Any ideas on migrating Anka templates to Tart?

[remarkov]

We have a heavily customized Anka template which would be nice to just migrate to Tart and avoid doing everything from scratch.
I totally understand that in most use cases this would be absolutely useless, especially if Packer is used for the VM setup but since both use the same Apple Virtualization I thought maybe there's a way to migrate a VM from Anka format to the one used by Tart?
Looking at the file structure I doubt so but would like to hear what Tart devs and advanced users think.

[fkorotkov]

At Cirrus Labs we hesitant to even look into Anka or Anka format in order to not break proprietary something. It seems to be a grey area since it's not a reverse engineering of the software per se but only looking into the output format... But I'm not a lawyer and would check with Veertu Labs to get some written permit to do such conversion from their format into Tart's format. Or consult with a knowledgable lawyer to see if reverse engineering of the output format is a potential violation of something.

But I can tell you what Virtualization.Framework needs to run a VM and where it's stored in Tart's VM directory (~/.tart/vms/VM_NAME) so one can construct it manually or automatically from any source that uses Virtualization.Framework.

To run a VM Virtualization.Framework uses 4 things:

1. The main disk image is located at ~/.tart/vms/VM_NAME/disk.img
2. Data used by the boot loader is stored in ~/.tart/vms/VM_NAME/nvram.bin.
3. ecid field in ~/.tart/vms/VM_NAME/config.json is just a base64 representation of VZMacMachineIdentifier.
4. hardwareModel field in ~/.tart/vms/VM_NAME/config.json is just a base64 representation of VZMacHardwareModel.

[remarkov]

Thank you Fedor, I totally understand and appreciate your point. Keep up the great work you do!