Improve prevent-fetch — return original url in response

[AdamWr]

• Return original URL
• Improve Response.type - Improve prevent-xhr — blob response, randomized responseText with given length #206 (comment) (first part of the comment)
• Fix issue with : in URL - Improve prevent-fetch — return original url in response #216 (comment)

---

At the moment, if prevent-fetch is used then Response.url returns empty string, but sometimes websites are checking Response.url to detect ad blockers.

Steps to reproduce

1. Add this rule to user rules:

example.org#%#//scriptlet('prevent-fetch', 'pagead2.googlesyndication.com')

2. Go to - https://example.org/
3. Run this script:

(async () => {
  const ads = "https://pagead2.googlesyndication.com/pagead/js/adsbygoogle.js";
  const getAds = await fetch(ads);

  if (getAds.status === 200 && getAds.url === ads) {
    let video =
      '<iframe width="560" height="315" src="https://www.youtube.com/embed/Fy2rtb95QhY" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe>';
    let createDiv = document.createElement("div");
    document.body.appendChild(createDiv);
    createDiv.innerHTML = video;
  } else {
    alert("Request blocked - AdBlock");
  }
})();

Alert is displayed because getAds.url returns '', if request is not blocked then video player should be loaded.

What if would add fetchData.url as another argument in:

Scriptlets/src/scriptlets/prevent-fetch.js

Line 133 in 2a72d26

return noopPromiseResolve(strResponseBody);

something like:

noopPromiseResolve(strResponseBody, fetchData.url); 

and add Object.defineProperty(response, 'url', { value: url, writable: false }); to:

Scriptlets/src/helpers/noop.js

Lines 59 to 70 in af1d5ad

	export const noopPromiseResolve = (responseBody = '{}') => {
	if (typeof Response === 'undefined') {
	return;
	}
	// eslint-disable-next-line compat/compat
	const response = new Response(responseBody, {
	status: 200,
	statusText: 'OK',
	});
	// eslint-disable-next-line compat/compat, consistent-return
	return Promise.resolve(response);
	};

For example:

export const noopPromiseResolve = (responseBody = '{}', url) => { 
     if (typeof Response === 'undefined') { 
         return; 
     } 
     // eslint-disable-next-line compat/compat 
     const response = new Response(responseBody, { 
         status: 200, 
         statusText: 'OK', 
     }); 
     Object.defineProperty(response, 'url', { value: url, writable: false });
     // eslint-disable-next-line compat/compat, consistent-return 
     return Promise.resolve(response); 
 };

It should causes that original url will be returned.

[AdamWr]

One more thing.
If there is : in URL which we want to "block" then rule doesn't work correctly.
Steps to reproduce

1. Add this rule:

example.org#%#//scriptlet("prevent-fetch", "/^https?:\/\/example\.org/")

2. Go to - https://example.org/
3. Run this script in console

fetch('https://example.org/', { method: 'GET' });

Request should be "blocked", but it isn't.
It works fine if I replace : with .:

example.org#%#//scriptlet("prevent-fetch", "/^https?.\/\/example\.org/")

As far as I understand, problem is in parseMatchProps function

Scriptlets/src/scriptlets/prevent-fetch.js

Line 114 in 3baf8d9

const parsedData = parseMatchProps(propsToMatch);

Scriptlets/src/helpers/fetch-utils.js

Lines 67 to 86 in 3baf8d9

	export const parseMatchProps = (propsToMatchStr) => {
	const PROPS_DIVIDER = ' ';
	const PAIRS_MARKER = ':';
	const propsObj = {};
	const props = propsToMatchStr.split(PROPS_DIVIDER);
	props.forEach((prop) => {
	const dividerInd = prop.indexOf(PAIRS_MARKER);
	if (dividerInd === -1) {
	propsObj.url = prop;
	} else {
	const key = prop.slice(0, dividerInd);
	const value = prop.slice(dividerInd + 1);
	propsObj[key] = value;
	}
	});
	return propsObj;
	};

because it splits /^https?:\/\/example\.org/ into /^https? and \/\/example\.org/, so it doesn't work like it should.
I suppose that it's for init options, like method or headers, but currently it also splits URL.

I guess that the same issue is with prevent-xhr.