Brancheneinstufung2/duplicate_checker.py

import os
import sys
import logging
import pandas as pd
from thefuzz import fuzz
from helpers import normalize_company_name, simple_normalize_url
from google_sheet_handler import GoogleSheetHandler

# duplicate_checker.py v2.9 (Bulletproof Name-Partial/SORT/SET + Bonus)
# Version: 2025-08-06_18-10

# --- Konfiguration ---
CRM_SHEET_NAME      = "CRM_Accounts"
MATCHING_SHEET_NAME = "Matching_Accounts"
SCORE_THRESHOLD     = 80   # Score-Schwelle
LOG_DIR             = "Log"
LOG_FILE            = "duplicate_check_v2.9.txt"

# --- Logging Setup ---
if not os.path.exists(LOG_DIR):
    os.makedirs(LOG_DIR, exist_ok=True)
log_path = os.path.join(LOG_DIR, LOG_FILE)
root = logging.getLogger()
root.setLevel(logging.DEBUG)
# Remove existing handlers
for h in list(root.handlers):
    root.removeHandler(h)
formatter = logging.Formatter("%(asctime)s - %(levelname)-8s - %(message)s")
# Console handler (INFO+)
ch = logging.StreamHandler(sys.stdout)
ch.setLevel(logging.INFO)
ch.setFormatter(formatter)
root.addHandler(ch)
# File handler (DEBUG+)
fh = logging.FileHandler(log_path, mode='a', encoding='utf-8')
fh.setLevel(logging.DEBUG)
fh.setFormatter(formatter)
root.addHandler(fh)
logger = logging.getLogger(__name__)
logger.info(f"Logging to console and file: {log_path}")
logger.info("Starting duplicate_checker.py v2.9 | Version: 2025-08-06_18-10")

# --- Ähnlichkeitsberechnung ---
def calculate_similarity(record1, record2):
    """Berechnet Score-Komponenten: Domain, Name (SET,PARTIAL,SORT), Ort und Bonus."""
    # Domain exact match
    dom1 = record1.get('normalized_domain', '')
    dom2 = record2.get('normalized_domain', '')
    domain_flag = 1 if dom1 and dom1 == dom2 else 0

    # Location exact match
    loc_flag = 1 if (record1.get('CRM Ort') == record2.get('CRM Ort') and
                     record1.get('CRM Land') == record2.get('CRM Land')) else 0

    # Name scores
    n1 = record1.get('normalized_name', '')
    n2 = record2.get('normalized_name', '')
    if n1 and n2:
        ts = fuzz.token_set_ratio(n1, n2)
        pr = fuzz.partial_ratio(n1, n2)
        ss = fuzz.token_sort_ratio(n1, n2)
        name_score = max(ts, pr, ss)
    else:
        name_score = 0

    # Bonus für reine Name-Matches
    bonus_flag = 1 if domain_flag == 0 and loc_flag == 0 and name_score >= 85 else 0

    # Gesamtscore
    total = domain_flag * 100 + name_score * 1.0 + loc_flag * 20 + bonus_flag * 20
    return round(total), domain_flag, name_score, loc_flag, bonus_flag

# --- Hauptfunktion ---
def main():
    logger.info("Starte Duplikats-Check v2.9 (Bulletproof)")
    # GoogleSheetHandler init
    try:
        sheet = GoogleSheetHandler()
        logger.info("GoogleSheetHandler initialisiert")
    except Exception as e:
        logger.critical(f"Init GoogleSheetHandler fehlgeschlagen: {e}")
        sys.exit(1)

    # Daten laden
    logger.info(f"Lade CRM-Daten aus '{CRM_SHEET_NAME}'...")
    crm_df = sheet.get_sheet_as_dataframe(CRM_SHEET_NAME)
    logger.info(f"{0 if crm_df is None else len(crm_df)} CRM-Datensätze geladen")
    logger.info(f"Lade Matching-Daten aus '{MATCHING_SHEET_NAME}'...")
    match_df = sheet.get_sheet_as_dataframe(MATCHING_SHEET_NAME)
    logger.info(f"{0 if match_df is None else len(match_df)} Matching-Datensätze geladen")

    if crm_df is None or crm_df.empty or match_df is None or match_df.empty:
        logger.critical("Leere Daten in einem der Sheets. Abbruch.")
        return

    # Normalisierung & Blocking-Key
    for df, label in [(crm_df, 'CRM'), (match_df, 'Matching')]:
        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()
        df['block_key']         = df['normalized_name'].apply(lambda x: x.split()[0] if x else None)
        logger.debug(f"{label}-Sample: {df.iloc[0][['normalized_name','normalized_domain','block_key']].to_dict()}")

    # Blocking-Index erzeugen
    crm_index = {}
    for _, row in crm_df.iterrows():
        key = row['block_key']
        if key:
            crm_index.setdefault(key, []).append(row)
    logger.info(f"Blocking-Index mit {len(crm_index)} Keys erstellt")

    # Matching
    results = []
    total = len(match_df)
    logger.info("Starte Matching-Prozess...")
    for i, mrow in match_df.iterrows():
        key = mrow['block_key']
        cands = crm_index.get(key, [])
        logger.info(f"Prüfe {i+1}/{total}: '{mrow['CRM Name']}' -> {len(cands)} Kandidaten")
        if not cands:
            results.append({'Match':'', 'Score':0})
            continue

        scored = []
        for crow in cands:
            sc, dm, ns, lm, bf = calculate_similarity(mrow, crow)
            scored.append((crow['CRM Name'], sc, dm, ns, lm, bf))
        # Top 3 loggen
        for name, sc, dm, ns, lm, bf in sorted(scored, key=lambda x: x[1], reverse=True)[:3]:
            logger.debug(f"  Kandidat: {name}, Score={sc}, Dom={dm}, Name={ns}, Ort={lm}, Bonus={bf}")

        best_name, best_score, dm, ns, lm, bf = max(scored, key=lambda x: x[1])
        if best_score >= SCORE_THRESHOLD:
            results.append({'Match':best_name, 'Score':best_score})
            logger.info(f" --> Match: '{best_name}' ({best_score}) [Dom={dm}, Name={ns}, Ort={lm}, Bonus={bf}]")
        else:
            results.append({'Match':'', 'Score':best_score})
            logger.info(f" --> Kein Match (Score={best_score}) [Dom={dm}, Name={ns}, Ort={lm}, Bonus={bf}]")

    # Ergebnisse zurückschreiben
    logger.info("Schreibe Ergebnisse ins Sheet...")
    out = pd.DataFrame(results)
    output = match_df[['CRM Name','CRM Website','CRM Ort','CRM Land']].copy()
    output = pd.concat([output.reset_index(drop=True), out], axis=1)
    data = [output.columns.tolist()] + output.values.tolist()
    if sheet.clear_and_write_data(MATCHING_SHEET_NAME, data):
        logger.info("Ergebnisse erfolgreich geschrieben")
    else:
        logger.error("Fehler beim Schreiben ins Google Sheet")

if __name__ == '__main__':
    main()