Issue 7296 - Implement lightweight URL parsing

test/url.js

 /*
  * This file is part of Adblock Plus <https://adblockplus.org/>,
  * Copyright (C) 2006-present eyeo GmbH
  *
  * Adblock Plus is free software: you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 3 as
  * published by the Free Software Foundation.
  *
  * Adblock Plus is distributed in the hope that it will be useful,
  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  * GNU General Public License for more details.
  *
  * You should have received a copy of the GNU General Public License
  * along with Adblock Plus.  If not, see <http://www.gnu.org/licenses/>.
  */
 
 "use strict";
 
 // Only starting NodeJS 10 that URL is in the global space.
 const {URL} = require("url");
 const {createSandbox} = require("./_common");
 
 const publicSuffixes = require("../data/publicSuffixList.json");
 
+let parseURL = null;
 let normalizeHostname = null;
 let domainSuffixes = null;
 let isThirdParty = null;
 let getBaseDomain = null;
 
 exports.setUp = function(callback)
 {
   let sandboxedRequire = createSandbox({
     extraExports: {
       domain: ["getBaseDomain"]
     }
   });
   (
-    {normalizeHostname, domainSuffixes, isThirdParty,
+    {parseURL, normalizeHostname, domainSuffixes, isThirdParty,
      getBaseDomain} = sandboxedRequire("../lib/url")
   );
 
   callback();
 };
 
 function hostnameToURL(hostname)
 {
   return new URL("http://" + hostname);
 }
 
+function testURLParsing(test, url)
+{
+  // The function expects a normalized URL.
+  url = new URL(url).href;
+
+  let urlInfo = parseURL(url);
+
+  // We need to ensure only that our implementation matches that of the URL
+  // object.
+  let urlObject = new URL(url);
+
+  test.equal(urlInfo.href, urlObject.href);
+  test.equal(urlInfo.protocol, urlObject.protocol);
+  test.equal(urlInfo.hostname, urlObject.hostname);
+
+  test.equal(urlInfo.toString(), urlObject.toString());
+  test.equal(String(urlInfo), String(urlObject));
+  test.equal(urlInfo + "", urlObject + "");
+}
+
 function testThirdParty(test, requestHostname, documentHostname, expected,
                         message)
 {
   test.equal(
     isThirdParty(
       hostnameToURL(requestHostname),
 
       // Chrome's URL object normalizes IP addresses. So some test
       // will fail if we don't normalize the document host as well.
       hostnameToURL(documentHostname).hostname
     ),
     expected,
     message
   );
 }
 
+exports.testParseURL = function(test)
+{
+  testURLParsing(test, "https://example.com");
+  testURLParsing(test, "https://example.com/");
+  testURLParsing(test, "https://example.com/foo");
+  testURLParsing(test, "https://example.com/foo/bar");
+  testURLParsing(
+    test,
+    "https://example.com/foo/bar?https://random/foo/bar"
+  );
+
+  testURLParsing(test, "https://example.com:8080");
+  testURLParsing(test, "https://example.com:8080/");
+  testURLParsing(test, "https://example.com:8080/foo");
+  testURLParsing(test, "https://example.com:8080/foo/bar");
+  testURLParsing(
+    test,
+    "https://example.com:8080/foo/bar?https://random/foo/bar"
+  );
+
+  testURLParsing(test, "http://localhost");
+  testURLParsing(test, "http://localhost/");
+  testURLParsing(test, "http://localhost/foo");
+  testURLParsing(test, "http://localhost/foo/bar");
+  testURLParsing(
+    test,
+    "http://localhost/foo/bar?https://random/foo/bar"
+  );
+
+  testURLParsing(test, "http://192.168.1.1");
+  testURLParsing(test, "http://192.168.1.1/");
+  testURLParsing(test, "http://192.168.1.1/foo");
+  testURLParsing(test, "http://192.168.1.1/foo/bar");
+  testURLParsing(
+    test,
+    "http://192.168.1.1/foo/bar?https://random/foo/bar"
+  );
+
+  testURLParsing(test, "https://user@example.com");
+  testURLParsing(test, "https://user@example.com/");
+  testURLParsing(test, "https://user@example.com/foo");
+  testURLParsing(test, "https://user@example.com/foo/bar");
+  testURLParsing(
+    test,
+    "https://user@example.com/foo/bar?https://random/foo/bar"
+  );
+
+  testURLParsing(test, "https://user@example.com:8080");
+  testURLParsing(test, "https://user@example.com:8080/");
+  testURLParsing(test, "https://user@example.com:8080/foo");
+  testURLParsing(test, "https://user@example.com:8080/foo/bar");
+  testURLParsing(
+    test,
+    "https://user@example.com:8080/foo/bar?https://random/foo/bar"
+  );
+
+  testURLParsing(test, "https://user:pass@example.com");
+  testURLParsing(test, "https://user:pass@example.com/");
+  testURLParsing(test, "https://user:pass@example.com/foo");
+  testURLParsing(test, "https://user:pass@example.com/foo/bar");
+  testURLParsing(
+    test,
+    "https://user:pass@example.com/foo/bar?https://random/foo/bar"
+  );
+
+  testURLParsing(test, "https://user:pass@example.com:8080");
+  testURLParsing(test, "https://user:pass@example.com:8080/");
+  testURLParsing(test, "https://user:pass@example.com:8080/foo");
+  testURLParsing(test, "https://user:pass@example.com:8080/foo/bar");
+  testURLParsing(
+    test,
+    "https://user:pass@example.com:8080/foo/bar?https://random/foo/bar"
+  );
+
+  testURLParsing(test, "https://us%40er:pa%40ss@example.com");
+  testURLParsing(test, "https://us%40er:pa%40ss@example.com/");
+  testURLParsing(test, "https://us%40er:pa%40ss@example.com/foo");
+  testURLParsing(test, "https://us%40er:pa%40ss@example.com/foo/bar");
+  testURLParsing(
+    test,
+    "https://us%40er:pa%40ss@example.com/foo/bar?https://random/foo/bar"
+  );
+
+  testURLParsing(test, "https://us%40er:pa%40ss@example.com:8080");
+  testURLParsing(test, "https://us%40er:pa%40ss@example.com:8080/");
+  testURLParsing(test, "https://us%40er:pa%40ss@example.com:8080/foo");
+  testURLParsing(test, "https://us%40er:pa%40ss@example.com:8080/foo/bar");
+  testURLParsing(
+    test,
+    "https://us%40er:pa%40ss@example.com:8080/foo/bar?https://random/foo/bar"
+  );
+
+  testURLParsing(test, "ftp://user:pass@example.com:8021");
+  testURLParsing(test, "ftp://user:pass@example.com:8021/");
+  testURLParsing(test, "ftp://user:pass@example.com:8021/foo");
+  testURLParsing(test, "ftp://user:pass@example.com:8021/foo/bar");
+
+  testURLParsing(test, "about:blank");
+  testURLParsing(test, "chrome://extensions");
+  testURLParsing(
+    test,
+    "chrome-extension://bhignfpcigccnlfapldlodmhlidjaion/options.html"
+  );
+  testURLParsing(test, "mailto:john.doe@mail.example.com");
+
+  testURLParsing(test, "news:newsgroup");
+  testURLParsing(test, "news:message-id");
+  testURLParsing(test, "nntp://example.com:8119/newsgroup");
+  testURLParsing(test, "nntp://example.com:8119/message-id");
+
+  testURLParsing(test, "data:,");
+  testURLParsing(
+    test,
+    "data:text/vnd-example+xyz;foo=bar;base64,R0lGODdh"
+  );
+  testURLParsing(
+    test,
+    "data:text/plain;charset=UTF-8;page=21,the%20data:1234,5678"
+  );
+
+  testURLParsing(test, "javascript:");
+  testURLParsing(test, "javascript:alert();");
+  testURLParsing(test, "javascript:foo/bar/");
+  testURLParsing(test, "javascript://foo/bar/");
+
+  testURLParsing(test, "file:///dev/random");
+
+  test.done();
+};
+
 exports.testNormalizeHostname = function(test)
 {
   test.equal(normalizeHostname("example.com"), "example.com");
   test.equal(normalizeHostname("example.com."), "example.com");
   test.equal(normalizeHostname("example.com.."), "example.com");
   test.equal(normalizeHostname("example.com..."), "example.com");
 
   test.equal(normalizeHostname("Example.com"), "example.com");
   test.equal(normalizeHostname("ExaMple.Com"), "example.com");
   test.equal(normalizeHostname("ExaMple.Com.."), "example.com");
@@ skipping 211 matching lines @@
     "us-east-1.amazonaws.com"
   );
   test.equal(getBaseDomain("example.amazonaws.com"), "amazonaws.com");
   test.equal(getBaseDomain("amazonaws.com"), "amazonaws.com");
 
   // Edge case.
   test.equal(getBaseDomain(""), "");
 
   test.done();
 };