Elasticsearch无法按单个词项搜索的问题解决咨询

问题背景 #

现有Elasticsearch环境信息如下：

1. 目标文档 #

{"_index" : "trigger","_type" : "_doc","_id" : "urn:trigger:(urn:analyticsplatform:uc4):acxiom_au_attr.done","_version" : 45,"_seq_no" : 420508,"_primary_term" : 20,"found" : true,"_source" : {"_boostScoreFactor" : 1.0,"type" : "trigger","title" : "acxiom_au_attr.done",}}

2. title字段映射 #

"title" : {
  "type" : "text",
  "fields" : {
    "keyword" : {
      "type" : "keyword",
      "ignore_above" : 256
    }
  }
}

3. 索引设置 #

{"trigger" : {"settings" : {"index" : {"routing" : {"allocation" : {"total_shards_per_node" : "2"}},"number_of_shards" : "1","provided_name" : "trigger","creation_date" : "1668526304683","unassigned" : {"node_left" : {"delayed_timeout" : "15m"}},"analysis" : {"filter" : {"discovery_stemmer" : {"type" : "stemmer","language" : "porter2"},"discovery_stemmer_possessive" : {"type" : "stemmer","language" : "minimal_english"},"discovery_synonym_abbr" : {"type" : "synonym_graph","synonyms" : ["BN => browse node"]}},"char_filter" : {"discovery_charfilter" : {"type" : "mapping","mappings" : [".=>|","_=>|"]}},"normalizer" : {"lowercase" : {"filter" : ["lowercase"],"type" : "custom","char_filter" : [ ]}},"analyzer" : {"discovery_index_analyzer" : {"filter" : ["lowercase","discovery_synonym_abbr"],"char_filter" : ["discovery_charfilter"],"type" : "custom","tokenizer" : "standard"}}},"number_of_replicas" : "2","uuid" : "oMzEq9L_Sum8A5ikwLyCdQ","version" : {"created" : "135238227"}}}}

4. 预期分词结果 #

使用discovery_index_analyzer分析acxiom_au_attr.done得到的词项：

{"tokens" : [{"token" : "acxiom","start_offset" : 0,"end_offset" : 6,"type" : "<ALPHANUM>","position" : 0},{"token" : "au","start_offset" : 7,"end_offset" : 9,"type" : "<ALPHANUM>","position" : 1},{"token" : "attr","start_offset" : 10,"end_offset" : 14,"type" : "<ALPHANUM>","position" : 2},{"token" : "done","start_offset" : 19,"end_offset" : 23,"type" : "<ALPHANUM>","position" : 4}]}

问题现象 #

• 执行单个词项查询时无结果返回：

GET trigger/_search
{"query": {"match": {"title": "acxiom"}}}

• 使用完整title值查询时可正常返回目标文档：

GET trigger/_search
{"query": {"match": {"title": "acxiom_au_attr.done"}}}

问题原因 #

title字段未明确指定使用自定义的discovery_index_analyzer作为索引和查询分析器，导致：

1. 索引时字段使用Elasticsearch默认的standard分析器，该分析器不会将_和.视为分隔符，因此acxiom_au_attr.done会被拆分为acxiom、au、attr.done三个词项，与预期的分词结果不符；
2. 查询时默认使用的分析器与索引时不一致，进一步导致词项无法匹配。

解决方案 #

方案1：修改字段映射并重新索引数据（推荐） #

由于已创建的text字段无法直接修改分析器，需重新创建索引并迁移数据：

1. 创建新索引（例如trigger_new），配置正确的分析器及字段映射：

PUT trigger_new
{
  "settings": {
    "index": {
      "routing": {
        "allocation": {
          "total_shards_per_node": "2"
        }
      },
      "number_of_shards": "1",
      "unassigned": {
        "node_left": {
          "delayed_timeout": "15m"
        }
      },
      "analysis": {
        "filter": {
          "discovery_stemmer": {
            "type": "stemmer",
            "language": "porter2"
          },
          "discovery_stemmer_possessive": {
            "type": "stemmer",
            "language": "minimal_english"
          },
          "discovery_synonym_abbr": {
            "type": "synonym_graph",
            "synonyms": ["BN => browse node"]
          }
        },
        "char_filter": {
          "discovery_charfilter": {
            "type": "mapping",
            "mappings": [".=>|", "_=>|"]
          }
        },
        "normalizer": {
          "lowercase": {
            "filter": ["lowercase"],
            "type": "custom",
            "char_filter": []
          }
        },
        "analyzer": {
          "discovery_index_analyzer": {
            "filter": ["lowercase", "discovery_synonym_abbr"],
            "char_filter": ["discovery_charfilter"],
            "type": "custom",
            "tokenizer": "standard"
          }
        }
      },
      "number_of_replicas": "2"
    }
  },
  "mappings": {
    "_doc": {
      "properties": {
        "title": {
          "type": "text",
          "analyzer": "discovery_index_analyzer",
          "search_analyzer": "discovery_index_analyzer",
          "fields": {
            "keyword": {
              "type": "keyword",
              "ignore_above": 256
            }
          }
        },
        "_boostScoreFactor": {
          "type": "float"
        },
        "type": {
          "type": "text"
        }
      }
    }
  }
}

2. 将原索引数据迁移至新索引：

POST _reindex
{
  "source": {
    "index": "trigger"
  },
  "dest": {
    "index": "trigger_new"
  }
}

3. 测试单个词项查询：

GET trigger_new/_search
{
  "query": {
    "match": {
      "title": "acxiom"
    }
  }
}

方案2：查询时指定分析器（临时解决） #

如果暂时无法重新索引，可在查询时强制指定使用discovery_index_analyzer，确保查询词的分词逻辑与预期一致：

GET trigger/_search
{
  "query": {
    "match": {
      "title": {
        "query": "acxiom",
        "analyzer": "discovery_index_analyzer"
      }
    }
  }
}

内容的提问来源于stack exchange，提问作者Jacky Guo