This document details the anatomy of a KingsIsle WAD file (KIWAD). To begin, a KIWAD file can be divided into three basic parts:
- the archive header, containing the magic bytes, WAD version, and file count;
- the file list, which provides info on every file in the archive; and
- the data section, which contains the data of all the files in the archive.
Every archive begins with a short header with the following structure:
| field | type | description |
|---|---|---|
| magic | string | KIWAD; used for identification |
| version | int32 | what version of the WAD format the file is |
| count | int32 | the number of files in the archive |
If the KIWAD is version 2 or greater, there will be a spacing byte appended at the end of the header.
After the header comes the file list, which is composed of many smaller chunks of data detailing each file in the archive. Each chunk is structured as follows:
| field | type | description |
|---|---|---|
| offset | int32 | the offset of the file’s data in the archive |
| size | int32 | how large the file is (in bytes) |
| comp_size | int32 | how large the file is when compressed (in bytes) |
| compressed | bool | whether the file is compressed or not |
| checksum | int32 | the CRC checksum of the file |
| path_len | int32 | how long (in bytes) the file’s path is |
| path | string | the path of the file in the archive |