Skip to content

Commit a89187b

Browse files
andrerfcsantosiHiDErikSchierboomjunedev
committed
Add pointer concept
Co-authored-by: Jeremy Walker <[email protected]> Co-authored-by: Erik Schierboom <[email protected]> Co-authored-by: June <[email protected]>
1 parent 131ca49 commit a89187b

File tree

15 files changed

+1018
-5
lines changed

15 files changed

+1018
-5
lines changed
Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -1,5 +1,5 @@
11
{
2-
"blurb": "TODO: add blurb",
3-
"authors": ["ErikSchierboom"],
2+
"blurb": "Pointers hold the memory address of a value.",
3+
"authors": ["andrerfcsantos"],
44
"contributors": []
55
}

concepts/pointers/about.md

Lines changed: 169 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -1,3 +1,171 @@
11
# About
22

3-
TODO: add information on pointers concept
3+
Like many other languages, Go has pointers.
4+
If you're new to pointers, they can feel a little mysterious but once you get used to them, they're quite straight-forward.
5+
They're a crucial part of Go, so take some time to really understand them.
6+
7+
Before digging into the details, it's worth understanding the use of pointers. Pointers are a way to share memory with other parts of our program, which is useful for two major reasons:
8+
1. When we have large amounts of data, making copies to pass between functions is very inefficient.
9+
By passing the memory location of where the data is stored instead, we can dramatically reduce the resource-footprint of our programs.
10+
2. By passing pointers between functions, we can access and modify the computer's memory directly, meaning that any changes made by one function are immediately visible to other parts of the program when the function ends.
11+
12+
## Variables and Memory
13+
14+
Let's say we have a regular integer variable `a`:
15+
16+
```go
17+
var a int
18+
```
19+
20+
When we declare a variable, Go has to find a place in memory to store its value. This is largely abstracted from us — when we need to fetch the value stored in that piece of memory, we can just refer to it by the variable name.
21+
22+
For instance, when we write `a + 2`, we are effectively fetching the value stored in the memory associated with the variable `a` and adding 2 to it.
23+
24+
Similarly, when we need to change the value in the piece of memory of `a`, we can use the variable name to do an assignment:
25+
26+
```go
27+
a = 3
28+
```
29+
30+
The piece of memory that is associated with `a` will now will now be storing the value `3`.
31+
32+
## Pointers
33+
34+
While variables allow us to refer to values in memory, sometimes it's useful to know the **memory address** to which the variable is pointing. **Pointers** hold the memory addresses of those values. You declare a variable with a pointer type by prefixing the underlying type with an asterisk:
35+
36+
```go
37+
var p *int // 'p' contains the memory address of an integer
38+
```
39+
40+
Here we declaring a variable `p` of type "pointer to int" (`*int`). This means that `p` will hold the memory address of an integer. The zero value of pointers is `nil` because a `nil` pointer holds no memory address.
41+
42+
### Getting a pointer to a variable
43+
44+
To find the memory address of the value of a variable, we can use the `&` operator.
45+
For example, if we want to find and store the memory address of variable `a` in the pointer `p`, we can do the following:
46+
47+
```go
48+
var a int
49+
a = 2
50+
51+
var p *int
52+
p = &a // the variable 'p' contains the memory address of 'a'
53+
```
54+
55+
### Accessing the value via a pointer (dereferencing)
56+
57+
When we have a pointer, we might want to know the value stored in the memory address the pointer represents. We can do this using the `*` operator:
58+
59+
```go
60+
var a int
61+
a = 2
62+
63+
var p *int
64+
p = &a // the variable 'p' contains the memory address of 'a'
65+
66+
var b int
67+
b = *p // b == 2
68+
```
69+
70+
The operation `*p` fetches the value stored at the memory address stored in `p`. This operation is often called "dereferencing".
71+
72+
We can also use the derefering operator to assign a new value to the memory address referenced by the pointer:
73+
74+
```go
75+
var a int
76+
a = 2 // declare int variable 'a' and assign it the value of 2
77+
78+
var pa *int
79+
pa = &a // 'pa' now contains to the memory address of 'a'
80+
*pa = *pa + 2 // increment by 2 the value at memory address 'pa'
81+
82+
fmt.Println(a) // Output: 4
83+
// 'a' will have the new value that was changed via the pointer!
84+
```
85+
86+
Assigning to `*pa` will change the value stored at the memory address `pa` holds. Since `pa` holds the memory address of `a`, by assigning to `*pa` we are effectively changing the value of `a`!
87+
88+
A note of caution however: always check if a pointer is not `nil` before dereferencing. Dereferencing a `nil` pointer will make the program crash at runtime!
89+
90+
```go
91+
var p *int // p is nil initially
92+
fmt.Println(*p)
93+
// panic: runtime error: invalid memory address or nil pointer dereference
94+
```
95+
96+
### Pointers to structs
97+
98+
So far we've only seen pointers to primitive values. We can also create pointers for structs:
99+
100+
```go
101+
type Person struct {
102+
Name string
103+
Age int
104+
}
105+
106+
var peter Person
107+
peter = Person{Name: "Peter", Age: 22}
108+
109+
var p *Person
110+
p = &peter
111+
```
112+
113+
We could have also created a new `Person` and immediately stored a pointer to it:
114+
115+
```go
116+
var p *Person
117+
p = &Person{Name: "Peter", Age: 22}
118+
```
119+
120+
When we have a pointer to a struct, we don't need to dereference the pointer before accessing one of the fields:
121+
122+
```go
123+
var p *Person
124+
p = &Person{Name: "Peter", Age: 22}
125+
126+
fmt.Println(p.Name) // Output: "Peter"
127+
// Go automatically dereferences 'p' to allow
128+
// access to the 'Name' field
129+
```
130+
131+
## Slices and maps are already pointers
132+
133+
Slices and maps are special types because they already have pointers in their implementation. This means that more often that not, we don't need to create pointers for these types to share the memory address for their values. Imagine we have a function that increments the value of a key in a map:
134+
135+
136+
```go
137+
func incrementPeterAge(m map[string]int){
138+
m["Peter"] += 1
139+
}
140+
```
141+
142+
If we create a map and call this function, the changes the function made to the map persist after the function ended. This is a similar behavior we get if we were using a pointer, but note how on this example we are not using any referencing/dereferencing or any of the pointer syntax:
143+
144+
```go
145+
ages := map[string]int{
146+
"Peter": 21
147+
}
148+
incrementPeterAge(ages)
149+
fmt.Println(ages)
150+
// Output: map[Peter:22]
151+
// The changes the function 'addPeterAge' made to the map are visible after the function ends!
152+
```
153+
154+
The same applies when changing an existing item in a slice.
155+
156+
However, actions that return a new slice like `append` are a special case and **might not** modify the slice outside of the function. This is due to the way slices work internally, but we won't cover this in detail in this exercise, as this is a more advanced topic. If you are really curious you can read more about this in [Go Blog: Mechanics of 'append'][mechanics-of-append]
157+
158+
## Pointer arithmetic
159+
160+
Unlike other languages, Go does not support pointer arithmetic. This means you cannot go from a memory address of one value to the address of another by incrementing the pointer value. So snippets like the following are invalid:
161+
162+
```go
163+
var a int
164+
a = 2
165+
166+
var pa *int
167+
pa = &a
168+
pa++ // NOT ALLOWED: incrementing the pointer or any other pointer arithmetic operations are invalid
169+
```
170+
171+
[mechanics-of-append]: https://go.dev/blog/slices

concepts/pointers/introduction.md

Lines changed: 156 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -1,3 +1,158 @@
11
# Introduction
22

3-
TODO: add introduction for pointers concept
3+
Like many other languages, Go has pointers.
4+
If you're new to pointers, they can feel a little mysterious but once you get used to them, they're quite straight-forward.
5+
They're a crucial part of Go, so take some time to really understand them.
6+
7+
Before digging into the details, it's worth understanding the use of pointers. Pointers are a way to share memory with other parts of our program, which is useful for two major reasons:
8+
1. When we have large amounts of data, making copies to pass between functions is very inefficient.
9+
By passing the memory location of where the data is stored instead, we can dramatically reduce the resource-footprint of our programs.
10+
2. By passing pointers between functions, we can access and modify the computer's memory directly, meaning that any changes made by one function are immediately visible to other parts of the program when the function ends.
11+
12+
## Variables and Memory
13+
14+
Let's say we have a regular integer variable `a`:
15+
16+
```go
17+
var a int
18+
```
19+
20+
When we declare a variable, Go has to find a place in memory to store its value. This is largely abstracted from us — when we need to fetch the value stored in that piece of memory, we can just refer to it by the variable name.
21+
22+
For instance, when we write `a + 2`, we are effectively fetching the value stored in the memory associated with the variable `a` and adding 2 to it.
23+
24+
Similarly, when we need to change the value in the piece of memory of `a`, we can use the variable name to do an assignment:
25+
26+
```go
27+
a = 3
28+
```
29+
30+
The piece of memory that is associated with `a` will now will now be storing the value `3`.
31+
32+
## Pointers
33+
34+
While variables allow us to refer to values in memory, sometimes it's useful to know the **memory address** to which the variable is pointing. **Pointers** hold the memory addresses of those values. You declare a variable with a pointer type by prefixing the underlying type with an asterisk:
35+
36+
```go
37+
var p *int // 'p' contains the memory address of an integer
38+
```
39+
40+
Here we declaring a variable `p` of type "pointer to int" (`*int`). This means that `p` will hold the memory address of an integer. The zero value of pointers is `nil` because a `nil` pointer holds no memory address.
41+
42+
### Getting a pointer to a variable
43+
44+
To find the memory address of the value of a variable, we can use the `&` operator.
45+
For example, if we want to find and store the memory address of variable `a` in the pointer `p`, we can do the following:
46+
47+
```go
48+
var a int
49+
a = 2
50+
51+
var p *int
52+
p = &a // the variable 'p' contains the memory address of 'a'
53+
```
54+
55+
### Accessing the value via a pointer (dereferencing)
56+
57+
When we have a pointer, we might want to know the value stored in the memory address the pointer represents. We can do this using the `*` operator:
58+
59+
```go
60+
var a int
61+
a = 2
62+
63+
var p *int
64+
p = &a // the variable 'p' contains the memory address of 'a'
65+
66+
var b int
67+
b = *p // b == 2
68+
```
69+
70+
The operation `*p` fetches the value stored at the memory address stored in `p`. This operation is often called "dereferencing".
71+
72+
We can also use the derefering operator to assign a new value to the memory address referenced by the pointer:
73+
74+
```go
75+
var a int
76+
a = 2 // declare int variable 'a' and assign it the value of 2
77+
78+
var pa *int
79+
pa = &a // 'pa' now contains to the memory address of 'a'
80+
*pa = *pa + 2 // increment by 2 the value at memory address 'pa'
81+
82+
fmt.Println(a) // Output: 4
83+
// 'a' will have the new value that was changed via the pointer!
84+
```
85+
86+
Assigning to `*pa` will change the value stored at the memory address `pa` holds. Since `pa` holds the memory address of `a`, by assigning to `*pa` we are effectively changing the value of `a`!
87+
88+
A note of caution however: always check if a pointer is not `nil` before dereferencing. Dereferencing a `nil` pointer will make the program crash at runtime!
89+
90+
```go
91+
var p *int // p is nil initially
92+
fmt.Println(*p)
93+
// panic: runtime error: invalid memory address or nil pointer dereference
94+
```
95+
96+
### Pointers to structs
97+
98+
So far we've only seen pointers to primitive values. We can also create pointers for structs:
99+
100+
```go
101+
type Person struct {
102+
Name string
103+
Age int
104+
}
105+
106+
var peter Person
107+
peter = Person{Name: "Peter", Age: 22}
108+
109+
var p *Person
110+
p = &peter
111+
```
112+
113+
We could have also created a new `Person` and immediately stored a pointer to it:
114+
115+
```go
116+
var p *Person
117+
p = &Person{Name: "Peter", Age: 22}
118+
```
119+
120+
When we have a pointer to a struct, we don't need to dereference the pointer before accessing one of the fields:
121+
122+
```go
123+
var p *Person
124+
p = &Person{Name: "Peter", Age: 22}
125+
126+
fmt.Println(p.Name) // Output: "Peter"
127+
// Go automatically dereferences 'p' to allow
128+
// access to the 'Name' field
129+
```
130+
131+
## Slices and maps are already pointers
132+
133+
Slices and maps are special types because they already have pointers in their implementation. This means that more often that not, we don't need to create pointers for these types to share the memory address for their values. Imagine we have a function that increments the value of a key in a map:
134+
135+
136+
```go
137+
func incrementPeterAge(m map[string]int){
138+
m["Peter"] += 1
139+
}
140+
```
141+
142+
If we create a map and call this function, the changes the function made to the map persist after the function ended. This is a similar behavior we get if we were using a pointer, but note how on this example we are not using any referencing/dereferencing or any of the pointer syntax:
143+
144+
```go
145+
ages := map[string]int{
146+
"Peter": 21
147+
}
148+
incrementPeterAge(ages)
149+
fmt.Println(ages)
150+
// Output: map[Peter:22]
151+
// The changes the function 'addPeterAge' made to the map are visible after the function ends!
152+
```
153+
154+
The same applies when changing an existing item in a slice.
155+
156+
However, actions that return a new slice like `append` are a special case and **might not** modify the slice outside of the function. This is due to the way slices work internally, but we won't cover this in detail in this exercise, as this is a more advanced topic. If you are really curious you can read more about this in [Go Blog: Mechanics of 'append'][mechanics-of-append]
157+
158+
[mechanics-of-append]: https://go.dev/blog/slices

concepts/pointers/links.json

Lines changed: 10 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -1 +1,10 @@
1-
[]
1+
[
2+
{
3+
"url": "https://tour.golang.org/moretypes/1",
4+
"description": "Tour of Go: Pointers"
5+
},
6+
{
7+
"url": "https://gobyexample.com/pointers",
8+
"description": "Go by Example: Pointers"
9+
}
10+
]

config.json

Lines changed: 22 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -282,6 +282,23 @@
282282
"basics"
283283
],
284284
"status": "beta"
285+
},
286+
{
287+
"name": "Election Day",
288+
"slug": "election-day",
289+
"uuid": "b9f43f40-2540-48be-b721-71117dee2250",
290+
"concepts": [
291+
"pointers"
292+
],
293+
"prerequisites": [
294+
"structs",
295+
"slices",
296+
"maps",
297+
"arithmetic-operators",
298+
"string-formatting",
299+
"packages"
300+
],
301+
"status": "beta"
285302
}
286303
],
287304
"practice": [
@@ -1928,6 +1945,11 @@
19281945
"name": "Comparison",
19291946
"slug": "comparison",
19301947
"uuid": "f7abb288-9526-447d-9eb8-45d756f2ae9b"
1948+
},
1949+
{
1950+
"name": "Pointers",
1951+
"slug": "pointers",
1952+
"uuid": "57477550-cecf-4334-903a-85a6c32d61d0"
19311953
}
19321954
],
19331955
"key_features": [

0 commit comments

Comments
 (0)