Skip to content


Switch branches/tags

Name already in use

A tag already exists with the provided branch name. Many Git commands accept both tag and branch names, so creating this branch may cause unexpected behavior. Are you sure you want to create this branch?

Latest commit


Git stats


Failed to load latest commit information.
Latest commit message
Commit time



nodejs-microdb is a tiny in-process database. It has very few methods, but does feature auto-flushing to disk (for file based stores). It can also be used in a memory-only mode.



var microdb = require('nodejs-microdb');
var myDB = new microdb({'file':'somefile.db'});


When making a new database, you have some choices:

var myDB = new microdb({
  'file': '',     // The filename to save to, or empty for memory only

  'savetime': 10, // In minutes, how often to flush to disk (approx)
                  // Set this to 0 to diable auto-save.

  'datatype': 0,  // Which data-type:
                  //  0 = Array-based, no keys. (useful for storing lists to disk)
                  //  1 = Object-based, with keys. (what most key/doc's do)
  'maxrec': 10,   // Maximum number of records (for datatype === 0 ONLY)

  'flushonexit': true,   // Auto-flush when program quits.
                         // I recommend you leave this on.

  'defalutClean': false, // Auto-remove incomplete records on sort opertaions.
                         // This is useful if your document type isn't consistent.
                         // It can also be turned on per-query.


Datatype === 1 Methods:

####MicroData.add(data, [ident]); Add an item to the store. If "ident" is not supplied, it will be created.
Return value is the "ident".

####MicroData.del(ident); Remove a named "ident" from the store.

####MicroData.find(key, value); Find a record "ident" by a named key/value pair. Returns first match, search order is arbitrary.

####MicroData.findAll(key, value); Return an array of record "ident"'s where key === value. Order is arbitrary.

####MicroData.findAllWithKey(key); Return an array of record "ident"'s the contain 'key'. Order is arbitrary.

####MicroData.sortByKey(key, [direction], [alpha], [cleanBad]); Returns an array of values and "ident"'s, sorted in "direction" (asc/desc). Set "alpha" to true for alphanumeric sort. If 'cleanBad' is true, results with one or more 'undefined' values for the specified keys will not be returned. NOTE: This is just a convience method to the below...

####MicroData.sortByKeys(sorts, [cleanBad]); Returns an array of values and "idents"'s sorted by "sorts" array - where sorts is an array of [key, direction, alpha] arrays. (see above). If 'cleanBad' is true, results with one or more 'undefined' values for the specified keys will not be returned.

Datatype === 0 Methods:


Add an item to the list.


Remove item number 'num' from list. 0-based.

Shared Methods:

####MicroData.load(); Load file from disk. Usually called automatically, but if you are before options.savetime, it might work as an undo.; Save file to disk. This is an anonomized asych method (no callback, it'll do it eventually. Used internally, I don't recommend it)

####MicroData.flush(); Save file to disk now. Synch method. If you think you need to flush the db yourself, chances are this is the method you want.

Current status

This module is in a development stage. It is broken horribly in places. There are quite a few features missing. And it has zero room for error - this will always be intended for internal persistent storage, not a real replacement for what you should use a real database for.



This module was only tested using node >= 0.8.8. There is no reason it shouldn't run under earlier versions though.


node-ansibuffer is licensed under the MIT license. Or the BSD license. Or no license if that's more convient for you.


A micro-sized key/document storage engine - with auto-flush on disk or memory only storage.






No releases published


No packages published