Incorrect disk format after upgrading to provider `0.9.1`

[guoard]

After upgrading to provider version 0.9.1, I updated my libvirt_volume configuration from:

resource "libvirt_volume" "debian_disk" {
  for_each = var.vms

  name   = "${each.key}.qcow2"
  pool   = libvirt_pool.pool.name
  format = "qcow2"

  create = {
    content = {
      url = "file://${var.debian_12_qcow_path}"
    }
  }
}

to the new schema:

resource "libvirt_volume" "debian_disk" {
  for_each = var.vms

  name   = "${each.key}.qcow2"
  pool   = libvirt_pool.pool.name
  target = { format = { type = "qcow2" } }

  create = {
    content = {
      url = "file://${var.debian_12_qcow_path}"
    }
  }
}

However, despite specifying qcow2, the generated domain XML shows the disk being created as raw:

<disk type='volume' device='disk'>
  <driver name='qemu' type='raw'/>
  <source pool='default' volume='vm1.qcow2' index='2'/>
  <backingStore/>
  <target dev='vda' bus='virtio'/>
  <alias name='virtio-disk0'/>
  <address type='pci' domain='0x0000' bus='0x00' slot='0x03' function='0x0'/>
</disk>

It seems the format value is being ignored or incorrectly interpreted after the upgrade. The resulting volume is created with a raw driver instead of qcow2.

[nevermindwv]

The "format" of the disk can be defined, as shown in your libvirt XML configuration by the "driver" context under domain.

i.e.

devices = {
    disks = [
      {
        driver = {
          type = "qcow2"
        }
        source = {
          file = {
            file = libvirt_volume.base.id
          }
        }
        target = {
          dev = "vda"
          bus = "sata"
        }
      }
    ]
}

Hope this helps.