Floke/Brancheneinstufung2
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

scrape_fotograf.py aktualisiert

Browse Source
This commit is contained in:
Floke
2025-07-16 14:48:13 +00:00
parent 8623850cd5
commit 15aae7f33c
1 changed files with 26 additions and 23 deletions
Download Patch File Download Diff File Expand all files Collapse all files

49
scrape_fotograf.py
View File

@@ -17,17 +17,19 @@ OUTPUT_FILE = os.path.join(OUTPUT_DIR, 'nutzer_ohne_logins.csv')
LOGIN_URL = 'https://app.fotograf.de/login/login'
# --- Selektoren (zentral verwaltet für einfache Anpassung) ---
# FINALE, KORREKTE VERSION mit Ihren IDs
# GEÄNDERT: Robuste Selektoren für die Alben-Übersicht
SELECTORS = {
"cookie_accept_button": "#CybotCookiebotDialogBodyLevelButtonLevelOptinAllowAll",
"login_user": "#login-email",
"login_pass": "#login-password",
"login_button": "#login-submit",
"job_name": "h1",
"album_rows": "div.table-row-group > div.table-row",
"album_link": "a.text-legacy-azure-200",
"child_name": "div:nth-of-type(4) strong",
"login_count": "div:nth-of-type(6) strong",
# NEU: Dieser Selector findet jede div-Zeile, die einen Link zu einer Galerie enthält. Das ist sehr robust.
"album_rows": "div:has(a[href*='/config_jobs_photos/gallery/'])",
# NEU: Der Link selbst, um ihn innerhalb der Zeile zu finden.
"album_link": "a[href*='/config_jobs_photos/gallery/']",
# NEU: Die Spalte "Zugänge" (Logins). Es ist die 7. Spalte in der Grid-Ansicht.
"login_count": "div:nth-child(7)",
"buyer_link": "a.block:has(span:contains('Käufer'))",
"buyer_email": "div.flex:nth-of-type(4) span"
}
@@ -64,10 +66,8 @@ def load_all_credentials():
with open(CREDENTIALS_FILE, 'r') as f:
return json.load(f)
except FileNotFoundError:
print(f"Fehler: {CREDENTIALS_FILE} nicht gefunden.")
return None
except json.JSONDecodeError:
print(f"Fehler: {CREDENTIALS_FILE} ist keine gültige JSON-Datei.")
return None
def login(driver, username, password):
@@ -75,7 +75,6 @@ def login(driver, username, password):
try:
driver.get(LOGIN_URL)
wait = WebDriverWait(driver, 10)
try:
print("Suche nach Cookie-Banner...")
cookie_button = wait.until(EC.element_to_be_clickable((By.CSS_SELECTOR, SELECTORS["cookie_accept_button"])))
@@ -84,14 +83,11 @@ def login(driver, username, password):
time.sleep(1)
except TimeoutException:
print("Kein Cookie-Banner gefunden, fahre fort.")
print("Fülle Anmeldeformular aus...")
wait.until(EC.visibility_of_element_located((By.CSS_SELECTOR, SELECTORS["login_user"]))).send_keys(username)
driver.find_element(By.CSS_SELECTOR, SELECTORS["login_pass"]).send_keys(password)
print("Klicke auf Login...")
driver.find_element(By.CSS_SELECTOR, SELECTORS["login_button"]).click()
print("Warte auf die nächste Seite...")
wait.until(EC.url_contains('/config_dashboard/index'))
print("Login erfolgreich!")
@@ -105,6 +101,7 @@ def login(driver, username, password):
take_error_screenshot(driver, "login_unexpected")
return False
# GEÄNDERT: Logik an die neue Seitenstruktur angepasst
def process_job(driver, job_url):
print(f"\nVerarbeite Job-URL: {job_url}")
job_id = job_url.split('/')[-1]
@@ -127,21 +124,32 @@ def process_job(driver, job_url):
albums_to_process = []
try:
# Warten, bis die Album-Zeilen geladen sind (mit dem neuen, robusten Selector)
album_rows = wait.until(EC.presence_of_all_elements_located((By.CSS_SELECTOR, SELECTORS["album_rows"])))
print(f"{len(album_rows)} Alben gefunden. Prüfe auf Logins...")
for row in album_rows:
try:
if int(row.find_element(By.CSS_SELECTOR, SELECTORS["login_count"]).text) == 0:
child_name = row.find_element(By.CSS_SELECTOR, SELECTORS["child_name"]).text
album_link = row.find_element(By.CSS_SELECTOR, SELECTORS["album_link"]).get_attribute('href')
albums_to_process.append({"child_name": child_name, "album_detail_url": album_link})
print(f" -> Gefunden: Kind '{child_name}' mit 0 Logins.")
except (NoSuchElementException, ValueError):
# Finde die Login-Anzahl in der 7. Spalte der Zeile
login_count_text = row.find_element(By.CSS_SELECTOR, SELECTORS["login_count"]).text
if int(login_count_text) == 0:
# Finde den Link innerhalb der Zeile
album_link_element = row.find_element(By.CSS_SELECTOR, SELECTORS["album_link"])
child_name = album_link_element.text
album_link = album_link_element.get_attribute('href')
albums_to_process.append({
"child_name": child_name,
"album_detail_url": album_link
})
print(f" -> Gefunden: Album '{child_name}' mit 0 Logins.")
except (NoSuchElementException, ValueError) as e:
# Ignorieren, falls eine Zeile nicht dem Format entspricht
pass
except TimeoutException:
print("Keine Alben auf der Seite gefunden.")
print("Keine Alben auf der Seite gefunden oder Timeout beim Warten.")
take_error_screenshot(driver, "album_list_timeout")
return []
@@ -151,17 +159,13 @@ def process_job(driver, job_url):
try:
print(f" Rufe Detailseite für '{album['child_name']}' auf...")
driver.get(album["album_detail_url"])
buyer_link_element = wait.until(EC.visibility_of_element_located((By.CSS_SELECTOR, SELECTORS["buyer_link"])))
buyer_name = buyer_link_element.text.replace('Käufer ', '').strip()
buyer_page_url = buyer_link_element.get_attribute('href')
print(f" Käufer gefunden: '{buyer_name}'. Rufe Käuferseite auf...")
driver.get(buyer_page_url)
buyer_email = wait.until(EC.visibility_of_element_located((By.CSS_SELECTOR, SELECTORS["buyer_email"]))).text
print(f" E-Mail gefunden: {buyer_email}")
results.append({
"Auftragsname": job_name,
"Kind Vorname": album["child_name"],
@@ -169,7 +173,6 @@ def process_job(driver, job_url):
"Käufer E-Mail": buyer_email,
})
time.sleep(1)
except TimeoutException:
print(f" Fehler: Timeout bei '{album['child_name']}'.")
take_error_screenshot(driver, f"detail_page_timeout_{album['child_name']}")