• 大致有三種解決方法
  • scan and scroll all documents
  • _update_by_query api
  • re-index and add new fields, use the /_reindex API
• 首先要設(shè)置允許script運行，修改elasticsearch.yml

script.engine.groovy.inline.aggs: on
script.engine.groovy.inline.search: on
script.engine.groovy.inline.update: on

或者
script.inline: true

• scan anc scroll all documents
  • use /_search?scroll to fetch the docs
  • perform your operation
  • send /_bulk update requests

POST bt/bt/_update
{
    "script" : "ctx._source.new_field = \"value_of_new_field\""
}

• _update_by_query API

POST bt/bt/_update_by_query
{
  "script": {
    "inline": "if (ctx._source.bt0 == null || ctx._source.bt1==null) { ctx._source.btNew1=null } else { ctx._source.btNew1 = ctx._source.bt0 + ctx._source.bt1 }"
  }
}

• reindex
  https://www.elastic.co/guide/en/elasticsearch/reference/master/docs-reindex.html

{
  "source": {
    "index": "bt",
    "type": "bt"
  },

  "dest": {
    "index": "new_bt1"
  },
  "script": {
    "inline": "if (ctx._source.bt0 == null || ctx._source.bt1==null) { ctx._source.btPlus=null } else { ctx._source.btPlus = ctx._source.bt0 + ctx._source.bt1 }"
  }
}

• Difference between _update_by_query and _reindex
  • Just like _update_by_query, _reindex gets a snapshot of the source index but its target must be a different index so version conflicts are unlikely.
  • Unlike _update_by_query, the script is allowed to modify the document’s metadata.

其實_update_and_query和_reindex的實現(xiàn)差不多，性能沒有測過，不過應(yīng)該差不多。所以elasticsearch還有task api可以用來檢測這兩個任務(wù)運行情況。