Change Details

As mentioned in D9536, we can now simplify the check in `rawMessageInfoForRowsAndRelatedMessages` to only filter out what is needed (`EDIT_MESSAGE` and `REACTION`). To do this, we'll call `isUnableToBeRenderedOnItsOwn`, which will check the message type's `messageSpec`, and then return the corresponding flag. This isolates the single source of truth to the individual message specs. There is a usage of `FlowFixMe` here. I'll copy my explanation for using it here in the code as well so it's closer to it's usage, but essentially: Flow doesn't realize that only certain message types have the flag set as false, so we never get the type refinement we're expecting to set the `rawMessageInfo` into the map. The `continue` will ensure we don't set an invalid type into the map anyways, but once again Flow doesn't realize that. The only workaround I found waas to basically `invariant` against every message type we don't expect to see (i.e. `REACˇION` and `EDIT_MESSAGE`), but I want to avoid that since that 1) removes the 'single source of truth' that we've been working to achieve with this stack, and 2) makes it easier for future changes to this code to forget to update this invariant and then something will inevitably go wrong. Resolves https://linear.app/comm/issue/ENG-5217/update-rawmessageinfoforrowsandrelatedmessages-to-check-for Depends on D9536

As mentioned in D9536, we can now simplify the check in `rawMessageInfoForRowsAndRelatedMessages` to only filter out what is needed (`EDIT_MESSAGE` and `REACTION`). To do this, we'll call `isUnableToBeRenderedOnItsOwn`, which will check the message type's `messageSpec`, and then return the corresponding flag. This isolates the single source of truth to the individual message specs. There is a usage of `FlowFixMe` here. I'll copy my explanation for using it here in the code as well so it's closer to it's usage, but essentially: Flow doesn't realize that only certain message types have the flag set as false, so we never get the type refinement we're expecting to set the `rawMessageInfo` into the map. The `continue` will ensure we don't set an invalid type into the map anyways, but once again Flow doesn't realize that. The only workaround I found was to basically `invariant` against every message type we don't expect to see (i.e. `REACTION` and `EDIT_MESSAGE`), but I want to avoid that since that 1) removes the 'single source of truth' that we've been working to achieve with this stack, and 2) makes it easier for future changes to this code to forget to update this invariant and then something will inevitably go wrong. Resolves https://linear.app/comm/issue/ENG-5217/update-rawmessageinfoforrowsandrelatedmessages-to-check-for Depends on D9536

As mentioned in D9536, we can now simplify the check in `rawMessageInfoForRowsAndRelatedMessages` to only filter out what is needed (`EDIT_MESSAGE` and `REACTION`). To do this, we'll call `isUnableToBeRenderedOnItsOwn`, which will check the message type's `messageSpec`, and then return the corresponding flag. This isolates the single source of truth to the individual message specs. There is a usage of `FlowFixMe` here. I'll copy my explanation for using it here in the code as well so it's closer to it's usage, but essentially: Flow doesn't realize that only certain message types have the flag set as false, so we never get the type refinement we're expecting to set the `rawMessageInfo` into the map. The `continue` will ensure we don't set an invalid type into the map anyways, but once again Flow doesn't realize that. The only workaround I found waas to basically `invariant` against every message type we don't expect to see (i.e. `REACˇTION` and `EDIT_MESSAGE`), but I want to avoid that since that 1) removes the 'single source of truth' that we've been working to achieve with this stack, and 2) makes it easier for future changes to this code to forget to update this invariant and then something will inevitably go wrong. Resolves https://linear.app/comm/issue/ENG-5217/update-rawmessageinfoforrowsandrelatedmessages-to-check-for Depends on D9536