Recursive Length Prefix (RLP) encoding.
The purpose of RLP is to encode arbitrarily nested arrays of binary data, and RLP is the main encoding method used to serialize objects in Ethereum. The only purpose of RLP is to encode structure; encoding specific atomic data types (eg. strings, integers, floats) is left up to higher-order protocols; in Ethereum the standard is that integers are represented in big endian binary form. If one wishes to use RLP to encode a dictionary, the two suggested canonical forms are to either use [[k1,v1],[k2,v2]...] with keys in lexicographic order or to use the higher-level Patricia Tree encoding as Ethereum does.
The RLP encoding function takes in an item. An item is defined as follows:
- A string (ie. byte array) is an item - A list of items is an item
For example, an empty string is an item, as is the string containing the word "cat", a list containing any number of strings, as well as more complex data structures like ["cat",["puppy","cow"],"horse",[[]],"pig",[""],"sheep"]. Note that in the context of the rest of this article, "string" will be used as a synonym for "a certain number of bytes of binary data"; no special encodings are used and no knowledge about the content of the strings is implied.
See:
https://github.com/ethereum/wiki/wiki/RLP