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
@fesch What do you think about the Andoid Auto rave, now that the worms are out of the can with the latest "hack"? I did a quick read up and while I fancy screen casting myself, it doesn't seem AA is "intended that way". Maybe I am misreading.
Screen mirror the phone's screen onto the car's display would be really great and solve my problem of having another device fixed somewhere in the car, but I'm am not sure if we really should encourage people to do this hack....
The hack described above can work for testing (it's still working given the xda-developpers thread on the subject), but there may be a way to make it available for general public if Google admits the app in the whitelist.
I'd like to volounteer for the AAuto SDK part, but I have like literally zero experience in Java (C/C++ coder here) and I'd need a hand with the "make the UI comply with what Google wants" part.
You may remember me from the comments about Zoe with Nissan Leaf battery (actually it's installed and working now) and getting CanZE to work on the car's screen would be a great help here since the BMS firmware is still in development.
This issue has been sleeping and I apologize for not getting back earlier. My flimsy defense is that I only enabled AA a few weeks ago.
I find the UI a bit confusing at the moment. (Try to cancel a destination in Waze using only the R-Link head-end).
From a UI perspective it looks like it's either pushing "cards", or doing a full screen app. If anything, we should limit ourselves to an equivalent to the consumption screen, more or less as @fesch has been doing the last couple of days for the "dashcam Android" device.
BUT...... I have peeked at the active installs the other day in the play store console and that triggers me to bring up another issue entirely. We need to be very careful where to put our energy. There is really no other incentive, as it seems, than for the two current developers to learn doing AA stuff. Same is going on with the DIY dongle. For neither I expect more than a handful of users, at most.
The way I look at it today is CanZE for AA would be a complete hobby project and the not inconsiderable energy required to make it A) work B) available is only justified by a developer "fun" factor. Which I lack at the moment. Unless somebody really is inclined to pick up the challenge, I propose to close this issue.
Unless somebody really is inclined to pick up the challenge
raises hand
I've played with the android auto support jar, sounds like I could do this on a hobby basis slowly, but from what it seems to be now (disclaimer: not an Android coder per se), the activities displayed on the car need to have separate activity classes deriving from a different hierarchy than the current phone ones. I.e. for code reuse most of the stuff needs to be converted to fragments and then pulled into 2 separate activities. Is that a feasible thing that can be done?
Other thing is that the jar is supposedly coming from a dubious source, I guess there is some diplomacy to be made with Google for them to provide the official one (after meeting the criteria)
Activity
yoh-there commentedon Jan 22, 2018
@fesch What do you think about the Andoid Auto rave, now that the worms are out of the can with the latest "hack"? I did a quick read up and while I fancy screen casting myself, it doesn't seem AA is "intended that way". Maybe I am misreading.
fesch commentedon Jan 22, 2018
Screen mirror the phone's screen onto the car's display would be really great and solve my problem of having another device fixed somewhere in the car, but I'm am not sure if we really should encourage people to do this hack....
landswellsong commentedon Jun 13, 2018
@fesch @yoh-there my 5¢. The way I see it, they made a whitelist of the apps that are able to screencast, but it's definitely possible to be on that list, as seen by having Waze. I think there is an application procedure for this, but Google requires a few guidelines about the UI:
https://developer.android.com/docs/quality-guidelines/auto-app-quality
https://developer.android.com/distribute/best-practices/launch/distribute-auto
The hack described above can work for testing (it's still working given the xda-developpers thread on the subject), but there may be a way to make it available for general public if Google admits the app in the whitelist.
I'd like to volounteer for the AAuto SDK part, but I have like literally zero experience in Java (C/C++ coder here) and I'd need a hand with the "make the UI comply with what Google wants" part.
You may remember me from the comments about Zoe with Nissan Leaf battery (actually it's installed and working now) and getting CanZE to work on the car's screen would be a great help here since the BMS firmware is still in development.
yoh-there commentedon Jan 28, 2019
This issue has been sleeping and I apologize for not getting back earlier. My flimsy defense is that I only enabled AA a few weeks ago.
I find the UI a bit confusing at the moment. (Try to cancel a destination in Waze using only the R-Link head-end).
From a UI perspective it looks like it's either pushing "cards", or doing a full screen app. If anything, we should limit ourselves to an equivalent to the consumption screen, more or less as @fesch has been doing the last couple of days for the "dashcam Android" device.
BUT...... I have peeked at the active installs the other day in the play store console and that triggers me to bring up another issue entirely. We need to be very careful where to put our energy. There is really no other incentive, as it seems, than for the two current developers to learn doing AA stuff. Same is going on with the DIY dongle. For neither I expect more than a handful of users, at most.
The way I look at it today is CanZE for AA would be a complete hobby project and the not inconsiderable energy required to make it A) work B) available is only justified by a developer "fun" factor. Which I lack at the moment. Unless somebody really is inclined to pick up the challenge, I propose to close this issue.
fesch commentedon Jan 28, 2019
I agree on that ...
landswellsong commentedon Jun 19, 2019
raises hand
I've played with the android auto support jar, sounds like I could do this on a hobby basis slowly, but from what it seems to be now (disclaimer: not an Android coder per se), the activities displayed on the car need to have separate activity classes deriving from a different hierarchy than the current phone ones. I.e. for code reuse most of the stuff needs to be converted to fragments and then pulled into 2 separate activities. Is that a feasible thing that can be done?
Other thing is that the jar is supposedly coming from a dubious source, I guess there is some diplomacy to be made with Google for them to provide the official one (after meeting the criteria)
landswellsong commentedon Nov 20, 2020
@fesch @yoh-there hey-hey
https://android-developers.googleblog.com/2020/10/introducing-android-for-cars-app-library.html
yoh-there commentedon Nov 20, 2020
Fun, thank you!