests/index/test_index_document_meilisearch.py
@@ -9,6 +9,7 @@
 
 import pytest
 
+from meilisearch.errors import MeilisearchApiError
 from meilisearch.models.document import Document
 from meilisearch.models.task import TaskInfo
 
@@ -207,7 +208,7 @@ def test_get_document_with_fields(index_with_documents):
 
 def test_get_document_inexistent(empty_index):
     """Tests getting one inexistent document from a populated index."""
-    with pytest.raises(Exception):
+    with pytest.raises(MeilisearchApiError):
         empty_index().get_document("123")
 
 
@@ -419,7 +420,7 @@ def test_delete_document(index_with_documents):
     assert isinstance(response, TaskInfo)
     assert response.task_uid is not None
     index.wait_for_task(response.task_uid)
-    with pytest.raises(Exception):
+    with pytest.raises(MeilisearchApiError):
         index.get_document("500682")
 
 
@@ -433,7 +434,7 @@ def test_delete_documents_by_id(index_with_documents):
         assert response.task_uid is not None
         index.wait_for_task(response.task_uid)
         for document in to_delete:
-            with pytest.raises(Exception):
+            with pytest.raises(MeilisearchApiError):
                 index.get_document(document)
         assert "The use of ids is depreciated" in str(w[0].message)
 
@@ -581,8 +582,8 @@ def test_add_documents_with_skip_creation_true(empty_index):
     task = index.add_documents(new_documents, skip_creation=True)
     index.wait_for_task(task.task_uid)
 
-    # New document should not exist
-    with pytest.raises(Exception):
+    # Document "2" should not exist because skip_creation=True prevents creation of new documents
+    with pytest.raises(MeilisearchApiError):
         index.get_document("2")
 
     # Existing document should still be there
@@ -647,8 +648,8 @@ def test_update_documents_with_skip_creation_true(empty_index):
     task = index.update_documents(new_documents, skip_creation=True)
     index.wait_for_task(task.task_uid)
 
-    # New document should not exist
-    with pytest.raises(Exception):
+    # Document "2" should not exist because skip_creation=True prevents creation of new documents
+    with pytest.raises(MeilisearchApiError):
         index.get_document("2")
 
     # Existing document should still be there and unchanged