summaryrefslogtreecommitdiffstats
path: root/.github/workflows
diff options
context:
space:
mode:
authorSilvan Mosberger <silvan.mosberger@tweag.io>2024-01-16 23:33:53 +0100
committerSilvan Mosberger <silvan.mosberger@tweag.io>2024-01-17 21:26:39 +0100
commit8cf3cd91b2b6052f99fbe78f80141e57fcc02230 (patch)
tree6b2d95d35fe782cca1104ae1b7554d773a0fbd88 /.github/workflows
parent50d57f623671afd2ecd4180b928a4e44cbabe399 (diff)
check-by-name: Update contributor docs
Now that the tooling is pinned instead of being fetched indirectly from the channel, the docs need some updating
Diffstat (limited to '.github/workflows')
-rw-r--r--.github/workflows/check-by-name.yml5
1 files changed, 3 insertions, 2 deletions
diff --git a/.github/workflows/check-by-name.yml b/.github/workflows/check-by-name.yml
index 983b10e4cd40..0093560cf3c2 100644
--- a/.github/workflows/check-by-name.yml
+++ b/.github/workflows/check-by-name.yml
@@ -4,8 +4,8 @@
# When you make changes to this workflow, also update pkgs/test/nixpkgs-check-by-name/scripts/run-local.sh adequately
name: Check pkgs/by-name
-# The pre-built tool is fetched from a channel,
-# making it work predictable on all PRs.
+# The tool is pinned to a pre-built version on Hydra,
+# see pkgs/test/nixpkgs-check-by-name/scripts/README.md
on:
# Using pull_request_target instead of pull_request avoids having to approve first time contributors
pull_request_target
@@ -92,6 +92,7 @@ jobs:
echo "base=$base" >> "$GITHUB_ENV"
- uses: cachix/install-nix-action@7ac1ec25491415c381d9b62f0657c7a028df52a7 # v24
- name: Fetching the pinned tool
+ # Update the pinned version using pkgs/test/nixpkgs-check-by-name/scripts/update-pinned-tool.sh
run: |
# Get the direct /nix/store path from the pin to avoid having to evaluate Nixpkgs
toolPath=$(jq -r '."ci-path"' pkgs/test/nixpkgs-check-by-name/scripts/pinned-tool.json)