Crate pure_riff 

Flow of parsing all chunks in a RIFF file

Keep track of a position. Read BUFFER_LEN bytes at the position. Give parse_chunk the data you read. Now you have the id and len of the chunk. You can then read the chunk data and do stuff with it. You can recursively go through all containers by checking if the chunk is a container chunk with RiffChunkHeader::container_info. You can parse the sub chunks of a container chunk. The end of the sub chunks is the start of the next sibling chunk of the container chunk (if present).

Handling possibly invalid data

This library does very minimal checking for invalid data, such as length fields that imply that data is out of bounds of the container it’s in. You have to do the checking yourself. If you feel like this library be better in terms of handling potentially invalid data, create an issue.

RIFF Format

There are chunks. Each chunk has a field with a 4-byte id, a field for the size of its data field, followed by a variable length data field. Each chunk is aligned by 2 bytes, so there is a padding byte after the data field if needed.

The type of a size in RIFF is a u32 and is encoded in little-endian. This limits the size of the RIFF file. There are extensions to RIFF that allow for larger sizes, but this library currently does not implement them.

A chunk can contain chunks inside of it. There are special chunks which contain chunks inside of their data field.

Some file extensions / formats, such as .wav, use RIFF as a container format. They contain a single chunk with id RIFF which contains the file’s data.

See https://en.wikipedia.org/wiki/Resource_Interchange_File_Format#Explanation for more information.

Re-exports

pub use zerocopy;

Structs

ContainerInfo

To read the container id, see CONTAINER_ID_OFFSET. To read the sub chunks, see SUB_CHUNKS_OFFSET.

ContainerInfoError

The length is too small for even the id

ParseChunkOutput

RiffChunkHeader

A RIFF header. You can use zerocopy to “parse” this.

Constants

BUFFER_LEN

CONTAINER_ID_OFFSET

The offset of the container id relative to the position of the container chunk.

SUB_CHUNKS_OFFSET

The offset of the sub chunks relative to the position of the container chunk.

Functions

parse_chunk

Parses the chunk header and returns it, along with the relative position of the next chunk header. You can then read BUFFER_LEN at the new position to iterate through all chunks.

Type Aliases

Id

Supposed to be ASCII (and contains spaces for strings that are smaller than 4 characters). Use str::from_utf8 to parse it.