elasticsearch 聚合显示占比代码

以下是一个使用 Elasticsearch 聚合显示占比的示例代码：

from elasticsearch import Elasticsearch

es = Elasticsearch()

# 构建聚合查询
aggs = {
    "group_by_state": {
        "terms": {"field": "state.keyword"},
        "aggs": {
            "group_by_city": {
                "terms": {"field": "city.keyword"}
            }
        }
    }
}

# 执行查询
result = es.search(
    index="your_index",
    body={"size": 0, "aggs": aggs}
)

# 解析结果并输出占比
total = result["hits"]["total"]["value"]
for state in result["aggregations"]["group_by_state"]["buckets"]:
    state_name = state["key"]
    state_count = state["doc_count"]
    state_percent = state_count / total * 100
    print(f"{state_name}: {state_percent:.2f}%")
    for city in state["group_by_city"]["buckets"]:
        city_name = city["key"]
        city_count = city["doc_count"]
        city_percent = city_count / total * 100
        print(f"- {city_name}: {city_percent:.2f}%")

这里的查询是以 state 和 city 两个字段分组，然后计算每个分组的文档数占总文档数的比例。输出的结果类似于：

California: 50.00%
- Los Angeles: 20.00%
- San Francisco: 30.00%
New York: 30.00%
- New York City: 20.00%
- Buffalo: 10.00%
Texas: 20.00%
- Austin: 10.00%
- Houston: 10.00%