[34588f42] Performance: Massive Beschleunigung der Analyse durch SQLite-Synchronisierung

- Neue Tabelle JobParticipant speichert detaillierte CSV-Daten von Fotograf.de.
- process_reminder_analysis und process_statistics nutzen nun die lokale Datenbank statt Selenium-Crawling.
- Neuer 'Daten abgleichen' Button im Vorbereitungs-Tab integriert.
- Automatischer Quick-Login Link-Generator basierend auf Zugangscodes.
This commit is contained in:
2026-04-18 13:49:03 +00:00
parent e6061868e6
commit 472f392107
4 changed files with 417 additions and 255 deletions

View File

@@ -1,6 +1,6 @@
# Fotograf.de Scraper & Management UI
**Status:** Production-Ready Microservice (Core Feature: PDF List Generation, QR Cards, Shooting Schedule, **Siblings List**, **Gmail API Integration** & **Automated Release Requests**)
**Status:** Production-Ready Microservice (Core Feature: PDF List Generation, QR Cards, Shooting Schedule, **SQLite Data Sync**, **Gmail API Integration** & **Automated Release Requests**)
Dieser Service modernisiert die alten `Fotograf.de` Skripte, indem er eine robuste, web-basierte UI zur Verwaltung und Automatisierung von Foto-Aufträgen bereitstellt. Er ist als eigenständiger Microservice konzipiert, der unabhängig vom Haupt-Stack läuft.
@@ -10,16 +10,22 @@ Der Service besteht aus zwei Hauptkomponenten:
1. **Backend (Python / FastAPI / Selenium / SQLAlchemy):**
* **Automatisierung:** Nutzt Selenium für das Scraping von `fotograf.de`.
* **Persistenz:** Eine SQLite-Datenbank (`fotograf_jobs.db`) speichert die Auftragsliste, OAuth-Tokens (`GmailToken`), Gutscheincodes (`DiscountCode`) und Teilnehmerdaten (`ReleaseParticipant`).
* **Persistenz:** Eine SQLite-Datenbank (`fotograf_jobs.db`) speichert die Auftragsliste, OAuth-Tokens (`GmailToken`), Gutscheincodes (`DiscountCode`), Teilnehmerdaten (`ReleaseParticipant`), **Auftragsteilnehmer (`JobParticipant`)** und die **Versand-Historie (`ReleaseHistory`)**.
* **PDF-Engine:** Nutzt WeasyPrint für Teilnehmerlisten und ReportLab/PyPDF2 für präzise PDF-Overlays (QR-Karten).
* **API-Integration:** Direkte Anbindung an die **Calendly API (v2)** sowie an die **Gmail API** für direkten E-Mail-Versand und automatisierte Webhook-Antworten.
2. **Frontend (TypeScript / React / Vite / TailwindCSS):**
* **Modernes UI:** Ein vollständig responsives Dashboard mit Tailwind CSS (Kachel-Layout, Tabs für Kiga/Schule).
* **Arbeitsfluss:** Tools sind direkt in der Detailansicht des jeweiligen Auftrags integriert.
* **Arbeitsfluss:** Tools sind in der Detailansicht eines Auftrags in logische Phasen (Vorbereitung, Follow-Up, Statistik) unterteilt.
## ✨ Core Features
### 🚀 Performance-Optimierung (SQLite Sync)
Statt wie früher jedes Mal mühsam durch alle Foto-Alben zu "crawlen", nutzt das System nun eine intelligente Synchronisierung:
* **One-Click Sync:** Über den Button "Daten von Fotograf.de abgleichen" lädt das System die detaillierte Namensliste (CSV) herunter.
* **Lokale Datenbank:** Alle relevanten Infos (E-Mail der Eltern, Login-Zahlen, Bestellstatus, Zugangscodes) werden in der Tabelle `job_participants` gespeichert.
* **Blitzschnelle Analyse:** Nachfass-Mails und Statistiken werden nun in Sekunden (statt Minuten) direkt aus der Datenbank generiert.
### Feature 1: Teilnehmerlisten (Vollständig)
Automatisierter Workflow zum Download und Formatieren der Anmeldelisten von `fotograf.de` als sortiertes PDF inkl. "Kinderfotos Erding" Branding.
@@ -28,43 +34,37 @@ Spezielles Modul für Familien-Mini-Shootings:
* **QR-Karten-Andruck:** Präzises Overlay von Name, Kinderanzahl und Uhrzeit inkl. automatischer **Einwilligungs-Checkbox (☑)** aus Calendly-Daten.
* **Termin-Übersichtsliste:** Generiert eine A4-Tabelle für den Shooting-Tag im 6-Minuten-Takt inkl. Lückenfüller.
### Feature 3: Nachfass-E-Mails & Gmail Direkt-Versand (Vollständig)
Identifizierung von Nicht-Käufern und automatisierter Massenversand personalisierter E-Mails via Gmail API.
### Feature 3: Nachfass-E-Mails & Gmail Direkt-Versand (Optimiert)
Identifizierung von Nicht-Käufern (0-1 Logins, keine Bestellung) basierend auf den synchronisierten Datenbank-Daten.
* **Vorschau-Modus:** Ermöglicht das Durchklicken der personalisierten E-Mails an jeden Empfänger vor dem eigentlichen Versand.
* **Quick-Login Automation:** Die Login-Links (`https://www.kinderfotos-erding.de/a/{code}`) werden automatisch generiert.
### Feature 4: Verkaufs-Statistiken (Vollständig)
Detaillierte Analyse des Kaufverhaltens pro Album mit Echtzeit-Fortschrittsanzeige.
### Feature 4: Verkaufs-Statistiken (Optimiert)
Detaillierte Analyse des Kaufverhaltens pro Gruppe/Klasse basierend auf den lokalen Datenbank-Einträgen.
### Feature 5: Geschwisterliste (Einrichtungsintern) (Vollständig)
Tool zur Identifizierung von Geschwistergruppen innerhalb einer Einrichtung inkl. Cross-Check mit Calendly-Buchungen und speziellen Geschwister-QR-Karten.
### Feature 6: Freigabeanfragen & Gutschein-Automation (Vollständig - Neu April 2026)
### Feature 6: Freigabeanfragen & Gutschein-Automation (Vollständig)
Vollautomatisierter DSGVO-Workflow zur Einholung von Veröffentlichungsgenehmigungen:
* **Schlanker Versand:** Manuelle Eingabe von Empfängern (E-Mail, Vorname, Kindernamen) für gezielte Anfragen.
* **Intelligente Personalisierung:** Automatische Bereinigung von Einrichtungsnamen (entfernt "Kindergarten" und Jahreszahlen).
* **Schlanker Versand:** Manuelle Eingabe von Empfängern (E-Mail, Vorname, Kindernamen) mit **E-Mail-Vorschau**.
* **Versand-Planung:** Einstellbare Versandzeit (Berlin Timezone) via Hintergrund-Tasks.
* **Webhook-Integration:** Direkte Anbindung an **Google Forms**. Bei Absenden des Freigabe-Formulars wird automatisch:
1. Ein freier Gutscheincode aus der DB reserviert.
2. Eine personalisierte Dankes-E-Mail mit dem Code und einer bebilderten Einlöse-Anleitung versendet.
* **Gutschein-Management:** UI zum Hochladen und Überwachen des Gutschein-Pools.
* **Webhook-Integration:** Direkte Anbindung an **Google Forms**. Bei Absenden des Freigabe-Formulars wird automatisch ein Gutscheincode reserviert und eine Dankes-E-Mail versendet.
* **Antwort-Übersicht:** Tabelle aller eingegangenen Freigaben inkl. zugewiesenem Code und Zeitstempel.
---
## 🛠️ Technische Details & Sicherheit
* **Sicherer Test-Modus:** Über die Umgebungsvariable `DEV_MODE_EMAIL_RECIPIENT` können alle ausgehenden E-Mails (Anfragen & Gutscheine) global an eine Test-Adresse umgeleitet werden.
* **Zeitzonen:** Durchgängige Verwendung von `Europe/Berlin` für alle zeitgesteuerten Operationen.
* **E-Mail Signatur:** Die offizielle HTML-Signatur von "Kinderfotos Erding" wird automatisch an alle ausgehenden E-Mails (auch vom Backend) angehängt.
* **Gmail OAuth:** Persistente Speicherung der Refresh-Tokens in der Datenbank ermöglicht dauerhaften Betrieb ohne erneutes Einloggen.
* **BCC-Kontrolle:** Jede vom System versendete E-Mail sendet automatisch eine Blindkopie (BCC) an `kontakt@kinderfotos-erding.de`.
* **Versand-Historie:** Alle Aussendungen (Anzahl Empfänger, Zeitpunkt) werden in der Tabelle `release_history` protokolliert.
* **Sicherer Test-Modus:** Über `DEV_MODE_EMAIL_RECIPIENT` können alle E-Mails global an eine Test-Adresse umgeleitet werden.
* **Zeitzonen:** Durchgängige Verwendung von `Europe/Berlin`.
* **Gmail OAuth:** Persistente Speicherung der Refresh-Tokens in der Datenbank.
## 🚀 Deployment & Konfiguration
Der Service wird über die Haupt-`docker-compose.yml` des Projekts verwaltet.
### Umgebungsvariablen (`.env`)
Wichtige neue Variablen in `/fotograf-de-scraper/.env`:
* `DEV_MODE_EMAIL_RECIPIENT`: (Optional) E-Mail für Umleitung im Testbetrieb.
* `google_fotograf_client_id` / `google_fotograf_secret`: OAuth Credentials.
* `CALENDLY_TOKEN`: API Zugriff.
### URLs
* **Frontend:** `https://floke-ai.duckdns.org/fotograf-de/`
* **Webhook für Google Forms:** `https://floke-ai.duckdns.org/fotograf-de-api/api/publish-request/webhook`

View File

@@ -49,6 +49,22 @@ class ReleaseHistory(Base):
recipient_count = Column(Integer)
scheduled_time = Column(String, nullable=True)
class JobParticipant(Base):
__tablename__ = "job_participants"
id = Column(Integer, primary_key=True)
job_id = Column(String, index=True)
child_id = Column(String, nullable=True)
vorname_kind = Column(String, nullable=True)
nachname_kind = Column(String, nullable=True)
vorname_eltern = Column(String, nullable=True)
nachname_eltern = Column(String, nullable=True)
email_eltern = Column(String, nullable=True)
zugangscode = Column(String, index=True)
gruppe = Column(String, nullable=True)
logins = Column(Integer, default=0)
has_orders = Column(Integer, default=0) # 0 for false, 1 for true
last_synced = Column(DateTime, default=datetime.datetime.utcnow)
Base.metadata.create_all(bind=engine)
def get_db():

View File

@@ -16,7 +16,7 @@ from fastapi.middleware.cors import CORSMiddleware
from fastapi.responses import FileResponse
from typing import List, Dict, Any, Optional
from sqlalchemy.orm import Session
from database import get_db, Job as DBJob, engine, Base
from database import get_db, Job as DBJob, engine, Base, JobParticipant, SessionLocal
import math
import uuid
@@ -141,6 +141,120 @@ def get_logo_base64():
logger.warning(f"Logo file not found at {logo_path}")
return None
def sync_job_participants(job_id: str, account_type: str, db: Session):
logger.info(f"Syncing participants for job {job_id} ({account_type})")
username = os.getenv(f"{account_type.upper()}_USER")
password = os.getenv(f"{account_type.upper()}_PW")
with tempfile.TemporaryDirectory() as temp_dir:
driver = setup_driver(download_path=temp_dir)
try:
if not login(driver, username, password):
raise Exception("Login failed during sync.")
# Navigate to job names list
job_url = f"https://app.fotograf.de/config_jobs_settings/index/{job_id}"
driver.get(job_url)
wait = WebDriverWait(driver, 20)
personen_tab = wait.until(EC.presence_of_element_located((By.CSS_SELECTOR, "[data-qa-id='link:photo-jobs-tabs-names_list']")))
driver.execute_script("arguments[0].click();", personen_tab)
export_btn = wait.until(EC.presence_of_element_located((By.CSS_SELECTOR, SELECTORS["export_dropdown"])))
driver.execute_script("arguments[0].click();", export_btn)
time.sleep(1)
csv_btn = wait.until(EC.presence_of_element_located((By.CSS_SELECTOR, SELECTORS["export_csv_link"])))
driver.execute_script("arguments[0].click();", csv_btn)
# Wait for download
csv_file = None
for _ in range(30):
files = [f for f in os.listdir(temp_dir) if f.endswith('.csv')]
if files:
csv_file = os.path.join(temp_dir, files[0])
break
time.sleep(1)
if not csv_file:
raise Exception("CSV download timeout during sync.")
# Parse CSV
df = None
for sep in [";", ","]:
try:
df = pd.read_csv(csv_file, sep=sep, encoding="utf-8-sig")
if len(df.columns) > 1: break
except: continue
if df is None:
df = pd.read_csv(csv_file, sep=";", encoding="latin1")
# Clean columns
df.columns = df.columns.str.strip().str.replace("\"", "")
# Map columns - based on user feedback
# Expected columns: Child ID, Email der Eltern (1), Vorname Eltern (1), Nachname Eltern (1), Vorname Kind, Zugangscode (1), Logins (1), Bestellungen
def get_col(df, patterns):
for p in patterns:
for col in df.columns:
if p.lower() in col.lower():
return col
return None
col_child_id = get_col(df, ["Child ID"])
col_email = get_col(df, ["Email der Eltern", "E-Mail der Eltern"])
col_parent_vn = get_col(df, ["Vorname Eltern", "Parent First Name"])
col_parent_nn = get_col(df, ["Nachname Eltern", "Parent Last Name"])
col_child_vn = get_col(df, ["Vorname Kind", "Child First Name"])
col_child_nn = get_col(df, ["Nachname Kind", "Child Last Name"])
col_code = get_col(df, ["Zugangscode", "Access Code"])
col_group = get_col(df, ["Gruppe", "Klasse", "Group", "Class"])
col_logins = get_col(df, ["Logins"])
col_orders = get_col(df, ["Bestellungen", "Orders"])
# Delete old entries for this job
db.query(JobParticipant).filter(JobParticipant.job_id == job_id).delete()
added = 0
for _, row in df.iterrows():
try:
logins_val = 0
try: logins_val = int(row[col_logins]) if col_logins and pd.notna(row[col_logins]) else 0
except: pass
orders_val = 0
if col_orders and pd.notna(row[col_orders]):
val = str(row[col_orders]).lower()
if val and val != "0" and val != "nein" and val != "false":
orders_val = 1
participant = JobParticipant(
job_id=job_id,
child_id=str(row[col_child_id]) if col_child_id and pd.notna(row[col_child_id]) else None,
vorname_kind=str(row[col_child_vn]) if col_child_vn and pd.notna(row[col_child_vn]) else None,
nachname_kind=str(row[col_child_nn]) if col_child_nn and pd.notna(row[col_child_nn]) else None,
vorname_eltern=str(row[col_parent_vn]) if col_parent_vn and pd.notna(row[col_parent_vn]) else None,
nachname_eltern=str(row[col_parent_nn]) if col_parent_nn and pd.notna(row[col_parent_nn]) else None,
email_eltern=str(row[col_email]).strip().lower() if col_email and pd.notna(row[col_email]) else None,
zugangscode=str(row[col_code]) if col_code and pd.notna(row[col_code]) else None,
gruppe=str(row[col_group]) if col_group and pd.notna(row[col_group]) else None,
logins=logins_val,
has_orders=orders_val
)
db.add(participant)
added += 1
except Exception as e:
logger.warning(f"Error adding participant row: {e}")
db.commit()
logger.info(f"Sync complete. {added} participants stored for job {job_id}")
return added
finally:
driver.quit()
def generate_pdf_from_csv(csv_path: str, institution: str, date_info: str, list_type: str, output_path: str):
logger.info(f"Generating PDF for {institution} from {csv_path}")
df = None
@@ -488,264 +602,142 @@ def get_jobs_list(driver) -> List[Dict[str, Any]]:
task_store: Dict[str, Dict[str, Any]] = {}
def process_statistics(task_id: str, job_id: str, account_type: str):
logger.info(f"Task {task_id}: Starting statistics calculation for job {job_id}")
task_store[task_id] = {"status": "running", "progress": "Initialisiere Browser...", "result": None}
username = os.getenv(f"{account_type.upper()}_USER")
password = os.getenv(f"{account_type.upper()}_PW")
driver = None
logger.info(f"Task {task_id}: Starting fast statistics calculation for job {job_id}")
task_store[task_id] = {"status": "running", "progress": "Synchronisiere Daten von Fotograf.de...", "result": None}
db = SessionLocal()
try:
driver = setup_driver()
if not driver or not login(driver, username, password):
task_store[task_id] = {"status": "error", "progress": "Login fehlgeschlagen. Überprüfe die Zugangsdaten."}
return
task_store[task_id]["progress"] = f"Lade Alben-Übersicht für Auftrag..."
albums_overview_url = f"https://app.fotograf.de/config_jobs_photos/index/{job_id}"
logger.info(f"Navigating to albums: {albums_overview_url}")
driver.get(albums_overview_url)
wait = WebDriverWait(driver, 15)
albums_to_visit = []
# 1. Sync data from CSV
try:
album_rows = wait.until(EC.presence_of_all_elements_located((By.XPATH, SELECTORS["album_overview_rows"])))
for row in album_rows:
try:
album_link = row.find_element(By.XPATH, SELECTORS["album_overview_link"])
albums_to_visit.append({"name": album_link.text, "url": album_link.get_attribute('href')})
except NoSuchElementException:
continue
except TimeoutException:
task_store[task_id] = {"status": "error", "progress": "Konnte die Album-Liste nicht finden."}
return
sync_participants(job_id, account_type, db)
except Exception as sync_err:
logger.error(f"Sync failed during statistics: {sync_err}")
count = db.query(JobParticipant).filter(JobParticipant.job_id == job_id).count()
if count == 0:
task_store[task_id] = {"status": "error", "progress": f"Synchronisierung fehlgeschlagen: {str(sync_err)}"}
return
total_albums = len(albums_to_visit)
task_store[task_id]["progress"] = f"{total_albums} Alben gefunden. Starte Auswertung..."
# 2. Query DB and group by 'gruppe'
task_store[task_id]["progress"] = "Berechne Statistiken..."
statistics = []
# Get all participants for this job
participants = db.query(JobParticipant).filter(JobParticipant.job_id == job_id).all()
for index, album in enumerate(albums_to_visit):
album_name = album['name']
task_store[task_id]["progress"] = f"Bearbeite Album {index + 1}/{total_albums}: '{album_name}'..."
driver.get(album['url'])
# Group by group
groups = {}
for p in participants:
g_name = p.gruppe or "Unbekannt"
if g_name not in groups:
groups[g_name] = {
"Album": g_name,
"Kinder_insgesamt": 0,
"Kinder_mit_Käufen": 0,
"Kinder_Alle_Bilder_gekauft": 0 # Not available in CSV, setting to 0 or estimates
}
groups[g_name]["Kinder_insgesamt"] += 1
if p.has_orders:
groups[g_name]["Kinder_mit_Käufen"] += 1
try:
total_codes_text = wait.until(EC.visibility_of_element_located((By.XPATH, SELECTORS["access_code_count"]))).text
num_pages = math.ceil(int(total_codes_text) / 20)
total_children_in_album = 0
children_with_purchase = 0
children_with_all_purchased = 0
for page_num in range(1, num_pages + 1):
task_store[task_id]["progress"] = f"Bearbeite Album {index + 1}/{total_albums}: '{album_name}' (Seite {page_num}/{num_pages})..."
if page_num > 1:
driver.get(album['url'] + f"?page_guest_accesses={page_num}")
person_rows = wait.until(EC.presence_of_all_elements_located((By.XPATH, SELECTORS["person_rows"])))
for person_row in person_rows:
total_children_in_album += 1
try:
photo_container = person_row.find_element(By.XPATH, "./following-sibling::div[1]")
num_total_photos = len(photo_container.find_elements(By.XPATH, SELECTORS["person_all_photos"]))
num_purchased_photos = len(photo_container.find_elements(By.XPATH, SELECTORS["person_purchased_photos"]))
num_access_cards = len(photo_container.find_elements(By.XPATH, SELECTORS["person_access_card_photo"]))
buyable_photos = num_total_photos - num_access_cards
if num_purchased_photos > 0:
children_with_purchase += 1
if buyable_photos > 0 and buyable_photos == num_purchased_photos:
children_with_all_purchased += 1
except NoSuchElementException:
continue
statistics.append({
"Album": album_name,
"Kinder_insgesamt": total_children_in_album,
"Kinder_mit_Käufen": children_with_purchase,
"Kinder_Alle_Bilder_gekauft": children_with_all_purchased
})
except Exception as e:
logger.error(f"Fehler bei Auswertung von Album '{album_name}': {e}")
continue
statistics = list(groups.values())
statistics.sort(key=lambda x: x["Album"])
task_store[task_id] = {
"status": "completed",
"progress": "Auswertung erfolgreich abgeschlossen!",
"progress": "Statistik erfolgreich berechnet!",
"result": statistics
}
except Exception as e:
logger.exception(f"Unexpected error in task {task_id}")
logger.exception(f"Unexpected error in statistics task {task_id}")
task_store[task_id] = {"status": "error", "progress": f"Unerwarteter Fehler: {str(e)}"}
finally:
if driver:
logger.debug(f"Task {task_id}: Closing driver.")
driver.quit()
db.close()
def process_reminder_analysis(task_id: str, job_id: str, account_type: str):
logger.info(f"Task {task_id}: Starting reminder analysis for job {job_id}")
task_store[task_id] = {"status": "running", "progress": "Initialisiere Browser...", "result": None}
username = os.getenv(f"{account_type.upper()}_USER")
password = os.getenv(f"{account_type.upper()}_PW")
driver = None
logger.info(f"Task {task_id}: Starting fast reminder analysis for job {job_id}")
task_store[task_id] = {"status": "running", "progress": "Synchronisiere Daten von Fotograf.de...", "result": None}
db = SessionLocal()
try:
driver = setup_driver()
if not driver or not login(driver, username, password):
task_store[task_id] = {"status": "error", "progress": "Login fehlgeschlagen."}
return
wait = WebDriverWait(driver, 15)
# 1. Navigate to albums overview
albums_overview_url = f"https://app.fotograf.de/config_jobs_photos/index/{job_id}"
task_store[task_id]["progress"] = "Lade Alben-Übersicht..."
driver.get(albums_overview_url)
albums_to_visit = []
# 1. Sync data from CSV (This takes ~20s and gets all parent emails, logins and orders)
try:
album_rows = wait.until(EC.presence_of_all_elements_located((By.XPATH, SELECTORS["album_overview_rows"])))
for row in album_rows:
try:
album_link = row.find_element(By.XPATH, SELECTORS["album_overview_link"])
albums_to_visit.append({"name": album_link.text, "url": album_link.get_attribute('href')})
except NoSuchElementException:
continue
except TimeoutException:
task_store[task_id] = {"status": "error", "progress": "Konnte die Album-Liste nicht finden."}
sync_participants(job_id, account_type, db)
except Exception as sync_err:
logger.error(f"Sync failed during reminder analysis: {sync_err}")
# Continue anyway if we have some data, or fail if we have none
count = db.query(JobParticipant).filter(JobParticipant.job_id == job_id).count()
if count == 0:
task_store[task_id] = {"status": "error", "progress": f"Synchronisierung fehlgeschlagen: {str(sync_err)}"}
return
# 2. Query DB for potential candidates (Logins <= 1 and No Orders)
task_store[task_id]["progress"] = "Analysiere Datenbank-Einträge..."
candidates = db.query(JobParticipant).filter(
JobParticipant.job_id == job_id,
JobParticipant.has_orders == 0,
JobParticipant.logins <= 1,
JobParticipant.email_eltern != "",
JobParticipant.email_eltern != None
).all()
if not candidates:
task_store[task_id] = {
"status": "completed",
"progress": "Keine passenden Empfänger (0-1 Logins, keine Bestellung) gefunden.",
"result": []
}
return
raw_results = []
total_albums = len(albums_to_visit)
for index, album in enumerate(albums_to_visit):
album_name = album['name']
task_store[task_id]["progress"] = f"Album {index+1}/{total_albums}: '{album_name}'..."
driver.get(album['url'])
try:
total_codes_text = wait.until(EC.visibility_of_element_located((By.XPATH, SELECTORS["access_code_count"]))).text
num_pages = math.ceil(int(total_codes_text) / 20)
for page_num in range(1, num_pages + 1):
task_store[task_id]["progress"] = f"Album {index+1}/{total_albums}: '{album_name}' (Seite {page_num}/{num_pages})..."
if page_num > 1:
driver.get(album['url'] + f"?page_guest_accesses={page_num}")
person_rows = wait.until(EC.presence_of_all_elements_located((By.XPATH, SELECTORS["person_rows"])))
num_persons = len(person_rows)
for i in range(num_persons):
# Re-locate rows to avoid stale element reference
person_rows = wait.until(EC.presence_of_all_elements_located((By.XPATH, SELECTORS["person_rows"])))
person_row = person_rows[i]
login_count_text = person_row.find_element(By.XPATH, ".//span[text()='Logins']/following-sibling::strong").text
# Only interested in people with 0 or 1 logins (potential reminders)
# Actually, if they haven't bought yet, they might need a reminder regardless of logins,
# but the legacy logic uses login_count <= 1.
# Let's stick to the legacy logic for now.
if int(login_count_text) <= 1:
vorname = person_row.find_element(By.XPATH, ".//span[text()='Vorname']/following-sibling::strong").text
try:
photo_container = person_row.find_element(By.XPATH, "./following-sibling::div[1]")
purchase_icons = photo_container.find_elements(By.XPATH, ".//img[@alt='Bestellungen mit diesem Foto']")
if len(purchase_icons) > 0:
continue
except NoSuchElementException:
pass
# Potential candidate
access_code_page_url = person_row.find_element(By.XPATH, ".//a[contains(@data-qa-id, 'guest-access-banner-access-code')]").get_attribute('href')
# Open in new tab or navigate back and forth?
# Scraper.py navigates back and forth.
driver.get(access_code_page_url)
try:
wait.until(EC.visibility_of_element_located((By.XPATH, "//a[@id='quick-login-url']")))
quick_login_url = driver.find_element(By.XPATH, "//a[@id='quick-login-url']").get_attribute('href')
potential_buyer_element = driver.find_element(By.XPATH, "//a[contains(@href, '/config_customers/view_customer')]")
buyer_name = potential_buyer_element.text
potential_buyer_element.click()
email = wait.until(EC.visibility_of_element_located((By.XPATH, "//span[contains(., '@')]"))).text
raw_results.append({
"child_name": vorname,
"buyer_name": buyer_name,
"email": email,
"quick_login": quick_login_url
})
except Exception as e:
logger.warning(f"Error getting details for {vorname}: {e}")
# Go back to the album page
driver.get(album['url'] + (f"?page_guest_accesses={page_num}" if page_num > 1 else ""))
wait.until(EC.presence_of_element_located((By.XPATH, SELECTORS["person_rows"])))
except Exception as e:
logger.error(f"Fehler bei Album '{album_name}': {e}")
continue
# Aggregate Results
task_store[task_id]["progress"] = "Aggregiere Ergebnisse..."
aggregated_data = {}
for res in raw_results:
email = res['email']
child_name = "Familienbilder" if res['child_name'] == "Familie" else res['child_name']
html_link = f'<a href="{res["quick_login"]}">Fotos von {child_name}</a>'
if email not in aggregated_data:
aggregated_data[email] = {
'buyer_first_name': res['buyer_name'].split(' ')[0],
'email': email,
'children': [child_name],
'links': [html_link]
# 3. Aggregate results by Email
aggregation = {}
for c in candidates:
email = c.email_eltern
if email not in aggregation:
aggregation[email] = {
"email": email,
"parent_name": c.vorname_eltern if c.vorname_eltern else "Liebe Eltern",
"children": [],
"links": []
}
else:
if child_name not in aggregated_data[email]['children']:
aggregated_data[email]['children'].append(child_name)
aggregated_data[email]['links'].append(html_link)
final_list = []
for email, data in aggregated_data.items():
names = data['children']
if len(names) > 2:
names_str = ', '.join(names[:-1]) + ' und ' + names[-1]
else:
names_str = ' und '.join(names)
# Add child name
child_name = c.vorname_kind or ""
child_label = "Familienbilder" if child_name.lower() == "familie" else child_name
if child_label and child_label not in aggregation[email]["children"]:
aggregation[email]["children"].append(child_label)
final_list.append({
'Name Käufer': data['buyer_first_name'],
'E-Mail-Adresse Käufer': email,
'Kindernamen': names_str,
'LinksHTML': '<br><br>'.join(data['links'])
# Add Quick Login Link
link = f"https://www.kinderfotos-erding.de/a/{c.zugangscode}"
html_link = f'<a href="{link}">Fotos von {child_label}</a>'
if html_link not in aggregation[email]["links"]:
aggregation[email]["links"].append(html_link)
# 4. Format for Supermailer/Gmail
final_result = []
for email, data in aggregation.items():
children_str = " und ".join(data["children"]) if len(data["children"]) > 1 else (data["children"][0] if data["children"] else "Eurem Kind")
links_html = "".join([f"{l}<br>" for l in data["links"]])
final_result.append({
"E-Mail-Adresse Käufer": email,
"Name Käufer": data["parent_name"],
"Kindernamen": children_str,
"Anzahl Kinder": len(data["children"]),
"LinksHTML": links_html
})
task_store[task_id] = {
"status": "completed",
"progress": "Analyse abgeschlossen!",
"result": final_list
"progress": f"Analyse fertig! {len(final_result)} Empfänger identifiziert.",
"result": final_result
}
except Exception as e:
logger.exception(f"Error in task {task_id}")
task_store[task_id] = {"status": "error", "progress": f"Fehler: {str(e)}"}
finally:
if driver: driver.quit()
db.close()
from fastapi import FastAPI, HTTPException, Depends, BackgroundTasks, UploadFile, File, Form
from fastapi.middleware.cors import CORSMiddleware
@@ -1092,6 +1084,124 @@ async def send_bulk_emails(request: BulkEmailRequest, db: Session = Depends(get_
"failed": failed_emails
}
def sync_participants(job_id: str, account_type: str, db: Session):
logger.info(f"Syncing participants for job {job_id} ({account_type})")
username = os.getenv(f"{account_type.upper()}_USER")
password = os.getenv(f"{account_type.upper()}_PW")
with tempfile.TemporaryDirectory() as temp_dir:
driver = setup_driver(download_path=temp_dir)
try:
if not login(driver, username, password):
raise Exception("Login failed.")
# Navigate to the Persons tab
job_url = f"https://app.fotograf.de/config_jobs_settings/index/{job_id}"
driver.get(job_url)
wait = WebDriverWait(driver, 30)
personen_tab = wait.until(EC.presence_of_element_located((By.CSS_SELECTOR, "[data-qa-id='link:photo-jobs-tabs-names_list']")))
driver.execute_script("arguments[0].click();", personen_tab)
# Click Export -> CSV
export_btn = wait.until(EC.presence_of_element_located((By.CSS_SELECTOR, SELECTORS["export_dropdown"])))
driver.execute_script("arguments[0].click();", export_btn)
time.sleep(1)
csv_btn = wait.until(EC.presence_of_element_located((By.CSS_SELECTOR, SELECTORS["export_csv_link"])))
driver.execute_script("arguments[0].click();", csv_btn)
# Wait for download
csv_file = None
for _ in range(45):
files = os.listdir(temp_dir)
csv_files = [f for f in files if f.endswith('.csv')]
if csv_files:
csv_file = os.path.join(temp_dir, csv_files[0])
break
time.sleep(1)
if not csv_file:
raise Exception("CSV download timed out.")
# Read CSV with pandas
df = None
for sep in [";", ","]:
try:
df = pd.read_csv(csv_file, sep=sep, encoding="utf-8-sig")
if len(df.columns) > 1: break
except: continue
if df is None: raise Exception("Could not parse CSV.")
# Clean columns
df.columns = df.columns.str.strip().str.replace("\"", "")
logger.debug(f"Sync CSV Columns: {list(df.columns)}")
# Column Mapping
mapping = {
"Child ID": "child_id",
"Email der Eltern (1)": "email_eltern",
"Vorname Eltern (1)": "vorname_eltern",
"Nachname Eltern (1)": "nachname_eltern",
"Vorname Kind": "vorname_kind",
"Nachname Kind": "nachname_kind",
"Zugangscode (1)": "zugangscode",
"Logins (1)": "logins",
"Bestellungen": "has_orders",
"Gruppe": "gruppe",
"Klasse": "gruppe"
}
# Upsert into database
for _, row in df.iterrows():
code = str(row.get("Zugangscode (1)", "")).strip()
if not code or code == "nan": continue
def clean_val(val):
v = str(val).strip()
return "" if v.lower() == "nan" else v
# Determine order status
orders_val = str(row.get("Bestellungen", "0")).lower()
has_orders = 1 if (orders_val != "0" and orders_val != "nan" and orders_val != "") else 0
# Determine logins
logins_val = row.get("Logins (1)", 0)
try: logins = int(float(logins_val))
except: logins = 0
participant = db.query(JobParticipant).filter(JobParticipant.job_id == job_id, JobParticipant.zugangscode == code).first()
if not participant:
participant = JobParticipant(job_id=job_id, zugangscode=code)
db.add(participant)
participant.child_id = clean_val(row.get("Child ID"))
participant.vorname_kind = clean_val(row.get("Vorname Kind"))
participant.nachname_kind = clean_val(row.get("Nachname Kind"))
participant.vorname_eltern = clean_val(row.get("Vorname Eltern (1)"))
participant.nachname_eltern = clean_val(row.get("Nachname Eltern (1)"))
participant.email_eltern = clean_val(row.get("Email der Eltern (1)")).lower()
participant.gruppe = clean_val(row.get("Gruppe", row.get("Klasse")))
participant.logins = logins
participant.has_orders = has_orders
participant.last_synced = datetime.datetime.utcnow()
db.commit()
logger.info(f"Successfully synced {len(df)} participants for job {job_id}")
return len(df)
finally:
driver.quit()
@app.post("/api/jobs/{job_id}/sync-participants")
async def sync_participants_api(job_id: str, account_type: str, db: Session = Depends(get_db)):
try:
count = sync_participants(job_id, account_type, db)
return {"status": "success", "count": count}
except Exception as e:
logger.exception("Sync failed")
raise HTTPException(status_code=500, detail=str(e))
@app.get("/api/jobs/{job_id}/generate-pdf")
async def generate_pdf(job_id: str, account_type: str, db: Session = Depends(get_db)):
logger.info(f"API Request: Generate PDF for job {job_id} ({account_type})")

View File

@@ -45,6 +45,24 @@ function App() {
const [isReminderRunning, setIsReminderRunning] = useState(false);
const [latestFile, setLatestFile] = useState<any>(null);
const [isGmailAuthenticated, setIsGmailAuthenticated] = useState(false);
const [isSyncing, setIsSyncing] = useState(false);
const handleSyncParticipants = async (job: Job) => {
setIsSyncing(true);
try {
const response = await fetch(`${API_BASE_URL}/api/jobs/${job.id}/sync-participants?account_type=${activeTab}`, {
method: 'POST'
});
if (response.ok) {
alert("Daten erfolgreich mit Fotograf.de synchronisiert!");
} else {
alert("Synchronisierung fehlgeschlagen.");
}
} catch (e) {
alert("Netzwerkfehler.");
}
setIsSyncing(false);
};
// Email States
const [reminderResult, setReminderResult] = useState<any[] | null>(null);
@@ -958,7 +976,25 @@ function App() {
<option key={et.uri} value={et.name}>{et.name}</option>
))}
</select>
<p className="text-xs text-gray-500 mt-2">Wird für QR-Karten und die Terminübersicht benötigt.</p>
<p className="text-xs text-gray-500 mt-2 mb-4">Wird für QR-Karten und die Terminübersicht benötigt.</p>
<div className="pt-4 border-t border-gray-100">
<button
onClick={() => handleSyncParticipants(selectedJob)}
disabled={isSyncing}
className="w-full px-3 py-2 bg-white border border-indigo-200 text-indigo-600 text-xs font-bold rounded-lg hover:bg-indigo-50 transition-colors flex items-center justify-center gap-2"
>
{isSyncing ? (
<>
<svg className="animate-spin h-3 w-3" viewBox="0 0 24 24" fill="none"><circle className="opacity-25" cx="12" cy="12" r="10" stroke="currentColor" strokeWidth="4" /><path className="opacity-75" fill="currentColor" d="M4 12a8 8 0 018-8V0C5.373 0 0 5.373 0 12h4zm2 5.291A7.962 7.962 0 014 12H0c0 3.042 1.135 5.824 3 7.938l3-2.647z" /></svg>
Sync läuft...
</>
) : (
<>🔄 Daten von Fotograf.de abgleichen</>
)}
</button>
<p className="text-[10px] text-gray-400 mt-2">Aktualisiert E-Mails, Logins & Bestellstatus.</p>
</div>
</div>
{/* Actions */}