z230

Covance_UCO3001/download_samples_report_v1.1.py

@@ -1,11 +1,11 @@
 # =============================================================================
 # Název:   download_samples_report_v1.1.py
 # Verze:   1.1
-# Datum:   2026-05-28
+# Datum:   2026-05-29
 # Popis:   Automatické stažení CSV reportu All Samples ze xsp.labcorp.com pro
 #          studie 77242113UCO3001 (study 36940) a 42847922MDD3003 (study 35472).
-#          Aplikuje country filtr CZ, date range od FROM_DATE, čeká na stabilní
-#          Record Count. Výstup do příslušné složky Source/ každé studie.
+#          Aplikuje country filtr CZ, date range od FROM_DATE, čeká na zmizení
+#          "Fetching Data" před exportem. Výstup do Source/ každé studie.
 # =============================================================================
 from playwright.sync_api import sync_playwright
 from datetime import datetime
@@ -31,6 +31,9 @@ TILE_SUFFIX = "allSamples"
 def login(page):
     page.goto(LOGIN_URL)
     page.wait_for_load_state("networkidle", timeout=120000)
+    if not page.get_by_label("Email").is_visible():
+        print(f"Session aktivni, prihlasen: {page.url}")
+        return
     page.get_by_label("Email").fill(EMAIL)
     page.get_by_role("button", name="Next").click()
     page.wait_for_load_state("networkidle", timeout=120000)
@@ -83,19 +86,14 @@ def export_tile(page, tile_label, file_suffix, timestamp, study_id, out_dir):
     page.wait_for_load_state("networkidle", timeout=120000)
     page.wait_for_timeout(3000)
 
-    # Čekej dokud:
-    # 1. se neobjeví "No Data Available" (= record count je 0), nebo
-    # 2. record count není nenulový
-    page.wait_for_function("""() => {
-        const noData = document.querySelector('div.table-row.no-data');
-        if (noData) return true;
-        const countEl = document.querySelector('div.grid-count span');
-        if (countEl) {
-            const n = parseInt(countEl.innerText.trim().replace(/,/g, ''), 10);
-            return !isNaN(n) && n > 0;
-        }
-        return false;
-    }""", timeout=30000)
+    # Čekej až zmizí "Fetching Data": po filtru 5s, pak opakuj kontrolu každých 5s
+    page.wait_for_timeout(5000)
+    for _ in range(24):  # max 2 minuty
+        if not page.get_by_text("Fetching Data").is_visible():
+            break
+        print("  Fetching Data... cekam 5s")
+        page.wait_for_timeout(5000)
+    page.wait_for_timeout(5000)  # extra buffer po zmizení
 
     if page.locator("div.table-row.no-data").is_visible():
         print(f"  Record Count: 0 — preskakuji.")
@@ -128,7 +126,12 @@ if __name__ == "__main__":
         context = p.chromium.launch_persistent_context(
             user_data_dir=PROFILE_DIR,
             headless=False,
-            args=["--disable-blink-features=AutomationControlled", "--start-maximized"],
+            args=[
+                "--disable-blink-features=AutomationControlled",
+                "--start-maximized",
+                "--disable-restore-session-state",
+                "--disable-session-crashed-bubble",
+            ],
             user_agent="Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/136.0.0.0 Safari/537.36",
             accept_downloads=True,
             no_viewport=True,