Elasticsearch给index新增字段:(1)固定值 (2)计算值

Elasticsearch给所有记录新增一个字段:(1)字段是固定值 (2)字段是其他字段计算值

1. 固定值: _update_by_query加固定值字段

新增一个speaker_bak字段,值是'HAMLET'

方法1:

POST hamlet/_update_by_query
{
 "script": {
 "source": "ctx._source.speaker_bak='HAMLET'",
 "lang": "painless"
 },
 "query": {
 "match_all": {}
 }
}

现在我们删掉这个字段, 用pipeline再试一次

POST hamlet/_update_by_query
{
 "script": {
 "source": "ctx._source.remove('speaker_bak')",
 "lang": "painless"
 }
}

扩展 如果存在字段 talked, 则将它的值增加increment;否则赋值为1:

POST hamlet/_update_by_query
{
 "script": {
 "lang": "painless",
 "source": """
 if(ctx._source.talked != null){
 ctx._source.talked += params.increment; 
 } else {
 ctx._source.talked = 1;
 }
 """,
 "params": {
 "increment": 3
 }
 }
}

方法2:通过 pipeline

PUT _ingest/pipeline/addSpeakerBak
{
 "description": "增加字段的pipeline",
 "processors": [
 {
 "set": {
 "field": "speaker_bak",
 "value": "HAMLET"
 }
 }
 ]
}
POST hamlet/_update_by_query?pipeline=addSpeakerBak

2. 新加一个计算值字段(通过其他字段)

新增一个entry_len字段,值是text_entry字段的长度

POST hamlet/_update_by_query
{
 "script": {
 "source": "ctx._source.entry_len=ctx._source.text_entry.length()",
 "lang": "painless"
 },
 "query": {
 "match_all": {}
 }
}

3.已有字段改名:旧字段改名

貌似只能通过pipeline

PUT _ingest/pipeline/renameField
{
 "description": "旧字段改名",
 "processors": [
 {
 "rename": {
 "field": "text_entry",
 "target_field": "textEntry"
 }
 }
 ]
}
POST hamlet/_update_by_query?pipeline=renameField