grakn.ai 图数据库学习笔记

目录

•  Grakn 下载安装

•  加载表结构及数据 、gql语句实现 增删改查

•  可视化工具Workbase 安装

•  Java连接grakn 查询数据

---

最近因业务需要在学习 grakn.ai 图数据库, 总结几项基础应用篇分享给大家.

grakn.ai 是一个开源知识图谱存储及查询系统，其自带 grakn console 等数据交互接口，具有自动推理功能，具有 CNN 等图谱深度学习扩展组件。

 

• Grakn 下载安装

1. 官网下载地址:  https://grakn.ai/download#core  选择linux版本的Grakn Core安装.

2. 将下载好的grakn.tar.gz包上传服务器, 解压到标准安装目录下。(不需要任何配置解压后既能使用)

3. cd 到 grakn-core-all-linux-1.5.7目录下通过 ./grakn server start 来启动grakn。

[root@nn1 grakn-core-all-linux-1.5.7]# ./grakn server status
====================================================================================================
      ________  _____     _______  __    __  __    __      _______  _______  _____     _______
     |   __   ||   _  \  |   _   ||  |  /  /|  \  |  |    |   _   ||   _   ||   _  \  |   ____|
     |  |  |__||  | |  | |  | |  ||  | /  / |   \ |  |    |  | |__||  | |  ||  | |  | |  |
     |  | ____ |  |_| /  |  |_|  ||  |/  /  |    \|  |    |  |     |  | |  ||  |_| /  |  |____
     |  ||_   ||   _  \  |   _   ||   _  \  |   _    |    |  |  __ |  | |  ||   _  \  |   ____|
     |  |__|  ||  | \  \ |  | |  ||  | \  \ |  | \   |    |  |_|  ||  |_|  ||  | \  \ |  |____
     |________||__|  \__\|__| |__||__|  \__\|__|  \__|    |_______||_______||__|  \__\|_______|

                                         THE KNOWLEDGE GRAPH
====================================================================================================

Starting Storage............SUCCESS
Starting Engine.....................SUCCESS
[root@nn1 grakn-core-all-linux-1.5.7]#

---

 

 

• 加载表结构及数据 、gql语句实现增删改查

     1. schema.gql + data.gql下载地址:

        https://dev.grakn.ai/docs/files/social-network/schema.gql

        https://dev.grakn.ai/docs/files/social-network/data.gql

 

     2. grakn建表+增删改查gql语句测试

1. 创建keyspace + 表结构
./grakn console --keyspace grakn_ example --file /home/schema.gql

2. 往grakn_ example中插入数据
./grakn console --keyspace grakn_ example --file /home/data.gql

3. 进到grakn_ example的控制台中
./grakn console -k grakn_ example

4. 提供以下4条gql查询语句测试
// 查询person这个实体
match $p isa person; get;
// 查询person中实体有full-name属性的
match $p isa person, has full-name $n; get;
// 查询有employer,employee这两个关系的
match $emp (employer: $x, employee: $y) isa employment; get;
// 查询person实体中nickname属性为 "Mitzi",并且phone-number中带"+44"的.
match $p isa person, has nickname "Mitzi", has phone-number $pn; $pn contains "+44"; get;
------------
5.提供以下3条插入语句
// 插入属性类型实例
insert $x isa emotion; $x "like";
// 插入实例类型实例
insert $p isa person, has full-name "John Parkson", has gender "male", has email "[email protected]", has phone-number "+44-1234=567890";
// 插入关系类型实例
match
  $org isa organisation, has name "Facelook";
  $person isa person, has email "[email protected]";
insert $new-employment (employer: $org, employee: $person) isa employment;
  $new-employment has reference-id "WGFTSH";
------------
6. 提供以下3条删除语句
// 删除实体类型实例
match $p isa person, has email "[email protected]"; delete $p;
// 删除关系类型实例
match
  $org isa organisation, has name "Pharos";
  $emp (employer: $org, employee: $p) isa employment;
delete $emp;

// 删除与属性的关联
match $t isa travel, has start-date 2013-12-22 via $r; delete $r;
------------
7. 提供以下3条更新语句
// 更新属性类型实例
## deleting the old
match $org isa organisation, has name "Medicely", has registration-number $rn via $r; delete $r;
## inserting the new
match $org isa organisation, has name "Medicely"; insert $org has registration-number "81726354";
// 更新给定属性的所有实例
## inserting the new
match
  $m isa media, has caption $c;
  $c contains "inappropriate word";
insert $m has caption "deleted";
## deleting the old
match $c isa caption; $c contains "inappropriate word"; delete $c;
// 更新关系的角色扮演者
## inserting the new
match
  $p isa person, has name "Amabo";
  $org isa organisation, has name "Etihw Esouh";
insert $emp (employer: $org, $employee: $p) isa employment;
## deleting the old
match
  $p isa person, has name "Prumt";
  $org isa organisation, has name "Etihw Esouh";
  $emp (employer: $org, employee: $p) isa employment;
delete $emp;

 

     3. gql语句支持sql中的大部分功能如聚合函数、分组、排序、or等, 以下是它的关键字

asc, by, compute, contains, count

delete, desc, from, label, limit

get, group, id, in, insert, or

match, max, mean, median, min

offset, order, regex, std, sum, to

---

 

• 可视化工具workbase安装

     1. 下载地址:  https://grakn.ai/download#workbase 选择windows版本的workbase安装.

     2. 启动并登录workbase;  Host改成安装grakn core那台ip, 端口不变.

     3. 在右上角选择要使用的keyspace, 输入gql语句查询.

---

 

 

• Java连接grakn查询数据

     1. 创建maven工程, 导入pom依赖.

<project xmlns="http://maven.apache.org/POM/4.0.0"     xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"     xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">     <modelVersion>4.0.0modelVersion>     <groupId>com.cape_7_post_officegroupId>     <artifactId>GraknartifactId>     <version>1.0version>     <packaging>jarpackaging>       <properties>         <project.build.sourceEncoding>UTF-8project.build.sourceEncoding>     properties>       <dependencies>         <dependency>             <groupId>junitgroupId>             <artifactId>junitartifactId>             <version>3.8.1version>             <scope>testscope>         dependency>           <dependency>             <groupId>io.grakn.clientgroupId>             <artifactId>grakn-clientartifactId>             <version>1.5.3version>         dependency>         <dependency>             <groupId>io.grakn.coregroupId>             <artifactId>conceptartifactId>             <version>1.5.6version>         dependency>         <dependency>             <groupId>io.graqlgroupId>             <artifactId>langartifactId>             <version>1.0.1version>         dependency>     dependencies> project>

 

     2. java代码

package com.grakn;

import static graql.lang.Graql.parse;
import java.util.ArrayList;
import java.util.Arrays;
import java.util.List;
import grakn.client.GraknClient;
import grakn.core.concept.answer.ConceptMap;
import graql.lang.query.GraqlGet;


public class FirstQuery {
         private static GraknClient client;
         private static GraknClient.Session session;
         private static GraknClient.Transaction transaction;

    public void connect(String node,String keyspace) {
        client = new GraknClient(node);
        session = client.session(keyspace);
    }

    public void querySchema() {
        transaction = session.transaction().write();
        List queryAsList = Arrays.asList(
            "match",
            "$tx (driver: $x, driveway: $y) isa driving ,has passTime $a1;",
            "$x isa car ,has plateNo $p1 ;",
            "$y isa cross ,has crossNo $c1;",
            "{$y has crossNo \"lk007\"; $a1>2019-01-14 ; $a1<2019-01-16  ;} or {$y has crossNo \"lk008\";$a1>2019-01-14 ; $a1<2019-01-16  ;} or {$y has crossNo \"lk009\";  $a1>2019-01-16 ; $a1<2019-01-18  ;}  ;",
            "get  $p1 ,$a1  ,$c1 ;"
        );
        System.out.println("\n------------------Query statement----------------\n" + String.join("\n", queryAsList) + "\n-------------------------------------------------\n");
        String query = String.join("", queryAsList);
        List result = new ArrayList<>();
        List answers = transaction.execute((GraqlGet) parse(query));
        for (ConceptMap answer : answers) {
            result.add(answer.get("p1").asAttribute().value().toString());
            result.add(answer.get("a1").asAttribute().value().toString());
            result.add(answer.get("c1").asAttribute().value().toString() + "\n");
        }
        System.out.println("\n------------------Results----------------\n" + String.join(", ", result) + "\n-----------------------------------------\n" + "\n Query Completed! \n");
    }

    public static void main(String[] args) {
        FirstQuery grakn = new FirstQuery();
        try {
            grakn.connect("192.168.1.1:48555", "keyspace_test");
            grakn.querySchema();
        } catch (Throwable t) {
            t.printStackTrace();
        } finally {
            transaction.close();
            session.close();
            client.close();
        }
    }
}

 

---

原文地址: https://blog.csdn.net/DelevinData/article/details/97385258