how to match a string of words using regex javascript

Question

Am trying to find a regex expression for this result:

string => should be matched (a single word or set of words at the beginning or the ending)
 string  => should be matched (a single word or set of words in the middle)
{{string}} -- should not be matched (a single word or set of words surrounded by two "{}" should not be matched)

am using this regex in this function :

text = text.replace(RegExp("([^{]{2})[^(\d:)]" + aTags[index].textContent + "\w* 
([^}]{2})", 'i'), "{{" + index + ":" + aTags[index].textContent + "}}");

the function should find the textContent of an 'a' tag in a 'text' string and replace it by adding a digit and ':' to the beginning of the textContent so that the result should be something like this :

some text => will became => {{1:some text}}

Try `text.replace(RegExp("{{.*?}}|(" + aTags[index].textContent + ")\\w*", "i"), function($0, $1) { return $1 ? "{{" + index + ":" + $1 + "}}" : $0;} )`, see http://jsfiddle.net/4bpkzhtd/1/. — Wiktor Stribiżew, Jul 01 '18 at 19:57
hi, thank you for your comment , the problem with your answer is that it match {{string}} witch i don't want to match it i need to skip it, this one : `text = text.replace(RegExp("{{[^}]+}}|(" + aTags[index].textContent + ")", 'gmi'), function (match, $1) { if($1) return $1 ? {{${index}:${$1}}} : match; });` worked on regex101 but in javascript i get this error `Invalid regular expression: /{{[^}]+}}|(*SKIP what's to avoid approach)/: Nothing to repeat` — , Jul 02 '18 at 12:11
You cannot "skip" anything in a **JavaScript** regex , it is not PCRE. You should match and capture it and then add some programming logic around it - see my snippet - doesn't it produce the results you need? In my snippet, I match `{{...}}`, but I do *skip* it in the code. — Wiktor Stribiżew, Jul 02 '18 at 12:13
I strongly advise to first understand the basics of JS regex before trying to meddle with the regex testers, and be very careful with options you select there. — Wiktor Stribiżew, Jul 02 '18 at 12:17
yes you right, i modified my code like this : `text = text.replace(RegExp("{{[^}]+}}|(" + aTags[index].textContent + ")", 'gmi'), function (match, $1) { return $1 ? `{{${index}:${$1}}}` : match; });` it worked now , thanks — , Jul 02 '18 at 12:17
Very frustrating to see it requires two people to explain things in every detail up to a point of providing a copy&pasteable solution. That essentially is a free coding service. — wp78de, Jul 02 '18 at 17:46
@wp78de Sorry, I did not even notice your answer (I think it is due to the one deleted that appeared on top). — Wiktor Stribiżew, Jul 03 '18 at 07:26

score 2 · Answer 1 · answered Jul 02 '18 at 00:29

2

We can apply the good old *SKIP what's to avoid approach and throw everything that does not need to be replaced in the full match and capture the desired output in group 1:

{{[^}]+}}|(string)

To make this work effectively in JavaScript we have to use a .replace callback function:

const regex = /{{[^}]+}}|(string)/gm;
const str = `string 
 string  
{{string}}`;

var index = 1; //this is your index var and is somehow set from outside
const result = str.replace(regex, function(m, group1) {
    if (group1) return `{{${index}:${group1}}}`;
    else return m;
});
console.log('Substitution result: ', result);

I had pseudo-coded this a bit since I cannot know where index and aTags[index].textContent is coming from. Adjust as needed.

answered Jul 02 '18 at 00:29

wp78de

18,207
7
43
71

hi , thank you for your answer, i modified my code to match you answer but i get this error : Invalid regular expression: /{{[^}]+}}|(*SKIP what's to avoid approach)/: Nothing to repea , here is the function : text = text.replace(RegExp("{{[^}]+}}|(" + aTags[index].textContent + ")", 'gmi'), function (match, $1) { if($1) return $1 ? `{{${index}:${$1}}}` : match; }); – Jul 02 '18 at 12:05
@j.sell sorry, you totally misinterpreted the pseudo-code. Fortunately, you are already covered. – wp78de Jul 02 '18 at 17:12

Wiktor Stribiżew · Accepted Answer · 2018-07-03T07:24:48.417

You cannot use PCRE verbs like (*SKIP)(*F) in a JavaScript regex, i.e. you cannot skip a matched portion of text with the regex means only. In JavaScript, you may match and capture a part of the string you want to later analyze in the replacement callback method (JS String#replace accepts a callback as the replacement argument).

So, in your case the solution will look like

text = text.replace(RegExp("{{.*?}}|(" + aTags[index].textContent + ")", "gi"),
    function ($0, $1) { 
        return $1 ? "{{" + index + ":" + $1 + "}}" : $0; 
    }
);

I understand the aTags[index].textContent value is alphanumeric, else, consider escaping it for use in a regex pattern.

The pattern will match a {{...}} substring having no } inside (with {{.*?}}) or (|) it will match and capture the text content ((aTags[index].textContent)) into Group 1. When you get a match, you need to pass 2 arguments to the callback, the whole match and Group 1 value. If Group 1 is not empty, you perform string manipulations, else, just insert the match back.

thank you for your answer, it worked finally, i need to read more on regex to understand them better — , Jul 02 '18 at 12:37

how to match a string of words using regex javascript

2 Answers2