These are local runs related to Byron/gitoxide#1429.
See also older test runs.
Unlike here, CI in that PR has failures. The main reason for this gist--since the above earlier gist is still relevant--is to show easily identiifable full lists of what tests are run, for comparison to CI.
| Test suite does not assert directory symlink creation |
| PermissionDenied checking Windows symlink target is misinterpreted as collision |
| Checkout fails when Windows symlinks have strangely named targets |
| C:\Users\ek\source\repos> mkdir are-symlinks-linearized | |
| Directory: C:\Users\ek\source\repos | |
| Mode LastWriteTime Length Name | |
| ---- ------------- ------ ---- | |
| d---- 6/23/2024 8:15 PM are-symlinks-linearized | |
| C:\Users\ek\source\repos> cd are-symlinks-linearized | |
| C:\Users\ek\source\repos\are-symlinks-linearized> cmd |
| ek@Glub MINGW64 ~/src/crossbeam (master) | |
| $ cargo nextest run --all --no-fail-fast | |
| warning: C:\Users\ek\src\crossbeam\Cargo.toml: unused manifest key: workspace.lints.rust.unexpected_cfgs.check-cfg | |
| Finished `test` profile [unoptimized + debuginfo] target(s) in 0.17s | |
| Starting 631 tests across 49 binaries (run ID: 63ca707b-a5e1-49ee-8a94-cf8d21fef6fd, nextest profile: default) | |
| PASS [ 0.074s] crossbeam::subcrates deque | |
| PASS [ 0.067s] crossbeam::subcrates epoch | |
| PASS [ 0.078s] crossbeam::subcrates channel | |
| PASS [ 0.061s] crossbeam::subcrates queue | |
| PASS [ 0.056s] crossbeam::subcrates utils |
| Finished `test` profile [unoptimized + debuginfo] target(s) in 4m 17s | |
| Starting 2351 tests across 127 binaries (9 skipped; run ID: f8326ed4-26fc-4711-a963-eb7ddcdac199, nextest profile: default) | |
| PASS [ 0.090s] gitoxide plumbing::main::tests::clap | |
| PASS [ 0.085s] gitoxide shared::value_parser_tests::rename_fraction | |
| PASS [ 0.081s] gix id::tests::size_of_oid | |
| PASS [ 0.076s] gix open::tests::size_of_options | |
| Canceling due to interrupt: 16 tests still running | |
| ABORT [ 23.634s] gix::gix clone::blocking_io::fetch_and_checkout | |
| Message [ ] code 0xc0000409: The system detected an overrun of a stack-based buffer in this application. This overrun could potentially allow a malicious user to gain control of this application. (os error 1282) | |
| ABORT [ 23.784s] gix remote::connection::fetch::refs::tests::update::checked_out_branches_in_worktrees_are_rejected_with_additional_information |
| Administrator@WIN-J648LOCBS5P MINGW64 ~/repos/gitoxide (main) | |
| $ RUST_BACKTRACE=1 cargo nextest run -p "path+file:///C:/Users/Administrator/repos/gitoxide/gix-discover" upwards::from_ | |
| dir_with_dot_dot | |
| Finished `test` profile [unoptimized + debuginfo] target(s) in 0.33s | |
| Starting 1 test across 3 binaries (46 skipped; run ID: d1571af1-5503-4a0f-b096-170a604ba2c7, nextest profile: default) | |
| FAIL [ 0.135s] gix-discover::discover upwards::from_dir_with_dot_dot | |
| --- STDOUT: gix-discover::discover upwards::from_dir_with_dot_dot --- | |
| running 1 test |
This is experiment D, examining the environments of the chain of processes created to clone submodules with different versions of git, to investigate the treatment of GIT_CONFIG_COUNT-related configuration in nested submodule clones.
Experiment B and experiment C are also of interest. In contrast, this experiment examines the environment in detail, and also does not make use of gitoxide tests or fixture scripts, though the motivating goal is still to improve the compatibility of gitoxide's test suite with different git versions (which are used in fixtures to produce repositories for testing gitoxide, and occasionally to compare git and gitoxide behavior).
License for this gist: CC0-1.0
These are the results of some Ubuntu 22.04 LTS test runs and manual fixture script runs, investigating failures with GIX_TEST_IGNORE_ARCHIVES=1 with the Ubuntu-provided git.
These augment, but do not supersede, https://gist.github.com/EliahKagan/53ebccab054a9b43e7f119bf1f71189d (experiment B).
The differences between experiment B and the this experiment (experiment C) are: