elasticsearch 6.4 不计算字段中的 search_analyzer答案

【问题标题】：elasticsearch 6.4 does not count search_analyzer in fieldselasticsearch 6.4 不计算字段中的 search_analyzer
【发布时间】：2019-02-06 13:25:52
【问题描述】：

我对 ElasticSearch 映射有疑问。例如字段name 的映射是：

{
    "name": {
        "type": "keyword",
        "fields": {
            "ngram": {
                "type": "text",
                "analyzer": "ngram_analyzer",
                "search_analyzer": "ngram_analyzer"
            },
            "word": {
                "type": "text",
                "analyzer": "word_analyzer",
                "search_analyzer": "word_analyzer"
            }
        }
    }
}

除了search_analyzer 之外，整个映射都有效，ElasticSearch 似乎忽略了这一点。

分析设置：

{
   "analysis":{
      "analyzer":{
         "ngram_analyzer":{
            "type":"custom",
            "char_filter":[
               "number_char_filter_map",
               "remove_duplicates"
            ],
            "tokenizer":"ngram_tokenizer_whitespace",
            "filter":[
               "lowercase",
               "english_stop"
            ]
         },
         "word_analyzer":{
            "type":"custom",
            "char_filter":[
               "number_char_filter_map",
               "remove_duplicates"
            ],
            "tokenizer":"word_tokenizer",
            "filter":[
               "lowercase",
               "english_stop"
            ]
         }
      },
      "char_filter":{
         "remove_duplicates":{
            "type":"pattern_replace",
            "pattern":"(.)(?=\\1)",
            "replacement":""
         },
         "remove_white_spaces":{
            "type":"pattern_replace",
            "pattern":"(\s)",
            "replacement":""
         }
      },
      "filter":{
         "english_stop":{
            "type":"stop",
            "ignore_case":true,
            "stopwords":"_english_"
         }
      },
      "tokenizer":{
         "ngram_tokenizer":{
            "type":"ngram",
            "min_gram":2,
            "max_gram":7
         },
         "ngram_tokenizer_whitespace":{
            "type":"ngram",
            "min_gram":2,
            "max_gram":7,
            "token_chars":[
               "letter",
               "digit",
               "punctuation",
               "symbol"
            ]
         },
         "word_tokenizer":{
            "type":"standard"
         }
      }
   }
}

根据 ElasticSearch 文档，我没有在字段中找到 search_analyzer 的任何定义。如果此方法不起作用，是否有任何替代结构来包含搜索分析器？

【问题讨论】：

这个怎么样？ elastic.co/guide/en/elasticsearch/reference/current/… 是什么让你觉得... ElasticSearch ignores that ？
@Val 谢谢。但这是一个带有单个分析器的普通字段，我需要的是带有多个分析器的多个字段，以获得更好的搜索结果。这种方式仅适用于 one 字段。并且通过这种方式，我必须多次克隆一个字段，这会影响结果的 _source 属性，这不是一个好看的结果。我在不接受 search_analyzer 的属性中遇到 fields 问题。
@Val 你有什么解决办法吗？如果您能分享您的想法，我将不胜感激

标签： elasticsearch analyzer elasticsearch-6 elasticsearch-analyzers

【解决方案1】：

几天后问题解决了... 问题是使用与search_analyzer 相同的分析器。我只是在设置中设置了另一个名称不同的分析器（使用与分析器相同的键，search_analyzer 似乎会导致问题并忽略 search_analyzer）。

克隆ngram_analyzer 到ngram_search_analyzer 将word_analyzer克隆为word_search_analyzer

以及创建映射请求：

PUT /suggestions HTTP/1.1
Host: localhost:9200
Content-Type: application/json

{
   "mappings":{
      "doc":{
         "properties":{
            "caption":{
               "type":"keyword",
               "fields":{
                  "ngram":{
                     "type":"text",
                     "analyzer":"ngram_analyzer",
                     "search_analyzer":"ngram_search_analyzer"
                  },
                  "word":{
                     "type":"text",
                     "analyzer":"word_analyzer",
                     "search_analyzer":"word_search_analyzer"
                  }
               }
            }
         }
      }
   },
   "settings":{
      "number_of_shards":1,
      "number_of_replicas":1,
      "routing_partition_size":1,
      "analysis":{
         "analyzer":{
            "ngram_analyzer":{
               "type":"custom",
               "char_filter":[
                  "number_char_filter_map",
                  "remove_duplicates"
               ],
               "tokenizer":"ngram_tokenizer_whitespace",
               "filter":[
                  "lowercase",
                  "english_stop"
               ]
            },
            "ngram_search_analyzer":{
               "type":"custom",
               "char_filter":[
                  "number_char_filter_map",
                  "remove_duplicates"
               ],
               "tokenizer":"ngram_tokenizer_whitespace",
               "filter":[
                  "lowercase",
                  "english_stop"
               ]
            },
            "word_analyzer":{
               "type":"custom",
               "char_filter":[
                  "number_char_filter_map",
                  "remove_duplicates"
               ],
               "tokenizer":"word_tokenizer",
               "filter":[
                  "lowercase",
                  "english_stop"
               ]
            },
            "word_search_analyzer":{
               "type":"custom",
               "char_filter":[
                  "number_char_filter_map",
                  "remove_duplicates"
               ],
               "tokenizer":"word_tokenizer",
               "filter":[
                  "lowercase",
                  "english_stop"
               ]
            }
         },
         "char_filter":{
            "number_char_filter_map":{
               "type":"mapping",
               "mappings":[
                  "\u0660 => 0",
                  "\u0661 => 1",
                  "\u0662 => 2",
                  "\u0663 => 3",
                  "\u0664 => 4",
                  "\u0665 => 5",
                  "\u0666 => 6",
                  "\u0667 => 7",
                  "\u0668 => 8",
                  "\u0669 => 9",
                  "\u06f0 => 0",
                  "\u06f1 => 1",
                  "\u06f2 => 2",
                  "\u06f3 => 3",
                  "\u06f4 => 4",
                  "\u06f5 => 5",
                  "\u06f6 => 6",
                  "\u06f7 => 7",
                  "\u06f8 => 8",
                  "\u06f9 => 9"
               ]
            },
            "remove_duplicates":{
               "type":"pattern_replace",
               "pattern":"(.)(?=\\1)",
               "replacement":""
            },
            "remove_white_spaces":{
               "type":"pattern_replace",
               "pattern":"(\\s)",
               "replacement":""
            }
         },
         "filter":{
            "english_stop":{
               "type":"stop",
               "ignore_case":true,
               "stopwords":"_english_"
            }
         },
         "tokenizer":{
            "ngram_tokenizer":{
               "type":"ngram",
               "min_gram":2,
               "max_gram":7
            },
            "ngram_tokenizer_whitespace":{
               "type":"ngram",
               "min_gram":2,
               "max_gram":7,
               "token_chars":[
                  "letter",
                  "digit",
                  "punctuation",
                  "symbol"
               ]
            },
            "word_tokenizer":{
               "type":"standard",
               "token_chars":[

               ]
            }
         }
      }
   }
}

现在我在映射中看到了搜索分析器：]

我认为拥有不同的分析器和搜索分析器以供以后自定义也是一件好事。

【讨论】：