bd1e50e410
- Add crawler_core/boss/sign.py: BossSign traceid generator (pure stdlib) - Add crawler_core/qcwy/sign.py: Job51Sign HMAC-SHA256 signing (pure stdlib) - Add crawler_core/zhilian/sign.py: ZhilianSign header/param signing (pure stdlib) - Add __init__.py for all three crawler_core platform directories - Updated module docstrings to reference crawler_core; all logic unchanged - No imports from spiderJobs or app; no HTTP dependencies
89 lines
2.6 KiB
Python
89 lines
2.6 KiB
Python
"""
|
||
前程无忧 (51Job) 签名算法 (crawler_core)
|
||
从小程序源码 utils/cupid.js + server/request/index.js 翻译而来
|
||
|
||
签名逻辑:
|
||
1. 构造 path = /{endpoint}?api_key=51job×tamp={ts}
|
||
2. GET: message = path + ¶m1=val1¶m2=val2
|
||
POST: message = path + JSON.stringify(body)
|
||
3. sign = HMAC-SHA256(message, SIGN_KEY).hex()
|
||
"""
|
||
|
||
from __future__ import annotations
|
||
|
||
import hmac
|
||
import hashlib
|
||
import time
|
||
import random
|
||
from urllib.parse import quote
|
||
|
||
|
||
# 与小程序硬编码一致（server/config/index.js）
|
||
SIGN_KEY = "abfc8f9dcf8c3f3d8aa294ac5f2cf2cc7767e5592590f39c3f503271dd68562b"
|
||
|
||
|
||
class Job51Sign:
|
||
"""
|
||
前程无忧请求签名
|
||
|
||
功能: 根据请求路径、方法、参数生成 sign 和带签名的 URL path
|
||
参数说明:
|
||
sign_key: HMAC 密钥（默认使用小程序内置 key）
|
||
"""
|
||
|
||
def __init__(self, *, sign_key: str = SIGN_KEY):
|
||
self.sign_key = sign_key
|
||
|
||
@staticmethod
|
||
def generate_uuid() -> str:
|
||
"""生成 UUID（13位时间戳 + 随机数），用作 distinct_id / uuid"""
|
||
ts = str(int(time.time() * 1000))
|
||
rand = str(random.randint(1000000000, 9999999999))
|
||
return ts + rand
|
||
|
||
def build_sign_path(
|
||
self,
|
||
endpoint: str,
|
||
method: str = "GET",
|
||
params: dict | None = None,
|
||
body: dict | None = None,
|
||
) -> tuple[str, str]:
|
||
"""
|
||
构建签名路径和 sign 值
|
||
|
||
Args:
|
||
endpoint: API 路径（不含域名，如 open/noauth/jobs/detail/base/170651439）
|
||
method: HTTP 方法（GET / POST）
|
||
params: GET 请求的额外 query 参数
|
||
body: POST 请求的 JSON body
|
||
|
||
Returns:
|
||
(url_path, sign_hex)
|
||
url_path: 完整的 URL path（含 api_key 和 timestamp）
|
||
sign_hex: HMAC-SHA256 签名
|
||
"""
|
||
import json
|
||
|
||
ts = int(time.time())
|
||
path = f"/{endpoint}?api_key=51job×tamp={ts}"
|
||
|
||
if method.upper() == "GET" and params:
|
||
query_parts = []
|
||
for k, v in params.items():
|
||
query_parts.append(f"{quote(str(k), safe='')}={quote(str(v), safe='')}")
|
||
if query_parts:
|
||
path += "&" + "&".join(query_parts)
|
||
|
||
# 签名字符串
|
||
message = path
|
||
if method.upper() == "POST" and body is not None:
|
||
message += json.dumps(body, ensure_ascii=False, separators=(",", ":"))
|
||
|
||
sign_hex = hmac.new(
|
||
self.sign_key.encode("utf-8"),
|
||
message.encode("utf-8"),
|
||
hashlib.sha256,
|
||
).hexdigest()
|
||
|
||
return path, sign_hex
|