Erster Wurf

core/data_processor.py

@@ -1,66 +1,83 @@
 import pandas as pd
 import requests
+import markdown
 from io import StringIO
 from datetime import datetime, timedelta
 from .config_loader import config
 
-_cache = {"events": None, "remarks": None}
+_cache = {"events": None, "remarks": None, "timestamp": None}
 HEADERS = {"User-Agent": "Mozilla/5.0"}
 
 def invalidate_cache():
     global _cache
-    _cache = {"events": None, "remarks": None}
-    print("DEBUG: Cache wurde manuell geleert.")
+    _cache = {"events": None, "remarks": None, "timestamp": None}
     return "Cache gelöscht"
 
-def get_upcoming_events():
-    if _cache["events"] is not None:
-        return _cache["events"]
-    
-    url = config['links']['times_csv']
-    print(f"DEBUG: Lade Zeiten von URL: {url}")
-    
-    response = requests.get(url, headers=HEADERS)
-    df = pd.read_csv(StringIO(response.text))
-    
-    # Debug: Spaltennamen prüfen
-    print(f"DEBUG: Spalten in Zeiten-Sheet gefunden: {df.columns.tolist()}")
+def _is_cache_valid():
+    if _cache["timestamp"] is None:
+        return False
+    return (datetime.now() - _cache["timestamp"]) < timedelta(hours=1)
+
+def get_upcoming_events(days_to_show=None, limit=None):
+    # Sofort Standardwert aus Config setzen, falls None oder 0
+    if not days_to_show:
+        days_to_show = config['processing']['days_to_show']
+
+    # 1. Daten laden (entweder aus Cache oder von Google)
+    if not _is_cache_valid() or _cache["events"] is None:
+        url = config['links']['times_csv']
+        response = requests.get(url, headers=HEADERS)
+        response.raise_for_status()
+        response.encoding = 'utf-8'
+        
+        df = pd.read_csv(StringIO(response.text))
+        
+        # Typografie: Bindestrich durch En-Dash (–) ersetzen
+        for col in ['Morgen', 'Nachmittag']:
+            df[col] = df[col].fillna('').astype(str).str.replace('-', '–', regex=False)
+            
+        date_col = config['google_sheet']['date_column']
+        df = df.dropna(subset=[date_col])
+        df[date_col] = pd.to_datetime(df[date_col], dayfirst=True, errors='coerce')
+        
+        wt_map = {0: 'Mo', 1: 'Di', 2: 'Mi', 3: 'Do', 4: 'Fr', 5: 'Sa', 6: 'So'}
+        df['Wochentag'] = df[date_col].dt.weekday.map(wt_map)
+        
+        _cache["events"] = df.sort_values(by=date_col).to_dict(orient='records')
+        _cache["timestamp"] = datetime.now()
 
-    date_col = config['google_sheet']['date_column']
-    df = df.dropna(subset=[date_col])
-    df[date_col] = pd.to_datetime(df[date_col], dayfirst=True, errors='coerce')
-    df = df.dropna(subset=[date_col]).fillna('')
-    
     heute = pd.Timestamp(datetime.now().date())
-    ende = heute + timedelta(days=config['processing']['days_to_show'])
+    date_col = config['google_sheet']['date_column']
+
+    # PRIORITÄT 1: Zeilen-Limit (gewinnt immer)
+    if limit and limit > 0:
+        return [e for e in _cache["events"] if e[date_col] >= heute][:limit]
     
-    mask = (df[date_col] >= heute) & (df[date_col] <= ende)
-    gefilterte = df.loc[mask].sort_values(by=date_col).copy()
-    
-    wt_map = {0: 'Mo', 1: 'Di', 2: 'Mi', 3: 'Do', 4: 'Fr', 5: 'Sa', 6: 'So'}
-    gefilterte['Wochentag'] = gefilterte[date_col].dt.weekday.map(wt_map)
-    
-    _cache["events"] = gefilterte.to_dict(orient='records')
-    return _cache["events"]
+    # PRIORITÄT 2: Tage-Logik
+    ende = heute + timedelta(days=int(days_to_show))
+    return [e for e in _cache["events"] if heute <= e[date_col] <= ende]
 
 def get_remarks():
-    if _cache["remarks"] is not None:
+    if _is_cache_valid() and _cache["remarks"] is not None:
         return _cache["remarks"]
     
     url = config['links']['remarks_csv']
-    print(f"DEBUG: Lade Bemerkungen von URL: {url}")
-    
     response = requests.get(url, headers=HEADERS)
-    
-    # Debug: Die ersten 100 Zeichen der Antwort sehen
-    print(f"DEBUG: Raw Data Bemerkungen (Anfang): {response.text[:100]}...")
+    response.raise_for_status()
+    response.encoding = 'utf-8'
     
     df = pd.read_csv(StringIO(response.text), skiprows=2, header=None)
     
     if not df.empty:
-        print(f"DEBUG: Bemerkungen-DF Head:\n{df.head(3)}")
-        remarks = df[0].dropna().astype(str).tolist()
-        _cache["remarks"] = [r.strip() for r in remarks if r.strip()]
+        raw_remarks = df[0].dropna().astype(str).tolist()
+        processed = []
+        for r in raw_remarks:
+            html = markdown.markdown(r.strip())
+            if html.startswith("<p>") and html.endswith("</p>"):
+                html = html[3:-4]
+            processed.append(html)
+        _cache["remarks"] = processed
+        _cache["timestamp"] = datetime.now()
     else:
         _cache["remarks"] = []