Improve query intent handling and preview playback

2026-03-16 09:55:14 +09:00
parent 6d9391bc2b
commit 9637b761bd
7 changed files with 142 additions and 12 deletions
@@ -141,6 +141,19 @@
 - Effect:
  - improves Artgrid recall in SearXNG result sets that favor canonical Artlist URLs over Artgrid URLs

+## Current Session Update (2026-03-16, Query / Preview Follow-up)
+- Search intent translation was updated to better preserve compound media phrases:
+  - added explicit normalization for terms like `사이버 펑크` -> `cyberpunk`
+  - added a guard that rejects over-compressed translations when the original query contains a richer multi-word intent
+- Artgrid page parsing was tightened:
+  - generic Artgrid homepage / challenge HTML should no longer be mistaken for a real clip page during enrichment
+  - this prevents homepage thumbnails/descriptions from overwriting real search result metadata
+- Hover preview playback was changed to lazy attach on hover:
+  - preview source is now attached on mouseenter
+  - playback waits for media readiness instead of trying to play immediately from the render path
+  - source is detached again on mouseleave
+- Self-test script search step now retries to reduce flaky startup timing failures during local smoke tests
+
 ## Local Self-Test Workflow
 - Primary command:
  - `bash scripts/selftest.sh`
@@ -229,6 +229,9 @@ func (s *SearchService) enrichArtgrid(result SearchResult) SearchResult {
 	if result.ThumbnailURL == "" || result.PreviewVideoURL == "" {
 		html, err := s.fetchText(result.Link)
 		if err == nil {
+			if !isMatchingArtgridClipPage(html, clipID) {
+				return result
+			}
 			result.Title = firstNonEmpty(
 				cleanArtgridTitle(extractMetaContent(html, "og:title")),
 				cleanArtgridTitle(extractMetaContent(html, "title")),
@@ -756,6 +759,33 @@ func cleanArtgridDescription(description string) string {
 	return strings.TrimSpace(description)
 }

+func isMatchingArtgridClipPage(html, clipID string) bool {
+	if clipID == "" {
+		return false
+	}
+	ogURL := extractMetaContent(html, "og:url")
+	canonical := extractCanonicalURL(html)
+	lowerHTML := strings.ToLower(html)
+	for _, candidate := range []string{ogURL, canonical} {
+		if strings.Contains(candidate, clipID) {
+			return true
+		}
+	}
+	if strings.Contains(lowerHTML, "main-clipvideo_"+clipID) || strings.Contains(lowerHTML, "/clip/"+clipID+"/") {
+		return true
+	}
+	return false
+}
+
+func extractCanonicalURL(html string) string {
+	pattern := regexp.MustCompile(`(?i)<link[^>]+rel=["']canonical["'][^>]+href=["']([^"']+)`)
+	matches := pattern.FindStringSubmatch(html)
+	if len(matches) == 2 {
+		return htmlUnescape(matches[1])
+	}
+	return ""
+}
+
 func deriveEnvatoPreviewFromThumbnail(thumbnail string) string {
 	candidate := htmlUnescape(strings.TrimSpace(thumbnail))
 	if candidate == "" {
@@ -81,6 +81,13 @@ func TestBuildArtgridQueriesIncludesArtlistCanonicalDomain(t *testing.T) {
 	}
 }

+func TestIsMatchingArtgridClipPageRejectsHomepage(t *testing.T) {
+	html := `<html><head><meta property="og:url" content="https://artgrid.io/"><link rel="canonical" href="https://artgrid.io/"></head></html>`
+	if isMatchingArtgridClipPage(html, "114756") {
+		t.Fatal("expected generic Artgrid homepage HTML to be rejected as a clip page")
+	}
+}
+
 func TestGeminiCandidateLimitNeverExceedsCandidates(t *testing.T) {
 	if got := GeminiCandidateLimit(9); got != 9 {
 		t.Fatalf("expected Gemini limit to stay within candidate count, got %d", got)
@@ -58,6 +58,10 @@ func (g *GeminiService) TranslateQuery(query string) string {
 	if trimmed == "" {
 		return ""
 	}
+	normalizedIntent := normalizeKnownMediaPhrases(trimmed)
+	if looksMostlyASCII(normalizedIntent) {
+		return strings.TrimSpace(normalizedIntent)
+	}
 	if looksMostlyASCII(trimmed) {
 		return trimmed
 	}
@@ -90,19 +94,19 @@ func (g *GeminiService) TranslateQuery(query string) string {
 		rawText, err := g.generateText(body)
 		if err == nil {
 			translated := sanitizePlainEnglishLine(rawText)
-			if translated != "" && !strings.EqualFold(translated, trimmed) {
+			if translated != "" && !strings.EqualFold(translated, trimmed) && !isOvercompressedTranslation(trimmed, translated) {
 				return translated
 			}
 		}
 	}

-	if translated, err := g.translateViaGoogle(trimmed); err == nil && translated != "" && isLikelyEnglishQuery(translated) {
+	if translated, err := g.translateViaGoogle(trimmed); err == nil && translated != "" && isLikelyEnglishQuery(translated) && !isOvercompressedTranslation(trimmed, translated) {
 		return translated
 	}
-	if translated := translateKoreanMediaTerms(trimmed); translated != "" && !strings.EqualFold(translated, trimmed) {
+	if translated := translateKoreanMediaTerms(normalizedIntent); translated != "" && !strings.EqualFold(translated, trimmed) {
 		return translated
 	}
-	return trimmed
+	return strings.TrimSpace(normalizedIntent)
 }

 func (g *GeminiService) generateText(body map[string]any) (string, error) {
@@ -493,6 +497,12 @@ func translateKoreanMediaTerms(query string) string {
 		korean  string
 		english string
 	}{
+		{korean: "사이버 펑크 도시", english: "cyberpunk city"},
+		{korean: "사이버펑크 도시", english: "cyberpunk city"},
+		{korean: "사이버 펑크", english: "cyberpunk"},
+		{korean: "사이버펑크", english: "cyberpunk"},
+		{korean: "네온 도시", english: "neon city"},
+		{korean: "미래 도시", english: "futuristic city"},
 		{korean: "숲속", english: "forest"},
 		{korean: "다정한", english: "affectionate"},
 		{korean: "항공샷", english: "aerial shot"},
@@ -528,6 +538,38 @@ func translateKoreanMediaTerms(query string) string {
 	return strings.TrimSpace(translated)
 }

+func normalizeKnownMediaPhrases(query string) string {
+	normalized := strings.TrimSpace(query)
+	replacements := []struct {
+		from string
+		to   string
+	}{
+		{from: "사이버 펑크 도시", to: "cyberpunk city"},
+		{from: "사이버펑크 도시", to: "cyberpunk city"},
+		{from: "사이버 펑크", to: "cyberpunk"},
+		{from: "사이버펑크", to: "cyberpunk"},
+	}
+	for _, replacement := range replacements {
+		normalized = strings.ReplaceAll(normalized, replacement.from, replacement.to)
+	}
+	return strings.Join(strings.Fields(normalized), " ")
+}
+
+func isOvercompressedTranslation(original, translated string) bool {
+	originalWords := len(strings.Fields(strings.TrimSpace(original)))
+	translatedWords := len(strings.Fields(strings.TrimSpace(translated)))
+	if originalWords < 2 || translatedWords >= 2 {
+		return false
+	}
+	lower := strings.ToLower(strings.TrimSpace(translated))
+	for _, allow := range []string{"cyberpunk", "nightlife", "cityscape"} {
+		if lower == allow {
+			return false
+		}
+	}
+	return true
+}
+
 func (g *GeminiService) translateViaGoogle(query string) (string, error) {
 	baseURL := g.TranslateEndpoint
 	if strings.TrimSpace(baseURL) == "" {
@@ -39,3 +39,10 @@ func TestTranslateQueryFallsBackToDictionaryWhenTranslateFails(t *testing.T) {
 		t.Fatalf("expected dictionary fallback translation, got %q", translated)
 	}
 }
+
+func TestNormalizeKnownMediaPhrases(t *testing.T) {
+	translated := translateKoreanMediaTerms("사이버 펑크 도시")
+	if translated != "cyberpunk city" {
+		t.Fatalf("expected cyberpunk city, got %q", translated)
+	}
+}
@@ -270,6 +270,33 @@ function attachVideoSource(video, src) {
  logEvent("preview:attach:file", { src });
 }

+function startHoverPreview(video, src) {
+  if (!src) {
+    return;
+  }
+  attachVideoSource(video, src);
+  video.classList.remove("hidden");
+  const attemptPlay = () => {
+    video.play().catch((error) => {
+      logEvent("preview:hover:play:error", { src, message: String(error) });
+    });
+  };
+  if (video.readyState >= 2) {
+    attemptPlay();
+    return;
+  }
+  const onReady = () => {
+    video.removeEventListener("loadeddata", onReady);
+    video.removeEventListener("canplay", onReady);
+    attemptPlay();
+  };
+  video.addEventListener("loadeddata", onReady, { once: true });
+  video.addEventListener("canplay", onReady, { once: true });
+  if (video.load) {
+    video.load();
+  }
+}
+
 function detachVideoSource(video) {
  const existing = hlsInstances.get(video);
  if (existing) {
@@ -299,21 +326,20 @@ function renderResults(results) {
    node.querySelector(".result-snippet").textContent = item.snippet || item.reason || item.source || "";
    node.querySelector(".result-reason").textContent = item.reason ? `AI note: ${item.reason}` : "";
    node.querySelector(".source-badge").textContent = item.source;
+    previewVideo.poster = item.thumbnailUrl || "";
    if (item.previewVideoUrl) {
-      attachVideoSource(previewVideo, item.previewVideoUrl);
-      previewVideo.poster = item.thumbnailUrl || "";
      const mediaArea = node.querySelector(".relative");
      mediaArea.addEventListener("mouseenter", () => {
        logEvent("preview:hover:start", { title: item.title, source: item.source, previewVideoUrl: item.previewVideoUrl });
        overlays.forEach((overlay) => overlay.classList.add("hidden"));
-        previewVideo.classList.remove("hidden");
-        previewVideo.play().catch(() => {});
+        startHoverPreview(previewVideo, item.previewVideoUrl);
      });
      mediaArea.addEventListener("mouseleave", () => {
        logEvent("preview:hover:end", { title: item.title, source: item.source });
        previewVideo.pause();
        previewVideo.currentTime = 0;
        previewVideo.classList.add("hidden");
+        detachVideoSource(previewVideo);
        overlays.forEach((overlay) => overlay.classList.remove("hidden"));
      });
    }
@@ -70,10 +70,15 @@ if payload.get("status") != "ok":
 PY

 echo "[selftest] verify search"
-curl -fsS \
-  -H "Content-Type: application/json" \
-  -d '{"query":"city rain","platforms":["envato","artgrid","google video"]}' \
-  "http://127.0.0.1:${APP_PORT}/api/search" >"${TMP_DIR}/search.json"
+for _ in $(seq 1 5); do
+  if curl -fsS \
+    -H "Content-Type: application/json" \
+    -d '{"query":"city rain","platforms":["envato","artgrid","google video"]}' \
+    "http://127.0.0.1:${APP_PORT}/api/search" >"${TMP_DIR}/search.json"; then
+    break
+  fi
+  sleep 1
+done
 python3 - "${TMP_DIR}/search.json" <<'PY'
 import json
 import sys