url: remove \t \n \r in url.parse() similar to WHATWG

WHATWG URL removes tab, new line, and carraige return characters before
processing URL strings. To narrow the differences between WHATWG URL and
url.parse(), and thus reduce opportunities for host spoofing etc. due to
differences between the two APIs, let's do the same with url.parse().

PR-URL: #45116
Reviewed-By: James M Snell <[email protected]>
Reviewed-By: Yagiz Nizipli <[email protected]>
Reviewed-By: Tobias Nießen <[email protected]>
Reviewed-By: Antoine du Hamel <[email protected]>

Author: Trott

lib/url.js

@@ -319,6 +319,10 @@ Url.prototype.parse = function parse(url, parseQueryString, slashesDenoteHost) {
         case CHAR_TAB:
         case CHAR_LINE_FEED:
         case CHAR_CARRIAGE_RETURN:
+          // WHATWG URL removes tabs, newlines, and carriage returns. Let's do that too.
+          rest = rest.slice(0, i) + rest.slice(i + 1);
+          i -= 1;
+          break;
         case CHAR_SPACE:
         case CHAR_DOUBLE_QUOTE:
         case CHAR_PERCENT:

test/parallel/test-url-parse-format.js

@@ -853,16 +853,16 @@ const parseTests = {
   'http://a\r" \t\n<\'b:b@c\r\nd/e?f': {
     protocol: 'http:',
     slashes: true,
-    auth: 'a\r" \t\n<\'b:b',
-    host: 'c',
+    auth: 'a" <\'b:b',
+    host: 'cd',
     port: null,
-    hostname: 'c',
+    hostname: 'cd',
     hash: null,
     search: '?f',
     query: 'f',
-    pathname: '%0D%0Ad/e',
-    path: '%0D%0Ad/e?f',
-    href: 'http://a%0D%22%20%09%0A%3C\'b:b@c/%0D%0Ad/e?f'
+    pathname: '/e',
+    path: '/e?f',
+    href: 'http://a%22%20%3C\'b:b@cd/e?f'
   },
 
   // Git urls used by npm
@@ -1023,7 +1023,7 @@ for (const u in parseTests) {
   assert.deepStrictEqual(
     actual,
     expected,
-    `expected ${inspect(expected)}, got ${inspect(actual)}`
+    `parsing ${u} and expected ${inspect(expected)} but got ${inspect(actual)}`
   );
   assert.deepStrictEqual(
     spaced,