grechabuf
v2.1.0
Published
Binary data serialization
Downloads
9
Readme
Grechabuf
binary data (de)serializer, highly inspired by Tsoding's multiplayer prototype
// just declare struct
const PersonStruct = grechabuf.createStruct({
name: grechabuf.string(),
roles: grechabuf.array(
grechabuf.string()
),
health: grechabuf.i8(),
food: grechabuf.i8(),
lostBrainCells: grechabuf.u32(),
})
// and just serialize!
const buffer = PersonStruct.serialize({
name: "John Doe",
roles: ["Admin", "TsodingSessionsEnjoyer"],
health: 100,
food: 77,
lostBrainCells: 69420
})
"why don't use my favourite Protobuf!?!??"
- Protobuf is bloated dependency (its like opening json in Visual Studio)
- Grechabuf has no dependencies (only
devDependencies
for building and testing) - Grechabuf is singlefile .ts/.cjs/.mjs to use
- Grechabuf is easy to expand (see custom fields)
- Grechabuf has TypeScript support
installation
- With NPM (with bundler on web)
npm install grechabuf yarn add grechabuf pnpm add grechabuf
- From Github:
- If you using Javascript: go to
/dist
folder and grab here .cjs/.mjs files - If you using Typescript: grab
grachabuf.ts
and copy it to your project
- If you using Javascript: go to
examples
custom fields
type Vector2 = { x: number, y: number }
// deserialized type ---vvvvvvv
const vector2 = (): Field<Vector2> => {
return {
// calculate size for certain value (useful for strings/arrays/other dynamic values)
size(value) {
return 4 + 4
},
// serialization
// takes DataView, position (byteOffset) and value (javascript value)
// returns count of written bytes
serialize(view, position, value) {
view.setFloat32(position, value.x)
view.setFloat32(position + 4, value.y)
return 4 + 4
},
// deserialization
// takes DataView, position (byteOffset)
// returns object
deserialize(view, position) {
return {
// deserialized data
data: {
x: view.getFloat32(position),
y: view.getFloat32(position + 4)
},
// count of read bytes
length: 4 + 4
}
}
}
}
// And then use it!
const SomeStruct = grechabuf.createStruct({
someField: vector2()
})
SomeStruct.serialize({
someField: { x: 123.4, y: 567.8 }
})