fix: make DocList.to_json() return str instead of bytes (#1769)

JohannesMessner · web-flow · commit 3dc525f46d8a · 2023-09-01T14:31:28.000+02:00
Signed-off-by: Johannes Messner &lt;messnerjo@gmail.com&gt;
diff --git a/docarray/array/doc_list/io.py b/docarray/array/doc_list/io.py
@@ -183,7 +183,7 @@ def _write_bytes(
             elif protocol == 'pickle-array':
                 f.write(pickle.dumps(self))
             elif protocol == 'json-array':
-                f.write(self.to_json())
+                f.write(self.to_json().encode())
             elif protocol in SINGLE_PROTOCOLS:
                 f.write(
                     b''.join(
@@ -327,11 +327,11 @@ def from_json(
         json_docs = orjson.loads(file)
         return cls([cls.doc_type(**v) for v in json_docs])
 
-    def to_json(self) -> bytes:
+    def to_json(self) -> str:
         """Convert the object into JSON bytes. Can be loaded via `.from_json`.
         :return: JSON serialization of `DocList`
         """
-        return orjson_dumps(self)
+        return orjson_dumps(self).decode('UTF-8')
 
     @classmethod
     def from_csv(
diff --git a/docs/user_guide/sending/serialization.md b/docs/user_guide/sending/serialization.md
@@ -75,15 +75,15 @@ dl = DocList[SimpleDoc]([SimpleDoc(text=f'doc {i}') for i in range(2)])
 with open('simple-dl.json', 'wb') as f:
     json_dl = dl.to_json()
     print(json_dl)
-    f.write(json_dl)
+    f.write(json_dl.encode())
 
 with open('simple-dl.json', 'r') as f:
     dl_load_from_json = DocList[SimpleDoc].from_json(f.read())
     print(dl_load_from_json)
 ```
 
 ```output
-b'[{"id":"5540e72d407ae81abb2390e9249ed066","text":"doc 0"},{"id":"fbe9f80d2fa03571e899a2887af1ac1b","text":"doc 1"}]'
+'[{"id":"5540e72d407ae81abb2390e9249ed066","text":"doc 0"},{"id":"fbe9f80d2fa03571e899a2887af1ac1b","text":"doc 1"}]'
 ```
 
 ### Protobuf
@@ -277,15 +277,15 @@ dv = DocVec[SimpleDoc](
 with open('simple-dv.json', 'wb') as f:
     json_dv = dv.to_json()
     print(json_dv)
-    f.write(json_dv)
+    f.write(json_dv.encode())
 
 with open('simple-dv.json', 'r') as f:
     dv_load_from_json = DocVec[SimpleDoc].from_json(f.read(), tensor_type=TorchTensor)
     print(dv_load_from_json)
 ```
 
 ```output
-b'{"tensor_columns":{},"doc_columns":{},"docs_vec_columns":{},"any_columns":{"id":["005a208a0a9a368c16bf77913b710433","31d65f02cb94fc9756c57b0dbaac3a2c"],"text":["doc 0","doc 1"]}}'
+'{"tensor_columns":{},"doc_columns":{},"docs_vec_columns":{},"any_columns":{"id":["005a208a0a9a368c16bf77913b710433","31d65f02cb94fc9756c57b0dbaac3a2c"],"text":["doc 0","doc 1"]}}'
 <DocVec[SimpleDoc] (length=2)>
 ```
 
diff --git a/tests/units/array/test_array_from_to_json.py b/tests/units/array/test_array_from_to_json.py
@@ -78,9 +78,9 @@ def _rand_vec_gen(tensor_type):
         return vec
 
     v = generate_docs(tensor_type)
-    bytes_ = v.to_json()
+    json_str = v.to_json()
 
-    v_after = DocVec[v.doc_type].from_json(bytes_, tensor_type=tensor_type)
+    v_after = DocVec[v.doc_type].from_json(json_str, tensor_type=tensor_type)
 
     assert v_after.tensor_type == v.tensor_type
     assert set(v_after._storage.columns.keys()) == set(v._storage.columns.keys())
@@ -125,9 +125,9 @@ class MyDoc(BaseDoc):
         return vec
 
     v = generate_docs()
-    bytes_ = v.to_json()
+    json_str = v.to_json()
 
-    v_after = DocVec[v.doc_type].from_json(bytes_, tensor_type=TensorFlowTensor)
+    v_after = DocVec[v.doc_type].from_json(json_str, tensor_type=TensorFlowTensor)
 
     assert v_after.tensor_type == v.tensor_type
     assert set(v_after._storage.columns.keys()) == set(v._storage.columns.keys())
