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 17:46:44 +00:00
parent b882124397
commit 57a8a4037e
1 changed files with 32 additions and 44 deletions
Download Patch File Download Diff File Expand all files Collapse all files

76
dealfront_enrichment.py
View File

@@ -113,55 +113,43 @@ class DealfrontScraper:
return False
def extract_current_page_results(self):
# Implicit-Wait kurz absenken für schnellen Fallback bei fehlenden Elementen
self.driver.implicitly_wait(1)
def extract_current_page_results(self):
# 1) Warte auf erste Ergebniszeile (bis zu 20 Sek.)
rows_locator = (By.CSS_SELECTOR, "table#t-result-table tbody tr[id]")
WebDriverWait(self.driver, 20).until(
EC.visibility_of_element_located(rows_locator)
)
# Erst auf das erste Daten-Element warten, dann optional kurzen Puffer
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)
rows = self.driver.find_elements(*rows_locator)
logger.info(f"{len(rows)} Firmen-Zeilen gefunden.")
try:
logger.info("Extrahiere Ergebnisse von der aktuellen Seite...")
results = []
results = []
for i, row in enumerate(rows, 1):
# 2) Name: erstes Profil-Link (<a href="/h/company/...">)
name_links = row.find_elements(
By.XPATH,
".//td[contains(@class,'sticky-column')]//a[contains(@href,'/h/company/')]"
)
if not name_links:
logger.warning(f"Zeile {i}: Kein Name-Link gefunden. Überspringe.")
continue
name_elem = name_links[0]
company_name = (name_elem.get_attribute("title") or name_elem.text).strip()
# Warten bis mindestens eine Daten-Zeile im DOM steht
rows_selector = (By.CSS_SELECTOR, "table#t-result-table tbody tr[id]")
self.wait.until(EC.presence_of_all_elements_located(rows_selector))
# 3) Website: erster externer Link (target="_blank")
ext_links = row.find_elements(By.XPATH, ".//a[@target='_blank']")
if ext_links:
href = ext_links[0].get_attribute("href")
website = href.replace("https://", "").replace("http://", "").rstrip("/")
else:
# Fallback: Text der 3. <td>
cells = row.find_elements(By.TAG_NAME, "td")
website = cells[2].text.strip() if len(cells) >= 3 else ""
rows = self.driver.find_elements(*rows_selector)
logger.info(f"{len(rows)} Firmen-Datenzeilen zur Verarbeitung gefunden.")
results.append({'name': company_name, 'website': website})
for i, row in enumerate(rows, 1):
# Name per find_elements (vermeidet lange Exceptions)
name_elems = row.find_elements(By.CSS_SELECTOR, "td.sticky-column a")
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 per find_elements aus dritter Spalte
web_elems = row.find_elements(By.CSS_SELECTOR, "td:nth-of-type(3) a")
if web_elems:
website = web_elems[0].get_attribute("href").split("://", 1)[1].rstrip("/")
else:
text_elems = row.find_elements(By.CSS_SELECTOR, "td:nth-of-type(3)")
website = text_elems[0].text.strip() if text_elems else ""
results.append({'name': company_name, 'website': website})
logger.info(f"Extraktion abgeschlossen. {len(results)} Firmen gefunden.")
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:
# Implicit-Wait wiederherstellen (Standard 10 s)
self.driver.implicitly_wait(10)
logger.info(f"Extraktion abgeschlossen: {len(results)} Firmen.")
return results
def close(self):