Crate numbat_codec

Modules§

DecodeError
EncodeError
Vec: Reexport needed by derive. A contiguous growable array type, written as Vec<T>, short for ‘vector’.

NestedDecode: Trait that allows zero-copy read of value-references from slices in LE format.
NestedDecodeInput: Trait that allows reading of data into a slice.
NestedEncode: Trait that allows zero-copy write of value-references to slices in LE format.
NestedEncodeNoErr: Most types will be encoded without any possibility of error. The trait is used to provide these implementations. This is currently not a substitute for implementing a proper TopEncode.
NestedEncodeOutput: Trait that allows appending bytes. Used especially by the NestedEncode trait to output data.
TopDecode: Trait that allows zero-copy read of values from an underlying API in big endian format.
TopDecodeInput: Trait that abstracts away an underlying API for a top-level object deserializer. The underlying API can provide pre-parsed i64/u64 or pre-bundled boxed slices.
TopEncode
TopEncodeOutput: Specifies objects that can receive the result of a TopEncode computation. in principle from NestedEncode performed on nested items.

boxed_slice_into_vec
bytes_to_number: Handles both signed and unsigned of any length. No generics here, because we want the executable binary as small as possible.
dep_decode_from_byte_slice: Convenience method, to avoid having to specify type when calling dep_decode. Especially useful in the macros. Also checks that the entire slice was used. The input doesn’t need to be mutable because we are not changing the underlying data.
dep_decode_from_byte_slice_or_exit
dep_encode_to_vec: Convenience function for getting an object nested-encoded to a Vec directly.
top_decode_from_nested: Top-decodes the result using the NestedDecode implementation.
top_decode_from_nested_or_exit: Top-decodes the result using the NestedDecode implementation. Uses the fast-exit mechanism in case of error.
top_encode_from_nested
top_encode_from_nested_or_exit
top_encode_number_to_output
top_encode_to_vec
using_encoded_number: Adds number to output buffer. No argument generics here, because we want the executable binary as small as possible. Smaller types need to be converted to u64 before using this function. TODO: there might be a quicker version of this using transmute + reverse bytes.
vec_into_boxed_slice