Update app.py

app.py

@@ -42,47 +42,68 @@ def log_failure(path, reason):
     with open(path, "a") as f:
         f.write(reason + "\n")
 
-def download_image_fast(url, out_path, fail_log, code, suf):
-    import requests, cloudscraper
-    from urllib.parse import urlparse
+import requests, cloudscraper
+from urllib.parse import quote, urlparse
 
+def download_image_fast(url, out_path, fail_log, code, suf):
+    """
+    1) Encode the URL for safe HTTP requests
+    2) Try a short requests.get
+    3) If that fails, try a short cloudscraper.get
+    4) Log *every* status/exception so nothing is silent
+    """
     def log(msg):
         with open(fail_log, "a") as f:
-            f.write(f"{code}{suf}: {msg}\n")
+            f.write(f"{code}{suf}: {msg}\\n")
+
+    # 1) URL‑encode (spaces, parens, etc.)
+    #    Keep scheme and netloc intact, encode path+query
+    parts = urlparse(url)
+    safe_path = quote(parts.path + ("?"+parts.query if parts.query else ""),
+                      safe="/?=&%")
+    safe_url  = f"{parts.scheme}://{parts.netloc}{safe_path}"
 
     headers = {
         "User-Agent": random.choice(USER_AGENTS),
-        "Referer":   urlparse(url).netloc,
+        "Referer":    parts.netloc,
     }
 
-    # 1) Quick requests pass
+    # 2) Try plain requests
     try:
-        r = requests.get(url, headers=headers, timeout=5, stream=True)
+        r = requests.get(safe_url, headers=headers,
+                         timeout=8, stream=True, allow_redirects=True)
+        log(f"requests status={r.status_code}")
         r.raise_for_status()
-        if "image" in r.headers.get("Content-Type", "").lower():
+        ct = r.headers.get("Content-Type","")
+        if "image" in ct.lower():
             with open(out_path, "wb") as f:
-                for chunk in r.iter_content(8192):
+                for chunk in r.iter_content(16384):
                     f.write(chunk)
             return True
-        log(f"requests non-image: {r.headers.get('Content-Type')}")
+        log(f"requests non-image: {ct}")
     except Exception as e:
         log(f"requests error: {e}")
 
-    # 2) cloudscraper fallback
+    # 3) cloudscraper fallback
     try:
-        scr = cloudscraper.create_scraper(browser={"browser":"chrome","platform":"windows"})
-        r   = scr.get(url, headers=headers, timeout=10, stream=True)
+        scr = cloudscraper.create_scraper(
+            browser={"browser":"chrome","platform":"windows"})
+        r   = scr.get(safe_url, headers=headers,
+                      timeout=8, stream=True, allow_redirects=True)
+        log(f"cloudscraper status={r.status_code}")
         r.raise_for_status()
-        if "image" in r.headers.get("Content-Type", "").lower():
+        ct = r.headers.get("Content-Type","")
+        if "image" in ct.lower():
             with open(out_path, "wb") as f:
-                for chunk in r.iter_content(8192):
+                for chunk in r.iter_content(16384):
                     f.write(chunk)
             return True
-        log(f"cloudscraper non-image: {r.headers.get('Content-Type')}")
+        log(f"cloudscraper non-image: {ct}")
     except Exception as e:
         log(f"cloudscraper error: {e}")
 
-    log("download failed")
+    # 4) All failed
+    log("download failed for all methods")
     return False
 
 
@@ -297,12 +318,16 @@ def handle_batch(workbook_path, fmt, w, h):
 
     # Download with tracking
     futures = {}
-    with ThreadPoolExecutor(max_workers=os.cpu_count() or 4) as ex:
+    with ThreadPoolExecutor(max_workers=min(32, CPU_CORES*2)) as ex:
+        futures = []
         for code, url, dl_path, suf in tasks:
-            fut = ex.submit(download_image_fast, url, dl_path, fail_log, code, suf)
-            futures[fut] = (code, suf)
+            futures.append(ex.submit(
+            download_image_fast, url, dl_path, fail_log, code, suf
+            ))
         for fut in as_completed(futures):
             success = fut.result()
+        # failures logged inside download_image_fast already
+
             code, suf = futures[fut]
             if not success:
                 log_failure(fail_log, f"{code}{suf}: download failed")