jevko / djevko.h

A byte-oriented variant of Jevko in C

[EXPERIMENTAL]

djevko.h

A variant of Jevko implemented in C which differs from the original in that:

• it is defined over bytes rather than Unicode code points
• it uses heredocs and length-prefixing instead of digraphs to "escape" special characters
• it automatically trims whitespace

Rationale

Features

• single-header library: simply #include "djevko.h" and use!
• provides utility functions to help with (de)serialization, escaping, checking, converting, etc.
• should be fairly fast (yet to be benchmarked)
• binary data can be mixed with text data
• human readable and writable
• super minimal and flexible

Example

An example piece of data like:

name [Jon Wąpierz 😀]
age [32]
is cool [true]
fav colors [
  [red]
  [green]
  [blue]
]

can be parsed into C structs like:

typedef struct {
  char** colors;
  Size len;
} FavColors;

typedef struct {
  char* name;
  int age;
  Bool is_cool;
  FavColors fav_colors;
} Person;

And the other way around.

Usage

All you need is djevko.h.

See example.c.

It shows how to (de)serialize C structs/arrays/strings/numbers/booleans to/from Djevko.

You can compile and run the example with:

gcc example.c && ./a.out