237899745/zsglpt-pc
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

feat: 添加依赖自动检测与安装、选项记忆、KDocs登录优化

Browse Source 
- 新增依赖检测模块：启动时自动检测wkhtmltoimage和Playwright Chromium
- 新增依赖安装对话框：缺失时提示用户一键下载安装
- 修复选项记忆功能：浏览类型、自动截图、自动上传选项现在会保存
- 优化KDocs登录检测：未登录时自动切换到金山文档页面并显示二维码
- 简化日志输出：移除debug信息，保留用户友好的状态提示
- 新增账号变化信号：账号管理页面的修改会自动同步到浏览任务页面

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
This commit is contained in:
237899745
2026-01-19 01:28:06 +08:00
parent 83fef6dff2
commit 9743186a9e
11 changed files with 713 additions and 506 deletions
Download Patch File Download Diff File Expand all files Collapse all files

36
core/api_browser.py
View File

@@ -107,10 +107,10 @@ class APIBrowser:
except Exception as e:
last_error = e
if attempt < max_retries:
self.log(f"[API] 请求超时，{retry_delay}秒后重试 ({attempt}/{max_retries})...")
self.log(f" 请求超时，{retry_delay}秒后重试 ({attempt}/{max_retries})...")
time.sleep(retry_delay)
else:
self.log(f"[API] 请求失败，已重试{max_retries}次: {str(e)}")
self.log(f" 请求失败，已重试{max_retries}次: {str(e)}")
raise last_error
@@ -125,7 +125,7 @@ class APIBrowser:
def login(self, username: str, password: str) -> bool:
"""登录"""
self.log(f"[API] 登录: {username}")
self.log(f" 登录: {username}")
self._username = username
try:
@@ -152,17 +152,17 @@ class APIBrowser:
if self.index_url_pattern in resp.url:
self.logged_in = True
self.log(f"[API] 登录成功")
self.log(f" 登录成功")
return True
else:
soup = BeautifulSoup(resp.text, "html.parser")
error = soup.find(id="lblMsg")
error_msg = error.get_text().strip() if error else "未知错误"
self.log(f"[API] 登录失败: {error_msg}")
self.log(f" 登录失败: {error_msg}")
return False
except Exception as e:
self.log(f"[API] 登录异常: {str(e)}")
self.log(f" 登录异常: {str(e)}")
return False
def get_real_name(self) -> Optional[str]:
@@ -217,10 +217,10 @@ class APIBrowser:
with open(cookies_path, "w", encoding="utf-8") as f:
f.write("\n".join(lines) + "\n")
self.log(f"[API] Cookies已保存供截图使用")
self.log(f" Cookies已保存供截图使用")
return True
except Exception as e:
self.log(f"[API] 保存cookies失败: {e}")
self.log(f" 保存cookies失败: {e}")
return False
def get_article_list_page(self, bz: int = 0, page: int = 1) -> tuple:
@@ -377,7 +377,7 @@ class APIBrowser:
# 根据浏览类型确定bz参数（网站更新后 bz=0 为应读）
bz = 0
self.log(f"[API] 开始浏览 '{browse_type}' (bz={bz})...")
self.log(f" 开始浏览 '{browse_type}' (bz={bz})...")
try:
total_items = 0
@@ -387,12 +387,12 @@ class APIBrowser:
articles, total_pages, _ = self.get_article_list_page(bz, 1)
if not articles:
self.log(f"[API] '{browse_type}' 没有待处理内容")
self.log(f" '{browse_type}' 没有待处理内容")
result.success = True
return result
total_records = self.last_total_records
self.log(f"[API] 共 {total_records} 条记录，开始处理...")
self.log(f" 共 {total_records} 条记录，开始处理...")
# 上报初始进度
if progress_callback:
@@ -404,7 +404,7 @@ class APIBrowser:
for iteration in range(max_iterations):
if should_stop_callback and should_stop_callback():
self.log("[API] 收到停止信号")
self.log(" 收到停止信号")
break
if not articles:
@@ -428,7 +428,7 @@ class APIBrowser:
try:
attachments, article_info = self.get_article_attachments(article_href)
except Exception as e:
self.log(f"[API] 获取文章失败: {title} | {str(e)}")
self.log(f" 获取文章失败: {title} | {str(e)}")
continue
total_items += 1
@@ -446,10 +446,10 @@ class APIBrowser:
for attach in attachments:
if self.mark_attachment_read(attach["id"], attach["channel_id"]):
total_attachments += 1
self.log(f"[API] [{total_items}] {title} - {len(attachments)}个附件")
self.log(f" [{total_items}] {title} - {len(attachments)}个附件")
else:
status = "已标记" if article_marked else "标记失败"
self.log(f"[API] [{total_items}] {title} - 无附件({status})")
self.log(f" [{total_items}] {title} - 无附件({status})")
# 上报进度
if progress_callback:
@@ -472,10 +472,10 @@ class APIBrowser:
if new_total_pages > 0:
total_pages = new_total_pages
except Exception as e:
self.log(f"[API] 获取第{current_page}页列表失败: {str(e)}")
self.log(f" 获取第{current_page}页列表失败: {str(e)}")
break
self.log(f"[API] 浏览完成: {total_items} 条内容，{total_attachments} 个附件")
self.log(f" 浏览完成: {total_items} 条内容，{total_attachments} 个附件")
result.success = True
result.total_items = total_items
result.total_attachments = total_attachments
@@ -483,7 +483,7 @@ class APIBrowser:
except Exception as e:
result.error_message = str(e)
self.log(f"[API] 浏览出错: {str(e)}")
self.log(f" 浏览出错: {str(e)}")
return result
def close(self):

215
core/kdocs_uploader.py
View File

@@ -41,6 +41,40 @@ class KDocsUploader:
if self._log_callback:
self._log_callback(msg)
def _find_visible_element(self, text: str, use_role: bool = False, role: str = "button"):
"""找到包含指定文本的可见元素
Args:
text: 要查找的文本
use_role: 是否使用role查找（默认False，使用文本查找）
role: 使用role查找时的角色类型
Returns:
可见的元素，或None
"""
if not self._page:
return None
try:
if use_role:
els = self._page.get_by_role(role, name=text)
else:
els = self._page.locator(f"text={text}")
count = els.count()
for i in range(count):
el = els.nth(i)
try:
if el.is_visible(timeout=500):
box = el.bounding_box()
if box and box.get('width', 0) > 0 and box.get('height', 0) > 0:
return el
except Exception:
pass
except Exception:
pass
return None
def _ensure_playwright(self, use_storage_state: bool = True) -> bool:
"""确保Playwright已启动"""
if sync_playwright is None:
@@ -53,9 +87,8 @@ class KDocsUploader:
if self._playwright is None:
self._playwright = sync_playwright().start()
if self._browser is None:
# 调试模式：有头模式，方便查看浏览器行为
# 生产环境改回 "true"
headless = os.environ.get("KDOCS_HEADLESS", "false").lower() != "false"
# 默认无头模式，设置环境变量 KDOCS_HEADLESS=false 可切换为有头模式调试
headless = os.environ.get("KDOCS_HEADLESS", "true").lower() != "false"
# 使用系统安装的Chrome浏览器（支持微信快捷登录）
# channel='chrome' 会使用系统Chrome，而不是Playwright自带的Chromium
chrome_args = [
@@ -69,10 +102,8 @@ class KDocsUploader:
channel='chrome', # 使用系统Chrome
args=chrome_args
)
self.log("[KDocs] 使用系统Chrome浏览器")
except Exception as e:
# 如果系统没有Chrome，回退到Chromium
self.log(f"[KDocs] 系统Chrome不可用({e})，使用Chromium")
self._browser = self._playwright.chromium.launch(headless=headless, args=chrome_args)
if self._context is None:
storage_state = str(KDOCS_LOGIN_STATE_FILE)
@@ -162,7 +193,6 @@ class KDocsUploader:
try:
join_btn = page.get_by_role("button", name="登录并加入编辑")
if join_btn.count() > 0 and join_btn.first.is_visible(timeout=500):
self.log("[KDocs] 点击加入编辑按钮")
join_btn.first.click()
time.sleep(1)
except Exception:
@@ -172,7 +202,6 @@ class KDocsUploader:
# 检查是否在登录页面
if self._is_login_url(url):
self.log(f"[KDocs] 检测到登录页面URL: {url}")
return True
# 只检查登录页面上的登录按钮（排除文档页面的邀请对话框）
@@ -181,7 +210,6 @@ class KDocsUploader:
try:
btn = page.get_by_role("button", name=text)
if btn.count() > 0 and btn.first.is_visible(timeout=500):
self.log(f"[KDocs] 检测到登录按钮: {text}")
return True
except Exception:
pass
@@ -198,7 +226,6 @@ class KDocsUploader:
if el.is_visible(timeout=200):
box = el.bounding_box()
if box and 80 <= box.get("width", 0) <= 400:
self.log(f"[KDocs] 检测到二维码元素: {selector}")
return True
except Exception:
pass
@@ -220,120 +247,60 @@ class KDocsUploader:
storage_state = str(KDOCS_LOGIN_STATE_FILE)
KDOCS_LOGIN_STATE_FILE.parent.mkdir(parents=True, exist_ok=True)
self._context.storage_state(path=storage_state)
self.log("[KDocs] 登录状态已保存")
except Exception as e:
self.log(f"[KDocs] 保存登录状态失败: {e}")
except Exception:
pass
def _ensure_login_dialog(self, use_quick_login: bool = False):
"""确保打开登录对话框
"""确保打开登录对话框并进入扫码页面"""
buttons_priority = [
"登录并加入编辑",
"立即登录",
"去登录",
]
Args:
use_quick_login: 是否尝试使用微信快捷登录
"""
agree_names = ["同意", "同意并继续", "我同意", "确定", "确认"]
# 循环处理登录流程
max_clicks = 8
for round_num in range(max_clicks):
max_clicks = 12
for _ in range(max_clicks):
clicked = False
current_url = self._page.url
# 检查是否已经到达文档页面（登录成功）
# 需要确保不是临时跳转，等待页面稳定
if "kdocs.cn/l/" in current_url or "www.kdocs.cn/l/" in current_url:
time.sleep(1) # 等待页面稳定
stable_url = self._page.url
if "kdocs.cn/l/" in stable_url and "account.wps.cn" not in stable_url:
self.log("[KDocs] 已到达文档页面，登录成功")
if "kdocs.cn/l/" in current_url and "account.wps.cn" not in current_url:
time.sleep(1)
has_login_elements = (
self._find_visible_element("立即登录", use_role=True) or
self._find_visible_element("登录并加入编辑", use_role=True) or
self._find_visible_element("微信扫码登录") or
self._find_visible_element("微信快捷登录")
)
if not has_login_elements:
return
# 1. 先检查是否有隐私协议同意按钮
for name in agree_names:
try:
btn = self._page.get_by_role("button", name=name)
if btn.count() > 0 and btn.first.is_visible(timeout=300):
self.log(f"[KDocs] 点击同意按钮: {name}")
btn.first.click()
time.sleep(1)
clicked = True
break
except Exception:
pass
if clicked:
continue
# 检查是否已经到达登录二维码页面
qr_page_indicators = ["微信扫码登录", "微信快捷登录"]
for indicator in qr_page_indicators:
if self._find_visible_element(indicator):
return
# 2. 如果启用快捷登录且在登录页面(account.wps.cn)，尝试点击"微信快捷登录"
if use_quick_login and "account.wps.cn" in current_url:
try:
quick_login = self._page.get_by_text("微信快捷登录", exact=False)
if quick_login.count() > 0 and quick_login.first.is_visible(timeout=500):
self.log("[KDocs] 点击微信快捷登录")
quick_login.first.click()
time.sleep(3) # 等待快捷登录处理
# 检查是否登录成功
if "kdocs.cn/l/" in self._page.url:
self.log("[KDocs] 微信快捷登录成功")
return
clicked = True
continue
except Exception:
pass
# 3. 点击"立即登录"进入登录页面
try:
btn = self._page.get_by_role("button", name="立即登录")
if btn.count() > 0 and btn.first.is_visible(timeout=500):
self.log("[KDocs] 点击立即登录")
btn.first.click()
# 按优先级点击登录按钮
for btn_name in buttons_priority:
el = self._find_visible_element(btn_name, use_role=True)
if el:
el.click(force=True)
time.sleep(2)
clicked = True
continue
except Exception:
pass
break
# 4. 点击"登录并加入编辑"（文档页面的邀请对话框）
try:
btn = self._page.get_by_role("button", name="登录并加入编辑")
if btn.count() > 0 and btn.first.is_visible(timeout=500):
self.log("[KDocs] 点击登录并加入编辑")
btn.first.click()
time.sleep(1.5)
clicked = True
continue
except Exception:
pass
# 如果没有点击到任何按钮，退出循环
if not clicked:
self.log("[KDocs] 未找到更多可点击的按钮")
break
for btn_name in buttons_priority:
el = self._find_visible_element(btn_name)
if el:
el.click(force=True)
time.sleep(2)
clicked = True
break
# 最后确保点击微信扫码登录（切换到扫码模式）
wechat_names = ["微信登录", "微信扫码登录", "扫码登录", "微信扫码"]
for name in wechat_names:
try:
btn = self._page.get_by_role("button", name=name)
if btn.is_visible(timeout=1000):
self.log(f"[KDocs] 点击微信登录: {name}")
btn.click()
time.sleep(1)
return
except Exception:
pass
# 尝试用文本查找微信登录
for name in wechat_names:
try:
el = self._page.get_by_text(name, exact=False).first
if el.is_visible(timeout=500):
self.log(f"[KDocs] 点击微信登录文本: {name}")
el.click()
time.sleep(1)
return
except Exception:
pass
self.log("[KDocs] 未找到登录按钮，可能页面已在登录状态或需要手动操作")
if not clicked:
time.sleep(1)
def _capture_qr_image(self) -> Optional[bytes]:
"""捕获登录二维码图片"""
@@ -436,31 +403,24 @@ class KDocsUploader:
return {"success": False, "error": self._last_error or "打开文档失败"}
# 检查是否已登录
self.log(f"[KDocs] 当前页面URL: {self._page.url}")
if not force and self._is_logged_in():
self._logged_in = True
self._save_login_state()
return {"success": True, "logged_in": True, "qr_image": ""}
# 需要登录，获取二维码
self.log("[KDocs] 需要登录，尝试打开登录对话框...")
self._ensure_login_dialog()
time.sleep(2) # 等待登录对话框加载
time.sleep(2)
self.log("[KDocs] 尝试捕获二维码...")
qr_image = None
for i in range(15): # 增加尝试次数
for _ in range(15):
qr_image = self._capture_qr_image()
if qr_image and len(qr_image) > 1024:
self.log(f"[KDocs] 二维码捕获成功，大小: {len(qr_image)} bytes")
break
self.log(f"[KDocs] 第{i+1}次尝试捕获二维码...")
time.sleep(1)
if not qr_image:
# 尝试截取整个页面帮助调试
self.log("[KDocs] 二维码捕获失败，当前页面可能没有显示二维码")
return {"success": False, "error": "二维码获取失败，请检查网络或手动打开金山文档链接确认"}
return {"success": False, "error": "二维码获取失败，请检查网络"}
return {
"success": True,
@@ -490,7 +450,6 @@ class KDocsUploader:
try:
confirm_btn = frame.get_by_role("button", name=name)
if confirm_btn.count() > 0 and confirm_btn.first.is_visible(timeout=200):
self.log(f"[KDocs] 找到确认按钮: {name}")
confirm_btn.first.click()
clicked_confirm = True
time.sleep(3)
@@ -504,7 +463,6 @@ class KDocsUploader:
try:
el = frame.get_by_text(name, exact=True)
if el.count() > 0 and el.first.is_visible(timeout=200):
self.log(f"[KDocs] 找到确认文本: {name}")
el.first.click()
clicked_confirm = True
time.sleep(3)
@@ -515,7 +473,6 @@ class KDocsUploader:
# 尝试用CSS选择器查找
if not clicked_confirm:
try:
# WPS登录页面的确认按钮可能的选择器
selectors = [
"button.ant-btn-primary",
"button[type='primary']",
@@ -537,7 +494,6 @@ class KDocsUploader:
if btn.is_visible(timeout=100):
btn_text = btn.inner_text() or ""
if any(kw in btn_text for kw in ["确认", "登录", "确定"]):
self.log(f"[KDocs] 找到按钮(CSS): {btn_text}")
btn.click()
clicked_confirm = True
time.sleep(3)
@@ -549,37 +505,28 @@ class KDocsUploader:
except Exception:
pass
# 如果点击了确认按钮，等待页面自动跳转（不要reload！）
if clicked_confirm:
self.log("[KDocs] 已点击确认，等待页面跳转...")
time.sleep(3) # 等待页面自动跳转
time.sleep(3)
# 检查当前URL是否已经到达文档页面
current_url = self._page.url
self.log(f"[KDocs] 当前URL: {current_url}")
# 直接检查URL判断是否已登录
if "kdocs.cn/l/" in current_url and "account.wps.cn" not in current_url:
# 已到达文档页面，登录成功
logged_in = True
# 尝试点击可能存在的"加入编辑"按钮
try:
join_btn = self._page.get_by_role("button", name="登录并加入编辑")
if join_btn.count() > 0 and join_btn.first.is_visible(timeout=500):
self.log("[KDocs] 点击加入编辑")
join_btn.first.click()
time.sleep(1)
except Exception:
pass
else:
# 还在登录页面或其他页面
logged_in = self._is_logged_in()
self._logged_in = logged_in
if logged_in:
self._save_login_state()
self.log("[KDocs] 登录状态检测：已登录")
return {"success": True, "logged_in": logged_in}
@@ -701,7 +648,6 @@ class KDocsUploader:
file_chooser.set_files(image_path)
time.sleep(2)
self.log(f"[KDocs] 图片已上传到 {cell_address}")
return True
except Exception as e:
@@ -764,7 +710,6 @@ class KDocsUploader:
pass
# 搜索姓名找到行
self.log(f"[KDocs] 搜索人员: {name}")
row_num = self._search_and_get_row(
name,
expected_col=kdocs_config.name_column,
@@ -775,8 +720,6 @@ class KDocsUploader:
if row_num < 0:
return {"success": False, "error": f"未找到人员: {name}"}
self.log(f"[KDocs] 找到人员在第 {row_num} 行")
# 上传图片
if self._upload_image_to_cell(row_num, image_path, kdocs_config.image_column):
return {"success": True}

1
core/screenshot.py
View File

@@ -312,7 +312,6 @@ def take_screenshot(
)
if success and os.path.exists(screenshot_path) and os.path.getsize(screenshot_path) > 1000:
log(f"[OK] 截图成功: {screenshot_filename}")
result.success = True
result.filename = screenshot_filename
result.filepath = screenshot_path