-
Notifications
You must be signed in to change notification settings - Fork 11
Conversation
f68af3c
to
c6e4687
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I don't see this being covered in the state machine tests, and this is touching a critical part of the state machine.
Also, I have to admit that I don't see a point in persisting a wallet safe mode state into a *.yaml
file when we have all the state covered without it?
If a user is having issues with the wallet and needs to run the wallet in safe mode, he should fix the underlying issues, not be awarded the safe mode option being run every time he runs the wallet.
This should be difficult for the user, simply because something is wrong.
@cleverca22 This is something I can get on board with if somebody has a really good reason for or some of the more experienced developers agree with you. |
The issue is electron doesn't support underlying GPU's that don't have GPU accceleration by default. A parameter needs passed to electron when it starts to tell it to use safe mode. We absolutely want to persist this as we don't want the user to have to get the blank screen every time they open Daedalus. |
Yes, I know.
This is the part that I find confusing. If Daedalus returns ExitCode 22, we enter safe mode? That was the agreed exit code for this. Also, the comments I made contain some issues that need fixing, like fixing the state machine tests if this is a serious PR since it changes the behavior of the launcher significantly. |
Other way around. The user gets the blank screen not in safe mode if they don't have GPU acceleration. exit code 22 instructs the launcher to restart daedalus process with safe mode flag which then shows the UI correctly. However, prior to this PR, when the user exits daedalus and restarts, it does not start in safe mode and they get the blank screen again. Note that FC5 release today is going to point at this branch. @cleverca22 can address the comments and make sure any interface changes end up in daedalus. We need to target getting this done before the mainnet daedalus release. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@disassembler This way of doing things will result in a failure of some kind, the state machine tests covered the specification nicely and this breaks that.
I hope that my comments will be addressed, approving since the version is around the corner.
5e71a60
to
979e34c
Compare
bors r+ |
979e34c
to
fcd827e
Compare
bors r- |
bors r+ |
bors ping |
pong |
bors r+ |
Whoops, saw bors in queue. |
bors was just added |
Build succeeded |
No description provided.