Merging of map and struct golang - go

I have 2 items, collections and accounts represented by 2 structs that I would like to merge into a single response.
collections, accounts, err := h.Service.Many(ctx, params)
The collection struct is defined as follows:
type Collection struct {
ID int64 `json:"id"`
Name *string `json:"name"`
Description *string `json:"description"`
Total *int64 `json:"total"`
}
And accounts is defined as a map as such accounts := make(map[int64][]string) and the data looks like this map[1:[19565 21423] 7:[]]
What I would like to do is merge these 2 something like the following:
// merge into single struct
type CollectionWithAccounts struct {
Collections []*collection.Collection
AccountIDs []string
}
// initialize struct
collectionsWithAccounts := make([]CollectionWithAccounts, 0)
// merge strucst in loop
for _, collection := range collections {
for _, account := range accounts {
collectionsWithAccounts.Collections = append(collectionsWithAccounts, collection)
collectionsWithAccounts.Accounts = append(collectionsWithAccounts, account)
}
}
How can I accomplish this merge?

You can do this even without any loops:
package main
import "fmt"
type Collection struct {
ID int64 `json:"id"`
Name *string `json:"name"`
Description *string `json:"description"`
Total *int64 `json:"total"`
}
type AccountID map[int64][]string
// merge into single struct
type CollectionWithAccounts struct {
Collections []*Collection
AccountIDs []AccountID
}
func main() {
// get the data
// []*Collections, []AccountID, err
collections, accounts, err := h.Service.Many(ctx, params)
// handle error
if err != nil {
fmt.Println(err.Error())
// more logic
}
collectionsWithAccounts := CollectionWithAccounts{
Collections: collections,
AccountIDs: accounts,
}
}

Related

Gorm Query Customized Join Extra Column

I am trying to get extra columns from a many2many relationships on Gorm. Example
Part
type Part struct {
Id unit
Name string
}
Diagram
type Diagram struct {
Id unit
Name string
Parts []Part `gorm:"many2many:diagram_parts;"`
}
DiagramPart
type DiagramPart struct{
DiagramId uint `gorm:"primaryKey"`
PartId uint `gorm:"primaryKey"`
PartDiagramNumber int
PartNumber string
PartDescription string
}
This is what I have done trying to retrieve PartNumber and PartDescription in Parts.
diagram := &Diagram{}
db := s.db.Where("id = ?", 1).
Preload("Parts", func(db *gorm.DB) *gorm.DB {
return db.Select("parts.*, diagram_parts.part_number, diagram_parts.part_description").
Joins("left join diagram_parts on diagram_parts.part_id = parts.id")
}).
First(diagram)
Unfortunately, I am not able to retrieve part_number, part_description. How should I go about it?
You can add field PartNumber and PartDescription on struct Part OR Diagram, then add tag gorm:"-:migration;->" on than fields to ignore migration and to readonly mode. But on your situation, you can add it in struct Part because you already preload it.
source: https://gorm.io/docs/models.html#Field-Level-Permission
here's the example:
package main
import (
"fmt"
"gorm.io/driver/sqlite"
"gorm.io/gorm"
)
type Part struct {
Id uint `gorm:"primaryKey"`
Name string
PartNumber string `gorm:"-:migration;->"`
PartDescription string `gorm:"-:migration;->"`
}
type Diagram struct {
Id uint `gorm:"primaryKey"`
Name string
Parts []Part `gorm:"many2many:diagram_parts;"`
// PartNumber string `gorm:"-:migration;->"`
// PartDescription string `gorm:"-:migration;->"`
}
type DiagramPart struct {
DiagramId uint `gorm:"primaryKey"`
PartId uint `gorm:"primaryKey"`
PartDiagramNumber int
PartNumber string
PartDescription string
}
func main() {
db, err := gorm.Open(sqlite.Open("test.db"), &gorm.Config{})
if err != nil {
panic("failed to connect database")
}
db.AutoMigrate(&Diagram{}, &Part{}, &DiagramPart{})
diagram := &Diagram{}
err = db.Debug().Where("id = ?", 1).
// Select("diagrams.*, diagram_parts.part_number, diagram_parts.part_description").
Preload("Parts", func(db *gorm.DB) *gorm.DB {
return db.Select("parts.*, diagram_parts.part_number, diagram_parts.part_description").
Joins("left join diagram_parts on diagram_parts.part_id = parts.id")
}).
// Joins("left join diagram_parts on diagram_parts.diagram_id = diagrams.id").
First(diagram).Error
if err != nil {
panic(err)
}
fmt.Printf("diagram: %v\n", diagram)
fmt.Println("part number:", diagram.Parts[0].PartNumber)
}

Golang Gorm reverse hasMany relation

We can Easy to get the child of hasMany relation in Golang Gorm with Preload.
But how to get reverse of relation.
type Owner struct {
ID int `gorm:"column:id" json:"id"`
Name string `gorm:"column:name" json:"name"`
Projects []Project `gorm:"foreignkey:OwnerID" json:"projects"`
}
type Project struct {
ID int `gorm:"column:id" json:"id"`
Name string `gorm:"column:name" json:"name"`
OwnerID int `gorm:"column:owner_id" json:"owner_id"`
Gallery []Gallery `gorm:"foreignkey:ProjectID" json:"gallery"`
}
type Gallery struct {
ID int `gorm:"column:id" json:"id"`
ProjectID int `gorm:"column:project_id" json:"project_id"`
Url string `gorm:"column:url" json:"url"`
Title string `gorm:"column:title" json:"title"`
Description string `gorm:"column:description" json:"description"`
}
we can populate Gallery inside Project with Preload, like this:
db.Preload("Gallery").Find(&project)
How to get reverse, we want load project from gallery or owner from project?
the result I want some thing like this, when get project in form of json:
{
"id": 1,
"name": "Name Of Project",
"owner": {
"id":1,
"name": "Owner 1"
},
"gallery":[]
}
I am a bit late to the game with this answer, but you can define inverse relationships (belongs to) on the child models by doing this:
type ParentModel struct {
Children []Child // This is a 'has many'
}
type Child struct {
ParentModelID int // These behave as a 'belongs to'
ParentModel ParentModel // You need both the model and the modelID
}
Full working example
This uses the models from your question. A note on performance, the preloading in gorm isn't done via joins. It's done via additional SQL queries, the more nesting the more queries. If you have commonly executed preloading, it could be worth writing a raw query and a customer scanner to use an optimised query.
package main
import (
"fmt"
"gorm.io/driver/sqlite"
"gorm.io/gorm"
)
type Owner struct {
gorm.Model
Projects []Project
}
type Project struct {
gorm.Model
OwnerID int
Owner Owner
Gallery []Gallery
}
type Gallery struct {
gorm.Model
ProjectID int
Project Project
}
func main() {
db, err := gorm.Open(sqlite.Open("many2many.db"), &gorm.Config{})
if err != nil {
panic("failed to connect database")
}
err = db.AutoMigrate(&Owner{}, &Project{}, &Gallery{})
if err != nil {
return
}
ownerOne := Owner{}
db.Create(&ownerOne)
projectOne := Project{Owner: ownerOne}
projectTwo := Project{Owner: ownerOne}
db.Create(&projectOne)
db.Create(&projectTwo)
galleryOne := Gallery{Project: projectOne}
galleryTwo := Gallery{Project: projectOne}
galleryThree := Gallery{Project: projectTwo}
galleryFour := Gallery{Project: projectTwo}
db.Create(&galleryOne)
db.Create(&galleryTwo)
db.Create(&galleryThree)
db.Create(&galleryFour)
// Find by project and preload owners
fetchedProject := Project{}
db.Preload("Owner").Find(&fetchedProject, projectOne.ID)
fmt.Println(fetchedProject.Owner)
// Find by a gallery and preload project and owner
fetchedGallery := Gallery{}
db.Preload("Project.Owner").Find(&fetchedGallery, galleryOne.ID)
fmt.Printf("Gallery.id = %d --> Project.id = %d --> Owner.id = %d\n", fetchedGallery.ID,
fetchedGallery.Project.ID, fetchedGallery.Project.Owner.ID)
}

sqlx Unable to use Struct Scan with nested structs

I have different types for a REST API which have some things in common:
All types have Id
All types have Type
All types have Attributes, but Attributes differ
As I would not like to have a big count of unnecessary types I came up with a type format but which does not seem to be scannable by sqlx.
type Pool struct {
Id *string `json:"id" db:"pool_id"`
Type *string `json:"type"`
Attributes struct {
Name *string `json:"name" db:"name"`
Description *string `json:"description" db:"description"`
} `json:"attributes"`
}
type User struct {
Id *string `json:"id" db:"user_id"`
Type *string `json:"type"`
Attributes struct {
FirstName *string `json:"first_name" db:"first_name"`
LastName *string `json:"last_name" db:"last_name"`
} `json:"attributes"`
}
When trying to load the SQL result into the variable I get an error: missing destination name name in *main.Pool
package main
import (
_ "github.com/jmoiron/sqlx"
)
func selectOne() interface{} {
pa := Pool{}
err := dbx.Get(&pa, "SELECT * FROM pools LIMIT 1")
if err != nil {
panic(err)
}
return pa
}
As name is present I do not see why this does not work. Any advice how to solve this is highly appreciated!
package main
import (
_ "github.com/jmoiron/sqlx"
)
func selectOne() interface{} {
pa := Pool{}
err := dbx.Get(&pa, `SELECT p.id, p.type, a.name AS "attributes.name" FROM pools p left join attributes a on a.p LIMIT 1`)
if err != nil {
panic(err)
}
return pa
}

reusing structs in another struct in golang

I have two structs in golang as below
type Data struct {
Name string
Description string
HasMore bool
}
type DataWithItems struct {
Name string
Description string
HasMore bool
Items []Items
}
At most DataWithItems struct can be rewritten as
type DataWithItems struct {
Info Data
Items []Items
}
But the above make it difficult when decoding a json object into DataWithItems. I know this can be solved with inheritance in other programming languages but Is there a way I can solve this in Go?
You can "embed" the one struct into the other:
type Items string
type Data struct {
Name string
Description string
HasMore bool
}
type DataWithItems struct {
Data // Notice that this is just the type name
Items []Items
}
func main() {
d := DataWithItems{}
d.Data.Name = "some-name"
d.Data.Description = "some-description"
d.Data.HasMore = true
d.Items = []Items{"some-item-1", "some-item-2"}
result, err := json.Marshal(d)
if err != nil {
panic(err)
}
println(string(result))
}
this prints
{"Name":"some-name","Description":"some-description","HasMore":true,"Items":["some-item-1","some-item-2"]}
Just use one struct - DataWithItems and sometimes leave items blank

Nested structs using gorm model

I have struct called User:
type User struct {
Email string
Name string
}
and struct called UserDALModel:
type UserDALModel struct {
gorm.Model
models.User
}
gorm Model looks like this:
type Model struct {
ID uint `gorm:"primary_key"`
CreatedAt time.Time
UpdatedAt time.Time
DeletedAt *time.Time `sql:"index"`
}
this is possible to make UserDALModel nested with gorm model and user model so the output will be:
{
ID
CreatedAt
UpdatedAt
DeletedAt
Email
Name
}
now the output is:
{
Model: {
ID
CreatedAt
UpdatedAt
DeletedAt
}
User: {
Name
Email
}
}
According to this test in gorm, I think you need to add an embedded tag to the struct.
type UserDALModel struct {
gorm.Model `gorm:"embedded"`
models.User `gorm:"embedded"`
}
You can also specify a prefix if you want with embedded_prefix.
I found the answer:
type UserModel struct {
Email string
Name string
}
type UserDALModel struct {
gorm.Model
*UserModal
}
------------------------------
user := UserModel{"name", "email#email.com"}
userDALModel := UserDALModel{}
userDal.UserModal = &user
be careful embedding two structs with the same column:
package tests
import (
"fmt"
"github.com/jinzhu/gorm"
_ "github.com/jinzhu/gorm/dialects/mysql"
"testing"
)
type A struct {
X string
Y string
}
type B struct {
X string
Y string
}
type AB struct {
B B `gorm:"embedded"` // Embedded struct B before struct A
A A `gorm:"embedded"`
}
var DB *gorm.DB
func connectDB() error {
var err error
spec := "slumberuser:password#tcp(localhost:3306)/slumber"
DB, err = gorm.Open("mysql", spec+"?parseTime=true&loc=UTC&charset=utf8")
DB.LogMode(true) // Print SQL statements
//defer DB.Close()
if err != nil {
return err
}
return nil
}
// cd tests
// go test -run TestGormEmbed
func TestGormEmbed(t *testing.T) {
if err := connectDB(); err != nil {
t.Errorf("error connecting to db %v", err)
}
values := []interface{}{&A{}, &B{}}
for _, value := range values {
DB.DropTable(value)
}
if err := DB.AutoMigrate(values...).Error; err != nil {
panic(fmt.Sprintf("No error should happen when create table, but got %+v", err))
}
DB.Save(&A{X: "AX1", Y: "AY1"})
DB.Save(&A{X: "AX2", Y: "AY2"})
DB.Save(&B{X: "BX1", Y: "BY1"})
DB.Save(&B{X: "BX2", Y: "BY2"})
//select * from `as` join `bs`;
// # x,y,x,y
// # AX1,AY1,BX1,BY1
// # AX2,AY2,BX1,BY1
// # AX1,AY1,BX2,BY2
// # AX2,AY2,BX2,BY2
var abs []AB
DB.Select("*").
Table("as").
Joins("join bs").
Find(&abs)
for _, ab := range abs {
fmt.Println(ab.A, ab.B)
}
// if it worked it should print
//{AX1 AY1} {BX1 BY1}
//{AX2 AY2} {BX1 BY1}
//{AX1 AY1} {BX2 BY2}
//{AX2 AY2} {BX2 BY2}
// but actually prints
//{BX1 BY1} {AX1 AY1}
//{BX1 BY1} {AX2 AY2}
//{BX2 BY2} {AX1 AY1}
//{BX2 BY2} {AX2 AY2}
}

Resources