范围过滤器不适用于"gt".运算符，但为"lt"工作 [英] Range filter not working for "gt" operator but working for "lt"
问题描述
我正在使用弹性搜索为我的文档建立索引，并希望根据特定属性来过滤文档.这是我的代码:
I am using elastic search to index my documents and wanted to filter documents based on a specific property. Here is my code:
filter.push({
range: {"audience.ethnicity.asian" : { gt: 50 } }
})
它不适用于gt运算符(发回不一致的结果)，但适用于lt运算符
It doesn't work with gt operator (sends back results that are not consistent) but works with lt operator
filter.push({
range: {"audience.ethnicity.asian" : { lt: 50 } }
})
这令人惊讶.我已经仔细检查过，属性值是一个整数.我尝试了许多不同的方法，但似乎无法弄清我想念的是什么. 映射:
This works surprisingly. I have double checked that the property value IS an integer. I have tried many different ways but can't seem to figure out what am I missing. Mapping:
"audience": {
"properties": {
"age": {
"properties": {
"13-17": {
"type": "text",
"fields": {
"keyword": {
"type": "keyword",
"ignore_above": 256
}
}
},
"18-24": {
"type": "text",
"fields": {
"keyword": {
"type": "keyword",
"ignore_above": 256
}
}
},
"25-34": {
"type": "text",
"fields": {
"keyword": {
"type": "keyword",
"ignore_above": 256
}
}
},
"35-44": {
"type": "text",
"fields": {
"keyword": {
"type": "keyword",
"ignore_above": 256
}
}
},
"45-64": {
"type": "text",
"fields": {
"keyword": {
"type": "keyword",
"ignore_above": 256
}
}
},
"eighteen": {
"type": "long"
},
"fortyfive": {
"type": "long"
},
"thirteen": {
"type": "long"
},
"thirtyfive": {
"type": "long"
},
"twentyfive": {
"type": "long"
}
}
},
"ages": {
"properties": {
"13-17": {
"type": "float"
},
"18-24": {
"type": "float"
},
"25-34": {
"type": "float"
},
"35-44": {
"type": "float"
},
"45-64": {
"type": "float"
}
}
},
"ethnicity": {
"properties": {
"African American": {
"type": "float"
},
"Asian": {
"type": "float"
},
"Hispanic": {
"type": "float"
},
"White / Caucasian": {
"type": "float"
},
"african": {
"type": "text",
"fields": {
"keyword": {
"type": "keyword",
"ignore_above": 256
}
}
},
"asian": {
"type": "text",
"fields": {
"keyword": {
"type": "keyword",
"ignore_above": 256
}
}
},
"hispanic": {
"type": "text",
"fields": {
"keyword": {
"type": "keyword",
"ignore_above": 256
}
}
},
"white": {
"type": "text",
"fields": {
"keyword": {
"type": "keyword",
"ignore_above": 256
}
}
}
}
},
"gender": {
"properties": {
"female": {
"type": "text",
"fields": {
"keyword": {
"type": "keyword",
"ignore_above": 256
}
}
},
"male": {
"type": "text",
"fields": {
"keyword": {
"type": "keyword",
"ignore_above": 256
}
}
}
}
},
"genders": {
"properties": {
"FEMALE": {
"type": "float"
},
"MALE": {
"type": "float"
}
}
},
"gendersPerAge": {
"properties": {
"13-17": {
"properties": {
"FEMALE": {
"type": "float"
},
"MALE": {
"type": "float"
}
}
},
"18-24": {
"properties": {
"FEMALE": {
"type": "float"
},
"MALE": {
"type": "float"
}
}
},
"25-34": {
"properties": {
"FEMALE": {
"type": "float"
},
"MALE": {
"type": "float"
}
}
},
"35-44": {
"properties": {
"FEMALE": {
"type": "float"
},
"MALE": {
"type": "float"
}
}
},
"45-64": {
"properties": {
"FEMALE": {
"type": "float"
},
"MALE": {
"type": "float"
}
}
}
}
}
}
} "audience": {
"properties": {
"age": {
"properties": {
"13-17": {
"type": "text",
"fields": {
"keyword": {
"type": "keyword",
"ignore_above": 256
}
}
},
"18-24": {
"type": "text",
"fields": {
"keyword": {
"type": "keyword",
"ignore_above": 256
}
}
},
"25-34": {
"type": "text",
"fields": {
"keyword": {
"type": "keyword",
"ignore_above": 256
}
}
},
"35-44": {
"type": "text",
"fields": {
"keyword": {
"type": "keyword",
"ignore_above": 256
}
}
},
"45-64": {
"type": "text",
"fields": {
"keyword": {
"type": "keyword",
"ignore_above": 256
}
}
},
"eighteen": {
"type": "long"
},
"fortyfive": {
"type": "long"
},
"thirteen": {
"type": "long"
},
"thirtyfive": {
"type": "long"
},
"twentyfive": {
"type": "long"
}
}
},
"ages": {
"properties": {
"13-17": {
"type": "float"
},
"18-24": {
"type": "float"
},
"25-34": {
"type": "float"
},
"35-44": {
"type": "float"
},
"45-64": {
"type": "float"
}
}
},
"ethnicity": {
"properties": {
"African American": {
"type": "float"
},
"Asian": {
"type": "float"
},
"Hispanic": {
"type": "float"
},
"White / Caucasian": {
"type": "float"
},
"african": {
"type": "text",
"fields": {
"keyword": {
"type": "keyword",
"ignore_above": 256
}
}
},
"asian": {
"type": "text",
"fields": {
"keyword": {
"type": "keyword",
"ignore_above": 256
}
}
},
"hispanic": {
"type": "text",
"fields": {
"keyword": {
"type": "keyword",
"ignore_above": 256
}
}
},
"white": {
"type": "text",
"fields": {
"keyword": {
"type": "keyword",
"ignore_above": 256
}
}
}
}
},
"gender": {
"properties": {
"female": {
"type": "text",
"fields": {
"keyword": {
"type": "keyword",
"ignore_above": 256
}
}
},
"male": {
"type": "text",
"fields": {
"keyword": {
"type": "keyword",
"ignore_above": 256
}
}
}
}
},
"genders": {
"properties": {
"FEMALE": {
"type": "float"
},
"MALE": {
"type": "float"
}
}
},
"gendersPerAge": {
"properties": {
"13-17": {
"properties": {
"FEMALE": {
"type": "float"
},
"MALE": {
"type": "float"
}
}
},
"18-24": {
"properties": {
"FEMALE": {
"type": "float"
},
"MALE": {
"type": "float"
}
}
},
"25-34": {
"properties": {
"FEMALE": {
"type": "float"
},
"MALE": {
"type": "float"
}
}
},
"35-44": {
"properties": {
"FEMALE": {
"type": "float"
},
"MALE": {
"type": "float"
}
}
},
"45-64": {
"properties": {
"FEMALE": {
"type": "float"
},
"MALE": {
"type": "float"
}
}
}
}
}
}
}
推荐答案
在您的映射中，audience.ethnicity.asian字段为文本类型，因此范围查询正在执行词法范围比较，而不是数字范围比较.
In your mapping, the audience.ethnicity.asian field is of type text, hence the range query is doing a lexical range comparison instead of a numerical range comparison.
从数字上看123> 50，但从词汇上看123< 50，因为ASCII字符1在5之前，这正是您在这里遇到的问题.
Numerically, 123 > 50, but lexically 123 < 50 because ASCII character 1 comes before 5, that's exactly the issue you're suffering from here.
我不确定是否可以使用audience.ethnicity.Asian字段，该字段是浮点数.如果不是，则需要将audience.ethnicity.asian的映射更改为数字类型(integer，float等)，然后重新索引数据.
I'm not sure if you can use the audience.ethnicity.Asian field, which is a float. If not, you'll need to change the mapping of audience.ethnicity.asian to a numeric type (integer, float, etc) and reindex your data.
这篇关于范围过滤器不适用于"gt".运算符，但为"lt"工作的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！
