I saw a lot of junior developper using forEach when they need to handle array.
I will show you and help you to over-abuse of forEach
by using the correct array methods depending on the current context!
forEach
When to use?
As I said before, you don't need to over-abuse of forEach
. In some case forEach is a very good method.
If you need to iterate an array in order to make a specific action (for example console log each item).
You shouldn't use forEach every time you need to fill a new array (before doing this, check if you can use other methods)
π
const items = [1, 2, 3, 4, 5]
items.forEach(item => console.log(item))
forEach
DOESN'T return any value !!
//
// ...
const toto = items.forEach(item => console.log(item))
toto // undefined
filter
When to use?
As named, it allow us to filter some value in your array depending on your condition.
Example π
For example if you want to remove odd number
β With forEach
const items = [1, 2, 3, 4]
const evenValue = []
items.forEach(item => {
if (item % 2 == 0) {
evenValue.push(item)
}
})
β
With filter
const items = [1, 2, 3, 4]
const evenValue = items.filter(currentValue => {
return currentValue % 2 == 0
})
π‘ When you are using filter
you should return a boolean (the condition of filter) in each iteration. (otherwise JS Engine will coerce the returned value into boolean!)
map
When to use?
When you need to transform items from array to another array!
Example π
For example if you want to multiply all value in an array by 2.
β With forEach
const items = [1, 2, 3, 4]
const result = []
items.forEach(item => {
result.push(item * 2)
})
β
With map
const items = [1, 2, 3, 4]
const result = items.map(item => {
return item * 2
})
π‘ When you are using map
you need to return an item (transformed item) in each iteration.
reduce
When to use?
When you need to get a single value from an array. This 'single value' can be an array.
Example π
For example if you want to sum all number in array.
β With forEach
const items = [1, 2, 3, 4]
let accumulator = 0
items.forEach(item => {
accumulator += item
})
β
With reduce
const items = [1, 2, 3, 4]
const sum = items.reduce((accumulator, currentValue) => {
return accumulator += currentValue
}, 0)
π‘ When you are using reduce
you need to return the accumulator (value that is returned by the reduce method) in each iteration and you should also init this accumulator (in the example above we init the accumulator to 0
)!
find
When to use?
When you need to find an item that fit a condition and you are using this item after.
const items = [1, 2, 3, 4]
const item = items.find(item => item === 3)
// ...
// Use item afterwards
some
When to use?
If you need to check if one of you item match a condition (like find
) but you don't need to use the item afterward.
Example π
For example if you want to check if you array has the number 2.
Using find
is bad if you don't use item afterward, or you just use item for checking if the value is undefined
or not (to know if the condition has matched)
β With find
const items = [1, 2, 3, 4]
const item = items.find(item => item === 2)
if (item !== undefined) {
// ...
}
β
With some
const items = [1, 2, 3, 4]
const hasNumber2 = items.some(item => item === 2)
if (hasNumber2) {
...
}
π‘ some
return boolean if at least one item matched your condition
every
When to use?
every
is similar to some
, it will check if ALL of your items match your condition. Instead of some
that will return true if only one item match the condition, every
will return true only if all items match the condition!
TLDR;
As you can see you can use a lot of array methods depending on the context, so you can avoid to over-abuse of forEach
!
It's also a good point to use methods depending on the current context for other developers that will read you code (for example if I see filter I know that we need to filter a value from this array)
A good point that I didn't show above, it's about the fact that you can combining array methods (except forEach
since forEach doesn't return value).
So you can make this π
You need to filter odd value from array and multiply event number by 2.
const items = [1, 2, 3, 4]
const result = items.filter(item => item % 2 == 0 ).map(item => item * 2)
I hope you like this reading!
π You can get my new book Underrated skills in javascript, make the difference
for FREE if you follow me on Twitter and send message to me π and SAVE 19$ π΅π΅
Or get it HERE
π«π·π₯ For french developper you can check my YoutubeChannel
π MY NEWSLETTER
βοΈ You can SUPPORT MY WORKS π
πββοΈ You can follow me on π
π Twitter : https://twitter.com/code__oz
π¨βπ» Github: https://github.com/Code-Oz
And you can mark π this article!
Top comments (24)
This is incorrect, you can omit the initial value for the accumulator. If you do this, the accumulator will be automatically initialised to the first item in the array, and the
reduce
will continue from the next item. Your example can therefore be written more efficiently as:Hey thanks for comment, in fact you but it's really not adice to not init your accumulator value!
I edtied the post about this fact thanks!
The ability to have the accumulator automatically initiated like this is a part of JS and allows for more efficient code. Telling people not to use it is bad advice. The same advice applies as per my other comment - how do you expect developers to improve at JS if there is an insistence on treating them as children who will not understand more than basic code. 'Readable' code is subjective, and should not come at the price of less efficient code, and the misleading of people who are learning.
I was about to write this comment
This isn't true. You can return anything you like - the value will be coerced to a boolean based upon its truthiness or falsiness. Examples:
Both of these filter functions return numbers, not booleans. Similarly, you could strip empty strings out of an array by just returning the string in the filter function:
This works because non-empty strings are truthy and empty strings are falsey
hey yes you can but it's not totally clear for every dev that will read your code! I prefer the option of making a good return about a condition ! I edited the post to add your comment thanks
If we want to make the code totally clear for every dev, then we need to write for the lowest common denominator of understanding... which is obviously a terrible idea since no developer will ever get any better, or understand how JS actually works
Coerce to boolean is a bad practice and you are not a better programmer because you use one less line of code, also developers that write clear code are not child
I very much did not say they were children. I said that by misleading them and falsely telling them they have to do something - you are treating them like a child - assuming they cannot understand the language feature and make the choice for themselves. You are denying them knowledge, to preserve the dogma of 'clean' code
From the MDN docs for
filter
:I know the docs, I know js and I know coertion, I also know this kind of things lead to bugs and thats why we have lint rules and typescript.
If you want to filter strings and abuse coertion you will lost one valid result from this array [null, '', 'valid'], just because one smart developer didnt want to write some extra text
You should probably email Mozilla to get them to change their docs - they're condoning abuse!
Seriously though, there's nothing wrong with doing it if you understand the ramifications. Sure, you gave an example with a very particular situation where there might be an issue, but there are also times when you KNOW what data is going to be coming in, and can code appropriately.
Appropriate code for appropriate situations
The docs are fine because they reflect reality. But js has many parts that work in weird ways that should be avoid to write maintanible code and make life a little easier for the person that is working in your code after you
If we always encourage developers to avoid the quirks, and sweep them under the carpet pretending they don't exist - knowledge of JS over the years will get progressively worse (something that I've actually seen happening over years of interviewing candidates), and you're denying the developers the chance to be better developers through a more complete understanding of the language. That is a bad idea. Learn about the quirks - learn why they're good, learn why they're bad, learn how they can help you, use them to your advantage when you can.
And then there is reduceRight which helped me a lot back in 2012 when we had input fields like
customer.billingadress.street
and had to parse that to{"customer":{"billingadress":{"street": $customer.billingadress.street.value}}}
Nice usage of reduceRight!
Thanks for this! Actually I use
map
Hi.
Interesting because i tend to forget "some" and "reduce".
Anyway note that those functions has a slower performance, and forEach is not great also, than a for in...
happy to help you ;D
Nice!
Nice article!
Can I post this in my blog and mention you in it ?
hey sorry I can't accept! I will put it in my own blog and it's bad to copy the same article in many website for the SEO ;D
πΉπΉ
Hey! I mean accumulator + currentValue but I use += since it will add current value to the current accumulator! it made the same thing but it return the accumulator directly!