video2.py

from moviepy.editor import *
from PIL import Image
import pytesseract
import numpy as np
from gtts import gTTS
from mutagen.mp3 import MP3
import uuid
import os
from pathlib import Path

# Use /app/data which we created with proper permissions
BASE_DIR = "/app/data"
IMAGE_DIR = "/tmp/images"
os.makedirs(IMAGE_DIR, exist_ok=True)
AUDIO_DIR = os.path.join(BASE_DIR, "sound")
CLIPS_DIR = os.path.join(BASE_DIR, "video")

# Create directories (no chmod needed)
for path in [BASE_DIR, AUDIO_DIR, CLIPS_DIR]:
    Path(path).mkdir(parents=True, exist_ok=True)
    

# Generate audio
def audio_func(id,lines):
  tts = gTTS(text=lines[id], lang='en', slow=False)
  audio_name = "audio"+str(id)+".mp3"
  audio_path=os.path.join(AUDIO_DIR,audio_name)
  tts.save(audio_path)
  if os.path.exists(audio_path):
      audio = MP3(audio_path)
      duration = audio.info.length
  return duration,audio_path
    
# --- CONFIGURATION ---
def video_func(id,lines):
  print(id,lines[id])
  duration,audio_path = audio_func(id,lines)
  IMAGE_PATH = os.path.join(IMAGE_DIR,f"slide{id}.png")   # Ensure this path is correct
  VIDEO_DURATION = duration               # seconds
  HIGHLIGHT_COLOR = (255, 255, 0)   # Yellow highlight
  HIGHLIGHT_OPACITY = 0.5           # Semi-transparent

  # --- OCR STEP ---
  img = Image.open(IMAGE_PATH)
  data = pytesseract.image_to_data(img, output_type=pytesseract.Output.DICT)

  # Extract words and their positions
  words = []
  for i in range(len(data['text'])):
      word = data['text'][i].strip()
      if word and int(data['conf'][i]) > 60:
          x, y, w, h = data['left'][i], data['top'][i], data['width'][i], data['height'][i]
          words.append({'text': word, 'box': (x, y, w, h)})

  # --- BASE IMAGE CLIP ---
  image_clip = ImageClip(IMAGE_PATH).set_duration(VIDEO_DURATION)

  # --- HIGHLIGHT WORDS ONE BY ONE ---
  n_words = len(words)
  highlight_duration = VIDEO_DURATION / n_words

  highlight_clips = []

  for i, word in enumerate(words):
      x, y, w, h = word['box']
      start = i * highlight_duration
      end = start + highlight_duration

      # Create highlight rectangle
      rect = ColorClip(size=(w, h), color=HIGHLIGHT_COLOR)
      rect = rect.set_opacity(HIGHLIGHT_OPACITY).set_position((x, y)).set_start(start).set_end(end)

      highlight_clips.append(rect)

  # --- FINAL VIDEO --

  final_clip = CompositeVideoClip([image_clip] + highlight_clips)
  audio = AudioFileClip(audio_path)
  final_clip = final_clip.set_audio(audio)
  clip_name = "clip"+str(id)+".mp4"
  video_path=os.path.join(CLIPS_DIR,clip_name)
  final_clip.write_videofile(video_path, fps=24,audio=True)

def video_com(lines):
    video_path = f"/tmp/video_{uuid.uuid4().hex}.mp4"
    for id in range(len(lines)):
        video_func(id,lines)
        print(id,lines[id])
    clips = []
    for id in range(len(lines)):
        clip = VideoFileClip(f"/app/data/video/clip{id}.mp4")
        clips.append(clip)
        
    final_video = concatenate_videoclips(clips,method="compose")
    final_video.write_videofile(video_path, fps=24)
    return video_path