Adding basic types and concept of sharing information between worker and app. Current type (PING/PONG) will be used probably only for development, will add more concrete ones in next diffs.
Depends on D6990
Details
Details
Run this code:
const webDbWorker = new SharedWorker('/worker/webDatabase');
webDbWorker.port.onmessage = function (e) {
console.log('message from worker: ', e.data);
};
const message: WorkerRequestMessage = {
type: workerRequestMessageTypes.PING,
text: 'hi from app',
};
webDbWorker.port.postMessage(message);and check results in browser, desktop and shared worker consoles.
Check error flow.
Diff Detail
Diff Detail
- Repository
- rCOMM Comm
- Lint
Lint Not Applicable - Unit
Tests Not Applicable
Event Timeline
| web/database/worker/web-db-worker.js | ||
|---|---|---|
| 40 ↗ | (On Diff #23509) | That's a problematic line but unfortunately messageEvent.data is typed in flow as mixed and there isn't a better way to cast this to our exact type. We can not override MessageEvent because even if we override all types there is a place where there will be a need to cast via any (there always will be a difference between what we have and flow's SharedWorker definition). I think this is the safest place because we will receive data from the other side of the pipe here |
| web/database/worker/web-db-worker.js | ||
|---|---|---|
| 40 ↗ | (On Diff #23509) | This is fine, and I think you're doing it well |
| web/database/worker/web-db-worker.js | ||
|---|---|---|
| 40 ↗ | (On Diff #23509) | Thanks for confirmation |
| web/database/worker/web-db-worker.js | ||
|---|---|---|
| 24 ↗ | (On Diff #23509) | Yes, it will, there will be messages (like INIT) for which resolving/rejecting promises will be the only information. We could avoid sending additional response data between app<->worker with success status when promise could do the same with fewer data transmitted. I know it's not logically consistent for this diff but the following diff in the stack (D6993) shows why this should not be mandatory. |
| 46–48 ↗ | (On Diff #23509) | Oops, I test that but didn't include it in the test plan, fixing |