Elasticsearch学习之的delimited_payloads使用

想要为每个token添加负载信息，例如标注词性

• elasticsearch中Token Filter：delimited_payloads
• 注意事项：在设置tokenizer时必须不能把delimited_payloads中的分界符去掉
• 实现：

PUT /20181105
{
  "settings": {
    "analysis": {
      "filter": {
        "payloads":{
          "type":"delimited_payload",
          "encoding":"int",
          "delimiter":"|"
        }
      },
      "analyzer": {
        "payloads":{
          "tokenizer":"whitespace",
          "filter":["payloads"],
          "char_filter":[]
          
        }
      }
    }
  },
  "mappings": {
    "doc":{
      "properties": {
        "text":{
          "type": "text",
          "analyzer": "payloads"
        }
      }
    }
  }
}

• 测试：

GET  20181105/_analyze 
{
  "analyzer": "payloads",
  "text": "the|1  Quick|2 fox|3"
  
}

• 结果：

"tokenfilters": [
      {
        "name": "payloads",
        "tokens": [
          {
            "token": "the",
            "start_offset": 0,
            "end_offset": 5,
            "type": "word",
            "position": 0,
            "bytes": "[74 68 65]",
            "payload": "[0 0 0 1]",
            "positionLength": 1,
            "termFrequency": 1
          },
          {
            "token": "Quick",
            "start_offset": 6,
            "end_offset": 13,
            "type": "word",
            "position": 1,
            "bytes": "[51 75 69 63 6b]",
            "payload": "[0 0 0 2]",
            "positionLength": 1,
            "termFrequency": 1
          },
          {
            "token": "fox",
            "start_offset": 14,
            "end_offset": 19,
            "type": "word",
            "position": 2,
            "bytes": "[66 6f 78]",
            "payload": "[0 0 0 3]",
            "positionLength": 1,
            "termFrequency": 1
          }
        ]
      }
    ]
  }
}