From dfcb270a7fb080b8a851f56521c5b27ae77456d1 Mon Sep 17 00:00:00 2001 From: Floke Date: Wed, 6 Aug 2025 09:06:07 +0000 Subject: [PATCH] duplicate_checker.py aktualisiert --- duplicate_checker.py | 248 ++++++++++++++++--------------------------- 1 file changed, 93 insertions(+), 155 deletions(-) diff --git a/duplicate_checker.py b/duplicate_checker.py index 6c17c6f4..8ea06b13 100644 --- a/duplicate_checker.py +++ b/duplicate_checker.py @@ -1,169 +1,107 @@ -import os -import re +duplicate_checker.py (v2.0 - mit Blocking-Strategie) import logging import pandas as pd -import numpy as np -import recordlinkage -from rapidfuzz import fuzz +from thefuzz import fuzz +from config import Config +from helpers import normalize_company_name, simple_normalize_url from google_sheet_handler import GoogleSheetHandler - -# --- Konfiguration --- -CRM_SHEET_NAME = "CRM_Accounts" +--- Konfiguration --- +CRM_SHEET_NAME = "CRM_Accounts" MATCHING_SHEET_NAME = "Matching_Accounts" -# Threshold gesenkt und konfigurierbar im Code -SCORE_THRESHOLD = 0.75 -WEIGHTS = { - 'domain': 0.5, - 'name': 0.4, - 'city': 0.1, -} -# Relativer Log-Ordner -LOG_DIR = 'log' -LOG_FILENAME = 'duplicate_check.log' - -# --- Logging Setup --- -if not os.path.exists(LOG_DIR): - try: - os.makedirs(LOG_DIR) - except Exception as e: - print(f"Warnung: Konnte Log-Ordner nicht anlegen: {e}") -log_path = os.path.join(LOG_DIR, LOG_FILENAME) - -logger = logging.getLogger(__name__) -logger.setLevel(logging.DEBUG) -formatter = logging.Formatter('%(asctime)s - %(levelname)-8s - %(name)s - %(message)s') - -# Console handler -console_handler = logging.StreamHandler() -console_handler.setLevel(logging.INFO) -console_handler.setFormatter(formatter) -logger.addHandler(console_handler) - -# File handler -try: - file_handler = logging.FileHandler(log_path, mode='a', encoding='utf-8') - file_handler.setLevel(logging.DEBUG) - file_handler.setFormatter(formatter) - logger.addHandler(file_handler) - logger.info(f"Logging auch in Datei: {log_path}") -except Exception as e: - logger.warning(f"Konnte keine Log-Datei schreiben: {e}") - -# --- Hilfsfunktionen --- -def normalize_company_name(name: str) -> str: - s = str(name).casefold() - for src, dst in [('ä','ae'), ('ö','oe'), ('ü','ue'), ('ß','ss')]: - s = s.replace(src, dst) - s = re.sub(r'[^a-z0-9\s]', ' ', s) - stops = ['gmbh','ag','kg','ug','ohg','holding','group','international'] - tokens = [t for t in s.split() if t and t not in stops] - return ' '.join(tokens) - - -def normalize_domain(url: str) -> str: - s = str(url).casefold().strip() - s = re.sub(r'^https?://', '', s) - s = s.split('/')[0] - if s.startswith('www.'): - s = s[4:] - return s - - +SCORE_THRESHOLD = 80 +logging.basicConfig(level=logging.INFO, format='%(asctime)s - %(levelname)s - %(message)s') +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(): - logger.info("Starte den Duplikats-Check mit Fallback-Blocking...") - # GoogleSheetHandler initialisieren - try: - sheet_handler = GoogleSheetHandler() - logger.info("GoogleSheetHandler initialisiert") - except Exception as e: - logger.critical(f"FEHLER bei Initialisierung des GoogleSheetHandler: {e}") - return +"""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 des GoogleSheetHandler: {e}") + return - # Daten laden - crm_df = sheet_handler.get_sheet_as_dataframe(CRM_SHEET_NAME) - match_df = sheet_handler.get_sheet_as_dataframe(MATCHING_SHEET_NAME) - if crm_df is None or crm_df.empty or match_df is None or match_df.empty: - logger.critical("CRM- oder Matching-Daten leer. Abbruch.") - return - logger.info(f"{len(crm_df)} CRM-Zeilen, {len(match_df)} Matching-Zeilen geladen") +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: + logging.critical(f"Konnte keine Daten aus '{CRM_SHEET_NAME}' laden. Breche ab.") + return - # Normalisierung - for df, label in [(crm_df, 'CRM'), (match_df, 'Matching')]: - df['norm_name'] = df['CRM Name'].fillna('').apply(normalize_company_name) - df['norm_domain'] = df['CRM Website'].fillna('').apply(normalize_domain) - df['city'] = df['CRM Ort'].fillna('').apply(lambda x: str(x).casefold().strip()) - df['name_token'] = df['norm_name'].apply(lambda x: x.split()[0] if x else np.nan) - # Leere Werte als NaN markieren - df['norm_domain'].replace('', np.nan, inplace=True) - df['city'].replace('', np.nan, inplace=True) - logger.debug(f"{label}-Normalisierung: norm_domain={df.iloc[0]['norm_domain']}, name_token={df.iloc[0]['name_token']}") +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: + logging.critical(f"Konnte keine Daten aus '{MATCHING_SHEET_NAME}' laden. Breche ab.") + return - # Blocking: Domain und Name-Token - index_dom = recordlinkage.Index() - index_dom.block('norm_domain') - pairs_dom = index_dom.index(crm_df, match_df) - index_name = recordlinkage.Index() - index_name.block('name_token') - pairs_name = index_name.index(crm_df, match_df) - # Union der Kandidatenpaare - candidate_pairs = pairs_dom.append(pairs_name).drop_duplicates() - logger.info(f"Blocking abgeschlossen: Dom-Paare={len(pairs_dom)}, Name-Paare={len(pairs_name)}, Gesamt={len(candidate_pairs)}") +logging.info("Normalisiere Daten für den Vergleich...") +for df in [crm_df, matching_df]: + df['normalized_name'] = df['CRM Name'].astype(str).apply(normalize_company_name) + 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() + # Blocking Key: Das erste Wort des normalisierten Namens + df['block_key'] = df['normalized_name'].apply(lambda x: x.split()[0] if x else None) - # Vergleichsregeln definieren - compare = recordlinkage.Compare() - compare.exact('norm_domain', 'norm_domain', label='domain', missing_value=0) - compare.string('norm_name', 'norm_name', method='jarowinkler', label='name_sim') - compare.exact('city', 'city', label='city', missing_value=0) - features = compare.compute(candidate_pairs, crm_df, match_df) - logger.debug(f"Features berechnet: {features.head()}...") +# --- NEUE, SCHNELLE BLOCKING-STRATEGIE --- +logging.info("Erstelle Index für CRM-Daten zur Beschleunigung...") +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) - # Score berechnen - features['score'] = ( - WEIGHTS['domain'] * features['domain'] + - WEIGHTS['name'] * features['name_sim'] + - WEIGHTS['city'] * features['city'] - ) - logger.info("Scores berechnet") +logging.info("Starte Matching-Prozess...") +results = [] +total_matches = len(matching_df) - # Per-Match Logging und Auswahl - results = [] - crm_map = crm_df.reset_index() - for match_idx, group in features.reset_index().groupby('level_1'): - logger.info(f"--- Prüfe Matching-Zeile {match_idx} ---") - df_block = group.sort_values('score', ascending=False).copy() - # CRM-Daten für Log - df_block['CRM Name'] = df_block['level_0'].map(crm_map.set_index('index')['CRM Name']) - # Log der Top-Kandidaten - for _, row in df_block.head(5).iterrows(): - logger.debug(f"Candidate [{int(row['level_0'])}]: score={row['score']:.3f}, name_sim={row['name_sim']:.3f}, dom={row['domain']}, city={row['city']} => {row['CRM Name']}") - top = df_block.iloc[0] - crm_idx = top['level_0'] if top['score'] >= SCORE_THRESHOLD else None - if crm_idx is not None: - logger.info(f" --> Match: {int(crm_idx)} ({top['CRM Name']}) mit Score {top['score']:.2f}") - else: - logger.info(f" --> Kein Match (höchster Score {top['score']:.2f})") - results.append((crm_idx, match_idx, top['score'])) +for index, match_row in matching_df.iterrows(): + best_score = 0 + best_match_name = "" + + logging.info(f"Prüfe {index + 1}/{total_matches}: {match_row['CRM Name']}...") - # Ausgabe zusammenstellen - match_map = match_df.reset_index() - output = match_map[['CRM Name','CRM Website','CRM Ort','CRM Land']].copy() - output[['Matched CRM Name','Matched CRM Website','Matched CRM Ort','Matched CRM Land','Score']] = '' - for crm_idx, match_idx, score in results: - if crm_idx is not None: - crm_row = crm_map[crm_map['index']==crm_idx].iloc[0] - output.at[match_idx, 'Matched CRM Name'] = crm_row['CRM Name'] - output.at[match_idx, 'Matched CRM Website'] = crm_row['CRM Website'] - output.at[match_idx, 'Matched CRM Ort'] = crm_row['CRM Ort'] - output.at[match_idx, 'Matched CRM Land'] = crm_row['CRM Land'] - output.at[match_idx, 'Score'] = round(score,3) - - # Zurückschreiben ins Google Sheet - data = [output.columns.tolist()] + output.values.tolist() - success = sheet_handler.clear_and_write_data(MATCHING_SHEET_NAME, data) - if success: - logger.info(f"Erfolgreich geschrieben: {len([r for r in results if r[0] is not None])} Matches") + # Finde den Block von Kandidaten + block_key = match_row['block_key'] + candidates = crm_index.get(block_key, []) + + # 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'] + + if best_score >= SCORE_THRESHOLD: + results.append({'Potenzieller Treffer im CRM': best_match_name, 'Ähnlichkeits-Score': best_score}) else: - logger.error("Fehler beim Schreiben ins Google Sheet.") + # 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}) -if __name__ == '__main__': - main() +logging.info("Matching abgeschlossen. Schreibe Ergebnisse zurück ins Sheet...") +result_df = pd.DataFrame(results) + +# 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 das Tabellenblatt '{MATCHING_SHEET_NAME}' geschrieben.") +else: + logging.error("FEHLER beim Schreiben der Ergebnisse ins Google Sheet.") +if name == "main": +main() \ No newline at end of file