Archived
snowy-yang
ad8c30d932
为什么：原先三处都在手写轮子——ffmpeg CLI 拼参数写标签、aiohttp 单连接手搓 下载、自己比对魔数判格式。手写部分既难测又易错；改走成熟库后测试从 144 涨到 175，并且彻底去掉了 ffmpeg 这个外部二进制依赖。 下载 → aria2c 常驻 JSON-RPC（aria2p 客户端），新增 _aria2.py - 单例后端：首次调用时拉起 aria2c，只监听 127.0.0.1 且强制随机 secret （无鉴权的 aria2 RPC 允许本机任意进程操纵队列、读写任意文件） - split=5 / max-connection-per-server=5 多连接 + 原生断点续传 - 异常语义与内置通道逐一对齐：取消/超限清残片抛 DownloadCanceled / DownloadException，任务失败/超时保留残片抛 RequestException 供续传 - 取不到 aria2c 时 get_backend() 返回 None，自动降级到内置 aiohttp 通道 （本机无 aria2c 的降级路径已实测） 标签 → mutagen 纯 Python 写入，新增 tagging.py - 取代 _write_metadata_with_ffmpeg + _build_format_args（226 行 shell 参数拼接） - MP3(ID3v2.3)/FLAC/M4A/OGG/Opus 分容器实现，无子进程、不重编码 - 就地改帧：TXXX 等未触及的标签原样保留（ffmpeg 整帧重写会丢） - 临时文件 + os.replace 原子替换，写坏不会留下半个文件 - 为什么不是 ffmpeg-python（原方案）：实测它用 **kwargs 建模选项，无法表达 重复键（-metadata ×N），list 值还会被字符串化成 "['0', '1:v']"；-map 尚有 内置的 stream 位置参数入口，但 -metadata 无解，且 overwrite_output=True 生成的 -y 被排到输出文件之后、可能失效 格式探测 → filetype，改 _http.py - 删除手写的 _AUDIO_MAGIC 魔数表，改用 filetype.guess() 判 MIME （audio/* 放行，video/mp4 作为容器显式放行） 配套 - Dockerfile：镜像不再装 ffmpeg，改装 aria2 - pyproject：新增 filetype / aria2p / mutagen；dev 加 httpx2（消除测试里的 StarletteDeprecationWarning，现 0 warning） - __main__.py / __init__.py / models.py：清理 ffmpeg 相关导出与文案 测试：新增 31 个用例（test_aria2.py 13、test_tagging.py 21，含最小合法 FLAC 样本构造）；全套 175 passed / 0 warning
615 lines
29 KiB
Python
615 lines
29 KiB
Python
# -*- coding: utf-8 -*-
|
||
"""下载主类：落盘、元数据写入与 ZIP 打包。
|
||
|
||
（自 music_downloader.py 拆出，行为与原实现一致。）
|
||
"""
|
||
|
||
from __future__ import annotations
|
||
|
||
import os
|
||
import re
|
||
import threading
|
||
import time
|
||
from pathlib import Path
|
||
|
||
from core import sync_http as requests # 同步门面（底层 aiohttp，替代 requests）
|
||
from core.async_bridge import run_sync
|
||
from music_api import APIException, NeteaseAPI, load_cookies
|
||
|
||
from ._aria2 import Aria2Backend, get_backend
|
||
from ._concurrency import (
|
||
CountingSemaphore,
|
||
get_file_lock,
|
||
)
|
||
from ._http import (
|
||
DL_CHUNK_SIZE,
|
||
_sniff_audio_header,
|
||
get_http_session,
|
||
)
|
||
from ._log import (
|
||
logger,
|
||
)
|
||
from .cover import (
|
||
fetch_cover_bytes,
|
||
)
|
||
from .models import (
|
||
API_TYPE_TO_EXT,
|
||
DownloadCanceled,
|
||
DownloadException,
|
||
DownloadResult,
|
||
MusicInfo,
|
||
_RetryFromScratch,
|
||
)
|
||
from .tagging import build_tags, write_tags
|
||
|
||
# 单文件下载体积硬上限：超限即弃，防止单文件拖垮磁盘
|
||
_MAX_DOWNLOAD_BYTES = 500 * 1024 * 1024
|
||
|
||
# ============ 核心实现 ============
|
||
|
||
class MusicDownloader:
|
||
"""网易云音乐下载器
|
||
|
||
用法:
|
||
downloader = MusicDownloader(download_dir="downloads")
|
||
result = downloader.download_music_file(music_id, quality="lossless")
|
||
"""
|
||
|
||
def __init__(
|
||
self,
|
||
download_dir: str = "downloads",
|
||
max_concurrent: int = 3,
|
||
global_max_concurrent: int = 10,
|
||
):
|
||
"""
|
||
Args:
|
||
download_dir: 下载目录
|
||
max_concurrent: 单个批量任务内的最大并发下载数
|
||
global_max_concurrent: 整个服务同时下载的文件数（含标签写入）。
|
||
所有请求共享该信号量，防止大量用户同时下载导致内存/CPU 耗尽。
|
||
"""
|
||
self.download_dir = Path(download_dir)
|
||
self.download_dir.mkdir(exist_ok=True)
|
||
self.max_concurrent = max_concurrent
|
||
|
||
self.api = NeteaseAPI()
|
||
|
||
# 全局并发信号量：所有请求共用，限制同时进行的下载数
|
||
# （带计数，供前端实时展示排队情况）
|
||
self._download_semaphore = CountingSemaphore(global_max_concurrent)
|
||
|
||
# 标签写入专用信号量：写标签会重写整个文件（IO 密集），
|
||
# 与下载分离限流 —— 写满时只阻塞后来者，不占下载名额、不拖慢其它文件的传输
|
||
cpu_count = os.cpu_count() or 2
|
||
self._tag_semaphore = threading.BoundedSemaphore(max(1, min(4, cpu_count)))
|
||
|
||
# ----------- 工具方法 -----------
|
||
|
||
def _sanitize_filename(self, filename: str, max_bytes: int = 180) -> str:
|
||
"""清理文件名非法字符。
|
||
换行/回车/制表等控制字符一律剔除（否则 send_file 头部 500）；
|
||
按 UTF-8 字节截断（Linux 文件系统 255 字节上限，中文按字符截 200 会超），
|
||
且从 _build_file_path 传入时保证尾部 [ID 音质] 标记不被截掉。"""
|
||
filename = re.sub(r'[\x00-\x1f\x7f]', '', filename)
|
||
filename = re.sub(r'[<>:"/\\|?*]', '_', filename)
|
||
filename = filename.strip(' .')
|
||
# 按字节截断（在字符边界上回退，不产生半个字符）
|
||
raw = filename.encode('utf-8')
|
||
if len(raw) > max_bytes:
|
||
cut = filename.encode('utf-8')[:max_bytes]
|
||
# 回退到合法 UTF-8 边界
|
||
while cut and (cut[-1] & 0xC0) == 0x80:
|
||
cut = cut[:-1]
|
||
filename = cut.decode('utf-8', errors='ignore').rstrip(' .')
|
||
return filename or "unknown"
|
||
|
||
def _determine_file_extension(
|
||
self,
|
||
url: str,
|
||
content_type: str = "",
|
||
api_type: str = "",
|
||
) -> str:
|
||
"""确定文件扩展名
|
||
|
||
优先级: API 返回的 type > URL 后缀 > HTTP Content-Type > 默认 mp3
|
||
"""
|
||
api_type = (api_type or "").lower()
|
||
if api_type in API_TYPE_TO_EXT:
|
||
return API_TYPE_TO_EXT[api_type]
|
||
|
||
url_lower = url.lower()
|
||
if ".flac" in url_lower:
|
||
return ".flac"
|
||
if ".mp3" in url_lower:
|
||
return ".mp3"
|
||
if ".m4a" in url_lower or ".mp4" in url_lower:
|
||
return ".m4a"
|
||
|
||
content_type = (content_type or "").lower()
|
||
if "flac" in content_type:
|
||
return ".flac"
|
||
if "mpeg" in content_type or "mp3" in content_type:
|
||
return ".mp3"
|
||
if "mp4" in content_type or "m4a" in content_type:
|
||
return ".m4a"
|
||
|
||
return ".mp3"
|
||
|
||
@staticmethod
|
||
def _unlink_quiet(path: Path) -> None:
|
||
"""安静地删除文件，忽略不存在 / 权限错误"""
|
||
try:
|
||
path.unlink()
|
||
except OSError:
|
||
pass
|
||
|
||
# ----------- 信息获取 -----------
|
||
|
||
def get_music_info(self, music_id: int, quality: str = "standard",
|
||
cookies: dict[str, str] | None = None) -> MusicInfo:
|
||
"""获取音乐详细信息（同步，URL/详情/歌词 3 个接口并行请求，加快单曲/批量启动）"""
|
||
try:
|
||
cookies = cookies if cookies is not None else load_cookies()
|
||
|
||
# 3 个接口并行请求，单曲下载/批量预取都能明显提速
|
||
# URL/详情拿到后立即取消未完成的歌词请求，已知失败不再被拖着等完
|
||
# 异步化：改用 asyncio.gather 并发（api 方法已异步），失败语义与旧线程池一致
|
||
import asyncio
|
||
|
||
async def _gather_info():
|
||
url_t = self.api.get_song_url(music_id, quality, cookies)
|
||
detail_t = self.api.get_song_detail(music_id)
|
||
lyric_t = self.api.get_lyric(music_id, cookies)
|
||
return await asyncio.gather(url_t, detail_t, lyric_t, return_exceptions=True)
|
||
|
||
url_result, detail_result, lyric_result = run_sync(_gather_info())
|
||
if isinstance(url_result, BaseException):
|
||
raise url_result
|
||
if isinstance(detail_result, BaseException):
|
||
raise detail_result
|
||
if isinstance(lyric_result, BaseException):
|
||
lyric_result = None # 歌词接口抖动不阻断解析/下载
|
||
|
||
if not url_result.get("data") or not url_result["data"]:
|
||
raise DownloadException(f"无法获取音乐ID {music_id} 的播放链接")
|
||
|
||
# 上游 data 数组元素可能为 null（同族）
|
||
song_data = url_result["data"][0] or {}
|
||
download_url = song_data.get("url", "") or ""
|
||
if not download_url:
|
||
raise DownloadException(f"音乐ID {music_id} 无可用的下载链接")
|
||
|
||
if not detail_result.get("songs") or not detail_result["songs"]:
|
||
raise DownloadException(f"无法获取音乐ID {music_id} 的详细信息")
|
||
|
||
song_detail = detail_result["songs"][0]
|
||
|
||
# 上游可选字段可能为 null（而非缺失），一律 (x or {}) / (x or []) 容错，
|
||
# 可选展示信息异常不再让整首有效歌曲解析失败
|
||
lyric = ((lyric_result or {}).get("lrc") or {}).get("lyric", "") or ""
|
||
tlyric = ((lyric_result or {}).get("tlyric") or {}).get("lyric", "") or ""
|
||
|
||
artists = "/".join(filter(None, ((a or {}).get("name") for a in (song_detail.get("ar") or []))))
|
||
|
||
al = song_detail.get("al") or {}
|
||
publish_time = al.get("publishTime", 0) or 0
|
||
publish_year = None
|
||
if publish_time:
|
||
import datetime
|
||
try:
|
||
publish_year = datetime.datetime.fromtimestamp(publish_time / 1000).year
|
||
except (OSError, ValueError, OverflowError):
|
||
publish_year = None
|
||
|
||
return MusicInfo(
|
||
id=music_id,
|
||
name=song_detail.get("name") or "未知歌曲",
|
||
artists=artists or "未知艺术家",
|
||
album=al.get("name") or "未知专辑",
|
||
pic_url=al.get("picUrl") or "",
|
||
duration=(song_detail.get("dt") or 0) // 1000,
|
||
track_number=song_detail.get("no", 0) or 0,
|
||
download_url=download_url,
|
||
file_type=(song_data.get("type") or "mp3").lower(), # null 容错
|
||
file_size=song_data.get("size", 0) or 0,
|
||
quality=quality,
|
||
lyric=lyric,
|
||
tlyric=tlyric,
|
||
publish_year=publish_year,
|
||
)
|
||
|
||
except APIException as e:
|
||
raise DownloadException(f"API调用失败: {e}") from e
|
||
except Exception as e:
|
||
raise DownloadException(f"获取音乐信息时发生错误: {e}") from e
|
||
|
||
# ----------- 文件路径 -----------
|
||
|
||
def _build_file_path(self, music_info: MusicInfo) -> Path:
|
||
"""根据 MusicInfo 生成下载文件路径
|
||
|
||
文件名附带歌曲 ID 与请求音质：不同歌曲同名（不同版本/重复收录）落盘互不冲突；
|
||
同一首歌不同音质也各自独立，避免低音质文件被当作高音质交付（调包）。
|
||
交付文件名由调用方另行决定，不受影响。
|
||
先截断主名（按字节），再拼接 [ID 音质] 标记——标记永不丢失，
|
||
杜绝截断后两首不同 ID 的歌落到同一路径。
|
||
"""
|
||
base = f"{music_info.artists} - {music_info.name}"
|
||
tag = f" [{music_info.id} {music_info.quality}]"
|
||
ext = self._determine_file_extension(
|
||
music_info.download_url,
|
||
api_type=music_info.file_type,
|
||
)
|
||
# 按完整路径反算文件名可用长度——旧口径只按 Linux 255 字节
|
||
# 截断文件名，未启用 LongPathsEnabled 的 Windows 上超长歌名会在 open 时
|
||
# OSError 白白重试 3 轮。预算 = 240 - 目录前缀 - 扩展名 - ".part"等余量；
|
||
# 中文名 UTF-8 字节数 ≥ 字符数，按字节截断天然保守。
|
||
# 目录过深时按比例压缩 tag 预算。注意下限保底仍可能超过 260 字节，
|
||
# 该病理场景需由部署侧控制 downloads 目录深度来规避。
|
||
budget = 240 - len(str(self.download_dir)) - len(ext) - 8
|
||
if budget < 124:
|
||
tag_budget = max(24, budget // 3)
|
||
else:
|
||
tag_budget = 60
|
||
safe_base = self._sanitize_filename(base, max_bytes=max(32, budget - tag_budget))
|
||
safe_tag = self._sanitize_filename(tag, max_bytes=tag_budget)
|
||
return self.download_dir / f"{safe_base}{safe_tag}{ext}"
|
||
|
||
# ----------- 下载（同步） -----------
|
||
|
||
def download_music_file(
|
||
self,
|
||
music_id: int,
|
||
quality: str = "standard",
|
||
write_meta: bool = True,
|
||
meta_fields: list[str] | None = None,
|
||
cookies: dict[str, str] | None = None,
|
||
retries: int = 2,
|
||
cancel_event: threading.Event | None = None,
|
||
music_info: MusicInfo | None = None,
|
||
progress_cb=None,
|
||
) -> DownloadResult:
|
||
"""同步下载音乐到本地
|
||
|
||
Args:
|
||
music_id: 音乐ID
|
||
quality: 音质
|
||
write_meta: 是否写入元数据
|
||
meta_fields: 写入字段列表，None 表示全部
|
||
cookies: 用于获取直链/歌词的 Cookie，None 表示全局账号
|
||
retries: 网络/写入失败时的重试次数（默认 2 次）
|
||
cancel_event: 取消事件，置位后立即中止下载（抛 DownloadCanceled）
|
||
music_info: 已预取的歌曲信息（跳过重复 API 调用，用于批量加速）
|
||
progress_cb: 可选进度回调 progress_cb(written_bytes, total_bytes)。
|
||
每写一个分块调用一次（含续传起点）；回调内部的异常会被吞掉，
|
||
绝不影响下载主流程。调用方自行限频。
|
||
|
||
说明:
|
||
- 流式下载到 .part 临时文件，完整下载后再原子改名，半截文件不会被当作已下载
|
||
- 断点续传：失败保留 .part，重试/下次下载从已写字节续传（服务器支持 Range 时）
|
||
- per-file 锁：同名歌曲并发下载互斥，防止 .part 被交叉写入损坏
|
||
- 连接复用：线程级 Session + keep-alive，避免每首歌重建 TCP/TLS 连接
|
||
- 网络传输与 ffmpeg 元数据写入分离限流，互不抢占对方资源
|
||
"""
|
||
if music_info is None:
|
||
music_info = self.get_music_info(music_id, quality, cookies)
|
||
|
||
file_path = self._build_file_path(music_info)
|
||
part_path0 = file_path.with_name(file_path.name + ".part")
|
||
last_error = ""
|
||
|
||
next_wait = 0.0
|
||
for attempt in range(retries + 1):
|
||
# 重试退避睡眠在文件锁外执行——同名并发任务不再陪睡
|
||
if next_wait > 0:
|
||
time.sleep(next_wait)
|
||
next_wait = 0.0
|
||
# 重试轮次顶部的取消检查也要清理 .part（原实现在 try 外抛出会绕过清理）
|
||
if cancel_event is not None and cancel_event.is_set():
|
||
self._unlink_quiet(part_path0)
|
||
raise DownloadCanceled("下载已取消")
|
||
|
||
# part_path 是 file_path 的 .part 别名。必须在进入下载 try 之前绑定：
|
||
# except 分支（取消/网络错误）要清理 .part，若此处未赋值，异常处理器
|
||
# 会以 NameError 二次崩溃、掩盖原始异常
|
||
part_path = part_path0
|
||
|
||
# 同一目标文件的下载者串行化（不同文件之间仍然并行）
|
||
with get_file_lock(file_path):
|
||
try:
|
||
# 其它线程可能刚好完成此文件，锁内复查
|
||
if file_path.exists():
|
||
# 缓存与本次期望不符（跨账号音质降级/上游变更都会让同 ID 同音质
|
||
# 的实际文件不同）：视为缓存 miss 删除重下，而不是把旧文件
|
||
# 当本次结果交付（原实现命中即返回、从不校验尺寸）。
|
||
# 期望未知（file_size=0）时至少拒绝 0 字节坏文件
|
||
st_size = file_path.stat().st_size
|
||
cache_bad = ((music_info.file_size and st_size != music_info.file_size)
|
||
or (not music_info.file_size and st_size == 0))
|
||
if cache_bad:
|
||
self._unlink_quiet(file_path)
|
||
else:
|
||
# 缓存命中也尊重本次的元数据请求（重跑一次 ffmpeg，幂等）
|
||
if write_meta:
|
||
with self._tag_semaphore:
|
||
self._write_metadata(file_path, music_info, meta_fields)
|
||
return DownloadResult(
|
||
success=True,
|
||
file_path=str(file_path),
|
||
file_size=file_path.stat().st_size,
|
||
music_info=music_info,
|
||
)
|
||
|
||
# part_path 已在文件锁外绑定为 part_path0（同一路径），此处不再重复计算
|
||
resume_from = 0
|
||
if part_path.exists():
|
||
existing = part_path.stat().st_size
|
||
if music_info.file_size > 0 and existing == music_info.file_size:
|
||
# 上次已下完但未改名：直接转正（转正后立即返回，
|
||
# 不再发起网络请求——原实现离线时会把现成完整文件报为失败）
|
||
# 改 >= 为 ==——超尺寸残片（上游多发货/坏分片）转正即交付坏文件
|
||
os.replace(part_path, file_path)
|
||
if write_meta:
|
||
with self._tag_semaphore:
|
||
self._write_metadata(file_path, music_info, meta_fields)
|
||
return DownloadResult(
|
||
success=True,
|
||
file_path=str(file_path),
|
||
file_size=file_path.stat().st_size,
|
||
music_info=music_info,
|
||
)
|
||
elif existing > 0:
|
||
if music_info.file_size > 0 and existing > music_info.file_size:
|
||
# 超尺寸残片没有续传价值（上游多发货/坏分片），
|
||
# 丢弃从头下——否则下轮被 >= 误转正交付坏文件
|
||
self._unlink_quiet(part_path)
|
||
else:
|
||
resume_from = existing # 断点续传起点
|
||
else:
|
||
self._unlink_quiet(part_path)
|
||
|
||
# 排队改用可取消等待（的 acquire_cancellable）——
|
||
# 原阻塞式 with 使取消后仍要等全局 10 名额腾出，任务卡死、退款延迟
|
||
if not self._download_semaphore.acquire_cancellable(cancel_event, poll=0.5):
|
||
raise DownloadCanceled("下载已取消")
|
||
try:
|
||
aria2_backend = get_backend()
|
||
if aria2_backend is not None:
|
||
# aria2c 通道：多连接 + 原生断点续传
|
||
self._fetch_via_aria2(aria2_backend, music_info, part_path,
|
||
cancel_event, progress_cb)
|
||
else:
|
||
# 内置通道：aiohttp 流式落盘 + Range 起点校验
|
||
self._fetch_via_http(music_info, part_path, resume_from,
|
||
cancel_event, progress_cb)
|
||
finally:
|
||
# 与 acquire_cancellable 配对释放（原 with 语义）
|
||
self._download_semaphore.release()
|
||
|
||
# 完整性校验（已知期望大小时必须精确相等；未知期望时至少非空）
|
||
actual_size = part_path.stat().st_size
|
||
if music_info.file_size > 0:
|
||
complete = (actual_size == music_info.file_size)
|
||
else:
|
||
complete = (actual_size > 0)
|
||
if not complete:
|
||
# 保留 .part 以便断点续传；下次续传从现有字节继续
|
||
last_error = (
|
||
f"下载不完整: 期望 {music_info.file_size or '未知'} 字节, "
|
||
f"实际 {actual_size} 字节"
|
||
)
|
||
if attempt < retries:
|
||
next_wait = 0.5 * (attempt + 1) # 锁外睡眠
|
||
continue
|
||
|
||
# 转正前内容嗅探——未知期望尺寸（"非空即通过"口径）
|
||
# 时 CDN 错误页/风控 JSON 曾可被存成音频交付
|
||
if music_info.file_size <= 0 and not _sniff_audio_header(part_path):
|
||
self._unlink_quiet(part_path)
|
||
raise DownloadException("下载内容校验失败（非音频数据）")
|
||
|
||
os.replace(part_path, file_path)
|
||
|
||
# 元数据写入移出下载信号量（CPU 操作不占网络名额），
|
||
# 但仍在文件锁内执行，保证其它请求看到的是已完成标记的最终文件
|
||
if write_meta:
|
||
# 元数据段入口补取消检查点——已取消的任务不再
|
||
# 进入信号量排队/ffmpeg 等待（有界但可达 120s），收尾不再延迟
|
||
if cancel_event is not None and cancel_event.is_set():
|
||
raise DownloadCanceled("下载已取消")
|
||
with self._tag_semaphore:
|
||
self._write_metadata(file_path, music_info, meta_fields)
|
||
|
||
return DownloadResult(
|
||
success=True,
|
||
file_path=str(file_path),
|
||
file_size=file_path.stat().st_size,
|
||
music_info=music_info,
|
||
)
|
||
|
||
except DownloadCanceled:
|
||
# 用户明确放弃：清理半截临时文件
|
||
self._unlink_quiet(part_path)
|
||
raise
|
||
except DownloadException:
|
||
# 确定性失败（如体积超上限）不重试——原落泛型
|
||
# handler 会删残片后从 0 重拉，3 轮白拉 ~1.5GB
|
||
raise
|
||
except _RetryFromScratch:
|
||
# .part 已清理，直接进入下一轮从头下载
|
||
last_error = "续传起点失效，重新下载"
|
||
if attempt < retries:
|
||
next_wait = 0.5 * (attempt + 1) # 锁外睡眠
|
||
continue
|
||
except (requests.RequestException, OSError) as e:
|
||
# 网络错误：保留 .part 支持断点续传；
|
||
# 但若任务已被取消，.part 没有任何续传价值，一并清理
|
||
if cancel_event is not None and cancel_event.is_set():
|
||
self._unlink_quiet(part_path)
|
||
raise DownloadCanceled("下载已取消") from e
|
||
last_error = f"下载请求失败: {e}"
|
||
if attempt < retries:
|
||
next_wait = 0.5 * (attempt + 1) # 锁外睡眠
|
||
continue
|
||
except Exception as e:
|
||
last_error = f"下载过程中发生错误: {e}"
|
||
if attempt < retries:
|
||
next_wait = 0.5 * (attempt + 1) # 锁外睡眠
|
||
continue
|
||
|
||
# 最后一次尝试仍失败：若任务已取消按取消收尾
|
||
if cancel_event is not None and cancel_event.is_set():
|
||
self._unlink_quiet(part_path0)
|
||
raise DownloadCanceled("下载已取消")
|
||
return DownloadResult(success=False, error_message=last_error or "下载失败")
|
||
|
||
|
||
# ----------- 传输通道（二选一，产物都是 part_path） -----------
|
||
|
||
def _fetch_via_http(
|
||
self,
|
||
music_info: MusicInfo,
|
||
part_path: Path,
|
||
resume_from: int,
|
||
cancel_event: threading.Event | None,
|
||
progress_cb,
|
||
) -> None:
|
||
"""内置流式通道：core.sync_http 逐块落盘，支持 Range 断点续传。
|
||
|
||
Raises:
|
||
DownloadCanceled: 取消事件置位
|
||
DownloadException: 体积超限、CDN 4xx 等确定性失败
|
||
_RetryFromScratch: 续传起点失效（残片已清，外层从头重下）
|
||
"""
|
||
headers = {}
|
||
mode = "wb"
|
||
if resume_from > 0:
|
||
headers["Range"] = f"bytes={resume_from}-"
|
||
mode = "ab"
|
||
|
||
if progress_cb is not None:
|
||
try:
|
||
progress_cb(resume_from, music_info.file_size or 0)
|
||
except Exception:
|
||
pass # 进度回调绝不影响下载主流程
|
||
resp: requests.SyncResponse | None = None
|
||
try:
|
||
resp = get_http_session().get(
|
||
music_info.download_url,
|
||
stream=True,
|
||
timeout=(10, 60),
|
||
headers=headers,
|
||
)
|
||
# 服务器忽略 Range 返回 200 → 从头重写而非追加
|
||
if resume_from > 0 and resp.status_code != 206:
|
||
mode = "wb"
|
||
# 206 必须从我们请求的起点开始；起点错位的分片拼起来是坏文件
|
||
if resume_from > 0 and resp.status_code == 206:
|
||
cr = resp.headers.get('Content-Range') or ''
|
||
m = re.match(r'bytes\s+(\d+)-', cr.strip())
|
||
if m and int(m.group(1)) != resume_from:
|
||
self._unlink_quiet(part_path)
|
||
raise _RetryFromScratch(
|
||
f"Content-Range 起点错位（{m.group(1)} != {resume_from}），已从头重新下载")
|
||
# 206 却解析不出起点（头被中间层剥离）——盲追加可能拼坏文件，
|
||
# 与起点错位同口径处理：清残片从头下
|
||
if not m:
|
||
self._unlink_quiet(part_path)
|
||
raise _RetryFromScratch("206 响应缺少可用 Content-Range，已从头重新下载")
|
||
resp.raise_for_status()
|
||
# 下载体积硬上限（配置 500MB），超限即弃，防单文件拖垮磁盘
|
||
max_bytes = _MAX_DOWNLOAD_BYTES
|
||
written = resume_from if mode == 'ab' else 0
|
||
over_limit = False
|
||
with open(part_path, mode) as f:
|
||
for chunk in resp.iter_content(chunk_size=DL_CHUNK_SIZE):
|
||
if cancel_event is not None and cancel_event.is_set():
|
||
raise DownloadCanceled("下载已取消")
|
||
if chunk:
|
||
f.write(chunk)
|
||
written += len(chunk)
|
||
if written > max_bytes:
|
||
# 超限残片没有任何续传价值。
|
||
# Windows 无法删除打开中的文件——置标志 break，
|
||
# with 关闭后再清理（原实现在打开中 unlink 被静默吞掉）
|
||
over_limit = True
|
||
break
|
||
if progress_cb is not None:
|
||
try:
|
||
progress_cb(written, music_info.file_size or 0)
|
||
except Exception:
|
||
pass # 进度回调绝不影响下载主流程
|
||
if over_limit:
|
||
self._unlink_quiet(part_path) # 文件已关闭，清理可达
|
||
raise DownloadException(
|
||
f"文件超过大小上限 {max_bytes // (1024*1024)}MB，已中止")
|
||
except requests.HTTPError as e:
|
||
# Range 起点无效（如 CDN 环节变化）：清除旧 part，从头重新下载
|
||
status = getattr(getattr(e, "response", None), "status_code", None)
|
||
if status == 416:
|
||
self._unlink_quiet(part_path)
|
||
raise _RetryFromScratch("Range 起点失效，已从头重新下载") from e
|
||
if status is not None and 400 <= status < 500 and status != 408:
|
||
# CDN 4xx（403 防盗链/429 风控等）是确定性
|
||
# 失败，对同一失效直链白重试只会加剧风控、拖慢失败返回
|
||
# （408 请求超时保留可重试）
|
||
self._unlink_quiet(part_path)
|
||
raise DownloadException(
|
||
f"下载失败（HTTP {status}），下载链接可能已失效") from e
|
||
raise
|
||
finally:
|
||
if resp is not None:
|
||
resp.close()
|
||
|
||
def _fetch_via_aria2(
|
||
self,
|
||
backend: Aria2Backend,
|
||
music_info: MusicInfo,
|
||
part_path: Path,
|
||
cancel_event: threading.Event | None,
|
||
progress_cb,
|
||
) -> None:
|
||
"""aria2c 通道：多连接加速 + 原生断点续传。
|
||
|
||
异常语义与内置通道对齐：取消抛 DownloadCanceled、体积超限抛
|
||
DownloadException（均清残片），网络类失败抛 RequestException
|
||
（保留残片，由外层退避重试续传）。
|
||
"""
|
||
backend.fetch(
|
||
music_info.download_url,
|
||
part_path,
|
||
cancel_event=cancel_event,
|
||
progress_cb=progress_cb,
|
||
max_bytes=_MAX_DOWNLOAD_BYTES,
|
||
)
|
||
|
||
|
||
# ============ 元数据写入（mutagen，纯 Python） ============
|
||
# download_music_to_memory / download_batch_async / get_download_progress
|
||
# 三个方法已删除——全服务端与验证脚本零调用（CLI 遗留），且 to_memory 不接受
|
||
# cookies 参数、真被启用会有行为差异；如需内存下载/异步批量再按当前接口重写
|
||
|
||
def _write_metadata(
|
||
self,
|
||
file_path: Path,
|
||
music_info: MusicInfo,
|
||
fields: list[str] | None = None,
|
||
) -> bool:
|
||
"""写入音频标签与封面（mutagen）；失败不影响交付，返回是否成功。
|
||
|
||
Args:
|
||
file_path: 已转正的音频文件
|
||
music_info: 歌曲信息（标题/歌手/专辑/年份/曲号/歌词/封面地址）
|
||
fields: 要写入的字段列表，None 表示全部写入
|
||
|
||
封面走共享缓存（与 ZIP 打包共用一次网络请求）；各容器的落盘细节见
|
||
tagging.py——按扩展名分派到 ID3v2.3 / Vorbis comment / iTunes ilst。
|
||
"""
|
||
include_cover = fields is None or "cover" in fields
|
||
cover: bytes | None = None
|
||
if include_cover and music_info.pic_url:
|
||
cover = fetch_cover_bytes(music_info.pic_url)
|
||
if not cover:
|
||
logger.warning("封面下载失败，跳过封面写入")
|
||
return write_tags(file_path, build_tags(music_info, fields), cover)
|