Update core/visual_engine.py

core/visual_engine.py

@@ -1,7 +1,7 @@
 # core/visual_engine.py
 from PIL import Image, ImageDraw, ImageFont, ImageOps
 import base64
-import mimetypes
+import mimetypes # For Data URI
 import numpy as np
 import os
 import openai # OpenAI v1.x.x+
@@ -11,36 +11,42 @@ import time
 import random
 import logging
 
+# --- MoviePy Imports ---
 from moviepy.editor import (ImageClip, VideoFileClip, concatenate_videoclips, TextClip,
                             CompositeVideoClip, AudioFileClip)
 import moviepy.video.fx.all as vfx
 
-try: # MONKEY PATCH for Pillow/MoviePy compatibility
-    if hasattr(Image, 'Resampling') and hasattr(Image.Resampling, 'LANCZOS'):
+# --- MONKEY PATCH for Pillow/MoviePy compatibility ---
+try:
+    if hasattr(Image, 'Resampling') and hasattr(Image.Resampling, 'LANCZOS'): # Pillow 9+
         if not hasattr(Image, 'ANTIALIAS'): Image.ANTIALIAS = Image.Resampling.LANCZOS
-    elif hasattr(Image, 'LANCZOS'):
+    elif hasattr(Image, 'LANCZOS'): # Pillow 8
          if not hasattr(Image, 'ANTIALIAS'): Image.ANTIALIAS = Image.LANCZOS
     elif not hasattr(Image, 'ANTIALIAS'):
-             print("WARNING: Pillow version lacks common Resampling or ANTIALIAS. MoviePy effects might fail.")
-except Exception as e_mp: print(f"WARNING: ANTIALIAS monkey-patch error: {e_mp}")
+             print("WARNING: Pillow version lacks common Resampling attributes or ANTIALIAS. MoviePy effects might fail or look different.")
+except Exception as e_monkey_patch:
+    print(f"WARNING: An unexpected error occurred during Pillow ANTIALIAS monkey-patch: {e_monkey_patch}")
 
 logger = logging.getLogger(__name__)
-# logger.setLevel(logging.DEBUG) # Uncomment for maximum verbosity
+# logger.setLevel(logging.DEBUG) # Uncomment for verbose debugging during development
 
+# --- External Service Client Imports ---
 ELEVENLABS_CLIENT_IMPORTED = False; ElevenLabsAPIClient = None; Voice = None; VoiceSettings = None
 try:
     from elevenlabs.client import ElevenLabs as ImportedElevenLabsClient
     from elevenlabs import Voice as ImportedVoice, VoiceSettings as ImportedVoiceSettings
     ElevenLabsAPIClient = ImportedElevenLabsClient; Voice = ImportedVoice; VoiceSettings = ImportedVoiceSettings
-    ELEVENLABS_CLIENT_IMPORTED = True; logger.info("ElevenLabs client components imported.")
-except Exception as e_11l_imp: logger.warning(f"ElevenLabs client import failed: {e_11l_imp}. Audio disabled.")
+    ELEVENLABS_CLIENT_IMPORTED = True; logger.info("ElevenLabs client components (SDK v1.x.x pattern) imported successfully.")
+except ImportError: logger.warning("ElevenLabs SDK not found (expected 'pip install elevenlabs>=1.0.0'). Audio generation will be disabled.")
+except Exception as e_eleven_import_general: logger.warning(f"General error importing ElevenLabs client components: {e_eleven_import_general}. Audio generation disabled.")
 
 RUNWAYML_SDK_IMPORTED = False; RunwayMLAPIClientClass = None
 try:
     from runwayml import RunwayML as ImportedRunwayMLAPIClientClass
     RunwayMLAPIClientClass = ImportedRunwayMLAPIClientClass; RUNWAYML_SDK_IMPORTED = True
-    logger.info("RunwayML SDK imported.")
-except Exception as e_rwy_imp: logger.warning(f"RunwayML SDK import failed: {e_rwy_imp}. RunwayML disabled.")
+    logger.info("RunwayML SDK (runwayml) imported successfully.")
+except ImportError: logger.warning("RunwayML SDK not found (pip install runwayml). RunwayML video generation will be disabled.")
+except Exception as e_runway_sdk_import_general: logger.warning(f"General error importing RunwayML SDK: {e_runway_sdk_import_general}. RunwayML features disabled.")
 
 
 class VisualEngine:
@@ -60,8 +66,7 @@ class VisualEngine:
         else: logger.warning("Preferred Pillow font not found. Default.")
         self.openai_api_key = None; self.USE_AI_IMAGE_GENERATION = False; self.dalle_model = "dall-e-3"; self.image_size_dalle3 = "1792x1024"
         self.video_frame_size = (1280, 720)
-        self.elevenlabs_api_key = None; self.USE_ELEVENLABS = False; self.elevenlabs_client_instance = None
-        self.elevenlabs_voice_id = default_elevenlabs_voice_id # Set initial voice ID from constructor
+        self.elevenlabs_api_key = None; self.USE_ELEVENLABS = False; self.elevenlabs_client_instance = None; self.elevenlabs_voice_id = default_elevenlabs_voice_id
         if VoiceSettings and ELEVENLABS_CLIENT_IMPORTED: self.elevenlabs_voice_settings_obj = VoiceSettings(stability=0.60, similarity_boost=0.80, style=0.15, use_speaker_boost=True)
         else: self.elevenlabs_voice_settings_obj = None
         self.pexels_api_key = None; self.USE_PEXELS = False
@@ -72,31 +77,13 @@ class VisualEngine:
         logger.info("VisualEngine initialized.")
 
     def set_openai_api_key(self, api_key_value): self.openai_api_key = api_key_value; self.USE_AI_IMAGE_GENERATION = bool(api_key_value); logger.info(f"DALL-E status: {'Ready' if self.USE_AI_IMAGE_GENERATION else 'Disabled'}")
-    
-    # <<< CORRECTED METHOD SIGNATURE AND LOGIC >>>
     def set_elevenlabs_api_key(self, api_key_value, voice_id_from_secret=None):
-        self.elevenlabs_api_key = api_key_value # Store the API key
-
-        if voice_id_from_secret: # If a specific voice ID is passed, update the instance's default
-            self.elevenlabs_voice_id = voice_id_from_secret
-            logger.info(f"ElevenLabs Voice ID updated to: {self.elevenlabs_voice_id} via set_elevenlabs_api_key.")
-        # If voice_id_from_secret is None, self.elevenlabs_voice_id retains the value from __init__
-
+        self.elevenlabs_api_key = api_key_value
+        if voice_id_from_secret: self.elevenlabs_voice_id = voice_id_from_secret; logger.info(f"11L Voice ID updated to: {self.elevenlabs_voice_id} via set_elevenlabs_api_key.")
         if api_key_value and ELEVENLABS_CLIENT_IMPORTED and ElevenLabsAPIClient:
-            try:
-                self.elevenlabs_client_instance = ElevenLabsAPIClient(api_key=api_key_value)
-                self.USE_ELEVENLABS = bool(self.elevenlabs_client_instance)
-                logger.info(f"ElevenLabs Client service status: {'Ready' if self.USE_ELEVENLABS else 'Failed Initialization'} (Using Voice ID: {self.elevenlabs_voice_id})")
-            except Exception as e_11l_setkey_init:
-                logger.error(f"ElevenLabs client initialization error during set_elevenlabs_api_key: {e_11l_setkey_init}. Service Disabled.", exc_info=True)
-                self.USE_ELEVENLABS = False
-                self.elevenlabs_client_instance = None
-        else:
-            self.USE_ELEVENLABS = False
-            self.elevenlabs_client_instance = None
-            if not api_key_value: logger.info(f"ElevenLabs Service Disabled (API key not provided).")
-            elif not (ELEVENLABS_CLIENT_IMPORTED and ElevenLabsAPIClient): logger.info(f"ElevenLabs Service Disabled (SDK issue).")
-
+            try: self.elevenlabs_client_instance = ElevenLabsAPIClient(api_key=api_key_value); self.USE_ELEVENLABS = bool(self.elevenlabs_client_instance); logger.info(f"11L Client: {'Ready' if self.USE_ELEVENLABS else 'Failed'} (Voice: {self.elevenlabs_voice_id})")
+            except Exception as e_11l_setkey_init: logger.error(f"11L client init error: {e_11l_setkey_init}. Disabled.", exc_info=True); self.USE_ELEVENLABS=False; self.elevenlabs_client_instance=None
+        else: self.USE_ELEVENLABS = False; logger.info(f"11L Disabled (key/SDK).")
     def set_pexels_api_key(self, api_key_value): self.pexels_api_key = api_key_value; self.USE_PEXELS = bool(api_key_value); logger.info(f"Pexels status: {'Ready' if self.USE_PEXELS else 'Disabled'}")
     def set_runway_api_key(self, api_key_value):
         self.runway_api_key = api_key_value
@@ -113,35 +100,38 @@ class VisualEngine:
             else: logger.warning("RunwayML SDK not imported. Service disabled."); self.USE_RUNWAYML = False
         else: self.USE_RUNWAYML = False; self.runway_ml_sdk_client_instance = None; logger.info("RunwayML Disabled (no API key).")
 
-    # ... (Rest of the methods: _image_to_data_uri, _map_resolution_to_runway_ratio, _get_text_dimensions,
-    #      _create_placeholder_image_content, _search_pexels_image, _generate_video_clip_with_runwayml,
-    #      _create_placeholder_video_content, generate_scene_asset, generate_narration_audio,
-    #      assemble_animatic_from_assets - keep these as they were in the last fully corrected version
-    #      that addressed the previous syntax errors and had robust image processing for MoviePy)
-
-    # For brevity, I'm re-pasting only the corrected _create_placeholder_image_content and _search_pexels_image
-    # and assuming the other long methods like generate_scene_asset and assemble_animatic_from_assets
-    # are taken from the previous "expertly crafted" full version which already had robust logic.
-    # Make sure to use the complete, most up-to-date versions of ALL methods.
-
     def _image_to_data_uri(self, image_path):
-        # (Implementation from before)
-        try: mime_type,_=mimetypes.guess_type(image_path)
-        if not mime_type:ext=os.path.splitext(image_path)[1].lower();mime_map={".png":"image/png",".jpg":"image/jpeg",".jpeg":"image/jpeg",".webp":"image/webp"};mime_type=mime_map.get(ext,"application/octet-stream");
-        if mime_type=="application/octet-stream":logger.warning(f"Unknown MIME for {image_path}, using {mime_type}.")
-        with open(image_path,"rb")as image_file:encoded_string=base64.b64encode(image_file.read()).decode('utf-8')
-        data_uri=f"data:{mime_type};base64,{encoded_string}";logger.debug(f"Data URI for {os.path.basename(image_path)} (MIME:{mime_type}): {data_uri[:100]}...");return data_uri
-        except FileNotFoundError:logger.error(f"Img not found {image_path} for data URI.");return None
-        except Exception as e:logger.error(f"Error converting {image_path} to data URI:{e}",exc_info=True);return None
+        # <<< CORRECTED METHOD >>>
+        try:
+            mime_type, _ = mimetypes.guess_type(image_path)
+            if not mime_type:
+                ext = os.path.splitext(image_path)[1].lower()
+                mime_map = {".png": "image/png", ".jpg": "image/jpeg", ".jpeg": "image/jpeg", ".webp": "image/webp"}
+                mime_type = mime_map.get(ext, "application/octet-stream")
+                if mime_type == "application/octet-stream":
+                    logger.warning(f"Could not determine MIME type for {image_path} from extension '{ext}', using default {mime_type}.")
+            
+            with open(image_path, "rb") as image_file_handle:
+                image_binary_data = image_file_handle.read()
+            
+            encoded_base64_string = base64.b64encode(image_binary_data).decode('utf-8')
+            
+            data_uri_string = f"data:{mime_type};base64,{encoded_base64_string}"
+            logger.debug(f"Generated data URI for {os.path.basename(image_path)} (MIME: {mime_type}). Data URI starts with: {data_uri_string[:100]}...")
+            return data_uri_string
+        except FileNotFoundError:
+            logger.error(f"Image file not found at path: '{image_path}' when trying to create data URI.")
+            return None
+        except Exception as e_data_uri_conversion:
+            logger.error(f"Error converting image '{image_path}' to data URI: {e_data_uri_conversion}", exc_info=True)
+            return None
 
     def _map_resolution_to_runway_ratio(self, width, height):
-        # (Implementation from before)
         ratio_str=f"{width}:{height}";supported_ratios_gen4=["1280:720","720:1280","1104:832","832:1104","960:960","1584:672"];
         if ratio_str in supported_ratios_gen4:return ratio_str
         logger.warning(f"Res {ratio_str} not in Gen-4 list. Default 1280:720.");return "1280:720"
 
     def _get_text_dimensions(self, text_content, font_object_pil):
-        # (Implementation from before)
         dch=getattr(font_object_pil,'size',self.active_font_size_pil);
         if not text_content:return 0,dch
         try:
@@ -271,7 +261,7 @@ class VisualEngine:
                 except Exception as e_cl_phv: logger.warning(f"Ignoring error closing placeholder TextClip: {e_cl_phv}")
     
     def generate_scene_asset(self, image_generation_prompt_text, motion_prompt_text_for_video,
-                             scene_data_dict, scene_identifier_fn_base, # Changed scene_data to scene_data_dict
+                             scene_data_dict, scene_identifier_fn_base,
                              generate_as_video_clip_flag=False, runway_target_dur_val=5):
         # (Corrected DALL-E loop from previous response)
         base_name_asset, _ = os.path.splitext(scene_identifier_fn_base)
@@ -312,9 +302,9 @@ class VisualEngine:
         return asset_info_result
 
     def generate_narration_audio(self, text_to_narrate, output_filename="narration_overall.mp3"):
-        # <<< CORRECTED VERSION OF THIS METHOD >>>
+        # (Corrected version from previous response)
         if not self.USE_ELEVENLABS or not self.elevenlabs_client_instance or not text_to_narrate:
-            logger.info("ElevenLabs conditions not met (service disabled, client not init, or no text). Skipping audio generation.")
+            logger.info("ElevenLabs conditions not met. Skipping audio generation.")
             return None
         audio_filepath_narration = os.path.join(self.output_dir, output_filename)
         try: