| 1 | #pragma once |
| 2 | |
| 3 | #include <cstdint> |
| 4 | |
| 5 | /** Common defines for compression */ |
| 6 | |
| 7 | #define DBMS_MAX_COMPRESSED_SIZE 0x40000000ULL /// 1GB |
| 8 | |
| 9 | /** one byte for method, 4 bytes for compressed size, 4 bytes for uncompressed size */ |
| 10 | #define 9 |
| 11 | |
| 12 | namespace DB |
| 13 | { |
| 14 | |
| 15 | /** The compressed block format is as follows: |
| 16 | * |
| 17 | * The first 16 bytes are the checksum from all other bytes of the block. Now only CityHash128 is used. |
| 18 | * In the future, you can provide other checksums, although it will not be possible to make them different in size. |
| 19 | * |
| 20 | * The next byte specifies the compression algorithm. Then everything depends on the algorithm. |
| 21 | * |
| 22 | * 0x82 - LZ4 or LZ4HC (they have the same format). |
| 23 | * Next 4 bytes - the size of the compressed data, taking into account the header; 4 bytes is the size of the uncompressed data. |
| 24 | * |
| 25 | * NOTE: Why is 0x82? |
| 26 | * Originally only QuickLZ was used. Then LZ4 was added. |
| 27 | * The high bit is set to distinguish from QuickLZ, and the second bit is set for compatibility, |
| 28 | * for the functions qlz_size_compressed, qlz_size_decompressed to work. |
| 29 | * Although now such compatibility is no longer relevant. |
| 30 | * |
| 31 | * 0x90 - ZSTD |
| 32 | * |
| 33 | * All sizes are little endian. |
| 34 | */ |
| 35 | |
| 36 | enum class CompressionMethodByte : uint8_t |
| 37 | { |
| 38 | NONE = 0x02, |
| 39 | LZ4 = 0x82, |
| 40 | ZSTD = 0x90, |
| 41 | Multiple = 0x91, |
| 42 | Delta = 0x92, |
| 43 | T64 = 0x93, |
| 44 | DoubleDelta = 0x94, |
| 45 | Gorilla = 0x95, |
| 46 | }; |
| 47 | |
| 48 | } |
| 49 | |