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

dealfront_enrichment.py aktualisiert

Browse Source
This commit is contained in:
Floke
2025-07-08 18:15:11 +00:00
parent a2b60e2137
commit ce885360a8
1 changed files with 41 additions and 31 deletions
Download Patch File Download Diff File Expand all files Collapse all files

72
dealfront_enrichment.py
View File

@@ -113,42 +113,52 @@ class DealfrontScraper:
return False
def extract_current_page_results(self):
# 1) Warten bis mindestens eine Ergebnis-Zeile sichtbar ist
rows_locator = (By.CSS_SELECTOR, "table#t-result-table tbody tr[id]")
WebDriverWait(self.driver, 20).until(
EC.visibility_of_element_located(rows_locator)
)
# 1) Kurz Implicit-Wait absenken, um nicht an Default-Timeouts zu hängen
self.driver.implicitly_wait(1)
rows = self.driver.find_elements(*rows_locator)
logger.info(f"{len(rows)} Firmen-Zeilen gefunden.")
# 2) Auf das erste Laden der Daten warten (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)
results = []
for i, row in enumerate(rows, 1):
# Name: <div class="t-highlight-text t-highlight-text-product" title="...">
name_divs = row.find_elements(
By.CSS_SELECTOR,
"div.t-highlight-text.t-highlight-text-product"
)
if not name_divs:
logger.warning(f"Zeile {i}: Kein Name-Element gefunden. Überspringe.")
continue
name_elem = name_divs[0]
company_name = (name_elem.get_attribute("title") or name_elem.text).strip() # :contentReference[oaicite:0]{index=0}
try:
logger.info("Extrahiere Ergebnisse von der aktuellen Seite...")
results = []
# Website: externer Link (target="_blank") in der dritten Spalte
ext_links = row.find_elements(By.CSS_SELECTOR, "td:nth-of-type(3) a[target='_blank']")
if ext_links:
href = ext_links[0].get_attribute("href")
website = href.replace("https://", "").replace("http://", "").rstrip("/")
else:
# Fallback: reiner Text in td[3]
cells = row.find_elements(By.TAG_NAME, "td")
website = cells[2].text.strip() if len(cells) >= 3 else ""
# 3) Auf mindestens eine Tabellenzeile warten
rows_selector = (By.CSS_SELECTOR, "table#t-result-table tbody tr[id]")
self.wait.until(EC.presence_of_all_elements_located(rows_selector))
results.append({'name': company_name, 'website': website})
rows = self.driver.find_elements(*rows_selector)
logger.info(f"{len(rows)} Firmen-Zeilen gefunden.")
logger.info(f"Extraktion abgeschlossen: {len(results)} Firmen.")
return results
# 4) Schleife ohne weitere Warte-Blocks
for i, row in enumerate(rows, 1):
# Name: 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()
# Website: bewährter Selector
web_elems = row.find_elements(By.CSS_SELECTOR, "a.text-gray-400.t-highlight-text")
website = web_elems[0].text.strip() if web_elems else ""
results.append({'name': company_name, 'website': website})
logger.info(f"Extraktion abgeschlossen: {len(results)} Firmen.")
return results
except Exception as e:
logger.error(f"Schwerwiegender Fehler bei der Extraktion: {type(e).__name__}", exc_info=True)
self._save_debug_artifacts()
return []
finally:
# 5) Implicit-Wait wieder auf Standard setzen (z.B. 10 s)
self.driver.implicitly_wait(10)
def close(self):