app.py

import os
import tempfile
import traceback
from flask import Flask, request, jsonify, send_file
from PIL import Image
from moviepy.editor import *
from gtts import gTTS
from mutagen.mp3 import MP3
import pytesseract
import numpy as np

app = Flask(__name__)

def video_func(id, lines):
    try:
        # Temp files
        temp_audio = tempfile.NamedTemporaryFile(delete=False, suffix='.mp3')
        temp_audio.close()

        # Text-to-Speech
        tts = gTTS(text=lines[id], lang='ta', slow=False)
        tts.save(temp_audio.name)

        audio = MP3(temp_audio.name)
        duration = audio.info.length

        image_path = f"/tmp/images/slide{id}.png"
        video_duration = duration
        highlight_color = (255, 255, 0)
        highlight_opacity = 0.5

        # OCR: Extract words
        img = Image.open(image_path)
        data = pytesseract.image_to_data(img, output_type=pytesseract.Output.DICT)

        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)})

        image_clip = ImageClip(image_path).set_duration(video_duration)

        # Highlight effect
        n_words = len(words)
        highlight_duration = video_duration / max(n_words, 1)
        highlight_clips = []

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

            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_clip = CompositeVideoClip([image_clip] + highlight_clips)
        audio = AudioFileClip(temp_audio.name)
        final_clip = final_clip.set_audio(audio)
        final_clip.write_videofile(f"/tmp/clip{id}.mp4", fps=24)

    except Exception as e:
        print(f"Error in video_func for ID {id}: {e}")
        traceback.print_exc()


@app.route("/generate", methods=["POST"])
def generate_video():
    try:
        data = request.get_json()
        prompt = data.get("duration", '').strip()
        prompts = prompt.replace("**", "")

        if not prompts:
            return jsonify({"error": "prompts must not be empty"}), 400

        raw_lines = prompts.splitlines()
        lines = []
        i = 0

        while i < len(raw_lines):
            line = raw_lines[i].strip()

            if line.startswith("#") and (line.endswith('?') or line.endswith(':')):
                block = line
                i += 1
                paragraph_lines = []

                while i < len(raw_lines):
                    next_line = raw_lines[i].strip()
                    if next_line.startswith("#") and (next_line.endswith('?') or next_line.endswith(':')):
                        break
                    paragraph_lines.append(next_line)
                    i += 1
                    if len(paragraph_lines) >= 5:
                        break

                if paragraph_lines:
                    block += '\n' + '\n'.join(paragraph_lines)
                lines.append(block)
            else:
                block_lines = []
                count = 0
                while i < len(raw_lines) and count < 5:
                    next_line = raw_lines[i].strip()
                    if next_line.startswith("#") and (next_line.endswith('?') or next_line.endswith(':')):
                        break
                    block_lines.append(next_line)
                    i += 1
                    count += 1
                if block_lines:
                    lines.append('\n'.join(block_lines))

        # Slide image creation
        image_olst = []
        for id in range(len(lines)):
            create_text_image(lines[id], id, image_olst)

        for i in range(len(lines)):
            video_func(i, lines)

        clips = [VideoFileClip(f"/tmp/clip{id}.mp4") for id in range(len(lines))]
        final_video = concatenate_videoclips(clips)
        output_path = "/tmp/final_output.mp4"
        final_video.write_videofile(output_path, fps=24)

        # Clean up
        for img in image_olst:
            os.remove(img)

        return send_file(output_path, mimetype='video/mp4')

    except Exception as e:
        traceback.print_exc()
        return jsonify({"error": str(e)}), 500


@app.route("/")
def home():
    return "Flask Video Generator is Running"


if __name__ == "__main__":
    app.run(host="0.0.0.0", port=7860)