Combine two JavaScript regular expressions into one

Question

Background:

I'm using director.js to match routes, and using regex for different parameters.

Problem:

Need to come up with a regex for a parameter which matches regex 1 while not matching regex 2 (I know it's bad design but there're various reasons which I have to do it this way)

[._a-zA-Z0-9-%!'*]+
[a-z0-9][a-zA-Z0-9]{3}(0[a-zA-Z0-9]{2}|[a-zA-Z0-9]00)[a-zA-Z0-9]{8}([a-zA-Z0-9]{3})?

What's the best way of doing this?

Update:

Thanks to gyre. I think something similiar to (?![a-z0-9][a-zA-Z0-9]{3}(0[a-zA-Z0-9]{2}|[a-zA-Z0-9]00)[a-zA-Z0-9]{8}([a-zA-Z0-9]{3})?)(?:[._a-zA-Z0-9-%!()'*]+) is what I want, but it doesn't work when I tried in my application, using this regex in https://regex101.com/, it can still find a match for string "00Bxx0000025e1UEA" --> Bxx0000025e1UEA which I think the library was confused somehow. Is there anyway to update the regex to not find a match for "00Bxx0000025e1UEA" at all?

I have tried to add ^ and $ for the regex: (?!^[a-z0-9][a-zA-Z0-9]{3}(0[a-zA-Z0-9]{2}|[a-zA-Z0-9]00)[a-zA-Z0-9]{8}([a-zA-Z0-9]{3})?$)(?:[._a-zA-Z0-9-%!()'*]+) but that's still incorrect.

What about this? ^(?![a-z0-9][a-zA-Z0-9]{3}(0[a-zA-Z0-9]{2}|[a-zA-Z0-9]00)[a-zA-Z0-9]{8}([a-zA-Z0-9]{3})?)[._a-zA-Z0-9-%!()'*]+ — trincot
– trincot, Commented Mar 2, 2017 at 20:00
This will exclude strings start with the second pattern but have something afterwards. — daily
– daily, Commented Mar 2, 2017 at 20:44
^(?![a-z0-9][a-zA-Z0-9]{3}(0[a-zA-Z0-9]{2}|[a-zA-Z0-9]00)[a-zA-Z0-9]{8}([a-zA-Z0-9]{3})?)[._a-zA-Z0-9-%!()'*]+$ this? Does't work with this string 00Bxx0000025gVvEAIddfdafdsfsf — daily
– daily, Commented Mar 2, 2017 at 20:47
Yes, I added a link. If the exclusion must not work as exclusion if something else follows that pattern, then also include a $ at the end of the exclusion pattern, just after the question mark. — trincot
– trincot, Commented Mar 2, 2017 at 20:48

gyre · Accepted Answer · 2017-03-02 20:01:46Z

2

You are essentially looking for this pattern, which uses a negative lookahead to ensure that the match for <first> is not also a match for <second>:

/(?! <second> )(?: <first> )/

var first = /[._a-zA-Z0-9-%!()'*]+/
var second = /[a-z0-9][a-zA-Z0-9]{3}(0[a-zA-Z0-9]{2}|[a-zA-Z0-9]00)[a-zA-Z0-9]{8}([a-zA-Z0-9]{3})?/

var third = new RegExp('(?!' + second.source + ')(?:' + first.source + ')')

console.log(third)

answered Mar 2, 2017 at 20:01

gyre

16.8k4 gold badges40 silver badges47 bronze badges

Sign up to request clarification or add additional context in comments.

1 Comment

Emperor Krauser Over a year ago

Voted this for code maintainability. You can separate parts of the regex by variables then combine it to the constructor RegExp.

mgul · Accepted Answer · 2017-03-02 20:00:57Z

0

Try this:

var str = "the string to test"
var re1 = "your first regex, [._a-zA-Z0-9-%!()'*]+"
var re2 = "your second regex, [a-z0-9][a-zA-Z0-9]{3}(0[a-zA-Z0-9]{2}|[a-zA-Z0-9]00)[a-zA-Z0-9]{8}([a-zA-Z0-9]{3})?"

if(str.search(re1) != -1 && str.search(r2) == -1) {
    // your action
}

search will return -1 if there is no match

answered Mar 2, 2017 at 20:00

mgul

74210 silver badges28 bronze badges

Comments

Nitesh · Accepted Answer · 2017-03-02 20:01:39Z

0

If you need to match regex1 and not match regex2 then why can't you check as,

var isTrue = regex1.test(str) && !regex2.test(str)

answered Mar 2, 2017 at 20:01

Nitesh

1,5151 gold badge13 silver badges21 bronze badges

1 Comment

daily Over a year ago

thx, I think I didn't make myself clear enough in the original post, I need a single regex to solve this.

trincot · Accepted Answer · 2017-03-02 21:57:40Z

You could do it in one regular expression like this:

^(?![a-z0-9][a-zA-Z0-9]{3}(0[a-zA-Z0-9]{2}|[a-zA-Z0-9]00)[a-zA-Z0-9]{8}([a-zA-Z0-9]{3})?$)[._a-zA-Z0-9-%!()'*]+$

The part that you want to reject comes first between (?! ), which means it looks ahead and will fail the match if that part matches.

The additional ^ and $ will make sure the match is done at the start of the string up until the very end of that string.

The pattern that you do want to match is placed at the end, again with an additional $ to prevent that other characters follow the pattern.

See it on regex101.com

Depending on your expectations, you might want to remove the first $. Remove it, if you also want to fail strings that have the exclusion pattern with some other characters following it. If however such suffixed characters make the string potentially acceptable, you should keep that $.

Collectives™ on Stack Overflow

Combine two JavaScript regular expressions into one

Background:

Problem:

Update:

4 Answers 4

1 Comment

Comments

1 Comment

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

Background:

Problem:

Update:

4 Answers 4

1 Comment

Comments

1 Comment

Comments

Your Answer

Sign up or log in

Post as a guest

Related