re-xxm/reaction_board/bilibili/auth.py

# -*- coding: utf-8 -*-
"""B 站 Cookie 与请求头工具。

从 ``favsort.py`` 抽取，供现有排序工具和看板共用，避免出现两套不一致的
身份 Cookie / 设备指纹处理逻辑。
"""

from __future__ import annotations

import random
import urllib.parse
import uuid

IDENTITY_KEYS = {
    "sessdata",
    "bili_jct",
    "dedeuserid",
    "dedeuserid__ckmd5",
    "bili_ticket",
    "bili_ticket_expires",
    "sid",
    "b_lsid",
}

DROP_STALE = {"sid", "b_lsid"}  # 会话级字段，刷新时应丢弃旧值


def strip_identity(cookie: str) -> str:
    """剥离账号身份类 cookie，保留 buvid3 / buvid4 / _uuid 等设备指纹。

    实测结论：
      - ids：完全无 cookie 也能读（连私密收藏夹都返回全量）；
      - infos：不需要账号身份，但缺设备指纹 cookie 会被 HTTP 412 拦；
      - sort：必须带完整 cookie + csrf，不能走这里。
    """
    kept = []
    for part in (cookie or "").split(";"):
        part = part.strip()
        if not part:
            continue
        key = part.split("=", 1)[0].strip().lower()
        if key in IDENTITY_KEYS:
            continue
        kept.append(part)
    return "; ".join(kept)


def _is_latin1(value: str) -> bool:
    try:
        value.encode("latin-1")
    except UnicodeEncodeError:
        return False
    return True


def cookie_header(cookie: str, *, auth: bool = False) -> str:
    """生成可放入 HTTP 头的 Cookie。占位中文、缺 ``=`` 的片段和非 latin-1 字段都会丢掉。"""
    source = cookie if auth else strip_identity(cookie)
    return "; ".join(
        f"{key}={value}"
        for key, value in parse_cookie(source)
        if _is_latin1(key) and _is_latin1(value)
    )


def make_headers(cfg: dict, auth: bool = False, *, referer: str | None = None) -> dict[str, str]:
    """构造与 ``data/raw-request.txt`` 对齐的 B 站 Web 请求头。"""
    return {
        "accept": "*/*",
        "accept-language": "zh-CN,zh;q=0.9,en-US;q=0.8,en;q=0.7",
        "cookie": cookie_header(str(cfg.get("cookie") or ""), auth=auth),
        "dnt": "1",
        "origin": "https://space.bilibili.com",
        "priority": "u=1, i",
        "referer": referer or "https://space.bilibili.com/",
        "sec-ch-ua": '"Google Chrome";v="155", "Chromium";v="155", "Not(A:Brand";v="24"',
        "sec-ch-ua-mobile": "?0",
        "sec-ch-ua-platform": '"Windows"',
        "sec-fetch-dest": "empty",
        "sec-fetch-mode": "cors",
        "sec-fetch-site": "same-site",
        "user-agent": (
            "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 "
            "(KHTML, like Gecko) Chrome/155.0.0.0 Safari/537.36"
        ),
    }


def parse_cookie(cookie: str) -> list[tuple[str, str]]:
    """把 cookie 字符串解析成 [(key, value)]，保留原顺序。"""
    pairs = []
    for part in (cookie or "").split(";"):
        part = part.strip()
        if not part or "=" not in part:
            continue
        k, v = part.split("=", 1)
        pairs.append((k.strip(), v.strip()))
    return pairs


def cookie_value(cookie: str, name: str) -> str:
    name = name.lower()
    for k, v in parse_cookie(cookie):
        if k.lower() == name:
            return v
    return ""


def cookies_from_set_cookie(headers) -> dict[str, str]:
    """从 Set-Cookie 头里取 name=value（忽略 Path/Domain/Expires 等属性）。"""
    pairs = {}
    for h in headers or []:
        first = h.split(";", 1)[0].strip()
        if "=" not in first:
            continue
        k, v = first.split("=", 1)
        pairs[k.strip()] = v.strip()
    return pairs


def merge_cookie(old: str, new_pairs: dict[str, str]) -> str:
    """只替换身份类 cookie，其它字段（设备指纹等）原样保留。

    新值里没有的身份字段不会被删除（例如服务端没下发新的 sid 时保留旧值），
    只有 DROP_STALE 里的会话字段会在刷新时清掉。
    """
    new_l = {
        k.lower(): (k, v)
        for k, v in new_pairs.items()
        if k.lower() in IDENTITY_KEYS
    }
    out, seen = [], set()
    for k, v in parse_cookie(old):
        kl = k.lower()
        if kl in new_l:
            nk, nv = new_l[kl]
            out.append((nk, nv))
            seen.add(kl)
        elif kl in DROP_STALE:
            continue
        else:
            out.append((k, v))
            seen.add(kl)
    for kl, (k, v) in new_l.items():
        if kl not in seen:
            out.append((k, v))
    return "; ".join(f"{k}={v}" for k, v in out)


def gen_uuid() -> str:
    """生成 B 站风格的 _uuid（前端本来就是本地生成的，服务端不下发）。"""
    suffix = "".join(random.choices("0123456789ABCDEF", k=7))
    return f"{str(uuid.uuid4()).upper()}{suffix}infoc"


def sessdata_expiry(cookie: str) -> int | None:
    """SESSDATA 形如 <值>%2C<过期时间戳>%2C<校验>，取中间的时间戳。"""
    value = urllib.parse.unquote(cookie_value(cookie, "SESSDATA"))
    parts = value.split(",")
    if len(parts) >= 2 and parts[1].isdigit():
        return int(parts[1])
    return None