Index

The Index record isolates a region of a Content record. This is especially useful when the Content’s blob is composed of heterogenous data that must be managed separately.

The Index record does not specify which Content record it refers to; applications may use custom logic to disambiguate the correct record (e.g., a stone may be encoded with Index records followed by the Content they refer to). The trivial case is a stone with only one Content record.

Field	Type	Size (bytes)	Description
start	uint	8	The offset in the Content blob where this region begins. This is a zero-based index: the first byte is addressed with 0.
end	uint	8	The offset in the Content blob where this region ends. This last byte is included in the range. This is a zero-based index: the first byte is addressed with 0.
hash	blob	16	Checksum of the content of the region, computed using XXH3_128bits.

Illustrated below is an example of two Index records addressing the same Content.

block
  columns 3

block
columns 1
  block:Index1:1
  columns 1
    start1["start = 1"]
    end1["end = 3"]
    cap1["hash"]
  end
space
  block:Index2:1
columns 1
    start2["start = 1"]
    end2["end = 6"]
    cap2["hash"]
  end
  end

space

  block:Content
  columns 1
    c0["0x41"]
    c1["0x65"]
    c2["0x72"]
    c3["0x79"]
    c4["0x6E"]
    c5["0x4F"]
    c6["0x53"]
  end

  start1 --> c1
  end1 --> c3

start2 --> c1
end2 --> c6