init
This commit is contained in:
@@ -0,0 +1 @@
|
||||
"""浏览器自动化包"""
|
||||
@@ -0,0 +1,340 @@
|
||||
"""
|
||||
投递自动化模块
|
||||
在后台线程运行,支持用户随时开始/结束。
|
||||
向云端服务端上报投递记录与统计。
|
||||
"""
|
||||
import random
|
||||
import re
|
||||
import time
|
||||
import threading
|
||||
from dataclasses import dataclass
|
||||
from loguru import logger
|
||||
|
||||
from .manager import BrowserManager
|
||||
from ..ai import AIClient
|
||||
from ..server_api import ServerAPI
|
||||
|
||||
|
||||
@dataclass
|
||||
class JobCard:
|
||||
job_id: str
|
||||
title: str
|
||||
company: str
|
||||
salary: str
|
||||
location: str
|
||||
hr_active: str
|
||||
can_chat: bool
|
||||
can_deliver: bool
|
||||
jd_text: str = ""
|
||||
|
||||
|
||||
class Applicant:
|
||||
"""投递执行器(可启停)"""
|
||||
|
||||
BASE_URL = "https://www.zhipin.com"
|
||||
|
||||
def __init__(
|
||||
self,
|
||||
manager: BrowserManager,
|
||||
server_api: ServerAPI,
|
||||
ai_client: AIClient | None,
|
||||
):
|
||||
self.manager = manager
|
||||
self.server = server_api
|
||||
self.ai = ai_client
|
||||
|
||||
# 运行控制
|
||||
self._stop_event = threading.Event()
|
||||
self._thread: threading.Thread | None = None
|
||||
self.is_running = False
|
||||
|
||||
# 运行时状态(供 Web 面板查询)
|
||||
self.session_id: int | None = None
|
||||
self.primary_resume_content: str = ""
|
||||
self.stats = {"applied": 0, "skipped": 0, "current": ""}
|
||||
|
||||
# ------------------------------------------------------------------
|
||||
# 启停控制
|
||||
# ------------------------------------------------------------------
|
||||
def start(self, keywords: list[str], city_code: str,
|
||||
salary_min: int | None, salary_max: int | None,
|
||||
greeting: str, min_score: int, delay_range: tuple[int, int]) -> None:
|
||||
"""在新线程中启动投递任务"""
|
||||
if self.is_running:
|
||||
logger.warning("已有投递任务在运行")
|
||||
return
|
||||
|
||||
# 创建服务端会话
|
||||
session = self.server.start_session(
|
||||
keywords=",".join(keywords),
|
||||
city_code=city_code,
|
||||
salary_min=salary_min,
|
||||
salary_max=salary_max,
|
||||
)
|
||||
self.session_id = session["id"]
|
||||
|
||||
# 获取主简历内容
|
||||
primary = self.server.get_primary_resume()
|
||||
self.primary_resume_content = primary["content"] if primary else ""
|
||||
|
||||
self._stop_event.clear()
|
||||
self.is_running = True
|
||||
self.stats = {"applied": 0, "skipped": 0, "current": "初始化..."}
|
||||
|
||||
self._thread = threading.Thread(
|
||||
target=self._run,
|
||||
args=(keywords, city_code, salary_min, salary_max,
|
||||
greeting, min_score, delay_range),
|
||||
daemon=True,
|
||||
)
|
||||
self._thread.start()
|
||||
logger.success(f"投递任务已启动,会话 ID:{self.session_id}")
|
||||
|
||||
def stop(self) -> None:
|
||||
"""请求停止投递任务"""
|
||||
logger.info("收到停止投递请求...")
|
||||
self._stop_event.set()
|
||||
|
||||
def _finish(self) -> None:
|
||||
"""任务结束后的收尾工作"""
|
||||
self.is_running = False
|
||||
if self.session_id:
|
||||
try:
|
||||
self.server.stop_session(self.session_id)
|
||||
except Exception as e:
|
||||
logger.warning(f"结束会话失败:{e}")
|
||||
self.stats["current"] = "已停止"
|
||||
logger.info(
|
||||
f"投递任务结束。本次投递 {self.stats['applied']} 个,"
|
||||
f"跳过 {self.stats['skipped']} 个"
|
||||
)
|
||||
|
||||
def _should_stop(self) -> bool:
|
||||
return self._stop_event.is_set()
|
||||
|
||||
# ------------------------------------------------------------------
|
||||
# 主循环
|
||||
# ------------------------------------------------------------------
|
||||
def _run(self, keywords, city_code, salary_min, salary_max,
|
||||
greeting, min_score, delay_range):
|
||||
try:
|
||||
for keyword in keywords:
|
||||
if self._should_stop():
|
||||
break
|
||||
self.stats["current"] = f"搜索关键词:{keyword}"
|
||||
self._process_keyword(
|
||||
keyword, city_code, salary_min, salary_max,
|
||||
greeting, min_score, delay_range,
|
||||
)
|
||||
if not self._should_stop():
|
||||
self._sleep(5, 9)
|
||||
except Exception as e:
|
||||
logger.error(f"投递任务异常:{e}")
|
||||
finally:
|
||||
self._finish()
|
||||
|
||||
def _process_keyword(self, keyword, city_code, salary_min, salary_max,
|
||||
greeting, min_score, delay_range, max_per_keyword: int = 15):
|
||||
page = 1
|
||||
applied_in_kw = 0
|
||||
while applied_in_kw < max_per_keyword and not self._should_stop():
|
||||
cards = self._fetch_job_list(keyword, city_code, salary_min, salary_max, page)
|
||||
if not cards:
|
||||
break
|
||||
for card in cards:
|
||||
if self._should_stop():
|
||||
return
|
||||
if applied_in_kw >= max_per_keyword:
|
||||
break
|
||||
applied_in_kw += self._handle_card(card, greeting, min_score, delay_range)
|
||||
page += 1
|
||||
|
||||
def _handle_card(self, card: JobCard, greeting: str, min_score: int,
|
||||
delay_range: tuple[int, int]) -> int:
|
||||
"""处理单个职位卡片,返回是否投递成功(1/0)"""
|
||||
# 1. 去重(服务端查询)
|
||||
try:
|
||||
if self.server.check_applied(card.job_id):
|
||||
return 0
|
||||
except Exception:
|
||||
pass
|
||||
|
||||
self.stats["current"] = f"评估:{card.company} - {card.title}"
|
||||
|
||||
# 2. AI 匹配度评分
|
||||
ai_score, ai_reason = -1, ""
|
||||
if self.ai and self.ai.available() and self.primary_resume_content:
|
||||
jd = card.jd_text or self._fetch_jd_detail(card)
|
||||
result = self.ai.score_job(jd, self.primary_resume_content)
|
||||
ai_score, ai_reason = result.score, result.reason
|
||||
if not result.is_worth_applying(min_score):
|
||||
self._report_skip(card, f"AI 评分低({ai_score})", ai_score, ai_reason)
|
||||
return 0
|
||||
|
||||
# 3. 执行投递
|
||||
success, apply_type = self._apply(card, greeting)
|
||||
if success:
|
||||
self._report_apply(card, apply_type, ai_score, ai_reason)
|
||||
self._sleep(*delay_range)
|
||||
return 1
|
||||
return 0
|
||||
|
||||
# ------------------------------------------------------------------
|
||||
# 服务端上报
|
||||
# ------------------------------------------------------------------
|
||||
def _report_apply(self, card: JobCard, apply_type: str, ai_score: int, ai_reason: str):
|
||||
try:
|
||||
self.server.report_application(self.session_id, {
|
||||
"job_id": card.job_id,
|
||||
"job_title": card.title,
|
||||
"company": card.company,
|
||||
"salary": card.salary,
|
||||
"location": card.location,
|
||||
"apply_type": apply_type,
|
||||
"ai_score": ai_score,
|
||||
"ai_reason": ai_reason[:255] if ai_reason else None,
|
||||
})
|
||||
self.server.update_session_stats(self.session_id, applied_delta=1, skipped_delta=0)
|
||||
self.stats["applied"] += 1
|
||||
logger.success(f"[{self.stats['applied']}] 投递成功({apply_type}):{card.company} - {card.title}")
|
||||
except Exception as e:
|
||||
logger.warning(f"上报投递记录失败:{e}")
|
||||
|
||||
def _report_skip(self, card: JobCard, reason: str, ai_score: int = -1, ai_reason: str = ""):
|
||||
try:
|
||||
self.server.report_application(self.session_id, {
|
||||
"job_id": card.job_id,
|
||||
"job_title": card.title,
|
||||
"company": card.company,
|
||||
"salary": card.salary,
|
||||
"apply_type": "skipped",
|
||||
"ai_score": ai_score,
|
||||
"ai_reason": ai_reason[:255] if ai_reason else None,
|
||||
"skip_reason": reason,
|
||||
})
|
||||
self.server.update_session_stats(self.session_id, applied_delta=0, skipped_delta=1)
|
||||
self.stats["skipped"] += 1
|
||||
logger.info(f"跳过({reason}):{card.company} - {card.title}")
|
||||
except Exception as e:
|
||||
logger.debug(f"上报跳过记录失败:{e}")
|
||||
|
||||
# ------------------------------------------------------------------
|
||||
# 页面操作
|
||||
# ------------------------------------------------------------------
|
||||
def _fetch_job_list(self, keyword, city_code, salary_min, salary_max, page) -> list[JobCard]:
|
||||
page_obj = self.manager.ensure_connected()
|
||||
url = f"{self.BASE_URL}/web/geek/job?query={keyword}&city={city_code}&page={page}"
|
||||
if salary_min or salary_max:
|
||||
url += f"&salary={(salary_min or 0) * 1000},{(salary_max or 0) * 1000}"
|
||||
|
||||
self.stats["current"] = f"加载职位列表 第{page}页"
|
||||
try:
|
||||
page_obj.goto(url, wait_until="domcontentloaded", timeout=30000)
|
||||
page_obj.wait_for_timeout(2500)
|
||||
page_obj.wait_for_selector(".job-card-wrapper", timeout=10000)
|
||||
except Exception:
|
||||
logger.warning(f"第 {page} 页无职位列表")
|
||||
return []
|
||||
|
||||
cards = []
|
||||
for el in page_obj.query_selector_all(".job-card-wrapper"):
|
||||
try:
|
||||
card = self._parse_card(el)
|
||||
if card:
|
||||
cards.append(card)
|
||||
except Exception:
|
||||
continue
|
||||
logger.info(f"第 {page} 页解析到 {len(cards)} 个职位")
|
||||
return cards
|
||||
|
||||
def _parse_card(self, el) -> JobCard | None:
|
||||
def text(sel: str) -> str:
|
||||
node = el.query_selector(sel)
|
||||
return node.inner_text().strip() if node else ""
|
||||
|
||||
link = el.query_selector("a.job-card-left")
|
||||
href = link.get_attribute("href") if link else ""
|
||||
m = re.search(r"/job_detail/(\w+)\.html", href or "")
|
||||
if not m:
|
||||
return None
|
||||
return JobCard(
|
||||
job_id=m.group(1),
|
||||
title=text(".job-name"),
|
||||
company=text(".company-name"),
|
||||
salary=text(".salary"),
|
||||
location=text(".job-area"),
|
||||
hr_active=text(".info-public"),
|
||||
can_chat=bool(el.query_selector(".start-chat-btn:not([disabled])")),
|
||||
can_deliver=bool(el.query_selector(".btn-deliver")),
|
||||
)
|
||||
|
||||
def _fetch_jd_detail(self, card: JobCard) -> str:
|
||||
page_obj = self.manager.page
|
||||
try:
|
||||
page_obj.goto(f"{self.BASE_URL}/job_detail/{card.job_id}.html",
|
||||
wait_until="domcontentloaded", timeout=20000)
|
||||
page_obj.wait_for_timeout(1500)
|
||||
el = page_obj.query_selector(".job-detail-section")
|
||||
jd = el.inner_text().strip() if el else f"{card.title} {card.company}"
|
||||
page_obj.go_back()
|
||||
page_obj.wait_for_timeout(1500)
|
||||
return jd
|
||||
except Exception:
|
||||
return f"{card.title} {card.company}"
|
||||
|
||||
def _apply(self, card: JobCard, greeting: str) -> tuple[bool, str]:
|
||||
if card.can_chat and self._do_chat(card, greeting):
|
||||
return True, "chat"
|
||||
if card.can_deliver and self._do_deliver(card):
|
||||
return True, "deliver"
|
||||
return False, ""
|
||||
|
||||
def _do_chat(self, card: JobCard, greeting: str) -> bool:
|
||||
page_obj = self.manager.page
|
||||
try:
|
||||
btn = page_obj.query_selector(f"[data-job-id='{card.job_id}'] .start-chat-btn")
|
||||
if not btn:
|
||||
return False
|
||||
btn.click()
|
||||
page_obj.wait_for_timeout(2000)
|
||||
if greeting:
|
||||
inp = page_obj.query_selector(".chat-input textarea")
|
||||
if inp:
|
||||
inp.fill(greeting)
|
||||
page_obj.wait_for_timeout(500)
|
||||
send = page_obj.query_selector(".chat-input .btn-send")
|
||||
if send:
|
||||
send.click()
|
||||
page_obj.wait_for_timeout(1500)
|
||||
close = page_obj.query_selector(".dialog-container .close-btn")
|
||||
if close:
|
||||
close.click()
|
||||
return True
|
||||
except Exception as e:
|
||||
logger.debug(f"立即沟通失败:{e}")
|
||||
return False
|
||||
|
||||
def _do_deliver(self, card: JobCard) -> bool:
|
||||
page_obj = self.manager.page
|
||||
try:
|
||||
btn = page_obj.query_selector(f"[data-job-id='{card.job_id}'] .btn-deliver")
|
||||
if not btn:
|
||||
return False
|
||||
btn.click()
|
||||
page_obj.wait_for_timeout(2000)
|
||||
confirm = page_obj.query_selector(".dialog-confirm .btn-sure")
|
||||
if confirm:
|
||||
confirm.click()
|
||||
page_obj.wait_for_timeout(1500)
|
||||
return True
|
||||
except Exception as e:
|
||||
logger.debug(f"投递简历失败:{e}")
|
||||
return False
|
||||
|
||||
def _sleep(self, min_s: float, max_s: float) -> None:
|
||||
# 分段睡眠以便及时响应停止信号
|
||||
total = random.uniform(min_s, max_s)
|
||||
elapsed = 0.0
|
||||
while elapsed < total and not self._should_stop():
|
||||
time.sleep(0.5)
|
||||
elapsed += 0.5
|
||||
@@ -0,0 +1,154 @@
|
||||
"""
|
||||
浏览器管理器
|
||||
统一管理 AdsPower 浏览器启动与 Playwright CDP 连接。
|
||||
登录、抓取简历、投递共用同一个浏览器会话(复用登录态)。
|
||||
"""
|
||||
import requests
|
||||
from loguru import logger
|
||||
from playwright.sync_api import sync_playwright, Playwright, Browser, BrowserContext, Page
|
||||
|
||||
from ..config import client_settings
|
||||
|
||||
BOSS_HOME = "https://www.zhipin.com/web/geek/job"
|
||||
BOSS_RESUME_PAGE = "https://www.zhipin.com/web/geek/resume"
|
||||
|
||||
|
||||
class BrowserManager:
|
||||
"""管理 AdsPower 浏览器 + Playwright 连接"""
|
||||
|
||||
def __init__(self):
|
||||
self.api_url = client_settings.ADSPOWER_API_URL.rstrip("/")
|
||||
self.user_id = client_settings.ADSPOWER_USER_ID
|
||||
self._playwright: Playwright | None = None
|
||||
self._browser: Browser | None = None
|
||||
self._context: BrowserContext | None = None
|
||||
self.page: Page | None = None
|
||||
self.cdp_url: str | None = None
|
||||
|
||||
# ------------------------------------------------------------------
|
||||
# AdsPower 本地 API
|
||||
# ------------------------------------------------------------------
|
||||
def is_adspower_running(self) -> bool:
|
||||
try:
|
||||
resp = requests.get(f"{self.api_url}/status", timeout=5)
|
||||
return resp.status_code == 200
|
||||
except Exception:
|
||||
return False
|
||||
|
||||
def start_adspower_browser(self) -> str:
|
||||
"""启动 AdsPower 浏览器,返回 CDP 地址"""
|
||||
url = f"{self.api_url}/api/v1/browser/start"
|
||||
resp = requests.get(url, params={"user_id": self.user_id, "open_tabs": 1}, timeout=30)
|
||||
data = resp.json()
|
||||
if data.get("code") != 0:
|
||||
raise RuntimeError(f"启动 AdsPower 浏览器失败:{data.get('msg')}")
|
||||
ws = data["data"]["ws"]
|
||||
self.cdp_url = ws.get("puppeteer") or ws.get("selenium")
|
||||
logger.success(f"AdsPower 浏览器已启动,CDP: {self.cdp_url}")
|
||||
return self.cdp_url
|
||||
|
||||
def stop_adspower_browser(self) -> None:
|
||||
try:
|
||||
requests.get(
|
||||
f"{self.api_url}/api/v1/browser/stop",
|
||||
params={"user_id": self.user_id}, timeout=10,
|
||||
)
|
||||
logger.info("AdsPower 浏览器已关闭")
|
||||
except Exception as e:
|
||||
logger.warning(f"关闭 AdsPower 浏览器失败:{e}")
|
||||
|
||||
# ------------------------------------------------------------------
|
||||
# Playwright 连接
|
||||
# ------------------------------------------------------------------
|
||||
def connect(self) -> Page:
|
||||
"""启动浏览器并通过 CDP 连接,返回可操作的 page"""
|
||||
if self.page is not None:
|
||||
return self.page
|
||||
|
||||
if not self.user_id:
|
||||
raise RuntimeError("未配置 AdsPower Profile ID(ADSPOWER_USER_ID)")
|
||||
|
||||
if not self.is_adspower_running():
|
||||
raise RuntimeError("AdsPower 软件未运行,请先打开 AdsPower")
|
||||
|
||||
self.start_adspower_browser()
|
||||
self._playwright = sync_playwright().start()
|
||||
self._browser = self._playwright.chromium.connect_over_cdp(self.cdp_url)
|
||||
contexts = self._browser.contexts
|
||||
self._context = contexts[0] if contexts else self._browser.new_context()
|
||||
pages = self._context.pages
|
||||
self.page = pages[0] if pages else self._context.new_page()
|
||||
logger.info("Playwright 已连接到 AdsPower 浏览器")
|
||||
return self.page
|
||||
|
||||
def ensure_connected(self) -> Page:
|
||||
"""确保已连接,返回 page"""
|
||||
if self.page is None:
|
||||
return self.connect()
|
||||
return self.page
|
||||
|
||||
# ------------------------------------------------------------------
|
||||
# 登录状态检测
|
||||
# ------------------------------------------------------------------
|
||||
def check_boss_login(self) -> bool:
|
||||
"""
|
||||
检测 Boss直聘 是否已登录。
|
||||
通过访问首页并检查是否存在登录后才有的元素判断。
|
||||
"""
|
||||
page = self.ensure_connected()
|
||||
try:
|
||||
page.goto(BOSS_HOME, wait_until="domcontentloaded", timeout=30000)
|
||||
page.wait_for_timeout(2000)
|
||||
# 已登录:页面上会显示用户头像/用户名;未登录:会显示"登录/注册"
|
||||
logged_in_indicators = [
|
||||
".user-nav .user-info",
|
||||
".nav-figure .user",
|
||||
"[class*='avatar']",
|
||||
".header-login .name",
|
||||
]
|
||||
for sel in logged_in_indicators:
|
||||
if page.query_selector(sel):
|
||||
logger.success("检测到已登录 Boss直聘")
|
||||
return True
|
||||
# 检查是否出现登录弹窗/登录按钮
|
||||
if page.query_selector(".btn-sign-up, .login-register, [class*='login-btn']"):
|
||||
return False
|
||||
# 兜底:检查 cookie 中是否有登录凭证
|
||||
cookies = page.context.cookies()
|
||||
cookie_names = {c["name"] for c in cookies}
|
||||
return "__zp_stoken__" in cookie_names or "wt2" in cookie_names
|
||||
except Exception as e:
|
||||
logger.warning(f"登录状态检测异常:{e}")
|
||||
return False
|
||||
|
||||
def get_boss_account_name(self) -> str | None:
|
||||
"""尝试获取登录用户名"""
|
||||
page = self.ensure_connected()
|
||||
try:
|
||||
for sel in [".user-nav .name", ".header-login .name", ".user-info .name"]:
|
||||
el = page.query_selector(sel)
|
||||
if el:
|
||||
return el.inner_text().strip()
|
||||
except Exception:
|
||||
pass
|
||||
return None
|
||||
|
||||
# ------------------------------------------------------------------
|
||||
# 关闭
|
||||
# ------------------------------------------------------------------
|
||||
def close(self) -> None:
|
||||
"""关闭 Playwright 连接(不关闭 AdsPower 浏览器,保留登录态)"""
|
||||
try:
|
||||
if self._playwright:
|
||||
self._playwright.stop()
|
||||
except Exception:
|
||||
pass
|
||||
self._playwright = None
|
||||
self._browser = None
|
||||
self._context = None
|
||||
self.page = None
|
||||
|
||||
def close_all(self) -> None:
|
||||
"""关闭连接并停止 AdsPower 浏览器"""
|
||||
self.close()
|
||||
self.stop_adspower_browser()
|
||||
@@ -0,0 +1,126 @@
|
||||
"""
|
||||
AI 简历抓取模块
|
||||
从已登录的 Boss直聘 账号中抓取所有简历,
|
||||
并用 AI 整理成结构化文本后返回。
|
||||
"""
|
||||
from loguru import logger
|
||||
|
||||
from .manager import BrowserManager, BOSS_RESUME_PAGE
|
||||
|
||||
|
||||
class ResumeScraper:
|
||||
"""抓取 Boss直聘 账号内的所有简历"""
|
||||
|
||||
def __init__(self, manager: BrowserManager, ai_client=None):
|
||||
self.manager = manager
|
||||
self.ai = ai_client
|
||||
|
||||
def scrape_all(self) -> list[dict]:
|
||||
"""
|
||||
抓取所有简历。
|
||||
返回格式:[{boss_resume_id, title, position, content}, ...]
|
||||
"""
|
||||
page = self.manager.ensure_connected()
|
||||
logger.info("正在打开 Boss直聘 简历页...")
|
||||
page.goto(BOSS_RESUME_PAGE, wait_until="domcontentloaded", timeout=30000)
|
||||
page.wait_for_timeout(3000)
|
||||
|
||||
resumes: list[dict] = []
|
||||
|
||||
# 尝试多种可能的简历卡片选择器(Boss直聘 页面结构可能变化)
|
||||
card_selectors = [
|
||||
".resume-item",
|
||||
".resume-card",
|
||||
"[class*='resume-list'] li",
|
||||
".geek-resume-item",
|
||||
]
|
||||
card_els = []
|
||||
for sel in card_selectors:
|
||||
card_els = page.query_selector_all(sel)
|
||||
if card_els:
|
||||
logger.info(f"使用选择器 {sel} 找到 {len(card_els)} 份简历")
|
||||
break
|
||||
|
||||
if not card_els:
|
||||
logger.warning("未通过卡片选择器找到简历,尝试整页抓取")
|
||||
return self._scrape_whole_page(page)
|
||||
|
||||
for i, card in enumerate(card_els):
|
||||
try:
|
||||
resume = self._parse_resume_card(card, i)
|
||||
if resume:
|
||||
resumes.append(resume)
|
||||
except Exception as e:
|
||||
logger.debug(f"解析第 {i} 份简历失败:{e}")
|
||||
|
||||
logger.success(f"共抓取到 {len(resumes)} 份简历")
|
||||
return resumes
|
||||
|
||||
def _parse_resume_card(self, card, index: int) -> dict | None:
|
||||
"""解析单个简历卡片"""
|
||||
def text(sel: str) -> str:
|
||||
el = card.query_selector(sel)
|
||||
return el.inner_text().strip() if el else ""
|
||||
|
||||
title = (
|
||||
text(".resume-title") or text(".title") or
|
||||
text(".name") or f"简历 {index + 1}"
|
||||
)
|
||||
position = text(".position") or text(".job-name") or text(".expect-position")
|
||||
|
||||
# 尝试点开详情获取完整内容
|
||||
content = self._extract_detail_content(card) or text(".resume-content") or text("p")
|
||||
|
||||
# 用 AI 整理内容
|
||||
if content and self.ai:
|
||||
content = self.ai.clean_resume(content)
|
||||
|
||||
return {
|
||||
"boss_resume_id": None, # Boss直聘 前端通常不暴露,后续可从 URL 提取
|
||||
"title": title,
|
||||
"position": position or None,
|
||||
"content": content or None,
|
||||
}
|
||||
|
||||
def _extract_detail_content(self, card) -> str | None:
|
||||
"""尝试点击简历卡片进入详情页抓取完整内容"""
|
||||
page = self.manager.page
|
||||
try:
|
||||
clickable = card.query_selector("a") or card
|
||||
clickable.click()
|
||||
page.wait_for_timeout(2000)
|
||||
detail_selectors = [
|
||||
".resume-detail",
|
||||
".geek-resume-detail",
|
||||
"[class*='resume-content']",
|
||||
".resume-preview",
|
||||
]
|
||||
for sel in detail_selectors:
|
||||
el = page.query_selector(sel)
|
||||
if el:
|
||||
content = el.inner_text().strip()
|
||||
page.go_back()
|
||||
page.wait_for_timeout(1500)
|
||||
return content
|
||||
page.go_back()
|
||||
page.wait_for_timeout(1500)
|
||||
except Exception as e:
|
||||
logger.debug(f"提取简历详情失败:{e}")
|
||||
return None
|
||||
|
||||
def _scrape_whole_page(self, page) -> list[dict]:
|
||||
"""兜底方案:抓取整个简历页主体文本作为一份简历"""
|
||||
body_selectors = [".main-wrap", ".geek-wrap", "main", "#main"]
|
||||
for sel in body_selectors:
|
||||
el = page.query_selector(sel)
|
||||
if el:
|
||||
content = el.inner_text().strip()
|
||||
if content:
|
||||
content = self.ai.clean_resume(content) if self.ai else content
|
||||
return [{
|
||||
"boss_resume_id": None,
|
||||
"title": "我的在线简历",
|
||||
"position": None,
|
||||
"content": content,
|
||||
}]
|
||||
return []
|
||||
Reference in New Issue
Block a user