From 64980a85d3ccd11dcf70e8677c987f00a0a2644f Mon Sep 17 00:00:00 2001
From: Floke <floke.com@gmail.com>
Date: Wed, 16 Jul 2025 15:20:26 +0000
Subject: [PATCH] scrape_fotograf.py aktualisiert

---
 scrape_fotograf.py | 34 +++++++++++++++++-----------------
 1 file changed, 17 insertions(+), 17 deletions(-)

diff --git a/scrape_fotograf.py b/scrape_fotograf.py
index 5f5cc159..8a173b0d 100644
--- a/scrape_fotograf.py
+++ b/scrape_fotograf.py
@@ -16,25 +16,24 @@ OUTPUT_DIR = 'output'
 OUTPUT_FILE = os.path.join(OUTPUT_DIR, 'nutzer_ohne_logins.csv')
 LOGIN_URL = 'https://app.fotograf.de/login/login'
 
-# --- Selektoren (FINALE, KORREKTE VERSION) ---
+# --- Selektoren (FINALE VERSION 2.0) ---
 SELECTORS = {
     "cookie_accept_button": "#CybotCookiebotDialogBodyLevelButtonLevelOptinAllowAll",
     "login_user": "#login-email",
     "login_pass": "#login-password",
     "login_button": "#login-submit",
     "job_name": "h1",
-    # Album-Übersicht (basierend auf Ihrem XPath, der <table> enthielt)
+    # Album-Übersicht (funktioniert)
     "album_overview_rows": "//table/tbody/tr",
-    "album_overview_link": ".//td[2]//a", # Link ist in der 2. Spalte
-    "album_overview_logins": ".//td[7]", # Logins sind in der 7. Spalte
+    "album_overview_link": ".//td[2]//a",
 
-    # Einzelpersonen-Ansicht (innerhalb eines Albums)
-    "person_rows": "//section[.//h3[contains(., 'Einzelfotos')]]//table/tbody/tr",
-    "person_vorname": ".//td[4]",
-    "person_logins": ".//td[6]",
-    "person_buyer_link": ".//td[7]//a",
+    # Einzelpersonen-Ansicht (basierend auf Ihrem finalen HTML)
+    "person_rows": "//div[contains(@class, 'border-legacy-silver-550') and .//span[text()='Logins']]",
+    "person_vorname": ".//span[text()='Vorname']/following-sibling::strong",
+    "person_logins": ".//span[text()='Logins']/following-sibling::strong",
+    "person_buyer_link": ".//a[contains(@data-qa-id, 'guest-access-banner-access-code')]",
 
-    # Käufer-Detailseite
+    # Käufer-Detailseite (funktioniert)
     "buyer_email": "//span[contains(., '@')]" 
 }
 
@@ -64,7 +63,6 @@ def setup_driver():
         print(f"Fehler bei der Initialisierung des WebDrivers: {e}")
         return None
 
-# HIER IST DIE FEHLENDE FUNKTION WIEDER EINGEFÜGT
 def load_all_credentials():
     try:
         with open(CREDENTIALS_FILE, 'r') as f:
@@ -127,8 +125,8 @@ def process_full_job(driver, job_url):
                 album_link = row.find_element(By.XPATH, SELECTORS["album_overview_link"]).get_attribute('href')
                 albums_to_visit.append({"name": album_name, "url": album_link})
             except NoSuchElementException:
-                continue # Ignoriere Zeilen, die kein Album-Link haben (z.B. Team-Bilder)
-        print(f"Sammeln der Album-Links abgeschlossen.")
+                continue
+        print(f"{len(albums_to_visit)} gültige Album-Links gesammelt.")
     except TimeoutException:
         print("Konnte die Album-Liste nicht finden.")
         take_error_screenshot(driver, "album_overview_error")
@@ -139,12 +137,14 @@ def process_full_job(driver, job_url):
         print(f"\n--- Betrete Album: {album['name']} ---")
         driver.get(album['url'])
         try:
+            # Warten, bis die Personen-DIVs geladen sind
             person_rows = wait.until(EC.presence_of_all_elements_located((By.XPATH, SELECTORS["person_rows"])))
             print(f"{len(person_rows)} Personen in diesem Album gefunden.")
 
             for person_row in person_rows:
                 try:
-                    login_count = int(person_row.find_element(By.XPATH, SELECTORS["person_logins"]).text)
+                    login_count_text = person_row.find_element(By.XPATH, SELECTORS["person_logins"]).text
+                    login_count = int(login_count_text)
                     
                     if login_count == 0:
                         vorname = person_row.find_element(By.XPATH, SELECTORS["person_vorname"]).text
@@ -153,7 +153,7 @@ def process_full_job(driver, job_url):
                         buyer_link_element = person_row.find_element(By.XPATH, SELECTORS["person_buyer_link"])
                         buyer_page_url = buyer_link_element.get_attribute('href')
                         
-                        current_window = driver.current_window_handle
+                        # E-Mail in einem neuen Tab holen, um den Kontext nicht zu verlieren
                         driver.execute_script("window.open(arguments[0]);", buyer_page_url)
                         driver.switch_to.window(driver.window_handles[-1])
                         
@@ -168,12 +168,12 @@ def process_full_job(driver, job_url):
                         })
                         
                         driver.close()
-                        driver.switch_to.window(current_window)
+                        driver.switch_to.window(driver.window_handles[0])
                         
                 except (ValueError, NoSuchElementException):
                     continue
         except TimeoutException:
-            print(f"  Keine Personen-Tabelle im Album '{album['name']}' gefunden. Überspringe.")
+            print(f"  Keine Personen-Daten im Album '{album['name']}' gefunden. Überspringe.")
             take_error_screenshot(driver, f"album_{album['name']}_error")
             continue