Update language tag canonicalisation per UTS35

Main changes:
- "cmn" is now canonicalised to "zh"
- "und-x-" is prepended before grandfathered tags without modern replacements
- "und-" is prepended before privateuse-only language tags

test/intl402/Intl/getCanonicalLocales/canonicalized-tags.js

@@ -30,12 +30,12 @@ var canonicalizedTags = {
   "de": "de",
   "DE-de": "de-DE",
   "de-DE": "de-DE",
-  "cmn": "cmn",
+  "cmn": "zh",
-  "CMN-hANS": "cmn-Hans",
+  "CMN-hANS": "zh-Hans",
-  "cmn-hans-cn": "cmn-Hans-CN",
+  "cmn-hans-cn": "zh-Hans-CN",
   "es-419": "es-419",
   "es-419-u-nu-latn": "es-419-u-nu-latn",
-  "cmn-hans-cn-u-ca-t-ca-x-t-u": "cmn-Hans-CN-t-ca-u-ca-x-t-u",
+  "cmn-hans-cn-u-ca-t-ca-x-t-u": "zh-Hans-CN-t-ca-u-ca-x-t-u",
   "de-gregory-u-ca-gregory": "de-gregory-u-ca-gregory",
   "no-nyn": "nn",
   "i-klingon": "tlh",
@@ -45,7 +45,7 @@ var canonicalizedTags = {
   "zh-hak-CN": "hak-CN",
   "sgn-ils": "ils",
   "in": "id",
-  "x-foo": "x-foo",
+  "x-foo": "und-x-foo",
   "sr-cyrl-ekavsk": "sr-Cyrl-ekavsk",
   "en-ca-newfound": "en-CA-newfound",
   "sl-rozaj-biske-1994": "sl-rozaj-biske-1994",

test/intl402/Intl/getCanonicalLocales/grandfathered.js

@@ -16,7 +16,7 @@ const regularGrandfathered = [
     },
     {
         tag: "zh-guoyu",
-        canonical: "cmn",
+        canonical: "zh",
     },
     {
         tag: "zh-hakka",

test/intl402/Intl/getCanonicalLocales/preferred-grandfathered.js

@@ -43,12 +43,12 @@ var canonicalizedTags = {
   "en-gb-oed": "en-GB-oxendict",
   "i-ami": "ami",
   "i-bnn": "bnn",
-  "i-default": "i-default",
+  "i-default": "und-x-i-default",
-  "i-enochian": "i-enochian",
+  "i-enochian": "und-x-i-enochian",
   "i-hak": "hak",
   "i-klingon": "tlh",
   "i-lux": "lb",
-  "i-mingo": "i-mingo",
+  "i-mingo": "und-x-i-mingo",
   "i-navajo": "nv",
   "i-pwn": "pwn",
   "i-tao": "tao",
@@ -60,12 +60,12 @@ var canonicalizedTags = {
 
   // Regular tags.
   "art-lojban": "jbo",
-  "cel-gaulish": "cel-gaulish",
+  "cel-gaulish": "und-x-cel-gaulish",
   "no-bok": "nb",
   "no-nyn": "nn",
-  "zh-guoyu": "cmn",
+  "zh-guoyu": "zh",
   "zh-hakka": "hak",
-  "zh-min": "zh-min",
+  "zh-min": "und-x-zh-min",
   "zh-min-nan": "nan",
   "zh-xiang": "hsn",
 };

test/intl402/Intl/getCanonicalLocales/weird-cases.js

@@ -13,13 +13,13 @@ includes: [compareArray.js]
 
 var weirdCases =
   [
-   "x-u-foo",
+   {locale: "x-u-foo", canonical: "und-x-u-foo"},
-   "en-x-u-foo",
+   {locale: "en-x-u-foo"},
-   "en-a-bar-x-u-foo",
+   {locale: "en-a-bar-x-u-foo"},
-   "en-x-u-foo-a-bar",
+   {locale: "en-x-u-foo-a-bar"},
-   "en-a-bar-u-baz-x-u-foo",
+   {locale: "en-a-bar-u-baz-x-u-foo"},
   ];
 
-weirdCases.forEach(function (weird) {
+weirdCases.forEach(function ({locale, canonical = locale}) {
-  assert(compareArray(Intl.getCanonicalLocales(weird), [weird]));
+  assert(compareArray(Intl.getCanonicalLocales(locale), [canonical]));
 });

test/intl402/language-tags-canonicalized.js

@@ -13,13 +13,13 @@ var canonicalizedTags = {
     "de": ["de"],
     "de-DE": ["de-DE", "de"],
     "DE-de": ["de-DE", "de"],
-    "cmn": ["cmn"],
+    "cmn": ["zh"],
-    "CMN-hANS": ["cmn-Hans", "cmn"],
+    "CMN-hANS": ["zh-Hans", "zh"],
-    "cmn-hans-cn": ["cmn-Hans-CN", "cmn-Hans", "cmn"],
+    "cmn-hans-cn": ["zh-Hans-CN", "zh-Hans", "zh"],
     "es-419": ["es-419", "es"],
     "es-419-u-nu-latn": ["es-419-u-nu-latn", "es-419", "es", "es-u-nu-latn"],
     // -u-ca is incomplete, so it will not show up in resolvedOptions().locale
-    "cmn-hans-cn-u-ca-t-ca-x-t-u": ["cmn-Hans-CN-t-ca-u-ca-x-t-u", "cmn-Hans-CN-t-ca-x-t-u", "cmn-Hans-CN-t-ca-x-t", "cmn-Hans-CN-t-ca", "cmn-Hans-CN", "cmn-Hans", "cmn"],
+    "cmn-hans-cn-u-ca-t-ca-x-t-u": ["zh-Hans-CN-t-ca-u-ca-x-t-u", "zh-Hans-CN-t-ca-x-t-u", "zh-Hans-CN-t-ca-x-t", "zh-Hans-CN-t-ca", "zh-Hans-CN", "zh-Hans", "zh"],
     "de-gregory-u-ca-gregory": ["de-gregory-u-ca-gregory", "de-gregory", "de-u-ca-gregory", "de"],
     "no-nyn": ["nn"],
     "i-klingon": ["tlh"],