You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
SessionStorage is per-tab (with navigation-inheritance from navigated-from tabs), while the rest of these storage systems are per-origin. This means that if you have multiple tabs / contexts, and they are all doing changes, then sessionstorage will always be wrong on all of them. Also, it will have to have unique data per tab / context, so you'll get a per-tab copy of the data in session storage on disk, which means (other than the already duplicate data being stored in other storage systems), tabs*data size extra data.
Also, the first get or put (on both SessionStorage and LocalStorage) actually does a synchronous read from disk. This is unfortunately by design of the very old spec, as the API is synchronous. So just keep that in mind, this isn't great for the user, as the page freezes, and it can be bad for machines that are old / spinning disk / slow.
(I'd also recommend just using IndexedDB by itself and not using any of these other storage backends... but that's a separate issue, and probably not a constructive one)
The text was updated successfully, but these errors were encountered:
SessionStorage is per-tab (with navigation-inheritance from navigated-from tabs), while the rest of these storage systems are per-origin. This means that if you have multiple tabs / contexts, and they are all doing changes, then sessionstorage will always be wrong on all of them. Also, it will have to have unique data per tab / context, so you'll get a per-tab copy of the data in session storage on disk, which means (other than the already duplicate data being stored in other storage systems), tabs*data size extra data.
Also, the first get or put (on both SessionStorage and LocalStorage) actually does a synchronous read from disk. This is unfortunately by design of the very old spec, as the API is synchronous. So just keep that in mind, this isn't great for the user, as the page freezes, and it can be bad for machines that are old / spinning disk / slow.
(I'd also recommend just using IndexedDB by itself and not using any of these other storage backends... but that's a separate issue, and probably not a constructive one)
The text was updated successfully, but these errors were encountered: