Chrome permission "window-management" not working

We are in the middle of changing the permission name, see https://crbug.com/1328581.

Meanwhile, here's a snippet (deployed here) that should work on all browsers:

document.querySelector('button').addEventListener('click', async () => {  
  let state;
  let permissionName;
  try {
    state = (
      await navigator.permissions.query({
        name: 'window-management',
      })
    ).state;
    permissionName = 'window-management'
  } catch (err) {    
    if (err.name === 'TypeError') {
      try {
        state = (
          await navigator.permissions.query({
            name: 'window-placement',
          })
        ).state;
        permissionName = 'window-placement'
      } catch (err) {
        if (err.name === 'TypeError') {
          state = 'Window management not supported';
        } else {
          state = `${err.name}: ${err.message}`;     
        }
      }
    } else {
      state = `${err.name}: ${err.message}`; 
    }
  }
  if ('request' in Permissions.prototype) {
    state = (await navigator.permissions.request({
      name: permissionName,
    })).state
  }
  document.querySelector('pre').textContent = state;
});

Slightly improved:

async function getWindowManagementPermissionState() {
  let state;
  // The new permission name.
  try {
    ({ state } = await navigator.permissions.query({
      name: "window-management",
    }));
  } catch (err) {
    if (err.name !== "TypeError") {
      return `${err.name}: ${err.message}`;
    }
    // The old permission name.
    try {
      ({ state } = await navigator.permissions.query({
        name: "window-placement",
      }));
    } catch (err) {
      if (err.name === "TypeError") {
        return "Window management not supported.";
      }
      return `${err.name}: ${err.message}`;
    }
  }
  return state;
}

document.querySelector("button").addEventListener("click", async () => {
  const state = await getWindowManagementPermissionState();
  document.querySelector("pre").textContent = state;
});

I'm guessing this is just a version issue, as window-management is rolling out in M111. @bradtriebwasser

@michaelwasserman thanks, that way it is the next update ;)

Current Version: 110.0.5481.78

Note that many popular Chromium derivates like Samsung Internet stay a long time on outdated Chromium versions. I highly recommend a defensive code pattern (also added to the article%3B)) if you use this API in production and you don't control what browsers people use. It's different in a production in-enterprise setting, where you can enforce a certain version.

@tomayac I completely understand what you are saying. I'm not going to disable either of the permission queries.

Currently my fallback is to query the single screen ( window.screen) and parse to a same object. I.e. window.screen as currentScreen and screen[0].

This way I don't have to implement multiple approaches.

@github-actions - 2024-08-30T02:27:04Z

This issue is stale because it has been open for 90 days with no activity.

• 2024-08-30T02:27:04Z @github-actions added the stale label

In case it's still an open question, this is the permission query that works today:

await navigator.permissions.query({name: 'window-management'});
// PermissionStatus {name: 'window-management', state: 'prompt', onchange: null}

@tomayac thanks for the update.

@Dinkh: i just tested it, and it works fine now.

See: https://developer.mozilla.org/en-US/docs/Web/API/Permissions/query

=> the queries are working inside dedicated & shared workers, so there is nothing to do here (just tested it).

no need for main thread based remote access (this would have been ugly anyways, since PermissionStatus is not spreadable & onchange can not get passed in postMessages).