1 files changed, 41 insertions, 0 deletions
diff --git a/testing/web-platform/tests/compression/decompression-with-detach.tentative.window.js b/testing/web-platform/tests/compression/decompression-with-detach.tentative.window.js
new file mode 100644
index 0000000000..1ff9c26983
--- /dev/null
+++ b/testing/web-platform/tests/compression/decompression-with-detach.tentative.window.js
@@ -0,0 +1,41 @@
+// META: global=window,worker,shadowrealm
+// META: script=resources/concatenate-stream.js
+
+'use strict';
+
+const kInputLength = 1000000;
+
+async function createLargeCompressedInput() {
+  const cs = new CompressionStream('deflate');
+  // The input has to be large enough that it won't fit in a single chunk when
+  // decompressed.
+  const writer = cs.writable.getWriter();
+  writer.write(new Uint8Array(kInputLength));
+  writer.close();
+  return concatenateStream(cs.readable);
+}
+
+promise_test(async () => {
+  const input = await createLargeCompressedInput();
+  const ds = new DecompressionStream('deflate');
+  const writer = ds.writable.getWriter();
+  writer.write(input);
+  writer.close();
+  // Object.prototype.then will be looked up synchronously when the promise
+  // returned by read() is resolved.
+  Object.defineProperty(Object.prototype, 'then', {
+    get() {
+      // Cause input to become detached and unreferenced.
+      try {
+        postMessage(undefined, 'nowhere', [input.buffer]);
+      } catch (e) {
+        // It's already detached.
+      }
+    }
+  });
+  const output = await concatenateStream(ds.readable);
+  // If output successfully decompressed and gave the right length, we can be
+  // reasonably confident that no data corruption happened.
+  assert_equals(
+      output.byteLength, kInputLength, 'output should be the right length');
+}, 'data should be correctly decompressed even if input is detached partway');