feat: integrate ApiPilot tools and add light-theme voice UI

2026-10-05 08:25:29 -04:00
parent 413ba08c21
commit 4d7c7a23df
9 changed files with 6930 additions and 171 deletions
@@ -12,6 +12,78 @@ import math
 import re
 import tempfile
 import unicodedata
+
+# 2026-09-25 SANITIZE APIPILOT MARKERS PRIMA DI TTS (anti-leak pronunciato da Qwen3).
+#   Lato server S2S Python abbiamo gia' gate + sanitize ma come DOPPIA RETE
+#   di sicurezza sanitizziamo DIRETTAMENTE il testo che entra in Qwen3-TTS.
+#   Formati osservati reali oggi:
+#     (a) <<<APIPILOT_CALL:NAME=tool,ARGUMENTS=json>>>    standard 3 parentesi
+#     (b) <APIPILOT_CALL:NAME=tool,ARGUMENTS={"format":"voice"}
+#                                                         singolo < > (caso user oggi)
+#     (c) <<<APIPILOT_CALL:...}>   \n>>                    spezzato piu' righe
+#     (d) >>>APIPILOT_CALL:...<<<                          invertito
+#   Regola: CALL -> placeholder spagnolo (frase naturale utente), RESULT -> strip.
+_APIPILOT_CALL_NORMAL_RE = re.compile(
+    r"[<]{1,3}\s*APIPILOT_CALL:NAME=([A-Za-z0-9_.-]+),ARGUMENTS=(.*?)[>]{1,3}",
+    re.IGNORECASE | re.DOTALL,
+)
+_APIPILOT_CALL_INVERTED_RE = re.compile(
+    r"[>]{1,3}\s*APIPILOT_CALL:NAME=([A-Za-z0-9_.-]+),ARGUMENTS=(.*?)[<]{1,3}",
+    re.IGNORECASE | re.DOTALL,
+)
+_APIPILOT_RESULT_RE = re.compile(
+    r"[<]{1,3}\s*APIPILOT_RESULT(?::PASS2)?[^<>\n]{0,2000}[>]{1,3}",
+    re.IGNORECASE | re.DOTALL,
+)
+_APIPILOT_SPANISH_PLACEHOLDER = "Voy a consultar tus datos, un momento por favor."
+
+
+def _tts_sanitize_apipilot_markers(text: str) -> tuple[str, bool, str]:
+    """Rimuove marker APIPILOT dal testo TTS.
+
+    Returns (sanitized_text, found_any, log_tag).
+    """
+    if not isinstance(text, str) or not text.strip():
+        return text, False, "empty"
+    found_call = False
+    found_result = False
+    tool_name = ""
+    # 1) marker CALL standard + invertito -> placeholder spagnolo
+    def _repl_call(m: re.Match) -> str:
+        nonlocal found_call, tool_name
+        found_call = True
+        try:
+            nm = m.group(1)
+            if isinstance(nm, str) and nm.strip():
+                tool_name = nm.strip()
+        except Exception:
+            pass
+        return f" {_APIPILOT_SPANISH_PLACEHOLDER} "
+    text2 = _APIPILOT_CALL_NORMAL_RE.sub(_repl_call, text)
+    if not found_call:
+        text2 = _APIPILOT_CALL_INVERTED_RE.sub(_repl_call, text2)
+    # 2) marker RESULT -> strip silenzioso (non sono pronunciabili)
+    if _APIPILOT_RESULT_RE.search(text2):
+        found_result = True
+        text2 = _APIPILOT_RESULT_RE.sub(" ", text2)
+    # 3) safety: FRAMMENTI marker singoli che non hanno wrapper completo
+    low = text2.lower()
+    if "apipilot" in low and ("call" in low or "result" in low):
+        import re as _re_sf
+        text2 = _re_sf.sub(r"(?i)\bAPIPILOT(?:_CALL|_RESULT)?\b", " ", text2)
+        text2 = _re_sf.sub(r"(?i)\b(?:NAME|ARGUMENTS|STATUS|PASS2)\s*[:=]", " ", text2)
+        found_call = True
+    if found_call or found_result:
+        text2 = re.sub(r"\s{2,}", " ", text2).strip()
+    tag = "-"
+    if found_call and found_result:
+        tag = f"CALL({tool_name or '?'})+RESULT"
+    elif found_call:
+        tag = f"CALL({tool_name or '?'})"
+    elif found_result:
+        tag = "RESULT"
+    return text2, (found_call or found_result), tag
+
 from collections.abc import Callable
 from pathlib import Path
 from sys import platform
@@ -709,7 +781,20 @@ class Qwen3TTSHandler(BaseHandler[TTSIn, TTSOut]):

        coalesced_text, _language_code, _saw_end_of_response = self._coalesce_pending_tts_input(tts_input)

-        text = coalesced_text or "Hello."
+        raw_text = coalesced_text or "Hello."
+        # 2026-09-25 SANITIZE APIPILOT PRIMA DI TTS (doppia rete sicurezza):
+        #   Anche se lato server S2s Python abbiamo gia' gate audio drop + sanitize,
+        #   Qwen3-TTS puo' ricevere marker per race / nuovi formati. Quindi QUI
+        #   eliminiamo / sostituiamo con placeholder PRIMA di chiamare generate().
+        text, found_marker, sanitize_tag = _tts_sanitize_apipilot_markers(raw_text)
+        if found_marker:
+            logger.warning(
+                "[TTS-APIPILOT-SANITIZE] marker=%s rawLen=%d sanitizedLen=%d → sostituito con placeholder spagnolo / strip.",
+                sanitize_tag, len(raw_text or ""), len(text or ""),
+            )
+        if not text.strip():
+            # Safety se il marker era l'unica cosa: ripetiamo placeholder invece di Hello.
+            text = _APIPILOT_SPANISH_PLACEHOLDER

        model_type = self._model_type()
        self._apply_session_voice_override(model_type, runtime_config, response)