Flutter and meta-flutter¶

Flutter is an open-source UI software development kit (SDK) created by Google for building natively compiled, multi-platform applications from a single codebase. Using the Dart programming language, it enables fast development for mobile (iOS, Android), web, desktop (Windows, macOS, Linux), and embedded devices. It features a "hot reload" function, allowing developers to view code changes instantly without losing app state.

Flutter in AGL¶

AGL embeds the meta-flutter embedder developed by Toyota (https://github.com/meta-flutter). It comes with its own workspace_automation tooling to setup a full development environment.

Setting up your workspace for AGL¶

If you clone AGL-repo, there will be a matching workspace-automation in ./external/workspace-automation and there will also be a matching ./external/workspace-config/ .

Quick start¶

The following steps are the gist of the full process. The detailed explanation follows.

#Use in-tree version in  ./external/workspace-automation/
#Call:
./flutter_workspace.py \
        --enable agl-ivi-demo-flutter-qemu-unagi \
        [ --flutter-version 3.38.3 ]
… wait …

source setup_env.sh         # all the magic ;)

# run AGL emulation
run-agl-ivi-demo-flutter-qemu

cd <path_to_your_flutter_project>

flutter run [-d agl-qemu-unagi]

Case 1: Basic Setup (no AGL APIs, same as 'Linux Desktop')¶

This will compile and run your flutter application on the Linux desktop (aka Ubuntu 22.x in our case).
You can develop, debug, run all on your PC.
No IDE
No specifics of the target environment!

./flutter_workspace.py --flutter-version 3.38.3
source setup_env.sh
cd <path_to_your_flutter_project>
flutter run -d linux

Pros and cons:

+ simple and straightforward setup
+ native performance running the app

- no target environment specifics / apis available
- integration to target environment necessary later-on
- no big difference to plain flutter development environment

Case 2: Setup with AGL/AGL APIs + Qemu¶

This will compile and run your flutter application for use within a VM that is running in the background.
You can develop, debug, run for the target environment.
No IDE, yet.

./flutter_workspace.py \ 
--enable agl-ivi-demo-flutter-qemu-unagi \ 
[ --flutter-version 3.38.3 ]
source setup_env.sh
run-agl-ivi-demo-flutter-qemu
cd <path_to_your_flutter_project>
flutter doctor -v 
should show: "AGL unagi-latest QEMU Image (mobile)" 
flutter run -d agl-qemu-unagi

Pros and cons:

+ simple and straightforward setup
+ target environment specifics / apis
+ no separate integration step lateron needed

- requires emulated target VM (qemu), thus slower

Case 3: AGL + QEMU + VScode integration¶

This will compile and run your flutter application for use within a VM that is running in the background.
You can develop, debug, run for the target environment.
Within IDE ;)

./flutter_workspace.py \ 
--enable agl-ivi-demo-flutter-qemu-unagi \
[--flutter-version 3.38.3]
source setup_env.sh
run-agl-ivi-demo-flutter-qemu
cd <path_to_your_flutter_project>
flutter doctor -v 
should show: "AGL unagi-latest QEMU Image (mobile)" 
code .
select run config under flutter and choose
"agl-unagi latest"

Pros and cons:

+ simple and straightforward setup
+ target environment specifics / apis
+ no separate integration step lateron needed
+ IDE integration

- requires emulated target VM (qemu), thus a bit slower