MongoDB 聚合错误 "each item in the pipeline must be a document"

Question

我有一个如下所示的 MongoDB 聚合:

[
    {
        "$match": [
            {"Created": {"$gte": ISODate("2014-01-10T00:00:00Z")}}
        ]
    },
    {
        "$group":
        {
            "_id": [
                {"year": {"$year": "Created"}},
                {"month": {"$month": "Created"}},
                {"day": {"$dayOfMonth": "Created"}}
            ],
            "count": {"$sum": 1}
        }
    }
]

当我在 MongoVUE 中运行此查询时，它返回以下错误:

Incorrect syntax in pipelineEach item in the pipeline must be a documentType: System.ExceptionStack:    at MangoUI.ComAggregate.kRemove_Click(Object sender, EventArgs e)

I'm completely stumped, and from my Googling so is the rest of the Intenet-at-large. The query worked correctly before I added the date created $match operator. It's possible that the error belongs to MongoVUE and not MongoDB itself.

The relevant portion of the schema looks like this:

{
  "_id" : new BinData(3, "m13wFpp9gUi09cRCuG43aw=="),
  "Created" : ISODate("2013-12-19T01:00:20.972Z")
}

谁能帮我找出这个错误的原因？恐怕我完全被难住了。

---

我不允许回答自己的问题(因为如果您的 StackOverflow 声誉太低，显然在 8 小时内完全不可能自己找到答案？嗯？)所以我在下面发布了答案:

好的，我找到了答案，据我所知，这确实是 MongoVUE 中的一个错误(或至少是一个遗漏和误导性错误消息)。以这种方式编写相同的查询(带有 orid 的修改)是可行的:

{
  "$match": {
    "Created": {
      "$gte": ISODate("2014-01-10T00:00:00Z")
    }
  }
},
{
  "$group": {
    "_id": {
      "year": {
        "$year": "$Created"
      },
      "month": {
        "$month": "$Created"
      },
      "day": {
        "$dayOfMonth": "$Created"
      }
    },
    "count": {
      "$sum": 1
    }
  }
}

(请注意，我真正所做的唯一更改是删除了它周围的数组标记。)

此外，当您在 MongoVUE 中的数组中放置具有多个操作的聚合时，任何都会出现该错误。如果您不使用数组，则操作可以正常工作。如果你添加它，你会突然得到“管道中的每个项目都必须是一个文档。”

Answer 1

pipeline = [{
  "$match": {
    "Created": {
      "$gte": ISODate("2014-01-10T00:00:00Z")
    }
  }
},
{
  "$group": {
    "_id": {
      "year": {
        "$year": "$Created"
      },
      "month": {
        "$month": "$Created"
      },
      "day": {
        "$dayOfMonth": "$Created"
      }
    },
    "count": {
      "$sum": 1
    }
  }
}]

Document.objects.aggregate(*pipeline)