chore(docs): refresh blog visual baseline after content change#3214
Conversation
There was a problem hiding this comment.
Wheels Bot — Reviewer
TL;DR — This PR refreshes the committed visual-regression baseline at web/tests/visual-baselines/blog.png after a blog-content merge, so that subsequent web/** PRs stop failing the Visual regression check on a stale-baseline diff (issue #2915). The diff is a single regenerated binary image with no code or logic surface. Verdict: approve (submitted as a comment because GitHub forbids the bot author from approving its own PR — HTTP 422 fallback).
I rendered the new baseline and confirmed it is a legitimate screenshot of the Wheels blog index, with the new post "Wheels + Claude: Building a Feature via the stdio MCP" (June 16, 2026) at the top — which matches the most recent content commit c81230f50. The baseline therefore accurately reflects the content change it claims to track, and the byte delta (Bin 122187 -> 118038) is consistent with a re-rendered page rather than corruption.
Commits
chore(docs): auto-refresh blog visual baseline after content change is a valid conventional-commit header — allowed type chore, optional scope docs, sentence-case subject, ~64 chars (well under the 100-char limit). The PR title chore(docs): refresh blog visual baseline after content change is likewise valid and is the squash-merge gate.
Nothing under Correctness, Cross-engine, Tests, Docs, or Security applies — this change touches no CFML, no views, and no runtime code. Clean.
Summary
A blog-content merge changed the blog index page, so the committed visual-regression baseline at
web/tests/visual-baselines/blog.pngis stale. Without this refresh, every subsequentweb/**-touching PR fails the Visual regression check on a deterministic diff until someone refreshes manually (issue #2915).develop's branch protection rejects direct pushes, so the refresh lands through this PR, opened and merged by the deploy run that published the blog change.
🤖 Opened automatically by .github/workflows/web-deploy.yml.