- Product: CompressKit - C++17 compression laboratory
- Repository:
LessUp/compress-kit - Default branch:
master
Four algorithms (Huffman, Arithmetic, Range, RLE) implemented in C++17. Binary format compatibility is the primary constraint.
Magic Numbers:
| Algorithm | Magic |
|---|---|
| Huffman | HFMN |
| Arithmetic | AENC |
| Range Coder | RCNC |
| RLE | RLE\x00 |
| Command | Purpose |
|---|---|
make build |
Build all C++ CLIs (CMake) |
make test |
Unit tests + CLI smoke tests |
make lint |
clang-format dry-run |
- Maintain binary format compatibility (magic bytes, frequency table LE layout, RLE pair layout)
- Security limits: 4 GiB input, 1 GiB output
- Error messages in code must be English
Binary format changes and new algorithms require careful design review. Internal refactors and bug fixes that preserve existing contract may be implemented directly.