Derive Macro parity_scale_codec_derive::Encode[−][src]

#[derive(Encode)]
{
    // Attributes available to this derive:
    #[codec]
}

Derive parity_scale_codec::Encode and parity_scale_codec::EncodeLike for struct and enum.

Struct

A struct is encoded by encoding each of its fields successively.

Fields can have some attributes:

#[codec(skip)]: the field is not encoded. It must derive Default if Decode is derived.
#[codec(compact)]: the field is encoded in its compact representation i.e. the field must implement parity_scale_codec::HasCompact and will be encoded as HasCompact::Type.
#[codec(encoded_as = "$EncodeAs")]: the field is encoded as an alternative type. $EncodedAs type must implement parity_scale_codec::EncodeAsRef<'_, $FieldType> with $FieldType the type of the field with the attribute. This is intended to be used for types implementing HasCompact as shown in the example.

#[derive(Encode)]
struct StructType {
	#[codec(skip)]
	a: u32,
	#[codec(compact)]
	b: u32,
	#[codec(encoded_as = "<u32 as HasCompact>::Type")]
	c: u32,
}

Enum

The variable is encoded with one byte for the variant and then the variant struct encoding. The variant number is:

if variant has attribute: #[codec(index = "$n")] then n
else if variant has discrimant (like 3 in enum T { A = 3 }) then the discrimant.
else its position in the variant set, excluding skipped variants, but including variant with discrimant or attribute. Warning this position does collision with discrimant or attribute index.

variant attributes:

#[codec(skip)]: the variant is not encoded.
#[codec(index = "$n")]: override variant index.

field attributes: same as struct fields attributes.

#[derive(Encode)]
enum EnumType {
	#[codec(index = "15")]
	A,
	#[codec(skip)]
	B,
	C = 3,
	D,
}

assert_eq!(EnumType::A.encode(), vec![15]);
assert_eq!(EnumType::B.encode(), vec![]);
assert_eq!(EnumType::C.encode(), vec![3]);
assert_eq!(EnumType::D.encode(), vec![2]);