Node Roundup: Express Form, streamline.js, long-stack-traces
Welcome to the Node Roundup. Send in your apps and libraries using our contact form or @dailyjs.
Express Form
Express Form by Dan Dean is a middleware data filtering and validation library for Express. It looks like this:
app.post(
// Route
'/user',
// Form filter and validation middleware
form(
filter("username").trim(),
validate("username").required().is(/^[a-z]+$/),
filter("password").trim(),
validate("password").required().is(/^[0-9]+$/)
),
// Express request-handler now receives filtered and validated data
function(req, res){
if (!req.form.isValid) {
// Handle errors
console.log(req.form.errors);
} else {
// Or, use filtered form data from the form object:
console.log("Username:", req.form.username);
console.log("Password:", req.form.password);
}
}
);
The form method is the middleware which will add req.form.errors and req.form.isValid. I haven’t actually used this yet but validation at the middleware layer is very intriguing.
streamline.js
streamline.js released by Sage is a kind of pre-processing library for Node that makes asynchronous code look synchronous.
The example given in the documentation of asynchronous API usage is asynchronous Node file system code (which could of course be written with the synchronous methods anyway):
function fileLength(path, callback) {
fs.stat(path, function(err, stat) {
if (err) { callback(err); return; }
if (stat.isFile()) {
fs.readFile(path, function(err, data) {
if (err) { callback(err); return; }
callback(null, data.length);
});
} else {
callback(new Error(path + " is not a file"));
}
});
}
The streamline.js version uses the underscore symbol as a plug for values that will be available later, so the previous example could be written like this:
function fileLength(_, path) {
if (fs.stat(path, _).isFile())
return fs.readFile(path, _).length;
else
throw new Error(path + " is not a file");
}
If you’re lost in callbacks and find promises and futures unsatisfactory you might want to try this library. It’s got a lot of examples and more features for parallelizing streamlined code.
long-stack-traces
long-stack-traces by Tom Robinson uses the V8 stack track API to generate long stack traces in Node or a Chromium-based browser. It’s available through npm as long-stack-traces.
Just loading the library will make it work in Node, so this long-stack-trace example will produce:
Uncaught Error: readFile
at Timer.secondTimeout (/Users/alex/example.js:6:15)
at node.js:772:9
----------------------------------------
at Object.setTimeout
at initSecondTimeout (/Users/alex/example.js:5:5)
at /Users/alex/example.js:23:7
at fs:82:13
at node.js:772:9
Rather than:
/Users/alex/example.js:5
throw new Error(tag);
^
Error: readFile
at Timer.secondTimeout [as callback] (/Users/alex/example.js:5:15)
at node.js:772:9
