Trait scrypto::prelude::radix_engine_common::Encode
source · pub trait Encode<X, E>where
X: CustomValueKind,
E: Encoder<X>,{
// Required methods
fn encode_value_kind(&self, encoder: &mut E) -> Result<(), EncodeError>;
fn encode_body(&self, encoder: &mut E) -> Result<(), EncodeError>;
}Expand description
A data structure that can be serialized into a byte array using SBOR.
Required Methods§
sourcefn encode_value_kind(&self, encoder: &mut E) -> Result<(), EncodeError>
fn encode_value_kind(&self, encoder: &mut E) -> Result<(), EncodeError>
Encodes the SBOR value’s kind to the encoder
sourcefn encode_body(&self, encoder: &mut E) -> Result<(), EncodeError>
fn encode_body(&self, encoder: &mut E) -> Result<(), EncodeError>
Encodes the SBOR body of the type to the encoder.
You may want to call encoder.encode_deeper_body instead of this method. See
the below section for details.
§Direct calls and SBOR Depth
In order to avoid SBOR depth differentials and disagreement about whether a payload is valid, typed codec implementations should ensure that the SBOR depth as measured during the encoding/decoding process agrees with the Value codec.
Each layer of the Value counts as one depth.
If the encoder you’re writing is embedding a child type (and is represented as such
in the Value type), then you should call encoder.encode_body to increment
the SBOR depth tracker.
You should only call value.encode_body directly when the encoding of that type
into an Value doesn’t increase the SBOR depth in the encoder, that is:
- When the wrapping type is invisible to the Value, ie:
- Smart pointers
- Transparent wrappers
- Where the use of the inner type is invisible to Value, ie:
- Where the use of
value.encode_bodyis coincidental / code re-use
- Where the use of