dokyu

A little ORM for mongodb: npm install dokyu

Basic Usage

const { Document } = require('dokyu');

Document.connect("mongodb://localhost:27017/example");

class MyDocument extends Document("my_collection") {
  constructor(name) {
    super();
    this.name = name;
  }
  greet() { return "Hello, "+this.name; }
}
MyDocument.createIndexes({ name: 1 }, { unique: true });

var doc = new MyDocument("Arthur");
doc.email = "arthur@camelot.com";
await doc.save(); // this will add the '_id' field to doc

var doc2 = await MyDocument.getOrCreate({ email: "arthur@camelot.com" })
doc2.name // will be "Arthur", from the database
doc2.greet() // will be "Hello, Arthur" from the prototype

API

• Document.connect( [name], url )

  The url should be something like: "mongodb://localhost:27017/db_name".

  Any connection options supported by mongo can be given as URL params, "?replSet=rs0"

• Document( collectionName, [opts] ) → class

  This creates a new base class, suitable only for extending.

  For instance, given class Foo extends Document("foos"), all instances of Foo will read/write to the "foos" collection, using whatever database you pass to Document.connect.

• MyDocument.count( query ) → Promise(number)

  The count value is the number of documents matching the query.

  const count = await MyDocument.count( query );
  assert('number' == typeof count)

• MyDocument.findOne( query ) → Promise(doc)

  The doc value is the first matching instance of MyDocument.

  const doc = await MyDocument.findOne({ name: "Adam" })
  doc.greet(); # "Hello, Adam"

• MyDocument.getOrCreate( fields ) → Promise(MyDocument)

  Example: const doc = await MyDocument.getOrCreate({ name: "Adam" });.

  In this example, if the query findOne({ name: "Adam" }) returns no matches, then the query is converted to a new MyDocument object and returned.

  The query should be limited to only simple matches, with names and values intended for direct inclusion in a new document. If you need more complex query features like ranges, use findOne(query) directly.

• MyDocument.find( query, opts ) → Promise(cursor)

  The cursor given here is a proxy cursor that emits objects of the proper type.

  • projection(fields) → cursor, limit the fields fetched on each item.
  • next() → MyDocument, return a Promise that resolves to the next item, once it's available, or null if the cursor is empty.
  • skip(n) → cursor, skip some items in the cursor.
  • limit(n) → cursor, read at most n items.
  • filter(qry) → cursor, only emit items that match the query object.
  • sort(keys) → cursor, sort the results of the cursor.
  • count() → number, return the number of items available to the cursor.
  • each(cb) → cursor, calls cb(doc) for every result, each doc is an instance of MyDocument.
  • toArray(cb) → cursor, calls cb(array), with an array of MyDocument instances.
  • [Symbol.asyncIterator], the cursor can be read in a for await (const item in cursor) loop.

• MyDocument.updateOne( query, update, [opts] ) → Promise(UpdateResult)

• MyDocument.updateMany( query, update, [opts] ) → Promise(UpdateResult)

• MyDocument.deleteOne( query, [opts] ) → Promise(DeleteResult)

• MyDocument.deleteMany( query, [opts] ) → Promise(DeleteResult)

• MyDocument.remove( query, [opts] ) → Promise(DeleteResult)

  Deprecated. Same as deleteMany.

• MyDocument.createIndexes( fields, opts ) → Promise(IndexResult)

  Example: MyDocument.createIndexes({ name: 1 }, { unique: true }).