MongoDB Advanced Chapter Dynamic Field Design

This article mainly introduces to you the relevant information of MongoDB advanced dynamic field design. The introduction in the article is very detailed and has certain reference and learning value for everyone. Friends who need it can follow the small Let’s learn together.

This article mainly introduces relevant information about MongoDB dynamic field design. It is shared for everyone's reference and study. Friends who need it can take a look at the detailed introduction.

Suitable for readers

• MongoDB developers
  

Basic requirements

Product: "We want to add a great feature to the existing form that allows users to add the fields they want"

Technical goal version 1

Storage dynamic form data (new fields do not need to modify the Schema)

Let’s talk about it first Let’s talk about what kinds of indexes are supported by MongoDB

Common field index

// 假如我们的文档长这样
{
 "name": "MongoDB",
 "age": 5
}

// 对age字段建立索引
{
 "age": 1
}

Embedded document index

// 假如我们的文档长成了Object
{
 "person": {
  "age": 2,
  "name": "MongoDB"
 }
}

//对person.age字段建立索引
{
 "person.age": 1
}

ArrayDocument Index

// 假如我们的文档长成了数组
{
 "persons": [
  { "name": "MongoDB", age: 5},
  { "name": "MySQL", age: 20}
 ]
}
//对persons.age字段建立索引
{
 "persons.age": 1
}

It seems that none of the above can achieve the function of dynamically adding fields

Programmer The following conversation took place between A and programmer S:

• Programmer A: "Then we need to add another collection to store dynamic content"

• programmerS: "But MongoDB's support for related queries is very weak, and it can't even be related to sorting. If future products say that they will add sorting and filtering functions, we will be confused☹️, ugh~ If we had known this, we would not have used MongoDB. "
  

Re-examine the requirements

• Store dynamic form data

• Need to support filtering and sorting

Technical goal version2

Add fields while also being able to index

Solution

• Use arrays to store dynamic fields

• Add description collection To record the user's form configuration
  

The storage structure is as follows:

//描述collection 
{
 "_id":"描述id",
 "type":"类型",
 "text": "订单名称",
 "default": "Default Name",
}

// 原本的表单增加字段form用来存储动态数据
{
 "_id": "",
 "name": "一个好名字",
 "form":[
 { "_id":"描述_id", "value": 10},
 { "_id":"描述_id", "value": "我的好伙伴"},
 ]
}

Attention!!! When the user When adding n fields, describe the collection and add n documents at the same time

How to query, sort and filter

// 比如用户增加了2个字段
// 现在要对字段1进行排序
db.items.find().sort({"form.0.value":1})
// 对字段2进行筛选
db.items.find({"form.1.value":"我的好伙伴"})

As can be seen from the above example, even if the user does not fill in the field value, but We still need to store null values ​​for it to ensure that the nth field under all our Document forms is the same Control, so that we can filter and sort the fields. , and indexable

Continue to drill down

products: "We need to allow users to add drop-down boxes and multi-select boxes, as well as Filter sorting"

Programmer: "Fxxx"

So how should such data be stored?

Solution As follows:

Our value is stored in the binary format of 1,2,4,8...

such as

If the user selects the first item of the radio button, 1 is stored, the second item is stored with 2, and the third item is stored with 4

The user selects the first item + third item of the multi-select box: Then save 5, (1+4)

MongoDB provides us with a powerful Aggregate function, which includes the Bitwise Query Operators function, including $bitsAllSet, $bitsAnySet, $bitsAllClear, $bitsAnyClear

db.items.aggregate([
 {
  "$match": {
  "$elemMatch": {
   "描述id": "id",
   "value": { $bitsAnySet: [ 1, 5 ] }
  }
 }}
])

The above completes the various requirements for using MongoDB dynamic field design

The above is the detailed content of MongoDB Advanced Chapter Dynamic Field Design. For more information, please follow other related articles on the PHP Chinese website!