mirror of
https://github.com/flibusta-apps/books_downloader.git
synced 2025-12-06 15:05:37 +01:00
315 lines
10 KiB
Python
315 lines
10 KiB
Python
import asyncio
|
|
from typing import Optional, AsyncIterator, cast
|
|
import zipfile
|
|
|
|
import aiofiles
|
|
import aiofiles.os
|
|
import asynctempfile
|
|
import httpx
|
|
|
|
from app.services.base import BaseDownloader
|
|
from app.services.book_library import BookLibraryClient
|
|
from app.services.exceptions import NotSuccess, ReceivedHTML, ConvertationError
|
|
from app.services.utils import (
|
|
zip,
|
|
unzip,
|
|
get_filename,
|
|
process_pool_executor,
|
|
async_retry,
|
|
)
|
|
from core.config import env_config, SourceConfig
|
|
|
|
|
|
class FLDownloader(BaseDownloader):
|
|
EXCLUDE_UNZIP = ["html"]
|
|
|
|
def __init__(self, book_id: int, file_type: str, source_id: int):
|
|
self.book_id = book_id
|
|
self.original_file_type = file_type
|
|
self.source_id = source_id
|
|
|
|
self.get_book_data_task = asyncio.create_task(self._get_book_data())
|
|
self.get_content_task = asyncio.create_task(self._get_content())
|
|
|
|
@property
|
|
def file_type(self):
|
|
return self.original_file_type.replace("zip", "")
|
|
|
|
@property
|
|
def need_zip(self):
|
|
return "zip" in self.original_file_type
|
|
|
|
async def get_filename(self) -> str:
|
|
if not self.get_book_data_task.done():
|
|
await asyncio.wait_for(self.get_book_data_task, None)
|
|
|
|
book = self.get_book_data_task.result()
|
|
if book is None:
|
|
raise ValueError("Book is None!")
|
|
|
|
return get_filename(self.book_id, book, self.file_type)
|
|
|
|
async def get_final_filename(self, force_zip: bool = False) -> str:
|
|
if self.need_zip or force_zip:
|
|
return (await self.get_filename()) + ".zip"
|
|
|
|
return await self.get_filename()
|
|
|
|
@async_retry(NotSuccess, times=5, delay=10)
|
|
async def _download_from_source(
|
|
self, source_config: SourceConfig, file_type: Optional[str] = None
|
|
) -> tuple[httpx.AsyncClient, httpx.Response, bool]:
|
|
basic_url: str = source_config.URL
|
|
proxy: Optional[str] = source_config.PROXY
|
|
|
|
file_type_ = file_type or self.file_type
|
|
|
|
if self.file_type in ("fb2", "epub", "mobi"):
|
|
url = basic_url + f"/b/{self.book_id}/{file_type_}"
|
|
else:
|
|
url = basic_url + f"/b/{self.book_id}/download"
|
|
|
|
client_kwargs = {
|
|
"timeout": httpx.Timeout(10 * 60, connect=15, read=60),
|
|
"follow_redirects": True,
|
|
}
|
|
|
|
if proxy is not None:
|
|
client = httpx.AsyncClient(proxies=httpx.Proxy(url=proxy), **client_kwargs)
|
|
else:
|
|
client = httpx.AsyncClient(**client_kwargs)
|
|
|
|
request = client.build_request(
|
|
"GET",
|
|
url,
|
|
)
|
|
try:
|
|
response = await client.send(request, stream=True)
|
|
except (asyncio.CancelledError, httpx.HTTPError) as e:
|
|
await client.aclose()
|
|
raise NotSuccess(str(e))
|
|
|
|
try:
|
|
if response.status_code != 200:
|
|
raise NotSuccess(f"Status code is {response.status_code}!")
|
|
|
|
content_type = response.headers.get("Content-Type")
|
|
content_disposition = response.headers.get("Content-Disposition", "")
|
|
|
|
if (
|
|
"text/html" in content_type
|
|
and self.file_type.lower() != "html"
|
|
and "html" not in content_disposition.lower()
|
|
):
|
|
raise ReceivedHTML()
|
|
|
|
return client, response, "application/zip" in content_type
|
|
except (asyncio.CancelledError, httpx.HTTPError, NotSuccess, ReceivedHTML) as e:
|
|
await response.aclose()
|
|
await client.aclose()
|
|
|
|
if isinstance(e, httpx.HTTPError):
|
|
raise NotSuccess(str(e))
|
|
else:
|
|
raise e
|
|
|
|
@classmethod
|
|
async def _close_other_done(
|
|
cls,
|
|
done_tasks: set[asyncio.Task[tuple[httpx.AsyncClient, httpx.Response, bool]]],
|
|
):
|
|
for task in done_tasks:
|
|
try:
|
|
data = await task
|
|
|
|
await data[0].aclose()
|
|
await data[1].aclose()
|
|
except (
|
|
NotSuccess,
|
|
ReceivedHTML,
|
|
ConvertationError,
|
|
FileNotFoundError,
|
|
ValueError,
|
|
asyncio.InvalidStateError,
|
|
asyncio.CancelledError,
|
|
):
|
|
continue
|
|
|
|
async def _wait_until_some_done(
|
|
self, tasks: set[asyncio.Task[tuple[httpx.AsyncClient, httpx.Response, bool]]]
|
|
) -> Optional[tuple[httpx.AsyncClient, httpx.Response, bool]]:
|
|
tasks_ = tasks
|
|
|
|
while tasks_:
|
|
done, pending = await asyncio.wait(
|
|
tasks_, return_when=asyncio.FIRST_COMPLETED
|
|
)
|
|
|
|
for task in done:
|
|
try:
|
|
data = task.result()
|
|
|
|
for t_task in pending:
|
|
t_task.cancel()
|
|
|
|
await self._close_other_done(pending)
|
|
await self._close_other_done(
|
|
{ttask for ttask in done if ttask != task}
|
|
)
|
|
|
|
return data
|
|
except:
|
|
continue
|
|
|
|
tasks_ = pending
|
|
|
|
return None
|
|
|
|
async def _write_response_content_to_ntf(self, temp_file, response: httpx.Response):
|
|
async for chunk in response.aiter_bytes(2048):
|
|
await temp_file.write(chunk)
|
|
|
|
await temp_file.flush()
|
|
await temp_file.seek(0)
|
|
|
|
async def _unzip(self, response: httpx.Response, file_type: str) -> Optional[str]:
|
|
async with asynctempfile.NamedTemporaryFile(delete=True) as temp_file:
|
|
try:
|
|
await self._write_response_content_to_ntf(temp_file, response)
|
|
except httpx.HTTPError:
|
|
return None
|
|
|
|
await temp_file.flush()
|
|
|
|
try:
|
|
return await asyncio.get_event_loop().run_in_executor(
|
|
process_pool_executor, unzip, temp_file.name, file_type
|
|
)
|
|
except (FileNotFoundError, zipfile.BadZipFile):
|
|
return None
|
|
|
|
async def _download_with_converting(
|
|
self,
|
|
) -> tuple[httpx.AsyncClient, httpx.Response, bool]:
|
|
tasks = set()
|
|
|
|
for source in env_config.FL_SOURCES:
|
|
tasks.add(
|
|
asyncio.create_task(self._download_from_source(source, file_type="fb2"))
|
|
)
|
|
|
|
data = await self._wait_until_some_done(tasks)
|
|
|
|
if data is None:
|
|
raise ValueError
|
|
|
|
client, response, is_zip = data
|
|
|
|
try:
|
|
if is_zip:
|
|
filename_to_convert = await self._unzip(response, "fb2")
|
|
else:
|
|
async with asynctempfile.NamedTemporaryFile(delete=False) as temp_file:
|
|
await self._write_response_content_to_ntf(temp_file, response)
|
|
filename_to_convert = temp_file.name
|
|
finally:
|
|
await response.aclose()
|
|
await client.aclose()
|
|
|
|
if filename_to_convert is None:
|
|
raise ValueError
|
|
|
|
form = {"format": self.file_type}
|
|
files = {"file": open(filename_to_convert, "rb")}
|
|
|
|
converter_client = httpx.AsyncClient(timeout=5 * 60)
|
|
converter_request = converter_client.build_request(
|
|
"POST", env_config.CONVERTER_URL, data=form, files=files
|
|
)
|
|
|
|
try:
|
|
converter_response = await converter_client.send(
|
|
converter_request, stream=True
|
|
)
|
|
except (httpx.ConnectError, httpx.ReadTimeout, asyncio.CancelledError):
|
|
await converter_client.aclose()
|
|
raise ConvertationError
|
|
finally:
|
|
await aiofiles.os.remove(filename_to_convert)
|
|
|
|
try:
|
|
if response.status_code != 200:
|
|
raise ConvertationError
|
|
|
|
return converter_client, converter_response, False
|
|
except (asyncio.CancelledError, ConvertationError):
|
|
await converter_response.aclose()
|
|
await converter_client.aclose()
|
|
await aiofiles.os.remove(filename_to_convert)
|
|
raise
|
|
|
|
async def _get_content(self) -> Optional[tuple[AsyncIterator[bytes], str]]:
|
|
tasks = set()
|
|
|
|
for source in env_config.FL_SOURCES:
|
|
tasks.add(asyncio.create_task(self._download_from_source(source)))
|
|
|
|
if self.file_type.lower() in ["epub", "mobi"]:
|
|
tasks.add(asyncio.create_task(self._download_with_converting()))
|
|
|
|
data = await self._wait_until_some_done(tasks)
|
|
|
|
if data is None:
|
|
return None
|
|
|
|
client, response, is_zip = data
|
|
|
|
try:
|
|
if is_zip and self.file_type.lower() not in self.EXCLUDE_UNZIP:
|
|
temp_filename = await self._unzip(response, self.file_type)
|
|
else:
|
|
async with asynctempfile.NamedTemporaryFile(delete=False) as temp_file:
|
|
temp_filename = temp_file.name
|
|
await self._write_response_content_to_ntf(temp_file, response)
|
|
finally:
|
|
await response.aclose()
|
|
await client.aclose()
|
|
|
|
if temp_filename is None:
|
|
return None
|
|
|
|
if self.need_zip:
|
|
content_filename = await asyncio.get_event_loop().run_in_executor(
|
|
process_pool_executor, zip, await self.get_filename(), temp_filename
|
|
)
|
|
await aiofiles.os.remove(temp_filename)
|
|
else:
|
|
content_filename = temp_filename
|
|
|
|
force_zip = is_zip and self.file_type.lower() in self.EXCLUDE_UNZIP
|
|
|
|
async def _content_iterator() -> AsyncIterator[bytes]:
|
|
try:
|
|
async with aiofiles.open(content_filename, "rb") as temp_file:
|
|
while chunk := await temp_file.read(2048):
|
|
yield cast(bytes, chunk)
|
|
finally:
|
|
await aiofiles.os.remove(content_filename)
|
|
|
|
return _content_iterator(), await self.get_final_filename(force_zip)
|
|
|
|
async def _get_book_data(self):
|
|
return await BookLibraryClient.get_remote_book(self.source_id, self.book_id)
|
|
|
|
async def _download(self) -> Optional[tuple[AsyncIterator[bytes], str]]:
|
|
await asyncio.wait([self.get_book_data_task, self.get_content_task])
|
|
|
|
return self.get_content_task.result()
|
|
|
|
@classmethod
|
|
async def download(
|
|
cls, remote_id: int, file_type: str, source_id: int
|
|
) -> Optional[tuple[AsyncIterator[bytes], str]]:
|
|
downloader = cls(remote_id, file_type, source_id)
|
|
return await downloader._download()
|