RPIC

Blocks

Square brackets group a sub-picture into one composite object: it has its own current position inside, and from the outside it behaves like a single box — with dimensions, compass anchors, and a center.

Open in Studio
.PS
A: [
circle rad 0.18 "cpu"
arrow right 0.4
box wid 0.5 ht 0.36 "ram"
]
box "one object" wid 1.1 ht 0.4 with .n at A.s + (0, -0.3)
arrow from A.s to last box.n
.PE
cpu ram one object

Placing blocks

Blocks accept the same placement attributes as any object — at, with, and they stack along the current direction like everything else:

Open in Studio
.PS
define unit {
[ B: box wid 0.5 ht 0.4 $1
  line down 0.18 from B.s
  circle rad 0.045 fill 0 at last line.end ]
}
unit("a"); move right 0.3; unit("b"); move right 0.3; unit("c")
.PE
a b c

Labels inside blocks

Objects labeled inside a block are reachable with dotted paths from outside — A.Inner — so blocks can expose terminals:

Open in Studio
.PS
Chip: [
B: box wid 0.9 ht 0.6 "chip"
In:  B.w
Out: B.e
]
arrow to Chip.In from Chip.In - (0.5, 0)
arrow from Chip.Out right 0.5
.PE
chip

This is how the circuit library models multi-terminal devices: bi_tr and opamp are blocks exposing their pins as labels.

Variables stay global

Unlike labels, plain variables are not scoped by blocks — an assignment inside propagates out (dpic semantics). Treat blocks as geometric grouping, not lexical scope.

Where to go next