TingoDB is an embedded JavaScript in-process filesystem or in-memory database upwards compatible with MongoDB at the v1.4 API level.
Note starting from version 0.6.x there is new configuration paramater apiLevel. Default value is 140 (1.4) but we started some adoptions for 2.x specific behaviour. For instance findAndModify will behave difeently if to use apiLevel:200 (2.x). Any contributions with making adoptions for API > 2.x are welcome.
Upwards compatible means that if you build an app that uses functionality implemented by TingoDB you can switch to MongoDB almost without code changes. This greatly reduces implementation risks and give you freedom to switch to a mature solution at any moment.
As a proof for upward compatibility, all tests designed to run against both MongoDB and TingoDB. Moreover, significant parts of tests contributed from MongoDB nodejs driver projects and are used as is without modifications.
For those folks who familiar with the Mongoose.js ODM, we suggest to look at Tungus, an experimental driver that allows using the famous ODM tool with our database.
TingoDB can be dropin replacement for existing apps and frameworks that are based on MongoDB. Please see some 3rd party integrations
For more details please visit http://www.tingodb.com
Goal of our project is to fully mimic MongoDB functionality. Which means that we will consider bug as bug only when you find something that is working with MongoDB but isn't working with TingoDB. It would be very helpful to get bugs in that case as pull requests to /test/misc-test.js file (or new file) which will contain code that reproduce issue.
To run test with MongoDB: ./test.sh --quick --single=misc-test --db=mongodb.
To run test with TingoDB: ./test.sh --quick --single=misc-test --db=tingodb
npm install tingodb
As stated, the API is fully compatible with the MongoDB v1.4 API. The only differences are the initialization and getting the Db object. Consider this MongoDB code:
var Db = require('mongodb').Db,
Server = require('mongodb').Server,
assert = require('assert');
var db = new Db('test', new Server('localhost', 27017));
var collection = db.collection("batch_document_insert_collection_safe");
collection.insert([{hello:'world_safe1'}
, {hello:'world_safe2'}], {w:1}, function(err, result) {
assert.equal(null, err);
collection.findOne({hello:'world_safe2'}, function(err, item) {
assert.equal(null, err);
assert.equal('world_safe2', item.hello);
})
});
The same example using TingoDB is as follows:
var Db = require('tingodb')().Db,
assert = require('assert');
var db = new Db('/some/local/path', {});
// Fetch a collection to insert document into
var collection = db.collection("batch_document_insert_collection_safe");
// Insert a single document
collection.insert([{hello:'world_safe1'}
, {hello:'world_safe2'}], {w:1}, function(err, result) {
assert.equal(null, err);
// Fetch the document
collection.findOne({hello:'world_safe2'}, function(err, item) {
assert.equal(null, err);
assert.equal('world_safe2', item.hello);
})
});
As you can see, the difference is in the require call and database object initialization.
In contrast to MongoDB, the module require call will not return a usable module. It will return a function that accepts configuration options. This function will return something similar to the MongoDB module. The extra step allows for passing some options that will control database behavior.
By specifiyng apiLevel you can expect that some functions will behave accordingly. For instance specifiing apiLevel 200 function findAndModify will return object {value:result,ok:1} as opposed to 1.4 which return just result.
Enable in memory (no file access) mode. Can be useful for unit tests. File path will be used as db identity.
Doing some experimentation we found that using integer keys we can get the database to work faster and save some space. Additionally, for in-process databases there are almost no drawbacks versus globally unique keys. However, at the same time, it is relatively hard to keep unique integer keys outside of the database engine, so we made it part of the database engine code. Generated keys will be unique in the collection scope.
When required, it is possible to switch to BSON ObjectID using the configuration option.
Maximum number of cached objects per collection.
Maximum size of objects that can be placed in the cache.
Globally enables support of search in nested arrays. MongoDB supports this unconditionally. For TingoDB, searching arrays when there are no arrays incurs a performance penalty. That's why this is switched off by default. Additionally, and this might be a better approach, nested array support can be enabled for individual indexes or search queries.
To enable nested arrays in individual indexed, use "_tiarr:true" option.
self._cash_transactions.ensureIndex("splits.accountId",{_tiarr:true},cb);
To enable nested arrays in individual queries for fields that do not use indexes, use "_tiarr." to prefix field names.
coll.find({'arr.num':10},{"_tiar.arr.num":0})
The only required parameter is the database path. It should be a valid path to an empty folder or a folder that already contains collection files.
From the initial release compactionation function was available internally. There were several requests to make this avilable through API and we did it. Please keep in mind that compactination is best called as the first operation with database. Using compactionation in the middle of work session is also possible, but all cursors obtained prior to that will be invalidated and will throw errors on data access.
It is possible to build applications that will transparently support both MongoDB and TingoDB. Here are some hints on how to do that:
require call into a helper module or make it part of the core object. This way you can control which engine is loaded in one place.Example below (please see the three files).
var fs = require('fs'),db,engine;
// load config
var cfg = JSON.parse(fs.readFileSync("./config.json"));
// load requestd engine and define engine-agnostic getDB function
if (cfg.app.engine=="mongodb") {
engine = require("mongodb");
module.exports.getDB = function () {
if (!db) db = new engine.Db(cfg.mongo.db,
new engine.Server(cfg.mongo.host, cfg.mongo.port, cfg.mongo.opts),
{native_parser: false, safe:true});
return db;
}
} else {
engine = require("tingodb")({});
module.exports.getDB = function () {
if (!db) db = new engine.Db(cfg.tingo.path, {});
return db;
}
}
// Depending on engine, this can be a different class
module.exports.ObjectID = engine.ObjectID;
var engine = require('./engine');
var db = engine.getDB();
console.time("sample")
db.open(function(err,db) {
db.collection("homes", function (err, homes) {
// it's fine to create ObjectID in advance
// NOTE!!! we get class through engine because its type
// can depends on database type
var homeId = new engine.ObjectID();
// but with TingoDB.ObjectID righ here it will be negative
// which means temporary. However it's unique and can be used for
// comparisons
console.log(homeId);
homes.insert({_id:homeId, name:"test"}, function (err, home) {
var home = home[0];
// here, homeID will change its value and will be in sync
// with the database
console.log(homeId,home);
db.collection("rooms", function (err, rooms) {
for (var i=0; i<5; i++) {
// it's ok also to not provide id, then it will be generated
rooms.insert({name:"room_"+i,_idHome:homeId}, function (err, room) {
console.log(room[0]);
i--;
if (i==0) {
// now lets assume we serving request like
// /rooms?homeid=_some_string_
var query = "/rooms?homeid="+homeId.toString();
// dirty code to get simulated GET variable
var getId = query.match("homeid=(.*)")[1];
console.log(query, getId)
// typical code to get id from external world
// and use it for queries
rooms.find({_idHome:new engine.ObjectID(getId)})
.count(function (err, count) {
console.log(count);
console.timeEnd("sample");
})
}
})
}
})
})
})
})
{
"app":{
"engine":"tingodb"
},
"mongo":{
"host":"127.0.0.1",
"port":27017,
"db":"data",
"opts":{
"auto_reconnect": true,
"safe": true
}
},
"tingo":{
"path":"./data"
}
}
-2
13 { _id: 13, name: 'test' }
{ name: 'room_0', _idHome: 13, _id: 57 }
{ name: 'room_1', _idHome: 13, _id: 58 }
{ name: 'room_2', _idHome: 13, _id: 59 }
{ name: 'room_3', _idHome: 13, _id: 60 }
{ name: 'room_4', _idHome: 13, _id: 61 }
/rooms?homeid=13 13
5
sample: 27ms
51b43a05f092a1c544000001
51b43a05f092a1c544000001 { _id: 51b43a05f092a1c544000001, name: 'test' }
{ name: 'room_3',
_idHome: 51b43a05f092a1c544000001,
_id: 51b43a05f092a1c544000005 }
{ name: 'room_2',
_idHome: 51b43a05f092a1c544000001,
_id: 51b43a05f092a1c544000004 }
{ name: 'room_1',
_idHome: 51b43a05f092a1c544000001,
_id: 51b43a05f092a1c544000003 }
{ name: 'room_0',
_idHome: 51b43a05f092a1c544000001,
_id: 51b43a05f092a1c544000002 }
{ name: 'room_4',
_idHome: 51b43a05f092a1c544000001,
_id: 51b43a05f092a1c544000006 }
/rooms?homeid=51b43a05f092a1c544000001 51b43a05f092a1c544000001
5
sample: 22ms
We maintain full API and functionality compatibility with MongoDB v1.4 BUT only for what we implemented support. I.e. if we support something it will work exactly the same way, but some features are not yet supported or support is limited.
We are open for contributions for this section. If you found or implemented some integration please open pull request :).
Sails.js
TingoDB adapter for Sails.js: http://github.com/andyhu/sails-tingo
Sails.js is a RoR like Node.js framework. It has a powerful ORM system called waterline, it supports all major databases (and web services) and provides an easy to use, unified interface for querying all different kind of databases. It also supports cross database (or different database engines) joins.
KeystoneJS
Running KeystoneJS with TingoDB: http://ifrederik.com/blog/2014/11/cms-without-db-running-keystonejs-without-mongodb/
Realistic app with dual database support
Web based GnuCash clone: https://github.com/sergeyksv/skilap
Copyright (c) PushOk Software
Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
FAQs
Embedded Node.js database upward compatible with MongoDB
The npm package tingodb receives a total of 2,054 weekly downloads. As such, tingodb popularity was classified as popular.
We found that tingodb demonstrated a not healthy version release cadence and project activity because the last version was released a year ago. It has 1 open source maintainer collaborating on the project.
Did you know?
Socket for GitHub automatically highlights issues in each pull request and monitors the health of all your open source dependencies. Discover the contents of your packages and block harmful activity before you install or update your dependencies.
Security News
Research
A malicious npm package disguised as a WhatsApp client is exploiting authentication flows with a remote kill switch to exfiltrate data and destroy files.
Security News
PyPI now supports digital attestations, enhancing security and trust by allowing package maintainers to verify the authenticity of Python packages.
Security News
GitHub removed 27 malicious pull requests attempting to inject harmful code across multiple open source repositories, in another round of low-effort attacks.