Add entries and length to MemoryHistory

[AlbertoSadoc]

What is the new or updated feature that you are suggesting?

The old MemoryHistory had those 2 properties that could be accessed in tests to check that the test navigated correctly through the paths provided

Why should this feature be included?

Those properties existed in v5 and provided a quick and easy way to check all the paths. It will also make the upgrading easier, even if you mark the properties as deprecated to be removed later on

[joanvf]

Hi, is there any plan to address this? Our use case requires to know when to enable or disable our app's back and forward buttons. As of today, users click on the back or forward buttons and nothing happens if there is no history. In this case, these buttons should have been disabled. How can we achieve this?

Many thanks

[brophdawg11]

Generally the way to do this is to look for a location.key of default which is the value RR will set on the very initial load. And navigations after that will have a unique key, and if the key is not default that is indicative of there being a location to navigate back to. Memory history is currently aligned to the API of window.history and I'm not sure we necessarily want to fork the API to differ between memory/browser/hash usages.

For testing purposes, it feels like you're trying to unit test internal RR behavior which is already well tested. But if you need to do some of your own tests, instead of relying on internal entries - could you assert against the rendered output? Even if that involves rendering the location.key?

[brophdawg11]

I'm going to convert this to a discussion so it can go through our new Open Development process. Please upvote the new Proposal if you'd like to see this considered!

[3alisaki]

i have same problem when there is no history the back or forward buttons should be disabled but i don't have any way to do that (the above way is only working for back button not for forward button).

[Donorlin]

Following on remix-run/history#939

In our company, we are running SPA in the SSR (company historical reasons). This means the SPA state is lost whenever a site is reloaded. In version history@4 we are able to save entries and index from the history object and pass it to initialEntries and initialIndex of the MemoryRouter to seed it back to its previous state (controlled behavior).

In react-router@6 we are still able to access the MemoryHistory object with UNSAFE_NavigationContext, but not its entries. So we are not able to implement a controlled MemoryRouter.

sadly because of this, we are not able to upgrade to v6
this change wasnt even mentioned in changelog (not even a deprecation warning)

[Nantris]

I'd really appreciate if this feature was restored. We've been stuck and unable to upgrade for years now.