Stop two lab tests reaching outside themselves - #247
Merged
Conversation
A lab run needs a -game dir under engine/, and a git add -A has now swept one in twice. Removes engine/argsession, which should never have been committed, and ignores engine/*/ with engine/argus/ negated so the shipped mod dir stays tracked and nothing else under engine/ can follow it in.
The lab suite fails roughly one run in ten on this box and took the lab job down on #246. Two causes, both mine, both from the tests rather than the code they cover. backup::two_installs_outside_the_root_both_survive called restore_backup, and install_paths() includes the rerelease Saved Games copy, which lives under USERPROFILE and not under the test's temp root. The test was writing to a real install on the developer's machine every run - the exact class #222 was about - and racing anything else that touched it. take_backup only reads, so it stays; the assertion moved to the manifest, which is where the #215 key collision actually shows. The GUI POST guard read the bound port from a process-global that the concurrent lib tests share. A security check should not depend on hidden process state, so the port is a parameter now and route() is a thin wrapper that supplies it. Neither change touches behaviour. A residual flake remains: two of ten full-suite runs still failed after both fixes and I could not catch the failing test name again in eight further runs, so it is rarer than it was and not yet root-caused.
|
You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard. |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
The lab suite fails roughly one run in ten on my box and took the lab job down on #246. Two causes, both mine, both in the tests rather than the code they cover.
backup::two_installs_outside_the_root_both_survivecalledrestore_backup, andinstall_paths()includes the rerelease Saved Games copy, which lives underUSERPROFILEand not under the test's temp root. The test was writing to a real install on the developer's machine every run - the exact class cargo test on a configured box installs progs.dat and changes what latest means #222 was about - and racing anything else touching it.take_backuponly reads, so it stays; the assertion moved to the manifest, which is where the Backup drops an install when two sources sit outside ARGUS_ROOT #215 key collision actually shows.route()is a thin wrapper.Neither change touches behaviour.
Residual flake, not fixed. Two of ten full-suite runs still failed after both fixes, and I could not catch the failing test name again across eight further runs. It is rarer than it was and not root-caused. Worth knowing before trusting a single green lab job.