Repository navigation
Conversation
…and coverage Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
| There are two test layers with different guarantees. Put a test in the | ||
| cheapest layer that can actually prove the thing, and don't restate the same | ||
| claim in both. | ||
| Put a test in the cheapest layer that can actually prove the thing, and don't |
There was a problem hiding this comment.
🛠️ Keep the testing choices clear: The old two-layer description could make Copilot treat the list as only two choices. The three existing test choices remain available, with the cheapest-sufficient placement and nonduplication duties applying across them.
| - **Go command test** (`cmd/**/*_test.go` with `httpmock`) — full CLI wiring | ||
| end to end, but with a stubbed resolver so it stays deterministic. Runs in | ||
| the blocking `test` CI job with no token. Reach for this to prove the | ||
| the blocking `go-test` CI job with no token. Reach for this to prove the |
There was a problem hiding this comment.
🛠️ Find the command-test gate: The old name directs Copilot to a job identifier that is not present. The workflow's Go job is go-test; it still runs the Go tests without a supplied API token.
|
|
||
| 1. Install Go [through download](https://go.dev/doc/install) | [through Homebrew](https://formulae.brew.sh/formula/go). The required version is pinned in [`go.mod`](go.mod). | ||
| 1. Some integration tests use Ruby; install it [through download](https://www.ruby-lang.org/en/documentation/installation/) | [through Homebrew](https://formulae.brew.sh/formula/ruby) if you plan to run them. | ||
| To recover from Ruby integration-harness load errors for `reline` or `webrick`, use Ruby providing `reline` and install `webrick` in that Ruby's gem environment if missing. |
There was a problem hiding this comment.
🛠️ Recover missing test libraries: Installing Ruby alone does not establish that the libraries loaded by the test runner are available. For those load errors, this directs Copilot to the Ruby environment actually running the tests, installing WEBrick only when missing.
| ## Setup | ||
|
|
||
| Requires an authenticated GitHub CLI. github.com and GitHub Enterprise Cloud with data residency | ||
| github.com and GitHub Enterprise Cloud with data residency |
There was a problem hiding this comment.
🛠️ Allow offline verification: The unconditional login prerequisite could stop Copilot from using the offline path, which runs before network setup. Online operations can still require authentication; removing the blanket prerequisite does not change those requirements.
|
|
||
| Run every command from the repository checkout. The host comes from `--hostname`, then `GH_HOST`, | ||
| then the repository remote, then github.com. | ||
| then the current repository (`GH_REPO` or a remote), then github.com. |
There was a problem hiding this comment.
🛠️ Choose the intended host: Remote-only wording could make Copilot predict the wrong host when GH_REPO is set. Current-repository resolution follows the explicit overrides, and a host-qualified GH_REPO supplies that repository host.
| JSON goes to stdout and progress to stderr; parse only stdout. `--json` alone selects | ||
| `valid,findings,workflows`. Pass a list to choose from `valid`, `findings`, `workflows`, | ||
| `dependencies`, and `pins`. Each finding has: | ||
| `dependencies`, and `pins`. Finding fields include: |
There was a problem hiding this comment.
🛠️ Accept valid findings: Requiring every listed key would reject valid findings with omitted optional fields. The list describes possible fields, not keys guaranteed in every finding.
| into the repository, then run `gh actions-lock` once so the new workflow is locked too. It runs | ||
| `gh actions-lock --verify` on pull requests, pushes to `main`, and a weekly schedule that catches | ||
| re-pointed tags. The job's `GITHUB_TOKEN` reads public actions and the repository itself; actions | ||
| `gh actions-lock --verify` on pull requests and pushes to `main`. |
There was a problem hiding this comment.
🛠️ Avoid assuming scheduled coverage: The weekly-run claim could make Copilot assume the copied workflow checks tag changes without new repository activity. Its triggers are pull requests and pushes to main; those checks remain, but the template provides no weekly run.
|
|
||
| `gh actions-lock --help` is authoritative for flags. Failed runs print the path of a saved run | ||
| log; attach it to bug reports. Start from these, plus each finding's `doc_url`: | ||
| `gh actions-lock --help` is authoritative for flags. Attach saved run logs to bug reports. |
There was a problem hiding this comment.
🛠️ Handle failures without a log path: The old promise could make Copilot treat a missing path as a separate defect. Offline and JSON failures do not print that path, and saving a log is best-effort; available saved logs should still accompany bug reports.
🧠 How this helps Copilot
Copilot could skip useful offline checks or assume verification runs when it does not. This guidance separates real prerequisites and verification coverage from guarantees the tools do not make, and gives a recovery route when the Ruby test runner cannot start.
Automated details
Source:
curate-memoriesDream, analyzingmainat113f91d.