Add download preview flow and search fallback

backend/handlers/api.go

@@ -64,12 +64,23 @@ func (h *Hub) Remove(conn *websocket.Conn) {
 	_ = conn.Close()
 }
 
+type PreviewResponse struct {
+	Title           string            `json:"title"`
+	Thumbnail       string            `json:"thumbnail"`
+	Duration        string            `json:"duration"`
+	DurationSeconds int               `json:"durationSeconds"`
+	StartDefault    string            `json:"startDefault"`
+	EndDefault      string            `json:"endDefault"`
+	Qualities       []map[string]any  `json:"qualities"`
+}
+
 func RegisterRoutes(router *gin.Engine, app *App) {
 	router.GET("/healthz", func(c *gin.Context) {
 		c.JSON(http.StatusOK, gin.H{"status": "ok"})
 	})
 	router.GET("/ws", app.handleWS)
 	router.GET("/api/history/check", app.checkDuplicate)
+	router.POST("/api/download/preview", app.previewDownload)
 	router.POST("/api/upload", app.uploadFile)
 	router.POST("/api/download", app.startDownload)
 	router.POST("/api/search", app.searchMedia)
@@ -132,6 +143,7 @@ func (a *App) startDownload(c *gin.Context) {
 		URL   string `json:"url"`
 		Start string `json:"start"`
 		End   string `json:"end"`
+		Quality string `json:"quality"`
 		Force bool   `json:"force"`
 	}
 	if err := c.ShouldBindJSON(&req); err != nil {
@@ -157,13 +169,46 @@ func (a *App) startDownload(c *gin.Context) {
 		return
 	}
 
-	go a.runDownload(recordID, req.URL, req.Start, req.End, outputPath)
+	quality := strings.TrimSpace(req.Quality)
+	if quality == "" {
+		quality = "best"
+	}
+
+	go a.runDownload(recordID, req.URL, req.Start, req.End, quality, outputPath)
 	c.JSON(http.StatusAccepted, gin.H{"message": "download started", "recordId": recordID})
 }
 
-func (a *App) runDownload(recordID int64, url, start, end, outputPath string) {
+func (a *App) previewDownload(c *gin.Context) {
+	var req struct {
+		URL string `json:"url"`
+	}
+	if err := c.ShouldBindJSON(&req); err != nil {
+		c.JSON(http.StatusBadRequest, gin.H{"error": err.Error()})
+		return
+	}
+	if strings.TrimSpace(req.URL) == "" {
+		c.JSON(http.StatusBadRequest, gin.H{"error": "url is required"})
+		return
+	}
+
+	cmd := exec.Command("python3", a.WorkerScript, "--mode", "probe", "--url", req.URL, "--output", filepath.Join(a.DownloadsDir, "probe.tmp"))
+	output, err := cmd.CombinedOutput()
+	if err != nil {
+		c.JSON(http.StatusBadGateway, gin.H{"error": strings.TrimSpace(string(output))})
+		return
+	}
+
+	var preview PreviewResponse
+	if err := json.Unmarshal(output, &preview); err != nil {
+		c.JSON(http.StatusBadGateway, gin.H{"error": "invalid probe response"})
+		return
+	}
+	c.JSON(http.StatusOK, preview)
+}
+
+func (a *App) runDownload(recordID int64, url, start, end, quality, outputPath string) {
 	a.Hub.Broadcast("progress", gin.H{"type": "download", "status": "queued", "progress": 0, "url": url})
-	cmd := exec.Command("python3", a.WorkerScript, "--url", url, "--start", start, "--end", end, "--output", outputPath)
+	cmd := exec.Command("python3", a.WorkerScript, "--url", url, "--start", start, "--end", end, "--quality", quality, "--output", outputPath)
 	stdout, err := cmd.StdoutPipe()
 	if err != nil {
 		a.Hub.Broadcast("progress", gin.H{"type": "download", "status": "error", "progress": 0, "message": err.Error()})
@@ -187,6 +232,9 @@ func (a *App) runDownload(recordID int64, url, start, end, outputPath string) {
 			a.Hub.Broadcast("progress", msg)
 		}
 	}
+	if err := scanner.Err(); err != nil {
+		a.Hub.Broadcast("progress", gin.H{"type": "download", "status": "error", "progress": 100, "message": err.Error()})
+	}
 
 	status := "completed"
 	if err := cmd.Wait(); err != nil {

backend/services/cse.go

@@ -3,6 +3,7 @@ package services
 import (
 	"encoding/json"
 	"fmt"
+	"io"
 	"net/http"
 	"net/url"
 	"strings"
@@ -45,61 +46,120 @@ func (s *SearchService) SearchMedia(query string) ([]SearchResult, error) {
 	values.Set("key", s.APIKey)
 	values.Set("cx", s.CX)
 	values.Set("q", fullQuery)
-	values.Set("searchType", "image")
 	values.Set("num", "10")
 	values.Set("safe", "off")
 
 	results := make([]SearchResult, 0, 30)
 	seen := map[string]bool{}
 	for _, start := range []string{"1", "11", "21"} {
-		values.Set("start", start)
+		pageResults, err := s.fetchPage(values, start, true)
-		endpoint := "https://www.googleapis.com/customsearch/v1?" + values.Encode()
-		resp, err := s.Client.Get(endpoint)
 		if err != nil {
-			return nil, err
+			pageResults, err = s.fetchPage(values, start, false)
+			if err != nil {
+				return nil, err
+			}
 		}
 
-		if resp.StatusCode >= 300 {
+		for _, item := range pageResults {
-			resp.Body.Close()
+			if item.Link == "" || item.ThumbnailURL == "" || seen[item.Link] {
-			return nil, fmt.Errorf("google cse returned status %d", resp.StatusCode)
-		}
-
-		var payload struct {
-			Items []struct {
-				Title       string `json:"title"`
-				Link        string `json:"link"`
-				DisplayLink string `json:"displayLink"`
-				Snippet     string `json:"snippet"`
-				Image       struct {
-					ThumbnailLink string `json:"thumbnailLink"`
-				} `json:"image"`
-			} `json:"items"`
-		}
-
-		if err := json.NewDecoder(resp.Body).Decode(&payload); err != nil {
-			resp.Body.Close()
-			return nil, err
-		}
-		resp.Body.Close()
-
-		for _, item := range payload.Items {
-			if item.Link == "" || seen[item.Link] {
 				continue
 			}
 			seen[item.Link] = true
-			results = append(results, SearchResult{
+			results = append(results, item)
-				Title:        item.Title,
-				Link:         item.Link,
-				DisplayLink:  item.DisplayLink,
-				Snippet:      item.Snippet,
-				ThumbnailURL: item.Image.ThumbnailLink,
-				Source:       inferSource(item.DisplayLink),
-			})
 		}
 	}
 	return results, nil
 }
 
+func (s *SearchService) fetchPage(values url.Values, start string, imageSearch bool) ([]SearchResult, error) {
+	pageValues := url.Values{}
+	for key, items := range values {
+		for _, item := range items {
+			pageValues.Add(key, item)
+		}
+	}
+	pageValues.Set("start", start)
+	if imageSearch {
+		pageValues.Set("searchType", "image")
+	}
+
+	endpoint := "https://www.googleapis.com/customsearch/v1?" + pageValues.Encode()
+	resp, err := s.Client.Get(endpoint)
+	if err != nil {
+		return nil, err
+	}
+	defer resp.Body.Close()
+
+	if resp.StatusCode >= 300 {
+		data, _ := io.ReadAll(io.LimitReader(resp.Body, 2048))
+		return nil, fmt.Errorf("google cse returned status %d: %s", resp.StatusCode, strings.TrimSpace(string(data)))
+	}
+
+	var payload struct {
+		Items []struct {
+			Title       string `json:"title"`
+			Link        string `json:"link"`
+			DisplayLink string `json:"displayLink"`
+			Snippet     string `json:"snippet"`
+			Image       struct {
+				ThumbnailLink string `json:"thumbnailLink"`
+			} `json:"image"`
+			Pagemap struct {
+				CSEImage []struct {
+					Src string `json:"src"`
+				} `json:"cse_image"`
+				CSEThumbnail []struct {
+					Src string `json:"src"`
+				} `json:"cse_thumbnail"`
+				Metatags []map[string]string `json:"metatags"`
+			} `json:"pagemap"`
+		} `json:"items"`
+	}
+
+	if err := json.NewDecoder(resp.Body).Decode(&payload); err != nil {
+		return nil, err
+	}
+
+	results := make([]SearchResult, 0, len(payload.Items))
+	for _, item := range payload.Items {
+		thumb := item.Image.ThumbnailLink
+		if thumb == "" {
+			thumb = extractThumbnail(item.Pagemap)
+		}
+		results = append(results, SearchResult{
+			Title:        item.Title,
+			Link:         item.Link,
+			DisplayLink:  item.DisplayLink,
+			Snippet:      item.Snippet,
+			ThumbnailURL: thumb,
+			Source:       inferSource(item.DisplayLink),
+		})
+	}
+	return results, nil
+}
+
+func extractThumbnail(pagemap struct {
+	CSEImage     []struct{ Src string "json:\"src\"" }            "json:\"cse_image\""
+	CSEThumbnail []struct{ Src string "json:\"src\"" }            "json:\"cse_thumbnail\""
+	Metatags     []map[string]string                              "json:\"metatags\""
+}) string {
+	if len(pagemap.CSEThumbnail) > 0 && pagemap.CSEThumbnail[0].Src != "" {
+		return pagemap.CSEThumbnail[0].Src
+	}
+	if len(pagemap.CSEImage) > 0 && pagemap.CSEImage[0].Src != "" {
+		return pagemap.CSEImage[0].Src
+	}
+	for _, tag := range pagemap.Metatags {
+		if value := tag["og:image"]; value != "" {
+			return value
+		}
+		if value := tag["twitter:image"]; value != "" {
+			return value
+		}
+	}
+	return ""
+}
+
 func inferSource(displayLink string) string {
 	switch {
 	case strings.Contains(displayLink, "youtube"):

frontend/app.js

@@ -13,6 +13,15 @@ const startTime = document.getElementById("startTime");
 const endTime = document.getElementById("endTime");
 const downloadResult = document.getElementById("downloadResult");
 const cardTemplate = document.getElementById("searchCardTemplate");
+const previewModal = document.getElementById("previewModal");
+const previewTitle = document.getElementById("previewTitle");
+const previewThumbnail = document.getElementById("previewThumbnail");
+const previewDuration = document.getElementById("previewDuration");
+const qualitySelect = document.getElementById("qualitySelect");
+const confirmDownload = document.getElementById("confirmDownload");
+const closePreviewModal = document.getElementById("closePreviewModal");
+
+let pendingDownload = null;
 
 function setStatus(label, progress) {
   statusLabel.textContent = label;
@@ -97,7 +106,35 @@ async function uploadFile(file) {
   const formData = new FormData();
   formData.append("file", file);
   uploadResult.textContent = "uploading...";
-  await api("/api/upload", { method: "POST", body: formData });
+  try {
+    await api("/api/upload", { method: "POST", body: formData });
+  } catch (error) {
+    uploadResult.textContent = error.message;
+  }
+}
+
+function openPreviewModal(preview) {
+  previewTitle.textContent = preview.title;
+  previewThumbnail.src = preview.thumbnail;
+  previewThumbnail.alt = preview.title;
+  previewDuration.textContent = preview.duration;
+  qualitySelect.innerHTML = "";
+  for (const item of preview.qualities || []) {
+    const option = document.createElement("option");
+    option.value = item.value;
+    option.textContent = item.label;
+    qualitySelect.appendChild(option);
+  }
+  startTime.value = preview.startDefault || "00:00:00";
+  endTime.value = preview.endDefault || "00:00:00";
+  previewModal.classList.remove("hidden");
+  previewModal.classList.add("flex");
+}
+
+function closeModal() {
+  previewModal.classList.add("hidden");
+  previewModal.classList.remove("flex");
+  pendingDownload = null;
 }
 
 dropzone.addEventListener("dragover", (event) => {
@@ -138,21 +175,49 @@ downloadForm.addEventListener("submit", async (event) => {
         return;
       }
     }
+    pendingDownload = { url: downloadUrl.value, force };
+    downloadResult.textContent = "loading preview...";
+    const preview = await api("/api/download/preview", {
+      method: "POST",
+      headers: { "Content-Type": "application/json" },
+      body: JSON.stringify({ url: downloadUrl.value }),
+    });
+    openPreviewModal(preview);
+    downloadResult.textContent = "preview loaded";
+  } catch (error) {
+    downloadResult.textContent = error.message;
+  }
+});
+
+confirmDownload.addEventListener("click", async () => {
+  if (!pendingDownload) {
+    return;
+  }
+  try {
     const data = await api("/api/download", {
       method: "POST",
       headers: { "Content-Type": "application/json" },
       body: JSON.stringify({
-        url: downloadUrl.value,
+        url: pendingDownload.url,
         start: startTime.value,
         end: endTime.value,
-        force,
+        quality: qualitySelect.value,
+        force: pendingDownload.force,
       }),
     });
+    closeModal();
     downloadResult.textContent = data.message;
   } catch (error) {
     downloadResult.textContent = error.message;
   }
 });
 
+closePreviewModal.addEventListener("click", closeModal);
+previewModal.addEventListener("click", (event) => {
+  if (event.target === previewModal) {
+    closeModal();
+  }
+});
+
 connectWS();
 setStatus("idle", 0);

frontend/index.html

@@ -13,7 +13,7 @@
         <div class="flex flex-col gap-4 lg:flex-row lg:items-end lg:justify-between">
           <div>
             <p class="text-xs uppercase tracking-[0.4em] text-zinc-500">AI Media Asset Ingest Hub</p>
-            <h1 class="mt-3 text-3xl font-semibold tracking-tight text-white">Multimodal Discovery, Drag Upload, Direct Clip Ingest</h1>
+            <h1 class="mt-3 text-3xl font-semibold tracking-tight text-white">SAVE THE NURSE AI Search</h1>
           </div>
           <div class="w-full max-w-md">
             <div class="mb-2 flex items-center justify-between text-xs uppercase tracking-[0.3em] text-zinc-500">
@@ -62,9 +62,9 @@
               <input id="downloadUrl" type="url" placeholder="https://..." class="w-full rounded-2xl border border-white/10 bg-black/40 px-4 py-3 text-sm text-white placeholder:text-zinc-500" />
               <div class="grid grid-cols-2 gap-3">
                 <input id="startTime" type="text" value="00:00:00" class="rounded-2xl border border-white/10 bg-black/40 px-4 py-3 text-sm text-white" />
-                <input id="endTime" type="text" value="00:00:10" class="rounded-2xl border border-white/10 bg-black/40 px-4 py-3 text-sm text-white" />
+                <input id="endTime" type="text" value="00:00:00" class="rounded-2xl border border-white/10 bg-black/40 px-4 py-3 text-sm text-white" />
               </div>
-              <button class="w-full rounded-2xl border border-white px-5 py-3 text-sm font-medium text-white transition hover:bg-white hover:text-black">Queue Clip Download</button>
+              <button class="w-full rounded-2xl border border-white px-5 py-3 text-sm font-medium text-white transition hover:bg-white hover:text-black">Preview & Queue</button>
             </form>
             <p id="downloadResult" class="mt-3 text-sm text-zinc-400"></p>
           </article>
@@ -72,6 +72,36 @@
       </section>
     </main>
 
+    <div id="previewModal" class="fixed inset-0 z-50 hidden items-center justify-center bg-black/80 px-4">
+      <div class="w-full max-w-2xl rounded-3xl border border-white/10 bg-zinc-950 p-5 shadow-2xl">
+        <div class="flex items-start justify-between gap-4">
+          <div>
+            <p class="text-xs uppercase tracking-[0.3em] text-zinc-500">Download Preview</p>
+            <h3 id="previewTitle" class="mt-2 text-xl font-semibold text-white"></h3>
+          </div>
+          <button id="closePreviewModal" class="rounded-full border border-white/10 px-3 py-2 text-xs uppercase tracking-[0.2em] text-zinc-300">Close</button>
+        </div>
+        <div class="mt-5 grid gap-5 md:grid-cols-[1.1fr_0.9fr]">
+          <div class="overflow-hidden rounded-3xl border border-white/10 bg-black/30">
+            <img id="previewThumbnail" class="aspect-video h-full w-full object-cover" alt="" />
+          </div>
+          <div class="space-y-4">
+            <div class="rounded-2xl border border-white/10 bg-white/[0.03] p-4">
+              <div class="flex items-center justify-between text-sm text-zinc-400">
+                <span>Detected duration</span>
+                <span id="previewDuration"></span>
+              </div>
+            </div>
+            <label class="block space-y-2">
+              <span class="text-sm text-zinc-400">Quality</span>
+              <select id="qualitySelect" class="w-full rounded-2xl border border-white/10 bg-black/40 px-4 py-3 text-sm text-white"></select>
+            </label>
+            <button id="confirmDownload" class="w-full rounded-2xl border border-white bg-white px-5 py-3 text-sm font-medium text-black transition hover:bg-zinc-200">Confirm Download</button>
+          </div>
+        </div>
+      </div>
+    </div>
+
     <template id="searchCardTemplate">
       <a target="_blank" rel="noreferrer" class="group overflow-hidden rounded-3xl border border-white/10 bg-black/30 transition hover:border-white/30">
         <div class="relative aspect-video overflow-hidden bg-zinc-900">

worker/downloader.py

@@ -5,6 +5,7 @@ import os
 import subprocess
 import sys
 import tempfile
+from typing import List
 
 
 def emit(status, progress, message="", output=""):
@@ -25,14 +26,74 @@ def run(cmd):
     return proc
 
 
+def parse_duration(value):
+    if value is None:
+        return "00:00:10"
+    total = int(float(value))
+    hours = total // 3600
+    minutes = (total % 3600) // 60
+    seconds = total % 60
+    return f"{hours:02d}:{minutes:02d}:{seconds:02d}"
+
+
+def format_label(height, ext):
+    if height:
+        return f"{height}p ({ext})"
+    return f"Best ({ext})"
+
+
+def build_quality_options(formats: List[dict]):
+    heights = []
+    for item in formats:
+        height = item.get("height")
+        if isinstance(height, int) and height > 0:
+            heights.append(height)
+
+    unique_heights = sorted(set(heights))
+    options = [{"value": "best", "label": "Best available"}]
+    for height in unique_heights:
+        options.append(
+            {
+                "value": f"bestvideo[height<={height}]+bestaudio/best[height<={height}]",
+                "label": f"Up to {height}p",
+            }
+        )
+    return options
+
+
+def probe(url):
+    cmd = ["yt-dlp", "--dump-single-json", "--no-playlist", url]
+    proc = run(cmd)
+    payload = json.loads(proc.stdout)
+    thumbnail = payload.get("thumbnail") or ""
+    duration = payload.get("duration")
+    formats = payload.get("formats") or []
+    preview = {
+        "title": payload.get("title") or "Untitled",
+        "thumbnail": thumbnail,
+        "durationSeconds": duration or 0,
+        "duration": parse_duration(duration),
+        "startDefault": "00:00:00",
+        "endDefault": parse_duration(duration),
+        "qualities": build_quality_options(formats),
+    }
+    print(json.dumps(preview), flush=True)
+
+
 def main():
     parser = argparse.ArgumentParser()
+    parser.add_argument("--mode", choices=["probe", "download"], default="download")
     parser.add_argument("--url", required=True)
     parser.add_argument("--start", default="00:00:00")
     parser.add_argument("--end", default="00:00:10")
     parser.add_argument("--output", required=True)
+    parser.add_argument("--quality", default="best")
     args = parser.parse_args()
 
+    if args.mode == "probe":
+        probe(args.url)
+        return
+
     os.makedirs(os.path.dirname(args.output), exist_ok=True)
     emit("starting", 5, "Resolving media stream")
 
@@ -42,7 +103,7 @@ def main():
             "yt-dlp",
             "--no-playlist",
             "-f",
-            "mp4/bestvideo*+bestaudio/best",
+            args.quality,
             "-o",
             source_path,
             args.url,