[go] go implements row column conversion of sets

One 、 Scene one

1、 Fixed column header row to column

Fixed column header row to column ： Explain it. , Converted column headers （ namely ： Yesterday's date 、 Today's date ） For fixed , We don't have to deal with it

demand ： user A There are 3 Accounts （account） Then each account generates a certain fee every day , Then we extract data and need to show Here's the picture ：

Here we will A Different users account A comparison is made between today's and yesterday's expenses
In this scenario , We need to perform an operation of transferring data from row to column

2、 Row to column method （map Array ）

If the data type we output is ：res := []map[string]interface{}{} namely ：map Array

Row to column conversion requires a unique reference column , As shown in the figure above account For reference columns
The input parameters are ： Output array 、map Of key
The return results are ：key Whether there is 、 There is map、 There is map The index of

// MapKeyExist  Based on the incoming map In the list , Determine whether there is a certain map Contains the key
func MapKeyExist(list []map[string]interface{
    }, key string) (bool, map[string]interface{
    }, int) {
    
	//fmt.Println(len(list))
	if len(list) > 0 {
    
		for i, m := range list {
    
			//fmt.Println(m)
			if _, ok := m[key]; ok {
    
				return true, m, i
			}
		}
	}
	return false, nil, 0
}

3、 How to use

Here is a pseudocode , Briefly introduce the idea of row column conversion
Please read each line carefully , If there are questions , Leave a comment in the comments section

res := []map[string]interface{
    }{
    }
	if len( Fetched from database list) > 0 {
    
		for _, item := range  Fetched from database list {
    

			//  Determine whether the unique value exists in the array （ Use here account）
			flag, data, index := maputil.MapKeyExist(res, item.Account)
			//  If there is 
			if flag == true {
    
				//  Get the currently existing object from the array 
				resTemp := res[index]
				//  Add the required fields below 
				....
				
			} else {
    
				//  Initialize a  map
				temp := make(map[string]interface{
    })
				//  Set a unique column here , As a reference for later consolidation 
				temp[item.Account] = item.Account
				//  The required columns can be completed during insertion 
				temp["xxx"] = 0 // int  Give default 
				temp["kkk"] = ""// string  Give default 
				//  Add to  map Array 
				res = append(res, temp)
			}

		}
	}

Two 、 Scene two

1、 Custom column header row to column

Custom column header row to column ： Explain it. , That is, we will see the left side of the figure below account number The row is converted to the column header on the right of the following figure

demand ： user 103 There are 2 Accounts , Then each account will incur a certain fee every day , Then we extract data to show the specific cost of each account according to the service Here's the picture ：

2、 Row to column method

The way of thinking is different from that of scenario 1 above
If we want to show the following data structure （ namely ： Do not show the user column above ）

3、 How to use

Analysis of the table
We need to put the header Th The above needs to dynamically obtain data
then table Of Tbody in

3.1 assemble th Ideas

First of all, we should make clear what to render th It's an array , Here we use the data structure of the array
Definition th := []string{}
A piece of pseudo code is as follows ：

th := []string{
    }
	if len(res) > 0 {
    
		for _, temp1 := range res {
    
			index := arrayUtil.In(temp1.Account, th)
			if index == false {
    
				th = append(th, temp1.Account)
			}
		}
	}
}

3.2 obtain Tbody Ideas

Clear the header , We will face that every line is a business name （ Name ：service Corresponding data ）, And the data of each business should correspond to account（ dynamic th Header ）, Finally, we choose map in string- Array data structure
Such as ：
{
“service1”:[1,2,3,4,5]
}
Definition tbody := map[string][]interface{}{}

if len(res) > 0 {
    
	for _, temp := range res {
    
		tbody[temp.Service] = append(tbody[temp.Service], totalNum)
	}
}

Best practices ：

1. above append I have tested , If tbody This map There is no such string Of key Will be automatically created and assigned
2. If tbody This map There is this in key, I'll tell you again key Add elements to the corresponding array
3. So you can use it safely

3.3 summary

above 2 There is a res Set
Why don't we consider th and tbody The order of , Because the same res, We don't have to think about
If there is no need for other special processing data , We can 3.1 and 3.2 Put them together