ChatGPT解决这个技术问题 Extra ChatGPT

Regex to replace multiple spaces with a single space

Given a string like:

"The dog      has a long   tail, and it     is RED!"

What kind of jQuery or JavaScript magic can be used to keep spaces to only one space max?

Goal:

"The dog has a long tail, and it is RED!"
Do you also want to match whitespacy tab characters?
@Chris, Yes please, great question.... With all these different answers, how is one supposed to know which is the most efficient solution?
Everyone below is right, but this is the most optimized regex: str.replace(/ +(?= )/g,''); you're not replacing anything you don't have to.
There's not going to be any noticeable difference in performance. You could always profile it, but I doubt it would be worth it. I would go for the clearest.
@EvanCarroll: Not true -- at least on Firefox. That version runs significantly slower. See the profiling results in my answer (below).

B
BalusC

Given that you also want to cover tabs, newlines, etc, just replace \s\s+ with ' ':

string = string.replace(/\s\s+/g, ' ');

If you really want to cover only spaces (and thus not tabs, newlines, etc), do so:

string = string.replace(/  +/g, ' ');

Does this also remove, whitespacy tab characters?
This is not working when a blank instead of a tab or newline is needed. Right? /\s+/ would be working.
it would might be better for you as a function such as function removeExtraSpaces(string){ return string.replace(/\s{2,}/g, ' ');}
@Ethan: JS has a builtin function for that: trim(). It's faster than regex. You could just do string.trim().replace(/\s\s+/g, ' '); or string.replace(/\s\s+/g, ' ').trim();.
/\s\s+/g and /\s{2,}/g do not match whitespacey characters unless there are at least two adjacent to each other, e.g. will match \t\t but will not match single \t. string.replace(/\s+/g, ' ') will match all single and multiple whitespacey character substrings and replace with single space.
S
Skippy le Grand Gourou

Since you seem to be interested in performance, I profiled these with firebug. Here are the results I got:

str.replace( /  +/g, ' ' )       ->  380ms
str.replace( /\s\s+/g, ' ' )     ->  390ms
str.replace( / {2,}/g, ' ' )     ->  470ms
str.replace( / +/g, ' ' )        ->  790ms
str.replace( / +(?= )/g, ' ')    -> 3250ms

This is on Firefox, running 100k string replacements.

I encourage you to do your own profiling tests with firebug, if you think performance is an issue. Humans are notoriously bad at predicting where the bottlenecks in their programs lie.

(Also, note that IE 8's developer toolbar also has a profiler built in -- it might be worth checking what the performance is like in IE.)


jsperf.com/removing-multiple-spaces Go forth and JSPerf! The last method; ( / +(?= )/g, ' '); fails in IE9, it leaves double spaces: "Foo Bar Baz".replace(/ +(?= )/g, ' '); --> "Foo Bar Baz"
@VivekPanday - I imagine that this is because the second line is only replacing occurrences of double spaces with a single space, while the first is replacing any space with a space as well. Whether this is time saved during the search or the actual replace, I don't know.
This doesn't remove the initial and trailing whitespaces. For that see this answer.
w
watain
var str = "The      dog        has a long tail,      and it is RED!";
str = str.replace(/ {2,}/g,' ');

EDIT: If you wish to replace all kind of whitespace characters the most efficient way would be like that:

str = str.replace(/\s{2,}/g,' ');

E
Ethan

A more robust method: This takes care of also removing the initial and trailing spaces, if they exist. Eg:

// NOTE the possible initial and trailing spaces
var str = "  The dog      has a long   tail, and it     is RED!  "

str = str.replace(/^\s+|\s+$|\s+(?=\s)/g, "");

// str -> "The dog has a long tail, and it is RED !"

Your example didn't have those spaces but they are a very common scenario too, and the accepted answer was only trimming those into single spaces, like: " The ... RED! ", which is not what you will typically need.


I used this pattern on PHP and works. $parts = preg_split("/^\s+|\s+$|\s+(?=\s)/", "Avenida Tancredo Neves, 745 Piso Térreo Sala");
This looks more inefficient than a simple preliminary .trim().
m
meder omuraliev

This is one solution, though it will target all space characters:

"The      dog        has a long tail,      and it is RED!".replace(/\s\s+/g, ' ')

"The dog has a long tail, and it is RED!"

Edit: This is probably better since it targets a space followed by 1 or more spaces:

"The      dog        has a long tail,      and it is RED!".replace(/  +/g, ' ')

"The dog has a long tail, and it is RED!"

Alternative method:

"The      dog        has a long tail,      and it is RED!".replace(/ {2,}/g, ' ')
"The dog has a long tail, and it is RED!"

I didn't use /\s+/ by itself since that replaces spaces that span 1 character multiple times and might be less efficient since it targets more than necessary.

I didn't deeply test any of these so lmk if there are bugs.

Also, if you're going to do string replacement remember to re-assign the variable/property to its own replacement, eg:

var string = 'foo'
string = string.replace('foo', '')

Using jQuery.prototype.text:

var el = $('span:eq(0)');
el.text( el.text().replace(/\d+/, '') )

The first one is totally pointless, \s\s+ means, an \s followed by one or more \s+, which can be reduced to a single \s+, the second example is more accurate because we only want to replace double spaces, not newlines, the third is more optimized because it only applies to examples with 2+ spaces. But str.replace(/ +(?= )/g,'');, only applies to examples with 2+ spaces but saves overwriting a space with a space step.
EvanCarroll you fail because \s\s+ is definitely different to \s+. \s\s+ would match '\t\t' or '\t\t\t' but NOT '\t'. And that's what it's all about, you don't want replace every f-en single whitespace character.
I do. Used for full-text search (and snippet-display): No random tabs, non-breakers or thingymajigs, please.
J
Joel Peltonen

I have this method, I call it the Derp method for lack of a better name.

while (str.indexOf("  ") !== -1) {
    str = str.replace(/  /g, " ");
}

Running it in JSPerf gives some surprising results where it beat some of the more sophisticated approaches EDIT Original JSPerf link http://jsperf.com/removing-multiple-spaces/3 seems to be dead at the time


I didn’t downvote this, but this looks inefficient and the JSPerf link is dead that I can’t check out the “surprising results.” So I ask you instead: was this surprising enough to be upvoted?
@КонстантинВан Thanks for pointing out the dead link!
j
james.garriss

More robust:

function trim(word)
{
    word = word.replace(/[^\x21-\x7E]+/g, ' '); // change non-printing chars to spaces
    return word.replace(/^\s+|\s+$/g, '');      // remove leading/trailing spaces
}

Why would you replace to trim when you can .trim()?
N
Narendra Jadhav

Here is an alternate solution if you do not want to use replace (replace spaces in a string without using replace javascript)

var str="The dog has a long tail, and it is RED!"; var rule=/\s{1,}/g; str = str.split(rule).join(" "); document.write(str);


If you do not want to use replace”? Why would one not want to use it? spliting and joining is obviously much slower than a replace call.
L
Leonard Meagher

I suggest

string = string.replace(/ +/g," ");

for just spaces OR

string = string.replace(/(\s)+/g,"$1");

for turning multiple returns into a single return also.


r
rfunduk

Also a possibility:

str.replace( /\s+/g, ' ' )

T
ToXic73

I know that I am late to the party, but I discovered a nice solution.

Here it is:

var myStr = myStr.replace(/[ ][ ]*/g, ' ');

And how is that different than / +/ and /[ ]+/?
s
semuzaboi

Comprehensive unencrypted answer for newbies et al.

This is for all of the dummies like me who test the scripts written by some of you guys which do not work.

The following 3 examples are the steps I took to remove special characters AND extra spaces on the following 3 websites (all of which work perfectly) {1. EtaVisa.com 2. EtaStatus.com 3. Tikun.com} so I know that these work perfectly.

We have chained these together with over 50 at a time and NO problems.

// This removed special characters + 0-9 and allows for just letters (upper and LOWER case)

function NoDoublesPls1()
{
var str=document.getElementById("NoDoubles1");
var regex=/[^a-z]/gi;
str.value=str.value.replace(regex ,"");
}

// This removed special characters and allows for just letters (upper and LOWER case) and 0-9 AND spaces

function NoDoublesPls2()
{
var str=document.getElementById("NoDoubles2");
var regex=/[^a-z 0-9]/gi;
str.value=str.value.replace(regex ,"");
}

// This removed special characters and allows for just letters (upper and LOWER case) and 0-9 AND spaces // The .replace(/\s\s+/g, " ") at the end removes excessive spaces // when I used single quotes, it did not work.

function NoDoublesPls3()
{    var str=document.getElementById("NoDoubles3");
var regex=/[^a-z 0-9]/gi;
str.value=str.value.replace(regex ,"") .replace(/\s\s+/g, " ");
}

::NEXT:: Save #3 as a .js // I called mine NoDoubles.js

::NEXT:: Include your JS into your page

 <script language="JavaScript" src="js/NoDoubles.js"></script>

Include this in your form field:: such as

<INPUT type="text" name="Name"
     onKeyUp="NoDoublesPls3()" onKeyDown="NoDoublesPls3()" id="NoDoubles3"/>

So that it looks like this

<INPUT type="text" name="Name" onKeyUp="NoDoublesPls3()" onKeyDown="NoDoublesPls3()" id="NoDoubles3"/>

This will remove special characters, allow for single spaces and remove extra spaces.


B
Brian Campbell
var string = "The dog      has a long   tail, and it     is RED!";
var replaced = string.replace(/ +/g, " ");

Or if you also want to replace tabs:

var replaced = string.replace(/\s+/g, " ");

using + seems cleaner but it will also replace single spaces with single spaces, a little bit redundant and I'm not sure but it may create performace problems with a much longer text.
I tend to use the shortest, simplest solution that will work, and only worry about that sort of optimization if I know that I need to be matching against a very large string, an at that point I will actually measure different solutions to see which will be faster. It can be hard to predict in advance what will be fastest without testing; for instance, in JavaScript interpreters, some complicated regular expressions will cause you to switch from a fast JIT compiled implementation to a slow interpreted one.
a
ahmetunal
var myregexp = new RegExp(/ {2,}/g);

str = str.replace(myregexp,' ');

T
Toolkit
var text = `xxx  df dfvdfv  df    
                     dfv`.split(/[\s,\t,\r,\n]+/).filter(x=>x).join(' ');

result:

"xxx df dfvdfv df dfv"

Great answer. It removes newline breaks too :)
First, why would you splitjoin when you can do a simple replace? Second, your character set also matches “,”.
x
xSachinx

I know we have to use regex, but during an interview, I was asked to do WITHOUT USING REGEX.

@slightlytyler helped me in coming with the below approach.

const testStr = "I LOVE STACKOVERFLOW LOL"; const removeSpaces = str => { const chars = str.split(''); const nextChars = chars.reduce( (acc, c) => { if (c === ' ') { const lastChar = acc[acc.length - 1]; if (lastChar === ' ') { return acc; } } return [...acc, c]; }, [], ); const nextStr = nextChars.join(''); return nextStr }; console.log(removeSpaces(testStr));


consider: console.log(testStr.split(" ").filter(s => s.length).join(" "))
@dpjanes thanks, mate, do we have any similar chain function to get "DISTINCT" element from an array, we will be using the same array, no new array
hahah I love a challenge. Not fully chained, but one line: console.log(Array.from(new Set("I LOVE STACKOVERFLOW LOL LOVE".split(" ").filter(s => s.length))).join(" "))
The core concept being Array.from(new Set()).
m
minhas23

We can use the following regex explained with the help of sed system command. The similar regex can be used in other languages and platforms.

Add the text into some file say test

manjeet-laptop:Desktop manjeet$ cat test
"The dog      has a long   tail, and it     is RED!"

We can use the following regex to replace all white spaces with single space

manjeet-laptop:Desktop manjeet$ sed 's/ \{1,\}/ /g' test
"The dog has a long tail, and it is RED!"

Hope this serves the purpose


j
jonathan klevin

Try this to replace multiple spaces with a single space.

<script type="text/javascript">
    var myStr = "The dog      has a long   tail, and it     is RED!";
    alert(myStr);  // Output 'The dog      has a long   tail, and it     is RED!'

    var newStr = myStr.replace(/  +/g, ' ');
    alert(newStr);  // Output 'The dog has a long tail, and it is RED!'
</script>

Read more @ Replacing Multiple Spaces with Single Space


A duplicate of the old answer.
j
jackotonye

For more control you can use the replace callback to handle the value.

value = "tags:HUNT  tags:HUNT         tags:HUNT  tags:HUNT"
value.replace(new RegExp(`(?:\\s+)(?:tags)`, 'g'), $1 => ` ${$1.trim()}`)
//"tags:HUNT tags:HUNT tags:HUNT tags:HUNT"

O
Oriol

This script removes any white space (multiple spaces, tabs, returns, etc) between words and trims:

// Trims & replaces any wihtespacing to single space between words
String.prototype.clearExtraSpace = function(){
  var _trimLeft  = /^\s+/,
      _trimRight = /\s+$/,
      _multiple  = /\s+/g;

  return this.replace(_trimLeft, '').replace(_trimRight, '').replace(_multiple, ' ');
};

Why would you replace twice when you can .trim()?
R
Ruwan Epage

' mouse pointer touch '.replace(/^\s+|\s+$|(\s)+/g, "$1") should do the trick!


佚名

my name is Edelcio Junior. If you want to prevent 2 or more white spaces, here it is a good solution to you:

<label">Name</label>
<input type="text" name="YourInputName">

<script>
  var field = document.querySelector('[name="YourInputName"]');

  field.addEventListener('keyup', function (event) {
    var userName = field.value;
    userName = userName.replace(/\s{2,}/g, ' ');
    field.value = userName;
  });
</script>

var field = document.querySelector('[name="YourInputName"]'); field.addEventListener('keyup', function (event) { var userName = field.value; userName = userName.replace(/\s{2,}/g, ' '); field.value = userName; }); Your-title


P
P K Dhee Taal
let nameCorrection = function (str) {
  let strPerfect = str.replace(/\s+/g, " ").trim();
  let strSmall = strPerfect.toLowerCase();
  let arrSmall = strSmall.split(" ");
  let arrCapital = [];
  for (let x of arrSmall.values()) {
    arrCapital.push(x[0].toUpperCase() + x.slice(1));
  }

  let result = arrCapital.join(" ");
  console.log(result);
};
nameCorrection("Pradeep kumar dHital");

the second line of code is helping to replace the multiple white space with just one. and the trim method at the end trims the white space out if there is any at the beginning of the string or at the end
I mean, this is not your notebook. Title-casing is irrelevant to the question.
V
Viraj

def removeblanks(text): return re.sub(r'\s\s+'," ",text) I was working on a large textual data with a lot of duplicate spaces. The above RE worked for me. All the duplicate blank spaces were replaced by a single space.


H
Haji Rahmatullah

Using nodepad++ function, below regex works fine for me,

Find: {1}\K\s+
Replace:leave it empty