Skip to content

Latest commit

 

History

History
106 lines (94 loc) · 4.88 KB

bytecode.md

File metadata and controls

106 lines (94 loc) · 4.88 KB

Direct Query Execution (Gremlin Bytecode)

Gremlin Traversal Steps are the best way to use a GraphDB in full capacity. This document shows you how to use Gremlin inside your repositories to run direct queries against AWS Neptune.

Gremlin for Javascript convert commands into bytecode. With that you can write complex and platform agnostic queries to efficiently navigate highly connected datasets, and use those data for social platforms, machine learning, and many other applications that requires complex related data.

Gremlin uses some stand-alone methods as part of its query language. In a standard Javascript file using Gremlin your code could look like this:

import gremlin from 'gremlin';
const { t: { id } } = gremlin.process;
const { P: { eq, neq, lt, lte, gt, gte, inside, outside, between, within, without } } = gremlin.process;
const { TextP: { startingWith, endingWith, containing, notStartingWith, notEndingWith, notContaining } } = gremlin.process;
const { order: { asc, desc } } = gremlin.process;
const { cardinality: { single } } = gremlin.process;
const __ = gremlin.process.statics;

Using Loopback Neptune Connector your repository will look like this:

import {inject} from '@loopback/core';
import {DefaultCrudRepository} from '@loopback/repository';
import {NeptuneDataSource} from '../datasources';
import {People, PeopleRelations} from '../models';

export class PeopleRepository extends DefaultCrudRepository<
  People,
  typeof People.prototype.id,
  PeopleRelations
> {
  g;
  P;
  order;

  constructor(
    @inject('datasources.neptune') dataSource: NeptuneDataSource
  ) {
    super(People, dataSource);

    this.g = this.dataSource.connector?.g;
    this.P = this.dataSource.connector?.process.P;
    this.order = this.dataSource.connector?.process.order;
  }

  async findFollower(userId: string) {
    return this.g.V(userId).hasLabel('People').out('Follow').path().toList();
  }
  
  async findFollowerGreaterThanAgePromise(age: number) {
    const promise = this.g.V().hasLabel('People').out("Follow").has('age', this.P.gt(age)).elementMap().dedup().order().by('age', this.order.asc).toList();
    const res = await this.execute(promise);
    return res;
  }

  async findFollowerGreaterThanAgeBytecode(age: number) {
    const bytecode = this.g.V().hasLabel('People').out("Follow").has('age', this.P.gt(age)).elementMap().dedup().order().by('age', this.order.asc);
    const res = await this.execute(bytecode, null, { method: 'toList' });
    return res;
  }

}

Here is the full list of classes you can pull into your repository to use in your Gremlin Queries:

this.g = this.dataSource.connector?.g;
this.statics = this.dataSource.connector?.process.statics;
this.EnumValue = this.dataSource.connector?.process.EnumValue;
this.P = this.dataSource.connector?.process.P;
this.TextP = this.dataSource.connector?.process.TextP;
this.Traversal = this.dataSource.connector?.process.Traversal;
this.TraversalSideEffects = this.dataSource.connector?.process.TraversalSideEffects;
this.TraversalStrategies = this.dataSource.connector?.process.TraversalStrategies;
this.TraversalStrategy = this.dataSource.connector?.process.TraversalStrategy;
this.Traverser = this.dataSource.connector?.process.Traverser;
this.barrier = this.dataSource.connector?.process.barrier;
this.cardinality = this.dataSource.connector?.process.cardinality;
this.column = this.dataSource.connector?.process.column;
this.direction = this.dataSource.connector?.process.direction;
this.operator = this.dataSource.connector?.process.operator;
this.order = this.dataSource.connector?.process.order;
this.pick = this.dataSource.connector?.process.pick;
this.pop = this.dataSource.connector?.process.pop;
this.scope = this.dataSource.connector?.process.scope;
this.t = this.dataSource.connector?.process.t;
this.GraphTraversal = this.dataSource.connector?.process.GraphTraversal;
this.GraphTraversalSource = this.dataSource.connector?.process.GraphTraversalSource;
this.traversal = this.dataSource.connector?.process.traversal;
this.withOptions = this.dataSource.connector?.process.withOptions;

To learn about what all these classes are for look at Gremlin Docs. Not all of these classes and methods are supported by AWS Neptune, but they are available in case you want to use this library to connect to any other Gremlin Compatible GraphDB. For more information please look at AWS Neptune Docs.

Keep reading

Back