You cannot select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
Clyne e58cd38132 | 4 years ago | |
---|---|---|
LICENSE | 4 years ago | |
README.md | 4 years ago | |
consteval_huffman.hpp | 4 years ago |
README.md
consteval-huffman
Allows for long string or data constants to be compressed at compile-time, with a small decoder routine for decompression at run-time.
Compression is achieved using Huffman coding, which works by creating codes for frequently-occuring characters.
Use cases
1. Text configurations (e.g. JSON)
A ~3.5kB string of JSON can be compressed down ~2.5kB (see it on Godbolt).
2. Scripts (e.g. Lisp)
A ~40 line commented sample of Lisp can be reduced from 1,662 bytes to 1,244 (418 bytes saved) (on Godbolt).
Compression will work best on not-small blocks of text or data. This is because a decoding tree must be stored with the compressed data, requiring three bytes per value.
How to Use
#include "consteval_huffman.hpp"
constexpr static const char some_data_raw[] = /* insert text here */;
constinit static const auto some_data = consteval_huffman<some_data_raw>();
// Or, with a set data length:
// ... some_data = consteval_huffman<some_data_raw, 1500>();
int main()
{
// Decompress and print out the data
for (auto decode = some_data.get_decoder(); decode; ++decode)
putchar(*decode);
}