Mongodb graphLookup

Question

自从 MongoDB 最近引入了 graphLookup，我试图找出它是否可以保存一个简单的社交关系图。我目前使用 neo4j 只是为了这个目的。

我将 graphLookup 理解为递归搜索，它只是通过每个文档的“connectFromField”进行更深入的搜索。

虽然我能够做基本的事情，但我想为每个关系提供更多属性。例如这里的第一个例子：（员工和报告层次结构）

https://docs.mongodb.com/manual/reference/operator/aggregation/graphLookup/

{ "_id" : 2, "name" : "Eliot", "reportsTo" : "Dev" }

如果我需要为“reportsTo”值添加开始日期，如下所示：

{ "_id" : 2, "name" : "Eliot", "reportsTo" : {"name": "Dev", "from": "date"  } }

恐怕不支持。

我想知道是否有人以这种方式使用过 MongoDB。

Answer 1

假设我们插入了以下文档：

> db.employees.insertMany([
... { "_id" : 1, "name" : "Dev" },
... { "_id" : 2, "name" : "Eliot", "reportsTo" : { name: "Dev", "from": ISODate("2016-01-01T00:00:00.000Z") } },
... { "_id" : 3, "name" : "Ron", "reportsTo" : { name: "Eliot", "from": ISODate("2016-01-01T00:00:00.000Z") } },
... { "_id" : 4, "name" : "Andrew", "reportsTo" : { name: "Eliot", "from": ISODate("2016-01-01T00:00:00.000Z") } },
... { "_id" : 5, "name" : "Asya", "reportsTo" : { name: "Ron", "from": ISODate("2016-01-01T00:00:00.000Z") } },
... { "_id" : 6, "name" : "Dan", "reportsTo" : { name: "Andrew", "from": ISODate("2016-01-01T00:00:00.000Z") } },
... ]);
{ "acknowledged" : true, "insertedIds" : [ 1, 2, 3, 4, 5, 6 ] }

然后，我们可以使用 . 使用以下聚合查询从嵌入文档中获取字段：

db.employees.aggregate([
{
   $graphLookup: {
      from: "employees",
      startWith: "Eliot",
      connectFromField: "reportsTo.name",
      connectToField: "name",
      as: "reportingHierarchy"
   }
}
])

这将返回以下结果：

{
        "_id" : 1,
        "name" : "Dev",
        "reportingHierarchy" : [
                {
                        "_id" : 1,
                        "name" : "Dev"
                },
                {
                        "_id" : 2,
                        "name" : "Eliot",
                        "reportsTo" : {
                                "name" : "Dev",
                                "from" : ISODate("2016-01-01T00:00:00Z")
                        }
                }
        ]
}
{
        "_id" : 2,
        "name" : "Eliot",
        "reportsTo" : {
                "name" : "Dev",
                "from" : ISODate("2016-01-01T00:00:00Z")
        },
        "reportingHierarchy" : [
                {
                        "_id" : 1,
                        "name" : "Dev"
                },
                {
                        "_id" : 2,
                        "name" : "Eliot",
                        "reportsTo" : {
                                "name" : "Dev",
                                "from" : ISODate("2016-01-01T00:00:00Z")
                        }
                }
        ]
}
{
        "_id" : 3,
        "name" : "Ron",
        "reportsTo" : {
                "name" : "Eliot",
                "from" : ISODate("2016-01-01T00:00:00Z")
        },
        "reportingHierarchy" : [
                {
                        "_id" : 1,
                        "name" : "Dev"
                },
                {
                        "_id" : 2,
                        "name" : "Eliot",
                        "reportsTo" : {
                                "name" : "Dev",
                                "from" : ISODate("2016-01-01T00:00:00Z")
                        }
                }
        ]
}
{
        "_id" : 4,
        "name" : "Andrew",
        "reportsTo" : {
                "name" : "Eliot",
                "from" : ISODate("2016-01-01T00:00:00Z")
        },
        "reportingHierarchy" : [
                {
                        "_id" : 1,
                        "name" : "Dev"
                },
                {
                        "_id" : 2,
                        "name" : "Eliot",
                        "reportsTo" : {
                                "name" : "Dev",
                                "from" : ISODate("2016-01-01T00:00:00Z")
                        }
                }
        ]
}
{
        "_id" : 5,
        "name" : "Asya",
        "reportsTo" : {
                "name" : "Ron",
                "from" : ISODate("2016-01-01T00:00:00Z")
        },
        "reportingHierarchy" : [
                {
                        "_id" : 1,
                        "name" : "Dev"
                },
                {
                        "_id" : 2,
                        "name" : "Eliot",
                        "reportsTo" : {
                                "name" : "Dev",
                                "from" : ISODate("2016-01-01T00:00:00Z")
                        }
                }
        ]
}
{
        "_id" : 6,
        "name" : "Dan",
        "reportsTo" : {
                "name" : "Andrew",
                "from" : ISODate("2016-01-01T00:00:00Z")
        },
        "reportingHierarchy" : [
                {
                        "_id" : 1,
                        "name" : "Dev"
                },
                {
                        "_id" : 2,
                        "name" : "Eliot",
                        "reportsTo" : {
                                "name" : "Dev",
                                "from" : ISODate("2016-01-01T00:00:00Z")
                        }
                }
        ]
}

然后我们还可以使用聚合管道的其余部分进行任何其他操作：

db.employees.aggregate([

   { $match: { "reportsTo.from": { $gt: ISODate("2016-01-01T00:00:00Z") } } },
   { $graphLookup: { ... } },
   { $project: { ... }
]);

有关流水线阶段，请参阅 https://docs.mongodb.com/v3.2/reference/operator/aggregation-pipeline/。