DailyJS

Node Roundup: Knockout, bignumber.js, r...e, Mongoose-Filter-Denormalize

Alex R. Young

Subscribe

@dailyjs

Facebook

Google+

node mongo number mongoose

Node Roundup: Knockout, bignumber.js, r...e, Mongoose-Filter-Denormalize

Posted by Alex R. Young on .
Featured

node mongo number mongoose

Node Roundup: Knockout, bignumber.js, r...e, Mongoose-Filter-Denormalize

Posted by Alex R. Young on .
You can send in your Node projects for review through our contact form or @dailyjs.

Node.js Knockout

Node.js Knockout site

Node.js Knockout is currently being judged, the winners will be announced on the 20th of November. The site is actually a small game in itself -- click around to move your character and type to say something.

bignumber.js

bignumber.js (License: MIT Expat, npm: bignumber.js) by Michael Mclaughlin is an arbitrary-precision arithmetic module. It works in Node, browsers, and is available as an AMD module. It comes with both tests and benchmarks, which is useful because one of Mile's goals was to write something faster and easier to use than JavaScript versions of Java's BigDecimal.

Objects created with BigNumber behave like the built-in Number type in that they have toExponential, toFixed, toPrecision, and toString methods.

Mike found the only other serious arbitrary-precision library for decimal arithmetic on npm is bigdecimal, which originates from the Google GWT project. Mike has written some examples of bigdecimal's problems to illustrate bugs he found while working with it, and offers bignumber.js as an alternative.

r...e

r...e (License: MIT, npm: r...e) by Veselin Todorov is a module for manipulating range expressions. Ranges are specified as separate arguments or strings, and a suitable array will be returned:

range(1, 3).toArray();  
range('1..3').toArray();  
// [1, 2, 3]

range('a', 'c').toArray();  
// ['a', 'b', 'c']

Stepped ranges are supported (0, 10, 5) as well as convenience methods like range(1, 3).include(2), map, join, sum, and so on. It works in browsers, and includes Mocha tests.

Mongoose-Filter-Denormalize

Mongoose-Filter-Denormalize (License: MIT) by Samuel Reed is a filtering and denormalization for Mongoose -- it essentially provides a way of preventing Mongoose from accidentally exposing sensitive data:

UserSchema.plugin(filter, {  
  readFilter: {
    'owner' : ['name', 'address', 'fb.id', 'fb.name', 'readOnlyField'],
    'public': ['name', 'fb.name']
  },
  writeFilter: {
    'owner' : ['name', 'address', 'fb.id', 'writeOnlyField']
  },
  // 'nofilter' is a built-in filter that does no processing, be careful with this
  defaultFilterRole: 'nofilter',
  sanitize: true // Escape HTML in strings
});

Now when passing the result of a findOne or other query to, say, res.send in your Express app, fields can be restricted based on user:

User.findOne({ name: 'Foo Bar' }, User.getReadFilterKeys('public')), function(err, user) {  
  if (err) next(err);
  res.send({ success: true, users: [user] });
});