From 3709e381ea85d2fa4a4845775be88228af5f67e3 Mon Sep 17 00:00:00 2001
From: Domenic Denicola <d@domenic.me>
Date: Tue, 24 Jan 2017 14:41:18 -0500
Subject: [PATCH] Spec update: add opaque hosts

Follows https://github.com/whatwg/url/pull/185.
---
 scripts/get-latest-platform-tests.js |  2 +-
 src/url-state-machine.js             | 35 ++++++++++++++++++++++++----
 2 files changed, 32 insertions(+), 5 deletions(-)

diff --git a/scripts/get-latest-platform-tests.js b/scripts/get-latest-platform-tests.js
index 63b12c1..20b5190 100644
--- a/scripts/get-latest-platform-tests.js
+++ b/scripts/get-latest-platform-tests.js
@@ -14,7 +14,7 @@ const request = require("request");
 // 1. Go to https://github.com/w3c/web-platform-tests/tree/master/url
 // 2. Press "y" on your keyboard to get a permalink
 // 3. Copy the commit hash
-const commitHash = "165824d1b5eae50004050c77ccbe243990b620fc";
+const commitHash = "eb26d340eb22a83cefe644a7ebe810e06682e326";
 
 const sourceURL = `https://raw.githubusercontent.com/w3c/web-platform-tests/${commitHash}/url/urltestdata.json`;
 const setterSourceURL = `https://raw.githubusercontent.com/w3c/web-platform-tests/${commitHash}/url/setters_tests.json`;
diff --git a/src/url-state-machine.js b/src/url-state-machine.js
index 4534bd6..ac7b46f 100644
--- a/src/url-state-machine.js
+++ b/src/url-state-machine.js
@@ -52,6 +52,10 @@ function isDoubleDot(buffer) {
   return buffer === ".." || buffer === "%2e." || buffer === ".%2e" || buffer === "%2e%2e";
 }
 
+function containsForbiddenHostCodePoint(string) {
+  return string.search(/\u0000|\u0009|\u000A|\u000D|\u0020|#|%|\/|:|\?|@|\[|\\|\]/) !== -1;
+}
+
 function isSpecialScheme(scheme) {
   return specialSchemes[scheme] !== undefined;
 }
@@ -382,7 +386,7 @@ function parseHost(input, isUnicode) {
     return failure;
   }
 
-  if (asciiDomain.search(/\u0000|\u0009|\u000A|\u000D|\u0020|#|%|\/|:|\?|@|\[|\\|\]/) !== -1) {
+  if (containsForbiddenHostCodePoint(asciiDomain)) {
     return failure;
   }
 
@@ -394,6 +398,23 @@ function parseHost(input, isUnicode) {
   return isUnicode ? tr46.toUnicode(asciiDomain, false).domain : asciiDomain;
 }
 
+function parseURLHost(input, isSpecialArg) {
+  if (isSpecialArg) {
+    return parseHost(input);
+  }
+
+  if (containsForbiddenHostCodePoint(input)) {
+    return failure;
+  }
+
+  let output = "";
+  const decoded = punycode.ucs2.decode(input);
+  for (let i = 0; i < decoded.length; ++i) {
+    output += encodeChar(decoded[i], isSimpleEncode);
+  }
+  return output;
+}
+
 function findLongestZeroSequence(arr) {
   let maxIdx = null;
   let maxLen = 1; // only find elements > 1
@@ -748,6 +769,10 @@ URLStateMachine.prototype["parse authority"] = function parseAuthority(c, cStr)
     this.buffer = "";
   } else if (isNaN(c) || c === p("/") || c === p("?") || c === p("#") ||
              (isSpecial(this.url) && c === p("\\"))) {
+    if (this.atFlag && this.buffer === "") {
+      this.parseError = true;
+      return failure;
+    }
     this.pointer -= countSymbols(this.buffer) + 1;
     this.buffer = "";
     this.state = "host";
@@ -761,11 +786,12 @@ URLStateMachine.prototype["parse authority"] = function parseAuthority(c, cStr)
 URLStateMachine.prototype["parse hostname"] =
 URLStateMachine.prototype["parse host"] = function parseHostName(c, cStr) {
   if (c === p(":") && !this.arrFlag) {
-    if (isSpecial(this.url) && this.buffer === "") {
+    if (this.buffer === "") {
+      this.parseError = true;
       return failure;
     }
 
-    const host = parseHost(this.buffer);
+    const host = parseURLHost(this.buffer, isSpecial(this.url));
     if (host === failure) {
       return failure;
     }
@@ -780,10 +806,11 @@ URLStateMachine.prototype["parse host"] = function parseHostName(c, cStr) {
              (isSpecial(this.url) && c === p("\\"))) {
     --this.pointer;
     if (isSpecial(this.url) && this.buffer === "") {
+      this.parseError = true;
       return failure;
     }
 
-    const host = parseHost(this.buffer);
+    const host = parseURLHost(this.buffer, isSpecial(this.url));
     if (host === failure) {
       return failure;
     }