Making code more generic - go

I have a program where many functionalities are similar across different structures, however, I end up writing these functions again and again, esp because the variable that are being dealt inside are of different structures.
I have written a sample code here.
In Go Playgroud
package main
import "fmt"
func (a *Match) Add(v Match) {
a.Runs += v.Runs
a.Points += v.Points
}
type Match struct {
Runs uint64
Points uint64
}
func (a *Activity) Add(v Activity) {
a.Walk += v.Walk
a.Jog += v.Jog
}
type Activity struct {
Walk uint64
Jog uint64
}
func GetDailyMatches() map[string]Match {
var dailyMatches map[string]Match
Match1, Match2 := Match{5, 10}, Match{1, 2}
dailyMatches = make(map[string]Match)
dailyMatches["01"] = Match1
dailyMatches["02"] = Match2
dailyMatches["03"] = Match1
dailyMatches["04"] = Match2
return dailyMatches
}
func GetDailyActivities() map[string]Activity {
var dailyActivities map[string]Activity
Activity1, Activity2 := Activity{5, 10}, Activity{1, 2}
dailyActivities = make(map[string]Activity)
dailyActivities["01"] = Activity1
dailyActivities["02"] = Activity2
dailyActivities["03"] = Activity1
dailyActivities["04"] = Activity2
return dailyActivities
}
func main() {
fmt.Println(CalculateMatchSummary("01", "03"))
fmt.Println(CalculateActivitySummary("02", "04"))
fmt.Println(CalculateMatchSummary("01", "03"))
fmt.Println(CalculateActivitySummary("02", "04"))
}
func CalculateMatchSummary(start, end string) (total Match) {
dailyMatches := GetDailyMatches()
for day, value := range dailyMatches {
if day < start {
continue
} else if day > end {
continue
} else {
total.Add(value)
}
}
return
}
func CalculateActivitySummary(start, end string) (total Activity) {
dailyActivities := GetDailyActivities()
for day, value := range dailyActivities {
if day < start {
continue
} else if day > end {
continue
} else {
total.Add(value)
}
}
return
}
If you notice, both Match and Activity has the same functions and same structures, except that internally they are of different structures.
Is there a easy way to make the code more generic (Go generics, which is not there in Go??) in Golang itself.

Go has a pretty package "reflect". You can not do genericity strictly speaking but you can get unification of code for the same behavior.
I've changed your playground a bit : https://play.golang.org/p/bfqZsFOgVQ
The main part :
func AddTwo(a, b interface{}) interface{} {
va := reflect.ValueOf(a)
vb := reflect.ValueOf(b)
res := reflect.New(reflect.TypeOf(a)).Elem()
if va.Kind() != reflect.Struct && vb.Kind() != reflect.Struct {
return nil
}
na, nb := va.NumField(), vb.NumField()
if na != nb {
return nil
}
for i := 0; i < na; i++ {
// additional verification needed here
fa := va.Field(i).Uint()
fb := vb.Field(i).Uint()
fr := fa + fb
res.Field(i).SetUint(fr)
}
return res.Interface()
}
I use reflect to check the fields of the struct I am given. If both are uint64, I can add them reflectively. If your structs contains many uint64, it can add them all !
Note that you must convert the resulting interface to the type of the struct given after calling this function. That is why this is not strictly generic, because the returning type is a interface, and not a Match or Activity.
EDIT: No need even to return a new struct. You can simply update the field of the "a" struct by calling .SetUint() method.

Related

Basic + Slice + Map Type Compatible Generics?

Is there a way to create a generic function that can adjust its operation when passed a map or a slice type vs a basic type?
Goal
Create a slice reading function generator with a flexible return type:
func ValueReader[T <probably something fancy>](i int) func ([]ProtoConvertable) T {
return func (row []ProtoConvertable) T {
return ...
}
}
row := []ProtoConvertable{
&Data[int]{Value: 333},
&ListData{Values: []ProtoConvertable{
&Data[string]{Value: "hello"},
&Data[string]{Value: "world"},
}},
&MapData{Values: map[ProtoConvertable]ProtoConvertable{
&Data[int]{Value: 22}: &Data[string]{Value: "world"},
&Data[int]{Value: 11}: &Data[string]{Value: "hello"},
}},
}
dataReader := ValueReader[int](0) // A function that converts the first element to an int
listDataReader := ValueReader[[]string](1) // A function that converts the second element to a slice
mapDataReader := ValueReader[map[int]string](2) // A function that converts the third element to a map
data := dataReader(row) // 333
listData := listDataReader(row) // []string{"hello", "world"}
mapData := mapDataReader(row) // map[int]string{11: "hello", 22: "world"}
Types
type ValueType interface {
int | string
}
type ProtoConvertable interface {
ToProto() *pb.GenericMessage
}
type Data[T ValueType] struct {
Value T
}
func (d *Data) ToProto() *pb.GenericMessage{
...
}
type ListData struct {
Values []ProtoConvertable
}
func (d *ListData) ToProto() *pb.GenericMessage {
...
}
type MapData struct {
Values map[ProtoConvertable]ProtoConvertable
}
func (d *MapData) ToProto() *pb.GenericMessage {
...
}
Current Solution
func ValueReader[T ValueType](i int) func([]ProtoConvertable) T {
return func(row []ProtoConvertable) T {
return row[i].(*Data[T]).Value
}
}
func ListValueReader[T ValueType](i int) func([]ProtoConvertable) []T {
return func(row []ProtoConvertable) []T {
vs := row[i].(*ListData).Values
res := make([]T, len(vs))
for i, v := range vs {
res[i] = v.(*Data[T]).Value
}
return res
}
}
func MapValueReader[K ValueType, V ValueType](i int) func([]ProtoConvertable) map[K]V {
return func(row []ProtoConvertable) map[K]V {
vs := row[i].(*MapData).Values
res := make(map[K]V, len(vs))
for k, v := range vs {
res[k.(*Data[K]).Value] = v.(*Data[V]).Value
}
return res
}
}
dataReader := ValueReader[int](0)
listDataReader := ListValueReader[string](1)
mapDataReader := MapValueReader[int, string](2)
Note: all of this code is an untested simplification of a more complicated library. It might need some tweaking to get to actually work.
The <probably something fancy> doesn't exist.
The main issue is that you want to model a type parameter that matches a base value and two composite types, one of which is a map type where you want to capture both K and V.
Even if it existed, the body of ValueReader would be a type-switch on T to return each specialized reader function, so your existing solution that involves a small amount of code duplication seems just a better strategy overall.
My advice is to use generics when the operations on the different concrete types of T are really identical. You can read more at: https://go.dev/blog/when-generics

Implement generic linked list in golang which disallows different types in the same linkedlist

I want to make a linked list which can hold the values of any type but the linked list must hold values of any one type only.
In general when I am using interfaces for achieving this - any type which implements the interface for the node can be added to the linked list.
I have written an implementation for this where whenever a new key is added to the linked list, the type of the key is checked against the type of the key at the head. I want to understand if this is the correct way of implementing it or is there a better way of doing it.
package main
import (
"errors"
"fmt"
"reflect"
"strings"
)
type MyNode struct {
value int
}
func (node *MyNode) PrintValue() {
fmt.Printf(" %d ", node.value)
}
type llNode struct {
key llNodeInterface
next *llNode
llNodeType reflect.Type
}
type llNodeInterface interface {
PrintValue()
}
type ComplexNode struct {
realValue int
imageValue int
}
func (node *ComplexNode) PrintValue() {
fmt.Printf(" %d + i%d", node.realValue, node.imageValue)
}
// Student type.
type Student struct {
name string
age int
}
// Student implements the PrintValue function - thus llNodeInterface is implemented.
func (node *Student) PrintValue() {
fmt.Printf("Name: %s | Age : %d ", node.name, node.age)
}
// Function which will check the of the new node before adding to the linked
// list. It checks the type of the new key against the type of the key in the
// head. If both are equal then it proceed else return error.
func (head *llNode) AddBeforeHeadTypeCheck(passedKey llNodeInterface) error {
if head.key == nil {
head.key = passedKey
head.llNodeType = reflect.TypeOf(head.key)
} else {
typeOfPassedKey := reflect.TypeOf(passedKey)
if typeOfPassedKey != head.llNodeType {
fmt.Printf("\nUnsupported type for the type %T", passedKey)
return errors.New("Type mistmatch")
}
temp := llNode{key: head.key, next: head.next}
head.key = passedKey
head.next = &temp
}
return nil
}
// Function which will not check the types and will simply add the new node to
// the linked list. Thus linked list will be able to have nodes of multiple
// types.
func (head *llNode) AddBeforeHead(passedKey llNodeInterface) {
if head.key == nil {
head.key = passedKey
head.llNodeType = reflect.TypeOf(head.key)
} else {
temp := llNode{key: head.key, next: head.next}
head.key = passedKey
head.next = &temp
}
}
func (head *llNode) Init() {
head.key = nil
head.next = nil
head.llNodeType = nil
}
// Print the linked list.
func (head *llNode) DisplayLL() {
temp := head
fmt.Printf("\n%s", strings.Repeat("#", 80))
fmt.Printf("\nPrinting the linked list\n")
for {
if temp.key == nil {
fmt.Println("Linked list is empty")
break
} else {
fmt.Printf("\n %T %v ", temp.key, temp.key)
key := temp.key
key.PrintValue()
if temp.next == nil {
break
} else {
temp = temp.next
}
}
}
fmt.Printf("\n%s", strings.Repeat("#", 80))
fmt.Printf("\n\n")
}
func testWithMixedType() {
head := llNode{}
head.Init()
for i := 1; i < 10; i++ {
temp := &ComplexNode{i, i * 10}
head.AddBeforeHeadTypeCheck(temp)
}
temps := &Student{"rishi", 20}
head.AddBeforeHeadTypeCheck(temps) // Will give error.
head.DisplayLL()
}
func testWithComplexNumber() {
head := llNode{}
head.Init()
for i := 1; i < 10; i++ {
temp := &ComplexNode{i, i * 10}
head.AddBeforeHeadTypeCheck(temp)
}
}
func main() {
testWithComplexNumber()
testWithMixedType()
}
The code is working fine - but I want to understand if there is a better or different way of doing this.
Also - what is the performance impact of the current checking of types using the reflect package. Is there a different way of achieving the same thing.
Unsupported type for the type *main.Student
Unsupported type for the type *main.Student
################################################################################
Printing the linked list
*main.ComplexNode &{9 90} 9 + i90
*main.ComplexNode &{8 80} 8 + i80
*main.ComplexNode &{7 70} 7 + i70
*main.ComplexNode &{6 60} 6 + i60
*main.ComplexNode &{5 50} 5 + i50
*main.ComplexNode &{4 40} 4 + i40
*main.ComplexNode &{3 30} 3 + i30
*main.ComplexNode &{2 20} 2 + i20
*main.ComplexNode &{1 10} 1 + i10
################################################################################
Good news, starting from Go 1.18, Generics are now supported in Go.
Following the example from the question, here is a simplified LinkedList using Generics. You can tinker with it at the playground here.
package main
import "fmt"
type MyNode[T any] struct {
next *MyNode[T]
value T
}
type MyLinkedList[T any] struct {
head *MyNode[T]
tail *MyNode[T]
}
func (list *MyLinkedList[T]) Add(t T) *MyLinkedList[T] {
// create node
node := &MyNode[T]{nil, t}
// if first node in list
if list.head == nil {
list.head = node
list.tail = node
} else {
list.tail.next = node
list.tail = list.tail.next
}
return list
}
func (list *MyLinkedList[T]) AddBeforeHead(t T) *MyLinkedList[T] {
node := &MyNode[T]{nil, t}
if list.head != nil {
node.next = list.head
list.head = node
} else {
// make head
list.head = node
list.tail = node
}
return list
}
// display the list
func DisplayList[T any](list *MyLinkedList[T]) string {
var out string = ""
iter := list.head
for iter != nil {
out += fmt.Sprintf("%v -> ", iter.value)
iter = iter.next
}
return out
}
func (list *MyLinkedList[T]) Display() string {
return DisplayList(list)
}
// for printing node value
// you could also implement Stringer
// but this is besides the point, you can ignore
func (node *MyNode[T]) String() string {
return fmt.Sprintf("<MyNode: %v>", node.value)
}
// helper func: create list from array
func CreateLinkedList[T any](arr []T) *MyLinkedList[T] {
list := &MyLinkedList[T]{}
for _, v := range arr {
list.Add(v)
}
return list
}
func main() {
// create a list from array of integers
intArr := []int{10, 20, 30, 40, 50, 60}
list1 := CreateLinkedList(intArr)
// create a list from array of strings
strArr := []string{"foo", "bar", "baz", "faz"}
list2 := CreateLinkedList(strArr)
// test inserting at the beginning
list2.AddBeforeHead("hello")
fmt.Println(list1.Display())
fmt.Println(list2.Display())
}
Some official references covering basics of Generics can be found here:
https://go.dev/doc/tutorial/generics
https://go.dev/blog/intro-generics
https://go.dev/tour/generics/1
You can do it either by using interfaces and runtime checks (as you've discovered), or by using code generation. These are the current options you have in Go for generic programming. The Go team is working to add generics to the language - it's a work in progress, and everyone is free to participate in the discussion. Once generics exist, they will provide the solution you seek here.
As for interfaces vs. code generation, there are the performance implications you've mentioned. Code generation will generate tighter code that doesn't need to do runtime checks for most operations; on the other hand, it adds a bit of complexity to the build process of your project. These are the usual tradeoffs of having something resolved at run-time vs. precomputing things at compile-time.

common function to create map[string]struct from slice of struct dynamically

I have two different struct as mentioned below A abd B and two process functions. Is there any way by means of which i can write a common function to generate the map[string]struct for the both the struct. Moreover, is there any way using reflection given the struct name i can create the object of the same?
type A struct {
name string
// more fields
}
type B struct {
name string
// more fields
}
func ProcessA(input []A) map[string]A {
output := make(map[string]A)
for _, v := range input {
output[v.name] = v
}
return output
}
func ProcessB(input []B) map[string]B {
output := make(map[string]B)
for _, v := range input {
output[v.name] = v
}
return output
}
Idiomatic way in Go would be to use interface.
type Named interface {
Name() string
}
type letter struct {
name string
}
func (l letter) Name() string {
return l.name
}
type A struct {
letter
// more fields
}
type B struct {
letter
// more fields
}
func ProcessNameds(input []Named) map[string]Named {
output := make(map[string]Named, len(input))
for _, v := range input {
output[v.Name()] = v
}
return output
}
Well, see if something like this would help:
package main
import (
"fmt"
"strconv"
)
type A struct {
name string
// more fields
}
type B struct {
name string
// more fields
}
func Process(x interface{}) interface{} {
ma := make(map[string]int)
mb := make(map[string]string)
if x == nil {
return nil
} else if a, ok := x.([]A); ok {
fmt.Printf("Type A argument passed %s\n", x)
ma[a[0].name] = 1
ma[a[1].name] = 2
return ma //you can return whatever type you want here
} else if b, ok := x.([]B); ok {
fmt.Printf("Type B argument passed %s\n", x)
mb[b[0].name] = "a"
mb[b[1].name] = "b"
return mb //you can return whatever type you want here
} else {
panic(fmt.Sprintf("Unexpected type %T: %v", x, x))
}
return nil
}
func main() {
a := make([]A, 5)
for i := 0; i < len(a); i++ {
a[i].name = strconv.Itoa(i) + "A"
}
b := make([]B, 7)
for i := 0; i < len(b); i++ {
b[i].name = strconv.Itoa(i) + "B"
}
fmt.Println(Process(a))
fmt.Println(Process(b))
//Uncomment line below to see the panic
//fmt.Println(Process(8))
}
https://play.golang.org/p/irdCsbpvUv_t

Golang convert list objects to string

I have two structs:
type A struct {
BankCode string `json:"bankCode"`
BankName string `json:"bankName"`
}
And:
type B struct {
A
extra string `json:" extra"`
}
And two slices:
listsA []A and listsB []B
I want to get bankCodes from listA and listB. bankcodes only contains bankcodes. It is a []string
It will be so easy as using two function.
func getBankCodes(data []A) []string {
res := make([]string, len(data))
for i := 0; i < len(data); i++ {
res[i] = data[i].BankCode
}
return res
}
func getBankCodes(data []B) []string {
res := make([]string, len(data))
for i := 0; i < len(data); i++ {
res[i] = data[i].BankCode
}
return res
}
How to use one common function ?
Well the clean solution would be to use an interface, since go doesn't support classic inheritance, so something like []parentclass can't work. Interfaces however can only describe functions not a common field, so you have to implement a Getter (essentially).
// GetBankCoder provides a function that gives the BankCode
type GetBankCoder interface {
getBankCode() string
}
// implement GetBankCoder for A (and indirectly for B)
func (a A) getBankCode() string {
return a.BankCode
}
and make your getBankCodes work on that interface type, notice the parameter of the function as well as the statement inside the loop:
func getBankCodes(data []GetBankCoder) []string { // <-- changed
res := make([]string, len(data))
for i := 0; i < len(data); i++ {
res[i] = data[i].getBankCode() // <-- changed
}
return res
}
There are other solutions where the function parameter is of interface{} type and then reflection is used to assure you can actually do .BankCode, but I don't like those, as they are not adding more clarity either.
... However, I couldn't get the golang playground to make this work correctly without putting it into a []GetBankCoder var first, before giving it to the function.
banks := make([]GetBankCoder, 0)
banks = append(banks, A{ BankCode: "ABC", BankName: "ABC Bank"})
getBankCodes(banks)
You may use one common function like so:
func BankCodes(data interface{}) []string {
if reflect.TypeOf(data).Kind() != reflect.Slice {
panic("err: data is not slice")
}
slice := reflect.Indirect(reflect.ValueOf(data))
res := make([]string, slice.Len())
for i := 0; i < slice.Len(); i++ {
a := slice.Index(i).Interface().(BankCoder)
res[i] = a.Bankcode()
}
return res
}
Code (try on The Go Playground):
package main
import (
"fmt"
"reflect"
)
func main() {
bs := []B{B{A{"BC1", "BN"}, "e"}, B{A{"BC2", "BN"}, "e"}}
strs := BankCodes(bs)
fmt.Println(strs)
as := []A{A{"AC1", "BN"}, A{"AC2", "BN"}}
strs2 := BankCodes(as)
fmt.Println(strs2)
}
func BankCodes(data interface{}) []string {
if reflect.TypeOf(data).Kind() != reflect.Slice {
panic("err: data is not slice")
}
slice := reflect.Indirect(reflect.ValueOf(data))
res := make([]string, slice.Len())
for i := 0; i < slice.Len(); i++ {
a := slice.Index(i).Interface().(BankCoder)
res[i] = a.Bankcode()
}
return res
}
type A struct {
BankCode string `json:"bankCode"`
BankName string `json:"bankName"`
}
type B struct {
A
extra string `json:" extra"`
}
type BankCoder interface {
Bankcode() string
}
func (a A) Bankcode() string {
return a.BankCode
}

Generic variadic argument in Go?

I know that Go doesn't support templates or overloaded functions, but I'm wondering if there's any way to do some kind of generic programming for variadic functions anyway?
I have many functions such as these:
func (this Document) GetString(name string, defaults ...string) string {
v, ok := this.GetValueFromDb(name)
if !ok {
if len(defaults) >= 1 {
return defaults[0]
} else {
return ""
}
}
return v.asString
}
func (this Document) GetInt(name string, defaults ...int) int {
v, ok := this.GetValueFromDb(name)
if !ok {
if len(defaults) >= 1 {
return defaults[0]
} else {
return 0
}
}
return v.asInt
}
// etc. for many different types
Is there any way to do this without having so much redundant code?
The most of what you can achieve is usage of interface{} type, something like this:
func (this Document) Get(name string, defaults ...interface{}) interface{} {
v, ok := this.GetValueFromDb(name)
if !ok {
if len(defaults) >= 1 {
return defaults[0]
} else {
return 0
}
}
return v
}
GetValueFromDb function should also be tweaked to return interface{} value and not some wrapper like now.
Then in the client code you can do the following:
value := document.Get("index", 1).(int) // Panics when the value is not int
or
value, ok := document.Get("index", 1).(int) // ok is false if the value is not int
This will yield some runtime overhead though. I'd better stick with separate functions and try to restructure the code somehow.
Here's a working example of how you could change your code.
package main
import (
"fmt"
)
type Document struct{
getSucceeds bool
}
func (d *Document) GetValueFromDb(name string) (interface{}, bool) {
return 1, d.getSucceeds
}
func (this Document) Get(name string, def ...int) interface{} {
v, ok := this.GetValueFromDb(name)
if !ok {
if len(def) >= 1 {
return def[0]
} else {
return 0
}
}
return v
}
func main() {
d1 := Document{true}
d2 := Document{false}
var int1, int2 int
int1 = d1.Get("foo", 2).(int)
int2 = d2.Get("foo", 2).(int)
fmt.Println(int1, int2)
}
Since you know what type you expect for the given name, you can write your Get method in a generic way, returning interface{}, and then assert the type at the call site. See the spec about type assertions.
There are different ways to emulate some aspects of generics in Go. There were lots of discussions on the mailing list. Often, there's a way to restructure code so it's less dependent on generics.
In the client code you can do like this :
res := GetValue("name", 1, 2, 3)
// or
// res := GetValue("name", "one", "two", "three")
if value, ok := res.(int); ok {
// process int return value
} else if value, ok := res.(string); ok {
// process string return value
}
// or
// res.(type) expression only work in switch statement
// and 'res' variable's type have to be interface type
switch value := res.(type) {
case int:
// process int return value
case string:
// process string return value
}

Resources