dao

JavaScript library for semantic graph traversal over POJOs. Dao is a collection of pure functions -- mix and match as you please.

Getting dao

npm install --save rhumbl-dao

Using dao

With ES6 imports:

import dao from 'dao'

dao.getEntityById(...);

With CommonJS

var dao = require('dao');

dao.getPathway(...);

As an SFX bundle:

<script src="dist/dao.js"></script>

Then anywhere in your JavaScript, you can call the globally-available dao object:

dao.getIncomingEntities(...)

Why dao?

When working with graph-like structures, we often need to query things like nodes, links and certain pathways. Dao makes it easy to search and select from your data through convenience methods.

Required dao data structure

Dao requires that you have a data object that represents a graph. This is nothing more than a POJO (plain-old JavaScript object and looks like this:

var graphData = {
  entities: [
    {id: 'Leonard Nimroy'},
    {id: 'Zachary Quinto'},
    {id: 'Gene Roddenberry'},
    {id: 'Star Trek'},
    {id: 'Heroes'}
    ...
  ],
  relationships: [
    {sourceId: '4', targetId: '8', type: 'IS_FRIENDS_WITH'},
    {sourceId: '4', targetId: '8', type: 'STARRED_IN'},
    {sourceId: '4', targetId: '8', type: 'DIRECTED'},
    ...
  ]
}

Entities must contain a required id attribute, and relationships have the 3 required attributes of type, sourceId and targetId. You can attach anything else you want to these objects.

API Reference

dao.getEntityById(entityId, entities)

Simple wrapper on top of lodash's _.find

Arguments

1. entityId (Number): The entity's id you're trying to get
2. entities (Array): The array of entities to search through

dao.getIncomingEntities(entityId, relationshipTypes, data)

Given entity A, get the entities B, C...,Z that have edges going towards entity A.

Example

var entities = dao.getIncomingEntities('Star Trek', ['STARRED_IN'], graphData);
// [{id: 'Leonard Nimroy'}, {id: 'Zachary Quinto'}]

Arguments

1. entityId (Number): The id of the entity for which you
2. relationshipTypes (Array): The relationship types that are admissible
3. data (Object): The required data structure to search through (see required dao data structure above)

dao.getIncomingEntitiesAll(entityId, relationshipTypes, data)

Given entity A, get the entities B, C...,Z that have edges going towards entity A, and all the entities that are connected to those entities.

Arguments

1. entityId (Number): The id of the entity for which you
2. relationshipTypes (Array): The relationship types that are admissible
3. data (Object): The required data structure to search through (see required dao data structure above)

dao.getOutgoingEntities(entityId, relationshipTypes, data)

Given entity A, get the entities B, C...,Z that have edges going from entity A to those entities

Example

var entities = dao.getOutgoingEntities('Zachary Quinto', ['STARRED_IN'], graphData);
// [{id: 'Star Trek'}, {id: 'Heroes'}]

Arguments

1. entityId (Number): The id of the entity for which you
2. relationshipTypes (Array): The relationship types that are admissible
3. data (Object): The required data structure to search through (see required dao data structure above)

dao.getOutgoingEntitiesAll(entityId, relationshipTypes, data)

Given entity A, get the entities B, C...,Z that have edges going from entity A to those entities, and all the entities that are connected to those entities.

Arguments

1. entityId (Number): The id of the entity for which you
2. relationshipTypes (Array): The relationship types that are admissible
3. data (Object): The required data structure to search through (see required dao data structure above)

dao.getPathway(entityId, relationshipTypes, algorithmName, data)

Given entity A, get the entities B, C...,Z that have edges going from entity A to those entities, and the edges comprising those pathways. The algorithmName lets you specify whether you want the immediately-connected nodes or if you want to traverse the entire pathway.

Arguments

1. entityId (Number): The id of the entity for which you
2. relationshipTypes (Array): The relationship types that are admissible
3. algorithmName (String): INCOMING | INCOMING_ALL | OUTGOING | OUTGOING_ALL
4. data (Object): The required data structure to search through (see required dao data structure above)

dao.rankDAG(dag, iterateeFn, maxIterations)

Given a collection of nodes and entities, compute the ranking of the graph.

Arguments

1. dag (Object): Must be an object of nodes. It looks like:

var dag = {
  nodes: [
    {id: 'my node'}
    ...
  ],
}

2. iterateeFn (function): For each node, computes the ranking for the node
3. maxIterations = 10 (number): Optional integer setting the threshold for when the function will trigger a warning. It is assumed that your iteratee function yields an acyclic graph, so this gives you a warning when you reach the specified depth threshold

dao.getEdgeSet(entityIds, relationshipTypes, data)

Given a set of entity ids, returns the collection of edges that connect the entities

Arguments

1. entityIds (Array): Array of entity ids
2. relationshipTypes (function): The relationship types that are admissible
3. data (Object): The required data structure to search through (see required dao data structure above)

About

Built by Rhumbl, the graph visualization engine for non-developers.

License

This content is released under the (http://opensource.org/licenses/MIT) MIT License.