Add limit parameter to client.data_object.get() REST query

integration/test_crud.py

@@ -3,7 +3,7 @@
 import time
 from datetime import datetime
 from datetime import timezone
-from typing import List
+from typing import List, Optional
 
 import pytest
 
@@ -78,6 +78,7 @@ def test_load_scheme(people_schema):
 @pytest.fixture(scope="module")
 def client(people_schema):
     client = weaviate.Client("http://localhost:8080")
+    client.schema.delete_all()
     client.schema.create(people_schema)
 
     yield client
@@ -104,6 +105,25 @@ def test_timeout(people_schema, timeout):
     client.schema.delete_all()
 
 
+@pytest.mark.parametrize("limit", [None, 1, 5, 20, 50])
+def test_query_get_with_limit(people_schema, limit: Optional[int]):
+    client = weaviate.Client("http://localhost:8080")
+    client.schema.delete_all()
+    client.schema.create(people_schema)
+
+    num_objects = 20
+    for i in range(num_objects):
+        with client.batch as batch:
+            batch.add_data_object({"name": f"name{i}"}, "Person")
+        batch.flush()
+    result = client.data_object.get(class_name="Person", limit=limit)
+    if limit is None or limit >= num_objects:
+        assert len(result["objects"]) == num_objects
+    else:
+        assert len(result["objects"]) == limit
+    client.schema.delete_all()
+
+
 def test_query_data(client):
     expected_name = "Sophie Scholl"
     client.data_object.create(

weaviate/batch/crud_batch.py

@@ -23,6 +23,7 @@
 from ..util import (
     _capitalize_first_letter,
     check_batch_result,
+    _check_positive_num,
 )
 from ..warnings import _Warnings
 
@@ -1521,33 +1522,6 @@ def connection_error_retries(self, value: int) -> None:
         self._connection_error_retries = value
 
 
-def _check_positive_num(value: Real, arg_name: str, data_type: type) -> None:
-    """
-    Check if the `value` of the `arg_name` is a positive number.
-
-    Parameters
-    ----------
-    value : Union[int, float]
-        The value to check.
-    arg_name : str
-        The name of the variable from the original function call. Used for error message.
-    data_type : type
-        The data type to check for.
-
-    Raises
-    ------
-    TypeError
-        If the `value` is not of type `data_type`.
-    ValueError
-        If the `value` has a non positive value.
-    """
-
-    if not isinstance(value, data_type) or isinstance(value, bool):
-        raise TypeError(f"'{arg_name}' must be of type {data_type}.")
-    if value <= 0:
-        raise ValueError(f"'{arg_name}' must be positive, i.e. greater that zero (>0).")
-
-
 def _check_non_negative(value: Real, arg_name: str, data_type: type) -> None:
     """
     Check if the `value` of the `arg_name` is a non-negative number.

weaviate/data/crud_data.py

@@ -19,6 +19,7 @@
     get_vector,
     get_valid_uuid,
     _capitalize_first_letter,
+    _check_positive_num,
 )
 
 
@@ -417,6 +418,7 @@ def get(
         additional_properties: List[str] = None,
         with_vector: bool = False,
         class_name: Optional[str] = None,
+        limit: Optional[int] = None,
     ) -> List[dict]:
         """
         Gets objects from weaviate, the maximum number of objects returned is 100.
@@ -433,11 +435,14 @@ def get(
         with_vector: bool
             If True the `vector` property will be returned too,
             by default False
-        class_name : Optional[str], optional
+        class_name: Optional[str], optional
             The class name of the object with UUID `uuid`. Introduced in Weaviate version v1.14.0.
             STRONGLY recommended to set it with Weaviate >= 1.14.0. It will be required in future
             versions of Weaviate Server and Clients. Use None value ONLY for Weaviate < v1.14.0,
             by default None
+        limit: Optional[int], optional
+            The maximum number of data objects to return.
+            by default None, which uses the weaviate default of 100 entries
 
         Returns
         -------
@@ -455,7 +460,6 @@ def get(
         weaviate.UnexpectedStatusCodeException
             If weaviate reports a none OK status.
         """
-
         is_server_version_14 = self._connection.server_version >= "1.14"
 
         if class_name is None and is_server_version_14 and uuid is not None:
@@ -488,6 +492,10 @@ def get(
         if uuid is not None:
             path += "/" + get_valid_uuid(uuid)
 
+        if limit is not None:
+            _check_positive_num(limit, "limit", int)
+            params["limit"] = limit
+
         try:
             response = self._connection.get(
                 path=path,

weaviate/util.py

@@ -6,6 +6,7 @@
 import os
 import uuid as uuid_lib
 from io import BufferedReader
+from numbers import Real
 from typing import Union, Sequence, Any, Optional
 
 import requests
@@ -488,3 +489,30 @@ def check_batch_result(results: dict) -> None:
             if "result" in result and "errors" in result["result"]:
                 if "error" in result["result"]["errors"]:
                     print(result["result"]["errors"])
+
+
+def _check_positive_num(value: Real, arg_name: str, data_type: type) -> None:
+    """
+    Check if the `value` of the `arg_name` is a positive number.
+
+    Parameters
+    ----------
+    value : Union[int, float]
+        The value to check.
+    arg_name : str
+        The name of the variable from the original function call. Used for error message.
+    data_type : type
+        The data type to check for.
+
+    Raises
+    ------
+    TypeError
+        If the `value` is not of type `data_type`.
+    ValueError
+        If the `value` has a non positive value.
+    """
+
+    if not isinstance(value, data_type) or isinstance(value, bool):
+        raise TypeError(f"'{arg_name}' must be of type {data_type}.")
+    if value <= 0:
+        raise ValueError(f"'{arg_name}' must be positive, i.e. greater that zero (>0).")