diff --git a/duplicate_checker.py b/duplicate_checker.py index a57a4ad5..51ff913c 100644 --- a/duplicate_checker.py +++ b/duplicate_checker.py @@ -1,4 +1,4 @@ -# duplicate_checker.py (v2.2 - Multi-Key Blocking & optimiertes Scoring) +# duplicate_checker.py (v2.0 - mit Blocking-Strategie) import logging import pandas as pd @@ -6,70 +6,48 @@ from thefuzz import fuzz from config import Config from helpers import normalize_company_name, simple_normalize_url from google_sheet_handler import GoogleSheetHandler -from collections import defaultdict # --- Konfiguration --- CRM_SHEET_NAME = "CRM_Accounts" MATCHING_SHEET_NAME = "Matching_Accounts" -SCORE_THRESHOLD = 85 # Etwas höherer Schwellenwert für bessere Präzision +SCORE_THRESHOLD = 80 logging.basicConfig(level=logging.INFO, format='%(asctime)s - %(levelname)s - %(message)s') -def calculate_similarity_details(record1, record2): - """Berechnet einen gewichteten Ähnlichkeits-Score und gibt die Details zurück.""" - scores = {'name': 0, 'location': 0, 'domain': 0} - - if record1.get('normalized_domain') and record1['normalized_domain'] != 'k.a.' and record1['normalized_domain'] == record2.get('normalized_domain'): - scores['domain'] = 100 - - # Höhere Gewichtung für den Namen, da die Website oft fehlt - if record1.get('normalized_name') and record2.get('normalized_name'): - scores['name'] = round(fuzz.token_set_ratio(record1['normalized_name'], record2['normalized_name']) * 0.85) - - if record1.get('CRM Ort') and record1['CRM Ort'] == record2.get('CRM Ort'): - if record1.get('CRM Land') and record1['CRM Land'] == record2.get('CRM Land'): - scores['location'] = 20 - - total_score = sum(scores.values()) - return {'total': total_score, 'details': scores} - -def create_blocking_keys(name): - """Erstellt mehrere Blocking Keys für einen Namen, um die Sensitivität zu erhöhen.""" - if not name: - return [] - - words = name.split() - keys = set() - - # 1. Erstes Wort - if len(words) > 0: - keys.add(words[0]) - # 2. Zweites Wort (falls vorhanden) - if len(words) > 1: - keys.add(words[1]) - # 3. Erste 4 Buchstaben des ersten Wortes - if len(words) > 0 and len(words[0]) >= 4: - keys.add(words[0][:4]) - - return list(keys) +def calculate_similarity(record1, record2): + """Berechnet einen gewichteten Ähnlichkeits-Score zwischen zwei Datensätzen.""" + total_score = 0 + if record1['normalized_domain'] and record1['normalized_domain'] == record2['normalized_domain']: + total_score += 100 + if record1['normalized_name'] and record2['normalized_name']: + name_similarity = fuzz.token_set_ratio(record1['normalized_name'], record2['normalized_name']) + total_score += name_similarity * 0.7 + if record1['CRM Ort'] and record1['CRM Ort'] == record2['CRM Ort']: + if record1['CRM Land'] and record1['CRM Land'] == record2['CRM Land']: + total_score += 20 + return round(total_score) def main(): - logging.info("Starte den Duplikats-Check (v2.2 mit Multi-Key Blocking)...") + """Hauptfunktion zum Laden, Vergleichen und Schreiben der Daten.""" + logging.info("Starte den Duplikats-Check (v2.0 mit Blocking)...") try: sheet_handler = GoogleSheetHandler() except Exception as e: - logging.critical(f"FEHLER bei Initialisierung: {e}") + logging.critical(f"FEHLER bei Initialisierung des GoogleSheetHandler: {e}") return logging.info(f"Lade Master-Daten aus '{CRM_SHEET_NAME}'...") crm_df = sheet_handler.get_sheet_as_dataframe(CRM_SHEET_NAME) - if crm_df is None or crm_df.empty: return + if crm_df is None or crm_df.empty: + logging.critical(f"Konnte keine Daten aus '{CRM_SHEET_NAME}' laden. Breche ab.") + return logging.info(f"Lade zu prüfende Daten aus '{MATCHING_SHEET_NAME}'...") matching_df = sheet_handler.get_sheet_as_dataframe(MATCHING_SHEET_NAME) - if matching_df is None or matching_df.empty: return - original_matching_df = matching_df.copy() + if matching_df is None or matching_df.empty: + logging.critical(f"Konnte keine Daten aus '{MATCHING_SHEET_NAME}' laden. Breche ab.") + return logging.info("Normalisiere Daten für den Vergleich...") for df in [crm_df, matching_df]: @@ -77,55 +55,58 @@ def main(): df['normalized_domain'] = df['CRM Website'].astype(str).apply(simple_normalize_url) df['CRM Ort'] = df['CRM Ort'].astype(str).str.lower().str.strip() df['CRM Land'] = df['CRM Land'].astype(str).str.lower().str.strip() - df['block_keys'] = df['normalized_name'].apply(create_blocking_keys) + # Blocking Key: Das erste Wort des normalisierten Namens + df['block_key'] = df['normalized_name'].apply(lambda x: x.split()[0] if x else None) + # --- NEUE, SCHNELLE BLOCKING-STRATEGIE --- logging.info("Erstelle Index für CRM-Daten zur Beschleunigung...") - crm_index = defaultdict(list) - for record in crm_df.to_dict('records'): - for key in record['block_keys']: - crm_index[key].append(record) + crm_index = {} + for index, row in crm_df.iterrows(): + key = row['block_key'] + if key: + if key not in crm_index: + crm_index[key] = [] + crm_index[key].append(row) logging.info("Starte Matching-Prozess...") results = [] + total_matches = len(matching_df) - for match_record in matching_df.to_dict('records'): - best_score_info = {'total': 0, 'details': {'name': 0, 'location': 0, 'domain': 0}} + for index, match_row in matching_df.iterrows(): + best_score = 0 best_match_name = "" - logging.info(f"Prüfe: {match_record['CRM Name']}...") + logging.info(f"Prüfe {index + 1}/{total_matches}: {match_row['CRM Name']}...") - candidate_pool = {} - for key in match_record['block_keys']: - for crm_record in crm_index.get(key, []): - candidate_pool[crm_record['CRM Name']] = crm_record + # Finde den Block von Kandidaten + block_key = match_row['block_key'] + candidates = crm_index.get(block_key, []) - if not candidate_pool: - logging.debug(" -> Keine Kandidaten im Index gefunden.") + # Führe den teuren Vergleich nur für die Kandidaten in diesem Block durch + for crm_row in candidates: + score = calculate_similarity(match_row, crm_row) + if score > best_score: + best_score = score + best_match_name = crm_row['CRM Name'] - for crm_record in candidate_pool.values(): - score_info = calculate_similarity_details(match_record, crm_record) - if score_info['total'] > best_score_info['total']: - best_score_info = score_info - best_match_name = crm_record['CRM Name'] - - results.append({ - 'Potenzieller Treffer im CRM': best_match_name if best_score_info['total'] >= SCORE_THRESHOLD else "", - 'Score (Gesamt)': best_score_info['total'], - 'Score (Name)': best_score_info['details']['name'], - 'Bonus (Standort)': best_score_info['details']['location'], - 'Bonus (Domain)': best_score_info['details']['domain'] - }) + if best_score >= SCORE_THRESHOLD: + results.append({'Potenzieller Treffer im CRM': best_match_name, 'Ähnlichkeits-Score': best_score}) + else: + # Wenn nichts im Block gefunden wurde, trotzdem den besten Treffer (kann 0 sein) anzeigen + results.append({'Potenzieller Treffer im CRM': '' if not best_match_name else best_match_name, 'Ähnlichkeits-Score': best_score}) logging.info("Matching abgeschlossen. Schreibe Ergebnisse zurück ins Sheet...") result_df = pd.DataFrame(results) - output_df = pd.concat([original_matching_df.reset_index(drop=True), result_df], axis=1) + # Die ursprünglichen Spalten aus matching_df für die Ausgabe nehmen + output_df = matching_df[['CRM Name', 'CRM Website', 'CRM Ort', 'CRM Land']].copy() + output_df = pd.concat([output_df.reset_index(drop=True), result_df], axis=1) data_to_write = [output_df.columns.values.tolist()] + output_df.values.tolist() success = sheet_handler.clear_and_write_data(MATCHING_SHEET_NAME, data_to_write) if success: - logging.info(f"Ergebnisse erfolgreich in '{MATCHING_SHEET_NAME}' geschrieben.") + logging.info(f"Ergebnisse erfolgreich in das Tabellenblatt '{MATCHING_SHEET_NAME}' geschrieben.") else: logging.error("FEHLER beim Schreiben der Ergebnisse ins Google Sheet.")