目录
向ES中添加文档时, ES会根据文档中各个字段的类型, 自动推测并创建映射(mapping), 这是比Solr灵活的地方;
这在大部分情况下是可用的, 但生产环境中不建议如此使用: 应该根据业务需求定制mapping映射. 关于定制mapping, 请参考后续文章.
1 创建、删除索引
1.1 创建索引
# 创建索引API: PUT test_index?pretty# 响应信息如下:#! Deprecation: the default number of shards will change from [5] to [1] in 7.0.0; # if you wish to continue using the default of [5] shards, you must manage this on the create index request or with an index template{ "acknowledged": true, "shards_acknowledged": true, "index": "test_index"}# 查看集群中的索引: health status index uuid pri rep docs.count docs.deleted store.size pri.store.sizeyellow open test_index hMeJ-M9pSHSXl0t39OedYw 5 1 0 0 1.2kb 1.2kbyellow open .kibana_1 4q7ELvdcTVilW3UwtMWqeg 1 0 18 0 78.5kb 78.5kb 过时说明:
在创建索引时, Elasticsearch提出过时警告: 从7.0.0版本开始, 默认的Shard个数将从[5]变为[1].
如果要继续使用默认的[5]个分片(Shard), 就需要在创建Index时指定, 或者通过索引模板创建Index. 关于创建Index时指定分片个数的方法, 参见后续的博文.
2.2 删除索引
# 删除索引API:DELETE test_index?pretty# 响应信息如下: { "acknowledged": true} 2 document的结构
ES是一款面向文档的数据搜索、分析引擎. document结构说明:
(1) 基于面向对象的开发思想, 应用系统中的数据结构都是很复杂的: 对象中嵌套对象, 如CRM系统中的客户对象中, 还会嵌入客户相关的企业对象.
(2) 对象数据存储到数据库中, 需要分解, 将嵌套对象分解为扁平的多张表数据, 每次操作时需要还原回对象格式, 过程繁琐.
(3) ES存储的是JSON格式的文档, 基于此, ES可以提供复杂的索引, 全文检索, 分析聚合等功能.
(4) document格式示例:
{ "id": "5220", "name": "张三", "sex": "男", "age": 25, "phone": 13312345678, "email": "zhangsan@163.com", "company": { "name": "Alibaba", "location": "杭州" }, "join_date": "2018/11/01"}
接下来以电商系统中的搜索子系统为例, 演示对文档的操作方法.
3 添加文档
(1) 添加API:
PUT index/type/id{ "JSON格式的文档数据"} 说明: ES会自动创建PUT API中指定的index和type, 不需要提前创建;
而且ES默认对document的每个field都建立倒排索引, 保证它们都可以被检索.
(2) 添加示例:
PUT book_shop/it_book/1{ "name": "Java编程思想", "author": "[美] Bruce Eckel", "category": "编程语言", "desc": "Java学习必读经典,殿堂级著作！", "price": 109.0, "publisher": "机械工业出版社", "date": "2007-06-01", "tags": [ "Java", "编程语言" ]} (3) 添加之后的响应信息:
{ "_index" : "book_shop", "_type" : "it_book", "_id" : "1", "_version" : 1, "result" : "created", # 操作结果: created(创建)了索引 "_shards" : { "total" : 2, "successful" : 1, "failed" : 0 }, "_seq_no" : 0, "_primary_term" : 1} (4) 再添加如下数据:
PUT book_shop/it_book/2{ "name": "深入理解Java虚拟机：JVM高级特性与最佳实践", "author": "周志明", "category": "编程语言", "desc": "Java图书领域公认的经典著作", "price": 79.0, "date": "2013-10-01", "publisher": "机械工业出版社", "tags": [ "Java", "虚拟机", "最佳实践" ]}PUT book_shop/it_book/3{ "name": "Java并发编程的艺术", "author": "方腾飞，魏鹏，程晓明", "category": "编程语言", "desc": "阿里系工程师的并发编程实践", "price": 59.0, "date": "2015-07-10", "publisher": "机械工业出版社", "tags": [ "Java", "并发编程" ]} 4 查询文档
(1) 检索API:
GET index/type/id
(2) 检索示例:
GET book_shop/it_book/2
(3) 检索的结果:
{ "_index" : "book_shop1", "_type" : "it_book", "_id" : "2", "_version" : 1, "_seq_no" : 0, "_primary_term" : 1, "found" : true, "_source" : { "name" : "深入理解Java虚拟机：JVM高级特性与最佳实践", "author" : "周志明", "category" : "编程语言", "desc" : "Java图书领域公认的经典著作", "price" : 79.0, "date" : "2013-10-01", "publisher" : "机械工业出版社", "tags" : [ "Java", "虚拟机", "最佳实践" ] }} 5 修改文档
5.1 替换文档
(1) 替换API - 与添加API相同, 只不过文档id要存在, 此时系统将判定为修改操作:
—— 无论文档数据是否存在修改, 对同一id的文档执行1次以上的PUT操作, 都是修改操作.
PUT index/type/id{ "JSON格式的文档数据"} 注意: 替换方式的不便之处: 必须填写要修改文档的所有field, 如果缺少, 修改后的文档中将丢失相关field.
(2) 替换示例 - 为name添加了"（第4版）"
PUT book_shop/it_book/1{ "name": "Java编程思想（第4版）", "author": "[美] Bruce Eckel", "category": "编程语言", "desc": "Java学习必读经典,殿堂级著作！", "price": 109.0, "date": "2007-06-01", "publisher": "机械工业出版社", "tags": [ "Java", "编程语言" ]} (3) 替换的结果信息:
{ "_index" : "book_shop", "_type" : "it_book", "_id" : "1", "_version" : 2, "result" : "updated", // 操作结果: updated(修改)了索引 "_shards" : { "total" : 2, "successful" : 1, "failed" : 0 }, "_seq_no" : 1, "_primary_term" : 1} 5.2 更新文档
(1) 更新API:
—— 通过POST和_update, 更新文档中的特定字段(field), 其他的字段不会改动.
POST index/type/id/_update{ "doc": { "field u want to update": "new value of ur update's field" }} 注意: 与替换方式相比, 更新方式的好处: 可以更新指定文档的指定field, 未指定的field也不会丢失.
(2) 更新示例 - 将name改为英文:
POST book_shop/it_book/1/_update{ "doc": { "name": "Thinking in Java(4th Edition) " }} (3) 更新的结果信息:
{ "_index" : "book_shop", "_type" : "it_book", "_id" : "1", "_version" : 3, "result" : "updated", // 操作结果: updated(修改)了索引 "_shards" : { "total" : 2, "successful" : 1, "failed" : 0 }, "_seq_no" : 2, "_primary_term" : 1} 此时查看该文档 GET book_shop/it_book/1, 可以发现更新成功.
6 删除文档
(1) 删除API:
DELETE index/type/id
(2) 删除示例:
DELETE book_shop/it_book/1
(3) 删除的结果信息:
{ "_index" : "book_shop", "_type" : "it_book", "_id" : "1", "_version" : 4, "result" : "deleted", // 操作结果: deleted(删除)了索引 "_shards" : { "total" : 2, "successful" : 1, "failed" : 0 }, "_seq_no" : 3, "_primary_term" : 1} (4) 删除确认: 再次查看删除的文档:
GET book_shop/it_book/1
(5) 确认的信息:
{ "_index" : "book_shop", "_type" : "it_book", "_id" : "1", "found" : false // 没有查找到相关文档} 为了后期演示的方便, 再次将该文档添加至索引中:
PUT book_shop/it_book/1{ "name": "Java编程思想（第4版）", "author": "[美] Bruce Eckel", "category": "编程语言", "desc": "Java学习必读经典,殿堂级著作！", "price": 109.0, "date": "2007-06-01", "publisher": "机械工业出版社", "tags": [ "Java", "编程语言" ]}
版权声明
作者:
出处: 博客园
感谢阅读, 如果文章有帮助或启发到你, 点个[好文要顶?] 或 [推荐?] 吧?
本文版权归博主所有, 欢迎转载, 但 [必须在文章页面明显位置标明原文链接], 否则博主保留追究相关人员法律责任的权利.