feat: pre-download filter for videos >= MIN_VIDEO_SIZE_MB (default 200 MB)

jd-webgui/app.py

@@ -57,6 +57,8 @@ BASIC_AUTH_USER = os.environ.get("BASIC_AUTH_USER", "")
 BASIC_AUTH_PASS = os.environ.get("BASIC_AUTH_PASS", "")
 
 POLL_SECONDS = float(os.environ.get("POLL_SECONDS", "5"))
+MIN_VIDEO_SIZE_MB = int(os.environ.get("MIN_VIDEO_SIZE_MB", "200"))
+MIN_VIDEO_BYTES = MIN_VIDEO_SIZE_MB * 1024 * 1024
 
 # JDownloader writes here inside container
 JD_OUTPUT_PATH = "/output"
@@ -830,6 +832,70 @@ def calculate_progress(links: List[Dict[str, Any]]) -> float:
         return 0.0
     return max(0.0, min(100.0, (loaded / total) * 100.0))
 
+# ============================================================
+# Linkgrabber filter (pre-download)
+# ============================================================
+def _filter_linkgrabber(dev, jobid: str) -> Tuple[int, int]:
+    """Wait for link crawler, then remove non-video and sub-minimum-size links.
+    Returns (accepted, rejected) counts."""
+    deadline = time.time() + 120
+    while time.time() < deadline:
+        try:
+            crawlers = dev.linkgrabber.query_link_crawlers([{"collectorInfo": True}]) or []
+            if not any(c.get("crawling") for c in crawlers):
+                break
+        except Exception:
+            break
+        time.sleep(2)
+
+    links = []
+    try:
+        links = dev.linkgrabber.query_links([{
+            "jobUUIDs": [int(jobid)] if str(jobid).isdigit() else [jobid],
+            "maxResults": -1,
+            "startAt": 0,
+            "name": True,
+            "size": True,
+            "uuid": True,
+            "packageUUID": True,
+        }]) or []
+    except Exception:
+        pass
+
+    to_remove_ids = []
+    keep_ids = []
+    keep_pkg_ids = set()
+    for link in links:
+        name = link.get("name", "")
+        size = link.get("size", -1)
+        _, ext = os.path.splitext(name.lower())
+        is_video = ext in VIDEO_EXTS
+        big_enough = size < 0 or size >= MIN_VIDEO_BYTES
+        if is_video and big_enough:
+            keep_ids.append(link.get("uuid"))
+            if link.get("packageUUID") is not None:
+                keep_pkg_ids.add(link.get("packageUUID"))
+        else:
+            to_remove_ids.append(link.get("uuid"))
+
+    if to_remove_ids:
+        try:
+            dev.linkgrabber.remove_links(link_ids=to_remove_ids, package_ids=[])
+        except Exception:
+            pass
+
+    if keep_ids:
+        try:
+            dev.linkgrabber.move_to_downloadlist(
+                link_ids=keep_ids,
+                package_ids=list(keep_pkg_ids),
+            )
+        except Exception:
+            pass
+
+    return len(keep_ids), len(to_remove_ids)
+
+
 # ============================================================
 # Worker
 # ============================================================
@@ -838,6 +904,26 @@ def worker(jobid: str):
         ensure_env()
         dev = get_device()
 
+        # Filter linkgrabber: keep only video files >= MIN_VIDEO_SIZE_MB
+        with lock:
+            job = jobs.get(jobid)
+            if job:
+                job.status = "collecting"
+                job.message = f"Filtere Links (nur Videos \u2265 {MIN_VIDEO_SIZE_MB} MB)\u2026"
+        accepted, rejected = _filter_linkgrabber(dev, jobid)
+        with lock:
+            job = jobs.get(jobid)
+            if job:
+                if accepted == 0:
+                    job.status = "failed"
+                    job.message = (
+                        f"Keine Video-Dateien \u2265 {MIN_VIDEO_SIZE_MB} MB gefunden "
+                        f"({rejected} Link(s) verworfen)."
+                    )
+                    job.progress = 0.0
+                    return
+                job.message = f"{accepted} Video(s) akzeptiert, {rejected} verworfen."
+
         while True:
             with lock:
                 job = jobs.get(jobid)
@@ -856,7 +942,7 @@ def worker(jobid: str):
             if not links:
                 with lock:
                     job.status = "collecting"
-                    job.message = "Warte auf Link-Crawler…"
+                    job.message = "Warte auf Link-Crawler\u2026"
                     job.progress = 0.0
                 time.sleep(POLL_SECONDS)
                 continue
@@ -866,7 +952,7 @@ def worker(jobid: str):
                 cancel_msg = cancel_job(dev, jobid)
                 with lock:
                     job.status = "failed"
-                    base_msg = "JDownloader lieferte das Demo-Video Big Buck Bunny statt des gewünschten Links."
+                    base_msg = "JDownloader lieferte das Demo-Video Big Buck Bunny statt des gew\u00fcnschten Links."
                     job.message = f"{base_msg} {cancel_msg}" if cancel_msg else base_msg
                     job.progress = 0.0
                 return
@@ -877,7 +963,7 @@ def worker(jobid: str):
                 with lock:
                     job.status = "downloading"
                     done = sum(1 for l in links if l.get("finished"))
-                    job.message = f"Download läuft… ({done}/{len(links)} fertig)"
+                    job.message = f"Download l\u00e4uft\u2026 ({done}/{len(links)} fertig)"
                     job.progress = progress
                 time.sleep(POLL_SECONDS)
                 continue
@@ -896,13 +982,13 @@ def worker(jobid: str):
             if not valid_videos:
                 with lock:
                     job.status = "failed"
-                    job.message = "ffprobe: keine gültige Video-Datei."
+                    job.message = "ffprobe: keine g\u00fcltige Video-Datei."
                     job.progress = 0.0
                 return
 
             with lock:
                 job.status = "upload"
-                job.message = f"Download fertig. MD5/Upload/Verify für {len(valid_videos)} Datei(en)…"
+                job.message = f"Download fertig. MD5/Upload/Verify f\u00fcr {len(valid_videos)} Datei(en)\u2026"
                 job.progress = 100.0
 
             ssh = ssh_connect()
@@ -1018,7 +1104,7 @@ def render_page(error: str = "") -> str:
     <head>
       <link rel="stylesheet" href="/static/style.css">
       <meta charset="utf-8">
-      <title>JD → Jellyfin</title>
+      <title>JD \u2192 Jellyfin</title>
       <script>
         async function refreshJobs() {{
           if (document.hidden) return;
@@ -1035,7 +1121,7 @@ def render_page(error: str = "") -> str:
       </script>
     </head>
     <body>
-      <h1>JD → Jellyfin</h1>
+      <h1>JD \u2192 Jellyfin</h1>
       {render_nav("downloads")}
       {err_html}
 
@@ -1099,7 +1185,7 @@ def render_logs_page() -> str:
     <head>
       <link rel="stylesheet" href="/static/style.css">
       <meta charset="utf-8">
-      <title>JD → Jellyfin (Logs)</title>
+      <title>JD \u2192 Jellyfin (Logs)</title>
       <script>
         async function refreshLogs() {{
           if (document.hidden) return;
@@ -1120,9 +1206,9 @@ def render_logs_page() -> str:
       </script>
     </head>
     <body>
-      <h1>JD → Jellyfin</h1>
+      <h1>JD \u2192 Jellyfin</h1>
       {render_nav("logs")}
-      <p class="hint">Verbindungs-Debugger (Echtzeit). Letzte {LOG_BUFFER_LIMIT} Einträge.</p>
+      <p class="hint">Verbindungs-Debugger (Echtzeit). Letzte {LOG_BUFFER_LIMIT} Eintr\u00e4ge.</p>
       <textarea id="log-body" class="log-area" rows="20" readonly></textarea>
     </body>
     </html>
@@ -1143,10 +1229,10 @@ def render_proxies_page(
     <head>
       <link rel="stylesheet" href="/static/style.css">
       <meta charset="utf-8">
-      <title>JD → Jellyfin (Proxies)</title>
+      <title>JD \u2192 Jellyfin (Proxies)</title>
     </head>
     <body>
-      <h1>JD → Jellyfin</h1>
+      <h1>JD \u2192 Jellyfin</h1>
       {render_nav("proxies")}
       {err_html}
       {msg_html}
@@ -1166,7 +1252,7 @@ def render_proxies_page(
       </form>
 
       <h2 style="margin-top:18px;">JDownloader Import-Liste</h2>
-      <p class="hint">Format: <code>socks5://IP:PORT</code>, <code>socks4://IP:PORT</code>. Keine Prüfung/Validierung.</p>
+      <p class="hint">Format: <code>socks5://IP:PORT</code>, <code>socks4://IP:PORT</code>. Keine Pr\u00fcfung/Validierung.</p>
 
       <div class="row">
         <textarea id="out" rows="12" readonly style="width:100%; max-width:860px; padding:10px; border:1px solid #ccc; border-radius:8px;">{esc(out_text)}</textarea>
@@ -1174,8 +1260,8 @@ def render_proxies_page(
 
       <button type="button" onclick="navigator.clipboard.writeText(document.getElementById('out').value)">Kopieren</button>
 
-      <h2 style="margin-top:18px;">Datei für Connection Manager</h2>
+      <h2 style="margin-top:18px;">Datei f\u00fcr Connection Manager</h2>
-      <p class="hint">Speichert die Liste als <code>.jdproxies</code> im Container, z. B. zum Import in JDownloader → Verbindungsmanager → Importieren.</p>
+      <p class="hint">Speichert die Liste als <code>.jdproxies</code> im Container, z. B. zum Import in JDownloader \u2192 Verbindungsmanager \u2192 Importieren.</p>
 
       <form method="post" action="/proxies/save">
         <textarea name="socks5_in" style="display:none;">{esc(socks5_in)}</textarea>
@@ -1217,7 +1303,7 @@ def submit(url: str = Form(...), package_name: str = Form(""), library: str = Fo
         return HTMLResponse(render_page(f"JDownloader nicht erreichbar: {e}"), status_code=503)
     resp = dev.linkgrabber.add_links([{
         "links": url,
-        "autostart": True,
+        "autostart": False,
         "assignJobID": True,
         "packageName": package_name,
     }])
@@ -1251,7 +1337,7 @@ def cancel(jobid: str):
         if job.status in {"finished", "failed", "canceled"}:
             return RedirectResponse(url="/", status_code=303)
         job.cancel_requested = True
-        job.message = "Abbruch angefordert…"
+        job.message = "Abbruch angefordert\u2026"
     return RedirectResponse(url="/", status_code=303)
 
 @app.post("/clear-finished")