mirror of
https://github.com/invoke-ai/InvokeAI
synced 2024-08-30 20:32:17 +00:00
feat(ui): only fetch TIs on first load, add comment
This commit is contained in:
parent
cd55c23713
commit
4bb5aba70e
@ -18,9 +18,6 @@ export const addSocketConnectedEventListener = (startAppListening: AppStartListe
|
||||
effect: async (action, { dispatch, getState, cancelActiveListeners, delay }) => {
|
||||
log.debug('Connected');
|
||||
|
||||
// query TIs so they are ready if user opens trigger phrases
|
||||
dispatch(modelsApi.endpoints.getTextualInversionModels.initiate());
|
||||
|
||||
/**
|
||||
* The rest of this listener has recovery logic for when the socket disconnects and reconnects.
|
||||
*
|
||||
@ -33,6 +30,12 @@ export const addSocketConnectedEventListener = (startAppListening: AppStartListe
|
||||
|
||||
// Bail on the recovery logic if this is the first connection - we don't need to recover anything
|
||||
if ($isFirstConnection.get()) {
|
||||
// The TI models are used in a component that is not always rendered, so when users open the prompt triggers
|
||||
// box has a delay while it does the initial fetch. We need to both pre-fetch the data and maintain an RTK
|
||||
// Query subscription to it, so the cache doesn't clear itself when the user closes the prompt triggers box.
|
||||
// So, we explicitly do not unsubscribe from this query!
|
||||
dispatch(modelsApi.endpoints.getTextualInversionModels.initiate());
|
||||
|
||||
$isFirstConnection.set(false);
|
||||
return;
|
||||
}
|
||||
|
Loading…
Reference in New Issue
Block a user