Milky - the Indent-based Language liKe pYthon

Introduction

Milky is a toy language that compiles into C, just like CoffeeScript compiles into JavaScript. As its name suggests, the syntax of milky is heavily based on indentations, and is very similar to Python. It is also inspired by Ruby, Go, Swift and many other programming languages.

Let's have a look at some examples. The following is a milky version of the famous FizzBuzz Test:

#include <stdio.h>
#include <stdbool.h>

int main(void):
    for int i = 1; i <= 100; i++:
        bool fizz = (i % 3 == 0)
        bool buzz = (i % 5 == 0)

        if fizz:
            fputs("Fizz", stdout)
        if buzz:
            fputs("Buzz", stdout)
        if fizz || buzz:
            putc('\n', stdout)
        else:
            printf("%d\n", i)

If you are new to milky, you can begin with a piece of C source code - basically, add proper indentations and colons, then remove redundant parentheses, braces and semicolons, and now you've got some code in milky.

For more examples on syntax details, don't forget the source code of milky compiler - it's also written in milky!

Build

As mentioned, the milky compiler is fully written in milky language itself. Hence we'll need C toolchain and a copy of compiled milky compiler (version 0.13.6+) to bootstrap the latest milky.

To build from scratch, you may build a milky compiler from the 'bootstrap' branch:

$ git clone https://github.com/Arnie97/milky
$ cd milky
$ git checkout bootstrap
$ make
$ sudo make install

Or, simply execute sudo make bootstrap to build from Github archive.

Then we can get back to the latest milky source:

$ git checkout master
$ make
$ sudo make install

Naming Conventions

Assume that you've already installed milky in your $PATH:

$ cat > hello.c.k << EOF
#include <stdio.h>
int main(int argc, char *argv[]):
    puts("Hello, World!")
EOF

$ milky hello.c.k           # prints "hello.c.k: Well done."
$ c99 -o hello hello.c
$ ./hello                   # prints "Hello, World!"

As shown above, milky source codes should be named with suffix .k (or .milk), i.e. foo.c.k for a C source file and bar.h.k for a C header file. Unless explicitly given, the output file will be named after the corresponding input, but without the .k suffix. You can specify the output file name with option -o:

$ milky in1.c.k -o out2.c in2.c.k -o out3.c in3.c.k

In this example, the milky compiler will compile in1.c.k into in1.c, in2.c.k into out2.c.k and in3.c.k into out3.c.k. As you can see, multiple inputs can be specified at the same time, hence the desired output name must be given just before the corresponding input file to prevent ambiguity.

Output Style

Milky outputs are designed to be as similar to the corresponding inputs as possible. Match on the line level is strictly guaranteed, which is useful when working with other tools, such as GDB or Gcov.

Taking hello.c.k above and Gcov as an example:

$ milky hello.c.k
$ gcc --coverage -o hello hello.c   # generates hello.gcno
$ ./hello                           # generates hello.gcna
$ cp hello.c.k hello.c              # uses milky source instead in Gcov
$ gcov hello                        # creates coverage report

If you prefer a cleaner programming style in your outputs, simply apply a code formatter. Astyle or Uncrustify might be handy.

C++ Support

Milky was initially designed for the C language. As C++ is almost a super set of C, and shares its basic syntaxes, milky should work with most C++ source codes, although not guaranteed.

Currently, the milky compiler is not deliberately designed to support C++ specific features, excepted for support of scope resolution operators (i.e. ::) since version 0.8.9.

In the future, milky might be developed to provide better C++ support.

License

GPL.

Thanks

The initial versions of milky compiler was based on sample codes from the book 「プログラミング言語を作る」.