fix: add transcription provider fallback

2026-06-10 11:47:47 +00:00
parent 6670b5adec
commit 53889da8f6
4 changed files with 183 additions and 7 deletions
@@ -1,6 +1,7 @@
 from __future__ import annotations

 import os
+import re
 from typing import Any


@@ -33,13 +34,39 @@ def summarize_error(body: Any, fallback: str) -> str:
        if isinstance(err, dict):
            msg = err.get("message")
            if isinstance(msg, str) and msg.strip():
-                return msg.strip()
+                return _compact_error_text(msg)

        msg = body.get("message")
        if isinstance(msg, str) and msg.strip():
-            return msg.strip()
+            return _compact_error_text(msg)

-    return fallback
+    return _compact_error_text(fallback)
+
+
+def _compact_error_text(text: str, limit: int = 240) -> str:
+    clean = text.strip()
+    if not clean:
+        return "Unknown provider error"
+
+    lower = clean.lower()
+    if "<!doctype html" in lower or "<html" in lower:
+        title_match = re.search(r"<title>\s*([^<]+?)\s*</title>", clean, flags=re.IGNORECASE | re.DOTALL)
+        title = re.sub(r"\s+", " ", title_match.group(1)).strip() if title_match else "HTML error page"
+        code_match = re.search(r"(?:error code\s*)?([45]\d{2})", clean, flags=re.IGNORECASE)
+        code = code_match.group(1) if code_match else None
+        phrase = "Bad gateway" if "bad gateway" in lower else title
+        if "cloudflare" in lower or "openrouter.ai" in lower:
+            if code:
+                return f"Cloudflare/OpenRouter returned HTTP {code} {phrase}"
+            return f"Cloudflare/OpenRouter returned {phrase}"
+        if code:
+            return f"Provider returned HTTP {code} HTML error page: {title}"
+        return f"Provider returned HTML error page: {title}"
+
+    clean = re.sub(r"\s+", " ", clean)
+    if len(clean) > limit:
+        return clean[: limit - 1].rstrip() + "…"
+    return clean


 def build_transcript_block(display_name: str, transcript: str) -> str: