From fab03eb8cd0a493589dce2e98d4dca7063694e3e Mon Sep 17 00:00:00 2001
From: Felix Fontein <felix@fontein.de>
Date: Fri, 14 Jan 2022 07:34:08 +0100
Subject: [PATCH] Validate generated HTML.

---
 .github/workflows/antsibull-docs.yml |  6 +++++-
 .github/workflows/validate-html.py   | 26 ++++++++++++++++++++++++++
 2 files changed, 31 insertions(+), 1 deletion(-)
 create mode 100644 .github/workflows/validate-html.py

diff --git a/.github/workflows/antsibull-docs.yml b/.github/workflows/antsibull-docs.yml
index 66e043e8..4603dee5 100644
--- a/.github/workflows/antsibull-docs.yml
+++ b/.github/workflows/antsibull-docs.yml
@@ -34,7 +34,7 @@ jobs:
       - name: Install dependencies
         run: |
           python -m pip install --upgrade pip
-          python -m pip install . coverage codecov
+          python -m pip install . coverage codecov html5lib
 
       - name: Use antsibull-docs sphinx-init
         run: |
@@ -59,6 +59,10 @@ jobs:
         run: |
           ./build.sh
 
+      - name: Validate HTML
+        run:
+          python .github/workflows/validate-html.py build/html/
+
       - name: Combine and upload coverage stats
         run: |
           coverage combine .coverage.*
diff --git a/.github/workflows/validate-html.py b/.github/workflows/validate-html.py
new file mode 100644
index 00000000..7af1c276
--- /dev/null
+++ b/.github/workflows/validate-html.py
@@ -0,0 +1,26 @@
+#!/usr/bin/python3
+
+import os
+import sys
+
+import html5lib
+
+
+errors = 0
+for path in sys.argv[1:]:
+    for dirname, _, files in os.walk(path):
+        for file in files:
+            if not file.endswith('.html'):
+                continue
+            path = os.path.join(dirname, file)
+            try:
+                parser = html5lib.HTMLParser(strict=True)
+                with open(path, 'rb') as f:
+                    document = parser.parse(f)
+            except Exception as e:
+                errors += 1
+                print(f'{path}: {e}')
+
+if errors > 0:
+    print(f'Found {errors} errors!')
+    sys.exit(1)