• Neo4j-Apoc


    Neo4j-Apoc

        <!-- 作者区域 -->
        
    
    
        <!-- 文章内容 -->
        <div data-note-content="" class="show-content">
          <div class="show-content-free">
            <p>APOC</p>
    

    https://neo4j-contrib.github.io/neo4j-apoc-procedures/#_virtual_nodes_rels

    提供的函数 存储过程应有尽有, 也可以自行实现添加

    CALL apoc.help("dijkstra")

    Apoc配置:

    apoc.trigger.enabled=false/true  :  Enable triggers
    

    apoc.ttl.enabled=false/true: Enable time to live background task

    apoc.jdbc.<key>.uri=jdbc-url-with-credentials : 配置数据库连接串

    apoc.es.<key>.uri=es-url-with-credentials: ES连接

    apoc.mongodb.<key>.uri=mongodb-url-with-credentials: mongodb连接

    apoc.couchbase.<key>.uri=couchbase-url-with-credentials: couchbase连接

    apoc.jobs.scheduled.num_threads=number-of-threads: APOC调度器的线程池

    apoc.jobs.pool.num_threads=number-of-threads: 执行器的线程池

    有用的函数方法:

    解析域名: WITH 'http://www.example.com/all-the-things' AS url RETURN apoc.data.domain(url) // will return 'www.example.com'
    

    日期函数:

    apoc.date.parse(<span class="hljs-string"><span class="hljs-string">'2015/03/25 03-15-59'</span></span>,[<span class="hljs-string"><span class="hljs-string">'s'</span></span>],[<span class="hljs-string"><span class="hljs-string">'yyyy/MM/dd HH/mm/ss'</span></span>])
    
    apoc.date.add(<span class="hljs-number"><span class="hljs-number">12345</span></span>, <span class="hljs-string"><span class="hljs-string">'ms'</span></span>, <span class="hljs-number"><span class="hljs-number">-365</span></span>, <span class="hljs-string"><span class="hljs-string">'d'</span></span>) 
    

    格式转换:

    <span class="hljs-keyword"><span class="hljs-keyword">return</span></span> apoc.number.format(<span class="hljs-number"><span class="hljs-number">12345.67</span></span>) <span class="hljs-keyword"><span class="hljs-keyword">as</span></span> value
    
    <span class="hljs-keyword"><span class="hljs-keyword">return</span></span> apoc.number.format(<span class="hljs-number"><span class="hljs-number">12345</span></span>, <span class="hljs-string"><span class="hljs-string">'#,##0.00;(#,##0.00)'</span></span>, <span class="hljs-string"><span class="hljs-string">'it'</span></span>) <span class="hljs-keyword"><span class="hljs-keyword">as</span></span> value
    

    数学运算:

    <span class="hljs-keyword"><span class="hljs-keyword">RETURN</span></span> apoc.number.exact.add(stringA,stringB)
    
    <span class="hljs-keyword"><span class="hljs-keyword">RETURN</span></span> apoc.number.exact.sub(stringA,stringB)
    
    ETURN apoc.number.exact.mul(stringA,stringB,[prec],[roundingModel]
    
    <span class="hljs-keyword"><span class="hljs-keyword">RETURN</span></span> apoc.number.exact.div(stringA,stringB,[prec],[roundingModel])
    
    <span class="hljs-keyword"><span class="hljs-keyword">RETURN</span></span> apoc.number.exact.toInteger(string,[prec],[roundingMode])
    
    <span class="hljs-keyword"><span class="hljs-keyword">RETURN</span></span> apoc.number.exact.toFloat(string,[prec],[roundingMode])
    
    <span class="hljs-keyword"><span class="hljs-keyword">RETURN</span></span> apoc.number.exact.toExact(number)
    

    比较节点不同:

    apoc.diff.nodes([leftNode],[rightNode])
    

    图算法:

    路径扩展(选择走哪些边, 哪些节点, 几层, 什么时候结束等等):
    
    CALL apoc.path.expand(startNode &lt;id&gt;|Node, relationshipFilter, labelFilter, minLevel, maxLevel )
    
    MATCH (user:User) WHERE user.id = <span class="hljs-number"><span class="hljs-number">460</span></span>
    CALL apoc.path.expandConfig(user,{relationshipFilter:<span class="hljs-string"><span class="hljs-string">"RATED"</span></span>,minLevel:<span class="hljs-number"><span class="hljs-number">3</span></span>,maxLevel:<span class="hljs-number"><span class="hljs-number">3</span></span>,bfs:<span class="hljs-keyword"><span class="hljs-keyword">false</span></span>,uniqueness:<span class="hljs-string"><span class="hljs-string">"NONE"</span></span>}) <span class="hljs-keyword"><span class="hljs-keyword">YIELD</span></span> path
    <span class="hljs-keyword"><span class="hljs-keyword">RETURN</span></span> count(*);
    
    
    apoc.path.subgraphAll(startNode &lt;id&gt;Node/<span class="hljs-keyword"><span class="hljs-keyword">list</span></span>, {maxLevel, relationshipFilter, labelFilter, bfs:<span class="hljs-keyword"><span class="hljs-keyword">true</span></span>, filterStartNode:<span class="hljs-keyword"><span class="hljs-keyword">true</span></span>, limit:<span class="hljs-number"><span class="hljs-number">-1</span></span>}) <span class="hljs-keyword"><span class="hljs-keyword">yield</span></span> nodes, relationships
    
    
    MATCH (user:User) WHERE user.id = <span class="hljs-number"><span class="hljs-number">460</span></span>
    CALL apoc.path.subgraphNodes(user, {}) <span class="hljs-keyword"><span class="hljs-keyword">YIELD</span></span> node
    <span class="hljs-keyword"><span class="hljs-keyword">RETURN</span></span> node;
    

    Closeness Centrality:

    CALL apoc.algo.closeness([<span class="hljs-string"><span class="hljs-string">'TYPE'</span></span>],nodes,<span class="hljs-string"><span class="hljs-string">'INCOMING'</span></span>) <span class="hljs-keyword"><span class="hljs-keyword">YIELD</span></span> node, score
    

    Betweenness Centrality:

    CALL apoc.algo.betweenness([<span class="hljs-string"><span class="hljs-string">'TYPE'</span></span>],nodes,<span class="hljs-string"><span class="hljs-string">'BOTH'</span></span>) <span class="hljs-keyword"><span class="hljs-keyword">YIELD</span></span> node, score
    

    pageRank:

    CALL apoc.algo.pageRank(nodes) <span class="hljs-keyword"><span class="hljs-keyword">YIELD</span></span> node, score
    
    <span class="hljs-comment"><span class="hljs-comment">// only compute over relationships of types TYPE_1 or TYPE_2</span></span>
    
    CALL apoc.algo.pageRankWithConfig(nodes,{types:<span class="hljs-string"><span class="hljs-string">'TYPE_1|TYPE_2'</span></span>}) <span class="hljs-keyword"><span class="hljs-keyword">YIELD</span></span> node, score
    
    <span class="hljs-comment"><span class="hljs-comment">// peroform 10 page rank iterations, computing only over relationships of type TYPE_1</span></span>
    
    CALL apoc.algo.pageRankWithConfig(nodes,{iterations:<span class="hljs-number"><span class="hljs-number">10</span></span>,types:<span class="hljs-string"><span class="hljs-string">'TYPE_1'</span></span>}) <span class="hljs-keyword"><span class="hljs-keyword">YIELD</span></span> node, score
    

    dijkstra:

    apoc.algo.dijkstra(startNode, endNode, <span class="hljs-string"><span class="hljs-string">'KNOWS|&lt;WORKS_WITH|IS_MANAGER_OF&gt;'</span></span>, <span class="hljs-string"><span class="hljs-string">'distance'</span></span>) <span class="hljs-keyword"><span class="hljs-keyword">YIELD</span></span> path, weight
    
    apoc.algo.dijkstraWithDefaultWeight(startNode, endNode, <span class="hljs-string"><span class="hljs-string">'KNOWS|&lt;WORKS_WITH|IS_MANAGER_OF&gt;'</span></span>, <span class="hljs-string"><span class="hljs-string">'distance'</span></span>, <span class="hljs-number"><span class="hljs-number">10</span></span>) <span class="hljs-keyword"><span class="hljs-keyword">YIELD</span></span> path, weight
    
    example:
    
        MATCH (from:Loc{name:<span class="hljs-string"><span class="hljs-string">'A'</span></span>}), (to:Loc{name:<span class="hljs-string"><span class="hljs-string">'D'</span></span>})
        CALL apoc.algo.dijkstra(from, to, <span class="hljs-string"><span class="hljs-string">'ROAD'</span></span>, <span class="hljs-string"><span class="hljs-string">'d'</span></span>) <span class="hljs-keyword"><span class="hljs-keyword">yield</span></span> path <span class="hljs-keyword"><span class="hljs-keyword">as</span></span> path, weight <span class="hljs-keyword"><span class="hljs-keyword">as</span></span> weight
        <span class="hljs-keyword"><span class="hljs-keyword">RETURN</span></span> path, weight
    

    community: 标签传播的社区发现算法

    apoc.algo.community(times,labels,partitionKey,type,direction,weightKey,batchSize)
    
    example:  遍历<span class="hljs-number"><span class="hljs-number">25</span></span>轮, 
    
        CALL apoc.algo.community(<span class="hljs-number"><span class="hljs-number">25</span></span>,<span class="hljs-keyword"><span class="hljs-keyword">null</span></span>,<span class="hljs-string"><span class="hljs-string">'partition'</span></span>,<span class="hljs-string"><span class="hljs-string">'X'</span></span>,<span class="hljs-string"><span class="hljs-string">'OUTGOING'</span></span>,<span class="hljs-string"><span class="hljs-string">'weight'</span></span>,<span class="hljs-number"><span class="hljs-number">10000</span></span>)
    

    aStar: A*遍历算法

    apoc.algo.aStar(startNode, endNode, <span class="hljs-string"><span class="hljs-string">'KNOWS|&lt;WORKS_WITH|IS_MANAGER_OF&gt;'</span></span>, <span class="hljs-string"><span class="hljs-string">'distance'</span></span>,<span class="hljs-string"><span class="hljs-string">'lat'</span></span>,<span class="hljs-string"><span class="hljs-string">'lon'</span></span>) <span class="hljs-keyword"><span class="hljs-keyword">YIELD</span></span> path, weight
    

    cliques: 聚类社区算法:

    apoc.algo.cliques(minSize) <span class="hljs-keyword"><span class="hljs-keyword">YIELD</span></span> clique
    
    apoc.algo.cliquesWithNode(startNode, minSize) <span class="hljs-keyword"><span class="hljs-keyword">YIELD</span></span> clique
    

    各种距离算法:

    apoc.algo.cosineSimilarity([vector1], [vector2])   cosin相似度
    
    apoc.algo.euclideanDistance([vector1], [vector2])  欧几里得距离
    
    apoc.algo.euclideanSimilarity([vector1], [vector2])  欧几里得相似度
    

    Virtual Nodes/Rels: 虚拟节点, 关系 类似视图的概念

    MATCH (a)-[r]->(b)
    WITH head(labels(a)) AS l, head(labels(b)) AS l2, type(r) AS rel_type, count(*) as count
    CALL apoc.create.vNode([l],{name:l}) yield node as a
    CALL apoc.create.vNode([l2],{name:l2}) yield node as b
    CALL apoc.create.vRelationship(a,rel_type,{count:count},b) yield rel
    RETURN *;
    

    CALL apoc.create.vPattern({_labels:['Person'],name:'Mary'},'KNOWS',{since:2012},{_labels:['Person'],name:'Michael'})

    CALL apoc.create.vPattern({_labels:['Person', 'Woman'],name:'Mary'},'KNOWS',{since:2012},{_labels:['Person', 'Man'],name:'Michael'})

    CALL apoc.create.vPatternFull(['British','Person'],{name:'James', age:28},'KNOWS',{since:2009},['Swedish','Person'],{name:'Daniel', age:30})

    Cypher Exectuion: 批量执行脚本

    CALL apoc.cypher.runFiles([files or urls],{config}) yield row, result   
    

    runs each statement in the files, all semicolon separated

    CALL apoc.cypher.runFile(file or url,{config}) yield row, result
    runs each statement in the file, all semicolon separated - currently no schema operations

    Manual Index: 手工索引, 默认不会自动更新

    synchronously同步更新 只有创建索引时指定autoUpdate:true, 才会真正生效, 更新图时性能上会有影响
    
    apoc.autoIndex.enabled=<span class="hljs-keyword"><span class="hljs-keyword">true</span></span>  
    

    asynchronously异步更新

    apoc.autoIndex.async=<span class="hljs-keyword"><span class="hljs-keyword">true</span></span>  
    
    默认的异步更新参数:<span class="hljs-number"><span class="hljs-number">50000</span></span> operations <span class="hljs-keyword"><span class="hljs-keyword">or</span></span> <span class="hljs-number"><span class="hljs-number">5000</span></span> milliseconds 触发
    
    apoc.autoIndex.queue_capacity=<span class="hljs-number"><span class="hljs-number">100000</span></span>
    apoc.autoIndex.async_rollover_opscount=<span class="hljs-number"><span class="hljs-number">50000</span></span>
    apoc.autoIndex.async_rollover_millis=<span class="hljs-number"><span class="hljs-number">5000</span></span>
    apoc.autoIndex.tx_handler_stopwatch=<span class="hljs-keyword"><span class="hljs-keyword">false</span></span>
    

    添加多个节点属性缩影 配合search用, 重复执行会先删除再创建: apoc.index.addAllNodes('index-name',{label1:['prop1',…​],…​}, {options})

    options的选择(map形式给入):

    type: fulltext/exact  全文索引/精确索引
    
    to_lower_case: <span class="hljs-keyword"><span class="hljs-keyword">false</span></span>/<span class="hljs-keyword"><span class="hljs-keyword">true</span></span>
    
    analyzer: classname   用哪种classname of lucene analyzer 
    
    similarity: classname   相似度计算的方式 classname <span class="hljs-keyword"><span class="hljs-keyword">for</span></span> lucene similarity 
    
    autoUpdate:  <span class="hljs-keyword"><span class="hljs-keyword">true</span></span>/<span class="hljs-keyword"><span class="hljs-keyword">false</span></span>  是否自动更新
    

    添加一个节点的属性索引(可以不存在这个属性字段) apoc.index.addNode(node,['prop1',…​])

    对于给定的标签,添加节点索引(索引名称的Label和node可以不一致): apoc.index.addNodeByLabel('Label',node,['prop1',…​])

    给索引命名 默认是Label作为名称: apoc.index.addNodeByName('name',node,['prop1',…​])

    apoc.index.addNodeMap(node,{key:value})

    apoc.index.addNodeMapByName(index, node,{key:value})

    关系索引: apoc.index.addRelationship(rel,['prop1',…​])

    apoc.index.addRelationshipByName('name',rel,['prop1',…​])

    apoc.index.addRelationshipMap(rel,{key:value})

    apoc.index.addRelationshipMapByName(index, rel,{key:value})

    删除节点索引 apoc.index.removeNodeByName('name',node) remove node from an index for the given name

    索引模糊查询(计算编辑距离) apoc.index.search('index-name', 'query') YIELD node, weight

    apoc.index.nodes('Label','prop:value*') YIELD node, weight

    apoc.index.relationships('TYPE','prop:value*') YIELD rel, weight

    没理解: apoc.index.between(node1,'TYPE',node2,'prop:value*') YIELD rel, weight

    示例:
    match (p:Person) call apoc.index.addNode(p,["name","age"]) RETURN count(*); // 129s for 1M People

    call apoc.index.nodes(<span class="hljs-string"><span class="hljs-string">'Person'</span></span>,<span class="hljs-string"><span class="hljs-string">'name:name100*'</span></span>) <span class="hljs-keyword"><span class="hljs-keyword">YIELD</span></span> node, weight <span class="hljs-keyword"><span class="hljs-keyword">return</span></span> * limit <span class="hljs-number"><span class="hljs-number">2</span></span>
    

    Index Management:

    CALL apoc.index.remove('Thing') 
    

    展示: CALL apoc.index.list()

    CALL apoc.index.forNodes('name',{config}) YIELD type,name,config

    CALL apoc.index.forRelationships('name',{config}) YIELD type,name,config

    Triggers : 触发器

    apoc.trigger.enabled=true
    

    Locking: 锁

    call apoc.lock.nodes([nodes])
    

    call apoc.lock.rels([relationships])

    call apoc.lock.all([nodes],[relationships])

    Meta Graph: 展示节点, 关系标签的概览图

    CALL apoc.meta.graphSample()
    

    CALL apoc.meta.graph

    有选择的展示一部分结果: CALL apoc.meta.subGraph({labels:[labels],rels:[rel-types],excludes:[label,rel-type,…​]})

    表格形式展示数据: CALL apoc.meta.data

    Map形式展示数据: CALL apoc.meta.schema

    快速查看图中各种存在的节点,边: CALL apoc.meta.stats yield labelCount, relTypeCount, propertyKeyCount, nodeCount, relCount, labels, relTypes, stats

    Sehema: 查看各种索引, 约束

    apoc.schema.assert({indexLabel:[indexKeys],…​},{constraintLabel:[constraintKeys],…​}, dropExisting : true) yield label, key, unique, action
    

    apoc.schema.nodes() yield name, label, properties, status, type

    apoc.schema.relationships() yield name, type, properties, status

    apoc.schema.node.indexExists(labelName, properties)

    apoc.schema.node.constraintExists(labelName, properties)

    apoc.schema.relationship.constraintExists(type, properties)

    Streaming Data to Gephi: 导出数据到Gephi

    apoc.gephi.add(url-or-key, workspace, data, weightproperty, ['exportproperty'])

          </div>
    

    原文地址:https://www.jianshu.com/p/851ab29420fd


  • 相关阅读:
    状态线程
    C++编译 C # 调用方法
    图像算法集合
    openmp 和 thread 性能实测
    RGB转YUV 各种库的性能比较
    ipp 实现图像空间的转换
    Eigen 学习笔记
    线性代数笔记
    凸优化 笔记
    Windows系统服务器中安装Redis服务
  • 原文地址:https://www.cnblogs.com/jpfss/p/11393456.html
  • Copyright © 2020-2023  润新知