如何在mangoose (物化路径类别)中获取类别的父项和子项的数组？

Question

大家好，我想在nodejs应用程序中实现mongodb的物化路径类别层次结构，如果我有这两个文档，我想像树一样获得它们的数组，这一点我将会解释

集合：

const categoriesCollection = [
  {
    "_id": ObjectId("54fd7392742abeef6186a68e")
    , "name": "electronics"
    , "parent": "/"
    , "category": "/electronics"
  },


  {
    "_id": ObjectId("54fd7392742abeef6186a68e")
    , "name": "embedded"
    , "parent": "/electronics"
    , "category": "/electronics/embedded"
  },
  {
    "_id": ObjectId("54fd7392742abeef6186a68e")
    , "name": "controllers"
    , "parent": "/electronics"
    , "category": "/electronics/controllers"
  },

  {
    "_id": ObjectId("54fd7392742abeef6186a68e")
    , "name": "cases"
    , "parent": "/electronics"
    , "category": "/electronics/cases"
  },

  {
    "_id": ObjectId("54fd7392742abeef6186a68e")
    , "name": "big"
    , "parent": "/electronics/cases"
    , "category": "/electronics/cases/big"
  },
  {
    "_id": ObjectId("54fd7392742abeef6186a68e")
    , "name": "small"
    , "parent": "/electronics/cases"
    , "category": "/electronics/cases/small"
  },


]

const ProductsCollection = [
  {
    "_id": ObjectId("54fd7392742abeef6186a68e")
    , "name": "product1"
    , "cost": 125
    , "currency": "USD"
    , "categories": ["/electronics/embedded"]
  },
  {
    "_id": ObjectId("54fd7392742abeef6186a68e")
    , "name": "product2"
    , "cost": 134
    , "currency": "USD"
    , "categories": ["/electronics/controllers"]
  },
  {
    "_id": ObjectId("54fd7392742abeef6186a68e")
    , "name": "product3"
    , "cost": 133
    , "currency": "USD"
    , "categories": ["/electronics/cases/big"]
  },

  {
    "_id": ObjectId("54fd7392742abeef6186a68e")
    , "name": "product4"
    , "cost": 188
    , "currency": "USD"
    , "categories": ["/electronics/cases/small", "/electronics/cases"]
  },

]

我希望在树状结构中拥有类似于父项、子项和产品的输出数组：

const allCategories = [
  {

    name: "electronics",
    categories: [
      {
        name: "embedded",
        categories: [],
        products: [{ name: "product1" }]
      },
      {
        name: "controllers",
        categories: [],
        products: [{ name: "product2" }]
      },
      {
        name: "cases",
        categories: [
          {
            name: "big",
            categories: [],
            products: [{ name: "product3" }]
          },
          {
            name: "small",
            categories: [],
            products: [{ name: "product4" }]
          }
        ],
        products: [{ name: "product1" }]
      },

    ]

    ,
    products: [{ "name": "product4" }]

  }
]

我如何通过mongoose查询在这个数组中获得这个数组？

Answer 1

在MongoDB中没有任何直接的方法来处理这种情况，您可以尝试$groupLookup和自定义逻辑，

只有深度的记录的

• $match过滤器为/
• $graphLookup，以获取categories数组中的子记录和深度数，并允许不删除具有products集合的空parent，并连接产品或categories
• $addFields，以在products
• $sort中仅选择depth level字段中的depth level字段，并在降序order
• $group by parent字段中重建depthField categories数组categories

db.category.aggregate([
  { $match: { parent: "/" } },
  {
    $graphLookup: {
      from: "category",
      startWith: "$category",
      connectFromField: "category",
      connectToField: "parent",
      depthField: "level",
      as: "categories"
    }
  },
  {
    $unwind: {
      path: "$categories",
      preserveNullAndEmptyArrays: true
    }
  },
  {
    $lookup: {
      from: "products",
      localField: "categories.category",
      foreignField: "categories",
      as: "categories.products"
    }
  },
  {
    $addFields: {
      "categories.products": {
        $map: {
          input: "$categories.products",
          in: { name: "$$this.name" }
        }
      }
    }
  },
  { $sort: { "categories.level": -1 } },
  {
    $group: {
      _id: "$category",
      parent: { $first: "$parent" },
      name: { $first: "$name" },
      categories: { $push: "$categories" }
    }
  },

• $addFields现在找到嵌套的级别类别并分配给它的级别，
  • $reduce to iterate categories
    • $addFields array.
    • initialize categories level默认值为-1，presentChild为[]，对于用于初始化字段的条件，purpose
    • $let为[]：
      • prev as per condition如果两个level相等，则返回prevChild；否则，返回
        • prev as per condition如果两个prevChild相等，则返回presentChild；否则返回presentChild

- `in` to return `level` field and `prevChild` field from initialized fields 
    - `presentChild` `$filter` `categories` from `prev` array and return, merge current objects with `categories` array using `$mergeObjects` and concat with `current` array of let using `$concatArrays`

• $addFields仅返回presentChild数组，因为我们只需要已处理的数组

  {
    $addFields: {
      categories: {
        $reduce: {
          input: "$categories",
          initialValue: { level: -1, presentChild: [], prevChild: [] },
          in: {
            $let: {
              vars: {
                prev: {
                  $cond: [
                    { $eq: ["$$value.level", "$$this.level"] },
                    "$$value.prevChild",
                    "$$value.presentChild"
                  ]
                },
                current: {
                  $cond: [{ $eq: ["$$value.level", "$$this.level"] }, "$$value.presentChild", []]
                }
              },
              in: {
                level: "$$this.level",
                prevChild: "$$prev",
                presentChild: {
                  $concatArrays: [
                    "$$current",
                    [
                      {
                        $mergeObjects: [
                          "$$this",
                          {
                            categories: {
                              $filter: {
                                input: "$$prev",
                                as: "e",
                                cond: { $eq: ["$$e.parent", "$$this.category"] }
                              }
                            }
                          }
                        ]
                      }
                    ]
                  ]
                }
              }
            }
          }
        }
      }
    }
  },
  { $addFields: { categories: "$categories.presentChild" } }
])

Playground

备注：这将导致海量数据的性能问题！