Merge pull request #1283 from Kiraprint/add-params2get_document_route

Add new params for get_document route

Author: sanders41

meilisearch_python_sdk/index.py

@@ -1288,12 +1288,22 @@ async def search_similar_documents(
 
         return SimilarSearchResults[self.hits_type](**response.json())  # type: ignore[name-defined]
 
-    async def get_document(self, document_id: str) -> JsonDict:
+    async def get_document(
+        self,
+        document_id: str,
+        *,
+        fields: list[str] | None = None,
+        retrieve_vectors: bool = False,
+    ) -> JsonDict:
         """Get one document with given document identifier.
 
         Args:
             document_id: Unique identifier of the document.
-
+            fields: Document attributes to show. If this value is None then all
+                attributes are retrieved. Defaults to None.
+            retrieve_vectors: If set to True the embedding vectors will be returned with the document.
+                Defaults to False. Note: This parameter can only be
+                used with Meilisearch >= v1.13.0
         Returns:
             The document information
 
@@ -1307,7 +1317,16 @@ async def get_document(self, document_id: str) -> JsonDict:
             >>>     index = client.index("movies")
             >>>     document = await index.get_document("1234")
         """
-        response = await self._http_requests.get(f"{self._documents_url}/{document_id}")
+        parameters: JsonDict = {}
+
+        if fields:
+            parameters["fields"] = ",".join(fields)
+        if retrieve_vectors:
+            parameters["retrieveVectors"] = "true"
+
+        url = _build_encoded_url(f"{self._documents_url}/{document_id}", parameters)
+
+        response = await self._http_requests.get(url)
 
         return response.json()
 
@@ -1318,6 +1337,7 @@ async def get_documents(
         limit: int = 20,
         fields: list[str] | None = None,
         filter: Filter | None = None,
+        retrieve_vectors: bool = False,
     ) -> DocumentsInfo:
         """Get a batch documents from the index.
 
@@ -1328,6 +1348,9 @@ async def get_documents(
                 attributes are retrieved. Defaults to None.
             filter: Filter value information. Defaults to None. Note: This parameter can only be
                 used with Meilisearch >= v1.2.0
+            retrieve_vectors: If set to True the vectors will be returned with each document.
+                Defaults to False. Note: This parameter can only be
+                used with Meilisearch >= v1.13.0
 
         Returns:
             Documents info.
@@ -1348,6 +1371,9 @@ async def get_documents(
             "limit": limit,
         }
 
+        if retrieve_vectors:
+            parameters["retrieveVectors"] = "true"
+
         if not filter:
             if fields:
                 parameters["fields"] = ",".join(fields)
@@ -5470,12 +5496,22 @@ def search_similar_documents(
 
         return SimilarSearchResults[self.hits_type](**response.json())  # type: ignore[name-defined]
 
-    def get_document(self, document_id: str) -> JsonDict:
+    def get_document(
+        self,
+        document_id: str,
+        *,
+        fields: list[str] | None = None,
+        retrieve_vectors: bool = False,
+    ) -> JsonDict:
         """Get one document with given document identifier.
 
         Args:
             document_id: Unique identifier of the document.
-
+            fields: Document attributes to show. If this value is None then all
+                attributes are retrieved. Defaults to None.
+            retrieve_vectors: If set to True the embedding vectors will be returned with the document.
+                Defaults to False. Note: This parameter can only be
+                used with Meilisearch >= v1.13.0
         Returns:
             The document information
 
@@ -5489,8 +5525,16 @@ def get_document(self, document_id: str) -> JsonDict:
             >>> index = client.index("movies")
             >>> document = index.get_document("1234")
         """
-        response = self._http_requests.get(f"{self._documents_url}/{document_id}")
+        parameters: JsonDict = {}
+
+        if fields:
+            parameters["fields"] = ",".join(fields)
+        if retrieve_vectors:
+            parameters["retrieveVectors"] = "true"
 
+        url = _build_encoded_url(f"{self._documents_url}/{document_id}", parameters)
+
+        response = self._http_requests.get(url)
         return response.json()
 
     def get_documents(
@@ -5500,6 +5544,7 @@ def get_documents(
         limit: int = 20,
         fields: list[str] | None = None,
         filter: Filter | None = None,
+        retrieve_vectors: bool = False,
     ) -> DocumentsInfo:
         """Get a batch documents from the index.
 
@@ -5510,6 +5555,9 @@ def get_documents(
                 attributes are retrieved. Defaults to None.
             filter: Filter value information. Defaults to None. Note: This parameter can only be
                 used with Meilisearch >= v1.2.0
+            retrieve_vectors: If set to True the vectors will be returned with each document.
+                Defaults to False. Note: This parameter can only be
+                used with Meilisearch >= v1.13.0
 
         Returns:
             Documents info.
@@ -5530,6 +5578,9 @@ def get_documents(
             "limit": limit,
         }
 
+        if retrieve_vectors:
+            parameters["retrieveVectors"] = "true"
+
         if not filter:
             if fields:
                 parameters["fields"] = ",".join(fields)

tests/test_async_documents.py

@@ -706,6 +706,20 @@ async def test_get_document_inexistent(async_empty_index):
         await index.get_document("123")
 
 
+async def test_get_document_with_fields(async_index_with_documents):
+    index = await async_index_with_documents()
+    response = await index.get_document("500682", fields=["title", "overview"])
+    assert len(response.keys()) == 2
+    assert "title" in response.keys()
+    assert "overview" in response.keys()
+
+
+async def test_get_document_with_vectors(async_index_with_documents):
+    index = await async_index_with_documents()
+    response = await index.get_document("500682", retrieve_vectors=True)
+    assert "_vectors" in response.keys()
+
+
 async def test_get_documents_populated(async_index_with_documents):
     index = await async_index_with_documents()
     response = await index.get_documents()
@@ -744,6 +758,14 @@ async def test_get_documents_filter_with_fields(async_index_with_documents):
     assert next(iter(genres)) == "action"
 
 
+async def test_get_documents_with_vectors(async_index_with_documents):
+    index = await async_index_with_documents()
+    response = await index.update_filterable_attributes(["genre"])
+    await async_wait_for_task(index.http_client, response.task_uid)
+    response = await index.get_documents(retrieve_vectors=True)
+    assert all("_vectors" in x.keys() for x in response.results)
+
+
 @pytest.mark.parametrize("compress", (True, False))
 async def test_update_documents(compress, async_index_with_documents, small_movies):
     index = await async_index_with_documents()

tests/test_documents.py

@@ -626,6 +626,20 @@ def test_get_document_inexistent(empty_index):
         index.get_document("123")
 
 
+def test_get_document_with_fields(index_with_documents):
+    index = index_with_documents()
+    response = index.get_document("500682", fields=["title", "overview"])
+    assert len(response.keys()) == 2
+    assert "title" in response.keys()
+    assert "overview" in response.keys()
+
+
+def test_get_document_with_vectors(index_with_documents):
+    index = index_with_documents()
+    response = index.get_document("500682", retrieve_vectors=True)
+    assert "_vectors" in response.keys()
+
+
 def test_get_documents_populated(index_with_documents):
     index = index_with_documents()
     response = index.get_documents()
@@ -652,6 +666,14 @@ def test_get_documents_filter(index_with_documents):
     assert next(iter(genres)) == "action"
 
 
+def test_get_documents_with_vectors(index_with_documents):
+    index = index_with_documents()
+    response = index.update_filterable_attributes(["genre"])
+    wait_for_task(index.http_client, response.task_uid)
+    response = index.get_documents(retrieve_vectors=True)
+    assert all("_vectors" in x.keys() for x in response.results)
+
+
 def test_get_documents_filter_with_fields(index_with_documents):
     index = index_with_documents()
     response = index.update_filterable_attributes(["genre"])