I`m trying to extract the src URL/path without the quotes, only in the case it is an image:
- src="/path/image.png" // should capture => /path/image.png
- src="/path/image.bmp" // should capture => /path/image.bmp
- src="/path/image.jpg" // should capture => /path/image.jpg
- src="https://www.site1.com" // should NOT capture
So far I have /src="(.*)"/g, but that obviously captures both, I have been looking at look behind and look ahead but just can`t put it together.
You can use a capture group, and you should prevent crossing the " using a negated character class.
If you want to match either href or src
\b(?:href|src)="([^\s"]*\.(?:png|jpg|bmp))"
Explanation
\b A word boundary to prevent a partial word match(?:href|src)=" match either href= or src=( Capture group 1
[^\s"]* Match optional chars other than a whitespace char or "\.(?:png|jpg|bmp) Match one of .png .jpg .bmp) Close group 1" Match literallyconst regex = /\b(?:href|src)="([^\s"]*\.(?:png|jpg|bmp))"/;
[
'src="/path/image.png" test "',
'src="/path/image.bmp"',
'src="/path/image.jpg"',
'src="https://www.site1.com"',
'href="image.png"'
].forEach(s => {
const m = s.match(regex);
if (m) {
console.log(m[1]);
}
})
Try /src="(.*(?:jpg|bmp|png))"/g
You'll need to enter in the list of extensions you consider valid images
[jpg|bmp|png] matches the same characters as [bgjmnp|], i.e. you cannot have string alternatives inside a class, it's only going to match the unique characters inside it. You probably meant to write something like (?:jpg|bmp|png) which is an example of actual alternation.If you want it to be a bit more fool proof you can use look behinds and look aheads. Expand the extension list png|bmp|jpg to test for more extensions.
/(?<=src=").*(png|bmp|jpg)(?=")/g
Try this src="(.*image.*)"
src="https://www.site1.com"might return an image if you request that link in browser.