I'm having issues with the following regex working in Safari and so am having to change how I approach this regex. I need the following regex to only select the text between: cp1= and |.
For example if I had this string:
firstName=John|lastName=Smith|cp1=/en-gb/|age=30
I'd only want to get /en-gb/ and nothing else.
The following JS works below in all browsers i've tested except for Safari:
// if myString = firstName=John|lastName=Smith|cp1=/en-gb/|age=30
let currentCP1Value = myString.match(/(?<=cp1=)(.*)(?=\|)|(?<=cp1=).*(?<!a)$/); // outputs /en-gb/
This is due to the positive/negative lookbehind's used in the regex. What would be the alternative to this to get this working in Safari?
About the pattern you tried:
This part of the pattern (.*)(?=\|) Matches until the end of the line and then backtracks until it can assert a | to the right, which can match too much if there are multiple occurrences of | following.
This part of the pattern |(?<=cp1=).*(?<!a)$ is the second alternative, which also matches until the end of the line and asserts that it does not end with an a char, which seems out of place as you want to match /en-gb/ ending on /
You don't need any lookarounds to get /en-gb/, you can use a capture group as you are already using in your pattern.
Use a negated character class [^|]* to match any character except | to not cross matching the next one.
\bcp1=([^|]*)
const myString = "firstName=John|lastName=Smith|cp1=/en-gb/|age=30"
let currentCP1Value = myString.match(/\bcp1=([^|]*)/);
if (currentCP1Value) {
console.log(currentCP1Value[1]);
}Why use regex at all? If your data doesn't contain text that can be interpreted as URL encoded segments, you can simply replace | with & and now you have a standard query string that JS has built-in parsing for through URLSearchParams:
const parsed = new URLSearchParams(input.replaceAll(`|`, `&`));
const cp = parsed.get(`cp1`);
// and many utility functions to perform further work
and you're done.