Builder: surface check failures and use a real Python #434
No reviewers
Labels
No labels
accessibility
autorelease: pending
autorelease: tagged
bug
dependencies
documentation
duplicate
enhancement
github_actions
good first issue
help wanted
invalid
question
testing wanted
wontfix
No milestone
No assignees
1 participant
Due date
No due date set.
Dependencies
No dependencies set
Reference
MagicBeanDev/Hookshot!434
Loading…
Reference in a new issue
No description provided.
Delete branch "nup-fix"
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?
Two source-tree Builder fixes found while testing with a
.nup.1. A check that stops now says why.
--checktests the compiler before it opens the package. A missing compiler raisedSystemExitfromtoolchain.need(), whichmake_game.pyre-raised without an@@errorline; the WPF builder then showed "The game files could not be read." and the Linux builder "The version could not be read.", with the reason only on stderr (whichRunner.Checkdiscarded).check_windows_sdk()/check_mingw()now report a missing compiler throughfail(), a textSystemExitbecomes@@error, and both builders show the last stderr line when a step stops without a protocol line.2. A real Python is found.
Tools.Find()usedpythonfromPATHfor source-tree runs. On Windows that is often the Store alias, which only opens the Store (exit 9009). It now prefers a realpython.exe, thenpy -3, and passes the launcher's arguments before-I(py -I -3made Python stop with "Unknown option: -3").Tests (game-free):
all pass;
dotnet build builder/HookshotBuilder/HookshotBuilder.csprojgives 0 warnings and 0 errors. WithHOOKSHOT_MINGWset,make_game.py --checkon a real.nuppasses the compiler check and reads the package (@@check ok ... build e0f3ccfa). Both tests are wired intosource-regressions.yml(run steps and path filters).Assisted-by: DeepSeek V4.1 Flash (OpenCode)