Awesome Open Source
Awesome Open Source


  1. (noun) the world or universe regarded as an orderly, harmonious system.

Cosmos is a modern logic programming language.

Cosmos 0.12 VM - Alpha version released.


The language is in testing stage.

This was published only to release the current code.

Ideally, it'll go:

alpha -> beta (most bugs fixed, no missing features)
beta -> full (decent performance)

This answers any "when is it done?" questions.

The focus is on fixing issues and making it easy to try the language instead of performance, so as to make it usable for simple scripts, however.

See details.


Coming soon.


Wait for it.

See test samples.



  • gcc

Use make cosmos or make (this will install additional utilities)

Linux will require sudo make

Tested using mingw/cygwin and gcc (7.4.0).

See readme.txt for dev info.


You can try out the language by opening the interpreter and making queries to the language.

$ cosmos -i
> x=1
| x = 1
> x=1 or 2=x
| x = 1
| x = 2

This will compile and run a file

$ cosmos -l test1


rel main()
    x =
    io.writeln("hello, "+x+'!')

The syntax is close to that of a typical scripting language (Python/JavaScript/Lua). The difference is that whereas traditional scripting languages tend to focus on imperative programming, Cosmos focuses on declarative (logic and functional) programming.

A Neutral Language

Although it is a logic programming language, code in this language can look and behave very conventionally: code might seem imperative or functional.

However, it's a declarative language. As such, variables are immutable. Instead of modifying a value we create a new one.

list.push(l, 55, l2) //instead of modifying l, we create a new variable l2
io.writeln(l)  //[1, 2, 3]
io.writeln(l2) //[1, 2, 3, 55]

Cosmos adopts principles from (procedural) scripting languages, logic programming and functional programming. You may write using an almost imperative style, this will however compile to logic code.


Instead of functions, Cosmos has relations.

Whereas functions have one output, relations may have zero, one or more outputs. You can check this by making queries at the interpreter.

cosmos -i
> x=1 or x=2 //this query has two answers (outputs)
| x = 1
| x = 2

If the system picks one answer and it turns out to be invalid, the system will backtrack and pick the other.

rel p(x)
    x=1 or x=2
rel main()
	io.writeln(x) //2

When a relation is nested, the last parameter is hidden.

//note that the there is no 'return' in the definition
//instead, the parameter y is explicit
//this is typically the 'output' parameter
rel double(x, y)
    y = x*2

double(4,x) //x is 8

Relations may adopt function syntax. When nesting, for example,

print(double(3)) //this will print 6

Logic-wise, double(4,x) is read as a statement: "the double of 4 is x".

double(4) reads as a function; "the double of 4".


Functors are composite data.

functor(F, Functor) //declares an object for creating functors
x = F(1, 2) //x is assigned to a functor F with the values 1 and 2
x = F(1, a) //uses pattern matching to match F(1, 2) against F(1, a)
print(a) //2

Lists are syntax sugar for the functor Cons. Here are two ways to define a list:

l = [1, 2]
l = Cons(1, Cons(2, Cons))

Relations such as first, map and filter can be used to manipulate lists.

require('io', io)
require('math', math)
require('list', list)

rel main()
    l = [1,2,3]
    list.first(l, head) //head is 1, tail) //tail is [2, 3],, l2) //l2 is [2, 3, 4]
    list.filter(l, rel(x) x!=3;, l4) //l4 is [1, 2]

Cosmos adopts many principles and features that are common in functional programming languages (although the principles apply to relations rather than functions).


Tables (also known as maps, dictionaries, etc.) are structures that map keys to values.

Table t = {x=1 and y=2}
table.set(t, 'a', 1, t2)

print(t) //{'x': 1, 'y': 2}
print(t2) //{'x': 1, 'y': 2, 'a': 1}


There is no boolean type. Instead, relations themselves are "booleans".

Code such as

if(s = 'a')
	x = 0
elseif(s = 'b')
	x = 1
	x = 2

is simply sugar for

(s = 'a' and x = 0) or (s = 'b' and x = 1) or x = 2

For a less imperative-looking code,

	s = 'a'
	x = 0
	s = 'b'
	x = 1
	x = 2


The language is whitespace sensitive.

rel p(x)

This could be a single line.

rel p(x) x!=1 and x<5;

It's possible to drop the whitespace semantics by writing the unnecessary characters, although this is not generally advisable.

Note that relations from different lines are separated by ands (semicolons are only used to end the indendation).



  • Added trace for debugging. Simply add this to your file,


With the advent of the Cosmos 0.12 VM,

  • It's still in alpha.
  • It does not rely on other languages to build.
  • C++ VM released. Parser is still built in Cosmos(!) itself.
  • The VM is based on the WAM, as it provides a set of bytecode instructions for Prolog. Temporary variables are not used. See original WAM paper.
  • The parser makes use of an intermediary language, "CWAM". Put(Var(0), 1) translates to put_variable 0,1.
  • This has missing features, however, it's already more than (raw) Prolog. Again, this was uploaded only to give the language a basis.
  • It has its own license.
  • Does not rely on other Prolog implementations.
  • For language samples, see test examples.

Get A Weekly Email With Trending Projects For These Topics
No Spam. Unsubscribe easily at any time.
c-plus-plus (18,715
language (435
interpreter (213
prolog (67
logic-programming (27
cosmos (22