quickly hack together a nodeschool adventure
This is an alternative to the workshopper module, which you should also look at.
workshopper is more convention-driven and fully-featured, but expects a
particular (configurable) filesystem organization for problems.
adventure is entirely api-driven and has fewer configuration options.
You can fork this tutorial from the example-adventure repo.
First make a runner.js. This is the file you can wire up to the package.json
"bin" field.
#!/usr/bin/env node
var adventure = require('adventure');
var shop = adventure('example-adventure');
shop.add('dinosaurs', function () { return require('./dinosaurs') });
shop.add('robots', function () { return require('./robots') });
shop.add('wowsers', function () { return require('./wowsers') });
shop.execute(process.argv.slice(2));You simply .add(name, fn) each of the adventures in your problem set and then
.execute() the adventure with the command-line arguments.
The interface to problem files is very simple. The simplest version of a problem
is just an object with a .problem string and .verify function.
Here's what we can put in dinosaurs/index.js:
exports.problem = 'Make a dinosaur sound.\n'
+ 'Use `$ADVENTURE_COMMAND verify YOUR_TEXT...` to make your sound.'
;
exports.verify = function (args, cb) {
if (/RAWR/.test(args)) {
console.log('Wow that is a convincing dinosaur.\n');
cb(true);
}
else if (/rawr/i.test(args)) {
console.log('Close, but too quiet. Try louder.\n');
cb(false);
}
else {
console.log("That doesn't sound like a dinosaur at all.\n");
cb(false);
}
};
You don't need to put this in a file necessarily even, you just need to return
an object with these properties from the function you pass to .add().
Your verify(args, cb) function will get the arguments passed to it on the
command-line and a callback that you can use to indicate whether the solution
was successful or not.
You can return many different kinds of objects in your .problem or .solution
functions: a string, a buffer, a stream, or a function that returns a string, a
buffer, or a stream.
Now in robots/index.js we can use streams for the problem and solution:
var fs = require('fs');
var path = require('path');
exports.problem = fs.createReadStream(__dirname + '/problem.txt');
exports.solution = fs.createReadStream(__dirname + '/solution.txt');
exports.verify = function (args, cb) {
var res = require(path.resolve(args[0]));
if (/beep/.test(res) && /boop/.test(res)) {
console.log('That sounds about right!\n');
cb(true);
}
else if (/beep/.test(res) || /boop/.test(res)) {
console.log('Hmm that sounds partly convincing but try harder.\n');
cb(false);
}
else {
console.log("That doesn't sound like a robot at all.\n");
cb(false);
}
};Finally, we can use adventure-verify to verify solutions using tape with friendly colorized tap output.
In wowsers/index.js we can use
adventure-verify to do:
var fs = require('fs');
var path = require('path');
var verify = require('adventure-verify');
exports.problem = fs.createReadStream(__dirname + '/problem.txt');
exports.solution = fs.createReadStream(__dirname + '/solution.txt');
exports.verify = verify({ modeReset: true }, function (args, t) {
var f = require(path.resolve(args[0]));
t.equal(typeof f, 'function', 'you exported a function');
t.equal(f(2,3), 6, '2 * 3 = 6');
t.equal(f(1,1), 1, '1 * 1 = 1');
t.equal(f(0.5,0.5), 0.25, '0.5 * 0.5 = 0.25');
t.end();
});Here we use modeReset so that when a user does console.log() or
console.error() in their solution, their text shows up as the terminal default
instead of getting mixed up with the TAP colors.
Now just fill in the problem.txt and solution.txt files and you will have a
working nodeschool-style adventure! Yay!
var adventure = require('adventure')Create a new nodeschool workshop adventure.
options are:
-
opts.name- name of your adventure (required) -
opts.command- the name of the adventure command (inferred fromopts.name) -
opts.title- title to use for your adventure (default:opts.name.toUpperCase()) -
opts.datadir- directory used to store the current level and the list of completed levels. default:'~/.config/' + opts.name -
opts.colors- object mapping color types to[r,g,b]arrays -
opts.colors.pass- show passing solution messages with this color -
opts.colors.fail- show failing solution messages with this color -
opts.colors.info- show extra info with this color -
opts.fg- menu foreground color -
opts.bg- menu background color
If opts is a string, it will be treated as the opts.name.
Your fn() should return a problem object in the format described below.
Run whatever commands are specified in the command-line args.
If you don't want to let .execute() show the menu, you can show the menu
yourself explicitly with .showMenu().
The options are:
opts.fb- foreground coloropts.bg- background coloropts.title- menu title text
You can explicitly select a level with this method if you don't want to rely on the user to select a menu for themselves from the graphical menu.
Problems must have a verify() function. All other fields are optional.
This function will be called when a user attempts to verify a problem with the
verify command on the command-line.
You will get an array of the arguments given after the verify command in
args.
You must call cb(ok) with ok, a boolean containing whether the solution was
acceptible.
Check out adventure-verify for a higher-level way of verifying solutions with tape.
This function will be called when the user uses the run command from the
command-line. You can implement this if you want to but it doesn't make sense
for all problems.
This message will be displayed when a user selects the problem from the menu.
problem.problem can be a string, a buffer, a stream, or a function that
returns a string, a buffer, or a stream.
This message will be displayed when a user successfully completes a problem, after the success notification.
problem.solution can be a string, a buffer, a stream, or a function that
returns a string, a buffer, or a stream.
This message will be displayed when a user successfully completes a level. The
default problem.pass is says YOUR SOLUTION IS CORRECT in a box of made of
@s.
problem.pass can be a string, a buffer, a stream, or a function that
returns a string, a buffer, or a stream.
This message will be displayed when a user's solution fails to pass all the
tests. The default problem.fail is says YOUR SOLUTION IS NOT CORRECT in a
box of made of #s.
problem.fail can be a string, a buffer, a stream, or a function that
returns a string, a buffer, or a stream.
This event fires when a solution passed.
This event fires when a solution failed.
This event fires when all the levels are completed.
These variables will be automatically replaced any time you use them in any of the problem messages, whether in a string, a buffer, a stream, or a function that returns a string, a buffer, or a stream.
$ADVENTURE_NAME- the name of the adventure$ADVENTURE_COMMAND- the name of the adventure command
The .execute(args) function accepts these commands:
$COMMAND
$COMMAND menu
Show the menu.
$COMMAND verify [ARGS...]
Verify the currently selected problem with ARGS.
$COMMAND run [ARGS...]
Run the currently selected problem with ARGS.
Not all problems support `run`.
$COMMAND selected
Print the currently selected level.
$COMMAND select LEVEL
Set the currently selected LEVEL.
$COMMAND list
List the available levels.
$COMMAND completed
List the completed levels.
$COMMAND reset
Reset the list of completed levels.
$COMMAND help
Show this message.
With npm do:
npm install adventure
MIT
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent