diff --git a/baidu_uploader.py b/baidu_uploader.py
index cee1cb3..6d175e1 100644
--- a/baidu_uploader.py
+++ b/baidu_uploader.py
@@ -1,21 +1,30 @@
 from __future__ import annotations
 
 import os
-import time
 import re
-from typing import Optional, Callable
+import time
+from typing import Callable, Optional
+
+from playwright.sync_api import TimeoutError as PlaywrightTimeoutError
 from playwright.sync_api import sync_playwright
 
+
 class BaiduUploader:
     """
     百度网盘上传工具 - 纯净上传版 (移除分享逻辑)
     """
-    def __init__(self, chrome_path: str, cookies_dir: str, log_callback: Optional[Callable[[str], None]] = None):
+
+    def __init__(
+        self,
+        chrome_path: str,
+        cookies_dir: str,
+        log_callback: Optional[Callable[[str], None]] = None,
+    ):
         self.chrome_path = chrome_path
         self.cookies_dir = cookies_dir
         self.log_callback = log_callback
         self.url = "https://pan.baidu.com/"
-        
+
         if not os.path.exists(self.cookies_dir):
             os.makedirs(self.cookies_dir, exist_ok=True)
 
@@ -24,7 +33,12 @@ class BaiduUploader:
         if self.log_callback:
             self.log_callback(message)
 
-    def upload_file(self, file_path: str, target_folder_name: str, root_path: str = "精品项目整理") -> bool:
+    def upload_file(
+        self,
+        file_path: str,
+        target_folder_name: str,
+        root_path: str = "精品项目整理",
+    ) -> bool:
         """
         上传文件到百度网盘
         """
@@ -33,145 +47,778 @@ class BaiduUploader:
             return False
 
         filename = os.path.basename(file_path)
-        
+        file_size_mb = os.path.getsize(file_path) / 1024 / 1024
+        context = None
+
         # 提取项目编号
         project_id = ""
-        match = re.search(r'(【[A-Za-z0-9]+】)', target_folder_name)
+        match = re.search(r"(【[A-Za-z0-9]+】)", target_folder_name)
         if match:
             project_id = match.group(1)
 
         try:
             with sync_playwright() as p:
-                self._log(f"正在启动浏览器...")
-                
+                self._log("正在启动浏览器...")
+
                 launch_args = {
                     "user_data_dir": self.cookies_dir,
                     "headless": False,
-                    "executable_path": self.chrome_path if self.chrome_path and os.path.exists(self.chrome_path) else None,
-                    "viewport": {"width": 1280, "height": 800}
+                    "executable_path": self.chrome_path
+                    if self.chrome_path and os.path.exists(self.chrome_path)
+                    else None,
+                    "viewport": {"width": 1280, "height": 800},
                 }
                 if not launch_args["executable_path"]:
                     launch_args.pop("executable_path")
 
                 context = p.chromium.launch_persistent_context(**launch_args)
-                context.add_init_script("Object.defineProperty(navigator, 'webdriver', {get: () => undefined})")
-                
+                context.add_init_script(
+                    "Object.defineProperty(navigator, 'webdriver', {get: () => undefined})"
+                )
+
                 page = context.new_page()
                 page.set_default_timeout(60000)
-                
+
                 # 1. 登录
                 self._log("检测登录状态...")
-                page.goto(self.url, wait_until="domcontentloaded")
-                
-                if not self._wait_for_login(page):
-                    self._log("请手动扫码登录...")
-                    page.wait_for_selector("text=全部文件", timeout=180000)
-                
+                page.goto(self.url, wait_until="domcontentloaded", timeout=60000)
+
+                if not self._wait_for_login(page, timeout=20):
+                    self._log("请手动扫码登录，登录后会自动继续...")
+                    if not self._wait_for_login(page, timeout=180):
+                        raise RuntimeError("等待百度网盘登录超时")
+
                 # 2. 导航
                 self._log("登录成功，进入目标路径...")
-                page.get_by_text("全部文件").first.click(force=True)
-                time.sleep(4) 
-                
-                if not self._scan_and_enter(page, root_path):
-                    self._create_folder(page, root_path)
-                    time.sleep(2); self._scan_and_enter(page, root_path)
-                
+                self._open_all_files(page)
+
+                self._log("等待文件列表加载...")
+                if not self._wait_for_file_list_loaded(page):
+                    self._log("警告：文件列表加载超时，继续尝试目录检查...")
+
+                self._log(f"准备进入根目录: {root_path}")
+                if not self._ensure_folder(page, root_path):
+                    raise RuntimeError(f"无法进入或创建根目录: {root_path}")
+
                 search_key = project_id if project_id else target_folder_name
-                if not self._scan_and_enter(page, search_key, is_fuzzy=True):
-                    self._create_folder(page, target_folder_name)
-                    time.sleep(2); self._scan_and_enter(page, search_key, is_fuzzy=True)
-                
+                self._log(f"准备进入项目目录: {target_folder_name}")
+                if not self._ensure_folder(
+                    page,
+                    search_key,
+                    create_name=target_folder_name,
+                    is_fuzzy=bool(project_id),
+                ):
+                    raise RuntimeError(f"无法进入或创建项目目录: {target_folder_name}")
+
                 # 3. 查重
                 if self._check_file_exists_robust(page, filename):
                     self._log(f"云端已存在文件 '{filename}'，跳过上传。")
                     time.sleep(2)
-                    context.close()
                     return True
-                
+
                 # 4. 执行上传
-                self._log(f"开始传输: {filename}")
-                page.set_input_files("input[type=file]", file_path)
-                
+                self._log(f"开始传输: {filename} ({file_size_mb:.2f} MB)")
+                self._clear_old_notifications(page)
+                self._select_upload_file(page, file_path)
+
+                if not self._wait_for_upload_started(page, filename):
+                    self._log("警告：未检测到明显的上传任务，继续监测文件列表。")
+                else:
+                    self._log("检测到上传任务已建立。")
+
                 # 5. 监测列表确认成功
                 self._log("上传已启动，监测云端状态中...")
-                success = self._wait_for_file_in_list(page, filename)
-                
+                success = self._wait_for_upload_finished(page, filename)
+
                 if success:
-                    self._log("\n" + "="*30)
+                    self._log("\n" + "=" * 30)
                     self._log("百度网盘上传圆满成功！")
-                    self._log("="*30 + "\n")
+                    self._log("=" * 30 + "\n")
                     time.sleep(3)
                 else:
                     self._log("警告：监测超时，未能在列表中看到文件，请手动核实。")
 
-                context.close()
                 return success
 
         except Exception as e:
             self._log(f"程序异常: {str(e)}")
             return False
+        finally:
+            if context:
+                try:
+                    context.close()
+                except Exception:
+                    pass
 
-    def _wait_for_login(self, page) -> bool:
-        for _ in range(10):
-            if page.get_by_text("全部文件").count() > 0: return True
-            time.sleep(2)
+    def _wait_for_login(self, page, timeout: float = 20) -> bool:
+        start_time = time.time()
+        while time.time() - start_time < timeout:
+            try:
+                body_text = self._body_text(page)
+                if "全部文件" in body_text or "我的文件" in body_text:
+                    return True
+            except Exception:
+                pass
+            time.sleep(1)
+        return False
+
+    def _open_all_files(self, page):
+        if self._click_text(page, ["全部文件", "我的文件"]):
+            time.sleep(1)
+            return
+        self._log("未找到“全部文件”入口，尝试直接在当前页面继续。")
+
+    def _wait_for_file_list_loaded(
+        self,
+        page,
+        timeout: float = 8,
+        poll_interval: float = 0.3,
+    ) -> bool:
+        """
+        等待目录页真正进入可操作状态。
+        百度网盘目录可能为空，所以不能只靠“列表里有项目”来判断。
+        """
+        start_time = time.time()
+        while time.time() - start_time < timeout:
+            try:
+                state = page.evaluate(
+                    """
+                    () => {
+                        const bodyText = document.body ? document.body.innerText : '';
+                        const itemCount = document.querySelectorAll(
+                            'span[title], a[title], .wp-s-file-list-drag-copy__item-title-text, .file-name'
+                        ).length;
+                        const breadcrumbCount = document.querySelectorAll(
+                            '.wp-s-file-list-drag-copy__header-breadcrumb-item'
+                        ).length;
+                        const hasCreateFolderButton = bodyText.includes('新建文件夹');
+                        const hasAllFilesText = bodyText.includes('全部文件');
+                        const hasEmptyState = ['暂无文件', '空文件夹', '拖拽文件到此处上传', '上传文件到百度网盘']
+                            .some(keyword => bodyText.includes(keyword));
+                        const hasLoadingText = ['加载中', '正在加载', '请稍候']
+                            .some(keyword => bodyText.includes(keyword));
+                        const hasLoadingMask = !!document.querySelector(
+                            '[class*="loading"], [class*="Loading"], [class*="skeleton"], [class*="Skeleton"], [class*="spin"], [class*="Spin"]'
+                        );
+                        return {
+                            itemCount,
+                            breadcrumbCount,
+                            hasCreateFolderButton,
+                            hasAllFilesText,
+                            hasEmptyState,
+                            hasLoadingText,
+                            hasLoadingMask,
+                        };
+                    }
+                    """
+                )
+
+                ready = (
+                    (state["breadcrumbCount"] > 0 or state["hasAllFilesText"])
+                    and state["hasCreateFolderButton"]
+                    and (
+                        state["itemCount"] > 0
+                        or state["hasEmptyState"]
+                        or (not state["hasLoadingText"] and not state["hasLoadingMask"])
+                    )
+                )
+                if ready:
+                    return True
+            except Exception:
+                pass
+            time.sleep(poll_interval)
+        return False
+
+    def _find_matching_item(
+        self,
+        page,
+        target_name: str,
+        is_fuzzy: bool = False,
+    ) -> Optional[str]:
+        try:
+            return page.evaluate(
+                """
+                (args) => {
+                    const { name, fuzzy } = args;
+                    const normalize = value => (value || '').replace(/\\s+/g, ' ').trim();
+                    const visible = element => {
+                        if (!element) return false;
+                        const style = window.getComputedStyle(element);
+                        const rect = element.getBoundingClientRect();
+                        return style.display !== 'none'
+                            && style.visibility !== 'hidden'
+                            && rect.width > 0
+                            && rect.height > 0;
+                    };
+                    const els = [
+                        ...document.querySelectorAll(
+                            'span[title], a[title], [aria-label], .wp-s-file-list-drag-copy__item-title-text, .file-name, [class*="name"], [class*="Name"], [class*="file"], [class*="File"]'
+                        ),
+                    ].filter(visible);
+                    const target = els.find(e => {
+                        const values = [
+                            normalize(e.getAttribute('title')),
+                            normalize(e.getAttribute('aria-label')),
+                            normalize(e.innerText || e.textContent),
+                        ].filter(Boolean);
+                        if (!values.length) {
+                            return false;
+                        }
+                        return values.some(value => fuzzy ? value.includes(name) : value === name);
+                    });
+                    if (!target) {
+                        return null;
+                    }
+                    return normalize(
+                        target.getAttribute('title')
+                        || target.getAttribute('aria-label')
+                        || target.innerText
+                        || target.textContent
+                    );
+                }
+                """,
+                {"name": target_name, "fuzzy": is_fuzzy},
+            )
+        except Exception:
+            return None
+
+    def _matches_name(self, source: str, target_name: str, is_fuzzy: bool = False) -> bool:
+        source = (source or "").strip()
+        if not source:
+            return False
+        return target_name in source if is_fuzzy else source == target_name
+
+    def _page_contains_text(self, page, text: str) -> bool:
+        try:
+            return bool(
+                page.evaluate(
+                    """
+                    (target) => {
+                        const normalize = value => (value || '').replace(/\\s+/g, ' ').trim();
+                        const bodyText = normalize(document.body ? document.body.innerText : '');
+                        return !!target && bodyText.includes(normalize(target));
+                    }
+                    """,
+                    text,
+                )
+            )
+        except Exception:
+            return False
+
+    def _get_navigation_context(self, page) -> dict:
+        try:
+            return page.evaluate(
+                """
+                () => {
+                    const normalize = text => (text || '').replace(/\\s+/g, ' ').trim();
+                    const breadcrumbs = [
+                        ...document.querySelectorAll(
+                            '.wp-s-file-list-drag-copy__header-breadcrumb-item, [class*="breadcrumb"] a, [class*="breadcrumb"] span, [class*="Breadcrumb"] a, [class*="Breadcrumb"] span'
+                        ),
+                    ]
+                        .map(node => normalize(node.getAttribute('title') || node.innerText || node.textContent))
+                        .filter(Boolean);
+                    const itemTitles = [
+                        ...document.querySelectorAll(
+                            'span[title], a[title], [aria-label], .wp-s-file-list-drag-copy__item-title-text, .file-name, [class*="name"], [class*="Name"]'
+                        ),
+                    ]
+                        .map(node => normalize(
+                            node.getAttribute('title')
+                            || node.getAttribute('aria-label')
+                            || node.innerText
+                            || node.textContent
+                        ))
+                        .filter(Boolean);
+                    return {
+                        url: window.location.href,
+                        breadcrumbs,
+                        itemTitles,
+                    };
+                }
+                """
+            )
+        except Exception:
+            return {
+                "url": "",
+                "breadcrumbs": [],
+                "itemTitles": [],
+            }
+
+    def _wait_for_folder_visible(
+        self,
+        page,
+        target_name: str,
+        is_fuzzy: bool = False,
+        exact_name: Optional[str] = None,
+        timeout: float = 8,
+        poll_interval: float = 0.3,
+    ) -> bool:
+        start_time = time.time()
+        while time.time() - start_time < timeout:
+            if exact_name and self._find_matching_item(page, exact_name, False):
+                return True
+            if self._find_matching_item(page, target_name, is_fuzzy):
+                return True
+            if exact_name and self._page_contains_text(page, exact_name):
+                return True
+            time.sleep(poll_interval)
+        return False
+
+    def _wait_for_folder_entered(
+        self,
+        page,
+        target_name: str,
+        is_fuzzy: bool = False,
+        timeout: float = 8,
+        before_url: str = "",
+        before_had_target: bool = False,
+        poll_interval: float = 0.25,
+    ) -> bool:
+        start_time = time.time()
+        while time.time() - start_time < timeout:
+            context = self._get_navigation_context(page)
+            breadcrumbs = context.get("breadcrumbs", [])
+            current_url = context.get("url", "")
+            item_titles = context.get("itemTitles", [])
+
+            breadcrumb_match = next(
+                (crumb for crumb in breadcrumbs if self._matches_name(crumb, target_name, is_fuzzy)),
+                "",
+            )
+            url_changed = bool(before_url and current_url and current_url != before_url)
+            list_no_longer_shows_target = not any(
+                self._matches_name(title, target_name, is_fuzzy) for title in item_titles
+            )
+
+            if breadcrumb_match or url_changed or (before_had_target and list_no_longer_shows_target):
+                if not self._wait_for_file_list_loaded(page, timeout=4, poll_interval=0.2):
+                    display_name = breadcrumb_match or target_name
+                    self._log(f"警告：已进入目录 {display_name}，但列表就绪确认超时，继续后续流程。")
+                return True
+            time.sleep(poll_interval)
+
+        context = self._get_navigation_context(page)
+        breadcrumbs = " > ".join(context.get("breadcrumbs", [])) or "无"
+        current_url = context.get("url", "") or "无"
+        self._log(
+            f"进入目录确认超时: 目标={target_name}, 当前面包屑={breadcrumbs}, 当前URL={current_url}"
+        )
+        return False
+
+    def _ensure_folder(
+        self,
+        page,
+        target_name: str,
+        create_name: Optional[str] = None,
+        is_fuzzy: bool = False,
+    ) -> bool:
+        create_name = create_name or target_name
+
+        if not self._wait_for_file_list_loaded(page, timeout=5, poll_interval=0.2):
+            self._log("警告：当前目录列表未完全确认加载，但仍继续检查目录。")
+
+        matched_name = self._find_matching_item(page, target_name, is_fuzzy)
+        if matched_name:
+            self._log(f"已找到目录: {matched_name}，准备进入...")
+            return self._wait_and_enter_folder(
+                page,
+                target_name,
+                is_fuzzy=is_fuzzy,
+                exact_name=matched_name,
+                timeout=18,
+            )
+
+        self._log(f"未找到目录，开始创建: {create_name}")
+        if not self._create_folder(page, create_name):
+            return False
+
+        if not self._wait_and_enter_folder(
+            page,
+            target_name,
+            is_fuzzy=is_fuzzy,
+            exact_name=create_name,
+            timeout=35,
+        ):
+            self._log(f"目录创建后仍未能进入: {create_name}")
+            return False
+
+        return True
+
+    def _wait_and_enter_folder(
+        self,
+        page,
+        target_name: str,
+        is_fuzzy: bool = False,
+        exact_name: Optional[str] = None,
+        timeout: float = 30,
+        poll_interval: float = 1,
+    ) -> bool:
+        start_time = time.time()
+        attempt = 0
+
+        while time.time() - start_time < timeout:
+            attempt += 1
+
+            if exact_name and self._find_matching_item(page, exact_name, False):
+                self._log(f"确认目录已出现: {exact_name}，尝试进入...")
+                if self._scan_and_enter(page, exact_name, is_fuzzy=False):
+                    return True
+
+            matched_name = self._find_matching_item(page, target_name, is_fuzzy)
+            if matched_name:
+                self._log(f"确认目录已出现: {matched_name}，尝试进入...")
+                if self._scan_and_enter(page, target_name, is_fuzzy=is_fuzzy):
+                    return True
+
+            if attempt % 5 == 0:
+                self._log(f"目录暂未稳定可进入，刷新当前目录后重试: {exact_name or target_name}")
+                self._refresh_current_folder(page)
+
+            time.sleep(poll_interval)
+
+        context = self._get_navigation_context(page)
+        item_titles = " | ".join(context.get("itemTitles", [])[:20]) or "无"
+        self._log(f"目录确认进入失败: {exact_name or target_name}，当前列表: {item_titles}")
         return False
 
     def _scan_and_enter(self, page, target_name: str, is_fuzzy: bool = False) -> bool:
         try:
             page.evaluate("window.scrollTo(0, 300)")
-            found = page.evaluate(f"""
-                (args) => {{
-                    const {{ name, fuzzy }} = args;
-                    const els = [...document.querySelectorAll('span[title], a[title], .wp-s-file-list-drag-copy__item-title-text')];
-                    const target = els.find(e => {{
-                        const title = e.getAttribute('title') || e.innerText.trim();
-                        return fuzzy ? title.includes(name) : title === name;
-                    }});
-                    if (target) {{
-                        target.click();
-                        const ev = new MouseEvent('dblclick', {{ 'view': window, 'bubbles': true, 'cancelable': true }});
-                        target.dispatchEvent(ev);
+            before_context = self._get_navigation_context(page)
+            before_url = before_context.get("url", "")
+            before_item_titles = before_context.get("itemTitles", [])
+            before_had_target = any(
+                self._matches_name(title, target_name, is_fuzzy) for title in before_item_titles
+            )
+            found = page.evaluate(
+                """
+                (args) => {
+                    const { name, fuzzy } = args;
+                    const normalize = value => (value || '').replace(/\\s+/g, ' ').trim();
+                    const visible = element => {
+                        if (!element) return false;
+                        const style = window.getComputedStyle(element);
+                        const rect = element.getBoundingClientRect();
+                        return style.display !== 'none'
+                            && style.visibility !== 'hidden'
+                            && rect.width > 0
+                            && rect.height > 0;
+                    };
+                    const els = [
+                        ...document.querySelectorAll(
+                            'span[title], a[title], [aria-label], .wp-s-file-list-drag-copy__item-title-text, .file-name, [class*="name"], [class*="Name"]'
+                        )
+                    ].filter(visible);
+                    const target = els.find(e => {
+                        const values = [
+                            normalize(e.getAttribute('title')),
+                            normalize(e.getAttribute('aria-label')),
+                            normalize(e.innerText || e.textContent),
+                        ].filter(Boolean);
+                        return values.some(value => fuzzy ? value.includes(name) : value === name);
+                    });
+                    if (target) {
+                        const row = target.closest('tr, li, [class*="item"], [class*="Item"], [class*="row"], [class*="Row"], [class*="file"], [class*="File"]') || target;
+                        row.scrollIntoView({block: 'center'});
+                        row.click();
+                        const ev = new MouseEvent('dblclick', { view: window, bubbles: true, cancelable: true });
+                        row.dispatchEvent(ev);
                         return true;
-                    }}
+                    }
                     return false;
-                }}
-            """, {"name": target_name, "fuzzy": is_fuzzy})
-            if found: time.sleep(3); return True
+                }
+                """,
+                {"name": target_name, "fuzzy": is_fuzzy},
+            )
+            if not found:
+                try:
+                    locator = page.get_by_text(target_name, exact=not is_fuzzy).first
+                    if locator.count() == 0:
+                        return False
+                    locator.click(force=True)
+                    locator.dblclick(force=True)
+                    found = True
+                except Exception:
+                    return False
+            time.sleep(0.2)
+            return self._wait_for_folder_entered(
+                page,
+                target_name,
+                is_fuzzy=is_fuzzy,
+                before_url=before_url,
+                before_had_target=before_had_target,
+                poll_interval=0.2,
+            )
+        except Exception:
             return False
-        except: return False
 
-    def _create_folder(self, page, folder_name: str):
+    def _create_folder(self, page, folder_name: str) -> bool:
         try:
             page.keyboard.press("Escape")
             btn = page.get_by_text("新建文件夹").first
-            if btn.count() > 0:
-                btn.click(force=True)
-                time.sleep(1.5)
-                page.keyboard.type(folder_name, delay=50)
-                page.keyboard.press("Enter")
-                time.sleep(3)
-        except: pass
+            if btn.count() == 0:
+                self._log("未找到“新建文件夹”按钮")
+                return False
+
+            btn.click(force=True)
+            time.sleep(0.8)
+            page.keyboard.type(folder_name, delay=50)
+            page.keyboard.press("Enter")
+
+            if self._wait_for_folder_visible(
+                page,
+                folder_name,
+                exact_name=folder_name,
+                timeout=8,
+                poll_interval=0.2,
+            ):
+                self._log(f"文件夹创建完成: {folder_name}")
+                return True
+
+            self._log(f"文件夹创建后未立即在列表中看到，刷新后继续确认: {folder_name}")
+            self._refresh_current_folder(page)
+            if self._wait_for_folder_visible(
+                page,
+                folder_name,
+                exact_name=folder_name,
+                timeout=8,
+                poll_interval=0.2,
+            ):
+                self._log(f"文件夹创建完成: {folder_name}")
+                return True
+
+            self._log(f"已提交文件夹创建请求，交给后续流程继续确认: {folder_name}")
+            return True
+        except Exception as exc:
+            self._log(f"创建文件夹失败: {folder_name}，原因: {exc}")
+            return False
 
     def _check_file_exists_robust(self, page, filename: str) -> bool:
-        return page.evaluate(f"""
-            (name) => {{
-                const els = [...document.querySelectorAll('span[title], a[title], .wp-s-file-list-drag-copy__item-title-text')];
-                return els.some(e => (e.getAttribute('title') || e.innerText.trim()) === name);
-            }}
-        """, filename)
+        try:
+            return bool(
+                page.evaluate(
+                    """
+                    (name) => {
+                        const normalize = value => (value || '').replace(/\\s+/g, ' ').trim();
+                        const visible = element => {
+                            if (!element) return false;
+                            const style = window.getComputedStyle(element);
+                            const rect = element.getBoundingClientRect();
+                            return style.display !== 'none'
+                                && style.visibility !== 'hidden'
+                                && rect.width > 0
+                                && rect.height > 0;
+                        };
+                        const els = [
+                            ...document.querySelectorAll(
+                                'span[title], a[title], [aria-label], .wp-s-file-list-drag-copy__item-title-text, .file-name'
+                            ),
+                        ].filter(visible);
+                        return els.some(e => {
+                            const title = normalize(e.getAttribute('title'));
+                            const label = normalize(e.getAttribute('aria-label'));
+                            const text = normalize(e.innerText || e.textContent);
+                            return title === name || label === name || text === name;
+                        });
+                    }
+                    """,
+                    filename,
+                )
+            )
+        except Exception:
+            try:
+                locator = page.get_by_text(filename, exact=True)
+                count = locator.count()
+                for index in range(min(count, 5)):
+                    if locator.nth(index).is_visible():
+                        return True
+            except Exception:
+                pass
+        return False
+
+    def _select_upload_file(self, page, file_path: str):
+        input_locator = page.locator("input[type=file]")
+        if input_locator.count() == 0:
+            if not self._click_text(page, ["上传", "上传文件"]):
+                self._log("未找到“上传”按钮，继续等待上传控件。")
+            page.wait_for_selector("input[type=file]", timeout=10000)
+
+        page.locator("input[type=file]").first.set_input_files(file_path)
+
+    def _wait_for_upload_started(
+        self,
+        page,
+        filename: str,
+        timeout: float = 20,
+    ) -> bool:
+        start_time = time.time()
+        while time.time() - start_time < timeout:
+            state = self._upload_state(page, filename)
+            if state["uploading"] or state["item_active"] or state["file_exists"]:
+                return True
+            time.sleep(0.5)
+        return False
+
+    def _wait_for_upload_finished(
+        self,
+        page,
+        filename: str,
+        timeout: float = 1800,
+    ) -> bool:
+        start_time = time.time()
+        stable_count = 0
+        last_active_log = 0.0
+        last_refresh = time.time()
+
+        while time.time() - start_time < timeout:
+            state = self._upload_state(page, filename)
+
+            if state["has_error"]:
+                self._log("检测到页面提示上传失败。")
+                return False
+
+            if state["file_exists"] and not state["item_active"] and not state["uploading"]:
+                stable_count += 1
+                if stable_count >= 3:
+                    self._log("文件已出现在列表中，且上传状态连续稳定。")
+                    time.sleep(3)
+                    return self._check_file_exists_robust(page, filename)
+
+                self._log(f"上传疑似完成，稳定性校验 ({stable_count}/3)...")
+            else:
+                stable_count = 0
+                now = time.time()
+                if (state["uploading"] or state["item_active"]) and now - last_active_log >= 15:
+                    self._log("监测到活跃传输流，继续等待...")
+                    last_active_log = now
+
+            if time.time() - last_refresh > 25:
+                self._refresh_current_folder(page)
+                last_refresh = time.time()
+
+            time.sleep(3)
+
+        return False
+
+    def _upload_state(self, page, filename: str) -> dict:
+        try:
+            state = page.evaluate(
+                """
+                (filename) => {
+                    const normalize = value => (value || '').replace(/\\s+/g, ' ').trim();
+                    const visible = element => {
+                        if (!element) return false;
+                        const style = window.getComputedStyle(element);
+                        const rect = element.getBoundingClientRect();
+                        return style.display !== 'none'
+                            && style.visibility !== 'hidden'
+                            && rect.width > 0
+                            && rect.height > 0;
+                    };
+                    const bodyText = normalize(document.body ? document.body.innerText : '');
+                    const nodes = [
+                        ...document.querySelectorAll(
+                            'span[title], a[title], [aria-label], .wp-s-file-list-drag-copy__item-title-text, .file-name, [class*="file"], [class*="File"]'
+                        )
+                    ].filter(visible);
+
+                    const matchingNode = nodes.find(node => {
+                        const title = normalize(node.getAttribute('title'));
+                        const label = normalize(node.getAttribute('aria-label'));
+                        const text = normalize(node.innerText || node.textContent);
+                        return title === filename || label === filename || text === filename;
+                    });
+
+                    const container = matchingNode
+                        ? matchingNode.closest('tr, li, [class*="item"], [class*="Item"], [class*="row"], [class*="Row"], [class*="file"], [class*="File"]')
+                        : null;
+                    const containerText = normalize(container ? container.innerText : '');
+                    const itemActiveWords = ['正在上传', '上传中', '等待上传', '传输队列', '传输中', '0B/s', '校验中', '处理中'];
+                    const pageActiveWords = ['正在上传', '上传中', '等待上传', '传输队列', '传输中', '校验中', '处理中'];
+                    const errorWords = ['上传失败', '传输失败', '网络异常', '上传出错', '文件上传失败'];
+                    const progressPattern = /(?:^|\\D)(?:100|\\d{1,2})(?:\\.\\d+)?%/;
+                    const itemHasProgress = container
+                        ? [...container.querySelectorAll('[class*="progress"], [class*="Progress"]')].some(visible)
+                        : false;
+
+                    const itemActive = itemActiveWords.some(word => containerText.includes(word))
+                        || progressPattern.test(containerText)
+                        || itemHasProgress;
+                    const uploading = itemActive
+                        || (bodyText.includes(filename)
+                            && (pageActiveWords.some(word => bodyText.includes(word))
+                                || progressPattern.test(bodyText)));
+                    const hasError = errorWords.some(word => bodyText.includes(word));
+
+                    return {
+                        file_exists: Boolean(matchingNode),
+                        item_active: itemActive,
+                        uploading,
+                        has_error: hasError,
+                    };
+                }
+                """,
+                filename,
+            )
+            return {
+                "file_exists": bool(state.get("file_exists")),
+                "item_active": bool(state.get("item_active")),
+                "uploading": bool(state.get("uploading")),
+                "has_error": bool(state.get("has_error")),
+            }
+        except Exception:
+            return {
+                "file_exists": self._check_file_exists_robust(page, filename),
+                "item_active": False,
+                "uploading": False,
+                "has_error": False,
+            }
+
+    def _refresh_current_folder(self, page):
+        try:
+            page.locator(".wp-s-file-list-drag-copy__header-breadcrumb-item").last.click()
+            time.sleep(2)
+        except Exception:
+            try:
+                page.keyboard.press("F5")
+                page.wait_for_load_state("domcontentloaded", timeout=10000)
+            except Exception:
+                pass
+
+    def _click_text(self, page, texts: list[str]) -> bool:
+        for text in texts:
+            for exact in (True, False):
+                try:
+                    locator = page.get_by_text(text, exact=exact)
+                    count = locator.count()
+                    for index in range(min(count, 5)):
+                        item = locator.nth(index)
+                        if item.is_visible():
+                            item.click(force=True, timeout=5000)
+                            return True
+                except PlaywrightTimeoutError:
+                    continue
+                except Exception:
+                    continue
+        return False
+
+    def _clear_old_notifications(self, page):
+        try:
+            page.evaluate(
+                """
+                () => {
+                    document
+                        .querySelectorAll('.ant-message-notice, .u-toast, [class*="toast"], [class*="Toast"]')
+                        .forEach(element => element.remove());
+                }
+                """
+            )
+        except Exception:
+            pass
+
+    def _body_text(self, page) -> str:
+        return page.evaluate("() => document.body ? document.body.innerText : ''")
 
     def _wait_for_file_in_list(self, page, filename: str) -> bool:
-        start_time = time.time(); last_refresh = time.time()
-        while time.time() - start_time < 1800:
-            if self._check_file_exists_robust(page, filename):
-                content = page.content()
-                # 如果文件在列表里，且页面内容没在报错或重试
-                if not any(x in content for x in ["正在上传", "传输队列", "0B/s"]):
-                    return True
-            if time.time() - last_refresh > 25:
-                try: page.locator(".wp-s-file-list-drag-copy__header-breadcrumb-item").last.click(); time.sleep(3)
-                except: pass
-                last_refresh = time.time()
-            time.sleep(5)
-        return False
+        return self._wait_for_upload_finished(page, filename)
diff --git a/gui.py b/gui.py
index dbc4e90..0fb3c2c 100644
--- a/gui.py
+++ b/gui.py
@@ -962,6 +962,8 @@ class YidaimaGUI:
         upload_frame.pack(fill=tk.X)
 
         ttk.Button(upload_frame, text="整理项目", command=self._ps_organize_project, width=15).pack(side=tk.LEFT, padx=(0, 5))
+        ttk.Button(upload_frame, text="整理并上传夸克", command=self._ps_organize_upload_cloud, width=18).pack(side=tk.LEFT, padx=(5, 5))
+        ttk.Button(upload_frame, text="整理并上传百度", command=self._ps_organize_upload_baidu, width=18).pack(side=tk.LEFT, padx=(5, 5))
         ttk.Button(upload_frame, text="上传夸克网盘", command=self._ps_upload_quark, width=15).pack(side=tk.LEFT, padx=(5, 5))
         ttk.Button(upload_frame, text="上传百度网盘", command=self._ps_upload_baidu, width=15).pack(side=tk.LEFT, padx=(5, 0))
 
@@ -1577,6 +1579,69 @@ class YidaimaGUI:
 
         threading.Thread(target=run, daemon=True).start()
 
+    def _ps_organize_upload_baidu(self):
+        """整理项目并上传到百度网盘"""
+        def run():
+            try:
+                self._ps_set_running(True)
+                self._ps_log("=" * 50)
+                self._ps_log("开始整理项目并打包...")
+                config = self._get_ps_config()
+                automation = ProjectScreenshotAutomation(config, log_callback=self._ps_log)
+
+                project_name = self.ps_project_name_var.get().strip()
+                if not project_name:
+                    project_name = None
+                else:
+                    self._ps_log(f"使用项目名称: {project_name}")
+
+                zip_file = automation.organize_and_zip_project(override_name=project_name)
+
+                if zip_file:
+                    self._ps_log("=" * 50)
+                    self._ps_log(f"项目整理完成！压缩包已生成：{zip_file}")
+                    self._ps_log("\n开始准备上传到百度网盘...")
+
+                    chrome_path = self.config.get("chrome.path", "")
+                    baidu_cookies_dir = self.config.get("baidu.cookies_dir", os.path.join(os.getcwd(), "data", "baidu_cookies"))
+                    baidu_root_path = self.config.get("baidu.root_path", "精品项目整理")
+
+                    if not project_name:
+                        project_name = os.path.basename(zip_file).replace(".zip", "")
+
+                    uploader = BaiduUploader(
+                        chrome_path=chrome_path,
+                        cookies_dir=baidu_cookies_dir,
+                        log_callback=self._ps_log
+                    )
+
+                    success = uploader.upload_file(
+                        file_path=zip_file,
+                        target_folder_name=project_name,
+                        root_path=baidu_root_path
+                    )
+
+                    if success:
+                        self._ps_log("=" * 50)
+                        self._ps_log("项目已成功整理并上传到百度网盘！")
+                        self.root.after(0, lambda: messagebox.showinfo("成功", "项目整理并上传百度网盘成功！"))
+                    else:
+                        self._ps_log("=" * 50)
+                        self._ps_log("上传百度网盘失败，请检查日志。")
+                        self.root.after(0, lambda: messagebox.showerror("错误", "上传百度网盘失败，请检查日志。"))
+                else:
+                    self._ps_log("=" * 50)
+                    self._ps_log("项目整理失败，请检查日志。")
+                    self.root.after(0, lambda: messagebox.showerror("错误", "项目整理并打包失败，请检查日志。"))
+
+            except Exception as e:
+                self._ps_log(f"错误: {str(e)}")
+                self.root.after(0, lambda: messagebox.showerror("错误", f"执行异常: {str(e)}"))
+            finally:
+                self._ps_set_running(False)
+
+        threading.Thread(target=run, daemon=True).start()
+
     def _ps_organize_upload_cloud(self):
         """整理项目并上传网盘"""
         def run():
diff --git a/quark_uploader.py b/quark_uploader.py
index 1e5f5c9..2eae106 100644
--- a/quark_uploader.py
+++ b/quark_uploader.py
@@ -2,19 +2,30 @@ from __future__ import annotations
 
 import os
 import time
-from typing import Optional, Callable
+from typing import Callable, Optional
+
+from playwright.sync_api import TimeoutError as PlaywrightTimeoutError
 from playwright.sync_api import sync_playwright
 
+
 class QuarkUploader:
     """
-    夸克网盘上传工具 - 终极稳健版 (双重状态锁)
+    夸克网盘上传工具。
+
+    通过真实 Chrome + 持久用户目录上传，避免依赖非公开接口。
     """
-    def __init__(self, chrome_path: str, cookies_dir: str, log_callback: Optional[Callable[[str], None]] = None):
+
+    def __init__(
+        self,
+        chrome_path: str,
+        cookies_dir: str,
+        log_callback: Optional[Callable[[str], None]] = None,
+    ):
         self.chrome_path = chrome_path
         self.cookies_dir = cookies_dir
         self.log_callback = log_callback
         self.url = "https://pan.quark.cn/"
-        
+
         if not os.path.exists(self.cookies_dir):
             os.makedirs(self.cookies_dir, exist_ok=True)
 
@@ -23,144 +34,403 @@ class QuarkUploader:
         if self.log_callback:
             self.log_callback(message)
 
-    def upload_file(self, file_path: str, target_folder_name: str, root_path: str = "精品项目整理") -> bool:
+    def upload_file(
+        self,
+        file_path: str,
+        target_folder_name: str,
+        root_path: str = "精品项目整理",
+    ) -> bool:
         """
-        上传文件到夸克网盘
+        上传文件到夸克网盘。
         """
         if not os.path.exists(file_path):
             self._log(f"错误: 本地文件不存在 {file_path}")
             return False
 
         filename = os.path.basename(file_path)
+        file_size_mb = os.path.getsize(file_path) / 1024 / 1024
+        context = None
 
         try:
             with sync_playwright() as p:
-                self._log(f"正在启动浏览器...")
+                self._log("正在启动浏览器...")
                 launch_args = {
                     "user_data_dir": self.cookies_dir,
                     "headless": False,
-                    "viewport": {"width": 1280, "height": 800}
+                    "viewport": {"width": 1280, "height": 800},
                 }
                 if self.chrome_path and os.path.exists(self.chrome_path):
                     launch_args["executable_path"] = self.chrome_path
-                
-                context = p.chromium.launch_persistent_context(**launch_args)
-                page = context.new_page()
-                
-                self._log("正在打开夸克网盘...")
-                page.goto(self.url, wait_until="networkidle")
-                
-                # 1. 登录检测
-                try:
-                    page.wait_for_selector("text=全部文件", timeout=10000)
-                except:
-                    self._log("请在浏览器中完成登录...")
-                    page.wait_for_selector("text=全部文件", timeout=180000)
-                
-                # 2. 导航到目录
-                self._log(f"定位目录: {root_path} > {target_folder_name}")
-                page.get_by_text("全部文件").first.click()
-                page.wait_for_timeout(2000)
-                
-                # 进入根目录
-                if not page.get_by_text(root_path, exact=True).first.is_visible():
-                    self._create_folder(page, root_path)
-                page.get_by_text(root_path, exact=True).first.dblclick()
-                page.wait_for_timeout(1000)
-                
-                # 进入项目目录
-                if not page.get_by_text(target_folder_name, exact=True).first.is_visible():
-                    self._create_folder(page, target_folder_name)
-                page.get_by_text(target_folder_name, exact=True).first.dblclick()
-                page.wait_for_timeout(2000)
-                
-                # 3. 查重
-                if page.get_by_text(filename, exact=True).count() > 0:
-                    self._log(f"云端已存在 '{filename}'，跳过上传。")
-                    context.close()
-                    return True
-                
-                # 4. 执行上传
-                self._log(f"开始上传: {filename}")
-                # 清除可能存在的旧“上传成功”提示
-                page.evaluate("() => { const els = document.querySelectorAll('.ant-notification-notice'); els.forEach(e => e.remove()); }")
-                
-                page.set_input_files("input[type=file]", file_path)
-                
-                # 5. 确认上传已启动 (关键一步)
-                self._log("等待上传任务初始化...")
-                try:
-                    # 等待出现任何进度指示元素
-                    page.wait_for_selector("text=% , .ant-progress-inner, text=正在上传", timeout=20000)
-                    self._log("检测到上传流已建立。")
-                except:
-                    self._log("警告：未检测到明显的进度条，可能由于文件较小或 UI 延迟，继续监测列表状态。")
 
-                # 6. 深度监测循环
-                self._log("进入深度监测模式，请保持浏览器前台运行...")
-                start_time = time.time()
-                timeout = 1800 # 30分钟
-                
-                # 连续稳定次数计数
-                stable_count = 0
-                
-                while time.time() - start_time < timeout:
-                    # A. 检查全局进度标志
-                    is_uploading = page.get_by_text("%").is_visible() or \
-                                   page.get_by_text("正在上传").is_visible() or \
-                                   page.locator(".ant-progress-inner").is_visible()
-                    
-                    # B. 检查列表中该行的具体状态 (精准匹配)
-                    # 找到包含文件名的那一行 tr，看里面是否有“正在上传”字样
-                    row_status_text = ""
-                    try:
-                        row = page.locator(f"tr:has-text('{filename}')").first
-                        if row.is_visible():
-                            row_status_text = row.inner_text()
-                    except:
-                        pass
-                    
-                    is_item_active = "正在上传" in row_status_text or "%" in row_status_text or "等待上传" in row_status_text
-                    
-                    # C. 确认列表里确实有这个文件
-                    in_list = page.get_by_text(filename, exact=True).count() > 0
-                    
-                    if not is_uploading and not is_item_active and in_list:
-                        stable_count += 1
-                        if stable_count >= 5: # 连续 5 次检测(约 15 秒)都处于稳定态
-                            self._log("检测到上传任务已从任务列表中消失，且列表文件状态正常。")
-                            self._log("为了绝对安全，最后等待 15 秒进行数据落盘同步...")
-                            page.wait_for_timeout(15000)
-                            self._log("上传确认圆满成功！")
-                            context.close()
-                            return True
-                        else:
-                            self._log(f"上传疑似完成，正在进行稳定性校验 ({stable_count}/5)...")
-                    else:
-                        if is_uploading or is_item_active:
-                            self._log("监测到活跃传输流...")
-                        stable_count = 0 # 只要发现还在传，重置计数
-                    
-                    time.sleep(3)
+                context = p.chromium.launch_persistent_context(**launch_args)
+                context.add_init_script(
+                    "Object.defineProperty(navigator, 'webdriver', {get: () => undefined})"
+                )
+
+                page = context.new_page()
+                page.set_default_timeout(60000)
+
+                self._log("正在打开夸克网盘...")
+                page.goto(self.url, wait_until="domcontentloaded", timeout=60000)
+
+                if not self._wait_for_login(page, timeout=10):
+                    self._log("请在浏览器中完成夸克登录，登录后会自动继续...")
+                    if not self._wait_for_login(page, timeout=180):
+                        raise RuntimeError("等待登录超时")
+
+                self._log(f"定位目录: {root_path} > {target_folder_name}")
+                self._open_all_files(page)
+                self._wait_for_file_list_loaded(page)
+
+                if not self._ensure_folder(page, root_path):
+                    raise RuntimeError(f"无法进入或创建根目录: {root_path}")
+
+                if not self._ensure_folder(page, target_folder_name):
+                    raise RuntimeError(f"无法进入或创建项目目录: {target_folder_name}")
+
+                if self._check_file_exists(page, filename):
+                    self._log(f"云端已存在 '{filename}'，跳过上传。")
+                    return True
+
+                self._clear_old_notifications(page)
+                self._log(f"开始上传: {filename} ({file_size_mb:.2f} MB)")
+                self._select_upload_file(page, file_path)
+
+                if not self._wait_for_upload_started(page, filename):
+                    self._log("警告：未检测到明显的上传任务，继续监测文件列表。")
+                else:
+                    self._log("检测到上传任务已建立。")
+
+                success = self._wait_for_upload_finished(page, filename)
+                if success:
+                    self._log("夸克网盘上传确认成功。")
+                    return True
 
                 self._log("传输任务监测超时，请确认网速或手动核实。")
-                context.close()
                 return False
 
-        except Exception as e:
-            self._log(f"代码执行异常: {str(e)}")
+        except Exception as exc:
+            self._log(f"代码执行异常: {exc}")
+            return False
+        finally:
+            if context:
+                try:
+                    context.close()
+                except Exception:
+                    pass
+
+    def _wait_for_login(self, page, timeout: float) -> bool:
+        start_time = time.time()
+        while time.time() - start_time < timeout:
+            try:
+                body_text = self._body_text(page)
+                if "全部文件" in body_text or "我的文件" in body_text:
+                    return True
+            except Exception:
+                pass
+            time.sleep(1)
+        return False
+
+    def _open_all_files(self, page):
+        if self._click_text(page, ["全部文件", "我的文件"]):
+            page.wait_for_timeout(1000)
+            return
+        self._log("未找到“全部文件”入口，尝试直接在当前页面继续。")
+
+    def _wait_for_file_list_loaded(
+        self,
+        page,
+        timeout: float = 12,
+        poll_interval: float = 0.4,
+    ) -> bool:
+        start_time = time.time()
+        while time.time() - start_time < timeout:
+            try:
+                state = page.evaluate(
+                    """
+                    () => {
+                        const text = document.body ? document.body.innerText : '';
+                        const loadingWords = ['加载中', '正在加载', '请稍候'];
+                        const hasLoadingText = loadingWords.some(word => text.includes(word));
+                        const itemCount = document.querySelectorAll(
+                            '[title], [aria-label], [class*="file"], [class*="File"]'
+                        ).length;
+                        const hasAction = ['新建', '上传', '全部文件', '我的文件']
+                            .some(word => text.includes(word));
+                        const hasEmptyState = ['暂无文件', '空文件夹', '拖拽文件']
+                            .some(word => text.includes(word));
+                        return {hasLoadingText, itemCount, hasAction, hasEmptyState};
+                    }
+                    """
+                )
+                if (
+                    state["hasAction"]
+                    and (state["itemCount"] > 0 or state["hasEmptyState"] or not state["hasLoadingText"])
+                ):
+                    return True
+            except Exception:
+                pass
+            time.sleep(poll_interval)
+        return False
+
+    def _ensure_folder(self, page, folder_name: str) -> bool:
+        self._wait_for_file_list_loaded(page)
+
+        if self._open_folder(page, folder_name):
+            return True
+
+        self._log(f"目录不存在，准备创建: {folder_name}")
+        if not self._create_folder(page, folder_name):
             return False
 
-    def _create_folder(self, page, folder_name: str):
-        """创建目录"""
+        end_time = time.time() + 20
+        while time.time() < end_time:
+            if self._open_folder(page, folder_name):
+                return True
+            time.sleep(1)
+
+        return False
+
+    def _open_folder(self, page, folder_name: str) -> bool:
         try:
-            page.get_by_text("新建").first.click()
+            locator = page.get_by_text(folder_name, exact=True)
+            count = locator.count()
+            for index in range(min(count, 5)):
+                item = locator.nth(index)
+                if item.is_visible():
+                    item.dblclick(force=True, timeout=5000)
+                    page.wait_for_timeout(1200)
+                    self._wait_for_file_list_loaded(page)
+                    self._log(f"已进入目录: {folder_name}")
+                    return True
+        except Exception:
+            pass
+
+        return False
+
+    def _create_folder(self, page, folder_name: str) -> bool:
+        try:
+            if not self._click_text(page, ["新建"]):
+                self._log("未找到“新建”按钮。")
+                return False
+
             page.wait_for_timeout(500)
-            page.get_by_text("新建文件夹").first.click()
+            if not self._click_text(page, ["新建文件夹", "文件夹"]):
+                self._log("未找到“新建文件夹”入口。")
+                return False
+
             page.wait_for_timeout(500)
             page.keyboard.type(folder_name)
             page.keyboard.press("Enter")
             self._log(f"已创建目录: {folder_name}")
             page.wait_for_timeout(1500)
-        except:
+            return True
+        except Exception as exc:
+            self._log(f"创建目录失败: {exc}")
+            return False
+
+    def _select_upload_file(self, page, file_path: str):
+        input_locator = page.locator("input[type=file]")
+        if input_locator.count() == 0:
+            self._click_text(page, ["上传", "上传文件"])
+            page.wait_for_selector("input[type=file]", timeout=10000)
+
+        page.locator("input[type=file]").first.set_input_files(file_path)
+
+    def _wait_for_upload_started(
+        self,
+        page,
+        filename: str,
+        timeout: float = 20,
+    ) -> bool:
+        start_time = time.time()
+        while time.time() - start_time < timeout:
+            state = self._upload_state(page, filename)
+            if state["uploading"] or state["item_active"] or state["file_exists"]:
+                return True
+            time.sleep(0.5)
+        return False
+
+    def _wait_for_upload_finished(
+        self,
+        page,
+        filename: str,
+        timeout: float = 1800,
+    ) -> bool:
+        start_time = time.time()
+        stable_count = 0
+        last_active_log = 0.0
+
+        self._log("进入上传监测模式，请保持浏览器前台运行...")
+        while time.time() - start_time < timeout:
+            state = self._upload_state(page, filename)
+
+            if state["has_error"]:
+                self._log("检测到页面提示上传失败。")
+                return False
+
+            if state["file_exists"] and not state["uploading"] and not state["item_active"]:
+                stable_count += 1
+                if stable_count >= 5:
+                    self._log("文件已出现在列表中，且上传状态连续稳定。")
+                    self._log("最后等待 15 秒进行网盘同步确认...")
+                    page.wait_for_timeout(15000)
+                    return self._check_file_exists(page, filename)
+
+                self._log(f"上传疑似完成，稳定性校验 ({stable_count}/5)...")
+            else:
+                stable_count = 0
+                now = time.time()
+                if (state["uploading"] or state["item_active"]) and now - last_active_log >= 15:
+                    self._log("监测到活跃传输流，继续等待...")
+                    last_active_log = now
+
+            time.sleep(3)
+
+        return False
+
+    def _upload_state(self, page, filename: str) -> dict:
+        try:
+            state = page.evaluate(
+                """
+                (filename) => {
+                    const normalize = value => (value || '').replace(/\\s+/g, ' ').trim();
+                    const visible = element => {
+                        if (!element) return false;
+                        const style = window.getComputedStyle(element);
+                        const rect = element.getBoundingClientRect();
+                        return style.display !== 'none'
+                            && style.visibility !== 'hidden'
+                            && rect.width > 0
+                            && rect.height > 0;
+                    };
+                    const bodyText = normalize(document.body ? document.body.innerText : '');
+                    const nodes = [
+                        ...document.querySelectorAll(
+                            '[title], [aria-label], [class*="file"], [class*="File"], [class*="name"], [class*="Name"]'
+                        )
+                    ].filter(visible);
+
+                    const matchingNode = nodes.find(node => {
+                        const title = normalize(node.getAttribute('title'));
+                        const label = normalize(node.getAttribute('aria-label'));
+                        const text = normalize(node.textContent);
+                        return title === filename || label === filename || text === filename;
+                    });
+
+                    const container = matchingNode
+                        ? matchingNode.closest('tr, li, [class*="item"], [class*="Item"], [class*="row"], [class*="Row"], [class*="file"], [class*="File"]')
+                        : null;
+                    const containerText = normalize(container ? container.innerText : '');
+                    const activeWords = ['正在上传', '上传中', '等待上传', '传输中', '解析中', '处理中'];
+                    const errorWords = ['上传失败', '传输失败', '网络异常', '上传出错'];
+                    const progressPattern = /(?:^|\\D)(?:100|\\d{1,2})(?:\\.\\d+)?%/;
+                    const itemHasProgress = container
+                        ? [...container.querySelectorAll(
+                            '.ant-progress-inner, [class*="progress"], [class*="Progress"]'
+                        )].some(visible)
+                        : false;
+
+                    const itemActive = activeWords.some(word => containerText.includes(word))
+                        || progressPattern.test(containerText)
+                        || itemHasProgress;
+                    const uploading = activeWords.some(word => bodyText.includes(word))
+                        || itemActive;
+                    const hasError = errorWords.some(word => bodyText.includes(word));
+
+                    return {
+                        file_exists: Boolean(matchingNode),
+                        item_active: itemActive,
+                        uploading,
+                        has_error: hasError,
+                    };
+                }
+                """,
+                filename,
+            )
+            return {
+                "file_exists": bool(state.get("file_exists")),
+                "item_active": bool(state.get("item_active")),
+                "uploading": bool(state.get("uploading")),
+                "has_error": bool(state.get("has_error")),
+            }
+        except Exception:
+            return {
+                "file_exists": self._check_file_exists(page, filename),
+                "item_active": False,
+                "uploading": False,
+                "has_error": False,
+            }
+
+    def _check_file_exists(self, page, filename: str) -> bool:
+        try:
+            locator = page.get_by_text(filename, exact=True)
+            count = locator.count()
+            for index in range(min(count, 5)):
+                if locator.nth(index).is_visible():
+                    return True
+        except Exception:
             pass
+
+        try:
+            names = page.evaluate(
+                """
+                () => {
+                    const normalize = value => (value || '').replace(/\\s+/g, ' ').trim();
+                    const visible = element => {
+                        const style = window.getComputedStyle(element);
+                        const rect = element.getBoundingClientRect();
+                        return style.display !== 'none'
+                            && style.visibility !== 'hidden'
+                            && rect.width > 0
+                            && rect.height > 0;
+                    };
+                    const result = [];
+                    for (const node of document.querySelectorAll('[title], [aria-label]')) {
+                        if (!visible(node)) continue;
+                        const title = normalize(node.getAttribute('title'));
+                        const label = normalize(node.getAttribute('aria-label'));
+                        if (title) result.push(title);
+                        if (label) result.push(label);
+                    }
+                    return result;
+                }
+                """
+            )
+            return filename in names
+        except Exception:
+            return False
+
+    def _click_text(self, page, texts: list[str]) -> bool:
+        for text in texts:
+            for exact in (True, False):
+                try:
+                    locator = page.get_by_text(text, exact=exact)
+                    count = locator.count()
+                    for index in range(min(count, 5)):
+                        item = locator.nth(index)
+                        if item.is_visible():
+                            item.click(force=True, timeout=5000)
+                            return True
+                except PlaywrightTimeoutError:
+                    continue
+                except Exception:
+                    continue
+        return False
+
+    def _clear_old_notifications(self, page):
+        try:
+            page.evaluate(
+                """
+                () => {
+                    document
+                        .querySelectorAll('.ant-notification-notice, .ant-message-notice')
+                        .forEach(element => element.remove());
+                }
+                """
+            )
+        except Exception:
+            pass
+
+    def _body_text(self, page) -> str:
+        return page.evaluate("() => document.body ? document.body.innerText : ''")