Update app.py

app.py

@@ -1,59 +1,39 @@
 import gradio as gr
 from internetarchive import search_items, get_item
-import requests
-import time
-import subprocess
-import json
-import re
+import requests, time, subprocess, json, re, tempfile
 from bs4 import BeautifulSoup
 from urllib.parse import urlparse
-import tempfile
 from pyvis.network import Network
 
 THEME = "gradio/soft"
-
-# --- Shared HTTP session for speed & headers ---
 session = requests.Session()
-session.headers.update({
-    "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64)"
-})
+session.headers.update({"User-Agent": "Mozilla/5.0"})
 
 def scan_url_vt(url, api_key):
-    """Return True if VirusTotal sees no malicious hits."""
     headers = {"x-apikey": api_key}
-    resp = session.post("https://www.virustotal.com/api/v3/urls",
-                        headers=headers, data={"url": url})
+    resp = session.post("https://www.virustotal.com/api/v3/urls", headers=headers, data={"url": url})
     resp.raise_for_status()
     analysis_id = resp.json()["data"]["id"]
-    # Poll until complete
     while True:
         time.sleep(5)
-        st = session.get(f"https://www.virustotal.com/api/v3/analyses/{analysis_id}",
-                         headers=headers)
+        st = session.get(f"https://www.virustotal.com/api/v3/analyses/{analysis_id}", headers=headers)
         st.raise_for_status()
         attr = st.json()["data"]["attributes"]
         if attr.get("status") == "completed":
             return attr.get("stats", {}).get("malicious", 0) == 0
 
 def extract_ffprobe_metadata(url_or_path):
-    """Run ffprobe and parse its JSON output."""
-    cmd = [
+    out = subprocess.check_output([
         "ffprobe", "-v", "error", "-print_format", "json",
         "-show_format", "-show_streams", url_or_path
-    ]
-    out = subprocess.check_output(cmd)
+    ])
     return json.loads(out)
 
 def fetch_page_metadata(url):
-    """Grab <title>, og: and twitter: meta tags from any page."""
     try:
-        r = session.get(url, timeout=5)
-        r.raise_for_status()
+        r = session.get(url, timeout=5); r.raise_for_status()
         soup = BeautifulSoup(r.text, "html.parser")
-        data = {
-            "url": url,
-            "title": soup.title.string if soup.title else None
-        }
+        data = {"url": url, "title": getattr(soup.title, "string", None)}
         for tag in soup.find_all("meta"):
             prop = tag.get("property") or tag.get("name")
             if prop and prop.startswith(("og:", "twitter:")):
@@ -63,7 +43,6 @@ def fetch_page_metadata(url):
         return {"url": url, "error": str(e)}
 
 def fetch_clean_videos(keywords, api_key, scan_enabled):
-    """Search IA, filter for common video formats, optional VT scan."""
     terms = [kw.strip().replace(" ", "+") for kw in keywords.split(",")]
     ia_query = f"mediatype:(movies) AND ({' OR '.join(terms)})"
     results = list(search_items(ia_query))[:50]
@@ -73,9 +52,10 @@ def fetch_clean_videos(keywords, api_key, scan_enabled):
         ident = r["identifier"]
         item = get_item(ident)
         for f in item.files:
-            fmt = f.get("format", "").lower()
-            if fmt.startswith(("mp4", "avi", "mov", "webm", "m4v")):
-                url = f"https://archive.org/download/{ident}/{f['name']}"
+            name = f.get("name", "")
+            ext  = name.lower().split(".")[-1]
+            if ext in ("mp4", "mkv", "avi", "mov", "webm", "m4v"):
+                url = f"https://archive.org/download/{ident}/{name}"
                 if scan_enabled and api_key:
                     try:
                         if not scan_url_vt(url, api_key):
@@ -86,19 +66,16 @@ def fetch_clean_videos(keywords, api_key, scan_enabled):
     return clean
 
 def get_favicon_url(page_url):
-    """Assume https://domain/favicon.ico exists."""
     dom = urlparse(page_url).netloc
     return f"https://{dom}/favicon.ico"
 
 def build_origin_graph(origins):
-    """Render a PyVis network; return its full HTML."""
     net = Network(height="400px", width="100%", directed=True)
     for i, m in enumerate(origins):
         fav = get_favicon_url(m["url"])
         label = urlparse(m["url"]).netloc
         title = json.dumps(m, indent=2)
-        net.add_node(i, label=label, title=title,
-                     shape="image", image=fav)
+        net.add_node(i, label=label, title=title, shape="image", image=fav)
         if i > 0:
             net.add_edge(i - 1, i)
     tmp = tempfile.NamedTemporaryFile(delete=False, suffix=".html").name
@@ -106,19 +83,18 @@ def build_origin_graph(origins):
     return open(tmp, encoding="utf-8").read()
 
 with gr.Blocks(theme=THEME) as demo:
-    gr.Markdown("## 📼 IA Drone-Strike Explorer")
+    gr.Markdown("## 📼 IA Drone‑Strike Explorer")
     with gr.Row():
-        kw = gr.Textbox(label="Search keywords",
-                        value="drone strike, military uav")
+        kw     = gr.Textbox(label="Search keywords", value="drone strike, military uav")
         vt_key = gr.Textbox(label="VirusTotal API Key", type="password")
-    scan = gr.Checkbox(label="Enable VT scan", value=True)
-    ff = gr.Checkbox(label="Enable FFprobe metadata", value=False)
-    btn = gr.Button("Search & Scan")
+    scan   = gr.Checkbox(label="Enable VT scan", value=True)
+    ff     = gr.Checkbox(label="Enable FFprobe metadata", value=False)
+    btn    = gr.Button("Search & Scan")
 
-    dropdown = gr.Dropdown(label="Clean Video URLs", choices=[])
-    graph_html = gr.HTML("<p>No origin graph yet.</p>")
+    dropdown    = gr.Dropdown(label="Clean Video URLs", choices=[])
+    graph_html  = gr.HTML("<p>No origin graph yet.</p>")
     video_player = gr.Video()
-    
+
     with gr.Tabs():
         with gr.TabItem("IA Metadata"):
             ia_json = gr.JSON()
@@ -135,11 +111,8 @@ with gr.Blocks(theme=THEME) as demo:
     def update_all(url_sel, ff_on, api_key):
         if not url_sel:
             return None, {}, {}, "<p>No origin graph.</p>", []
-
-        # --- IA metadata for just this file ---
         parts = url_sel.split("/")
-        ident = parts[4]
-        fn = parts[-1]
+        ident = parts[4]; fn = parts[-1]
         ia_data = {"identifier": ident, "file": {}}
         try:
             item = get_item(ident)
@@ -150,33 +123,22 @@ with gr.Blocks(theme=THEME) as demo:
         except Exception as e:
             ia_data["error"] = str(e)
 
-        # --- FFprobe if requested ---
         ff_data = {}
         if ff_on:
-            try:
-                ff_data = extract_ffprobe_metadata(url_sel)
-            except Exception as e:
-                ff_data = {"error": str(e)}
+            try:    ff_data = extract_ffprobe_metadata(url_sel)
+            except Exception as e: ff_data = {"error": str(e)}
 
-        # --- Origin tracing: all links in description ---
         desc = item.metadata.get("description", "") or ""
         found = re.findall(r"https?://[^\s\"'<]+", desc)
         origins = [fetch_page_metadata(u) for u in found]
+        graph   = build_origin_graph(origins) if origins else "<p>No origins.</p>"
 
-        # --- Build the interactive graph HTML ---
-        graph = build_origin_graph(origins) if origins else "<p>No origins.</p>"
         return url_sel, ia_data, ff_data, graph, origins
 
-    btn.click(
-        fn=search_and_populate,
-        inputs=[kw, vt_key, scan],
-        outputs=[dropdown]
-    )
-    dropdown.change(
-        fn=update_all,
-        inputs=[dropdown, ff, vt_key],
-        outputs=[video_player, ia_json, ff_json, graph_html, orig_json]
-    )
+    btn.click(fn=search_and_populate, inputs=[kw, vt_key, scan], outputs=[dropdown])
+    dropdown.change(fn=update_all,
+                    inputs=[dropdown, ff, vt_key],
+                    outputs=[video_player, ia_json, ff_json, graph_html, orig_json])
 
 if __name__ == "__main__":
     demo.launch()