)]}'
{"swift/common/exceptions.py":[{"author":{"_account_id":15343,"name":"Tim Burke","email":"tburke@nvidia.com","username":"tburke"},"change_message_id":"093ca6f4c797b66ea55ee27ab00591720dc8ab04","unresolved":true,"context_lines":[{"line_number":20,"context_line":"    from eventlet import Timeout"},{"line_number":21,"context_line":"else:"},{"line_number":22,"context_line":"    class Timeout(BaseException):"},{"line_number":23,"context_line":"        \"\"\"Dummy eventlet.Timeout-like that should never get raised\"\"\""},{"line_number":24,"context_line":""},{"line_number":25,"context_line":"        def __init__(self, seconds\u003dNone, exception\u003dNone):"},{"line_number":26,"context_line":"            self.start()"}],"source_content_type":"text/x-python","patch_set":3,"id":"844859f5_525020bd","line":23,"range":{"start_line":23,"start_character":51,"end_line":23,"end_character":67},"updated":"2025-09-25 18:30:35.000000000","message":"OK, so *everywhere* in the codebase that we currently have something like `with Timeout(...):`, we need to start thinking about where/how we\u0027re going to plumb in proper timeouts, yeah? `grep -r \u0027with \\w*Timeout\u0027 swift | wc -l` gives me ~80 which... I guess isn\u0027t *so* bad?\n\nBut until we get that plumbing in, most/all code will wait indefinitely, right?\n\nI wonder how many places we\u0027ll find we use timeouts in ways that the plumbing available to us doesn\u0027t expect. I remember [a while back](https://github.com/eventlet/eventlet/pull/911#discussion_r1478637054) noticing that `requests`, for example, doesn\u0027t have a way to limit total response time...","commit_id":"99c7a8b2c5ef104f0b17c736a9915e918fb10589"}]}