get_live_commentary

Instructions

Implementation Reference

• cricket_server.py:536-678 (handler)

  The main handler function for the get_live_commentary tool. It is decorated with @mcp.tool() for automatic registration in FastMCP. The function fetches live commentary primarily from the Cricbuzz JSON API, falling back to HTML scraping if needed. It extracts match title, commentary events, and handles errors gracefully.

  def get_live_commentary(match_url: str, limit: int = 20) -> dict:
      """
      Get recent live commentary events for a Cricbuzz match.
  
      Args:
          match_url (str): Cricbuzz match URL. Can be a general match page; the commentary tab will be resolved automatically.
          limit (int): Maximum number of recent commentary items to return.
  
      Returns:
          dict: {"title": str, "commentary_url": str, "events": [{"text": str}]}
      """
      if not match_url or "cricbuzz.com" not in match_url:
          return {"error": "A valid Cricbuzz match URL is required."}
  
      # Try official JSON commentary API first (more reliable than HTML scraping)
      match_id_match = re.search(r"/(\d{5,7})/", match_url)
      if not match_id_match:
          return {"error": "Could not extract match id from URL."}
      match_id = match_id_match.group(1)
  
      api_url = f"https://www.cricbuzz.com/api/cricket-match/commentary/{match_id}"
      try:
          resp = requests.get(api_url, headers=HEADERS, timeout=15)
          resp.raise_for_status()
          data = resp.json()
      except Exception as e:
          data = None
  
      def _clean_comm_text(text: str) -> str:
          # Remove formatting markers like "B0$", "I0$" etc.
          t = re.sub(r"[A-Z]\d\$", "", text)
          # Normalize whitespace
          t = re.sub(r"\s+", " ", t).strip()
          return t
  
      if isinstance(data, dict) and data.get("commentaryList"):
          header = data.get("matchHeader", {})
          miniscore = data.get("miniscore", {})
          title_parts = []
          if header.get("matchDescription"):
              title_parts.append(header.get("matchDescription"))
          if header.get("status"):
              title_parts.append(header.get("status"))
          title = " - ".join(title_parts) if title_parts else None
  
          events = []
          for item in data.get("commentaryList", [])[: max(0, limit)]:
              text = _clean_comm_text(str(item.get("commText", "")).strip())
              if not text:
                  continue
              ev = {"text": text}
              if item.get("event"):
                  ev["event"] = item.get("event")
              if item.get("ballNbr") is not None:
                  ev["ball"] = item.get("ballNbr")
              events.append(ev)
  
          return {
              "title": title,
              "commentary_url": match_url,
              "events": events,
          }
  
      # If JSON API fails, fall back to HTML scraping heuristics
      def _fetch(url: str) -> BeautifulSoup | None:
          try:
              resp = requests.get(url, headers=HEADERS, timeout=15)
              resp.raise_for_status()
              return BeautifulSoup(resp.text, "lxml")
          except Exception:
              return None
  
      page = _fetch(match_url)
      if not page:
          return {"error": "Failed to load match page. The match might not be live or the URL may be incorrect."}
  
      commentary_url = None
      try:
          nav = page.find("div", class_=re.compile("cb-nav-pills"))
          if nav:
              for a in nav.find_all("a", href=True):
                  if "commentary" in a.get("href", "").lower() or (a.text and "commentary" in a.text.lower()):
                      commentary_url = ("https://www.cricbuzz.com" + a["href"]) if a["href"].startswith("/") else a["href"]
                      break
      except Exception:
          pass
  
      if not commentary_url:
          commentary_url = match_url.rstrip("/") + "/commentary"
  
      cpage = _fetch(commentary_url)
      if not cpage:
          return {"error": "Failed to load commentary page. This match may not have live commentary available."}
  
      result: dict = {"title": None, "commentary_url": commentary_url, "events": []}
      title_tag = cpage.find("h1", class_=re.compile("cb-nav-hdr"))
      if title_tag:
          result["title"] = title_tag.text.strip()
  
      candidates = []
      candidates.extend(cpage.find_all("div", class_=re.compile(r"cb-col\s+cb-col-90\s+cb-com-ln")))
      if not candidates:
          lst = cpage.find("div", class_=re.compile("cb-com-lst"))
          if lst:
              candidates.extend(lst.find_all("div", class_=re.compile(r"cb-col\s+cb-col-90")))
      if not candidates:
          candidates.extend(cpage.find_all("p", class_=re.compile("cb-com-ln")))
      if not candidates:
          candidates.extend(cpage.find_all("div", class_=re.compile("cb-com-ln")))
      if not candidates:
          for div in cpage.find_all("div"):
              text = div.get_text(" ", strip=True)
              if text and len(text) > 20 and ("ball" in text.lower() or "over" in text.lower() or "wicket" in text.lower()):
                  candidates.append(div)
  
      events = []
      for node in candidates:
          try:
              text = node.get_text(" ", strip=True)
              if not text:
                  continue
              if text.lower().startswith("commentary"):
                  continue
              if len(text) < 10:
                  continue
              events.append({"text": text})
              if len(events) >= limit:
                  break
          except Exception:
              continue
  
      result["events"] = events
      if not events:
          result["note"] = "No commentary items found. This match may not be live or commentary may not be available."
          try:
              match_info = get_match_details(match_url)
              if "error" not in match_info:
                  result["fallback"] = "Commentary not available, but here's the match details:"
                  result["match_details"] = match_info
          except Exception:
              pass
  
      return result