explicitly set utf8 for Windows (#17664)

utils/check_repo.py

@@ -725,7 +725,7 @@ def check_docstrings_are_in_md():
     """Check all docstrings are in md"""
     files_with_rst = []
     for file in Path(PATH_TO_TRANSFORMERS).glob("**/*.py"):
-        with open(file, "r") as f:
+        with open(file, encoding="utf-8") as f:
             code = f.read()
         docstrings = code.split('"""')
 

utils/custom_init_isort.py

@@ -167,7 +167,7 @@ def sort_imports(file, check_only=True):
     """
     Sort `_import_structure` imports in `file`, `check_only` determines if we only check or overwrite.
     """
-    with open(file, "r") as f:
+    with open(file, encoding="utf-8") as f:
         code = f.read()
 
     if "_import_structure" not in code:
@@ -227,7 +227,7 @@ def sort_imports(file, check_only=True):
             return True
         else:
             print(f"Overwriting {file}.")
-            with open(file, "w") as f:
+            with open(file, "w", encoding="utf-8") as f:
                 f.write("\n".join(main_blocks))
 
 

utils/notification_service_doc_tests.py

@@ -289,7 +289,7 @@ def retrieve_artifact(name: str):
         files = os.listdir(name)
         for file in files:
             try:
-                with open(os.path.join(name, file)) as f:
+                with open(os.path.join(name, file), encoding="utf-8") as f:
                     _artifact[file.split(".")[0]] = f.read()
             except UnicodeDecodeError as e:
                 raise ValueError(f"Could not open {os.path.join(name, file)}.") from e