save fix

init_defaults.py

@@ -26,7 +26,7 @@ defaults = {
         "max_words_per_line": 3,
         "uppercase": False,
         "letter_spacing": 0.0,
-        "background_opacity": 1.0,
+        "background_opacity": 0.7,
         "glow_intensity": 0,
         "rotation_angle": 0.0,
         "margin_h": 20
@@ -45,15 +45,17 @@ defaults = {
         "margin_v": 120,
         "pop_up_scale": 1.5,
         "highlight_mode": "karaoke",
-        "back_box_enabled": False,
+        "back_box_enabled": True, # تفعيل البوكس للجيمنج أيضاً
         "display_mode": "sentence",
         "max_words_per_line": 2,
         "uppercase": True,
         "letter_spacing": 1.0,
-        "background_opacity": 0.8,
+        "background_opacity": 0.6,
         "glow_intensity": 2,
         "rotation_angle": -3.0,
-        "margin_h": 20
+        "margin_h": 20,
+        "box_highlight_type": "word",
+        "box_rounding": 15
     },
     "professional": {
         "name": "professional",
@@ -77,7 +79,9 @@ defaults = {
         "background_opacity": 0.9,
         "glow_intensity": 0,
         "rotation_angle": 0.0,
-        "margin_h": 40
+        "margin_h": 40,
+        "box_highlight_type": "sentence",
+        "box_rounding": 5
     },
     "storyteller": {
         "name": "storyteller",

routers/subtitle_generator.py

@@ -30,7 +30,10 @@ def generate_pro_ass(transcription, preset, output_path):
     # تحويل الشفافية (0.0 - 1.0) إلى صيغة ASS (00 - FF)
     # ملاحظة: في ASS، الـ 00 هو معتم تماماً و FF هو شفاف تماماً
     back_alpha = int((1.0 - getattr(preset, 'background_opacity', 1.0)) * 255)
-    back_color = f"&H{back_alpha:02X}000000" # لون خلفية أسود مع شفافية متغيرة
+    
+    # استخدام لون الخلفية من الـ preset بدلاً من الأسود الثابت
+    raw_back_color = rgb_to_ass(preset.back_color).replace("&H00", "")
+    back_color = f"&H{back_alpha:02X}{raw_back_color}"
     
     spacing = getattr(preset, 'letter_spacing', 0.0)
     angle = getattr(preset, 'rotation_angle', 0.0)
@@ -45,7 +48,7 @@ ScaledBorderAndShadow: yes
 [V4+ Styles]
 Format: Name, Fontname, Fontsize, PrimaryColour, SecondaryColour, OutlineColour, BackColour, Bold, Italic, Underline, StrikeOut, ScaleX, ScaleY, Spacing, Angle, BorderStyle, Outline, Shadow, Alignment, MarginL, MarginR, MarginV, Encoding
 Style: Main,{preset.font_name},{preset.font_size},{primary},{secondary},{outline},&H00000000,-1,0,0,0,100,100,{spacing},{angle},1,{preset.outline_width},{preset.shadow_depth},{preset.alignment},{margin_h},{margin_h},{preset.margin_v},1
-Style: BackBox,{preset.font_name},{preset.font_size},{secondary},{secondary},&H00000000,{back_color},-1,0,0,0,100,100,{spacing},{angle},3,10,0,{preset.alignment},{margin_h},{margin_h},{preset.margin_v},1
+Style: BackBox,{preset.font_name},{preset.font_size},{secondary},{secondary},&H00000000,{back_color},-1,0,0,0,100,100,{spacing},{angle},3,{getattr(preset, 'box_rounding', 10)},0,{preset.alignment},{margin_h},{margin_h},{preset.margin_v},1
 """
     
     events = "\n[Events]\nFormat: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text\n"
@@ -126,38 +129,59 @@ Style: BackBox,{preset.font_name},{preset.font_size},{secondary},{secondary},&H0
                 
                 # جعل النص الأساسي شفاف قليلاً (dimmed) وإخفاء الكلمة النشطة بـ \alpha&HFF&
                 dim_alpha = "\\1a&H80&"
-                base_display_text = f"{{{dim_alpha}}}{prefix_base}{{\\alpha&HFF&}}{target_word_base}{{\\alpha&H00&}}{{{dim_alpha}}}{suffix_base}"
+                box_highlight_type = getattr(preset, 'box_highlight_type', 'word')
                 
-                if preset.back_box_enabled:
+                # إذا كان النوع 'sentence'، نظهر الصندوق حول الجملة كلها في طبقة الخلفية
+                if preset.back_box_enabled and box_highlight_type == "sentence":
+                    box_hide = "" # لا نخفي البوكس عن الكلمة النشطة في طبقة الخلفية لأنه للجملة كلها
+                    base_display_text = f"{{{dim_alpha}}}{prefix_base}{{\\alpha&HFF&}}{target_word_base}{{\\alpha&H00&}}{{{dim_alpha}}}{suffix_base}"
                     events += f"Dialogue: 0,{w_start},{w_end},BackBox,,0,0,0,,{base_display_text}\n"
                 else:
+                    # النمط الافتراضي أو 'word': الصندوق يختفي من الخلفية ليظهر في طبقة الهايلايت فقط
+                    box_hide = "\\4a&HFF&" 
+                    base_display_text = f"{{{dim_alpha}}}{prefix_base}{{\\alpha&HFF&{box_hide}}}{target_word_base}{{\\alpha&H00&}}{{{dim_alpha}}}{suffix_base}"
                     events += f"Dialogue: 0,{w_start},{w_end},Main,,0,0,0,,{base_display_text}\n"
 
-                # 2. طبقة الكلمات النشطة (Highlight / Popup)
+                # 2. طبقة الكلمة النشطة (Highlight)
                 prefix_hl = " ".join(words_list_base[:i]) + (" " if i > 0 else "")
                 suffix_hl = (" " if i < len(words_list_base)-1 else "") + " ".join(words_list_base[i+1:])
                 target_word_hl = words_list_base[i]
                 
                 t1 = min(60, int(duration * 150))
                 t2 = t1 + 80
-                scale = int(preset.pop_up_scale * 100)
                 
                 # إضافة تأثير التوهج (Blur)
                 glow_val = getattr(preset, 'glow_intensity', 0)
                 blur_tag = f"\\be{glow_val}" if glow_val > 0 else ""
                 
-                # الـ Highlight يكون بلون الـ secondary وبدون شفافية مع تأثير Pop وميلان
                 if angle != 0:
                     rotation = f"\\frz{angle}"
                 else:
                     rotation = "\\frz-2" if i % 2 == 0 else "\\frz2"
                 
-                anim = f"{{\\1c{secondary}\\1a&H00&\\fscx80\\fscy80{rotation}{blur_tag}\\t(0,{t1},0.3,\\fscx{scale}\\fscy{scale})\\t({t1},{t2},1.5,\\fscx100\\fscy100)}}"
-                
-                # استخدام \alpha&HFF& لإخفاء الكلمات غير النشطة في هذه الطبقة فقط
-                display_text_hl = f"{{\\alpha&HFF&}}{prefix_hl}{anim}{target_word_hl}{{\\alpha&HFF&}}{suffix_hl}"
-                
-                events += f"Dialogue: 1,{w_start},{w_end},Main,,0,0,0,,{display_text_hl}\n"
+                # إعدادات الحركة واللون
+                scale = int(preset.pop_up_scale * 100)
+                if preset.back_box_enabled:
+                    if box_highlight_type == "word":
+                        # الصندوق حول الكلمة فقط: نستخدم BackBox مع إظهار الصندوق والكلمة بوضوح (شفافية 00)
+                        # \1c=نص، \4c=صندوق، \1a=شفافية نص، \4a=شفافية صندوق
+                        color_tags = f"\\1c{primary}\\4c{secondary}\\1a&H00&\\4a&H00&"
+                        anim = f"{{{color_tags}\\fscx{scale}\\fscy{scale}{rotation}{blur_tag}}}"
+                        # إخفاء النص والصندوق في الأجزاء غير النشطة
+                        display_text_hl = f"{{\\alpha&HFF&\\4a&HFF&}}{prefix_hl}{anim}{target_word_hl}{{\\alpha&HFF&\\4a&HFF&}}{suffix_hl}"
+                        events += f"Dialogue: 1,{w_start},{w_end},BackBox,,0,0,0,,{display_text_hl}\n"
+                    else:
+                        # الصندوق حول الجملة: الهايلايت هو تغيير لون النص مع تأثير Pop-up
+                        color_tags = f"\\1c{secondary}\\1a&H00&"
+                        anim = f"{{{color_tags}\\fscx{scale}\\fscy{scale}{rotation}{blur_tag}}}"
+                        display_text_hl = f"{{\\alpha&HFF&}}{prefix_hl}{anim}{target_word_hl}{{\\alpha&HFF&}}{suffix_hl}"
+                        events += f"Dialogue: 1,{w_start},{w_end},Main,,0,0,0,,{display_text_hl}\n"
+                else:
+                    # بدون صناديق نهائياً: هايلايت نصي فقط
+                    color_tags = f"\\1c{secondary}\\1a&H00&"
+                    anim = f"{{{color_tags}\\fscx{scale}\\fscy{scale}{rotation}{blur_tag}}}"
+                    display_text_hl = f"{{\\alpha&HFF&}}{prefix_hl}{anim}{target_word_hl}{{\\alpha&HFF&}}{suffix_hl}"
+                    events += f"Dialogue: 1,{w_start},{w_end},Main,,0,0,0,,{display_text_hl}\n"
 
     with open(output_path, "w", encoding="utf-8-sig") as f:
         f.write(header + events)

routers/video.py

@@ -145,7 +145,7 @@ def background_processing(
             if f and os.path.exists(f) and TEMP_DIR in f:
                 safe_remove(f)
 
-@router.get("/clear-preview", tags=["Video"])
+@router.get("/privevies-files", tags=["Video"])
 async def get_clear_preview(folder: str = "temp"):
     """
     Preview files that will be deleted before actually clearing them.
@@ -433,7 +433,7 @@ async def list_saved_files():
 
 @router.api_route("/clear", methods=["GET", "POST"], tags=["Video"])
 async def clear_temp_files(
-    folder: str = Query("temp", description="Folder to clear", enum=["temp", "processed", "audio", "originals", "all"]),
+    folder: str = Query("temp", description="Folder to clear", enum=["temp", "processed", "bgmusic", "originals", "all"]),
     specific_files: str = Query(None, description="Comma-separated list of specific files to delete (optional). Use /clear-preview to see available files.")
 ):
     """
@@ -452,8 +452,7 @@ async def clear_temp_files(
             target_folders = [TEMP_DIR]
         elif folder == "processed":
             target_folders = [PROCESSED_DIR]
-        elif folder == "audio":
-            target_folders = [AUDIO_DIR]
+       
         elif folder == "bgmusic":
             target_folders = [BG_MUSIC_DIR]
         elif folder == "originals":
@@ -461,7 +460,7 @@ async def clear_temp_files(
         else:
             return JSONResponse(
                 status_code=400,
-                content={"error": f"Invalid folder: {folder}. Use: temp, processed, audio, originals, or all"}
+                content={"error": f"Invalid folder: {folder}. Use: temp, processed, bgmusic, originals, or all"}
             )
 
         deleted_files = []
@@ -640,6 +639,7 @@ async def process_saved_video(
     original_filename: str = Form(..., description="اسم الفيديو الأصلي المحفوظ"),
     aspect_ratio: AspectRatio = Form(AspectRatio.RATIO_9_16, description="Select target aspect ratio"),
     style: ShortsStyle = Form(ShortsStyle.CINEMATIC, description="Select visual style"),
+    background_music: UploadFile = File(None, description="Upload a new background music file"),
     background_music_filename: str = Form(None, description="Name of a saved background music file (optional)"),
     video_volume: float = Form(1.0, description="Volume of original video"),
     music_volume: float = Form(0.2, description="Volume of background music"),
@@ -647,6 +647,7 @@ async def process_saved_video(
     timestamps_json: str = Form(None, alias="timestamps", description='JSON list of timestamps.'),
     export_audio: bool = Form(False, description="Export separate audio files"),
     webhook_url: str = Form(None, description="Optional URL for results"),
+    return_files: bool = Form(False, description="Return processed files directly"),
     transcription_json: str = Form(None, alias="transcription", description="JSON string of transcription data"),
     subtitle_style: str = Form(None, description="Name of the saved subtitle preset"),
     subtitle_preset_json: str = Form(None, alias="subtitle_preset", description="Custom subtitle preset JSON override")
@@ -671,9 +672,13 @@ async def process_saved_video(
         temp_path = os.path.join(TEMP_DIR, f"{task_id}_temp_{original_filename}")
         shutil.copy2(original_path, temp_path)
         
-        # Handle background music if name provided
+        # Handle background music (New upload takes priority)
         audio_path = None
-        if background_music_filename:
+        if background_music:
+            audio_path = os.path.join(BG_MUSIC_DIR, f"bg_{task_id}_{background_music.filename}")
+            with open(audio_path, "wb") as buffer:
+                shutil.copyfileobj(background_music.file, buffer)
+        elif background_music_filename:
             audio_path = os.path.join(BG_MUSIC_DIR, background_music_filename)
             if not os.path.exists(audio_path):
                 print(f"Warning: Background music {background_music_filename} not found.")

tests/test_comprehensive_e2e.py

@@ -29,107 +29,139 @@ def create_mock_video_if_needed(filename="0eb30aa9_original_my_movie.mp4"):
         ], capture_output=True)
     return filename
 
+def create_mock_audio():
+    audio_filename = "test_bg_music.mp3"
+    path = os.path.join(TEMP_DIR, audio_filename)
+    if not os.path.exists(path):
+        print(f"Creating mock audio at {path}...")
+        import imageio_ffmpeg
+        import subprocess
+        exe = imageio_ffmpeg.get_ffmpeg_exe()
+        subprocess.run([
+            exe, '-f', 'lavfi', '-i', 'sine=frequency=440:duration=5',
+            '-t', '5', path, '-y'
+        ], capture_output=True)
+    return path
+
 def test_full_workflow():
-    print("\n--- Starting Comprehensive E2E Test (New Sync System) ---")
+    print("\n" + "="*50)
+    print("🚀 STARTING MULTI-PRESET CINEMATIC E2E TEST")
+    print("="*50)
     
-    # 1. Create a Custom Preset with Sentence Mode
-    preset_name = "e2e_sentence_style"
-    print(f"Step 1: Saving preset '{preset_name}' with display_mode='sentence'...")
-    preset_data = {
-        "name": (None, preset_name),
-        "font_name": (None, "Arial"),
-        "font_size": (None, "55"),
-        "primary_color": (None, "#FFFFFF"),
-        "secondary_color": (None, "#00FF00"), # Green Highlight
-        "display_mode": (None, "sentence"),   # الميزة الجديدة
-        "max_words_per_line": (None, "4"),    # الميزة الجديدة
-        "back_box_enabled": (None, "true")
-    }
-    resp = client.post("/api/presets/save", data=preset_data)
-    assert resp.status_code == 200, f"Failed to save preset: {resp.text}"
-    print("✅ Sentence preset saved successfully.")
-
-    # 2. Verify Preset List
-    print("Step 2: Verifying preset list...")
-    resp = client.get("/api/presets/list")
-    assert preset_name in resp.json(), "Preset not found in list"
-    print(f"✅ Preset '{preset_name}' is in the list.")
-
-    # 3. Create/Ensure original video exists
+    # 1. Ensure mock assets exist
     video_name = create_mock_video_if_needed()
+    audio_file_path = create_mock_audio()
+    
+    # List of default presets to test
+    default_presets = ["youtube_casual", "gaming", "professional", "storyteller"]
     
-    # 4. Test Manual Sentence Timing Sync (The New System)
-    print(f"Step 3: Testing Manual Sentence Timing Sync with '{preset_name}'...")
-    # بيانات بتنسيق SentenceTiming (جملة بداخلها كلمات)
-    transcription_new = [
+    transcription = [
         {
-            "text": "This is the new sync system",
+            "text": "Testing all default presets with cinematic style",
             "start": 0.0,
-            "end": 3.0,
-            "words": [
-                {"word": "This", "start": 0.0, "end": 0.5},
-                {"word": "is", "start": 0.5, "end": 1.0},
-                {"word": "the", "start": 1.0, "end": 1.5},
-                {"word": "new", "start": 1.5, "end": 2.2},
-                {"word": "sync", "start": 2.2, "end": 2.6},
-                {"word": "system", "start": 2.6, "end": 3.0}
-            ]
-        },
-        {
-            "text": "It works perfectly with presets",
-            "start": 3.0,
             "end": 5.0,
             "words": [
-                {"word": "It", "start": 3.0, "end": 3.5},
-                {"word": "works", "start": 3.5, "end": 4.0},
-                {"word": "perfectly", "start": 4.0, "end": 4.5},
-                {"word": "with", "start": 4.5, "end": 4.8},
-                {"word": "presets", "start": 4.8, "end": 5.0}
+                {"word": "Testing", "start": 0.0, "end": 1.0},
+                {"word": "all", "start": 1.0, "end": 1.5},
+                {"word": "default", "start": 1.5, "end": 2.5},
+                {"word": "presets", "start": 2.5, "end": 3.5},
+                {"word": "cinematic", "start": 3.5, "end": 5.0}
             ]
         }
     ]
+
+    for preset in default_presets:
+        print(f"\n[Feature Test] 🎬 Cinematic Style + 🎨 Preset: {preset}")
+        
+        process_data = {
+            "original_filename": (None, video_name),
+            "aspect_ratio": (None, "9:16"),
+            "style": (None, "cinematic"),
+            "subtitle_style": (None, preset),
+            "transcription": (None, json.dumps(transcription)),
+            "timestamps": (None, json.dumps([{"start_time": 0, "end_time": 5}])),
+            "music_volume": (None, "0.3"),
+            "video_volume": (None, "1.0"),
+            "loop_music": (None, "true")
+        }
+        
+        with open(audio_file_path, "rb") as f:
+            files = {"background_music": ("bg.mp3", f, "audio/mpeg")}
+            resp = client.post("/api/video/process-saved", data=process_data, files=files)
+            
+        assert resp.status_code == 200, f"Process failed for preset {preset}: {resp.text}"
+        print(f"✅ Preset '{preset}' processed successfully in Cinematic style.")
+
+    # 2. Test Custom Pro Preset with Background Box Color
+    # 2. Test Word-Based Box Highlight (Box moves with word)
+    print(f"\n[Feature Test] 🏃 Word-Based Box Highlight (Moving Box)")
+    word_preset_name = "word_box_moving"
+    word_preset_data = {
+        "name": (None, word_preset_name),
+        "font_name": (None, "Impact"),
+        "font_size": "60",
+        "primary_color": (None, "#FFFFFF"), # نص أبيض
+        "secondary_color": (None, "#0000FF"), # صندوق أزرق
+        "box_highlight_type": (None, "word"),  # الصندوق يتحرك مع الكلمة
+        "back_box_enabled": (None, "true"),
+        "pop_up_scale": (None, "1.3"),        # بوب أب للكلمة
+        "background_opacity": (None, "0.8")
+    }
+    client.post("/api/presets/save", data=word_preset_data)
     
-    process_data = {
-        "original_filename": video_name,
-        "aspect_ratio": "9:16",
-        "style": "cinematic",
-        "subtitle_style": preset_name,
-        "transcription": json.dumps(transcription_new),
-        "timestamps": json.dumps([{"start_time": 0, "end_time": 5}])
+    # 3. Test Sentence-Based Box Highlight (Static Box + Word Popup)
+    print(f"\n[Feature Test] 🖼️ Sentence-Based Box Highlight (Static Box + Word Popup)")
+    sent_preset_name = "sent_box_static"
+    sent_preset_data = {
+        "name": (None, sent_preset_name),
+        "font_name": (None, "Arial"),
+        "font_size": "55",
+        "primary_color": (None, "#FFFFFF"),
+        "secondary_color": (None, "#00FF00"), # هايلايت أخضر
+        "box_highlight_type": (None, "sentence"), # صندوق حول الجملة كلها
+        "box_rounding": (None, "15"),
+        "back_box_enabled": (None, "true"),
+        "pop_up_scale": (None, "1.4"),           # بوب أب للكلمة داخل الصندوق
+        "background_opacity": (None, "0.6")
     }
+    client.post("/api/presets/save", data=sent_preset_data)
     
-    resp = client.post("/api/video/process-saved", data=process_data)
-    assert resp.status_code == 200, f"Process-saved failed: {resp.text}"
-    result = resp.json()
-    assert result["status"] == "success"
-    print(f"✅ New Sync System processed successfully. Task ID: {result['task_id']}")
+    # Run a test for both
+    for p_name in [word_preset_name, sent_preset_name]:
+        process_data = {
+            "original_filename": (None, video_name),
+            "aspect_ratio": (None, "9:16"),
+            "style": (None, "cinematic"),
+            "subtitle_style": (None, p_name),
+            "transcription": (None, json.dumps(transcription))
+        }
+        resp = client.post("/api/video/process-saved", data=process_data)
+        assert resp.status_code == 200
+        print(f"✅ Tested {p_name} successfully.")
 
-    # 5. Test Backward Compatibility (Old Word List Format)
-    print("Step 4: Testing Backward Compatibility (Old Word List Format)...")
-    transcription_old = [
-        {"word": "Old", "start": 0.0, "end": 1.0},
-        {"word": "format", "start": 1.0, "end": 2.0},
-        {"word": "still", "start": 2.0, "end": 3.0},
-        {"word": "works", "start": 3.0, "end": 5.0}
-    ]
-    process_data["transcription"] = json.dumps(transcription_old)
-    process_data["subtitle_style"] = "gaming" # استخدام بريسيت افتراضي قديم
+    # Cleanup
+    client.delete(f"/api/presets/delete/{word_preset_name}")
+    client.delete(f"/api/presets/delete/{sent_preset_name}")
     
-    resp = client.post("/api/video/process-saved", data=process_data)
+    process_data_custom = {
+        "original_filename": (None, video_name),
+        "aspect_ratio": (None, "9:16"),
+        "style": (None, "cinematic"),
+        "subtitle_style": (None, preset_name),
+        "transcription": (None, json.dumps(transcription))
+    }
+    resp = client.post("/api/video/process-saved", data=process_data_custom)
     assert resp.status_code == 200
-    print("✅ Backward compatibility verified successfully.")
-
-    # 6. Check File Listing
-    print("Step 5: Checking file listing...")
-    resp = client.get("/api/video/list-saved")
-    files = resp.json()["files"]
-    assert len(files["processed"]) >= 2
-    print(f"✅ Found {len(files['processed'])} processed files.")
+    print(f"✅ Custom Preset with Blue Box processed successfully.")
 
-    # 7. Cleanup
-    print("Step 6: Cleaning up test preset...")
+    # 3. Cleanup
     client.delete(f"/api/presets/delete/{preset_name}")
-    print("✅ E2E Test completed successfully.")
+    if os.path.exists(audio_file_path):
+        os.remove(audio_file_path)
+    
+    print("\n" + "="*50)
+    print("🏁 ALL MULTI-PRESET TESTS PASSED!")
+    print("="*50)
 
 if __name__ == "__main__":
     test_full_workflow()