import os
import asyncio
from model.trainer import train_on_message
from model.scheduler import set_next_action
from reader.filter import is_valid_line
import json

BOOK_DIR = "data/books"
PROGRESS_FILE = "data/memory/book_progress.json"
READ_DELAY = 10  # seconds between lines


def get_books():
    return [f for f in os.listdir(BOOK_DIR) if f.endswith(".txt")]


def load_progress():
    if os.path.exists(PROGRESS_FILE):
        with open(PROGRESS_FILE, "r", encoding="utf-8") as f:
            return json.load(f)
    return {}


def save_progress(prog):
    with open(PROGRESS_FILE, "w", encoding="utf-8") as f:
        json.dump(prog, f, indent=2)


async def read_books_forever():
    books = get_books()
    progress = load_progress()
    buffered_lines = []

    while True:
        for book in books:
            path = os.path.join(BOOK_DIR, book)
            if not os.path.exists(path):
                continue

            with open(path, "r", encoding="utf-8") as f:
                lines = f.readlines()

            idx = progress.get(book, 0)
            while idx < len(lines):
                line = lines[idx].strip()
                idx += 1
                progress[book] = idx
                save_progress(progress)

                if is_valid_line(line):
                    buffered_lines.append(line)

                    # If we have enough lines buffered, combine and train
                    if len(buffered_lines) >= 3:
                        combined_text = " ".join(buffered_lines)
                        train_on_message(combined_text, source="book")
                        buffered_lines.clear()

                set_next_action(READ_DELAY, "Reading")
                await asyncio.sleep(READ_DELAY)

            # End of a book: train whatever lines are left buffered
            if buffered_lines:
                combined_text = " ".join(buffered_lines)
                train_on_message(combined_text, source="book")
                buffered_lines.clear()