asjadsyed opened a new pull request, #3121: URL: https://github.com/apache/thrift/pull/3121
<!-- Explain the changes in the pull request below: --> Normally I include a check like this before closing Thrift connections: ``` if transport and transport.isOpen(): transport.close() ``` However, for Tornado, `TTornadoStreamTransport.isOpen()` is not implemented, so the call falls back to `TTransportBase`'s empty implementation which just returns `None`. As a result, the above condition to close the connection never gets satisfied, and my application accumulates open connections until eventually it crashes with error message `OSError: [Errno 24] Too many open files`. To fix this, this patch implements `TTornadoStreamTransport.isOpen()`, using Tornado's [`BaseIOStream.closed()`](https://www.tornadoweb.org/en/stable/iostream.html#tornado.iostream.BaseIOStream.closed ), which returns `True` if the stream has been closed. Before checking if the stream is open, we need to verify that the stream exists. To test the patch, we can add these print statements into [`thrift/tutorial/py.tornado/PythonClient.py`](https://github.com/apache/thrift/blob/2e00c9998f1aa316c8d0168488887fb957845230/tutorial/py.tornado/PythonClient.py#L46) in the appropriate places: ``` print(f"transport.isOpen() before transport.open(): {transport.isOpen()}") print(f"transport.isOpen() after transport.open(): {transport.isOpen()}") print(f"transport.isOpen() after TTransportException: {transport.isOpen()}") print(f"transport.isOpen() before transport.close(): {transport.isOpen()}") print(f"transport.isOpen() after transport.close(): {transport.isOpen()}") ``` Without the patch applied, `isOpen()` always returns `None`, regardless of whether the transport is opened or closed: ``` transport.isOpen() before transport.open(): None transport.isOpen() after transport.open(): None transport.isOpen() before transport.close(): None transport.isOpen() after transport.close(): None ``` ``` transport.isOpen() before transport.open(): None ERROR:root:could not connect to 127.0.0.1:9090 (Stream is closed) transport.isOpen() after TTransportException: None ``` With the patch applied, `isOpen()` accurately reflects the transport state: ``` transport.isOpen() before transport.open(): False transport.isOpen() after transport.open(): True transport.isOpen() before transport.close(): True transport.isOpen() after transport.close(): False ``` ``` transport.isOpen() before transport.open(): False ERROR:root:could not connect to localhost:9090 (Stream is closed) transport.isOpen() after TTransportException: False ``` It also worked neatly with reopening the transport but I didn't include that as it didn't fit in neatly with the tutorial's outline. <!-- We recommend you review the checklist/tips before submitting a pull request. --> - [x] Did you create an [Apache Jira](https://issues.apache.org/jira/projects/THRIFT/issues/) ticket? ([Request account here](https://selfserve.apache.org/jira-account.html), not required for trivial changes) - [THRIFT-5861](https://issues.apache.org/jira/browse/THRIFT-5861) - [x] If a ticket exists: Does your pull request title follow the pattern "THRIFT-NNNN: describe my issue"? - Yes - [x] Did you squash your changes to a single commit? (not required, but preferred) - Yes - [x] Did you do your best to avoid breaking changes? If one was needed, did you label the Jira ticket with "Breaking-Change"? - Yes - [x] If your change does not involve any code, include `[skip ci]` anywhere in the commit message to free up build resources. - Involves a code change <!-- The Contributing Guide at: https://github.com/apache/thrift/blob/master/CONTRIBUTING.md has more details and tips for committing properly. --> -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: dev-unsubscr...@thrift.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org
