ElasticSearch入门：使用ES来实现模糊查询功能
需求描述方案设计代码设计测试中遇到的问题总结与心得需求描述
本文针对在工作中遇到的需求：通过es来实现模糊查询来进行总结；模糊查询的具体需求是：查询基金/A股/港股等金融数据，要求可以根据字段，拼音首字母，部分拼音全称进行联想查询；需要注意的是，金融数据名称中可能不止包含汉字，还有英文，数字，特殊字符等。
方案设计
常用的es模糊查询出于性能问题，官方建议是慎重使用的，但一般针对于与其他es查询相比，如果和其他搜索工具相比，es的模糊查询性能还是不错的；常见的模糊查询相关函数，例如wildcard，fuzzy，query_string等均不完全适配现有的业务需求，因此从另一个角度思考问题，拟采用更加灵活的分词器来解决多条件模糊查询问题。
ngram分词器与传统的standard分词器或者是ik分词器相比，他的优点是可以分词出特殊字符，因此，在对字段查询时，可以采用ngram分词器；而对拼音全称以及首字母查询时，可以使用keyword与拼音结合的自定义分词。
代码设计
根据上述的方案设计，我们可以在es中定义这样一个索引：
settings:{"analysis":{"analyzer":{"my_ngram_analyzer":{"tokenizer":"my_ngram_tokenizer"},"my_pinyin_analyzer":{"tokenizer":"keyword","filter":"py"}},"tokenizer":{"my_ngram_tokenizer":{"type":"ngram","min_ngram":1,"max_ngram":1}},"filter":{"py":{"type":"pinyin","first_letter":"prefix","keep_separate_first_letter":true,"keep_full_pinyin":true,"keep_joined_full_pinyin":true,"keep_original":true,"limit_first_letter_length":16,"lowercase":true,"remove_duplicated_term":true}}}}mapping:{"properties":{"name":{"type":"text","analyzer":"my_ngram_analyzer"},"fields":{"PY":{"type":"text","analyzer":"my_pinyin_analyzer","term_vector":"with_positions_offsets","boost":10.0}}}}
以text = "恒生电子"为例，它的自定义拼音分词器my_pinyin_analyzer效果如下：
{"tokens": [{"token": "h","start_offset": 0,"end_offset": 4,"type": "word","position": 0},{"token": "heng","start_offset": 0,"end_offset": 4,"type": "word","position": 0},{"token": "恒生电子","start_offset": 0,"end_offset": 4,"type": "word","position": 0},{"token": "hengshengdianzi","start_offset": 0,"end_offset": 4,"type": "word","position": 0},{"token": "hsdz","start_offset": 0,"end_offset": 4,"type": "word","position": 0},{"token": "s","start_offset": 0,"end_offset": 4,"type": "word","position": 1},{"token": "sheng","start_offset": 0,"end_offset": 4,"type": "word","position": 1},{"token": "d","start_offset": 0,"end_offset": 4,"type": "word","position": 2},{"token": "dian","start_offset": 0,"end_offset": 4,"type": "word","position": 2},{"token": "z","start_offset": 0,"end_offset": 4,"type": "word","position": 3},{"token": "zi","start_offset": 0,"end_offset": 4,"type": "word","position": 3}]}
而在对应的代码层面，出于对输入词的关联精确性与词语顺序的考虑，从match , match phrase 以及 match phrase prefix中选择match phrase来进行查询：
// 直接的字段匹配优先级大于拼音匹配BoolQueryBuilder boolQueryBuilderKeyWord = QueryBuilders.boolQuery().matchPhraseQuery("name", imageStr).boost(2.0f)).should(QueryBuilders.matchPhraseQuery("name.PY", imageStr).boost(1.0f));
测试中遇到的问题
应用上述的代码于项目中，经过测试会发现一个问题：输入汉字会查询得出与汉字不相关但缩写一致的数据，例如：关键字录入"恒生电子"，接口返回结果如下：
{"error_code": "0","error_info": "success","data": [{"en_prod_code": "600570.SH","secu_code": "600570","secu_abbr": "恒生电子","type": "A_stock","modification_time": "-04-22T19:47:37.000+00:00","en_abbr": "HSDZ"},{"en_prod_code": "007685.OF","secu_code": "007685.OF","secu_abbr": "华商电子","type": "fund","modification_time": "-04-22T19:41:38.000+00:00","en_abbr": "HSDZ"}]}
通过检查发现，是代码中设置的查询语句有问题，将字段查询与拼音首字母查询隔离即可，即通过中文查询则只查询name字段，通过非中文查询则只查询name.PY，Java代码修改如下：
if (!imageStr.matches("(.*)[\u4e00-\u9fa5](.*)")) {BoolQueryBuilder boolQueryBuilderKeyWord = QueryBuilders.boolQuery().matchPhraseQuery("name.PY", imageStr);} else {BoolQueryBuilder boolQueryBuilderKeyWord = QueryBuilders.boolQuery().matchPhraseQuery("name", imageStr);}
最后，再次查询关键词“恒生电子"，接口返回结果为：
{"error_code": "0","error_info": "success","data": [{"en_prod_code": "600570.SH","secu_code": "600570","secu_abbr": "恒生电子","type": "A_stock","modification_time": "-04-22T19:47:37.000+00:00","en_abbr": "HSDZ"}]}
