Validate meetup image sources at build time (#142)

.github/workflows/ci.yml

@@ -33,6 +33,9 @@ jobs:
       - name: Generate tag and category data
         run: make generate
 
+      - name: Validate meetup image references
+        run: make validate
+
       - name: Build Jekyll site
         run: bundle exec jekyll build
 

CLAUDE.md

@@ -60,6 +60,10 @@ Tags and categories are not manually maintained — they're generated from colle
 
 The CI build (`.github/workflows/pages.yml`) always runs `make generate` before `jekyll build`. Do not manually edit files in `tags/`, `categories/`, or `_data/tags.yml`/`_data/categories.yml` — they will be overwritten.
 
+### Image Validation
+
+`_scripts/validate_images.sh` (exposed as `make validate`) checks that every meetup's `image:` front matter resolves to a file in `archives/images/events/`, skipping external `http(s)` URLs. It exits non-zero on a missing source and runs on pull requests via `ci.yml`. It is intentionally **not** part of `make generate` or `pages.yml`, so a missing image never blocks the production deploy. Run `bash _scripts/test_validate_images.sh` to exercise the validator against fixtures.
+
 ### Navigation and Site Data
 
 Site-wide data is in `_data/`:

Makefile

@@ -40,6 +40,10 @@ generate-pages:
 # Generate all content
 generate: generate-data generate-pages
 
+# Validate that meetup image references resolve to source files
+validate:
+	$(SCRIPTS_DIR)/validate_images.sh
+
 # Update Git submodules
 update:
 	git submodule update --remote --merge
@@ -48,4 +52,4 @@ update:
 clean:
 	rm -rf _site .jekyll-metadata
 
-.PHONY: all bundle serve serve-incremental generate-data generate-pages generate update-submodules clean
+.PHONY: all bundle serve serve-incremental generate-data generate-pages generate validate update-submodules clean

README.md

@@ -88,6 +88,9 @@ Ensure your Ruby environment matches the `Gemfile` requirements. Older gem versi
 **Jekyll-GitHub Pages Compatibility**
 GitHub Pages restricts Jekyll plugins for security. See [supported versions](https://docs.github.com/en/pages/setting-up-a-github-pages-site-with-jekyll/about-github-pages-and-jekyll#plugins) for details.
 
+**Validating Meetup Images**
+Run `make validate` to check that every meetup's `image:` front matter points at a real file in `archives/images/events/`. CI runs this on pull requests to `main` (including the daily submodule-update PR) and fails if a referenced image is missing, so broken references are caught before they reach the site.
+
 **Image Derivative Caching**
 CI caches generated image thumbnails (`_site/assets/thumbs`) keyed on the contents of `archives/images/`, so unchanged images are not regenerated. Locally, `_site` persists between builds, so repeat builds are already incremental — running `make clean` wipes `_site` and forces a full regeneration on the next build.
 

_scripts/test_validate_images.sh

@@ -0,0 +1,54 @@
+#!/bin/bash
+# Fixture-based tests for validate_images.sh. Each case builds a temp ARCHIVES_DIR.
+set -euo pipefail
+
+SCRIPT_DIR="$(cd "$(dirname "$0")" && pwd)"
+VALIDATOR="$SCRIPT_DIR/validate_images.sh"
+fails=0
+
+setup_fixture() {
+  mkdir -p "$1/_meetups" "$1/images/events"
+}
+
+# Case 1: source present -> exit 0
+t1=$(mktemp -d); setup_fixture "$t1"
+printf 'image: present.jpg\n' > "$t1/_meetups/ok.md"
+: > "$t1/images/events/present.jpg"
+if ARCHIVES_DIR="$t1" "$VALIDATOR" >/dev/null 2>&1; then
+  echo "✅ case 1 (present) exits 0"
+else
+  echo "❌ case 1 (present) should exit 0"; fails=$((fails + 1))
+fi
+
+# Case 2: source missing -> exit 1, names the file and image
+t2=$(mktemp -d); setup_fixture "$t2"
+printf 'image: missing.avif\n' > "$t2/_meetups/bad.md"
+out=$(ARCHIVES_DIR="$t2" "$VALIDATOR" 2>&1) && rc=0 || rc=$?
+if [ "${rc:-0}" -eq 1 ] && printf '%s' "$out" | grep -q "bad.md" && printf '%s' "$out" | grep -q "missing.avif"; then
+  echo "✅ case 2 (missing) exits 1 and names bad.md/missing.avif"
+else
+  echo "❌ case 2 (missing) should exit 1 naming bad.md/missing.avif (rc=${rc:-0})"; fails=$((fails + 1))
+fi
+
+# Case 3: external URL -> not flagged, exit 0
+t3=$(mktemp -d); setup_fixture "$t3"
+printf 'image: https://example.com/x.jpg\n' > "$t3/_meetups/url.md"
+if ARCHIVES_DIR="$t3" "$VALIDATOR" >/dev/null 2>&1; then
+  echo "✅ case 3 (external url) exits 0"
+else
+  echo "❌ case 3 (external url) should exit 0"; fails=$((fails + 1))
+fi
+
+# Case 4: CRLF line endings on the image value -> still resolves -> exit 0
+t4=$(mktemp -d); setup_fixture "$t4"
+printf 'image: present.jpg\r\n' > "$t4/_meetups/crlf.md"
+: > "$t4/images/events/present.jpg"
+if ARCHIVES_DIR="$t4" "$VALIDATOR" >/dev/null 2>&1; then
+  echo "✅ case 4 (CRLF) exits 0"
+else
+  echo "❌ case 4 (CRLF) should exit 0"; fails=$((fails + 1))
+fi
+
+rm -rf "$t1" "$t2" "$t3" "$t4"
+if [ "$fails" -gt 0 ]; then echo "❌ $fails test(s) failed"; exit 1; fi
+echo "✅ all validate_images tests passed"

_scripts/validate_images.sh

@@ -0,0 +1,41 @@
+#!/bin/bash
+set -euo pipefail
+
+ARCHIVES_DIR="${ARCHIVES_DIR:-archives}"
+MEETUPS_DIR="$ARCHIVES_DIR/_meetups"
+EVENTS_DIR="$ARCHIVES_DIR/images/events"
+
+echo "🔍 Validating meetup image references in $MEETUPS_DIR..."
+
+if [ ! -d "$MEETUPS_DIR" ]; then
+  echo "❌ Meetups directory not found: $MEETUPS_DIR"
+  exit 1
+fi
+
+missing=0
+checked=0
+
+shopt -s nullglob
+for file in "$MEETUPS_DIR"/*.md; do
+  # First `image:` value, with surrounding whitespace and any quotes stripped.
+  img=$(grep -m1 '^image:' "$file" 2>/dev/null | sed "s/^image:[[:space:]]*//; s/[\"']//g; s/[[:space:]]*\$//" || true)
+
+  # Skip entries with no image or an external URL.
+  [ -z "$img" ] && continue
+  case "$img" in
+    http://*|https://*) continue ;;
+  esac
+
+  checked=$((checked + 1))
+  if [ ! -f "$EVENTS_DIR/$img" ]; then
+    echo "❌ $(basename "$file") → image '$img' not found in $EVENTS_DIR/"
+    missing=$((missing + 1))
+  fi
+done
+
+if [ "$missing" -gt 0 ]; then
+  echo "❌ $missing meetup image reference(s) missing."
+  exit 1
+fi
+
+echo "✅ All $checked meetup image references resolve."