From a622bf203f01d1ddb37b423b5cabd91045e16a48 Mon Sep 17 00:00:00 2001
From: Mark Eaton <MarkEEaton@users.noreply.github.com>
Date: Wed, 17 Sep 2025 14:23:26 -0400
Subject: [PATCH] feat: add script to audit guides for unique WCAG 2.1 AA
 issues using pa11y

Co-authored-by: aider (o3) <aider@aider.chat>
---
 check_wcag_unique.py | 100 +++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 100 insertions(+)
 create mode 100644 check_wcag_unique.py

diff --git a/check_wcag_unique.py b/check_wcag_unique.py
new file mode 100644
index 0000000..507d310
--- /dev/null
+++ b/check_wcag_unique.py
@@ -0,0 +1,100 @@
+#!/usr/bin/env python3
+"""
+Audit each guide URL in guides.csv for WCAG 2.1 AA accessibility issues
+(using pa11y).  Writes a report that omits issues appearing in *every*
+guide.
+
+Output file: output2.txt
+Usage:
+    python check_wcag_unique.py
+Prerequisite:
+    npm install -g pa11y            # or have pa11y available via npx
+"""
+from __future__ import annotations
+
+import csv
+import json
+import shutil
+import subprocess
+import sys
+from pathlib import Path
+from typing import Dict, List, Set
+
+GUIDES_CSV = Path(__file__).with_name("guides.csv")
+OUTPUT_FILE = Path("output2.txt")
+
+
+def run_pa11y(url: str) -> List[dict]:
+    """Run pa11y on the URL and return its JSON‐parsed list of issues."""
+    cmd_base = ["pa11y"] if shutil.which("pa11y") else ["npx", "pa11y"]
+    result = subprocess.run(
+        [*cmd_base, url, "--standard", "WCAG2AA", "--reporter", "json"],
+        capture_output=True,
+        text=True,
+        check=False,
+    )
+
+    # pa11y exits 0 (no issues) or 1 (issues found). Treat both as success.
+    if result.returncode not in (0, 1):
+        print(f"[ERROR] pa11y failed with code {result.returncode} for {url}", file=sys.stderr)
+        return []
+
+    try:
+        return json.loads(result.stdout)
+    except json.JSONDecodeError:
+        print(f"[WARN] Could not parse pa11y JSON output for {url}", file=sys.stderr)
+        return []
+
+
+def read_guides(csv_path: Path) -> List[Dict[str, str]]:
+    with csv_path.open(newline="", encoding="utf-8-sig") as fh:
+        return list(csv.DictReader(fh))
+
+
+def compute_common_codes(issues_by_guide: Dict[str, List[dict]]) -> Set[str]:
+    """Return set of issue codes present in *every* guide."""
+    common: Set[str] | None = None
+    for issues in issues_by_guide.values():
+        codes = {i["code"] for i in issues}
+        common = codes if common is None else common & codes
+    return common or set()
+
+
+def format_issue(issue: dict) -> str:
+    selector = (issue.get("selector") or "").strip() or "N/A"
+    return f"[{issue['type']}] {issue['code']}: {issue['message']}  (selector: {selector})"
+
+
+def main() -> None:
+    guides = read_guides(GUIDES_CSV)
+    issues_by_guide: Dict[str, List[dict]] = {}
+
+    print("Auditing guides with pa11y…\n")
+    for guide in guides:
+        name, url = guide["Name"], guide["URL"]
+        print(f"• {name} — {url}")
+        issues_by_guide[name] = run_pa11y(url)
+
+    common_codes = compute_common_codes(issues_by_guide)
+
+    lines: List[str] = []
+    lines.append("Accessibility report (WCAG 2.1 AA — unique issues only)")
+    lines.append("=" * 50)
+
+    for name, issues in issues_by_guide.items():
+        unique = [i for i in issues if i["code"] not in common_codes]
+        lines.append(f"\nGuide: {name}")
+        if not unique:
+            lines.append("  ✔ No unique issues found.")
+            continue
+        for issue in unique:
+            lines.append(f"  • {format_issue(issue)}")
+
+    report = "\n".join(lines)
+    print("\n" + report)
+    OUTPUT_FILE.write_text(report, encoding="utf-8")
+    print(f"\nReport written to {OUTPUT_FILE.resolve()}")
+
+
+if __name__ == "__main__":
+    main()