242 lines
9.6 KiB
Python
242 lines
9.6 KiB
Python
import time
|
||
import os
|
||
import requests
|
||
from io import BytesIO
|
||
import undetected_chromedriver as uc
|
||
|
||
# 注意：不要禁用析构函数，否则会导致Chrome进程泄漏
|
||
# if os.name == 'nt':
|
||
# try:
|
||
# uc.Chrome.__del__ = lambda self: None
|
||
# except Exception:
|
||
# pass
|
||
from selenium.webdriver.common.by import By
|
||
from selenium.webdriver.support.ui import WebDriverWait
|
||
from selenium.webdriver.support import expected_conditions as EC
|
||
from bs4 import BeautifulSoup
|
||
from reportlab.lib.pagesizes import A3
|
||
from reportlab.lib import colors
|
||
from reportlab.platypus import SimpleDocTemplate, Paragraph, Spacer, Image, PageBreak
|
||
from reportlab.lib.styles import getSampleStyleSheet
|
||
from reportlab.pdfbase.ttfonts import TTFont
|
||
from reportlab.pdfbase import pdfmetrics
|
||
from datetime import datetime
|
||
from PIL import Image as PILImage
|
||
import re
|
||
from PyPDF2 import PdfReader, PdfWriter
|
||
from loguru import logger
|
||
|
||
|
||
def download_image(url, session):
|
||
"""使用同步的 session 下载图片，确保 Cookie 一致"""
|
||
try:
|
||
if not url.lower().endswith(('.jpg', '.jpeg', '.png')):
|
||
return None
|
||
response = session.get(url, timeout=15)
|
||
response.raise_for_status()
|
||
return BytesIO(response.content)
|
||
except Exception as e:
|
||
logger.warning(f"下载图片失败: {e}")
|
||
return None
|
||
|
||
|
||
def add_pdf_encryption(pdf_file, password="4000"):
|
||
try:
|
||
pdf_writer = PdfWriter()
|
||
pdf_reader = PdfReader(pdf_file)
|
||
for page in pdf_reader.pages:
|
||
pdf_writer.add_page(page)
|
||
pdf_writer.encrypt(password)
|
||
with open(pdf_file, "wb") as output_pdf:
|
||
pdf_writer.write(output_pdf)
|
||
logger.debug("PDF加密成功")
|
||
except Exception as e:
|
||
logger.error(f"PDF加密失败: {e}")
|
||
|
||
|
||
def fetch_and_create_pdf(url):
|
||
driver = None
|
||
service = None
|
||
try:
|
||
options = uc.ChromeOptions()
|
||
# 规避检测的关键配置
|
||
# 在Linux服务器上使用headless模式
|
||
if os.name != 'nt':
|
||
options.headless = True
|
||
options.add_argument('--headless=new') # 使用新版headless模式
|
||
else:
|
||
options.headless = False
|
||
|
||
options.add_argument('--no-sandbox')
|
||
options.add_argument('--disable-gpu')
|
||
options.add_argument('--disable-dev-shm-usage')
|
||
options.add_argument('--disable-extensions')
|
||
options.add_argument('--disable-background-networking')
|
||
# 确保进程能被正确清理
|
||
options.add_argument('--disable-crash-reporter')
|
||
options.add_argument('--disable-in-process-stack-traces')
|
||
options.add_argument('--disable-logging')
|
||
options.add_argument('--disable-dev-shm-usage')
|
||
|
||
# 创建driver实例
|
||
# 让 undetected_chromedriver 自动检测浏览器版本并下载匹配的 ChromeDriver
|
||
# 强制指定版本为144，以匹配服务器当前的 Chrome 版本
|
||
driver = uc.Chrome(options=options, version_main=144)
|
||
|
||
logger.info(f"正在访问: {url}")
|
||
driver.get(url)
|
||
|
||
# 等待 Cloudflare 5秒盾结束，并处理“满18岁”按钮
|
||
time.sleep(8)
|
||
|
||
try:
|
||
enter_btn = WebDriverWait(driver, 10).until(
|
||
EC.element_to_be_clickable((By.XPATH, '//a[contains(text(), "满18岁，请点此进入")]'))
|
||
)
|
||
enter_btn.click()
|
||
logger.debug("点击了年龄确认按钮")
|
||
time.sleep(3)
|
||
except Exception:
|
||
logger.debug("未发现年龄验证按钮，可能已过检测")
|
||
|
||
# 确保列表加载
|
||
WebDriverWait(driver, 20).until(
|
||
EC.presence_of_element_located((By.CSS_SELECTOR, 'tbody[id^="normalthread"]'))
|
||
)
|
||
|
||
# 提取数据
|
||
soup = BeautifulSoup(driver.page_source, 'html.parser')
|
||
posts = [p for p in soup.find_all('tbody', {'id': lambda x: x and x.startswith('normalthread')}) if
|
||
p.find('span', {'class': 'xi1'})]
|
||
today_posts = posts[::-1]
|
||
|
||
# 字体注册
|
||
pdfmetrics.registerFont(TTFont('SimHei', 'fonts/simhei.ttf'))
|
||
styles = getSampleStyleSheet()
|
||
title_style = styles['Heading1']
|
||
title_style.fontName = 'SimHei'
|
||
title_style.textColor = colors.red
|
||
normal_style = styles['Normal']
|
||
normal_style.fontName = 'SimHei'
|
||
|
||
# 路径逻辑 - 保存到 temp/JAV 目录
|
||
base_dir = os.path.dirname(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
|
||
save_path = os.path.join(base_dir, 'temp', 'JAV')
|
||
if not os.path.exists(save_path):
|
||
os.makedirs(save_path)
|
||
pdf_filename = os.path.join(save_path, f"JAV-{datetime.now().strftime('%Y-%m-%d')}-{len(today_posts)}.pdf")
|
||
|
||
doc = SimpleDocTemplate(pdf_filename, pagesize=A3)
|
||
content = []
|
||
max_w, max_h = (A3[0] - 72) * 0.95, (A3[1] - 72) * 0.7
|
||
|
||
# 同步 Session
|
||
session = requests.Session()
|
||
ua = driver.execute_script("return navigator.userAgent")
|
||
session.headers.update({'User-Agent': ua, 'Referer': 'https://www.sehuatang.net/'})
|
||
for c in driver.get_cookies():
|
||
session.cookies.set(c['name'], c['value'])
|
||
|
||
# 循环帖子
|
||
for post in today_posts:
|
||
title_tag = post.find('a', {'class': 's xst'})
|
||
if not title_tag: continue
|
||
|
||
p_title = title_tag.get_text()
|
||
p_url = 'https://www.sehuatang.net/' + title_tag.get('href')
|
||
logger.info(f"详情页: {p_title}")
|
||
|
||
try:
|
||
resp = session.get(p_url, timeout=15)
|
||
p_soup = BeautifulSoup(resp.text, 'html.parser')
|
||
div = p_soup.find('div', {'class': 't_fsz'})
|
||
|
||
if div:
|
||
content.append(Paragraph(f" {p_title}", title_style))
|
||
magnets = re.findall(r'magnet:\?[^ \u4e00-\u9fff]+', div.get_text())
|
||
for m in magnets:
|
||
content.append(Paragraph(f"<b>{m}</b>", normal_style))
|
||
|
||
for img_tag in div.find_all('img'):
|
||
src = img_tag.get('zoomfile')
|
||
if src and 'http' in src:
|
||
img_io = download_image(src, session)
|
||
if img_io:
|
||
with PILImage.open(img_io) as p_img:
|
||
iw, ih = p_img.size
|
||
sc = min(max_w / iw, max_h / ih, 1.0)
|
||
img_io.seek(0)
|
||
content.append(Image(img_io, width=iw * sc, height=ih * sc))
|
||
|
||
if post != today_posts[-1]: content.append(PageBreak())
|
||
except Exception as e:
|
||
logger.error(f"帖子处理失败: {e}")
|
||
|
||
doc.build(content)
|
||
add_pdf_encryption(pdf_filename)
|
||
return pdf_filename
|
||
|
||
except Exception as e:
|
||
logger.exception(f"抓取异常: {e}")
|
||
return ""
|
||
finally:
|
||
# --- 确保Chrome进程被完全关闭 ---
|
||
if driver:
|
||
try:
|
||
logger.debug("正在安全关闭浏览器...")
|
||
# 先关闭所有标签页和窗口
|
||
try:
|
||
driver.close()
|
||
except Exception as e:
|
||
logger.warning(f"关闭浏览器窗口时出错: {e}")
|
||
|
||
# 强制退出所有Chrome进程
|
||
driver.quit()
|
||
logger.debug("浏览器已完全关闭")
|
||
except Exception as e:
|
||
logger.error(f"关闭浏览器时出错: {e}")
|
||
|
||
# 额外保险：强制清理残留的Chrome进程（仅Linux）
|
||
if os.name != 'nt':
|
||
try:
|
||
import psutil
|
||
current_user = os.getlogin()
|
||
for proc in psutil.process_iter(['pid', 'name', 'cmdline', 'username']):
|
||
try:
|
||
if proc.info['name'] and 'chrome' in proc.info['name'].lower():
|
||
if proc.info['username'] == current_user:
|
||
# 检查是否是本次启动的chrome进程（通过命令行参数判断）
|
||
cmdline = proc.info.get('cmdline', [])
|
||
if cmdline and any('--user-data-dir=/tmp/playwright' in str(cmd) for cmd in cmdline):
|
||
logger.info(f"强制终止残留Chrome进程: PID={proc.info['pid']}")
|
||
proc.kill()
|
||
except (psutil.NoSuchProcess, psutil.AccessDenied, psutil.ZombieProcess):
|
||
pass
|
||
except ImportError:
|
||
logger.debug("未安装psutil，跳过强制清理")
|
||
except Exception as e:
|
||
logger.warning(f"强制清理Chrome进程时出错: {e}")
|
||
|
||
|
||
def pdf_file_path_undetected():
|
||
try:
|
||
url = 'https://www.sehuatang.net/forum.php?mod=forumdisplay&fid=103&filter=typeid&typeid=481'
|
||
pdf_path = fetch_and_create_pdf(url)
|
||
if pdf_path:
|
||
logger.info(f"返回的PDF文件路径：{pdf_path}")
|
||
return True, pdf_path
|
||
else:
|
||
# 如果生成失败，返回一个默认的PDF路径
|
||
default_path = os.path.join(os.path.dirname(os.path.abspath(__file__)), "default.pdf")
|
||
logger.info(f"PDF生成失败，返回默认路径: {default_path}")
|
||
return False, default_path
|
||
except Exception as e:
|
||
logger.error(f"生成PDF路径时出错: {e}")
|
||
# 返回一个默认路径
|
||
default_path = os.path.join(os.path.dirname(os.path.abspath(__file__)), "default.pdf")
|
||
return False, default_path
|
||
|
||
|
||
if __name__ == "__main__":
|
||
pdf_file_path_undetected()
|