move the comments for schema generation settings to the settings docs page

Author: ghazi-git

docs/settings.md

@@ -19,5 +19,46 @@ DRF_STANDARDIZED_ERRORS = {
     # {field}{NESTED_FIELD_SEPARATOR}{nested_field}
     # for example: 'shipping_address.zipcode'
     "NESTED_FIELD_SEPARATOR": ".",
+
+    # The below settings are for OpenAPI 3 schema generation
+
+    # ONLY the responses that correspond to these status codes will appear
+    # in the API schema.
+    "ALLOWED_ERROR_STATUS_CODES": [
+        "400",
+        "401",
+        "403",
+        "404",
+        "405",
+        "406",
+        "415",
+        "429",
+        "500",
+    ],
+
+    # A mapping used to override the default serializers used to describe
+    # the error response. The key is the status code and the value is a string
+    # that represents the path to the serializer class that describes the
+    # error response.
+    "ERROR_SCHEMAS": None,
+
+    # When there is a validation error in list serializers, the "attr" returned
+    # will be sth like "0.email", "1.email", "2.email", ... So, to describe
+    # the error codes linked to the same field in a list serializer, the field
+    # will appear in the schema with the name "INDEX.email"
+    "LIST_INDEX_IN_API_SCHEMA": "INDEX",
+
+    # When there is a validation error in a DictField with the name "extra_data",
+    # the "attr" returned will be sth like "extra_data.<key1>", "extra_data.<key2>",
+    # "extra_data.<key3>", ... Since the keys of a DictField are not predetermined,
+    # this setting is used as a common name to be used in the API schema. So, the
+    # corresponding "attr" value for the previous example will be "extra_data.KEY"
+    "DICT_KEY_IN_API_SCHEMA": "KEY",
+
+    # should be unique to error components since it is used to identify error
+    # components generated dynamically to exclude them from being processed by
+    # the postprocessing hook. This avoids raising warnings for "code" and "attr"
+    # which can have the same choices across multiple serializers.
+    "ERROR_COMPONENT_NAME_SUFFIX": "ErrorComponent",
 }
 ```

drf_standardized_errors/settings.py

@@ -76,26 +76,9 @@ def reload(self):
         "429",
         "500",
     ],
-    # A mapping used to override the default serializers used to describe
-    # the error response. The key is the status code and the value is a string
-    # that represents the path to the serializer class that describes the
-    # error response.
     "ERROR_SCHEMAS": None,
-    # When there is a validation error in list serializers, the "attr" returned
-    # will be sth like "0.email", "1.email", "2.email", ... So, to describe
-    # the error codes linked to the same field in a list serializer, the field
-    # will appear in the schema with the name "INDEX.email"
     "LIST_INDEX_IN_API_SCHEMA": "INDEX",
-    # When there is a validation error in a DictField with the name "extra_data",
-    # the "attr" returned will be sth like "extra_data.<key1>", "extra_data.<key2>",
-    # "extra_data.<key3>", ... Since the keys of a DictField are not predetermined,
-    # this setting is used as a common name to be used in the API schema. So, the
-    # corresponding "attr" value for the previous example will be "extra_data.KEY"
     "DICT_KEY_IN_API_SCHEMA": "KEY",
-    # should be unique to error components since it is used to identify error
-    # components generated dynamically to exclude them from being processed by
-    # the postprocessing hook. This avoids raising warnings for "code" and "attr"
-    # which can have the same choices across multiple serializers.
     "ERROR_COMPONENT_NAME_SUFFIX": "ErrorComponent",
 }