navigation über die seiten

dealfront_enrichment.py

@@ -113,64 +113,94 @@ class DealfrontScraper:
             return False
 
     def extract_current_page_results(self):
-        # 1) Kurz Implicit-Wait absenken
+    def extract_current_page_results(self):
+        # 1) Kurzes Absenken des Implicit-Waits
         self.driver.implicitly_wait(1)
 
-        # 2) Auf das erste Daten-Element warten
+        # 2) Warten auf erstes Firmen-Element
         first_row_locator = (By.CSS_SELECTOR, ".sticky-column a.t-highlight-text")
         self.wait.until(EC.visibility_of_element_located(first_row_locator))
         time.sleep(1)
 
-        try:
-            logger.info("Extrahiere Ergebnisse von der aktuellen Seite...")
-            results = []
+        logger.info("Extrahiere Ergebnisse von der aktuellen Seite...")
+        results = []
 
-            # 3) Warten auf mindestens eine Tabellen-Zeile 
-            rows_selector = (By.CSS_SELECTOR, "table#t-result-table tbody tr[id]")
-            self.wait.until(EC.presence_of_all_elements_located(rows_selector))
-            rows = self.driver.find_elements(*rows_selector)
-            logger.info(f"{len(rows)} Firmen-Zeilen gefunden.")
+        # 3) Warten auf mindestens eine Tabellenzeile
+        rows_selector = (By.CSS_SELECTOR, "table#t-result-table tbody tr[id]")
+        self.wait.until(EC.presence_of_all_elements_located(rows_selector))
+        rows = self.driver.find_elements(*rows_selector)
+        logger.info(f"{len(rows)} Firmen-Zeilen gefunden.")
 
-            for i, row in enumerate(rows, 1):
-                # Name-Extraktion (bewährter Selector)
-                name_elems = row.find_elements(By.CSS_SELECTOR, ".sticky-column a.t-highlight-text")
-                if not name_elems:
-                    logger.warning(f"Zeile {i}: Kein Name-Element gefunden. Überspringe.")
-                    continue
-                name_elem = name_elems[0]
-                company_name = (name_elem.get_attribute("title") or name_elem.text).strip()
+        # 4) Daten sammeln, ohne weitere Sleeps/Exceptions
+        for i, row in enumerate(rows, 1):
+            name_elems = row.find_elements(By.CSS_SELECTOR, ".sticky-column a.t-highlight-text")
+            if not name_elems:
+                logger.warning(f"Zeile {i}: Kein Name-Element gefunden. Überspringe.")
+                continue
+            name_elem = name_elems[0]
+            company_name = (name_elem.get_attribute("title") or name_elem.text).strip()
 
-                # Website-Extraktion aus 3. Spalte
-                web_elems = row.find_elements(By.CSS_SELECTOR, "td:nth-of-type(3) a")
-                if web_elems:
-                    # Link-Text ist der Domain-Name
-                    website = web_elems[0].text.strip()
-                else:
-                    # Fallback: reiner Zellen-Text
-                    cell = row.find_elements(By.CSS_SELECTOR, "td:nth-of-type(3)")
-                    website = cell[0].text.strip() if cell else ""
+            web_elems = row.find_elements(By.CSS_SELECTOR, "td:nth-of-type(3) a")
+            if web_elems:
+                website = web_elems[0].text.strip()
+            else:
+                cell = row.find_elements(By.CSS_SELECTOR, "td:nth-of-type(3)")
+                website = cell[0].text.strip() if cell else ""
 
-                results.append({'name': company_name, 'website': website})
+            results.append({'name': company_name, 'website': website})
 
-            logger.info(f"Extraktion abgeschlossen: {len(results)} Firmen.")
-            return results
+        logger.info(f"Extraktion abgeschlossen: {len(results)} Firmen.")
+        return results
 
-        finally:
-            # 4) Implicit-Wait auf Standard zurücksetzen
-            self.driver.implicitly_wait(10)
+    def click_next_page(self) -> bool:
+        """
+        Klickt auf den 'Next'-Paginator-Button.
+        Gibt False zurück, wenn kein Next-Button (mehr) klickbar ist.
+        """
+        # alle Buttons (Prev, Seiten, Next) abgreifen
+        buttons = self.driver.find_elements(By.CSS_SELECTOR, "nav.eb-pagination a.eb-pagination-button")
+        next_btn = buttons[-1]  # letzter ist Next
+        # Ist er deaktiviert?
+        if not next_btn.is_enabled() or "disabled" in next_btn.get_attribute("class"):
+            return False
 
-            
-    def close(self):
-        if self.driver:
-            logger.info("Schließe den WebDriver.")
-            self.driver.quit()
+        # Merke aktuelle Seite
+        current = self.driver.find_element(
+            By.CSS_SELECTOR,
+            "nav.eb-pagination a.eb-pagination-button.active"
+        ).text
+        next_btn.click()
+
+        # Warte, bis die aktive Seite sich ändert
+        WebDriverWait(self.driver, 10).until(
+            lambda d: d.find_element(
+                By.CSS_SELECTOR,
+                "nav.eb-pagination a.eb-pagination-button.active"
+            ).text != current
+        )
+        return True
+
+    def run(self, search_name):
+        # 1) Login & Suche laden
+        self.login_and_find_list(search_name)
+
+        # 2) Alle Seiten durchgehen
+        all_results = []
+        while True:
+            page_results = self.extract_current_page_results()
+            all_results.extend(page_results)
+            if not self.click_next_page():
+                break
+
+        return all_results
 
 
 if __name__ == "__main__":
     logger.info("Starte Dealfront Automatisierung - DEBUG-MODUS")
     scraper = None
     try:
-        scraper = DealfrontScraper()
+        scraper = DealfrontScraper(driver, wait)
+        results = scraper.run("Facility Management")
         if not scraper.driver:
             raise Exception("WebDriver konnte nicht initialisiert werden.")