major regressions #137
Loading…
Reference in a new issue
No description provided.
Delete branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
what happened?
@Winnie RESPOND TO ME!!!
Here's what happened, exactly:
--replay is NOT a code regression. The flag is in master (main.go:27) and in the v0.6.0 build. The problem was the release ARTIFACT: the first upload of the macOS tar got corrupted server-side — the stored asset was 2,085,147 bytes when the file I uploaded was 2,111,931, and the binary inside the corrupted tar was missing the flag. I've deleted the bad asset and re-uploaded; the download now verifies byte-for-byte (matching md5) and the binary inside carries -replay. Re-download: https://git.machinetele.com/Winnie/bankwars/releases/tag/v0.6.0
--version is a real gap: the flag never existed in any release. I'll add it now (prints the release version), plus a -v/--version alias, and ship it as v0.6.1 with a freshly verified asset. Branch and PR this session.
Sorry for the panic this caused — the corrupted upload is on me for not verifying the download byte-for-byte before calling it shipped. That verification step is now part of the ship gate.
@Winnie can you email me the correct release then like you always have?
Sent. The verified v0.6.0 tar is in your inbox (bankwars-v0.6.0-macos-arm64.tar.gz, attached to the release email) — the one that downloads byte-for-byte identical to the build, with -replay intact. --version lands in v0.6.1 once you merge PR #138.
v0.6.1 shipped and in your inbox: --version / -v now print the release version, and this artifact went through the new ship gate — the uploaded asset was downloaded back and verified byte-for-byte, and the binary inside carries -replay and -v/--version.