Encoding/HTML: cross-document encoding inheritance

Make Encoding tests no longer rely on an interoperability issue and simultaneously add tests for that issue. The latter tests align with the HTML Standard, but that might not necessarily be the path we want to take. whatwg/html#8789 can be used for further discussion.
web-platform-tests · Apr 4, 2023 · a58bbf6 · a58bbf6
1 parent cbd294b
commit a58bbf6
Show file tree

Hide file tree

Showing 7 changed files with 105 additions and 4 deletions.
diff --git a/encoding/resources/text-html-meta-charset.py b/encoding/resources/text-html-meta-charset.py
@@ -0,0 +1,3 @@
+def main(request, response):
+  response.headers.set(b"Content-Type", b"text/html")
+  response.content = b"<meta charset=\"" + request.GET.first(b"label") + b"\">hello encoding"
diff --git a/encoding/unsupported-labels.window.js b/encoding/unsupported-labels.window.js
@@ -167,11 +167,13 @@
     t.add_cleanup(() => {
       frame.remove();
     });
-    frame.src = "resources/text-plain-charset.py?label=" + label;
+    // Intentionally use <meta> as Content-Type results in browser differences
+    // See /html/syntax/charset/inheritance-bogus-meta.html
+    frame.src = "resources/text-html-meta-charset.py?label=" + label;
     frame.onload = t.step_func_done(() => {
-      // If we ever change this default this needs adjusting accordingly.
-      assert_equals(frame.contentDocument.characterSet, "windows-1252");
-      assert_equals(frame.contentDocument.inputEncoding, "windows-1252");
+      // UTF-8 as it inherits from the parent document when unrecognized
+      assert_equals(frame.contentDocument.characterSet, "UTF-8");
+      assert_equals(frame.contentDocument.inputEncoding, "UTF-8");
     });
     document.body.append(frame);
   }, `${label} is not supported by the Encoding Standard`);

diff --git a/html/syntax/charset/inheritance-bogus-meta-utf-8.html b/html/syntax/charset/inheritance-bogus-meta-utf-8.html
@@ -0,0 +1,35 @@
+<!doctype html>
+<meta charset="utf-8" />
+<title>Inheriting from UTF-8 parent</title>
+<script src=/resources/testharness.js></script>
+<script src=/resources/testharnessreport.js></script>
+<div id=log></div>
+<script>
+[
+  {
+    "title": "Child with bogus <meta charset>",
+    "url": "resources/bogus-charset.html",
+    "expected": "\uFFFD\n" // 0x00A2 in windows-1252 is 0xFFFD in UTF-8
+  },
+  {
+    "title": "Child with bogus Content-Type charset",
+    "url": "resources/bogus-charset-http.py",
+    "expected": "\uFFFD\n"
+  },
+  {
+    "title": "Child with bogus Content-Type charset, but valid <meta charset>",
+    "url": "resources/bogus-charset-http-valid-meta.py",
+    "expected": "\u045E\n"
+  }
+].forEach(({ title, url, expected }) => {
+  async_test(t => {
+    const frame = document.createElement("iframe");
+    t.add_cleanup(() => frame.remove());
+    frame.src = url;
+    frame.onload = t.step_func_done(() => {
+      assert_equals(frame.contentDocument.body.textContent, expected);
+    });
+    document.body.append(frame);
+  }, title);
+});
+</script>
diff --git a/html/syntax/charset/inheritance-bogus-meta.html b/html/syntax/charset/inheritance-bogus-meta.html
@@ -0,0 +1,46 @@
+<!doctype html>
+<meta charset=windows-1253>
+<title>Inheriting from windows-1253 parent</title>
+<script src=/resources/testharness.js></script>
+<script src=/resources/testharnessreport.js></script>
+<script src=/common/get-host-info.sub.js></script>
+<div id=log></div>
+<script>
+[
+  {
+    "title": "Child with bogus <meta charset>",
+    "url": "resources/bogus-charset.html",
+    "expected": "¢\n" // 0x00A2 in windows-1252 is at the same position as 0x0386 in windows-1253
+  },
+  {
+    "title": "Child with bogus Content-Type charset",
+    "url": "resources/bogus-charset-http.py",
+    "expected": "¢\n"
+  },
+  {
+    "title": "Child with bogus Content-Type charset, but valid <meta charset>",
+    "url": "resources/bogus-charset-http-valid-meta.py",
+    "expected": "\u045E\n"
+  }
+].forEach(({ title, url, expected }) => {
+  async_test(t => {
+    const frame = document.createElement("iframe");
+    t.add_cleanup(() => frame.remove());
+    frame.src = url;
+    frame.onload = t.step_func_done(() => {
+      assert_equals(frame.contentDocument.body.textContent, expected);
+    });
+    document.body.append(frame);
+  }, title);
+});
+
+async_test(t => {
+  self.onmessage = t.step_func_done(({ data }) => {
+    assert_equals(data, "\u00A2\n");
+  });
+  const frame = document.createElement("iframe");
+  t.add_cleanup(() => frame.remove());
+  frame.src = get_host_info().HTTP_REMOTE_ORIGIN + new URL("resources/bogus-charset.html", location).pathname + "?postMessage";
+  document.body.append(frame);
+}, "Cross-origin child with bogus <meta charset>");
+</script>
diff --git a/html/syntax/charset/resources/bogus-charset-http-valid-meta.py b/html/syntax/charset/resources/bogus-charset-http-valid-meta.py
@@ -0,0 +1,4 @@
+def main(request, response):
+  response.headers.set(b"Content-Type", b"text/html;charset=this-is-not-a-charset")
+  # ¢
+  response.content = b"<meta charset=windows-1251>\xA2\n"
diff --git a/html/syntax/charset/resources/bogus-charset-http.py b/html/syntax/charset/resources/bogus-charset-http.py
@@ -0,0 +1,4 @@
+def main(request, response):
+  response.headers.set(b"Content-Type", b"text/html;charset=this-is-not-a-charset")
+  # ¢
+  response.content = b"\xA2\n"
diff --git a/html/syntax/charset/resources/bogus-charset.html b/html/syntax/charset/resources/bogus-charset.html
@@ -0,0 +1,7 @@
+<!doctype html><meta charset=this-is-not-a-charset><script>
+onload = () => {
+  if (location.search === "?postMessage") {
+    parent.postMessage(document.body.textContent, "*");
+  }
+}
+</script>¢