Video search via Piped API with normalized engagement ranking + instance failover; Data API/yt-dlp fallback

app.py

@@ -104,11 +104,8 @@ def _cookiefile(workdir: str) -> str | None:
     return path
 
 
-def _resolve_proxy(ui_proxy: str | None = None) -> str | None:
-    """A per-user proxy pasted in the UI wins over the shared YT_PROXY secret, so each
-    user can route through their own residential proxy (see tools/home_proxy_panel.py).
-    It also backs screenshot downloads (capture_shots falls back to it)."""
-    proxy = (ui_proxy or "").strip() or os.environ.get("YT_PROXY", "").strip()
+def _resolve_proxy() -> str | None:
+    proxy = os.environ.get("YT_PROXY", "").strip()
     return proxy or None
 
 
@@ -125,15 +122,14 @@ def _resolve_api_key(ui_key: str | None) -> str | None:
     return key or None
 
 
-def check_access(yt_proxy=None):
+def check_access():
     """Health check: which secrets are set, the egress IP, and YouTube reachability."""
     import requests
 
-    proxy = _resolve_proxy(yt_proxy)
+    proxy = _resolve_proxy()
     proxies = {"http": proxy, "https": proxy} if proxy else None
-    src = "your proxy field" if (yt_proxy or "").strip() else "YT_PROXY secret"
     lines = [
-        f"- **Proxy**: {('set ✅ (' + src + ')') if proxy else 'not set ⚪'}",
+        f"- **Proxy** (`YT_PROXY`): {'set ✅' if proxy else 'not set ⚪'}",
         f"- **Media proxy** (`YT_MEDIA_PROXY`, screenshots): "
         f"{'set ✅' if _resolve_media_proxy() else 'not set ⚪ (falls back to YT_PROXY)'}",
         f"- **Cookies** (`YT_COOKIES`): {'set ✅' if os.environ.get('YT_COOKIES') else 'not set ⚪'}",
@@ -222,7 +218,7 @@ def _collect_keywords(primary_kw, secondary_kw) -> dict:
     return {"primary": primary, "secondary": secondary}
 
 
-def run_pipeline(topic, hf_token, yt_api_key, yt_proxy, llm_model, vlm_model,
+def run_pipeline(topic, hf_token, yt_api_key, llm_model, vlm_model,
                  w_llm, w_whisper, lead, max_shots, primary_kw, secondary_kw,
                  content_brief="", progress=gr.Progress()):
     """Generator yielding (status_md, ranking_df, transcript, docx_file)."""
@@ -243,17 +239,18 @@ def run_pipeline(topic, hf_token, yt_api_key, yt_proxy, llm_model, vlm_model,
     try:
         api_key = _resolve_api_key(yt_api_key)
         cookiefile = _cookiefile(workdir)
-        proxy = _resolve_proxy(yt_proxy)
+        proxy = _resolve_proxy()
         if proxy:
-            # Never log the proxy URL itself — it carries user:pass credentials.
-            src = "your proxy" if (yt_proxy or "").strip() else "the `YT_PROXY` secret"
-            log.append(f"🔐 Routing transcript + screenshot requests through {src}.")
+            log.append("🔐 `YT_PROXY` is set — transcript + stream requests route through it.")
 
         # 1. Search ------------------------------------------------------------------
         progress(0.03, desc="Searching")
-        yield status(f"🔍 Searching top videos for **{topic}**…"), gr.update(), gr.update(), gr.update()
-        videos = search_mod.search_top5(topic)
-        yield status(f"Found {len(videos)} candidate videos."), gr.update(), gr.update(), gr.update()
+        yield status(f"🔍 Searching top videos for **{topic}** (engagement-ranked)…"), gr.update(), gr.update(), gr.update()
+        videos = search_mod.search_top5(topic, api_key=api_key, proxy=proxy)
+        eng_note = ""
+        if videos and "engagement" in videos[0]:
+            eng_note = " • ranked by likes+comments+subscribers−dislikes (normalized)"
+        yield status(f"Found {len(videos)} candidate videos{eng_note}."), gr.update(), gr.update(), gr.update()
 
         # 2. Sentiment ranking (YouTube Data API comments) ---------------------------
         if api_key:
@@ -339,12 +336,7 @@ def build_ui():
             "Enter a topic, your **Hugging Face token** (LLM + vision model, billed to you) "
             "and a **YouTube Data API key** (for comments). The Space picks the best video by "
             "comment sentiment, pulls its transcript, writes an AEO-friendly tutorial, grabs "
-            "real screenshots at the right moments, and builds a **.docx**.\n\n"
-            "> 🏠 **YouTube blocks this Space's datacenter IP.** Run your own residential proxy "
-            "with the **Home Proxy Panel** "
-            "([download / source](https://github.com/vivekchakraverty/tutorialmaker-home-proxy-panel)) "
-            "and paste its URL into **Your proxy URL** below — transcript + screenshots then route "
-            "through your home IP. Each user brings their own proxy."
+            "real screenshots at the right moments, and builds a **.docx**."
         )
         with gr.Row():
             with gr.Column(scale=2):
@@ -353,13 +345,6 @@ def build_ui():
                                       placeholder="hf_…  (Inference Providers permission)")
                 yt_api_key = gr.Textbox(label="YouTube Data API key", type="password",
                                         placeholder="for comments (or set the YOUTUBE_API_KEY secret)")
-                yt_proxy = gr.Textbox(
-                    label="Your proxy URL (recommended)", type="password",
-                    placeholder="http://user:pass@bore.pub:12345  — from the Home Proxy Panel",
-                    info="Run the Home Proxy Panel (github.com/vivekchakraverty/"
-                         "tutorialmaker-home-proxy-panel) on your own machine and paste its "
-                         "proxy URL here so YouTube requests exit from your residential IP. "
-                         "Leave blank to use the Space's shared proxy, if configured.")
             with gr.Column(scale=1):
                 llm_model = gr.Dropdown(LLM_CHOICES, value=LLM_CHOICES[0],
                                         label="Tutorial LLM", allow_custom_value=True)
@@ -411,10 +396,10 @@ def build_ui():
         transcript_box = gr.Textbox(label="Transcript preview", lines=10, max_lines=20)
         docx_file = gr.File(label="Download tutorial (.docx)")
 
-        health_btn.click(check_access, inputs=[yt_proxy], outputs=health_md)
+        health_btn.click(check_access, outputs=health_md)
         run_btn.click(
             run_pipeline,
-            inputs=[topic, hf_token, yt_api_key, yt_proxy, llm_model, vlm_model,
+            inputs=[topic, hf_token, yt_api_key, llm_model, vlm_model,
                     w_llm, w_whisper, lead, max_shots, primary_kw, secondary_kw,
                     content_brief],
             outputs=[status_md, ranking_df, transcript_box, docx_file],