# SimpleSerialiZe (SSZ)

This is a **work in progress** describing typing, serialization and Merkleization of Ethereum 2.0 objects.

## Table of contents

- [Constants](#constants)
- [Typing](#typing)
    - [Basic types](#basic-types)
    - [Composite types](#composite-types)
    - [Aliases](#aliases)
- [Serialization](#serialization)
    - [`"uintN"`](#uintn)
    - [`"bool"`](#bool)
    - [Vectors, containers, lists](#composite-types-vectors-containers-and-lists)
- [Deserialization](#deserialization)
- [Merkleization](#merkleization)
- [Self-signed containers](#self-signed-containers)
- [Implementations](#implementations)

## Constants

| Name | Value | Description |
|-|-|-|
| `BYTES_PER_CHUNK` | `32` | Number of bytes per chunk.
| `BYTES_PER_LENGTH_PREFIX` | `4` | Number of bytes per serialized length prefix. |

## Typing
### Basic types

* `"uintN"`: `N`-bit unsigned integer (where `N in [8, 16, 32, 64, 128, 256]`)
* `"bool"`: `True` or `False`

### Composite types

* **container**: ordered heterogenous collection of values
    * key-pair curly bracket notation `{}`, e.g. `{"foo": "uint64", "bar": "bool"}`
* **vector**: ordered fixed-length homogeneous collection of values
    * angle bracket notation `[type, N]`, e.g. `["uint64", N]`
* **list**: ordered variable-length homogenous collection of values
    * angle bracket notation `[type]`, e.g. `["uint64"]`

We recursively define "variable-size" types to be lists and all types that contains a variable-size type. All other types are said to be "fixed-size".

### Aliases

For convenience we alias:

* `"byte"` to `"uint8"` (this is a basic type)
* `"bytes"` to `["byte"]` (this is *not* a basic type)
* `"bytesN"` to `["byte", N]` (this is *not* a basic type)

## Serialization

We recursively define the `serialize` function which consumes an object `value` (of the type specified) and returns a bytestring of type `"bytes"`.

> *Note*: In the function definitions below (`serialize`, `hash_tree_root`, `signed_root`, `is_fixed_size`, `is_variable_size` etc.) objects implicitly carry their type.

### Basic Types

For basic types the `serialize` function is defined as follows.

#### `"uintN"`

A byte string of width  `N // 8` containing the little-endian encode integer.

```python
assert N in [8, 16, 32, 64, 128, 256]
return value.to_bytes(N // 8, "little")
```

#### `"bool"`

* The byte `\x00` **if** the value is `False`
* The byte `\x01` **if** the value is `True`

```python
assert value in (True, False)
return b"\x01" if value is True else b"\x00"
```

### Composite Types (Vectors, Containers and Lists)

The serialized representation of composite types is comprised of two binary sections.

* The first section is *fixed size* for all types, containing the concatenation of
    - The serialized representation for each of the *fixed size* elements from the value
    - The `"uint32"` serialized offset where the serialized representation of the *variable sized* elements from the value are located in the second section.
* The second section contains the concatenation of the serialized representations of **only** the *variable size* types.
    - This section is empty in the case of a purely *fixed size* type.

> **NOTE**: Offsets are relative to the beginning of the beginning of the entire serialized representation (the start of the first section)


#### `"vector"`, `"container"` and `"list"`

Below is an illustrative implementation of the `serialize` function for `"Vector"`,
`"Container"` and `"List"` types.

```python
# The second section is just the concatenation of the serialized *variable size* elements
section_2_parts = [
    serialize(element) if is_variable_size(element)
    else ''
    for element in value
]
section_2_lengths = [len(part) for part in section_2_parts]
section_2 = ''.join(section_2_parts)

# Compute the length of the first section (can also be extracted from the type directly)
section_1_length = sum(
    len(serialize(element)) if is_fixed_size(element)
    else 4
    for element in value
)

# Compute the offset values for each part of the second section
section_1_offsets = [
    section_1_length + sum(section_2_lengths[:element_index]) if is_variable_size(element)
    else None
    for element_index, element in enumerate(value)
]
assert all(offset is None or offset < 2**32 for offset in section_1_offsets)

# The first section is the concatenation of the serialized fixed size elements and offsets
section_1_parts = [
    serialize(element) if is_fixed_size(element)
    else serialize(section_1_offsets[element_index])
    for element_index, element in enumerate(value)
]
section_1 = ''.join(section_1_parts)

return ''.join([section_1, section_2])
```


## Deserialization

Because serialization is an injective function (i.e. two distinct objects of the same type will serialize to different values) any bytestring has at most one object it could deserialize to. Efficient algorithms for computing this object can be found in [the implementations](#implementations).


## Merkleization

We first define helper functions:

* `pack`: Given ordered objects of the same basic type, serialize them, pack them into `BYTES_PER_CHUNK`-byte chunks, right-pad the last chunk with zero bytes, and return the chunks.
* `merkleize`: Given ordered `BYTES_PER_CHUNK`-byte chunks, if necessary append zero chunks so that the number of chunks is a power of two, Merkleize the chunks, and return the root.
* `mix_in_length`: Given a Merkle root `root` and a length `length` (`"uint256"` little-endian serialization) return `hash(root + length)`.

We now define Merkleization `hash_tree_root(value)` of an object `value` recursively:

* `merkleize(pack(value))` if `value` is a basic object or a vector of basic objects
* `mix_in_length(merkleize(pack(value)), len(value))` if `value` is a list of basic objects
* `merkleize([hash_tree_root(element) for element in value])` if `value` is a vector of composite objects or a container
* `mix_in_length(merkleize([hash_tree_root(element) for element in value]), len(value))` if `value` is a list of composite objects

## Self-signed containers

Let `value` be a self-signed container object. The convention is that the signature (e.g. a `"bytes96"` BLS12-381 signature) be the last field of `value`. Further, the signed message for `value` is `signed_root(value) = hash_tree_root(truncate_last(value))` where `truncate_last` truncates the last element of `value`.

## Implementations

| Language | Project | Maintainer | Implementation |
|-|-|-|-|
| Python | Ethereum 2.0 | Ethereum Foundation | [https://github.com/ethereum/py-ssz](https://github.com/ethereum/py-ssz) |
| Rust | Lighthouse | Sigma Prime | [https://github.com/sigp/lighthouse/tree/master/beacon_chain/utils/ssz](https://github.com/sigp/lighthouse/tree/master/beacon_chain/utils/ssz) |
| Nim | Nimbus | Status | [https://github.com/status-im/nim-beacon-chain/blob/master/beacon_chain/ssz.nim](https://github.com/status-im/nim-beacon-chain/blob/master/beacon_chain/ssz.nim) |
| Rust | Shasper | ParityTech | [https://github.com/paritytech/shasper/tree/master/util/ssz](https://github.com/paritytech/shasper/tree/master/util/ssz) |
| Javascript | Lodestart | Chain Safe Systems | [https://github.com/ChainSafeSystems/ssz-js/blob/master/src/index.js](https://github.com/ChainSafeSystems/ssz-js/blob/master/src/index.js) |
| Java | Cava | ConsenSys | [https://www.github.com/ConsenSys/cava/tree/master/ssz](https://www.github.com/ConsenSys/cava/tree/master/ssz) |
| Go | Prysm | Prysmatic Labs | [https://github.com/prysmaticlabs/prysm/tree/master/shared/ssz](https://github.com/prysmaticlabs/prysm/tree/master/shared/ssz) |
| Swift | Yeeth | Dean Eigenmann | [https://github.com/yeeth/SimpleSerialize.swift](https://github.com/yeeth/SimpleSerialize.swift) |
| C# | | Jordan Andrews | [https://github.com/codingupastorm/csharp-ssz](https://github.com/codingupastorm/csharp-ssz) |
| C++ | | | [https://github.com/NAKsir-melody/cpp_ssz](https://github.com/NAKsir-melody/cpp_ssz) |