Base entity ids on English for languages not using Latin script (#91357)

script/languages.py

@@ -15,10 +15,61 @@ req = requests.get(
 data = json.loads(req.content)
 languages = set(data.keys())
 
+# Languages which can be used for entity IDs.
+# Languages in the set are those which use a writing system based on the Latin
+# script. Languages not in this set will instead base the entity ID on English.
+
+# Note: Although vietnamese writing is based on the Latin script, it's too ambiguous
+# after accents and diacritics have been removed by slugify
+NATIVE_ENTITY_IDS = {
+    "af",  # Afrikaans
+    "bs",  # Bosanski
+    "ca",  # Català
+    "cs",  # Čeština
+    "cy",  # Cymraeg
+    "da",  # Dansk
+    "de",  # Deutsch
+    "en",  # English
+    "en-GB",  # English (GB)
+    "eo",  # Esperanto
+    "es",  # Español
+    "es-419",  # Español (Latin America)
+    "et",  # Eesti
+    "eu",  # Euskara
+    "fi",  # Suomi
+    "fr",  # Français
+    "fy",  # Frysk
+    "gl",  # Galego
+    "gsw",  # Schwiizerdütsch
+    "hr",  # Hrvatski
+    "hu",  # Magyar
+    "id",  # Indonesia
+    "is",  # Íslenska
+    "it",  # Italiano
+    "ka",  # Kartuli
+    "lb",  # Lëtzebuergesch
+    "lt",  # Lietuvių
+    "lv",  # Latviešu
+    "nb",  # Nederlands
+    "nl",  # Norsk Bokmål
+    "nn",  # Norsk Nynorsk"
+    "pl",  # Polski
+    "pt",  # Português
+    "pt-BR",  # Português (BR)
+    "ro",  # Română
+    "sk",  # Slovenčina
+    "sl",  # Slovenščina
+    "sr-Latn",  # Srpski
+    "sv",  # Svenska
+    "tr",  # Türkçe
+}
+
 Path("homeassistant/generated/languages.py").write_text(
     format_python_namespace(
         {
+            "DEFAULT_LANGUAGE": "en",
             "LANGUAGES": languages,
+            "NATIVE_ENTITY_IDS": NATIVE_ENTITY_IDS,
         },
         generator="script.languages [frontend_tag]",
     )