From 190cf5c730258ca9d2f94d6b33eee5e01aca3d0c Mon Sep 17 00:00:00 2001
From: Ms2ger <Ms2ger@gmail.com>
Date: Fri, 27 Apr 2018 11:49:37 +0200
Subject: [PATCH] Extract list of structurally invalid tags from
 invalid-tags.js.

---
 harness/testIntl.js                           | 60 +++++++++++++++++++
 .../Intl/getCanonicalLocales/invalid-tags.js  | 53 +---------------
 2 files changed, 61 insertions(+), 52 deletions(-)

diff --git a/harness/testIntl.js b/harness/testIntl.js
index 690791d7d0..3b9e62d7ce 100644
--- a/harness/testIntl.js
+++ b/harness/testIntl.js
@@ -162,6 +162,66 @@ function getLocaleSupportInfo(Constructor) {
 }
 
 
+/**
+ * Returns an array of strings for which IsStructurallyValidLanguageTag() returns false
+ */
+function getInvalidLanguageTags() {
+  var invalidLanguageTags = [
+    "", // empty tag
+    "i", // singleton alone
+    "x", // private use without subtag
+    "u", // extension singleton in first place
+    "419", // region code in first place
+    "u-nu-latn-cu-bob", // extension sequence without language
+    "hans-cmn-cn", // "hans" could theoretically be a 4-letter language code,
+                   // but those can't be followed by extlang codes.
+    "cmn-hans-cn-u-u", // duplicate singleton
+    "cmn-hans-cn-t-u-ca-u", // duplicate singleton
+    "de-gregory-gregory", // duplicate variant
+    "*", // language range
+    "de-*", // language range
+    "中文", // non-ASCII letters
+    "en-ß", // non-ASCII letters
+    "ıd", // non-ASCII letters
+
+    // underscores in different parts of the language tag
+    "de_DE",
+    "DE_de",
+    "cmn_Hans",
+    "cmn-hans_cn",
+    "es_419",
+    "es-419-u-nu-latn-cu_bob",
+    "i_klingon",
+    "cmn-hans-cn-t-ca-u-ca-x_t-u",
+    "enochian_enochian",
+    "de-gregory_u-ca-gregory",
+
+    "en\u0000", // null-terminator sequence
+    " en", // leading whitespace
+    "en ", // trailing whitespace
+    "it-IT-Latn", // country before script tag
+    "de-u", // incomplete Unicode extension sequences
+    "de-u-",
+    "de-u-ca-",
+    "de-u-ca-gregory-",
+    "si-x", // incomplete private-use tags
+    "x-",
+    "x-y-",
+  ];
+
+  // make sure the data above is correct
+  for (var i = 0; i < invalidLanguageTags.length; ++i) {
+    var invalidTag = invalidLanguageTags[i];
+    assert(
+      !isCanonicalizedStructurallyValidLanguageTag(invalidTag),
+      "Test data \"" + invalidTag + "\" is a canonicalized and structurally valid language tag."
+    );
+  }
+
+  return invalidLanguageTags;
+}
+
+
 /**
  * @description Tests whether locale is a String value representing a
  * structurally valid and canonicalized BCP 47 language tag, as defined in
diff --git a/test/intl402/Intl/getCanonicalLocales/invalid-tags.js b/test/intl402/Intl/getCanonicalLocales/invalid-tags.js
index e218f21f1b..b131603a50 100644
--- a/test/intl402/Intl/getCanonicalLocales/invalid-tags.js
+++ b/test/intl402/Intl/getCanonicalLocales/invalid-tags.js
@@ -21,58 +21,7 @@ info: |
 includes: [testIntl.js]
 ---*/
 
-var invalidLanguageTags = [
-  "", // empty tag
-  "i", // singleton alone
-  "x", // private use without subtag
-  "u", // extension singleton in first place
-  "419", // region code in first place
-  "u-nu-latn-cu-bob", // extension sequence without language
-  "hans-cmn-cn", // "hans" could theoretically be a 4-letter language code,
-                 // but those can't be followed by extlang codes.
-  "cmn-hans-cn-u-u", // duplicate singleton
-  "cmn-hans-cn-t-u-ca-u", // duplicate singleton
-  "de-gregory-gregory", // duplicate variant
-  "*", // language range
-  "de-*", // language range
-  "中文", // non-ASCII letters
-  "en-ß", // non-ASCII letters
-  "ıd", // non-ASCII letters
-
-  // underscores in different parts of the language tag
-  "de_DE",
-  "DE_de",
-  "cmn_Hans",
-  "cmn-hans_cn",
-  "es_419",
-  "es-419-u-nu-latn-cu_bob",
-  "i_klingon",
-  "cmn-hans-cn-t-ca-u-ca-x_t-u",
-  "enochian_enochian",
-  "de-gregory_u-ca-gregory",
-
-  "en\u0000", // null-terminator sequence
-  " en", // leading whitespace
-  "en ", // trailing whitespace
-  "it-IT-Latn", // country before script tag
-  "de-u", // incomplete Unicode extension sequences
-  "de-u-",
-  "de-u-ca-",
-  "de-u-ca-gregory-",
-  "si-x", // incomplete private-use tags
-  "x-",
-  "x-y-",
-];
-
-// make sure the data above is correct
-for (var i = 0; i < invalidLanguageTags.length; ++i) {
-  var invalidTag = invalidLanguageTags[i];
-  assert(
-    !isCanonicalizedStructurallyValidLanguageTag(invalidTag),
-    "Test data \"" + invalidTag + "\" is a canonicalized and structurally valid language tag."
-  );
-}
-
+var invalidLanguageTags = getInvalidLanguageTags();
 for (var i = 0; i < invalidLanguageTags.length; ++i) {
   var invalidTag = invalidLanguageTags[i];
   assert.throws(RangeError, function() {