app.py

from asyncio import create_subprocess_shell, create_task, gather, run, sleep
from json import dumps
from logging import ERROR, INFO, basicConfig, getLogger
from pathlib import Path
from shutil import rmtree
from subprocess import CalledProcessError, PIPE
from typing import Any, List
from uuid import uuid4

from PIL import Image
from aiorentry.client import Client as RentryClient
from fastapi import FastAPI, HTTPException
from fastapi.responses import PlainTextResponse
from httpx import AsyncClient, HTTPStatusError, RequestError
from lxml import html
from markdown import markdown
from pydantic import BaseModel, HttpUrl
from uvicorn import run as uvicorn_run

proxy_endpoint = 'https://telegraphprxver.vercel.app'

need_logging = False

basicConfig(level=INFO if need_logging else ERROR)
logger = getLogger(__name__)

oxipng_bin = Path(__file__).parent / 'oxipng'
if not oxipng_bin.stat().st_mode & 0o111:
    oxipng_bin.chmod(0o755)

tokens = [
    # мне в общем-то все равно на эти токены
    '7e0ea3da6a73d77003c1abba7f0ea13c',
    'bc2e68b5918e5bb59ebca6c05d73daf9',
    'fecbfbe0938bcd1df27b7a9be1702cc9',
    '04e9981d4d0981964cb4c9753173244d',
    'dee75b07981c7aa211628ea7c7cbc03d',
]


class UserInfo:
    def __init__(self, token: str, short_name: str, author_name: str, author_url: str, auth_link: str):
        self.token = token
        self.short_name = short_name
        self.author_name = author_name
        self.author_url = author_url
        self.auth_link = auth_link

    def get_token(self) -> str:
        return self.token

    def get_short_name(self) -> str:
        return self.short_name

    def get_author_name(self) -> str:
        return self.author_name

    def get_author_url(self) -> str:
        return self.author_url

    def get_auth_link(self) -> str:
        return self.auth_link

    def __repr__(self):
        return f'UserAuthInfo(token={self.token}, short_name={self.short_name}, author_name={self.author_name}, author_url={self.author_url}, auth_link={self.auth_link})'


async def create_account(short_name: str, author_name: str, author_url: str) -> UserInfo:
    params = {
        'short_name': short_name,
        'author_name': author_name,
        'author_url': author_url
    }
    async with AsyncClient(verify=False, follow_redirects=True, timeout=30.0) as client:
        response = await client.get(f'{proxy_endpoint}/createAccount', params=params)
        json_response: dict = response.json()
        if json_response.get('ok'):
            result = json_response.get('result', {})
            return UserInfo(result['access_token'], result['short_name'], result['author_name'], result['author_url'], result['auth_url'])


tgph_acc = run(create_account('prolapse', 'prolapse', ''))


def md_to_dom(markdown_text: str) -> list[dict[str, str | list | dict | None]]:
    def handle_heading(element) -> dict[str, str | list | dict | None]:
        if element.tag == 'h1':
            return {'tag': 'h3', 'children': parse_children(element)}
        elif element.tag == 'h2':
            return {'tag': 'h4', 'children': parse_children(element)}
        else:
            return {'tag': 'p', 'children': [{'tag': 'strong', 'children': parse_children(element)}]}

    def handle_list(element) -> dict[str, str | list | dict | None]:
        return {'tag': element.tag, 'children': [{'tag': 'li', 'children': parse_children(li)} for li in element.xpath('./li')]}

    def handle_link(element) -> dict[str, str | list | dict | None]:
        return {'tag': 'a', 'attrs': {'href': element.get('href')}, 'children': parse_children(element)}

    def handle_media(element) -> dict[str, str | list | dict | None]:
        return {'tag': element.tag, 'attrs': {'src': element.get('src')}}

    def parse_children(element) -> list[str | dict[str, str | list | dict | None]]:
        children = []
        for child in element.iterchildren():
            if child.tag:
                children.append(parse_element(child))
            elif isinstance(child, str):
                children.append(child.strip())
        if element.text and element.text.strip():
            children.insert(0, element.text.strip())
        if element.tail and element.tail.strip():
            children.append(element.tail.strip())
        return children

    def parse_element(element) -> dict[str, str | list | dict | None]:
        handlers = {'h1': handle_heading, 'h2': handle_heading, 'h3': handle_heading, 'h4': handle_heading, 'h5': handle_heading, 'h6': handle_heading, 'ul': handle_list, 'ol': handle_list, 'a': handle_link, 'img': handle_media, 'iframe': handle_media}
        handler = handlers.get(element.tag, lambda e: {'tag': e.tag, 'children': parse_children(e)})
        return handler(element)
    html_content = markdown(markdown_text, extensions=['extra', 'sane_lists'])
    tree = html.fromstring(html_content)
    try:
        return [parse_element(element) for element in tree.body]
    except:
        return [parse_element(element) for element in tree.xpath('/*/*')]

async def tgph_create_page(token: str, title: str, markdown_text: str) -> str:
    content = dumps(md_to_dom(markdown_text))
    params = {
        'access_token': token,
        'title': title,
        'content': content,
        'return_content': False
    }
    async with AsyncClient(verify=False, follow_redirects=True, timeout=30.0) as client:
        response = await client.get(f'{proxy_endpoint}/createPage', params=params)
        json_response: dict = response.json()
        if json_response.get('ok'):
            result = json_response.get('result', {})
        else:
            result = {}
            print(f'ошибка создания страницы: {json_response}')
            print(markdown_text)
            print(content)
        return result.get('path')


async def tgph_edit_page(token: str, page: str, title: str, markdown_text: str) -> str:
    content = dumps(md_to_dom(markdown_text))
    data = {
        'access_token': token,
        'path': page,
        'title': title,
        'content': content,
        'return_content': False
    }
    async with AsyncClient(verify=False, follow_redirects=True, timeout=30.0) as client:
        response = await client.post(f'{proxy_endpoint}/editPage', data=data)
        json_response = response.json()
        if json_response.get('ok'):
            result = json_response.get('result', {})
        else:
            result = {}
        return result.get('path', '')


async def download_png(url: str, folder: str, client: AsyncClient, retries: int = 5) -> Path:
    # print(f'загрузка изображения: {url}')
    for attempt in range(retries):
        try:
            response = await client.get(url)
            response.raise_for_status()
            file_path = Path(__file__).parent / folder / f'{uuid4()}.png'
            file_path.parent.mkdir(parents=True, exist_ok=True)
            file_path.write_bytes(response.content)
            return file_path
        except (HTTPStatusError, RequestError) as e:
            if attempt < retries - 1:
                await sleep(2 ** attempt)
            else:
                raise e


async def download_pngs(urls: str | list[str]) -> list[Any]:
    urls = [urls] if isinstance(urls, str) else urls
    # print(f'скачивается список список из {len(urls)}: {urls}')
    # бот coze имеет баг, и из воркфлоу прибавляет предыдущий ответ к ссылкам, если включен контекст чата:
    valid_urls = [url for url in urls if url and '\n' not in url and '\\n' not in url and url.strip() != '']
    if len(valid_urls) != len(urls):
        print(f'некорректные ссылки удалены из списка: {set(urls) - set(valid_urls)}')
    async with AsyncClient(verify=False, follow_redirects=True, timeout=30.0) as client:
        tasks = [download_png(url, str(uuid4()), client) for url in valid_urls]
        return list(await gather(*tasks))


async def optimize_png(image_path: Path, retries: int = 3) -> None:
    command = f'{oxipng_bin.resolve()} --opt 2 --strip safe --out {image_path} {image_path}'
    # print(f'оптимизация картинки {image_path}')
    for attempt in range(retries):
        try:
            process = await create_subprocess_shell(command, stdout=PIPE, stderr=PIPE)
            stdout, stderr = await process.communicate()
            if process.returncode == 0:
                return
            else:
                raise CalledProcessError(process.returncode, command, output=stdout, stderr=stderr)
        except CalledProcessError as e:
            print(f'ошибка при оптимизации {image_path}')
            if attempt < retries - 1:
                await sleep(2 ** attempt)
            else:
                raise e


async def convert_to_jpeg(image_path: Path) -> Path:
    # print(f'конвертируется {image_path}')
    try:
        image = Image.open(image_path)
        output_path = image_path.with_suffix('.jpg')
        image.save(output_path, 'JPEG', quality=98, optimize=True)
        image_path.unlink(missing_ok=True)
        return output_path
    except:
        print(f'ошибка при конвертации {image_path}')
        return image_path


async def convert_to_jpegs(image_paths: list[str | Path] | str | Path) -> tuple[Path]:
    image_paths = [Path(image_file) for image_file in ([image_paths] if not isinstance(image_paths, list) else image_paths)]
    # print(f'оптимизируется список список из {len(image_paths)}: {image_paths}')
    tasks = [convert_to_jpeg(image_path) for image_path in image_paths]
    return await gather(*tasks)


async def optimize_pngs(image_paths: list[str | Path] | str | Path) -> None:
    image_paths = [Path(image_file) for image_file in ([image_paths] if not isinstance(image_paths, list) else image_paths)]
    # print(f'оптимизируется список список из {len(image_paths)}: {image_paths}')
    tasks = [optimize_png(image_path) for image_path in image_paths]
    await gather(*tasks)


async def upload_image_to_imgbb(file_path: Path, file_type: str = 'png') -> str | None:
    for token in tokens:
        url = f'https://api.imgbb.com/1/upload?key={token}'
        try:
            with file_path.open('rb') as file:
                files = {'image': (file_path.name, file, f'image/{file_type}')}
                data = {}
                async with AsyncClient(verify=False, follow_redirects=True, timeout=30.0) as client:
                    response = await client.post(url, files=files, data=data, timeout=30)
                    response.raise_for_status()
                    json = response.json()
                    if json.get('success'):
                        return json['data']['url']
        except Exception as e:
            print(f"ошибка при загрузке с {token}: {e}")
            continue
    return None


async def upload_image_to_freeimagehost(image_path: Path, file_type: str = 'png') -> str | None:
    try:
        async with AsyncClient(verify=False, follow_redirects=True, timeout=30.0) as client:
            with image_path.open("rb") as image_file:
                files = {'source': (image_path.name, image_file, f'image/{file_type}')}
                payload = {'key': '6d207e02198a847aa98d0a2a901485a5', 'action': 'upload', 'format': 'json'}
                response = await client.post('https://freeimage.host/api/1/upload', data=payload, files=files)
            response.raise_for_status()
            response_data = response.json()
            return response_data['image']['url']
    except Exception as e:
        print(f'ошибка при загрузке {image_path}: {e}')
        return None


async def upload_image(file_path: Path | str, file_type: str = 'png') -> str | None:
    file_path = Path(file_path)
    return await upload_image_to_freeimagehost(file_path, file_type) or upload_image_to_imgbb(file_path, file_type)


async def process_image(old_url: str, image_path: Path, convert: bool) -> tuple[str, Path]:
    new_url = await upload_image(image_path, 'png' if not convert else 'jpeg')
    if new_url:
        # print(f'загружено изображение {image_path} в {new_url}')
        pass
    else:
        new_url = old_url
        print(f'не удалось загрузить изображение {image_path}, оставим старую ссылку: {old_url}')

    try:
        image_path.unlink()
    except Exception as e:
        print(f'не удалось удалить файл {image_path}: {e}')

    return new_url, image_path


async def optimize_and_upload(images_urls: List[str] | str, convert: bool = False) -> List[str]:
    images_urls = [images_urls] if isinstance(images_urls, str) else images_urls
    ph_link = None
    if convert:
        try:
            ph_link = await tgph_create_page(tgph_acc.get_token(), 'DAll-E v3', '*изображения скоро появятся, обнови страницу...*')
        except Exception as e:
            print(f'не получилось создать страницу на телеграфе: {e}')
        async with RentryClient('https://rentry.org') as client:
            page = await client.new_page('изображения скоро появятся, обнови страницу...')
            page_id, code = page.url, page.edit_code
        continue_task = create_task(continue_optimizing_and_uploading(images_urls, page_id, code, ph_link))
        print(f'https://telegra.ph/{ph_link}', f'https://rentry.co/{page_id}', f'https://rentry.org/{page_id}')
        return [f'https://telegra.ph/{ph_link}', f'https://rentry.co/{page_id}', f'https://rentry.org/{page_id}'] if ph_link else [f'https://rentry.co/{page_id}', f'https://rentry.org/{page_id}']
    else:
        return await continue_optimizing_and_uploading(images_urls)


async def continue_optimizing_and_uploading(images_urls: list[str], page_id: str = None, code: str = None, ph_link: str = None) -> list[str]:
    images_paths = await download_pngs(images_urls)

    if not page_id:  # convert=False
        await optimize_pngs(images_paths)

    new_images_urls = []
    images_paths = images_paths if not page_id else await convert_to_jpegs(images_paths)

    tasks = []
    for old_url, image_path in zip(images_urls, images_paths):
        tasks.append(process_image(old_url, image_path, page_id is not None))

    results = await gather(*tasks)
    new_images_urls = [result[0] for result in results]
    print(f'новые ссылки: ({len(new_images_urls)}): {new_images_urls}')

    try:
        rmtree(images_paths[0].parent)
    except Exception as e:
        print(f'не удалось удалить файл {images_paths[0].parent}: {e}')

    content = '\n\n'.join([f'![]({url})' for url in new_images_urls])
    print(content)
    if ph_link:
        try:
            await tgph_edit_page(tgph_acc.get_token(), ph_link, 'DAll-E v3', content)
        except Exception as e:
            print(f'не удалось отредактировать на телеграфе: {e}')
    else:
        print(f'страница телеграф не предоставлена: {ph_link}')
    if page_id and code:
        try:
            async with RentryClient('https://rentry.org') as client:
                await client.edit_page(text=content, url=page_id, edit_code=code)
        except Exception as e:
            print(f'не удалось создать страницу в rentry: {e}')

    return new_images_urls if len(new_images_urls) > 0 and new_images_urls[0] else images_urls


app = FastAPI()


class ImageURLs(BaseModel):
    urls: List[HttpUrl]


@app.get('/')
async def read_root():
    return PlainTextResponse('ну пролапс, ну и что', status_code=200)


@app.post('/pngopt_by_urls/')
async def optimize_images_endpoint(image_urls: ImageURLs):
    try:
        optimized_urls = await optimize_and_upload([str(url) for url in image_urls.urls])
        return {"optimized_urls": optimized_urls}
    except Exception as e:
        raise HTTPException(status_code=500, detail=str(e))


@app.post('/jpegs_by_urls/')
async def optimize_images_endpoint(image_urls: ImageURLs):
    try:
        optimized_urls = await optimize_and_upload([str(url) for url in image_urls.urls], convert=True)
        return {"optimized_urls": optimized_urls}
    except Exception as e:
        raise HTTPException(status_code=500, detail=str(e))


if __name__ == "__main__":
    uvicorn_run(app, host='0.0.0.0', port=7860, timeout_keep_alive=90, log_level='info', use_colors=False)