Update app.py

app.py

@@ -6,38 +6,35 @@ import subprocess
 
 allowed_medias = [".png", ".jpg", ".jpeg", ".bmp", ".gif", ".tiff"]
 
-def generate_slideshow_with_text_ypos(images, duration, texts=None, y_positions=None, fade_duration=0.7):
+def generate_slideshow_with_fixed_text_and_padding(images, duration, texts=None, y_pos=0.5, fade_duration=0.7, font_size=60):
     """
-    Slideshow mit smooth Fade-In/Fade-Out Text, Y-Position pro Bild einstellbar.
-    texts: Liste der Strings pro Bild
-    y_positions: Liste der Y-Positionen pro Bild (0=oben, 0.5=mitte, 1=unten)
+    Slideshow:
+    - Bilder behalten ihre Proportionen, schwarze Balken falls nötig
+    - Text Overlay mit Y-Position, Fade-In/Out und fester Größe
     """
     if not images:
         return None, "❌ Keine Bilder ausgewählt"
 
+    y_pos = min(max(0.0, y_pos), 0.9)  # Max 0.9 für unten
     temp_dir = tempfile.mkdtemp()
     clips = []
 
-    # Standardwerte
     if texts is None:
         texts = [""] * len(images)
-    if y_positions is None:
-        y_positions = [0.5] * len(images)
 
     for i, img_path in enumerate(images):
         clip_path = Path(temp_dir) / f"clip_{i}.mp4"
         text = texts[i] if i < len(texts) else ""
-        y_pos = y_positions[i] if i < len(y_positions) else 0.5
 
         drawtext = ""
         if text:
-            # Smooth Fade-In / Fade-Out, Y-Position dynamisch
             drawtext = (
-                f"drawtext=text='{text}':fontcolor=white:fontsize=40:borderw=2:"
+                f"drawtext=text='{text}':fontcolor=white:fontsize={font_size}:borderw=2:"
                 f"x=(w-text_w)/2:y=(h-text_h)*{y_pos}:"
                 f"alpha='if(lt(t,{fade_duration}), t/{fade_duration}, if(lt(t,{duration}-{fade_duration}), 1, ({duration}-t)/{fade_duration}))'"
             )
 
+        # FFmpeg scale + pad für schwarzen Rand
         cmd = [
             "ffmpeg",
             "-y",
@@ -45,7 +42,7 @@ def generate_slideshow_with_text_ypos(images, duration, texts=None, y_positions=
             "-i", str(img_path),
             "-t", str(duration),
             "-vf",
-            f"scale=trunc(iw/2)*2:trunc(ih/2)*2,fps=25,format=yuv420p{','+drawtext if drawtext else ''}",
+            f"scale='min(1280,iw)':-2,pad=1280:720:(1280-iw)/2:(720-ih)/2:color=black,fps=25,format=yuv420p{','+drawtext if drawtext else ''}",
             str(clip_path)
         ]
         subprocess.run(cmd, check=True)
@@ -72,31 +69,32 @@ def generate_slideshow_with_text_ypos(images, duration, texts=None, y_positions=
     subprocess.run(cmd_concat, check=True)
 
     if output_file.exists():
-        return str(output_file), f"✅ Slideshow erstellt ({len(images)} Bilder mit Text + Y-Position)"
+        return str(output_file), f"✅ Slideshow erstellt ({len(images)} Bilder, Text fixiert, Bildproportionen erhalten)"
     else:
         return None, "❌ Video konnte nicht erstellt werden"
 
+
 # Gradio UI
 with gr.Blocks() as demo:
-    gr.Markdown("# Slideshow mit Text und Y-Position")
+    gr.Markdown("# Slideshow mit Text-Overlay & schwarzem Rand")
 
     img_input = gr.Files(label="Bilder auswählen (mehrere)", file_types=allowed_medias)
     duration_input = gr.Number(value=3, label="Dauer pro Bild in Sekunden", precision=1)
     fade_input = gr.Number(value=0.7, label="Fade Dauer in Sekunden", precision=1)
     text_input = gr.Textbox(label="Texte pro Bild (mit Komma trennen)", placeholder="Text1, Text2, Text3 ...")
-    ypos_input = gr.Textbox(label="Y-Position pro Bild (0=oben,0.5=mitte,1=unten, mit Komma trennen)", placeholder="0.2,0.5,0.8")
+    ypos_input = gr.Slider(minimum=0.0, maximum=1.0, step=0.01, value=0.5, label="Y-Position für alle Texte (0=oben, 0.5=mitte, 1=unten, max 0.9)")
+    font_size_input = gr.Number(value=60, label="Textgröße (px)")
     out_video = gr.Video(interactive=False, label="Generiertes Video")
     status = gr.Textbox(interactive=False, label="Status")
 
-    def wrapper(images, duration, fade_duration, text_str, ypos_str):
+    def wrapper(images, duration, fade_duration, text_str, y_pos, font_size):
         texts = [t.strip() for t in text_str.split(",")] if text_str else None
-        y_positions = [float(y.strip()) for y in ypos_str.split(",")] if ypos_str else None
-        return generate_slideshow_with_text_ypos(images, duration, texts, y_positions, fade_duration)
+        return generate_slideshow_with_fixed_text_and_padding(images, duration, texts, y_pos, fade_duration, font_size)
 
     btn = gr.Button("Video erstellen")
     btn.click(
         fn=wrapper,
-        inputs=[img_input, duration_input, fade_input, text_input, ypos_input],
+        inputs=[img_input, duration_input, fade_input, text_input, ypos_input, font_size_input],
         outputs=[out_video, status]
     )