Elasticsearch 基于脚本进行 partial update

关注

发布于: 2021 年 01 月 27 日

文字内容整理自 B 站中华石杉的 Elasticsearch 顶尖高手系列课程核心知识篇，教程原本是基于 Groovy 脚本，现在的 Elasticsearch 似乎已经不怎么支持了，我把脚本改为了 painless 的，过程一点都不 painless。

Elasticsearch 有个内置的脚本支持，可以基于 ~~groovy~~ painless 脚本实现各种各样的复杂操作。

先生成一条测试数据。

PUT test_index/_doc/12{  "num": 0,  "tags": []}
{  "_index" : "test_index",  "_type" : "_doc",  "_id" : "12",  "_version" : 1,  "result" : "created",  "_shards" : {    "total" : 2,    "successful" : 2,    "failed" : 0  },  "_seq_no" : 4,  "_primary_term" : 1}

复制代码

内置脚本 inline script

POST test_index/_update/11{  "script": "ctx._source.num+=1"}
{  "error" : {    "root_cause" : [      {        "type" : "illegal_argument_exception",        "reason" : "failed to execute script"      }    ],    "type" : "illegal_argument_exception",    "reason" : "failed to execute script",    "caused_by" : {      "type" : "script_exception",      "reason" : "runtime error",      "script_stack" : [        "ctx._source.num+=1",        "                 ^---- HERE"      ],      "script" : "ctx._source.num+=1",      "lang" : "painless",      "position" : {        "offset" : 17,        "start" : 0,        "end" : 18      },      "caused_by" : {        "type" : "null_pointer_exception",        "reason" : null      }    }  },  "status" : 400}
POST test_index/_update/11{  "script": "ctx._source.num+=1",  "lang": "groovy"}
{  "error" : {    "root_cause" : [      {        "type" : "x_content_parse_exception",        "reason" : "[3:3] [UpdateRequest] unknown field [lang]"      }    ],    "type" : "x_content_parse_exception",    "reason" : "[3:3] [UpdateRequest] unknown field [lang]"  },  "status" : 400}

复制代码

Elasticsearch 似乎已经不支持 groovy 了，从官方文档看大概是从 5.6 版本，默认的脚本语言从 groovy 改为了 painless。

POST test_index/_update/11{  "script": {    "lang": "painless",    "source": "ctx._source.num+=1"  }}
{  "_index" : "test_index",  "_type" : "_doc",  "_id" : "11",  "_version" : 7,  "result" : "updated",  "_shards" : {    "total" : 2,    "successful" : 2,    "failed" : 0  },  "_seq_no" : 8,  "_primary_term" : 1}

复制代码

外部脚本

POST _scripts/test-add-tags{  "script": {    "lang": "groovy",    "source": "ctx._source.tags+=new_tag"  }}
{  "error" : {    "root_cause" : [      {        "type" : "illegal_argument_exception",        "reason" : "unable to put stored script with unsupported lang [groovy]"      }    ],    "type" : "illegal_argument_exception",    "reason" : "unable to put stored script with unsupported lang [groovy]"  },  "status" : 400}
POST _scripts/test-add-tags{  "script": {    "lang": "painless",    "source": "ctx._source.tags+=new_tag"  }}
{  "acknowledged" : true}

复制代码

虽然 painless 的脚本写进去了，但是后来执行的时候却发生了编译错误，似乎是不支持"+=" 操作。于是我把脚本改的简单了一点。

POST _scripts/test-add-tags{  "script": {    "lang": "painless",    "source": "ctx._source.num += params.my_val"  }}
POST test_index/_update/12{  "script": {    "id": "test-add-tags",    "params": {      "my_val": 100    }  }}

复制代码

用脚本删除文档

PUT test_index/_doc/1{  "num": 1}
POST _scripts/test-delete-document{  "script": {    "lang": "painless",    "source": "ctx.op = ctx._source.num == params.count ? 'delete':'none'"  }}
POST test_index/_update/1{  "script": {    "id": "test-delete-document",    "params": {      "count": 12    }  }}
{  "_index" : "test_index",  "_type" : "_doc",  "_id" : "1",  "_version" : 3,  "result" : "updated",  "_shards" : {    "total" : 2,    "successful" : 2,    "failed" : 0  },  "_seq_no" : 24,  "_primary_term" : 1}
POST test_index/_update/1{  "script": {    "id": "test-delete-document",    "params": {      "count": 1    }  }}
{  "_index" : "test_index",  "_type" : "_doc",  "_id" : "1",  "_version" : 4,  "result" : "updated",  "_shards" : {    "total" : 2,    "successful" : 2,    "failed" : 0  },  "_seq_no" : 25,  "_primary_term" : 1}
# Elasticsearch 5.2# groovy# ctx.op = ctx._source.num == count ? 'delete' : 'none' 
POST /test_index/test_type/11/_update{ct  "script": {    "lang": "groovy",    "file": "test-delete-document",    "params": {      "count": 1    }  }}

复制代码

upsert 操作

POST test_index/_update/13{  "doc": {    "num": 13  }}
{  "error" : {    "root_cause" : [      {        "type" : "document_missing_exception",        "reason" : "[_doc][13]: document missing",        "index_uuid" : "itFrdXbPQ-ukBl1sp5EbOA",        "shard" : "0",        "index" : "test_index"      }    ],    "type" : "document_missing_exception",    "reason" : "[_doc][13]: document missing",    "index_uuid" : "itFrdXbPQ-ukBl1sp5EbOA",    "shard" : "0",    "index" : "test_index"  },  "status" : 404}

复制代码

如果指定的 document 不存在，就执行 upsert 中的初始化操作；如果指定的 document 存在，就执行 doc 或者 script 指定的 partial update 操作。

POST test_index/_update/13{  "script": "ctx._source.num+=1",  "upsert": {    "num": 13,    "tags": []  }}
{  "_index" : "test_index",  "_type" : "_doc",  "_id" : "13",  "_version" : 1,  "result" : "created",  "_shards" : {    "total" : 2,    "successful" : 2,    "failed" : 0  },  "_seq_no" : 35,  "_primary_term" : 1}

复制代码

如果再次执行，就可以看到执行了内置脚本中的 +=1 操作。

发布于: 2021 年 01 月 27 日阅读数: 26

原文链接:【http://xie.infoq.cn/article/2be3e3ff62def0bfe58760649】。

escray

关注

Let's Go 2017.11.19 加入

在学 Elasticsearch 的项目经理

发布

暂无评论

创作场景

Elasticsearch 基于脚本进行 partial update

内置脚本 inline script

外部脚本

用脚本删除文档

upsert 操作

escray

评论