目录
HugeGraph 索引介绍
二级索引
组合索引
范围索引
全文索引
HugeGraph 索引介绍
二级索引
创建schema和添加数据
schema.propertyKey("name").asText().ifNotExist().create();schema.propertyKey("uid").asLong().ifNotExist().create();schema.propertyKey("city").asText().ifNotExist().create();schema.propertyKey("mid").asText().ifNotExist().create();schema.propertyKey("url").asText().ifNotExist().create();schema.propertyKey("date").asText().ifNotExist().create();schema.vertexLabel("article").properties("name", "uid", "city","mid","url").useCustomizeStringId().ifNotExist().create(); schema.indexLabel("articleByNameSecondary").onV("article").by("name").secondary().ifNotExist().create();graph.addVertex(T.label, "article",T.id,"id7", "mid", "lakjdflkjalskdjfl", "name", "北京三联WiFi熟练度附近", "uid", 1234567897,"city","未知","url","http://test.com//IodUhh2bV");graph.addVertex(T.label, "article",T.id,"id2", "mid", "lakjdflkjalskdjfl", "name", "北京三联法律文件而来反馈", "uid", 1234567891,"city","未知","url","http://test.com//IodUhh2bV");graph.addVertex(T.label, "article",T.id,"id3", "mid", "lakjdflkjalskdjfl", "name", "深圳广州WiFi熟练度附近", "uid", 1234567892,"city","未知","url","http://test.com//IodUhh2bV");graph.addVertex(T.label, "article",T.id,"id4", "mid", "lakjdflkjalskdjfl", "name", "深圳WiFi熟练度附近", "uid", 1234567893,"city","未知","url","http://test.com//IodUhh2bV");graph.addVertex(T.label, "article",T.id,"id5", "mid", "lakjdflkjalskdjfl", "name", "广州测试", "uid", 1234567894,"city","未知","url","http://test.com//IodUhh2bV");graph.addVertex(T.label, "article",T.id,"id6", "mid", "lakjdflkjalskdjfl", "name", "数据", "uid", 1234567895,"city","未知","url","http://test.com//IodUhh2bV");
根据name查询，该方法只能全不值匹配，不能模糊匹配
模糊匹配使用该方法，不推荐这么使用，filter过滤数据，当数据量很大的时候性能会下降
g.V().hasLabel("article").filter{it.get().property("name").value().contains("北京")};
或者使用下面这个方法进行正则匹配
g.V().hasLabel("article").filter{it.get().property("name").value().matches("北京(.*)")};可以对结果进行过滤
组合索引
待补充
范围索引
待补充
全文索引
创建索引标签
设置定点article的name字段为全文索引（可以模糊检索）
schema.indexLabel("articleByNameSearch").onV("article").by("name").search().ifNotExist().create();其他schema设置
schema.propertyKey("name").asText().ifNotExist().create();schema.propertyKey("uid").asLong().ifNotExist().create();schema.propertyKey("city").asText().ifNotExist().create();schema.propertyKey("mid").asText().ifNotExist().create();schema.propertyKey("url").asText().ifNotExist().create();schema.propertyKey("date").asText().ifNotExist().create();schema.vertexLabel("article").properties("name", "uid", "city","mid","url")// .primaryKeys("name").useCustomizeStringId().ifNotExist().create();添加数据
graph.addVertex(T.label, "article",T.id,"id7", "mid", "lakjdflkjalskdjfl", "name", "北京三联WiFi熟练度附近", "uid", 1234567897,"city","未知","url","http://test.com//IodUhh2bV");
graph.addVertex(T.label, "article",T.id,"id2", "mid", "lakjdflkjalskdjfl", "name", "北京三联法律文件而来反馈", "uid", 1234567891,"city","未知","url","http://test.com//IodUhh2bV");
graph.addVertex(T.label, "article",T.id,"id3", "mid", "lakjdflkjalskdjfl", "name", "深圳广州WiFi熟练度附近", "uid", 1234567892,"city","未知","url","http://test.com//IodUhh2bV");
graph.addVertex(T.label, "article",T.id,"id4", "mid", "lakjdflkjalskdjfl", "name", "深圳WiFi熟练度附近", "uid", 1234567893,"city","未知","url","http://test.com//IodUhh2bV");
graph.addVertex(T.label, "article",T.id,"id5", "mid", "lakjdflkjalskdjfl", "name", "广州测试", "uid", 1234567894,"city","未知","url","http://test.com//IodUhh2bV");
graph.addVertex(T.label, "article",T.id,"id6", "mid", "lakjdflkjalskdjfl", "name", "数据", "uid", 1234567895,"city","未知","url","http://test.com//IodUhh2bV");
数据添加成功，查看数据
| id | label | properties |
|---|---|---|
| id2 | article | {"name":"北京三联法律文件而来反馈","uid":1234567891,"city":"未知","mid":"lakjdflkjalskdjfl","url":"http://test.com//IodUhh2bV"} |
| id3 | article | {"name":"深圳广州WiFi熟练度附近","uid":1234567892,"city":"未知","mid":"lakjdflkjalskdjfl","url":"http://test.com//IodUhh2bV"} |
| id4 | article | {"name":"深圳WiFi熟练度附近","uid":1234567893,"city":"未知","mid":"lakjdflkjalskdjfl","url":"http://test.com//IodUhh2bV"} |
| id5 | article | {"name":"广州测试","uid":1234567894,"city":"未知","mid":"lakjdflkjalskdjfl","url":"http://test.com//IodUhh2bV"} |
| id6 | article | {"name":"数据","uid":1234567895,"city":"未知","mid":"lakjdflkjalskdjfl","url":"http://test.com//IodUhh2bV"} |
| id7 | article | {"name":"北京三联WiFi熟练度附近","uid":1234567897,"city":"未知","mid":"lakjdflkjalskdjfl","url":"http://test.com//IodUhh2bV"} |
查询
添加索引的数据会进行分词，我们在HugeGraph中可以看到配置文件 hugegraph.properties 中有配置分词项，这里使用jieba分词
search.text_analyzer=jieba
search.text_analyzer_mode=INDEX执行查询
g.V().hasLabel("article").has("name", Text.contains("熟练度"))查询结果：
| id | label | properties |
|---|---|---|
| id3 | article | {"name":"深圳广州WiFi熟练度附近","uid":1234567892,"city":"未知","mid":"lakjdflkjalskdjfl","url":"http://test.com//IodUhh2bV"} |
| id4 | article | {"name":"深圳WiFi熟练度附近","uid":1234567893,"city":"未知","mid":"lakjdflkjalskdjfl","url":"http://test.com//IodUhh2bV"} |
| id7 | article | {"name":"北京三联WiFi熟练度附近","uid":1234567897,"city":"未知","mid":"lakjdflkjalskdjfl","url":"http://test.com//IodUhh2bV"} |
我们可以看出可以模糊查询了
我们再查询“熟练”，发现数据也可以出来
当查询“熟”一个字的时候发现没有数据，搜索“练度”也没有数据，原因是索引将数据进行分词，将分词后的每个词进行索引，如果你搜索的词在分词的时候没有分词出来，最终搜索的结果就会搜索不到，“熟”，“练度”这两个字（词），在JieBa分词中没有，所以搜索不到。
所以查询使用使用Text.contains()，也并非我们开发中判断字符串中是否包含该字符串的意思。
