OML

Model mapping

OML is not a codec in the same sense as the other four. Its syntax is the Document model, so there is no mapping to describe — only a spelling. Every Document shape round-trips with zero adjustments: any nesting, any repeated label, any interleaving, in original order.

OML	Document
a: 1 / b: 2	[(a,1),(b,2)]
m: A / x: X / m: B	[(m,A),(x,X),(m,B)] — interleaving preserved
a: { b: "hi" }	[(a,[(b,"hi")])]

There is no list syntax, because there is no list in the model. An array is a label written more than once. There is no wrapper, no separator token between repetitions, and no way to distinguish a one-element array from a single value — they are the same Document.

The grammar is chapter 4; this page does not restate it.

The one gap. OML has no native temporal literal. A temporal value written to OML text becomes a string, and reading it back requires a schema to recover the type. On that single point TOML is more capable.

Worked example

The schema:

record Address  { "street": string, "city": string }
record LineItem { "sku": string, "qty": integer, "price": number }

record Order {
    "id":           string,
    "status":       string,
    "total":        number,
    "address":      Address,
    "items" [1,]:   LineItem,
    "coupon" [0,1]: string,
}

record Root { "order": Order }
root Root

The same order in OML:

order: {
    id: "A1"
    status: "shipped"
    total: 29.97
    address: { street: "1 Main"; city: "London" }
    items: { sku: "W"; qty: 3; price: 9.99 }
    items: { sku: "G"; qty: 1; price: 9.99 }
}

reads to:

[ (order, [ (id,      "A1"),
            (status,  "shipped"),
            (total,   29.97),
            (address, [ (street, "1 Main"), (city, "London") ]),
            (items,   [ (sku, "W"), (qty, 3), (price, 9.99) ]),
            (items,   [ (sku, "G"), (qty, 1), (price, 9.99) ]) ]) ]

The items label simply appears twice. That is the whole array mechanism, and it is the one place where reading the format text and reading the edge list are the same exercise — compare this block against the Document above line by line.

The Root wrapper is not needed by OML: OML carries several top-level edges without complaint. It is kept here so the example is one Document across all five formats rather than two (XML is the constraint).

Parity gaps

Chapter 9's status table (§9.3) is the authority. As of spec v0.1, "OML read" is complete in all three implementations (Core and Extended alike), and "OML canonical write" is complete in Python and Rust; TypeScript's canonical writer is partial.

D-5, previously open in this area, is now closed: all three implementations read OML-Extended (raw and multiline strings). Only the writer restriction remains one-sided by design — a canonical writer MUST NOT emit Extended spellings even though a reader MUST accept them.

Canonical output is not permitted to vary at all. Chapter 9's §9.2 puts the exact bytes an OML writer emits, and which OML texts parse, in the forbidden-variation list. A partial writer is a gap; a writer that emits different bytes is a conformance failure.