A tiny TOML parser in C, intended for the memory footprint conscious and easy drop-in usage, it is thus only guaranteed to work on ~32kB documents as it limits it's own local memory space to 16-bit although most files up to 65kB should work unless they are intentionally crafted to be problematic. The resulting data structure is deterministic and can be cached as a single file to avoid re-parsing an unmodified document and since all the memory is contiguous, there is very little wasted space or time for small allocations and deallocations. I hope to eventually support strict and permissive parsing modes(WIP). In strict mode it will abort parsing as soon as an error occurs and report the error, however in permissive mode, it will attempt to parse as much as it can and even attempt to recover after a malformed token if possible, some errors may even parse the intent correctly but still issue a warning that something was wrong to comply with the specification.