基本格式如下所示

  1. PUT /my-index/_settings
  2. {
  3.   "index": {
  4.     "analysis": {
  5.       "analyzer": {
  6.         "customHTMLSnowball": {
  7.          "type": "custom",
  8.           "char_filter": [
  9.             "html_strip"
  10.           ],
  11.           "tokenizer": "standard",
  12.           "filter": [
  13.             "lowercase",
  14.             "stop",
  15.             "snowball"
  16.           ]
  17.         }}}}}

案例 1

  1. {
  2.     "settings": {
  3.         "refresh_interval": "5s", 
  4.         "number_of_shards": 1, 
  5.         "number_of_replicas": 1, 
  6.         "analysis": {
  7.             "filter": {
  8.                 "edge_ngram_filter": {
  9.                     "type": "edge_ngram", 
  10.                     "min_gram": 1, 
  11.                     "max_gram": 50
  12.                 }, 
  13.                 "pinyin_simple_filter": {
  14.                     "type": "pinyin", 
  15.                     "keep_first_letter": true, 
  16.                     "keep_separate_first_letter": false, 
  17.                     "keep_full_pinyin": false, 
  18.                     "keep_original": false, 
  19.                     "limit_first_letter_length": 50, 
  20.                     "lowercase": true
  21.                 }, 
  22.                 "pinyin_full_filter": {
  23.                     "type": "pinyin", 
  24.                     "keep_first_letter": false, 
  25.                     "keep_separate_first_letter": false, 
  26.                     "keep_full_pinyin": true, 
  27.                     "none_chinese_pinyin_tokenize": true, 
  28.                     "keep_original": false, 
  29.                     "limit_first_letter_length": 50, 
  30.                     "lowercase": true
  31.                 }, 
  32.                 "t2s_convert": {
  33.                     "type": "stconvert", 
  34.                     "delimiter": ",", 
  35.                     "convert_type": "t2s"
  36.                 }
  37.             }, 
  38.             "char_filter": {
  39.                 "charconvert": {
  40.                     "type": "mapping", 
  41.                     "mappings_path": "char_filter_text.txt"
  42.                 }
  43.             }, 
  44.             "tokenizer": {
  45.                 "ik_smart": {
  46.                     "type": "ik", 
  47.                     "use_smart": true
  48.                 }
  49.             }, 
  50.             "analyzer": {
  51.                 "ngramIndexAnalyzer": {
  52.                     "type": "custom", 
  53.                     "tokenizer": "keyword", 
  54.                     "filter": [
  55.                         "edge_ngram_filter", 
  56.                         "lowercase"
  57.                     ], 
  58.                     "char_filter": [
  59.                         "charconvert"
  60.                     ]
  61.                 }, 
  62.                 "ngramSearchAnalyzer": {
  63.                     "type": "custom", 
  64.                     "tokenizer": "keyword", 
  65.                     "filter": [
  66.                         "lowercase"
  67.                     ], 
  68.                     "char_filter": [
  69.                         "charconvert"
  70.                     ]
  71.                 }, 
  72.                 "ikIndexAnalyzer": {
  73.                     "type": "custom", 
  74.                     "tokenizer": "ik", 
  75.                     "char_filter": [
  76.                         "charconvert"
  77.                     ]
  78.                 }, 
  79.                 "ikSearchAnalyzer": {
  80.                     "type": "custom", 
  81.                     "tokenizer": "ik", 
  82.                     "char_filter": [
  83.                         "charconvert"
  84.                     ]
  85.                 }, 
  86.                 "pinyiSimpleIndexAnalyzer": {
  87.                     "tokenizer": "keyword", 
  88.                     "filter": [
  89.                         "pinyin_simple_filter", 
  90.                         "edge_ngram_filter", 
  91.                         "lowercase"
  92.                     ]
  93.                 }, 
  94.                 "pinyiSimpleSearchAnalyzer": {
  95.                     "tokenizer": "keyword", 
  96.                     "filter": [
  97.                         "pinyin_simple_filter", 
  98.                         "lowercase"
  99.                     ]
  100.                 }, 
  101.                 "pinyiFullIndexAnalyzer": {
  102.                     "tokenizer": "keyword", 
  103.                     "filter": [
  104.                         "pinyin_full_filter", 
  105.                         "lowercase"
  106.                     ]
  107.                 }, 
  108.                 "pinyiFullSearchAnalyzer": {
  109.                     "tokenizer": "keyword", 
  110.                     "filter": [
  111.                         "pinyin_full_filter", 
  112.                         "lowercase"
  113.                     ]
  114.                 }
  115.             }
  116.         }
  117.     }
  118. }

案例 2

  1. {
  2.     "settings":{
  3.         "analysis":{
  4.             "analyzer":{
  5.                 "myanalyzer":{
  6.                     "tokenizer":"mytokenizer"
  7.                 }
  8.             },
  9.             "tokenizer":{
  10.                 "mytokenizer":{
  11.                     "type":"ngram",
  12.                     "min_gram":1,
  13.                     "max_gram":2,
  14.                     "token_chars":[
  15.                         "letter",
  16.                         "digit",
  17.                         "whitespace",
  18.                         "punctuation",
  19.                         "symbol"
  20.                     ]
  21.                 }
  22.             }
  23.         }
  24.     }
  25. }

案例 3

  1. {
  2.   "settings": {
  3.     "analysis": {
  4.       "filter": {   # filter过滤器设置
  5.         "edge_ngram_filter":{
  6.           "type":"edge_ngram",
  7.           "min_gram":1,
  8.           "max_gram":50
  9.         },
  10.         "pinyin_simple_filter":{
  11.           "type":"pinyin",
  12.           "keep_first_letter":true,
  13.           "keep_separate_first_letter":false,
  14.           "keep_full_pinyin":false,
  15.           "keep_original":false,
  16.           "limit_first_letter_length":50,
  17.           "lowercase":true
  18.         },
  19.         "pinyin_full_filter":{
  20.           "type":"pinyin",
  21.           "keep_first_letter":false,
  22.           "keep_separate_first_letter":false,
  23.           "keep_full_pinyin":true,
  24.           "none_chinese_pinyin_tokenize":true,
  25.           "keep_original":false,
  26.           "limit_first_letter_length":50,
  27.           "lowercase":true
  28.         }
  29.       },
  30.       "tokenizer": {  # tokenizer 分词器设置
  31.         "ik_max_word":{
  32.           "type":"ik_max_word",
  33.           "use_smart":true
  34.         }
  35.       },
  36.       "analyzer": {  #analyzer 分析器设置
  37.         "ngramIndexAnalyzer":{
  38.           "type":"custom",
  39.           "tokenizer":"keyword",
  40.           "filter":[
  41.             "edge_ngram_filter",
  42.             "lowercase"
  43.             ]
  44.         },
  45.         "ikIndexAnalyzer":{
  46.           "type":"custom",
  47.           "tokenizer":"ik_max_word"
  48.         },
  49.         "pinyiSimpleIndexAnalyzer":{
  50.           "tokenizer":"keyword",
  51.           "filter":[
  52.             "pinyin_simple_filter",
  53.             "edge_ngram_filter",
  54.             "lowercase"
  55.             ]
  56.         },
  57.         "pinyiFullIndexAnalyzer":{
  58.           "tokenizer":"keyword",
  59.           "filter":[
  60.             "pinyin_full_filter",
  61.             "lowercase"
  62.             ]
  63.         }
  64.       }
  65.     }
  66.   }
  67. }

案例 4

这是我们综合的实战的代码

  1. {
  2.   "index": {
  3.     "number_of_replicas": "1",
  4.     "analysis": {
  5.       "char_filter": {
  6.         "ue_char_filter": {
  7.           "type": "mapping",
  8.           "mappings": [
  9.             "- => ,",
  10.             "— => ,"
  11.           ]
  12.         }
  13.       },
  14.       "tokenizer": {
  15.         "ngram_tokenizer": {
  16.           "type": "ngram",
  17.           "min_gram": 2,
  18.           "max_gram": 3,
  19.           "token_chars": [
  20.             "letter",
  21.             "digit"
  22.           ]
  23.         }
  24.       },
  25.       "filter": {
  26.         "my_pinyin": {
  27.           "type": "pinyin",
  28.           "keep_none_chinese": true,
  29.           "keep_none_chinese_together": true,
  30.           "none_chinese_pinyin_tokenize": false,
  31.           "lowercase": true,
  32.           "trim_whitespace": true,
  33.           "keep_first_letter": true,
  34.           "keep_full_pinyin": false,
  35.           "keep_joined_full_pinyin": true,
  36.           "keep_original": true
  37.         }
  38.       },
  39.       "analyzer": {
  40.         "ue_ik_pinyin_analyzer": {
  41.           "type": "custom",
  42.           "char_filter": [
  43.             "html_strip",
  44.             "ue_char_filter"
  45.           ],
  46.           "tokenizer": "ik_max_word",
  47.           "filter": [
  48.             "my_pinyin"
  49.           ]
  50.         },
  51.         "ue-ngram":{
  52.           "type": "custom",
  53.           "char_filter": [
  54.             "html_strip",
  55.             "ue_char_filter"
  56.           ],
  57.           "tokenizer": "ngram_tokenizer"
  58.         }
  59.       }
  60.     }
  61.   }
  62. }