feat(python): 完成 Python 全量迁移——剩余平台、登录通道与公共层修复
- 新增 5sing/apple/xm/me/qt/lr 平台包与取流实现（含 xm-sign、MissEvan 签名、蜻蜓 HMAC、Apple developer token） - 新增 wy/kg/tx/lr 登录：扫码、手机号、账密（kg 20028 风控指纹）、cookie 绑定、kg/tx 续期 - request 层：保留多值 Set-Cookie；JSON body 紧凑化（修复 kg 网关验签）；直连不经系统代理（对齐 Node） - 登录路由统一 Body(embed=True)，修复扫码轮询 422；lifespan 接入登录续期调度 - ruff/format/pyright 全绿；14 平台真实联网与 HTTP 路由验证通过
This commit is contained in:
snowy-yang committed
1 parent
800871ef6e
commit
dfc06bf4de
182 files changed
+26620
No files matched your search
@@ -0,0 +1 @@
|
||||
"""共享基础设施（请求层、工具函数）。"""
|
||||
@@ -0,0 +1,36 @@
|
||||
"""二维码生成 —— 把登录 URL 渲染成 data URL，供 /login/* 的 qrimg 字段直接展示。
|
||||
|
||||
对应原 TS 里 `import qrcode from 'qrcode'` 的 `toDataURL` 用法（bili 扫码登录）。
|
||||
用 Pillow 后端输出 PNG，再 base64 成 data URL，前端 <img src> 可直接用。
|
||||
"""
|
||||
|
||||
from __future__ import annotations
|
||||
|
||||
import base64
|
||||
import io
|
||||
|
||||
import qrcode as _qrcode
|
||||
from qrcode.constants import ERROR_CORRECT_M
|
||||
|
||||
|
||||
def to_data_url(text: str, *, box_size: int = 6, border: int = 2) -> str:
|
||||
"""生成 `data:image/png;base64,...`。"""
|
||||
qr = _qrcode.QRCode(version=None, error_correction=ERROR_CORRECT_M, box_size=box_size, border=border)
|
||||
qr.add_data(text)
|
||||
qr.make(fit=True)
|
||||
image = qr.make_image(fill_color="black", back_color="white")
|
||||
buffer = io.BytesIO()
|
||||
image.save(buffer, "PNG")
|
||||
encoded = base64.b64encode(buffer.getvalue()).decode("ascii")
|
||||
return f"data:image/png;base64,{encoded}"
|
||||
|
||||
|
||||
def to_png_bytes(text: str, *, box_size: int = 6, border: int = 2) -> bytes:
|
||||
"""生成 PNG 原始字节（需要直接返回图片响应时用）。"""
|
||||
qr = _qrcode.QRCode(version=None, error_correction=ERROR_CORRECT_M, box_size=box_size, border=border)
|
||||
qr.add_data(text)
|
||||
qr.make(fit=True)
|
||||
image = qr.make_image(fill_color="black", back_color="white")
|
||||
buffer = io.BytesIO()
|
||||
image.save(buffer, "PNG")
|
||||
return buffer.getvalue()
|
||||
@@ -0,0 +1,224 @@
|
||||
"""网络请求层。
|
||||
|
||||
对应原 sdk/common/request.ts，但把 Node 的 `{ promise, cancelHttp }` 句柄收敛为
|
||||
原生 `async def` —— 取消语义由调用方用 `asyncio.wait_for` / task 取消表达。
|
||||
|
||||
保持与原文一致的行为细节：
|
||||
* 不跟随重定向（Node http 默认行为；需要跟随的调用方自行处理 Location）
|
||||
* 不保留 cookie（每次请求独立；各平台自己管理 Cookie 头，签名依赖这一点）
|
||||
* body：dict/list 自动 JSON 序列化，str/bytes 原样发送；form 走 urlencoded
|
||||
* 未显式给 content-type 时按 body 形式补 application/json / x-www-form-urlencoded
|
||||
* 响应体：无论 content-type 都先尝试 JSON 解析，失败返回文本
|
||||
* 64MB 响应上限
|
||||
* 错误文案与原文一致（超时 / 无法连接到服务器 / 请求失败）
|
||||
"""
|
||||
|
||||
from __future__ import annotations
|
||||
|
||||
import asyncio
|
||||
import json
|
||||
from collections.abc import Mapping
|
||||
from dataclasses import dataclass, field
|
||||
from typing import Any
|
||||
|
||||
import aiohttp
|
||||
|
||||
MAX_BUFFER_SIZE = 64 * 1024 * 1024
|
||||
|
||||
DEFAULT_HEADERS = {
|
||||
"User-Agent": (
|
||||
"Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/69.0.3497.100 Safari/537.36"
|
||||
),
|
||||
}
|
||||
|
||||
MSG_CANCEL = "请求已取消"
|
||||
MSG_FAIL = "请求失败"
|
||||
MSG_TIMEOUT = "请求超时"
|
||||
MSG_NOT_CONNECT = "无法连接到服务器"
|
||||
|
||||
|
||||
@dataclass
|
||||
class HttpResponse:
|
||||
status_code: int
|
||||
headers: dict[str, str] = field(default_factory=dict)
|
||||
body: Any = None
|
||||
raw: bytes = b""
|
||||
#: 原始多值 Set-Cookie（headers 折叠同名头后只留最后一个，登录链路必须按数组读）
|
||||
set_cookies: list[str] = field(default_factory=list)
|
||||
|
||||
@property
|
||||
def statusCode(self) -> int: # 兼容原文命名
|
||||
return self.status_code
|
||||
|
||||
def header(self, name: str, default: str = "") -> str:
|
||||
return self.headers.get(name.lower(), default)
|
||||
|
||||
def json(self) -> Any:
|
||||
"""把 body 当 JSON 取；body 已是结构时原样返回。"""
|
||||
return self.body
|
||||
|
||||
|
||||
_session: aiohttp.ClientSession | None = None
|
||||
_session_loop: asyncio.AbstractEventLoop | None = None
|
||||
|
||||
|
||||
def _get_session() -> aiohttp.ClientSession:
|
||||
"""进程级连接池。禁用 cookie jar 与自动重定向，语义对齐 Node http。
|
||||
|
||||
trust_env=False：Node 的 http/https 不读系统代理，而 aiohttp 在 Windows 上会通过
|
||||
注册表拿到「Internet 选项」里的代理（本机是 127.0.0.1:7890）。走该代理会让部分域名
|
||||
（如 xm 的 hdaa.shuzilm.cn）连不上，且所有请求的出口 IP/风控画像与 TS 版不一致，
|
||||
因此这里显式直连，与 Node 行为保持一致。
|
||||
"""
|
||||
global _session, _session_loop
|
||||
loop = asyncio.get_running_loop()
|
||||
if _session is None or _session.closed or _session_loop is not loop:
|
||||
_session = aiohttp.ClientSession(
|
||||
cookie_jar=aiohttp.DummyCookieJar(),
|
||||
auto_decompress=True,
|
||||
trust_env=False,
|
||||
connector=aiohttp.TCPConnector(limit=200, ttl_dns_cache=300),
|
||||
)
|
||||
_session_loop = loop
|
||||
return _session
|
||||
|
||||
|
||||
async def close_session() -> None:
|
||||
global _session
|
||||
if _session is not None and not _session.closed:
|
||||
await _session.close()
|
||||
_session = None
|
||||
|
||||
|
||||
def _parse_body(raw: bytes) -> Any:
|
||||
if not raw:
|
||||
return None
|
||||
text = raw.decode("utf-8", errors="replace")
|
||||
try:
|
||||
return json.loads(text)
|
||||
except (ValueError, TypeError):
|
||||
return text
|
||||
|
||||
|
||||
def _build_body(options: Mapping[str, Any]) -> tuple[bytes | None, bool]:
|
||||
form = options.get("form")
|
||||
if form:
|
||||
from urllib.parse import urlencode
|
||||
|
||||
pairs: list[tuple[str, str]] = []
|
||||
for key, value in form.items():
|
||||
if value is None:
|
||||
continue
|
||||
# 对象值对齐 URLSearchParams 的 JSON.stringify（紧凑、无空格）
|
||||
pairs.append(
|
||||
(
|
||||
key,
|
||||
json.dumps(value, ensure_ascii=False, separators=(",", ":"))
|
||||
if isinstance(value, (dict, list))
|
||||
else str(value),
|
||||
)
|
||||
)
|
||||
return urlencode(pairs).encode(), True
|
||||
body = options.get("body")
|
||||
if body is None:
|
||||
return None, False
|
||||
if isinstance(body, bytes):
|
||||
return body, False
|
||||
if isinstance(body, str):
|
||||
return body.encode(), False
|
||||
# 必须与 JS 的 JSON.stringify 逐字节一致（紧凑分隔符）：
|
||||
# 部分平台（kg 网关）的 signature 就是按紧凑 JSON 计算的，空格会导致验签失败。
|
||||
return json.dumps(body, ensure_ascii=False, separators=(",", ":")).encode(), False
|
||||
|
||||
|
||||
def _flatten_headers(headers: Mapping[str, Any]) -> dict[str, str]:
|
||||
return {str(k): str(v) for k, v in headers.items()}
|
||||
|
||||
|
||||
async def http_fetch(url: str, **options: Any) -> HttpResponse:
|
||||
"""发起 HTTP 请求并返回完整响应。
|
||||
|
||||
options: method / headers / body / form / timeout(毫秒) / lookup / family
|
||||
"""
|
||||
method = str(options.get("method") or "GET").upper()
|
||||
payload, is_form = _build_body(options)
|
||||
headers = {**DEFAULT_HEADERS, **_flatten_headers(options.get("headers") or {})}
|
||||
|
||||
lowered = {k.lower() for k in headers}
|
||||
if payload is not None:
|
||||
if "content-type" not in lowered:
|
||||
headers["content-type"] = "application/x-www-form-urlencoded" if is_form else "application/json"
|
||||
if "content-length" not in lowered:
|
||||
headers["content-length"] = str(len(payload))
|
||||
|
||||
timeout_ms = options.get("timeout")
|
||||
timeout = aiohttp.ClientTimeout(total=(timeout_ms / 1000) if timeout_ms else 60)
|
||||
|
||||
session = _get_session()
|
||||
try:
|
||||
async with session.request(
|
||||
method,
|
||||
url,
|
||||
headers=headers,
|
||||
data=payload,
|
||||
timeout=timeout,
|
||||
allow_redirects=False,
|
||||
) as resp:
|
||||
chunks: list[bytes] = []
|
||||
size = 0
|
||||
async for chunk in resp.content.iter_chunked(64 * 1024):
|
||||
size += len(chunk)
|
||||
if size > MAX_BUFFER_SIZE:
|
||||
raise RuntimeError("Response too large")
|
||||
chunks.append(chunk)
|
||||
raw = b"".join(chunks)
|
||||
return HttpResponse(
|
||||
status_code=resp.status,
|
||||
headers={k.lower(): v for k, v in resp.headers.items()},
|
||||
body=_parse_body(raw),
|
||||
raw=raw,
|
||||
set_cookies=list(resp.headers.getall("set-cookie", [])),
|
||||
)
|
||||
except TimeoutError as err:
|
||||
raise TimeoutError(MSG_TIMEOUT) from err
|
||||
except aiohttp.ClientConnectorDNSError as err:
|
||||
raise ConnectionError(MSG_NOT_CONNECT) from err
|
||||
except aiohttp.ClientConnectionError as err:
|
||||
if isinstance(err, aiohttp.ServerDisconnectedError):
|
||||
raise ConnectionError(MSG_FAIL) from err
|
||||
raise ConnectionError(MSG_NOT_CONNECT) from err
|
||||
except aiohttp.ClientError as err:
|
||||
raise RuntimeError(MSG_FAIL) from err
|
||||
|
||||
|
||||
async def http_probe(url: str, *, headers: Mapping[str, Any] | None = None, timeout_ms: int = 8000) -> tuple[int, str]:
|
||||
"""只读响应头的探测请求：返回 ``(status, location)``，**不消费 body**。
|
||||
|
||||
取流 CDN 探测专用（bili）：CDN 忽略 Range 时会尝试把整段音频推下来，
|
||||
等 http_fetch 读完 body 就等于把文件下载完（点歌卡死）。拿到响应头即退出上下文，
|
||||
aiohttp 随即中止并关闭连接。不跟随重定向，调用方用 location 自行跟跳（对应 Node 的 manual）。
|
||||
"""
|
||||
timeout_obj = aiohttp.ClientTimeout(total=timeout_ms / 1000)
|
||||
session = _get_session()
|
||||
try:
|
||||
async with session.request(
|
||||
"GET",
|
||||
url,
|
||||
headers={**DEFAULT_HEADERS, **_flatten_headers(headers or {})},
|
||||
timeout=timeout_obj,
|
||||
allow_redirects=False,
|
||||
) as resp:
|
||||
return resp.status, resp.headers.get("location", "")
|
||||
except TimeoutError as err:
|
||||
raise TimeoutError(MSG_TIMEOUT) from err
|
||||
except aiohttp.ClientConnectorError as err:
|
||||
raise ConnectionError(MSG_NOT_CONNECT) from err
|
||||
except aiohttp.ClientError as err:
|
||||
raise RuntimeError(MSG_FAIL) from err
|
||||
|
||||
|
||||
async def http_get_json(url: str, **options: Any) -> Any:
|
||||
"""GET 并返回解析后的 body。"""
|
||||
options.setdefault("method", "get")
|
||||
resp = await http_fetch(url, **options)
|
||||
return resp.body
|
||||
@@ -0,0 +1,158 @@
|
||||
"""通用工具函数（对应原 sdk/common/utils.ts 与 sdk/utils.ts）。"""
|
||||
|
||||
from __future__ import annotations
|
||||
|
||||
import hashlib
|
||||
import math
|
||||
import re
|
||||
import time
|
||||
from collections.abc import Iterable, Mapping
|
||||
from datetime import datetime
|
||||
from typing import Any
|
||||
|
||||
|
||||
def to_md5(text: str) -> str:
|
||||
return hashlib.md5(text.encode("utf-8")).hexdigest()
|
||||
|
||||
|
||||
def as_dict(value: Any) -> dict[str, Any]:
|
||||
"""上游 ``x?.field ?? {}`` 的安全取法：非 dict 一律当空 dict。"""
|
||||
return value if isinstance(value, dict) else {}
|
||||
|
||||
|
||||
def as_list(value: Any) -> list[Any]:
|
||||
"""上游 ``x ?? []`` 的安全取法：非 list 一律当空列表。"""
|
||||
return value if isinstance(value, list) else []
|
||||
|
||||
|
||||
def size_formate(size: float | int | None) -> str:
|
||||
"""将字节数格式化为可读字符串。"""
|
||||
if not size:
|
||||
return "0 B"
|
||||
units = ["B", "KB", "MB", "GB", "TB"]
|
||||
number = math.floor(math.log(size) / math.log(1024))
|
||||
number = min(number, len(units) - 1)
|
||||
return f"{size / math.pow(1024, number):.2f} {units[number]}"
|
||||
|
||||
|
||||
def format_play_time(seconds: float | int | str | None) -> str:
|
||||
"""秒数 → mm:ss；0 返回 --/--。"""
|
||||
try:
|
||||
value = float(seconds) # type: ignore[arg-type]
|
||||
except (TypeError, ValueError):
|
||||
return "--/--"
|
||||
if math.isnan(value) or value < 0:
|
||||
value = 0
|
||||
minutes = int(value // 60)
|
||||
secs = int(value % 60)
|
||||
if minutes == 0 and secs == 0:
|
||||
return "--/--"
|
||||
return f"{minutes:02d}:{secs:02d}"
|
||||
|
||||
|
||||
def date_format(value: Any, fmt: str = "Y-M-D h:m:s") -> str:
|
||||
"""时间对象/时间戳/字符串 → 日期字符串（Y/M/D/h/m/s 占位）。"""
|
||||
if not value:
|
||||
return ""
|
||||
if isinstance(value, datetime):
|
||||
date = value
|
||||
elif isinstance(value, (int, float)):
|
||||
date = datetime.fromtimestamp(value / 1000 if value > 1e11 else value)
|
||||
elif isinstance(value, str):
|
||||
raw = value
|
||||
if "T" not in raw:
|
||||
raw = raw.split(".")[0].replace("-", "/")
|
||||
parsed: datetime | None = None
|
||||
for pattern in ("%Y/%m/%d %H:%M:%S", "%Y/%m/%d", "%Y-%m-%dT%H:%M:%S", "%Y-%m-%dT%H:%M:%S.%f"):
|
||||
try:
|
||||
parsed = datetime.strptime(raw[: len(pattern) + 10], pattern)
|
||||
break
|
||||
except ValueError:
|
||||
continue
|
||||
if parsed is None:
|
||||
try:
|
||||
parsed = datetime.fromisoformat(value.replace("Z", "+00:00"))
|
||||
except ValueError:
|
||||
return ""
|
||||
date = parsed
|
||||
else:
|
||||
return ""
|
||||
return (
|
||||
fmt.replace("Y", str(date.year))
|
||||
.replace("M", f"{date.month:02d}")
|
||||
.replace("D", f"{date.day:02d}")
|
||||
.replace("h", f"{date.hour:02d}")
|
||||
.replace("m", f"{date.minute:02d}")
|
||||
.replace("s", f"{date.second:02d}")
|
||||
)
|
||||
|
||||
|
||||
def date_format2(timestamp_ms: float | int) -> str:
|
||||
"""相对时间格式化（用于评论时间）。"""
|
||||
differ = int((time.time() * 1000 - float(timestamp_ms)) / 1000)
|
||||
if differ < 60:
|
||||
return f"{differ}秒前"
|
||||
if differ < 3600:
|
||||
return f"{differ // 60}分钟前"
|
||||
if differ < 86400:
|
||||
return f"{differ // 3600}小时前"
|
||||
if differ < 2592000:
|
||||
return f"{differ // 86400}天前"
|
||||
return date_format(timestamp_ms)
|
||||
|
||||
|
||||
def format_play_count(num: float | int | None) -> str:
|
||||
if num is None:
|
||||
return "0"
|
||||
if num > 100000000:
|
||||
return f"{int(num / 10000000) / 10}亿"
|
||||
if num > 10000:
|
||||
return f"{int(num / 1000) / 10}万"
|
||||
return str(num)
|
||||
|
||||
|
||||
_ENTITY_MAP = {
|
||||
"&": "&",
|
||||
"<": "<",
|
||||
">": ">",
|
||||
""": '"',
|
||||
"'": "'",
|
||||
"'": "'",
|
||||
" ": " ",
|
||||
}
|
||||
|
||||
|
||||
def decode_name(text: Any = "") -> str:
|
||||
"""解码 HTML 实体。"""
|
||||
if not text:
|
||||
return ""
|
||||
if not isinstance(text, str):
|
||||
text = str(text)
|
||||
for entity, char in _ENTITY_MAP.items():
|
||||
text = text.replace(entity, char)
|
||||
return re.sub(r"&#(\d+);", lambda m: chr(int(m.group(1))), text)
|
||||
|
||||
|
||||
def format_singer_name(
|
||||
singers: Iterable[Mapping[str, Any]] | str | None,
|
||||
name_key: str = "name",
|
||||
join: str = "、",
|
||||
) -> str:
|
||||
"""歌手数组或字符串 → 展示名（数组按 name_key 提取后 join）。"""
|
||||
if isinstance(singers, (list, tuple)):
|
||||
names = [str(item[name_key]) for item in singers if isinstance(item, Mapping) and item.get(name_key)]
|
||||
return decode_name(join.join(names))
|
||||
return decode_name(str(singers or ""))
|
||||
|
||||
|
||||
def pick(data: Any, *keys: str, default: Any = None) -> Any:
|
||||
"""沿键路径取值（dict 缺失返回 default），迁移时统一空值访问。"""
|
||||
cur = data
|
||||
for key in keys:
|
||||
if isinstance(cur, Mapping):
|
||||
cur = cur.get(key)
|
||||
else:
|
||||
return default
|
||||
if cur is None:
|
||||
return default
|
||||
return cur
|
||||
Reference in new issue
Block a user