test(release): the git-cliff stand-in is a .cmd on Windows, which can start it #145

d4nkn3ss · 2026-11-13T06:34:16Z

d4nkn3ss commented

2026-11-13 06:34:16 +00:00

Why

tests/test_release.py's test_the_first_notes_give_git_cliff_no_range (added in #116) errors on every Windows run: its git-cliff stand-in is a #!/bin/sh script, which Windows can't start.

What changed

tests: in tests/test_release.py, on Windows the stand-in is git-cliff-stub.cmd (@echo %*), which echoes its arguments on one line; the test reads them with .split() either way. Elsewhere the shell script stays as it was.

Flags and settings

None

Measurements

main: e28236b4 (2026-11-12)
this PR: fix/test-release-cliff-stub-windows @ ef339b94
Hardware: Windows 11, Python 3.12 from python.org
Scene: n/a (game-free)
Method: python -I -B tests/test_release.py Bump.test_the_first_notes_give_git_cliff_no_range on both sides
How to reproduce: on Windows, the command above

	main @ `e28236b4`	this PR @ `ef339b94`
`test_the_first_notes_give_git_cliff_no_range`	ERROR: `OSError: [WinError 193] %1 is not a valid Win32 application`	OK

Testing

Tested: the row above on Windows 11.

Full suite: python -I -B tests/run_all.py on this branch @ ef339b94: 500 passed, 3 failed, 35 skipped, the same three as main @ e28236b4: tests/test_check_third_party.py (#80), tests/test_setup_runtime.py (#118) and tests/test_release.py, whose other failures are #119 (the SHA256SUMS fixture) and, from PowerShell, the missing openssl (#144).

Not tested: Linux and macOS: the shell-script branch is the code as it was.

Needs other testers: None

Screenshots

Nothing on screen

Risk and rollback

Tests only; reverting the commit is enough.

Issues

None

Checklist

For everyone

One topic.
The title is a Conventional Commit (type(scope): description).
Rebased on current main.
python -I -B tests/run_all.py run and passing (any failure that also fails on main is named above).
Tested in the game, or what wasn't is named under Testing. (Test-only change: nothing in the game.)
Testing filled in: what was tested, what wasn't, and what needs other testers (with test requests).
Measured: main and this PR at named commits, on the same PC, scene and settings, at least 2 interleaved runs each, the median reported. (A test result: the same on every run.)
No game files, dumps, build output or generated files.
third_party/ untouched (a new folder with a PROVENANCE.md instead).
The license is OK: GPL-3.0-or-later, nothing incompatible and nothing decompiled.

If applicable

AI assistance recorded with Assisted-by: / Co-Authored-By: trailers.

🤖 Generated with Claude Code

## Why `tests/test_release.py`'s `test_the_first_notes_give_git_cliff_no_range` (added in #116) errors on every Windows run: its git-cliff stand-in is a `#!/bin/sh` script, which Windows can't start. ## What changed - tests: in `tests/test_release.py`, on Windows the stand-in is `git-cliff-stub.cmd` (`@echo %*`), which echoes its arguments on one line; the test reads them with `.split()` either way. Elsewhere the shell script stays as it was. ## Flags and settings None ## Measurements main: e28236b4 (2026-11-12) this PR: fix/test-release-cliff-stub-windows @ ef339b94 Hardware: Windows 11, Python 3.12 from python.org Scene: n/a (game-free) Method: `python -I -B tests/test_release.py Bump.test_the_first_notes_give_git_cliff_no_range` on both sides How to reproduce: on Windows, the command above | | main @ e28236b4 | this PR @ ef339b94 | |---|---|---| | `test_the_first_notes_give_git_cliff_no_range` | ERROR: `OSError: [WinError 193] %1 is not a valid Win32 application` | OK | ## Testing **Tested:** the row above on Windows 11. **Full suite:** `python -I -B tests/run_all.py` on this branch @ ef339b94: 500 passed, 3 failed, 35 skipped, the same three as main @ e28236b4: `tests/test_check_third_party.py` (#80), `tests/test_setup_runtime.py` (#118) and `tests/test_release.py`, whose other failures are #119 (the SHA256SUMS fixture) and, from PowerShell, the missing `openssl` (#144). **Not tested:** Linux and macOS: the shell-script branch is the code as it was. **Needs other testers:** None ## Screenshots Nothing on screen ## Risk and rollback Tests only; reverting the commit is enough. ## Issues None ## Checklist ### For everyone - [x] One topic. - [x] The title is a Conventional Commit (`type(scope): description`). - [x] Rebased on current `main`. - [x] `python -I -B tests/run_all.py` run and passing (any failure that also fails on `main` is named above). - [x] Tested in the game, or what wasn't is named under Testing. (Test-only change: nothing in the game.) - [x] Testing filled in: what was tested, what wasn't, and what needs other testers (with test requests). - [x] Measured: main and this PR at named commits, on the same PC, scene and settings, at least 2 interleaved runs each, the median reported. (A test result: the same on every run.) - [x] No game files, dumps, build output or generated files. - [x] `third_party/` untouched (a new folder with a `PROVENANCE.md` instead). - [x] The license is OK: GPL-3.0-or-later, nothing incompatible and nothing decompiled. ### If applicable - [x] AI assistance recorded with `Assisted-by:` / `Co-Authored-By:` trailers. 🤖 Generated with [Claude Code](https://claude.com/claude-code)

d4nkn3ss added 1 commit

2026-11-13 06:34:17 +00:00

test(release): the git-cliff stand-in is a .cmd on Windows, which can start it

CI / Conventional Commit title (pull_request) Blocked by required conditions

Details

CI / Python source checks (ubuntu-24.04, part 1/2) (pull_request) Blocked by required conditions

Details

CI / Python source checks (ubuntu-24.04, part 2/2) (pull_request) Blocked by required conditions

Details

CI / SPIR-V fixtures (global narrow load/store, direct surface atomics, F2I immediate) (pull_request) Blocked by required conditions

Details

CI / Workflow lint (actionlint) (pull_request) Blocked by required conditions

Details

Third-party parts / As their notes say (pull_request) Blocked by required conditions

Details

Third-party parts / The same as upstream (pull_request) Blocked by required conditions

Details

PR description / Description follows the template (pull_request_target) Successful in 9s

Details

ef339b9442

test_the_first_notes_give_git_cliff_no_range (#116) gave release.notes() a #!/bin/sh script as GIT_CLIFF; Windows
can't start one (OSError: [WinError 193] %1 is not a valid Win32 application), so the test errored on Windows. There
the stand-in is a .cmd that echoes its arguments on one line, which the test's .split() reads the same way.

Co-Authored-By: Claude Opus 5.5 <[email protected]>

d4nkn3ss referenced this pull request

2026-11-13 06:34:25 +00:00

test(release): run_all finds openssl (Git for Windows' own too) and skips test_release without one #144

d4nkn3ss referenced this pull request

2026-11-13 07:24:12 +00:00

fix(release): the Windows zip carries tools/version.txt, so a release's game and builder know their version #149

d4nkn3ss referenced this pull request

2026-11-13 07:30:02 +00:00

fix(setup): the setup copies the Builder from a zip unpacked deeper than MAX_PATH #151

d4nkn3ss referenced this pull request

2026-11-13 07:51:20 +00:00

docs(design): fog and exposure use the game's own values; the removed OoT factors are no longer needed #159

d4nkn3ss referenced this pull request

2026-11-13 08:35:53 +00:00

test(release): the publish test writes its SHA256SUMS fixture as bytes, so it passes on Windows #119

HumanBean referenced this pull request from a commit

2026-11-13 09:06:04 +00:00

fix(release): the Windows zip carries tools/version.txt, so a release's game and builder know their version (#149)

d4nkn3ss referenced this pull request

2026-11-13 11:00:25 +00:00

fix(vk): Auto (this PC) reads the primary monitor when there's no window, so an update no longer lowers the resolution #176

d4nkn3ss referenced this pull request

2026-11-13 11:05:08 +00:00

test(builder): the DLSS update fixture no longer requires the removed self-update's download to be cleared #179

HumanBean referenced this pull request from a commit

2026-11-13 11:09:25 +00:00

fix(vk): Auto (this PC) reads the primary monitor when there's no window, so an update no longer lowers the resolution (#176)

HumanBean referenced this pull request from a commit

2026-11-13 11:10:00 +00:00

test(builder): the DLSS update fixture no longer requires the removed self-update's download to be cleared (#179)

d4nkn3ss referenced this pull request

2026-11-13 12:35:39 +00:00

test(input): the display-pointer fixture defines rx_option, which mouse_capture now asks (#178) #187

d4nkn3ss referenced this pull request

2026-11-13 13:09:59 +00:00

test(vk): the texture-staging fixture defines logf_at and LogLevel, which the cut code now calls #195

HumanBean referenced this pull request from a commit

2026-11-13 16:38:19 +00:00

test(input): the display-pointer fixture defines rx_option, which mouse_capture now asks (#178) (#187)

HumanBean referenced this pull request from a commit

2026-11-13 17:01:44 +00:00

docs(design): fog and exposure use the game's own values; the removed OoT factors are no longer needed (#159)

monsterlag requested changes

2026-11-13 20:02:10 +00:00

monsterlag left a comment

Requesting changes at ef339b94. The .cmd does fix WinError193 for an ASCII temporary path, but its output is not the UTF-8 that release.notes requires.

[P2] Emit UTF-8 from the Windows stand-in (tests/test_release.py:220). On native Windows Python3.14.7, set TEMP/TMP to an authored folder containing Zoë and run only Bump.test_the_first_notes_give_git_cliff_no_range. This head fails: cmd echo writes byte0x89 for ë in the echoed --config path, the release.notes UTF-8 reader raises UnicodeDecodeError, then p.stdout is None and .strip raises AttributeError. The same case passes under an ASCII TEMP path. A Windows user with a non-ASCII account/TEMP path therefore still cannot run this check. Use a stand-in that emits explicit UTF-8 and add a Unicode-path regression while keeping the range assertions. PR255's follow-up01943afa addresses both Unicode paths and output encoding; its same case passes in the accented TEMP here, and may be useful when reconciling the overlap.

This is a focused authored-fixture check; no real release creation, token operation or game run. I did not repeat the whole suite or change its assertions.

Requesting changes at ef339b94. The .cmd does fix WinError193 for an ASCII temporary path, but its output is not the UTF-8 that release.notes requires. 1. [P2] Emit UTF-8 from the Windows stand-in (tests/test_release.py:220). On native Windows Python3.14.7, set TEMP/TMP to an authored folder containing Zoë and run only Bump.test_the_first_notes_give_git_cliff_no_range. This head fails: cmd echo writes byte0x89 for ë in the echoed --config path, the release.notes UTF-8 reader raises UnicodeDecodeError, then p.stdout is None and .strip raises AttributeError. The same case passes under an ASCII TEMP path. A Windows user with a non-ASCII account/TEMP path therefore still cannot run this check. Use a stand-in that emits explicit UTF-8 and add a Unicode-path regression while keeping the range assertions. PR255's follow-up01943afa addresses both Unicode paths and output encoding; its same case passes in the accented TEMP here, and may be useful when reconciling the overlap. This is a focused authored-fixture check; no real release creation, token operation or game run. I did not repeat the whole suite or change its assertions.

tests/test_release.py

					
				@ -220,1 +218,3 @@

				        stub.chmod(0o755)

				        if os.name == "nt":   # (Windows starts a .cmd, not a #! script: WinError 193. It echoes its arguments on one line)

				            stub = self.dir / "git-cliff-stub.cmd"

				            stub.write_text("@echo %*\r\n", encoding="utf-8")

[P2] This echoes in the Windows console code page, but release.notes decodes UTF-8. Reproduced with TEMP/TMP containing Zoë: byte 0x89 in the --config path causes UnicodeDecodeError, then p.stdout.strip fails because stdout is None. The identical case passes with an ASCII TEMP. Please emit explicit UTF-8 and cover a Unicode path; PR255 follow-up01943afa already does this and passes the same probe.

CI / Conventional Commit title (pull_request) Blocked by required conditions

Details

CI / Python source checks (ubuntu-24.04, part 1/2) (pull_request) Blocked by required conditions

Details

CI / Python source checks (ubuntu-24.04, part 2/2) (pull_request) Blocked by required conditions

Details

CI / SPIR-V fixtures (global narrow load/store, direct surface atomics, F2I immediate) (pull_request) Blocked by required conditions

Details

CI / Workflow lint (actionlint) (pull_request) Blocked by required conditions

Details

Third-party parts / As their notes say (pull_request) Blocked by required conditions

Details

Third-party parts / The same as upstream (pull_request) Blocked by required conditions

Details

PR description / Description follows the template (pull_request_target) Successful in 9s

Details

This pull request doesn't have enough approvals yet. 0 of 1 approvals granted.

Some workflows are waiting to be reviewed.

This branch is out-of-date with the base branch

You are not authorized to merge this pull request.

View command line instructions

Checkout

From your project repository, check out a new branch and test the changes.

git fetch -u 'fix/test-release-cliff-stub-windows:d4nkn3ss-fix/test-release-cliff-stub-windows'

git switch 'd4nkn3ss-fix/test-release-cliff-stub-windows'

Sign in to join this conversation.

No reviewers

No milestone

No project

No assignees

2 participants

Notifications

Due date

The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set

Reference

MagicBeanDev/Longshot!145

Rows
Columns