core/visual_engine.py

# core/visual_engine.py
from PIL import Image, ImageDraw, ImageFont, ImageOps
import base64
import mimetypes
import numpy as np
import os
import openai # OpenAI v1.x.x+
import requests
import io
import time
import random
import logging

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'):
        if not hasattr(Image, 'ANTIALIAS'): Image.ANTIALIAS = Image.Resampling.LANCZOS
    elif hasattr(Image, 'LANCZOS'):
         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}")

logger = logging.getLogger(__name__)
# logger.setLevel(logging.DEBUG) # Uncomment for maximum verbosity

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.")

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.")


class VisualEngine:
    DEFAULT_FONT_SIZE_PIL = 10; PREFERRED_FONT_SIZE_PIL = 20
    VIDEO_OVERLAY_FONT_SIZE = 30; VIDEO_OVERLAY_FONT_COLOR = 'white'
    DEFAULT_MOVIEPY_FONT = 'DejaVu-Sans-Bold'; PREFERRED_MOVIEPY_FONT = 'Liberation-Sans-Bold'

    def __init__(self, output_dir="temp_cinegen_media", default_elevenlabs_voice_id="Rachel"):
        self.output_dir = output_dir; os.makedirs(self.output_dir, exist_ok=True)
        self.font_filename_pil_preference = "DejaVuSans-Bold.ttf"
        font_paths = [ self.font_filename_pil_preference, f"/usr/share/fonts/truetype/dejavu/{self.font_filename_pil_preference}", f"/usr/share/fonts/truetype/liberation/LiberationSans-Bold.ttf", f"/System/Library/Fonts/Supplemental/Arial.ttf", f"C:/Windows/Fonts/arial.ttf", f"/usr/local/share/fonts/truetype/mycustomfonts/arial.ttf"]
        self.resolved_font_path_pil = next((p for p in font_paths if os.path.exists(p)), None)
        self.active_font_pil = ImageFont.load_default(); self.active_font_size_pil = self.DEFAULT_FONT_SIZE_PIL; self.active_moviepy_font_name = self.DEFAULT_MOVIEPY_FONT
        if self.resolved_font_path_pil:
            try: self.active_font_pil = ImageFont.truetype(self.resolved_font_path_pil, self.PREFERRED_FONT_SIZE_PIL); self.active_font_size_pil = self.PREFERRED_FONT_SIZE_PIL; logger.info(f"Pillow font: {self.resolved_font_path_pil} sz {self.active_font_size_pil}."); self.active_moviepy_font_name = 'DejaVu-Sans-Bold' if "dejavu" in self.resolved_font_path_pil.lower() else ('Liberation-Sans-Bold' if "liberation" in self.resolved_font_path_pil.lower() else self.DEFAULT_MOVIEPY_FONT)
            except IOError as e_font: logger.error(f"Pillow font IOError '{self.resolved_font_path_pil}': {e_font}. Default.")
        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
        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
        self.runway_api_key = None; self.USE_RUNWAYML = False; self.runway_ml_sdk_client_instance = None
        if RUNWAYML_SDK_IMPORTED and RunwayMLAPIClientClass and os.getenv("RUNWAYML_API_SECRET"):
            try: self.runway_ml_sdk_client_instance = RunwayMLAPIClientClass(); self.USE_RUNWAYML = True; logger.info("RunwayML Client init from env var at startup.")
            except Exception as e_rwy_init: logger.error(f"Initial RunwayML client init failed: {e_rwy_init}"); self.USE_RUNWAYML = False
        logger.info("VisualEngine initialized.")

    def set_openai_api_key(self, k): self.openai_api_key=k; self.USE_AI_IMAGE_GENERATION=bool(k); logger.info(f"DALL-E: {'Ready' if self.USE_AI_IMAGE_GENERATION else 'Disabled'}")
    def set_elevenlabs_api_key(self, k, vid=None):
        self.elevenlabs_api_key=k;
        if vid: self.elevenlabs_voice_id = vid
        if k and ELEVENLABS_CLIENT_IMPORTED and ElevenLabsAPIClient:
            try: self.elevenlabs_client_instance = ElevenLabsAPIClient(api_key=k); self.USE_ELEVENLABS=True; logger.info(f"11L Client: Ready (Voice:{self.elevenlabs_voice_id})")
            except Exception as e: logger.error(f"11L client init err: {e}. 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, k): self.pexels_api_key=k; self.USE_PEXELS=bool(k); logger.info(f"Pexels: {'Ready' if self.USE_PEXELS else 'Disabled'}")
    def set_runway_api_key(self, k):
        self.runway_api_key = k
        if k:
            if RUNWAYML_SDK_IMPORTED and RunwayMLAPIClientClass:
                if not self.runway_ml_sdk_client_instance:
                    try:
                        orig_secret = os.getenv("RUNWAYML_API_SECRET")
                        if not orig_secret: os.environ["RUNWAYML_API_SECRET"]=k; logger.info("Temp set RUNWAYML_API_SECRET for SDK.")
                        self.runway_ml_sdk_client_instance=RunwayMLAPIClientClass(); self.USE_RUNWAYML=True; logger.info("RunwayML Client init via set_key.")
                        if not orig_secret: del os.environ["RUNWAYML_API_SECRET"]; logger.info("Cleared temp RUNWAYML_API_SECRET.")
                    except Exception as e: logger.error(f"RunwayML Client init in set_key fail: {e}", exc_info=True); self.USE_RUNWAYML=False;self.runway_ml_sdk_client_instance=None
                else: self.USE_RUNWAYML=True; logger.info("RunwayML Client already init.")
            else: logger.warning("RunwayML SDK not imported. Disabled."); self.USE_RUNWAYML=False
        else: self.USE_RUNWAYML=False; self.runway_ml_sdk_client_instance=None; logger.info("RunwayML Disabled (no key).")

    def _image_to_data_uri(self, img_path):
        try:
            mime, _ = mimetypes.guess_type(img_path)
            if not mime: ext=os.path.splitext(img_path)[1].lower(); mime_map={".png":"image/png",".jpg":"image/jpeg",".jpeg":"image/jpeg",".webp":"image/webp"}; mime=mime_map.get(ext,"application/octet-stream");
            if mime=="application/octet-stream": logger.warning(f"Unknown MIME for {img_path}, using {mime}.")
            with open(img_path,"rb") as f_img: enc_str=base64.b64encode(f_img.read()).decode('utf-8')
            uri=f"data:{mime};base64,{enc_str}"; logger.debug(f"Data URI for {os.path.basename(img_path)} (MIME:{mime}): {uri[:100]}..."); return uri
        except FileNotFoundError: logger.error(f"Img not found {img_path} for data URI."); return None
        except Exception as e: logger.error(f"Error converting {img_path} to data URI:{e}",exc_info=True); return None

    def _map_resolution_to_runway_ratio(self, w, h):
        r_str=f"{w}:{h}"; supp_r=["1280:720","720:1280","1104:832","832:1104","960:960","1584:672"];
        if r_str in supp_r: return r_str
        logger.warning(f"Res {r_str} not in Gen-4 list. Default 1280:720."); return "1280:720"

    def _get_text_dimensions(self, txt, font):
        dh=getattr(font,'size',self.active_font_size_pil);
        if not txt: return 0,dh
        try:
            if hasattr(font,'getbbox'):b=font.getbbox(txt);w=b[2]-b[0];h=b[3]-b[1];return w,h if h>0 else dh
            elif hasattr(font,'getsize'):w,h=font.getsize(txt);return w,h if h>0 else dh
            else: return int(len(txt)*dh*0.6),int(dh*1.2)
        except Exception as e:logger.warning(f"Err _get_text_dimensions:{e}");return int(len(txt)*self.active_font_size_pil*0.6),int(self.active_font_size_pil*1.2)

    def _create_placeholder_image_content(self, desc, fname, sz=None):
        # (Corrected and robust placeholder generation)
        if sz is None: sz=self.video_frame_size; img=Image.new('RGB',sz,color=(20,20,40));drw=ImageDraw.Draw(img);pad=25;maxw=sz[0]-(2*pad);lns=[]
        if not desc: desc="(Placeholder)"
        wds=desc.split();curr_ln=""
        for idx,w in enumerate(wds):
            prosp_add=w+(" "if idx<len(wds)-1 else"");test_ln=curr_ln+prosp_add
            curr_w,_=self._get_text_dimensions(test_ln,self.active_font_pil)
            if curr_w==0 and test_ln.strip():curr_w=len(test_ln)*(self.active_font_size_pil*0.6)
            if curr_w<=maxw:curr_ln=test_ln
            else:
                if curr_ln.strip():lns.append(curr_ln.strip())
                curr_ln=prosp_add
        if curr_ln.strip():lns.append(curr_ln.strip())
        if not lns and desc:
            avg_cw,_=self._get_text_dimensions("W",self.active_font_pil);avg_cw=avg_cw or(self.active_font_size_pil*0.6)
            cpl=int(maxw/avg_cw)if avg_cw>0 else 20;lns.append(desc[:cpl]+("..."if len(desc)>cpl else""))
        elif not lns:lns.append("(PH Error)")
        _,slh=self._get_text_dimensions("Ay",self.active_font_pil);slh=slh if slh>0 else self.active_font_size_pil+2
        maxl=min(len(lns),(sz[1]-(2*pad))//(slh+2))if slh>0 else 1;maxl=max(1,maxl)
        yp=pad+(sz[1]-(2*pad)-maxl*(slh+2))/2.0
        for i in range(maxl):
            lt=lns[i];lw,_=self._get_text_dimensions(lt,self.active_font_pil)
            if lw==0 and lt.strip():lw=len(lt)*(self.active_font_size_pil*0.6)
            xp=(sz[0]-lw)/2.0
            try:drw.text((xp,yp),lt,font=self.active_font_pil,fill=(200,200,180))
            except Exception as e:logger.error(f"Pillow d.text err:{e} for '{lt}'")
            yp+=slh+2
            if i==6 and maxl>7:
                try:drw.text((xp,yp),"...",font=self.active_font_pil,fill=(200,200,180))
                except Exception as e:logger.error(f"Pillow ellipsis err:{e}");break
        fpath=os.path.join(self.output_dir,fname)
        try:img.save(fpath);return fpath
        except Exception as e:logger.error(f"Save PH img '{fpath}' err:{e}",exc_info=True);return None

    def _search_pexels_image(self, q_str, out_fn_base):
        # (Corrected from before)
        if not self.USE_PEXELS or not self.pexels_api_key: return None
        h={"Authorization":self.pexels_api_key};p={"query":q_str,"per_page":1,"orientation":"landscape","size":"large2x"}
        base_n_px,_=os.path.splitext(out_fn_base);px_fn=base_n_px+f"_pexels_{random.randint(1000,9999)}.jpg";fp_px=os.path.join(self.output_dir,px_fn)
        try:
            logger.info(f"Pexels: Search '{q_str}'");eff_q=" ".join(q_str.split()[:5]);p["query"]=eff_q
            resp_px=requests.get("https://api.pexels.com/v1/search",headers=h,params=p,timeout=20);resp_px.raise_for_status();data_px=resp_px.json()
            if data_px.get("photos") and len(data_px["photos"]) > 0:
                ph_det=data_px["photos"][0];ph_url=ph_det.get("src",{}).get("large2x")
                if not ph_url:logger.warning(f"Pexels: 'large2x' URL missing for '{eff_q}'.");return None
                img_resp=requests.get(ph_url,timeout=60);img_resp.raise_for_status();img_pil=Image.open(io.BytesIO(img_resp.content))
                if img_pil.mode!='RGB':img_pil=img_pil.convert('RGB')
                img_pil.save(fp_px);logger.info(f"Pexels: Saved to {fp_px}");return fp_px
            else:logger.info(f"Pexels: No photos for '{eff_q}'.");return None
        except requests.exceptions.RequestException as e:logger.error(f"Pexels ReqExc '{q_str}':{e}",exc_info=False);return None
        except Exception as e:logger.error(f"Pexels GenErr '{q_str}':{e}",exc_info=True);return None

    def _generate_video_clip_with_runwayml(self, motion_prompt, input_img_path, scene_id_base_fn, duration_s=5):
        # (Updated RunwayML integration logic)
        if not self.USE_RUNWAYML or not self.runway_ml_sdk_client_instance: logger.warning("RunwayML skip: Not enabled/client not init."); return None
        if not input_img_path or not os.path.exists(input_img_path): logger.error(f"Runway Gen-4 needs input img. Invalid: {input_img_path}"); return None
        img_data_uri = self._image_to_data_uri(input_img_path)
        if not img_data_uri: return None
        rwy_dur = 10 if duration_s >= 8 else 5; rwy_ratio = self._map_resolution_to_runway_ratio(self.video_frame_size[0],self.video_frame_size[1])
        rwy_base_name,_=os.path.splitext(scene_id_base_fn);rwy_out_fn=rwy_base_name+f"_runway_gen4_d{rwy_dur}s.mp4";rwy_out_fp=os.path.join(self.output_dir,rwy_out_fn)
        logger.info(f"Runway Gen-4 task: motion='{motion_prompt[:70]}...', img='{os.path.basename(input_img_path)}', dur={rwy_dur}s, ratio='{rwy_ratio}'")
        try:
            rwy_task_sub = self.runway_ml_sdk_client_instance.image_to_video.create(model='gen4_turbo',prompt_image=img_data_uri,prompt_text=motion_prompt,duration=rwy_dur,ratio=rwy_ratio)
            rwy_task_id = rwy_task_sub.id; logger.info(f"Runway task ID: {rwy_task_id}. Polling...")
            poll_s=10;max_p_count=36;poll_t_start=time.time()
            while time.time()-poll_t_start < max_p_count*poll_s:
                time.sleep(poll_s);rwy_task_det=self.runway_ml_sdk_client_instance.tasks.retrieve(id=rwy_task_id)
                logger.info(f"Runway task {rwy_task_id} status: {rwy_task_det.status}")
                if rwy_task_det.status=='SUCCEEDED':
                    rwy_out_url=getattr(getattr(rwy_task_det,'output',None),'url',None) or (getattr(rwy_task_det,'artifacts',None)and rwy_task_det.artifacts and hasattr(rwy_task_det.artifacts[0],'url')and rwy_task_det.artifacts[0].url) or (getattr(rwy_task_det,'artifacts',None)and rwy_task_det.artifacts and hasattr(rwy_task_det.artifacts[0],'download_url')and rwy_task_det.artifacts[0].download_url)
                    if not rwy_out_url:logger.error(f"Runway task {rwy_task_id} SUCCEEDED, no output URL. Details:{vars(rwy_task_det)if hasattr(rwy_task_det,'__dict__')else rwy_task_det}");return None
                    logger.info(f"Runway task {rwy_task_id} SUCCEEDED. Downloading: {rwy_out_url}")
                    vid_resp=requests.get(rwy_out_url,stream=True,timeout=300);vid_resp.raise_for_status()
                    with open(rwy_out_fp,'wb')as f:
                        for chk in vid_resp.iter_content(chunk_size=8192):f.write(chk)
                    logger.info(f"Runway Gen-4 video saved: {rwy_out_fp}");return rwy_out_fp
                elif rwy_task_det.status in['FAILED','ABORTED','ERROR']:
                    rwy_err_msg=getattr(rwy_task_det,'error_message',None)or getattr(getattr(rwy_task_det,'output',None),'error',"Unknown Runway error.")
                    logger.error(f"Runway task {rwy_task_id} status:{rwy_task_det.status}. Error:{rwy_err_msg}");return None
            logger.warning(f"Runway task {rwy_task_id} timed out.");return None
        except AttributeError as e:logger.error(f"RunwayML SDK AttrError:{e}. SDK methods changed?",exc_info=True);return None
        except Exception as e:logger.error(f"Runway Gen-4 API error:{e}",exc_info=True);return None

    def _create_placeholder_video_content(self, text_desc, fname, duration=4, size=None):
        # (Corrected from previous)
        if size is None: size = self.video_frame_size
        fpath_ph_vid = os.path.join(self.output_dir, fname) # Renamed
        txt_clip_ph_vid = None # Renamed
        try:
            txt_clip_ph_vid = TextClip(text_desc, fontsize=50, color='white', font=self.video_overlay_font,
                                bg_color='black', size=size, method='caption').set_duration(duration)
            txt_clip_ph_vid.write_videofile(fpath_ph_vid, fps=24, codec='libx264', preset='ultrafast', logger=None, threads=2)
            logger.info(f"Generic placeholder video created: {fpath_ph_vid}")
            return fpath_ph_vid
        except Exception as e_phv: # Renamed
            logger.error(f"Failed to create generic placeholder video '{fpath_ph_vid}': {e_phv}", exc_info=True)
            return None
        finally:
            if txt_clip_ph_vid and hasattr(txt_clip_ph_vid, 'close'):
                try: txt_clip_ph_vid.close()
                except Exception as e_cl_phv: logger.warning(f"Ignoring error closing placeholder TextClip: {e_cl_phv}")
    
    def generate_scene_asset(self, img_prompt, motion_prompt, scene_dict, scene_id_fn_base, gen_as_vid=False, rwy_dur=5):
        # (Corrected DALL-E loop from previous, renamed variables for clarity)
        asset_base_name,_=os.path.splitext(scene_id_fn_base); asset_info_obj={'path':None,'type':'none','error':True,'prompt_used':img_prompt,'error_message':'Asset gen init failed'}; base_img_path_for_rwy=None
        base_img_fn = asset_base_name + ("_base_for_video.png" if gen_as_vid else ".png"); base_img_fp = os.path.join(self.output_dir, base_img_fn)
        if self.USE_AI_IMAGE_GENERATION and self.openai_api_key:
            max_r,att_c=2,0
            for att_idx in range(max_r): # Renamed att_n_dalle to att_idx
                att_c=att_idx+1
                try:
                    logger.info(f"Att {att_c} DALL-E (base img): {img_prompt[:70]}...");oai_client=openai.OpenAI(api_key=self.openai_api_key,timeout=90.0);oai_resp=oai_client.images.generate(model=self.dalle_model,prompt=img_prompt,n=1,size=self.image_size_dalle3,quality="hd",response_format="url",style="vivid");oai_url=oai_resp.data[0].url;oai_rev_p=getattr(oai_resp.data[0],'revised_prompt',None)
                    if oai_rev_p:logger.info(f"DALL-E revised: {oai_rev_p[:70]}...")
                    oai_img_get_resp=requests.get(oai_url,timeout=120);oai_img_get_resp.raise_for_status();oai_pil_img=Image.open(io.BytesIO(oai_img_get_resp.content))
                    if oai_pil_img.mode!='RGB':oai_pil_img=oai_pil_img.convert('RGB')
                    oai_pil_img.save(base_img_fp);logger.info(f"DALL-E base img saved: {base_img_fp}");base_img_path_for_rwy=base_img_fp;asset_info_obj={'path':base_img_fp,'type':'image','error':False,'prompt_used':img_prompt,'revised_prompt':oai_rev_p};break
                except openai.RateLimitError as e:logger.warning(f"OpenAI RateLimit Att {att_c}:{e}.Retry...");time.sleep(5*att_c);asset_info_obj['error_message']=str(e)
                except openai.APIError as e:logger.error(f"OpenAI APIError Att {att_c}:{e}");asset_info_obj['error_message']=str(e);break
                except requests.exceptions.RequestException as e:logger.error(f"Requests Err DALL-E Att {att_c}:{e}");asset_info_obj['error_message']=str(e);break
                except Exception as e:logger.error(f"General DALL-E Err Att {att_c}:{e}",exc_info=True);asset_info_obj['error_message']=str(e);break
            if asset_info_obj['error']:logger.warning(f"DALL-E failed after {att_c} attempts for base img.")
        if asset_info_obj['error'] and self.USE_PEXELS:
            logger.info("Trying Pexels for base img.");px_q=scene_dict.get('pexels_search_query_감독',f"{scene_dict.get('emotional_beat','')} {scene_dict.get('setting_description','')}");px_p=self._search_pexels_image(px_q,base_img_fn)
            if px_p:base_img_path_for_rwy=px_p;asset_info_obj={'path':px_p,'type':'image','error':False,'prompt_used':f"Pexels:{px_q}"}
            else:curr_err=asset_info_obj.get('error_message',"");asset_info_obj['error_message']=(curr_err+" Pexels failed for base.").strip()
        if asset_info_obj['error']:
            logger.warning("Base img (DALL-E/Pexels) failed. Using placeholder.");ph_p_txt=asset_info_obj.get('prompt_used',img_prompt);ph_img_p=self._create_placeholder_image_content(f"[Base Placeholder]{ph_p_txt[:70]}...",base_img_fn)
            if ph_img_p:base_img_path_for_rwy=ph_img_p;asset_info_obj={'path':ph_img_p,'type':'image','error':False,'prompt_used':ph_p_txt}
            else:curr_err=asset_info_obj.get('error_message',"");asset_info_obj['error_message']=(curr_err+" Base placeholder failed.").strip()
        if gen_as_vid:
            if not base_img_path_for_rwy:logger.error("RunwayML video: base img failed.");asset_info_obj['error']=True;asset_info_obj['error_message']=(asset_info_obj.get('error_message',"")+" Base img miss, Runway abort.").strip();asset_info_obj['type']='none';return asset_info_obj
            if self.USE_RUNWAYML:
                rwy_vid_p=self._generate_video_clip_with_runwayml(motion_prompt,base_img_path_for_rwy,asset_base_name,rwy_dur)
                if rwy_vid_p and os.path.exists(rwy_vid_p):asset_info_obj={'path':rwy_vid_p,'type':'video','error':False,'prompt_used':motion_prompt,'base_image_path':base_img_path_for_rwy}
                else:logger.warning(f"RunwayML video failed for {asset_base_name}. Fallback to base img.");asset_info_obj['error']=True;asset_info_obj['error_message']=(asset_info_obj.get('error_message',"Base img ok.")+" RunwayML video fail; use base img.").strip();asset_info_obj['path']=base_img_path_for_rwy;asset_info_obj['type']='image';asset_info_obj['prompt_used']=img_prompt
            else:logger.warning("RunwayML selected but disabled. Use base img.");asset_info_obj['error']=True;asset_info_obj['error_message']=(asset_info_obj.get('error_message',"Base img ok.")+" RunwayML disabled; use base img.").strip();asset_info_obj['path']=base_img_path_for_rwy;asset_info_obj['type']='image';asset_info_obj['prompt_used']=img_prompt
        return asset_info_obj

    def generate_narration_audio(self, narration_text, output_fn="narration_overall.mp3"):
        # (Corrected version from previous response)
        if not self.USE_ELEVENLABS or not self.elevenlabs_client_instance or not narration_text: logger.info("11L conditions not met. Skip audio."); return None
        narration_fp = os.path.join(self.output_dir, output_fn)
        try:
            logger.info(f"11L audio (Voice:{self.elevenlabs_voice_id}): \"{narration_text[:70]}...\"")
            stream_method = None
            if hasattr(self.elevenlabs_client_instance,'text_to_speech') and hasattr(self.elevenlabs_client_instance.text_to_speech,'stream'): stream_method=self.elevenlabs_client_instance.text_to_speech.stream; logger.info("Using 11L .text_to_speech.stream()")
            elif hasattr(self.elevenlabs_client_instance,'generate_stream'): stream_method=self.elevenlabs_client_instance.generate_stream; logger.info("Using 11L .generate_stream()")
            elif hasattr(self.elevenlabs_client_instance,'generate'):
                logger.info("Using 11L .generate() (non-streaming).")
                voice_p = Voice(voice_id=str(self.elevenlabs_voice_id),settings=self.elevenlabs_voice_settings_obj) if Voice and self.elevenlabs_voice_settings_obj else str(self.elevenlabs_voice_id)
                audio_b = self.elevenlabs_client_instance.generate(text=narration_text,voice=voice_p,model="eleven_multilingual_v2")
                with open(narration_fp,"wb") as f_audio: f_audio.write(audio_b); logger.info(f"11L audio (non-stream): {narration_fp}"); return narration_fp
            else: logger.error("No recognized 11L audio method."); return None
            if stream_method:
                voice_stream_params={"voice_id":str(self.elevenlabs_voice_id)}
                if self.elevenlabs_voice_settings_obj:
                    if hasattr(self.elevenlabs_voice_settings_obj,'model_dump'): voice_stream_params["voice_settings"]=self.elevenlabs_voice_settings_obj.model_dump()
                    elif hasattr(self.elevenlabs_voice_settings_obj,'dict'): voice_stream_params["voice_settings"]=self.elevenlabs_voice_settings_obj.dict()
                    else: voice_stream_params["voice_settings"]=self.elevenlabs_voice_settings_obj
                audio_iter = stream_method(text=narration_text,model_id="eleven_multilingual_v2",**voice_stream_params)
                with open(narration_fp,"wb") as f_audio_stream:
                    for chunk_item in audio_iter:
                        if chunk_item: f_audio_stream.write(chunk_item)
                logger.info(f"11L audio (stream): {narration_fp}"); return narration_fp
        except AttributeError as e_11l_attr: logger.error(f"11L SDK AttrError: {e_11l_attr}. SDK/methods changed?", exc_info=True); return None
        except Exception as e_11l_gen: logger.error(f"11L audio gen error: {e_11l_gen}", exc_info=True); return None

    def assemble_animatic_from_assets(self, asset_data_list, overall_narration_path=None, output_filename="final_video.mp4", fps=24):
        # (Keep the version with robust image processing, C-contiguous array, debug saves, and pix_fmt)
        # ... (This extensive method from the previous full rewrite is assumed to be mostly correct for the "blank video" debugging)
        if not asset_data_list: logger.warning("No assets for animatic."); return None
        processed_clips_list = []; narration_audio_clip_mvpy = None; final_video_output_clip = None
        logger.info(f"Assembling from {len(asset_data_list)} assets. Target Frame: {self.video_frame_size}.")
        for i_asset, asset_info_item_loop in enumerate(asset_data_list):
            path_of_asset, type_of_asset, duration_for_scene = asset_info_item_loop.get('path'), asset_info_item_loop.get('type'), asset_info_item_loop.get('duration', 4.5)
            num_of_scene, action_in_key = asset_info_item_loop.get('scene_num', i_asset + 1), asset_info_item_loop.get('key_action', '')
            logger.info(f"S{num_of_scene}: Path='{path_of_asset}', Type='{type_of_asset}', Dur='{duration_for_scene}'s")
            if not (path_of_asset and os.path.exists(path_of_asset)): logger.warning(f"S{num_of_scene}: Not found '{path_of_asset}'. Skip."); continue
            if duration_for_scene <= 0: logger.warning(f"S{num_of_scene}: Invalid duration ({duration_for_scene}s). Skip."); continue
            active_scene_clip = None
            try:
                if type_of_asset == 'image':
                    opened_pil_img = Image.open(path_of_asset); logger.debug(f"S{num_of_scene}: Loaded img. Mode:{opened_pil_img.mode}, Size:{opened_pil_img.size}")
                    converted_img_rgba = opened_pil_img.convert('RGBA') if opened_pil_img.mode != 'RGBA' else opened_pil_img.copy()
                    thumbnailed_img = converted_img_rgba.copy(); resample_f = Image.Resampling.LANCZOS if hasattr(Image.Resampling,'LANCZOS') else Image.BILINEAR; thumbnailed_img.thumbnail(self.video_frame_size,resample_f)
                    rgba_canvas = Image.new('RGBA',self.video_frame_size,(0,0,0,0)); pos_x,pos_y=(self.video_frame_size[0]-thumbnailed_img.width)//2,(self.video_frame_size[1]-thumbnailed_img.height)//2
                    rgba_canvas.paste(thumbnailed_img,(pos_x,pos_y),thumbnailed_img)
                    final_rgb_img_pil = Image.new("RGB",self.video_frame_size,(0,0,0)); final_rgb_img_pil.paste(rgba_canvas,mask=rgba_canvas.split()[3])
                    debug_path_img_pre_numpy = os.path.join(self.output_dir,f"debug_PRE_NUMPY_S{num_of_scene}.png"); final_rgb_img_pil.save(debug_path_img_pre_numpy); logger.info(f"DEBUG: Saved PRE_NUMPY_S{num_of_scene} to {debug_path_img_pre_numpy}")
                    numpy_frame_arr = np.array(final_rgb_img_pil,dtype=np.uint8);
                    if not numpy_frame_arr.flags['C_CONTIGUOUS']: numpy_frame_arr=np.ascontiguousarray(numpy_frame_arr,dtype=np.uint8)
                    logger.debug(f"S{num_of_scene}: NumPy for MoviePy. Shape:{numpy_frame_arr.shape}, DType:{numpy_frame_arr.dtype}, C-Contig:{numpy_frame_arr.flags['C_CONTIGUOUS']}")
                    if numpy_frame_arr.size==0 or numpy_frame_arr.ndim!=3 or numpy_frame_arr.shape[2]!=3: logger.error(f"S{num_of_scene}: Invalid NumPy array for MoviePy. Skip."); continue
                    base_image_clip = ImageClip(numpy_frame_arr,transparent=False).set_duration(duration_for_scene)
                    debug_path_moviepy_frame=os.path.join(self.output_dir,f"debug_MOVIEPY_FRAME_S{num_of_scene}.png"); base_image_clip.save_frame(debug_path_moviepy_frame,t=0.1); logger.info(f"DEBUG: Saved MOVIEPY_FRAME_S{num_of_scene} to {debug_path_moviepy_frame}")
                    fx_image_clip = base_image_clip
                    try: scale_end_kb=random.uniform(1.03,1.08); fx_image_clip=base_image_clip.fx(vfx.resize,lambda t_val:1+(scale_end_kb-1)*(t_val/duration_for_scene) if duration_for_scene>0 else 1).set_position('center')
                    except Exception as e_kb_fx: logger.error(f"S{num_of_scene} Ken Burns error: {e_kb_fx}",exc_info=False)
                    active_scene_clip = fx_image_clip
                elif type_of_asset == 'video':
                    source_video_clip_obj=None
                    try:
                        source_video_clip_obj=VideoFileClip(path_of_asset,target_resolution=(self.video_frame_size[1],self.video_frame_size[0])if self.video_frame_size else None, audio=False)
                        temp_video_clip_obj_loop=source_video_clip_obj
                        if source_video_clip_obj.duration!=duration_for_scene:
                            if source_video_clip_obj.duration>duration_for_scene:temp_video_clip_obj_loop=source_video_clip_obj.subclip(0,duration_for_scene)
                            else:
                                if duration_for_scene/source_video_clip_obj.duration > 1.5 and source_video_clip_obj.duration>0.1:temp_video_clip_obj_loop=source_video_clip_obj.loop(duration=duration_for_scene)
                                else:temp_video_clip_obj_loop=source_video_clip_obj.set_duration(source_video_clip_obj.duration);logger.info(f"S{num_of_scene} Video clip ({source_video_clip_obj.duration:.2f}s) shorter than target ({duration_for_scene:.2f}s).")
                        active_scene_clip=temp_video_clip_obj_loop.set_duration(duration_for_scene)
                        if active_scene_clip.size!=list(self.video_frame_size):active_scene_clip=active_scene_clip.resize(self.video_frame_size)
                    except Exception as e_vid_load_loop:logger.error(f"S{num_of_scene} Video load error '{path_of_asset}':{e_vid_load_loop}",exc_info=True);continue
                    finally:
                        if source_video_clip_obj and source_video_clip_obj is not active_scene_clip and hasattr(source_video_clip_obj,'close'):source_video_clip_obj.close()
                else: logger.warning(f"S{num_of_scene} Unknown asset type '{type_of_asset}'. Skip."); continue
                if active_scene_clip and action_in_key:
                    try:
                        dur_text_overlay=min(active_scene_clip.duration-0.5,active_scene_clip.duration*0.8)if active_scene_clip.duration>0.5 else active_scene_clip.duration; start_text_overlay=0.25
                        if dur_text_overlay > 0:
                            text_clip_for_overlay=TextClip(f"Scene {num_of_scene}\n{action_in_key}",fontsize=self.VIDEO_OVERLAY_FONT_SIZE,color=self.VIDEO_OVERLAY_FONT_COLOR,font=self.active_moviepy_font_name,bg_color='rgba(10,10,20,0.7)',method='caption',align='West',size=(self.video_frame_size[0]*0.9,None),kerning=-1,stroke_color='black',stroke_width=1.5).set_duration(dur_text_overlay).set_start(start_text_overlay).set_position(('center',0.92),relative=True)
                            active_scene_clip=CompositeVideoClip([active_scene_clip,text_clip_for_overlay],size=self.video_frame_size,use_bgclip=True)
                        else: logger.warning(f"S{num_of_scene}: Text overlay duration zero. Skip text.")
                    except Exception as e_txt_comp:logger.error(f"S{num_of_scene} TextClip error:{e_txt_comp}. No text.",exc_info=True)
                if active_scene_clip:processed_moviepy_clips_list.append(active_scene_clip);logger.info(f"S{num_of_scene} Processed. Dur:{active_scene_clip.duration:.2f}s.")
            except Exception as e_asset_loop_main:logger.error(f"MAJOR Error processing asset for S{num_of_scene} ({path_of_asset}):{e_asset_loop_main}",exc_info=True)
            finally:
                if active_scene_clip and hasattr(active_scene_clip,'close'):
                    try: active_scene_clip.close()
                    except: pass
        if not processed_moviepy_clips_list:logger.warning("No clips processed for animatic. Aborting.");return None
        transition_duration_val=0.75
        try:
            logger.info(f"Concatenating {len(processed_moviepy_clips_list)} clips for final animatic.");
            if len(processed_moviepy_clips_list)>1:final_video_output_clip=concatenate_videoclips(processed_moviepy_clips_list,padding=-transition_duration_val if transition_duration_val>0 else 0,method="compose")
            elif processed_moviepy_clips_list:final_video_output_clip=processed_moviepy_clips_list[0]
            if not final_video_output_clip:logger.error("Concatenation resulted in a None clip. Aborting.");return None
            logger.info(f"Concatenated animatic duration:{final_video_output_clip.duration:.2f}s")
            if transition_duration_val>0 and final_video_output_clip.duration>0:
                if final_video_output_clip.duration>transition_duration_val*2:final_video_output_clip=final_video_output_clip.fx(vfx.fadein,transition_duration_val).fx(vfx.fadeout,transition_duration_val)
                else:final_video_output_clip=final_video_output_clip.fx(vfx.fadein,min(transition_duration_val,final_video_output_clip.duration/2.0))
            if overall_narration_path and os.path.exists(overall_narration_path) and final_video_output_clip.duration>0:
                try:narration_audio_clip_mvpy=AudioFileClip(overall_narration_path);final_video_output_clip=final_video_output_clip.set_audio(narration_audio_clip_mvpy);logger.info("Overall narration added to animatic.")
                except Exception as e_narr_add:logger.error(f"Error adding narration to animatic:{e_narr_add}",exc_info=True)
            elif final_video_output_clip.duration<=0:logger.warning("Animatic has no duration. Audio not added.")
            if final_video_output_clip and final_video_output_clip.duration>0:
                final_output_path_str=os.path.join(self.output_dir,output_filename);logger.info(f"Writing final animatic video to:{final_output_path_str} (Duration:{final_video_output_clip.duration:.2f}s)")
                final_video_output_clip.write_videofile(final_output_path_str,fps=fps,codec='libx264',preset='medium',audio_codec='aac',temp_audiofile=os.path.join(self.output_dir,f'temp-audio-{os.urandom(4).hex()}.m4a'),remove_temp=True,threads=os.cpu_count()or 2,logger='bar',bitrate="5000k",ffmpeg_params=["-pix_fmt", "yuv420p"])
                logger.info(f"Animatic video created successfully:{final_output_path_str}");return final_output_path_str
            else:logger.error("Final animatic clip is invalid or has zero duration. Cannot write video file.");return None
        except Exception as e_vid_write_final:logger.error(f"Error during final animatic video file writing or composition:{e_vid_write_final}",exc_info=True);return None
        finally:
            logger.debug("Closing all MoviePy clips in `assemble_animatic_from_assets` main finally block.")
            all_clips_to_close_list = processed_moviepy_clips_list + ([narration_audio_clip_mvpy] if narration_audio_clip_mvpy else []) + ([final_video_output_clip] if final_video_output_clip else [])
            for clip_item_to_close in all_clips_to_close_list:
                if clip_item_to_close and hasattr(clip_item_to_close, 'close'):
                    try: clip_item_to_close.close()
                    except Exception as e_final_close: logger.warning(f"Ignoring error while closing a MoviePy clip: {type(clip_item_to_close).__name__} - {e_final_close}")