Details
This whole stack was tested with the following steps:
- On native, attempting to log in with an ETH account that hasn’t been registered yet
- On native, attemping to register with an ETH account that has already been registered
- On native, log in with an ETH account that has already been registered
- On native, register with an ETH account that hasn’t already been registered
- On native, register with a password account
- On native, log in with a password account
- On web, attempting to log in with an ETH account that hasn’t been registered yet
- On web, log in with an ETH account that has already been registered
- On web, log in with a password account
Diff Detail
- Repository
- rCOMM Comm
- Lint
Lint Not Applicable - Unit
Tests Not Applicable
Event Timeline
lib/hooks/login-hooks.js | ||
---|---|---|
57–59 ↗ | (On Diff #39451) | In D11763 we check if step is inactive. Do we need to do that here? |
63 ↗ | (On Diff #39451) | This probably shouldn't be named "registration". And, same as in D11763, I think this shouldn't be named "waiting for", when later it represents the state after registration has succeeded. |
lib/hooks/login-hooks.js | ||
---|---|---|
57–59 ↗ | (On Diff #39451) | Good call! Yes, I'll make that change |
63 ↗ | (On Diff #39451) |
Good point, I'll make this change
As described there, we're really waiting for data to appear in Redux, but I thought that name was a bit of a mouthful. Open to other suggestions, but opposed to identity_registration here because we're really doing keyserver auth during this step. |
- waiting_for_identity_registration -> waiting_for_identity_login_in_redux
- Add a check that step is inactive in order to avoid triggering auth multiple times