literate-jasmine

literate-jasmine

The idea is to write markdown that gets translated to Jasmine describe and it blocks. Because we want to be able to annotate in between parts of what would become a single it, we make use of markdown hierarchy to separate one test from another and to give the it (and describe blocks names).

This README.md has this markdown structure (which includes the main header above and the other parts below):

# literate-jasmine
## Mathematices
### add can add numbers (level 3 header)
### add can add numbers
## Strings
### appending works with +

Which is parsed into a tree:

literate-jasmine
  Mathematices
    add can add numbers (level 3 header)
    add can add numbers
  Strings
    appending works with +

Which is then written to disk as FILENAME_spec.js (so README_spec.js):

describe('literate-jasmine', function() {
  describe('Mathematics', function() {
    it('add can add numbers', function() {
      // test code
    });
    it('can divide numbers', function() {
      // test code
    });
  });
  describe('Strings', function() {
    it('appending works with +', function() {
      // test code
    });
  });
});

The command literate-jasmine is used to convert the markdown to JavaScript (assuming you ran npm install -g literate-jasmine):

literate-jasmine README.md

(If you're working on this project, run ./bin/literate-jasmine instead.)

Then run the jasmine tests:

> jasmine-node README_spec.js
1 2 3
....

Finished in 0.006 seconds
4 tests, 4 assertions, 0 failures, 0 skipped

Mathematics

add can add numbers

var a = 1,
    b = 2;

console.log(a, b, a + b);

expect(a + b).toBe(3);

can divide numbers

var a = 6,
    b = 2;

And a comment here doesn't break things:

expect(a/b).toBe(3);

Strings

appending works with +

var text = "abc";

expect(text + "d").toBe("abcd");