Skip to content

Take #69, #70 and #71, each with a test - #77

Merged
hsbt merged 6 commits into
masterfrom
claude/great-vaughan-8eb8d4
Sep 17, 2026
Merged

hsbt merged 6 commits into
masterfrom
claude/great-vaughan-8eb8d4

Conversation

@hsbt

@hsbt hsbt commented Sep 17, 2026

Copy link
Copy Markdown
Member

I cherry-picked the three fixes from #69, #70 and #71 unchanged, so they keep Oskar Eichler as their author. I reproduced each symptom on master first, then reviewed the patches against it. Only #71 has room to go wrong, since it restructures five methods, and its success paths come out byte for byte identical. None of the three needed a change.

What those PRs left out, deliberately, is tests, so I wrote them. Each one fails with its own fix reverted. The :wait_readable fake caps its attempts, so a regression there fails instead of hanging the suite.

write_message_by_block had no coverage at all, so I also pinned its normal return value and its break case, neither of which the restructure changes.

Generated with Claude Code

OskarEichler and others added 6 commits September 17, 2026 12:04
Before the check, read(-1) walked @rbuf_offset backwards and a later
read_all handed back bytes an earlier read had already consumed.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The case had no branch for it, so the retry loop spun without waiting
and without ever reaching the write_timeout. The fake refuses the write
until it has been waited on, which turns a regression into a failure
rather than a hang.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
LOG_off had no counterpart on the exception path, so a raise from the
block or from the source silenced the connection's debug_output for
good. The two success cases go with them because the restructure
rewrote the method that had no coverage at all.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
@hsbt
hsbt merged commit 1c2ce69 into master Sep 17, 2026
72 checks passed
@hsbt
hsbt deleted the claude/great-vaughan-8eb8d4 branch September 17, 2026 03:25
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants