elasticsearch - 嵌套字段值的聚集Elasticsearch索引

Question

我需要使用用户和文章对网站进行建模，使每个用户可以与任何文章多次互动(阅读，打开e.t.c)。我想通过遵循嵌套映射在一个Elasticsearch索引中对该数据建模:

{
    "mappings": {
        "user": {
            "properties": {
                "user_id": {"type": "string"},
                "interactions": {
                    "type": "nested",
                    "properties": {
                        "article_id": {"type": "string"},
                        "interact_date": {"type": "date"}
                    }
                }
            }
        }
    }
}

索引文件示例:

{
    "user_id": 20,
    "interactions": [
        {"article_id": "111", "interact_date": "2015-01-01"},
        {"article_id": "111", "interact_date": "2015-01-02"},
        {"article_id": "222", "interact_date": "2015-01-01"}
     ]
}

我需要对数据进行以下汇总:

每天通过嵌套聚合完成的互动总数:

GET /_search
{
    "size": 0,
    "aggs": {
        "by_date": {
            "nested": {
                "path": "interactions"
            },
            "aggs": {
                "m_date": {"terms": {"field": "interactions.interact_date"}}
            }
        }
    }
}

每天的唯一身份用户互动数。如果特定用户在同一日期范围内与几篇文章进行了互动，则该用户应仅被计数一次。
在postgres中，它是简单的查询:
用于具有3列的表格[user_id，article_id，interact_date]

SELECT dt, count(uid)
FROM (SELECT interact_date::TIMESTAMP::DATE dt, user_id uid FROM interactions
        GROUP BY interact_date::TIMESTAMP::DATE, user_id) by_date
GROUP BY dt;

如何在Elasticsearch索引中做同样的事情？

如何在不重新索引整个文档的情况下通过_update添加交互？

如何按特定文章过滤用户-仅当用户与指定文章之一互动时，才按日期对用户计数一次？

谢谢

Answer 1

Number of unique users interactions per day.

{
  "size": 0,
  "aggs": {
    "nested_agg": {
      "nested": {
        "path": "interactions"
      },
      "aggs": {
        "per_day": {
          "date_histogram": {
            "field": "interactions.interact_date",
            "interval": "day",
            "min_doc_count": 1
          },
          "aggs": {
            "users_count": {
              "reverse_nested": {},
              "aggs": {
                "uniques": {
                  "cardinality": {
                    "field": "user_id"
                  }
                }
              }
            }
          }
        }
      }
    }
  }
}

How to add interactions by _update without re-indexing whole document?

这不可能。这就是 definition of a nested object:要更新，添加或删除嵌套对象，我们必须为整个文档重新编制索引。

How to filter users by specific articles - count user once in aggregation by date only if he interacted with one of specified articles?

{
  "size": 0,
  "query": {
    "nested": {
      "path": "interactions",
      "query": {
        "term": {
          "interactions.article_id": {
            "value": "222"
          }
        }
      }
    }
  },
  "aggs": {
    "nested_agg": {
      "nested": {
        "path": "interactions"
      },
      "aggs": {
        "filtered": {
          "filter": {
            "term": {
              "interactions.article_id": {
                "value": "222"
              }
            }
          },
          "aggs": {
            "per_day": {
              "date_histogram": {
                "field": "interactions.interact_date",
                "interval": "day",
                "min_doc_count": 1
              },
              "aggs": {
                "users_count": {
                  "reverse_nested": {},
                  "aggs": {
                    "uniques": {
                      "cardinality": {
                        "field": "user_id"
                      }
                    }
                  }
                }
              }
            }
          }
        }
      }
    }
  }
}