Skip to content

Latest commit

 

History

History
32 lines (29 loc) · 1.25 KB

HEX.md

File metadata and controls

32 lines (29 loc) · 1.25 KB

Hexadecimal coding

Hex coding for any binary data.

Redefined mostly as an example how drain/feed semantics works between buffers in different formats (here, hex and binary). The conventions:

  1. Subroutine name starts with the module name, which is all-caps up to 4 chars (like a ticker symbol) The next word (lowercase) is preferably a verb.
  2. All subroutines that can fail return ok64.
  3. The returned result goes as the 1st parameter (as a pointer, standard C).
  4. Slices are consumed, unless we imply the parameter must be consumed fully. Then, it is untouched. Either way, that is signalled through the const-ness of the slice type, e.g.
    • $cu8c untouched -- not consumed, not changed,
    • $u8c consumed, content unchanged,
    • $cu8 not consumed, content changed,
    • $u8 consumed, changed.
  5. Subroutines do not know whether the memory was allocated, mapped or it is a part of the stack. They are given a generic memory range. There is no memory-owning "HexBuffer" or "HexStream" or anything. Those would impede composition.
fun ok64 HEXfeed($u8 hex, $u8c bin);
fun ok64 HEXdrain($u8 bin, $u8c hex);
fun ok64 HEXfeedall($u8 hex, $cu8c bin);
fun ok64 HEXdrainall($u8 bin, $cu8c hex);