Pronounce IP segments instead of passing numerals to TTS (#54)

* Pronounce IP segments instead of passing numerals to TTS

* Add missing `load_language` call

* Fix TypeError in spoken IP address parsing

* Update dialog rendering for translations

* Refactor to minimize extra resource rendering

---------

Co-authored-by: Daniel McKnight <[email protected]>

__init__.py

@@ -42,6 +42,7 @@
 
 from typing import List
 from ifaddr import get_adapters
+from lingua_franca import load_language
 from neon_utils.user_utils import get_user_prefs
 from requests import get
 from adapt.intent import IntentBuilder
@@ -51,6 +52,7 @@
 from ovos_utils.process_utils import RuntimeRequirements
 from ovos_bus_client.message import Message
 from ovos_workshop.decorators import intent_handler
+from lingua_franca.format import pronounce_number
 
 
 # TODO: Add something equivalent to neon_utils.net_utils
@@ -101,29 +103,32 @@ def handle_query_ip(self, message):
  Handle a user request for the IP Address
  :param message: Message associated with request
  """
+ load_language(self.lang)
  if message.data.get("public"):
  public = True
  addr = {'public': self._get_public_ip_address(message)}
  else:
  public = False
  addr = get_ifaces(message=message)
 
- dot = self.dialog_renderer.render("dot")
-
  if len(addr) == 0: # No IP Address found
  if not get_user_prefs(message)["response_mode"].get(
  "limit_dialog"):
  self.speak_dialog("no network connection", private=True)
  else:
  self.speak("I'm not connected to a network", private=True)
  return
+
+ dot = self.resources.render_dialog("dot")
+
  if len(addr) == 1: # Single IP Address to speak
  iface, ip = addr.popitem()
- ip_spoken = ip.replace(".", f" {dot} ")
+ ip_spoken = f" {dot} ".join([pronounce_number(int(part))
+ for part in ip.split('.')])
  if public:
- say_ip = self.translate("word_public")
+ say_ip = self.resources.render_dialog("word_public")
  else:
- say_ip = self.translate("word_local")
+ say_ip = self.resources.render_dialog("word_local")
  self.speak_dialog("my address is",
  {'ip': ip_spoken,
  'pub': say_ip}, private=True)