I think parted only accepts absolutes, not x+y.
You could do it like this: (+ interpreted by shell)
# start=1
# size=512
# parted /dev/loop0 unit mib mkpart primary $start $(($start+$size))
# parted /dev/loop0 unit mib print free
Model: Loopback device (loopback)
Disk /dev/loop0: 1000MiB
Sector size (logical/physical): 512B/512B
Partition Table: gpt
Disk Flags:
Number Start End Size File system Name Flags
0.02MiB 1.00MiB 0.98MiB Free Space
1 1.00MiB 513MiB 512MiB primary
513MiB 1000MiB 487MiB Free Space
...and if you want to append more, then
# start=$(($start+$size+1))
# size=128
# parted /dev/loop0 unit mib mkpart primary $start $(($start+$size))
# parted /dev/loop0 unit mib print free
Model: Loopback device (loopback)
Disk /dev/loop0: 1000MiB
Sector size (logical/physical): 512B/512B
Partition Table: gpt
Disk Flags:
Number Start End Size File system Name Flags
0.02MiB 1.00MiB 0.98MiB Free Space
1 1.00MiB 513MiB 512MiB primary
513MiB 514MiB 1.00MiB Free Space
2 514MiB 642MiB 128MiB primary
642MiB 1000MiB 358MiB Free Space
The +1 is entirely optional, I just like leaving 1MiB free "bumpers" between partitions, for some reason.
If you want to deal with percentages, you can obtain the device size from the shell also, using blockdev --getsize64 /dev/device, so you can move all the math out of parted into a shell script. I do this to guarantee MiB-aligned partitions, as I'm never too sure about what parted means exactly by -a optimal.