Update src/data_sources.py

src/data_sources.py

@@ -1,64 +1,34 @@
-import cbpy as cb
-import pandas as pd
-from typing import Dict, Any
+import requests
+from bs4 import BeautifulSoup
 
-def list_live_matches() -> pd.DataFrame:
-    """Return minimal live match list (id, title)."""
-    try:
-        matches = cb.Cricbuzz().matches()  # pulls cricbuzz matches.json
-        rows = []
-        for m in matches.get("matches", []):
-            if m.get("mchstate") in {"preview","inprogress","innings break","toss"}:
-                rows.append({
-                    "match_id": m.get("id"),
-                    "title": f'{m.get("srs", "")} - {m.get("mnum", "")} | {m.get("team1", {}).get("name")} vs {m.get("team2", {}).get("name")}',
-                    "state": m.get("mchstate"),
-                })
-        return pd.DataFrame(rows)
-    except Exception:
-        return pd.DataFrame([])
+# Example: scrape live matches (ESPNCricinfo or Cricbuzz)
+def list_live_matches():
+    url = "https://www.espncricinfo.com/live-cricket-match-results"
+    response = requests.get(url)
+    soup = BeautifulSoup(response.text, "html.parser")
 
-def scorecard(match_id: str) -> Dict[str, Any]:
-    """Fetch a compact scorecard/summary for a match id."""
-    c = cb.Cricbuzz()
-    try:
-        sc = c.scorecard(match_id)  # team info, batsmen, bowlers, overs
-        return sc
-    except Exception:
-        return {}
+    matches = []
+    for item in soup.select(".ds-px-4.ds-py-3"):
+        title = item.get_text(" ", strip=True)
+        matches.append(title)
 
-def extract_simple_context(sc: Dict[str, Any]) -> Dict[str, Any]:
-    """Minimal, robust context from scorecard response."""
-    if not sc or "scorecard" not in sc:
-        return {}
-    # Grab the latest innings block
-    inn = sc["scorecard"][-1]
-    return {
-        "batting_team": inn.get("batteam", {}).get("name"),
-        "bowling_team": inn.get("bowlteam", {}).get("name"),
-        "score": inn.get("runs"),
-        "wickets": inn.get("wickets"),
-        "overs": inn.get("overs"),
-        "rr": inn.get("rr"),
-        "crr": inn.get("crr"),
-        "players": {
-            "batsmen": [b.get("name") for b in inn.get("batcard", [])],
-            "bowlers": [b.get("name") for b in inn.get("bowlcard", [])],
-        }
-    }
+    return matches
 
-def striker_form_last_overs(sc: Dict[str, Any], striker_name: str) -> Dict[str, Any]:
-    """Very light ‘recent pattern’ proxy using scorecard strike rates and runs."""
-    if not sc or "scorecard" not in sc:
-        return {}
-    inn = sc["scorecard"][-1]
-    bats = {b.get("name"): b for b in inn.get("batcard", [])}
-    s = bats.get(striker_name, {})
-    return {
-        "runs": s.get("runs"),
-        "balls": s.get("balls"),
-        "sr": s.get("sr"),
-        "fours": s.get("fours"),
-        "sixes": s.get("sixers") or s.get("six", None),
-        "note": "Approximation from live card; ball-by-ball upgrades later."
-    }
+
+def scorecard(match_url: str):
+    response = requests.get(match_url)
+    soup = BeautifulSoup(response.text, "html.parser")
+
+    # (MVP: extract only scores; later we can refine for player stats)
+    score = soup.select_one(".ds-text-compact-xxs.ds-font-medium")
+    return score.get_text(strip=True) if score else "No score found"
+
+
+def extract_simple_context(match_url: str):
+    # For MVP: return score only
+    return scorecard(match_url)
+
+
+def striker_form_last_overs(player_name: str):
+    # Placeholder – later we’ll integrate CricAPI or Stats API
+    return f"Recent form data for {player_name} (to be implemented)"