Conversation
Greptile SummaryThis PR adds a new compatibility scraper for argo-workflows, following the same conventions as sibling scrapers in
Confidence Score: 4/5Safe to merge; all changes are additive and follow established codebase conventions. The scraper closely mirrors existing scrapers (argo-rollouts, argo-cd) in structure and delegation to shared utilities. No critical logic errors were found. The three flagged items are all style/best-practice suggestions: per-tag HTTP request volume, a page-count asymmetry that may silently skip old versions, and a missing eolApiSlug that other addons include. None of these prevent the scraper from producing correct output for recent versions. utils/compatibility/scrapers/argo-workflows.py — review the per-tag HTTP call pattern and the page-count mismatch with get_github_releases_timestamps; static/compatibilities/argo-workflows.yaml — consider adding eolApiSlug
|
| Filename | Overview |
|---|---|
| utils/compatibility/scrapers/argo-workflows.py | New scraper for argo-workflows; fetches K8s versions from hack/k8s-versions.sh per tag with a timestamp-based fallback; makes one unthrottled HTTP request per tag and has a page-count asymmetry with get_github_releases_timestamps |
| static/compatibilities/argo-workflows.yaml | New compatibility metadata file with correct icon, URLs, and helm chart info; missing eolApiSlug field present in other addon YAMLs |
| static/compatibilities/manifest.yaml | Correctly adds argo-workflows to the scraper manifest in the Argo-tools grouping |
Reviews (1): Last reviewed commit: "added argo-workflows scraper" | Re-trigger Greptile
![greptile-apps[bot]](https://avatars.githubusercontent.com/in/867647?s=60&v=4){kind=small}
| def fetch_github_tags(): | ||
| """Fetch release tags from GitHub API.""" | ||
| tags = [] | ||
| for page in range(1, 5): | ||
| url = f"https://api.github.com/repos/{github_repo_owner}/{github_repo_name}/tags" | ||
| response = requests.get(url, params={"page": page, "per_page": 100}) | ||
| if response.status_code != 200: | ||
| print_error(f"Failed to fetch GitHub tags. Status code: {response.status_code}") | ||
| break | ||
| page_tags = [tag["name"] for tag in response.json()] | ||
| if not page_tags: | ||
| break | ||
| tags.extend(page_tags) | ||
| return tags |
There was a problem hiding this comment.
fetch_k8s_versions_from_tag is called for every tag that passes the "-" in tag filter and has a matching chart version. With up to 400 tags fetched across 4 pages, this can generate hundreds of sequential unauthenticated HTTP requests to raw.githubusercontent.com with no rate limiting or delay between calls.
The argo-rollouts.py scraper follows the same pattern, so this is consistent with the codebase — but given the larger number of argo-workflows releases, it may be worth adding a short delay or pre-checking whether the version already exists in the YAML before fetching.
|
|
||
| chart_versions = get_chart_versions(app_name) | ||
| kube_releases = get_kube_release_info() | ||
| argo_releases = list(reversed(list(get_github_releases_timestamps(github_repo_owner, github_repo_name)))) | ||
| pruned_argo_releases = {r.lstrip("v"): ts for r, ts in argo_releases if "-" not in r} |
There was a problem hiding this comment.
fetch_github_tags fetches up to 4 pages (≤ 400 tags), but get_github_releases_timestamps in utils.py only fetches 2 pages (≤ 200 releases). For any tag that falls outside the 200-release window and does not have a hack/k8s-versions.sh file, the fallback at lines 85–91 will find no matching entry in pruned_argo_releases and silently skip that version:
print_warning(f"No K8s version info found for {tag}, skipping.")
continueIn practice the most recent releases are always within the first 200 entries, so current data should be complete. For completeness of historical data, consider aligning the two page counts or documenting the intentional cap.
| icon: https://avatars.githubusercontent.com/u/30269780?s=200&v=4 | ||
| git_url: https://github.com/argoproj/argo-workflows | ||
| release_url: https://github.com/argoproj/argo-workflows/releases/tag/v{vsn} | ||
| helm_repository_url: https://argoproj.github.io/argo-helm | ||
| chart_name: argo-workflows | ||
| versions: [] |
There was a problem hiding this comment.
eolApiSlug field
Several other addon YAML files (e.g. cert-manager.yaml, cilium.yaml) include an eolApiSlug field, which main.py uses to enrich version entries with EOL dates from endoflife.date. Argo Workflows has an entry there under the slug argo-workflows.
Omitting this field means the EOL enrichment step will be skipped entirely for this addon. Consider adding:
| icon: https://avatars.githubusercontent.com/u/30269780?s=200&v=4 | |
| git_url: https://github.com/argoproj/argo-workflows | |
| release_url: https://github.com/argoproj/argo-workflows/releases/tag/v{vsn} | |
| helm_repository_url: https://argoproj.github.io/argo-helm | |
| chart_name: argo-workflows | |
| versions: [] | |
| icon: https://avatars.githubusercontent.com/u/30269780?s=200&v=4 | |
| git_url: https://github.com/argoproj/argo-workflows | |
| release_url: https://github.com/argoproj/argo-workflows/releases/tag/v{vsn} | |
| helm_repository_url: https://argoproj.github.io/argo-helm | |
| chart_name: argo-workflows | |
| eolApiSlug: argo-workflows | |
| versions: [] |
| release_url: https://github.com/argoproj/argo-workflows/releases/tag/v{vsn} | ||
| helm_repository_url: https://argoproj.github.io/argo-helm | ||
| chart_name: argo-workflows | ||
| versions: [] |
There was a problem hiding this comment.
this should have a list of versions discovered, did the scraper not run or is it just not yet functional?
2 participants
Test Plan
tested via CLI commands
Checklist
Plural Flow: console