文章目錄

• • 一、安裝Janusgraph
  • 二、連接gremlin
  • 三、增刪查改方法封裝
  • 四、其他方法封裝
  • 五、致謝

一、安裝Janusgraph

1、下載安裝包

前往janusgraph的官方發(fā)布頁(yè)下載離線安裝包，如下圖所示，目前最新的版本是0.4.0，點(diǎn)擊janusgraph-0.4.0-hadoop2.zip下載即可。請(qǐng)注意，該安裝包只能在linux系統(tǒng)下安裝！！！

2、解壓運(yùn)行

打開(kāi)終端，使用unzip命令解壓下載好的zip安裝包，然后進(jìn)入到janusgraph-0.4.0目錄下，執(zhí)行 bin/janusgraph.sh start ，會(huì)得到如下的運(yùn)行結(jié)果，說(shuō)明janusgraph運(yùn)行成功。

$ bin/janusgraph.sh startForking Cassandra...Running `nodetool statusthrift`.. OK (returned exit status 0 and printed string "running").Forking Elasticsearch...Connecting to Elasticsearch (127.0.0.1:9300)... OK (connected to 127.0.0.1:9300).Forking Gremlin-Server...Connecting to Gremlin-Server (127.0.0.1:8182)... OK (connected to 127.0.0.1:8182).Run gremlin.sh to connect.

3、其他操作

在janusgraph-0.4.0目錄下，執(zhí)行 bin/janusgraph.sh status ，可以查看janusgraph的運(yùn)行狀態(tài)及進(jìn)程ID，而執(zhí)行 bin/janusgraph.sh stop ，則關(guān)閉janusgraph及其相關(guān)的所有進(jìn)程。

$ ./bin/janusgraph.sh statusGremlin-Server (org.apache.tinkerpop.gremlin.server.GremlinServer) is running with pid 31841Elasticsearch (org.elasticsearch.bootstrap.Elasticsearch) is running with pid 31668Cassandra (org.apache.cassandra.service.CassandraDaemon) is running with pid 31336$ bin/janusgraph.sh stopKilling Gremlin-Server (pid 31841)...Killing Elasticsearch (pid 31668)...Killing Cassandra (pid 31336)...

二、連接gremlin

1、使用gremlin控制臺(tái)連接

在janusgraph-0.4.0目錄下，執(zhí)行 bin/gremlin.sh 則可啟動(dòng)gremlin控制臺(tái)，如下所示。然后可以進(jìn)行g(shù)remlin的相關(guān)操作，如:remote命令告訴控制臺(tái)使用conf/remote.yaml配置文件與gremlin服務(wù)進(jìn)行遠(yuǎn)程連接，:>是“提交”命令，比如 :> graph.addVertex("name", "test")是添加了一個(gè)name為test的節(jié)點(diǎn)，然后 :> g.V().values('name')則是查詢?cè)搱D中所有節(jié)點(diǎn)中含有name屬性的值。

$ bin/gremlin.sh\,,,/(o o)-----oOOo-(3)-oOOo-----plugin activated: tinkerpop.serverplugin activated: tinkerpop.hadoopplugin activated: tinkerpop.utilitiesplugin activated: janusgraph.importsplugin activated: tinkerpop.tinkergraphgremlin> :remote connect tinkerpop.server conf/remote.yaml==>Connected - localhost/127.0.0.1:8182gremlin> :> graph.addVertex("name", "test")==>v[4726]gremlin> :> g.V().values('name')==>test

2、使用gremlin_python連接

首先需要安裝gremlin_python這個(gè)依賴包，使用pip命令安裝 pip install gremlinpython==3.4.1 ，這里要注意的是 janusgraph0.4.0支持的gremlin最高版本為3.4.1，所以安裝的時(shí)候需指定版本號(hào)。
安裝完成后，便可在python程序中進(jìn)行連接了，注意此時(shí)需要保證janusgraph處于運(yùn)行狀態(tài)。

from gremlin_python.driver.driver_remote_connection import DriverRemoteConnectionfrom gremlin_python.process.anonymous_traversal import traversalconnection = DriverRemoteConnection('ws://127.0.0.1:8182/gremlin', 'g')graph = traversal().withRemote(connection)graph.addV('person').property('name', 'test').next() # 添加節(jié)點(diǎn)print(graph.V().values('name').toList()) # 輸出為 ['test']

三、增刪查改方法封裝

1、新增節(jié)點(diǎn)

只需傳入新增節(jié)點(diǎn)的 label 以及 properties（dict，可選），返回 Vertex(id, label)類(lèi)型。

def add_vertex(graph, label, properties=None):"""add vertex:param graph: graph, type: GraphTraversalSource:param label: label, type: str:param properties: property dict, like {'p1': 'value1', 'p2': 'value2'}:return: vertex, Vertex(id, label)"""vert = graph.addV(label)if properties:for key in properties.keys():vert.property(key, properties.get(key))return vert.next()

2、新增邊

傳入新增邊的 label 和 properties（dict，可選），以及需要添加邊的兩節(jié)點(diǎn)（或其ID）v_from和v_to。

def add_edge(graph, label, v_from, v_to, properties=None):"""add edge:param graph: graph, type: GraphTraversalSource:param label: label, type: str:param v_from: long vertex id or Vertex(id, label) of from:param v_to: long vertex id or Vertex(id, label) of to:param properties: property dict, like {'p1': 'value1', 'p2': 'value2'}:return: None"""if isinstance(v_from, int):v_from = graph.V().hasId(v_from).next()if isinstance(v_to, int):v_to = graph.V().hasId(v_to).next()edge = graph.V(v_from).addE(label).to(v_to)if properties:for key in properties.keys():edge.property(key, properties.get(key))edge.next()

3、刪除節(jié)點(diǎn)

可以根據(jù)要求來(lái)刪除特定節(jié)點(diǎn)，如根據(jù)節(jié)點(diǎn)（或其ID）、label、properties。如果不傳入其他參數(shù)，則默認(rèn)刪除所有節(jié)點(diǎn)。

def drop_vertex(graph, v_id=None, label=None, properties=None):"""drop all vertex or specific vertex:param graph: graph, type: GraphTraversalSource:param v_id: long vertex id or Vertex(id, label):param label: label, type: str:param properties: property list, like ['p1', 'p2', {'p3': 'value'}]:return: None"""if isinstance(v_id, int):v_id = graph.V().hasId(v_id).next()travel = graph.V(v_id) if v_id else graph.V()if label:travel = travel.hasLabel(label)if properties:for p in properties:if isinstance(p, dict):key = list(p.keys())[0]travel = travel.has(key, p.get(key))else:travel = travel.has(p)travel.drop().iterate()

4、刪除邊

可以根據(jù)要求來(lái)刪除特定邊，如根據(jù)邊ID、label、properties。如果不傳入其他參數(shù)，則默認(rèn)刪除所有邊。

def drop_edge(graph, e_id=None, label=None, properties=None):"""drop all edges or specific edge:param graph: graph, type: GraphTraversalSource:param e_id: edge id, type str:param label: label, type: str:param properties: property list, like ['p1', 'p2', {'p3': 'value'}]:return: None"""travel = graph.E(e_id) if e_id else graph.E()if label:travel = travel.hasLabel(label)if properties:for p in properties:if isinstance(p, dict):key = list(p.keys())[0]travel = travel.has(key, p.get(key))else:travel = travel.has(p)travel.drop().iterate()

5、查詢節(jié)點(diǎn)

首先，可以根據(jù)節(jié)點(diǎn)（或其ID）查詢?cè)摴?jié)點(diǎn)的所有屬性值，此時(shí)需使用return travel.valueMap().toList()。
其次，可以通過(guò) label 或 properties 來(lái)查詢符合條件的所有節(jié)點(diǎn)，此時(shí)使用return travel.toList()。

def query_vertex(graph, v_id=None, label=None, properties=None):"""query graph vertex (value) list:param graph: graph, type: GraphTraversalSource:param v_id: long vertex id or Vertex(id, label):param label: label, type: str:param properties: property list, like ['p1', 'p2', {'p3': 'value'}]:return: vertex list or vertex value list"""if isinstance(v_id, int):v_id = graph.V().hasId(v_id).next()travel = graph.V(v_id) if v_id else graph.V()if label:travel = travel.hasLabel(label)if properties:for p in properties:if isinstance(p, dict):key = list(p.keys())[0]travel = travel.has(key, p.get(key))else:travel = travel.has(p)# return travel.valueMap().toList()return travel.toList()

6、查詢邊

根據(jù)邊的ID、label 或 properties 來(lái)查詢符合條件的邊的所有屬性值。

def query_edge(graph, e_id=None, label=None, properties=None):"""query graph edge value list:param graph: graph, type: GraphTraversalSource:param e_id: edge id, type str:param label: label, type: str:param properties: property list, like ['p1', 'p2', {'p3': 'value'}]:return: valueMap list"""travel = graph.E(e_id) if e_id else graph.E()if label:travel = travel.hasLabel(label)if properties:for p in properties:if isinstance(p, dict):key = list(p.keys())[0]travel = travel.has(key, p.get(key))else:travel = travel.has(p)return travel.valueMap().toList()

7、查詢所有與節(jié)點(diǎn)相連的邊

根據(jù)節(jié)點(diǎn)（或其ID）查詢與該節(jié)點(diǎn)相連的所有邊。

def query_edges_of_vertex(graph, v_id):"""query all edges of vertex:param graph: graph, type: GraphTraversalSource:param v_id: v_id: long vertex id or Vertex(id, label):return: edge list"""if isinstance(v_id, int):v_id = graph.V().hasId(v_id).next()result = []in_edges = graph.V(v_id).inE().toList()out_edges = graph.V(v_id).outE().toList()result.extend(in_edges)result.extend(out_edges)return result

8、查詢所有與節(jié)點(diǎn)相連的節(jié)點(diǎn)

根據(jù)節(jié)點(diǎn)（或其ID）查詢與該節(jié)點(diǎn)相連的所有節(jié)點(diǎn)。

def query_near_vertex(graph, v_id):"""query near vertices of vertex:param graph: graph, type: GraphTraversalSource:param v_id: v_id: long vertex id or Vertex(id, label):return: vertex list"""if isinstance(v_id, int):v_id = graph.V().hasId(v_id).next()result = []out_v = graph.V(v_id).out().toList()in_v = graph.V(v_id).in_().toList()result.extend(out_v)result.extend(in_v)return result

四、其他方法封裝

1、獲取邊的ID

def get_edge_id(edge):"""get edge id:param edge: Egde(id, label, outV, inV):return: edge id, type str"""return edge.id.get('@value').get('relationId')

2、節(jié)點(diǎn)屬性轉(zhuǎn)字典

def vertex_to_dict(graph, vertex):"""transfer Vertex's info to dict:param graph: graph, type: GraphTraversalSource:param vertex: vertex, Vertex(id, label):return: vertex info dict"""properties = graph.V(vertex).valueMap().toList()[0]for key in properties.keys():properties[key] = properties.get(key)[0]return {'id': vertex.id,'label': vertex.label,'properties': properties}

3、邊屬性轉(zhuǎn)dict

def edge_to_dict(graph, edge):"""transfer Edge's info to dict:param graph: graph, type: GraphTraversalSource:param edge: edge, Edge(id, label, outV, inV):return: edge info dict"""e_id = get_edge_id(edge)properties = graph.E(e_id).valueMap().toList()[0]return {'id': e_id,'label': edge.label,'properties': properties}

4、判斷節(jié)點(diǎn)是否在圖中

對(duì)于已知節(jié)點(diǎn)的 label 和 properties 值，判斷該節(jié)點(diǎn)是否在圖中；如果在，返回該節(jié)點(diǎn)，否則返回None。

def judge_vertex_in_graph(graph, vertex_dict):"""judge a vertex whether in graph:param graph: graph, type: GraphTraversalSource:param vertex_dict: vertex dict, like {'label': 'value1', 'properties': {'p1': 'v1', ...}}:return: None or Vertex(id,label)"""label = vertex_dict.get('label')properties = vertex_dict.get('properties')travel = graph.V()if label:travel = travel.hasLabel(label)if properties:for k in properties.keys():travel = travel.has(k, properties.get(k))if travel.hasNext():return travel.next()return None

5、獲取子圖

根據(jù)特定的節(jié)點(diǎn)信息或者邊信息來(lái)獲取子圖。

def get_sub_graph(graph, vertices=None, edges=None, vertex_properties=None):"""get sub graph:param graph: graph, type: GraphTraversalSource:param vertices: hasLabel('label').has('property').has('age', gt(20)):param edges: hasLabel('label').has('property'):param vertex_properties::return: sub_graph, type: GraphTraversalSource"""strategy = SubgraphStrategy(vertices=vertices, edges=edges, vertex_properties=vertex_properties)return graph.withStrategies(strategy)

五、致謝

以上所有代碼都供參考學(xué)習(xí)，非常感謝大家的閱讀。如果有任何問(wèn)題，都可在下方留言交流，希望大家一起學(xué)習(xí)，共同進(jìn)步。

總結(jié)

以上是生活随笔為你收集整理的gremlin_python使用及增删查改方法封装的全部?jī)?nèi)容，希望文章能夠幫你解決所遇到的問(wèn)題。

如果覺(jué)得生活随笔網(wǎng)站內(nèi)容還不錯(cuò)，歡迎將生活随笔推薦給好友。