)]}'
{"openstack/object_store/v1/_proxy.py":[{"author":{"_account_id":8257,"name":"Brian Curtin","email":"brian@python.org","username":"briancurtin"},"change_message_id":"66d97138bbb84d0d80f66a5a807b3518aa21eb16","unresolved":false,"context_lines":[{"line_number":159,"context_line":"        return self._get(_obj.Object, object,"},{"line_number":160,"context_line":"                         path_args\u003d{\"container\": container_name})"},{"line_number":161,"context_line":""},{"line_number":162,"context_line":"    def download_object(self, object, container\u003dNone, path\u003dNone):"},{"line_number":163,"context_line":"        \"\"\"Download the data contained inside an object to disk."},{"line_number":164,"context_line":""},{"line_number":165,"context_line":"        :param object: The value can be the name of an object or a"}],"source_content_type":"text/x-python","patch_set":1,"id":"9a68dd71_db134a5b","line":162,"updated":"2016-01-21 02:10:17.000000000","message":"Can path go first? It feels slightly unconventional that way, but it\u0027s surely not an optional parameter so I think it\u0027s better if we flip it to the front.","commit_id":"848450abb6569187d811631cb7dd5ed1fba4f547"},{"author":{"_account_id":1112,"name":"Everett Toews","email":"everett.toews@rackspace.com","username":"everett-toews"},"change_message_id":"b7598e794e6fa459408aacc17a02a3fb57341b5f","unresolved":false,"context_lines":[{"line_number":159,"context_line":"        return self._get(_obj.Object, object,"},{"line_number":160,"context_line":"                         path_args\u003d{\"container\": container_name})"},{"line_number":161,"context_line":""},{"line_number":162,"context_line":"    def download_object(self, object, container\u003dNone, path\u003dNone):"},{"line_number":163,"context_line":"        \"\"\"Download the data contained inside an object to disk."},{"line_number":164,"context_line":""},{"line_number":165,"context_line":"        :param object: The value can be the name of an object or a"}],"source_content_type":"text/x-python","patch_set":1,"id":"9a68dd71_ff4f8731","line":162,"in_reply_to":"9a68dd71_db134a5b","updated":"2016-01-24 21:08:02.000000000","message":"Ya. That\u0027s the most painful part of this patch. For download_object/get_object to have a consistent API, container must come first. But that comes at the cost of have the required path appear to be \"optional\" at the end.\n\nI honestly think having a consistent API is the lesser of the two evils. It\u0027s just too surprising to have have such similar methods have different APIs. \n\nHaving path appear optional but be required is a one time annoyance for devs. They get an error, fix it, and move on. The inconsistent API will forever be a annoyance because there\u0027s nothing a dev can do to \"fix\" it.\n\nI could probably improve the docstrings to make this more clear.","commit_id":"848450abb6569187d811631cb7dd5ed1fba4f547"}]}