⚠️ Work in progress
This website (and Lutra project as a whole) is a work in progress. It describes what we want Lutra to become, not what it is yet. The content might be outdated, incomplete or might contain typos.
See project status.
Specification
Each type has a defined encoding with head and an optional body.
Head has static size. Body has dynamic size.
Buffer that contains the encoded type, has the its head at the beginning of the buffer. If there is a body, head contains a pointer to the body. Head and body might not be adjacent in the buffer - they might be separated by data belonging to other types.
For example:
int,floatandboolonly have a head,texthas pointer and length in head and UTF-8 encoded bytes in the body.
Primitives
bool- 8 bit, only first bit is used,int{n}- 64 bit integer, little-endian signed,uint{n}- 64 bit integer, little-endian unsigned,float{n}- 64 bit float, little-endian IEEE 754,text- head contains offset & len, body contains the text encoded as UTF8,
Tuple
Head contains heads of each of the fields. Body is empty.
Array
Head contains offset of the body (32 bit le) and length of the array (32 bit le).
Body contains consecutive heads of each of the items of the array.
Enum
The selected variant is encoded with a tag which uses S bits, where S = ceil(log2(number_of_variants)).
Then, the size of inner head I is determined: H = variants.map(|v| v.size_of_head).max()
If H <= 32 bits, head of the enum contains the tag followed by the head of the inner value of the selected variant. The head of the inner value is padded, such that its total length is H. The size of the head is S + H.
If H > 32 bits, head of the enum contains the tag followed by the pointer to the body. Body contains the inner value of the selected variant. The size of the head is S + 32 bits.