Merge pull request #324 from gonX/add-sitemap

gonX · web-flow · commit d213b83c1039 · 2026-01-19T06:36:44.000+01:00
Add XML sitemap
diff --git a/.github/workflows/jekyll.yml b/.github/workflows/jekyll.yml
@@ -20,19 +20,6 @@ jobs:
       - name: Run tests
         run: ./run-tests.sh || exit 1
 
-  internal-links:
-    name: "Absolute Internal Links"
-    needs: build
-    runs-on: ubuntu-latest
-    steps:
-      - name: Checkout
-        uses: actions/checkout@v4
-
-      - uses: ./.github/workflows/site-workflow
-
-      - name: Check for internal absolute links
-        run: bash -c "! grep -REi 'https?://(www.)?opentabletdriver.net' site/_site || exit 1"
-
   html-validation:
     name: "Validate HTML"
     needs: build
diff --git a/site/_layouts/default.html b/site/_layouts/default.html
@@ -9,6 +9,7 @@
     <title>{{ site.title }}</title>
     {%- endif -%}
     {% include styles.html %}
+    <link rel="canonical" href="{{ site.url }}{{ page.url | replace:'index.html',''}}">
   </head>
   <body>
     <header class="mb-4">
diff --git a/site/sitemap.xml b/site/sitemap.xml
@@ -0,0 +1,25 @@
+---
+layout:
+---
+
+<?xml version="1.0" encoding="UTF-8"?>
+<urlset xmlns="http://www.sitemaps.org/schemas/sitemap/0.9">
+  {%- for page in site.pages -%}
+    {%- if page.url contains '.xml' or page.url contains 'assets' %}{%- else %}
+    <url>
+        <loc>{{ site.url }}{{ page.url }}</loc>
+    </url>
+    {%- endif -%}
+  {% endfor %}
+  {%- for page in site.sections -%}
+    {%- if page.layout == "redirect" -%}{%- continue -%}{%- endif %}
+    <url>
+        <loc>{{ site.url }}{{ page.url | replace: 'index.html', '' }}</loc>
+    </url>
+  {%- endfor -%}
+  {%- for page in site.wiki %}
+    <url>
+        <loc>{{ site.url }}{{ page.url | replace: 'index.html', '' }}</loc>
+    </url>
+  {%- endfor %}
+</urlset>
diff --git a/tests/check-for-internal-absolute-links/check-for-internal-absolute-links.sh b/tests/check-for-internal-absolute-links/check-for-internal-absolute-links.sh
@@ -0,0 +1,28 @@
+#!/bin/bash
+#
+# usage:
+# ./$0 "jekyll root" "html root"
+#
+# for use in CI to ensure absolute links aren't being written to html files
+#
+set -e # exit on any error
+
+if [[ -z "$1" ]]; then
+  echo "Usage: $0 HTML_ROOT"
+  exit 127
+fi
+
+HTML_ROOT="$1"
+
+! grep -REi 'https?://(www.)?opentabletdriver.net' \
+  --exclude=sitemap.xml \
+  "${HTML_ROOT}" \
+  | grep -vE 'canonical'
+
+rv="$?"
+
+if [ "$rv" -eq 0 ]; then
+  echo "OK"
+else
+  echo "FAIL"
+fi
