我试图在一个flattened字段中的几个字段上执行must查询，该字段具有一个对象数组。我想在数组的一个元素上匹配所有条件，但它匹配数组中的任何元素。
我有一组文档，如下所示：
[
{
"_source": {
"metadata": {
"name": "banana",
"seasonalities": [
{
"date": "2023-01-01",
"country": "AA"
},
{
"date": "2023-01-22",
"country": "BB"
}
]
}
},
{
"_source": {
"metadata": {
"name": "potato",
"seasonalities": [
{
"date": "2023-01-01",
"country": "AA"
},
{
"date": "2023-01-01",
"country": "BB"
},
{
"date": "2023-01-01",
"country": "CC"
},
{
"date": "2023-01-22",
"country": "DD"
},
{
"date": "2023-01-23",
"country": "EE"
}
]
}
}
]下面的查询工作并检索单个文档：
{
"query": {
"bool": {
"filter": [
{
"bool": {
"must": [
{
"term": {
"metadata.seasonalities.date": {"value":"2023-01-22"}
}
},
{
"term": {
"metadata.name": {"value":"banana"}
}
}
]
}
}
]
}
}
}这个更简单的查询也可以工作：
{
"query": {
"bool": {
"must": [
{
"term": {
"metadata.seasonalities.date": "2023-01-22"
}
},
{
"term": {
"metadata.name": "banana"
}
}
]
}
}
}这些查询都像预期的那样返回一个值，但当我尝试这样做时：
{
"query": {
"bool": {
"must": [
{
"term": {
"metadata.seasonalities.date": "2023-01-22"
}
},
{
"term": {
"metadata.seasonalities.country": "BB"
}
}
]
}
}
}我得到了两个文件，而不仅仅是第一个。
下面是索引的Map：
"mappings": {
"properties": {
"metadata": {
"type": "flattened"
}
}
},当我用explain在我的实际索引中运行搜索时，我得到了这个（我还有一些字段和文档）：
{
"took": 7,
"timed_out": false,
"_shards": {
"total": 5,
"successful": 5,
"skipped": 0,
"failed": 0
},
"hits": {
"total": {
"value": 943,
"relation": "eq"
},
...
{
"_source": {
"metadata": {
"name": "XXX",
"seasonalities": [
{
"date": "2023-01-22",
"country": "MM",
},
{
"date": "2023-03-01",
"country": "BB",
},
...
"_explanation": {
"value": 9.256371,
"description": "sum of:",
"details": [
{
"value": 5.61615,
"description": "weight(metadata._keyed:seasonalities.date\u00002023-01-22 in 1) [PerFieldSimilarity], result of:",
"details": [
{
"value": 5.61615,
"description": "score(freq=1.0), computed as boost * idf * tf from:",
"details": [
{
"value": 2.2,
"description": "boost",
"details": []
},
{
"value": 3.3873937,
"description": "idf, computed as log(1 + (N - n + 0.5) / (n + 0.5)) from:",
"details": [
{
"value": 177,
"description": "n, number of documents containing term",
"details": []
},
{
"value": 5251,
"description": "N, total number of documents with field",
"details": []
}
]
},
{
"value": 0.75361645,
"description": "tf, computed as freq / (freq + k1 * (1 - b + b * dl / avgdl)) from:",
"details": [
{
"value": 1,
"description": "freq, occurrences of term within document",
"details": []
},
{
"value": 1.2,
"description": "k1, term saturation parameter",
"details": []
},
{
"value": 0.75,
"description": "b, length normalization parameter",
"details": []
},
{
"value": 1,
"description": "dl, length of field",
"details": []
},
{
"value": 33.413826,
"description": "avgdl, average length of field",
"details": []
}
]
}
]
}
]
},
{
"value": 3.640221,
"description": "weight(metadata._keyed:seasonalities.country\u0000BB in 1) [PerFieldSimilarity], result of:",
"details": [
{
"value": 3.640221,
"description": "score(freq=1.0), computed as boost * idf * tf from:",
"details": [
{
"value": 2.2,
"description": "boost",
"details": []
},
{
"value": 2.1956074,
"description": "idf, computed as log(1 + (N - n + 0.5) / (n + 0.5)) from:",
"details": [
{
"value": 584,
"description": "n, number of documents containing term",
"details": []
},
{
"value": 5251,
"description": "N, total number of documents with field",
"details": []
}
]
},
{
"value": 0.75361645,
"description": "tf, computed as freq / (freq + k1 * (1 - b + b * dl / avgdl)) from:",
"details": [
{
"value": 1,
"description": "freq, occurrences of term within document",
"details": []
},
{
"value": 1.2,
"description": "k1, term saturation parameter",
"details": []
},
{
"value": 0.75,
"description": "b, length normalization parameter",
"details": []
},
{
"value": 1,
"description": "dl, length of field",
"details": []
},
{
"value": 33.413826,
"description": "avgdl, average length of field",
"details": []
}
]
}
]
}
]
}
]
}
},我得到了943个结果，但看看解释它确实说：2023年1月22日：177 n, number of documents containing term，然后对于BB：584 n, number of documents containing term在交叉点处，我希望只有一个或两个文档，我如何才能只得到这些文档？我尝试在查询中将term更改为match，但没有效果。我在ES 7.13.3。
1条答案
按热度按时间k2fxgqgv1#
当你需要匹配一个对象字段的两个字段时，你需要使用
nested字段类型和查询。您的Map应该如下所示：
然后，第一个查询将像预期的那样仅返回第一个文档：
当你对季节性对象有两个约束时，你可以这样做，以便也只返回第一个文档：