mirrors/arti.git/crates/arti-rpc-client-core/src/conn.rs, branch arti-v1.4.0 mirror of https://gitlab.torproject.org/tpo/core/arti http://git.dilluti0n.com/mirrors/arti.git/atom?h=arti-v1.4.0 2025-02-04T19:41:15Z rpclib: Improve output for ConnectFailure 2025-02-04T19:41:15Z Nick Mathewson nickm@torproject.org 2025-02-04T17:28:02Z urn:sha1:dffe9330ab57d90f844f2aac950e4cbc2fc7a70b We want to explain better what has happened, not only with the final fatal error, but with any nonfatal errors that occurred in the middle. Closes #1826. rpclib: Obey error Display convention wrt sources. 2025-02-04T19:41:15Z Nick Mathewson nickm@torproject.org 2025-02-04T15:20:43Z urn:sha1:c8f614faf5084c300a71b6e607a0b189df8830da In general, we try to obey the convention that an error's Display method does not display that error's sources. Part of #1650. rpclib: Move socketpair to a testing module; use it in llconn tests 2025-02-03T18:41:15Z Nick Mathewson nickm@torproject.org 2025-02-03T18:41:15Z urn:sha1:7686f3a460ae511410f9e51f54e457635a972289 rpclib: In tests, use an ersatz socketpair on windows. 2025-02-03T17:00:20Z Nick Mathewson nickm@torproject.org 2025-02-03T17:00:20Z urn:sha1:624b3b4bcb833954a34bdae4c8f1215ac9a44652 Fixes #1831. I'm doing this because switching from the socketpair crate to an ersatz connect-to-localhost socketpair made the test pass. My windows skills are weak, but it appears possible that you aren't supposed to simultaneously have a blocking ReadFile and a blocking WriteFile to a named pipe from separate threads? At least, that's consistent with our findings in !2758. I first saw this hack in perl, when I needed it in 2003. I am displeased that it is still apparently needed in 2025. rpclib: make conn::complex use fewer iterations. 2025-02-03T16:57:39Z Nick Mathewson nickm@torproject.org 2025-02-03T16:57:39Z urn:sha1:150dbc30a6c9d85a51847ad77193040fae3e0506 This was the slowest test in our suite. fix: fix typos 2025-01-30T08:39:30Z Dimitris Apostolou dimitris.apostolou@icloud.com 2025-01-30T08:39:30Z urn:sha1:6371829c976f6a5b77d46bf977f369a915f2ced3 rpclib: Implement accessors for ConnectFailure 2025-01-29T14:17:21Z Nick Mathewson nickm@torproject.org 2025-01-28T15:18:57Z urn:sha1:55b074e904c0affc8b5e08425cf42c7a2b15e26a rpclib: New combined error type for RPC connect failures. 2025-01-29T14:17:21Z Nick Mathewson nickm@torproject.org 2025-01-28T14:52:32Z urn:sha1:28cc171b91d315412286b52297e0973fd2431f41 This type explains where every error came from, and explains why each declined connect point was declined. rpclib: More explanations about internal method usage. 2025-01-23T18:06:23Z Nick Mathewson nickm@torproject.org 2025-01-23T18:05:37Z urn:sha1:c19fc3779ce073c520a04b5228f004eb1a380620 arti-rpc-client-core: Rename Authentication{Rejected=>Failed} 2025-01-23T18:06:23Z Nick Mathewson nickm@torproject.org 2025-01-23T17:37:45Z urn:sha1:2d4872c41eec4d4ec27f022ede306b68d1583492 Also change corresponding message. This way, it will be clear that we're just reporting that Arti gave us an error—not that the error necessarily means that the authentication itself was "wrong".