初始化：办公工具平台

.devcontainer/devcontainer.json

@@ -0,0 +1,33 @@
+{
+  "name": "Python 3",
+  // Or use a Dockerfile or Docker Compose file. More info: https://containers.dev/guide/dockerfile
+  "image": "mcr.microsoft.com/devcontainers/python:1-3.11-bookworm",
+  "customizations": {
+    "codespaces": {
+      "openFiles": [
+        "README.md",
+        "app.py"
+      ]
+    },
+    "vscode": {
+      "settings": {},
+      "extensions": [
+        "ms-python.python",
+        "ms-python.vscode-pylance"
+      ]
+    }
+  },
+  "updateContentCommand": "[ -f packages.txt ] && sudo apt update && sudo apt upgrade -y && sudo xargs apt install -y <packages.txt; [ -f requirements.txt ] && pip3 install --user -r requirements.txt; pip3 install --user streamlit; echo '✅ Packages installed and Requirements met'",
+  "postAttachCommand": {
+    "server": "streamlit run app.py --server.enableCORS false --server.enableXsrfProtection false"
+  },
+  "portsAttributes": {
+    "8501": {
+      "label": "Application",
+      "onAutoForward": "openPreview"
+    }
+  },
+  "forwardPorts": [
+    8501
+  ]
+}

.gitignore

@@ -0,0 +1,27 @@
+# Python
+__pycache__/
+*.py[cod]
+*$py.class
+*.so
+.Python
+venv/
+env/
+*.egg-info/
+dist/
+build/
+
+# IDE
+.vscode/
+.idea/
+*.swp
+*.swo
+
+# OS
+.DS_Store
+Thumbs.db
+
+# Streamlit secrets (contains sensitive data)
+.streamlit/secrets.toml
+
+# Uploads
+uploads/

README.md

@@ -0,0 +1,64 @@
+---
+title: Office Tools
+emoji: 🛠️
+colorFrom: blue
+colorTo: green
+sdk: streamlit
+sdk_version: 1.28.0
+app_file: app.py
+pinned: false
+---
+
+# 🛠️ 办公工具平台
+
+专为公司内部打造的自助办公工具集合，帮助员工高效处理日常文档和数据工作。
+
+## 功能模块
+
+### 📊 Excel处理
+- 多Excel文件合并
+- 数据去重与清洗
+- 数据格式转换（CSV/Excel/JSON）
+- 透视表生成
+
+### 📄 PDF处理
+- PDF合并/拆分
+- PDF加水印/密码
+
+### 📝 文档生成
+- 合同模板生成
+- 批量邮件合并
+- 金额大写转换
+- 条码/二维码生成
+
+### 🔧 文本处理
+- 文本对比（Diff）
+- 批量文本替换
+- JSON格式化
+- 字数统计
+
+## 本地运行
+
+```bash
+# 安装依赖
+pip install -r requirements.txt
+
+# 运行应用
+streamlit run app.py
+```
+
+## 部署到 Streamlit Cloud
+
+1. 将代码推送到 GitHub
+2. 访问 [share.streamlit.io](https://share.streamlit.io)
+3. 连接 GitHub 仓库
+4. 选择 `app.py` 作为主文件
+5. 点击 Deploy
+
+## 技术栈
+
+- Streamlit
+- Pandas
+- PyPDF2
+- python-docx
+- qrcode

app.py

@@ -0,0 +1,65 @@
+import streamlit as st
+
+st.set_page_config(
+    page_title="办公工具平台",
+    page_icon="🛠️",
+    layout="wide",
+    initial_sidebar_state="expanded"
+)
+
+st.title("🛠️ 办公工具平台")
+st.markdown("---")
+
+st.markdown("""
+### 欢迎使用办公工具平台
+
+这是一个专为公司内部打造的自助办公工具集合，帮助您高效处理日常文档和数据工作。
+
+#### 📊 Excel处理
+- 多Excel文件合并
+- 数据去重与清洗
+- 数据格式转换（CSV/Excel/JSON）
+- 透视表生成
+
+#### 📄 PDF处理
+- PDF合并/拆分
+- PDF加水印/密码
+- PDF转Word / Word转PDF
+
+#### 🤖 AI文档处理
+- 长文档摘要 - 上传文档，AI提取核心要点
+- 文档翻译 - 支持多语言互译，保持格式
+- 会议纪要整理 - 杂乱笔记变结构化纪要
+- 合同条款解读 - 用大白话解释专业术语
+
+#### 📑 PPT生成
+- 智能PPT生成 - 输入主题，AI自动生成演示文稿
+- 多种模板可选 - 商务、科技、教育等风格
+- 自定义内容 - 支持编辑和调整
+
+#### 🔧 文本处理
+- 文本对比（Diff）
+- 批量文本替换
+- JSON格式化
+- 字数统计
+
+#### 🖼️ 图片处理
+- 多图拼接（横向/纵向/宫格）
+- 自定义间距和背景颜色
+
+#### 📐 Excel公式助手
+- 自然语言描述需求，AI生成Excel公式
+- 支持中英文函数名、新旧版本兼容
+
+#### 🎙️ 语音脚本
+- 文本自动生成带停顿/强调标记的朗读脚本
+- 支持上传文件，估算朗读时长
+
+#### 🔊 语音合成
+- 输入文本，AI直接生成可播放的语音
+- 多种音色可选，支持调节语速
+- 生成MP3文件可直接下载
+""")
+
+st.markdown("---")
+st.markdown("请从左侧菜单选择需要使用的工具。")

pages/1_excel_tools.py

@@ -0,0 +1,140 @@
+import streamlit as st
+import pandas as pd
+import sys
+sys.path.append('..')
+from utils.excel_utils import *
+
+st.set_page_config(page_title="Excel处理", page_icon="📊", layout="wide")
+
+st.title("📊 Excel处理工具")
+st.markdown("---")
+
+# 功能选择
+tab1, tab2, tab3, tab4 = st.tabs(["合并Excel", "数据清洗", "格式转换", "透视表"])
+
+with tab1:
+    st.header("合并多个Excel文件")
+    uploaded_files = st.file_uploader("上传Excel/CSV文件", type=['xlsx', 'xls', 'csv'], accept_multiple_files=True, key="merge")
+
+    if uploaded_files:
+        merge_type = st.radio("合并方式", ["按行合并（上下拼接）", "按列合并（左右拼接）"])
+
+        if st.button("开始合并", key="merge_btn"):
+            try:
+                result_df = merge_excel_files(uploaded_files, "rows" if "按行" in merge_type else "columns")
+                st.success(f"合并完成！共 {len(result_df)} 行数据")
+                st.dataframe(result_df.head(100))
+
+                # 下载按钮
+                csv = result_df.to_csv(index=False, encoding='utf-8-sig')
+                st.download_button("下载CSV", csv, "merged_data.csv", "text/csv")
+            except Exception as e:
+                st.error(f"合并失败: {str(e)}")
+
+with tab2:
+    st.header("数据清洗")
+    uploaded_file = st.file_uploader("上传文件", type=['xlsx', 'xls', 'csv'], key="clean")
+
+    if uploaded_file:
+        if uploaded_file.name.endswith('.csv'):
+            df = pd.read_csv(uploaded_file)
+        else:
+            df = pd.read_excel(uploaded_file)
+
+        st.write(f"原始数据: {len(df)} 行 × {len(df.columns)} 列")
+        st.dataframe(df.head())
+
+        col1, col2 = st.columns(2)
+        with col1:
+            remove_dup = st.checkbox("去除重复行", value=True)
+        with col2:
+            remove_empty = st.checkbox("去除空白行", value=True)
+
+        if st.button("开始清洗", key="clean_btn"):
+            result_df = df.copy()
+            if remove_dup:
+                result_df = remove_duplicates(result_df)
+            if remove_empty:
+                result_df = clean_data(result_df, remove_empty_rows=True, remove_empty_cols=False)
+
+            st.success(f"清洗完成！剩余 {len(result_df)} 行数据")
+            st.dataframe(result_df.head(100))
+
+            csv = result_df.to_csv(index=False, encoding='utf-8-sig')
+            st.download_button("下载清洗后数据", csv, "cleaned_data.csv", "text/csv")
+
+with tab3:
+    st.header("数据格式转换")
+    uploaded_file = st.file_uploader("上传文件", type=['xlsx', 'xls', 'csv', 'json'], key="convert")
+
+    if uploaded_file:
+        target_format = st.selectbox("目标格式", ["Excel (.xlsx)", "CSV (.csv)", "JSON (.json)"])
+
+        if st.button("开始转换", key="convert_btn"):
+            try:
+                # 读取文件
+                if uploaded_file.name.endswith('.csv'):
+                    df = pd.read_csv(uploaded_file)
+                elif uploaded_file.name.endswith('.json'):
+                    df = pd.read_json(uploaded_file)
+                else:
+                    df = pd.read_excel(uploaded_file)
+
+                # 转换格式
+                format_map = {
+                    "Excel (.xlsx)": "excel",
+                    "CSV (.csv)": "csv",
+                    "JSON (.json)": "json"
+                }
+                result = convert_format(df, format_map[target_format])
+
+                # 下载
+                ext_map = {"Excel (.xlsx)": "xlsx", "CSV (.csv)": "csv", "JSON (.json)": "json"}
+                mime_map = {
+                    "Excel (.xlsx)": "application/vnd.openxmlformats-officedocument.spreadsheetml.sheet",
+                    "CSV (.csv)": "text/csv",
+                    "JSON (.json)": "application/json"
+                }
+                st.download_button(
+                    "下载转换结果",
+                    result,
+                    f"converted.{ext_map[target_format]}",
+                    mime_map[target_format]
+                )
+                st.success("转换完成！")
+            except Exception as e:
+                st.error(f"转换失败: {str(e)}")
+
+with tab4:
+    st.header("生成透视表")
+    uploaded_file = st.file_uploader("上传文件", type=['xlsx', 'xls', 'csv'], key="pivot")
+
+    if uploaded_file:
+        if uploaded_file.name.endswith('.csv'):
+            df = pd.read_csv(uploaded_file)
+        else:
+            df = pd.read_excel(uploaded_file)
+
+        st.write("原始数据预览:")
+        st.dataframe(df.head())
+
+        columns = df.columns.tolist()
+        col1, col2, col3 = st.columns(3)
+        with col1:
+            index_col = st.selectbox("行字段", columns, key="pivot_index")
+        with col2:
+            columns_col = st.selectbox("列字段", columns, key="pivot_columns")
+        with col3:
+            values_col = st.selectbox("值字段", columns, key="pivot_values")
+
+        aggfunc = st.selectbox("聚合函数", ["sum", "mean", "count", "max", "min"])
+
+        if st.button("生成透视表", key="pivot_btn"):
+            try:
+                result = create_pivot_table(df, index_col, columns_col, values_col, aggfunc)
+                st.dataframe(result)
+
+                csv = result.to_csv(encoding='utf-8-sig')
+                st.download_button("下载透视表", csv, "pivot_table.csv", "text/csv")
+            except Exception as e:
+                st.error(f"生成失败: {str(e)}")

pages/2_pdf_tools.py

@@ -0,0 +1,108 @@
+import streamlit as st
+import sys
+sys.path.append('..')
+from utils.pdf_utils import *
+
+st.set_page_config(page_title="PDF处理", page_icon="📄", layout="wide")
+
+st.title("📄 PDF处理工具")
+st.markdown("---")
+
+tab1, tab2, tab3, tab4, tab5 = st.tabs(["PDF合并", "PDF拆分", "添加水印", "添加密码", "格式转换"])
+
+with tab1:
+    st.header("合并多个PDF文件")
+    uploaded_files = st.file_uploader("上传PDF文件", type=['pdf'], accept_multiple_files=True, key="merge_pdf")
+
+    if uploaded_files and len(uploaded_files) >= 2:
+        st.write(f"已上传 {len(uploaded_files)} 个文件")
+
+        if st.button("开始合并", key="merge_pdf_btn"):
+            try:
+                result = merge_pdfs(uploaded_files)
+                st.success("合并完成！")
+                st.download_button("下载合并后的PDF", result, "merged.pdf", "application/pdf")
+            except Exception as e:
+                st.error(f"合并失败: {str(e)}")
+    elif uploaded_files:
+        st.warning("请至少上传2个PDF文件")
+
+with tab2:
+    st.header("拆分PDF文件")
+    uploaded_file = st.file_uploader("上传PDF文件", type=['pdf'], key="split_pdf")
+
+    if uploaded_file:
+        reader = PyPDF2.PdfReader(uploaded_file)
+        total_pages = len(reader.pages)
+        st.write(f"文件共 {total_pages} 页")
+
+        split_mode = st.radio("拆分方式", ["按页码范围", "每页一个文件"])
+
+        if split_mode == "按页码范围":
+            pages_input = st.text_input("输入页码范围（如: 1-3,5,7-9）", "1-3")
+            if st.button("拆分", key="split_range_btn"):
+                try:
+                    results = split_pdf(uploaded_file, pages_input)
+                    for i, result in enumerate(results):
+                        st.download_button(f"下载第{i+1}部分", result, f"split_{i+1}.pdf", "application/pdf")
+                except Exception as e:
+                    st.error(f"拆分失败: {str(e)}")
+        else:
+            if st.button("拆分为单页", key="split_each_btn"):
+                try:
+                    results = split_pdf(uploaded_file)
+                    for i, result in enumerate(results):
+                        st.download_button(f"下载第{i+1}页", result, f"page_{i+1}.pdf", "application/pdf")
+                except Exception as e:
+                    st.error(f"拆分失败: {str(e)}")
+
+with tab3:
+    st.header("添加文字水印")
+    uploaded_file = st.file_uploader("上传PDF文件", type=['pdf'], key="watermark_pdf")
+
+    if uploaded_file:
+        watermark_text = st.text_input("水印文字", "内部文件")
+
+        if st.button("添加水印", key="watermark_btn"):
+            try:
+                result = add_watermark(uploaded_file, watermark_text)
+                st.success("水印添加完成！")
+                st.download_button("下载带水印的PDF", result, "watermarked.pdf", "application/pdf")
+            except Exception as e:
+                st.error(f"添加水印失败: {str(e)}")
+
+with tab4:
+    st.header("添加密码保护")
+    uploaded_file = st.file_uploader("上传PDF文件", type=['pdf'], key="password_pdf")
+
+    if uploaded_file:
+        user_password = st.text_input("设置打开密码", type="password")
+        owner_password = st.text_input("设置权限密码（可选）", type="password")
+
+        if st.button("添加密码", key="password_btn"):
+            if user_password:
+                try:
+                    result = add_password(uploaded_file, user_password, owner_password)
+                    st.success("密码添加完成！")
+                    st.download_button("下载加密后的PDF", result, "encrypted.pdf", "application/pdf")
+                except Exception as e:
+                    st.error(f"添加密码失败: {str(e)}")
+            else:
+                st.warning("请输入密码")
+
+with tab5:
+    st.header("PDF格式转换")
+
+    conversion_type = st.radio("转换类型", ["PDF转Word", "Word转PDF"])
+
+    if conversion_type == "PDF转Word":
+        st.info("注意：PDF转Word功能需要安装额外依赖，复杂排版可能无法完美转换")
+        uploaded_file = st.file_uploader("上传PDF文件", type=['pdf'], key="pdf_to_word")
+        if uploaded_file:
+            if st.button("转换为Word", key="pdf2word_btn"):
+                st.warning("该功能需要安装 pdf2docx 库，当前版本暂不支持")
+    else:
+        uploaded_file = st.file_uploader("上传Word文件", type=['docx', 'doc'], key="word_to_pdf")
+        if uploaded_file:
+            if st.button("转换为PDF", key="word2pdf_btn"):
+                st.warning("该功能需要安装 LibreOffice 或 docx2pdf 库，当前版本暂不支持")

pages/3_ai_doc.py

@@ -0,0 +1,120 @@
+import streamlit as st
+import sys
+sys.path.append('..')
+from utils.ai_utils import *
+
+st.set_page_config(page_title="AI文档处理", page_icon="🤖", layout="wide")
+
+st.title("🤖 AI文档处理")
+st.markdown("---")
+
+tab1, tab2, tab3, tab4 = st.tabs(["文档摘要", "文档翻译", "会议纪要", "合同解读"])
+
+with tab1:
+    st.header("长文档摘要")
+    st.markdown("上传文档或粘贴文本，AI自动提取核心要点")
+
+    input_method = st.radio("输入方式", ["粘贴文本", "上传文件"], key="summary_input")
+
+    text = ""
+    if input_method == "粘贴文本":
+        text = st.text_area("粘贴文档内容", height=300, key="summary_text")
+    else:
+        uploaded_file = st.file_uploader("上传文档", type=['txt', 'md'], key="summary_file")
+        if uploaded_file:
+            text = uploaded_file.read().decode('utf-8')
+            st.text_area("文档内容预览", text[:500] + "..." if len(text) > 500 else text, height=100)
+
+    max_length = st.slider("摘要字数", 100, 1000, 500, step=50)
+
+    if st.button("生成摘要", key="summary_btn"):
+        if text:
+            with st.spinner("AI正在分析文档..."):
+                result = summarize_document(text, max_length)
+                st.markdown("### 摘要结果")
+                st.markdown(result)
+        else:
+            st.warning("请输入或上传文档内容")
+
+with tab2:
+    st.header("文档翻译")
+    st.markdown("支持多语言互译，保持原文格式")
+
+    input_method = st.radio("输入方式", ["粘贴文本", "上传文件"], key="translate_input")
+
+    text = ""
+    if input_method == "粘贴文本":
+        text = st.text_area("粘贴文档内容", height=300, key="translate_text")
+    else:
+        uploaded_file = st.file_uploader("上传文档", type=['txt', 'md'], key="translate_file")
+        if uploaded_file:
+            text = uploaded_file.read().decode('utf-8')
+
+    col1, col2 = st.columns(2)
+    with col1:
+        source_lang = st.selectbox("源语言", ["自动检测", "中文", "英文", "日文", "韩文", "法文", "德文"])
+    with col2:
+        target_lang = st.selectbox("目标语言", ["英文", "中文", "日文", "韩文", "法文", "德文"])
+
+    if st.button("开始翻译", key="translate_btn"):
+        if text:
+            with st.spinner("AI正在翻译..."):
+                result = translate_document(text, target_lang)
+                st.markdown("### 翻译结果")
+                st.markdown(result)
+        else:
+            st.warning("请输入或上传文档内容")
+
+with tab3:
+    st.header("会议纪要整理")
+    st.markdown("将杂乱的会议笔记整理成结构化的会议纪要")
+
+    input_method = st.radio("输入方式", ["粘贴文本", "上传文件"], key="meeting_input")
+
+    text = ""
+    if input_method == "粘贴文本":
+        text = st.text_area("粘贴会议笔记", height=300, key="meeting_text", placeholder="请输入会议记录，可以是语音转文字的原始内容...")
+    else:
+        uploaded_file = st.file_uploader("上传会议记录", type=['txt', 'md'], key="meeting_file")
+        if uploaded_file:
+            text = uploaded_file.read().decode('utf-8')
+
+    if st.button("整理纪要", key="meeting_btn"):
+        if text:
+            with st.spinner("AI正在整理会议纪要..."):
+                result = organize_meeting_notes(text)
+                st.markdown("### 会议纪要")
+                st.markdown(result)
+
+                # 下载按钮
+                st.download_button("下载纪要", result, "会议纪要.md", "text/markdown")
+        else:
+            st.warning("请输入或上传会议记录")
+
+with tab4:
+    st.header("合同条款解读")
+    st.markdown("用通俗易懂的语言解释专业合同条款")
+
+    input_method = st.radio("输入方式", ["粘贴文本", "上传文件"], key="contract_input")
+
+    text = ""
+    if input_method == "粘贴文本":
+        text = st.text_area("粘贴合同条款", height=300, key="contract_text", placeholder="请粘贴需要解读的合同条款...")
+    else:
+        uploaded_file = st.file_uploader("上传合同文档", type=['txt', 'md', 'pdf'], key="contract_file")
+        if uploaded_file:
+            if uploaded_file.name.endswith('.pdf'):
+                import PyPDF2
+                reader = PyPDF2.PdfReader(uploaded_file)
+                text = "\n".join([page.extract_text() for page in reader.pages])
+            else:
+                text = uploaded_file.read().decode('utf-8')
+
+    if st.button("解读条款", key="contract_btn"):
+        if text:
+            with st.spinner("AI正在分析合同条款..."):
+                result = explain_contract_terms(text)
+                st.markdown("### 条款解读")
+                st.markdown(result)
+        else:
+            st.warning("请输入或上传合同内容")

pages/4_ppt_generator.py

@@ -0,0 +1,119 @@
+import streamlit as st
+import sys
+sys.path.append('..')
+from utils.ppt_utils import create_ppt, create_ppt_from_outline
+from utils.ai_utils import generate_ppt_content
+
+st.set_page_config(page_title="PPT生成", page_icon="📑", layout="wide")
+
+st.title("📑 PPT生成器")
+st.markdown("---")
+
+tab1, tab2 = st.tabs(["AI智能生成", "手动编辑"])
+
+with tab1:
+    st.header("AI智能生成PPT")
+    st.markdown("输入主题，AI自动生成演示文稿内容")
+
+    topic = st.text_input("输入PPT主题", placeholder="例如：2024年度工作总结")
+
+    col1, col2 = st.columns(2)
+    with col1:
+        slides_count = st.slider("幻灯片数量", 5, 20, 10)
+    with col2:
+        template_style = st.selectbox("模板风格", ["商务", "科技", "教育", "简约"])
+
+    if st.button("生成PPT", key="ai_ppt_btn"):
+        if topic:
+            with st.spinner("AI正在生成PPT内容..."):
+                # 生成内容
+                content = generate_ppt_content(topic, slides_count, template_style)
+
+                if content:
+                    # 生成PPT
+                    ppt_bytes = create_ppt(
+                        content.get("title", topic),
+                        content.get("slides", []),
+                        template_style
+                    )
+
+                    # 预览
+                    st.success("PPT生成完成！")
+                    st.markdown("### 内容预览")
+                    st.markdown(f"**标题：** {content.get('title', topic)}")
+                    for i, slide in enumerate(content.get("slides", []), 1):
+                        with st.expander(f"第{i}页: {slide.get('title', '')}"):
+                            for point in slide.get("content", []):
+                                st.markdown(f"- {point}")
+
+                    # 下载
+                    st.download_button(
+                        "下载PPT",
+                        ppt_bytes,
+                        f"{topic}.pptx",
+                        "application/vnd.openxmlformats-officedocument.presentationml.presentation"
+                    )
+                else:
+                    st.error("AI生成内容失败，请检查配置")
+        else:
+            st.warning("请输入PPT主题")
+
+with tab2:
+    st.header("手动编辑PPT")
+    st.markdown("手动输入PPT大纲，系统自动生成演示文稿")
+
+    template_style = st.selectbox("模板风格", ["商务", "科技", "教育", "简约"], key="manual_style")
+
+    outline = st.text_area(
+        "输入PPT大纲",
+        height=400,
+        placeholder="""公司年度总结
+
+# 第一部分：业绩回顾
+- 营收增长30%
+- 新增客户100家
+- 市场份额提升5%
+
+# 第二部分：团队建设
+- 新增员工50人
+- 培训覆盖率100%
+- 员工满意度95%
+
+# 第三部分：未来规划
+- 2025年营收目标
+- 新产品发布计划
+- 市场拓展策略""",
+        help="使用 # 或数字开头表示页面标题，- 表示要点"
+    )
+
+    if st.button("生成PPT", key="manual_ppt_btn"):
+        if outline:
+            with st.spinner("正在生成PPT..."):
+                ppt_bytes = create_ppt_from_outline(outline, template_style)
+
+                st.success("PPT生成完成！")
+                st.download_button(
+                    "下载PPT",
+                    ppt_bytes,
+                    "演示文稿.pptx",
+                    "application/vnd.openxmlformats-officedocument.presentationml.presentation"
+                )
+        else:
+            st.warning("请输入PPT大纲")
+
+# 使用说明
+st.markdown("---")
+with st.expander("使用说明"):
+    st.markdown("""
+    ### AI智能生成
+    1. 输入PPT主题
+    2. 选择幻灯片数量和模板风格
+    3. 点击"生成PPT"
+    4. AI会自动生成内容并创建PPT
+
+    ### 手动编辑
+    1. 使用 `#` 或数字开头表示页面标题
+    2. 使用 `-` 表示要点内容
+    3. 第一行作为PPT标题
+    4. 点击"生成PPT"即可下载
+    """)

pages/4_text_tools.py

@@ -0,0 +1,101 @@
+import streamlit as st
+import sys
+sys.path.append('..')
+from utils.text_utils import *
+
+st.set_page_config(page_title="文本处理", page_icon="🔧", layout="wide")
+
+st.title("🔧 文本处理工具")
+st.markdown("---")
+
+tab1, tab2, tab3, tab4 = st.tabs(["文本对比", "批量替换", "JSON格式化", "字数统计"])
+
+with tab1:
+    st.header("文本对比（Diff）")
+
+    col1, col2 = st.columns(2)
+    with col1:
+        text1 = st.text_area("原始文本", height=300, key="text1")
+    with col2:
+        text2 = st.text_area("修改后文本", height=300, key="text2")
+
+    if st.button("开始对比", key="diff_btn"):
+        if text1 and text2:
+            diff_result = compare_texts(text1, text2)
+            if diff_result:
+                st.code(diff_result, language="diff")
+            else:
+                st.success("两段文本完全相同！")
+        else:
+            st.warning("请输入两段文本")
+
+with tab2:
+    st.header("批量文本替换")
+
+    text_input = st.text_area("输入文本", height=200, key="replace_input")
+
+    st.markdown("**替换规则**（每行一个，格式：原文 → 替换为）")
+    rules_input = st.text_area("替换规则", "旧文本 → 新文本\n错误 → 正确", height=100)
+
+    if st.button("开始替换", key="replace_btn"):
+        if text_input and rules_input:
+            # 解析替换规则
+            replacements = {}
+            for line in rules_input.strip().split('\n'):
+                if '→' in line:
+                    old, new = line.split('→', 1)
+                    replacements[old.strip()] = new.strip()
+
+            if replacements:
+                result = batch_replace(text_input, replacements)
+                st.text_area("替换结果", result, height=200)
+            else:
+                st.warning("未找到有效的替换规则")
+        else:
+            st.warning("请输入文本和替换规则")
+
+with tab3:
+    st.header("JSON格式化")
+
+    json_input = st.text_area("输入JSON", height=200, key="json_input")
+
+    col1, col2 = st.columns(2)
+    with col1:
+        if st.button("格式化（美化）", key="format_btn"):
+            if json_input:
+                result = format_json(json_input)
+                st.code(result, language="json")
+            else:
+                st.warning("请输入JSON")
+    with col2:
+        if st.button("压缩", key="minify_btn"):
+            if json_input:
+                result = minify_json(json_input)
+                st.code(result, language="json")
+            else:
+                st.warning("请输入JSON")
+
+with tab4:
+    st.header("字数统计")
+
+    text_input = st.text_area("输入文本", height=200, key="count_input")
+
+    if st.button("统计", key="count_btn"):
+        if text_input:
+            result = count_chars(text_input)
+
+            col1, col2, col3 = st.columns(3)
+            with col1:
+                st.metric("总字符数", result['总字符数'])
+                st.metric("中文字数", result['中文字数'])
+            with col2:
+                st.metric("英文字母", result['英文字母'])
+                st.metric("数字", result['数字'])
+            with col3:
+                st.metric("空格", result['空格'])
+                st.metric("总行数", result['总行数'])
+
+            st.markdown("---")
+            st.json(result)
+        else:
+            st.warning("请输入文本")

pages/5_image_tools.py

@@ -0,0 +1,65 @@
+import streamlit as st
+import sys
+sys.path.append('..')
+from utils.image_utils import merge_images
+
+st.set_page_config(page_title="图片处理", page_icon="🖼️", layout="wide")
+
+st.title("🖼️ 图片处理工具")
+st.markdown("---")
+
+tab1, = st.tabs(["图片拼接"])
+
+with tab1:
+    st.header("多图拼接")
+    st.markdown("上传多张图片，选择拼接方式，一键生成拼接图")
+
+    uploaded_files = st.file_uploader(
+        "上传图片", type=['png', 'jpg', 'jpeg', 'webp', 'bmp'],
+        accept_multiple_files=True, key="merge_img"
+    )
+
+    if uploaded_files:
+        st.info(f"已上传 {len(uploaded_files)} 张图片")
+
+        col1, col2, col3 = st.columns(3)
+        with col1:
+            direction = st.radio("拼接方向", ["横向拼接", "纵向拼接", "宫格排列"], key="direction")
+        with col2:
+            gap = st.slider("间距（像素）", 0, 50, 5, key="gap")
+        with col3:
+            bg_color = st.color_picker("背景颜色", "#FFFFFF", key="bg_color")
+
+        # 预览上传的图片
+        with st.expander("预览上传的图片"):
+            preview_cols = st.columns(min(len(uploaded_files), 4))
+            for i, f in enumerate(uploaded_files):
+                with preview_cols[i % 4]:
+                    st.image(f, caption=f.name, use_container_width=True)
+
+        if st.button("开始拼接", key="merge_img_btn"):
+            with st.spinner("拼接中..."):
+                try:
+                    direction_map = {
+                        "横向拼接": "horizontal",
+                        "纵向拼接": "vertical",
+                        "宫格排列": "grid"
+                    }
+                    # 解析颜色
+                    r, g, b = int(bg_color[1:3], 16), int(bg_color[3:5], 16), int(bg_color[5:7], 16)
+
+                    result = merge_images(
+                        uploaded_files,
+                        direction=direction_map[direction],
+                        gap=gap,
+                        bg_color=(r, g, b)
+                    )
+
+                    st.success("拼接完成！")
+                    st.image(result, caption="拼接结果", use_container_width=True)
+
+                    st.download_button(
+                        "下载拼接图", result, "merged_image.png", "image/png"
+                    )
+                except Exception as e:
+                    st.error(f"拼接失败: {str(e)}")

pages/6_excel_formula.py

@@ -0,0 +1,81 @@
+import streamlit as st
+import sys
+sys.path.append('..')
+from utils.ai_utils import get_mimo_client
+
+st.set_page_config(page_title="Excel公式助手", page_icon="📐", layout="wide")
+
+st.title("📐 Excel公式助手")
+st.markdown("---")
+
+st.markdown("用自然语言描述你的需求，AI帮你生成Excel公式")
+
+# 常用场景快捷按钮
+st.subheader("常用场景")
+quick_scenarios = [
+    "查找A列中重复的值",
+    "统计B列中大于100的个数",
+    "根据姓名查找对应的工资",
+    "将日期格式从YYYYMMDD转为YYYY-MM-DD",
+    "提取单元格中的数字部分",
+    "合并A1到A10的内容，用逗号分隔",
+]
+
+cols = st.columns(3)
+for i, scenario in enumerate(quick_scenarios):
+    with cols[i % 3]:
+        if st.button(scenario, key=f"quick_{i}", use_container_width=True):
+            st.session_state.formula_input = scenario
+
+# 输入区域
+user_input = st.text_area(
+    "描述你的需求",
+    value=st.session_state.get("formula_input", ""),
+    height=100,
+    placeholder="例如：我想在Sheet2中根据工号查找Sheet1中的姓名和部门",
+    key="formula_text"
+)
+
+col1, col2 = st.columns([1, 1])
+with col1:
+    excel_version = st.selectbox("Excel版本", ["Excel 365/2021", "Excel 2019及更早"], key="version")
+with col2:
+    language = st.selectbox("函数语言", ["中文函数名", "英文函数名"], key="lang")
+
+if st.button("生成公式", key="gen_formula"):
+    if not user_input.strip():
+        st.warning("请描述你的需求")
+    else:
+        client = get_mimo_client()
+        if not client:
+            st.error("AI服务未配置，请设置 MIMO_API_KEY 环境变量")
+        else:
+            with st.spinner("AI正在思考..."):
+                lang_hint = "使用中文函数名（如 求和、如果、垂直查找）" if "中文" in language else "use English function names (SUM, IF, VLOOKUP)"
+                ver_hint = "可以使用 XLOOKUP、FILTER、UNIQUE 等新函数" if "365" in excel_version else "请使用兼容旧版的函数，避免 XLOOKUP 等新函数"
+
+                prompt = f"""你是一个Excel公式专家。用户会描述一个需求，请生成对应的Excel公式。
+
+要求：
+1. 直接给出公式，不要多余的解释
+2. 如果需要辅助列或特殊操作，简要说明
+3. {lang_hint}
+4. {ver_hint}
+5. 如果一个公式不够，给出完整的解决方案
+
+用户需求：{user_input}"""
+
+                try:
+                    response = client.chat.completions.create(
+                        model="mimo-v2.5-pro",
+                        messages=[{"role": "user", "content": prompt}],
+                        max_tokens=1000
+                    )
+                    result = response.choices[0].message.content
+
+                    st.subheader("生成结果")
+                    st.code(result, language="text")
+
+                    st.button("复制公式", on_click=lambda: st.write("已复制"))
+                except Exception as e:
+                    st.error(f"生成失败: {str(e)}")

pages/7_tts_script.py

@@ -0,0 +1,73 @@
+import streamlit as st
+import sys
+sys.path.append('..')
+from utils.tts_utils import generate_tts_script
+
+st.set_page_config(page_title="语音脚本", page_icon="🎙️", layout="wide")
+
+st.title("🎙️ 文本转语音脚本")
+st.markdown("---")
+
+st.markdown("输入文本，自动生成带停顿、语速、强调标记的朗读脚本，可直接用于TTS工具")
+
+# 输入区域
+tab1, tab2 = st.tabs(["手动输入", "上传文件"])
+
+text = ""
+with tab1:
+    text = st.text_area(
+        "粘贴文本内容",
+        height=300,
+        placeholder="在这里粘贴需要转换为语音脚本的文本...",
+        key="tts_text"
+    )
+
+with tab2:
+    uploaded = st.file_uploader("上传文本文件", type=['txt', 'md'], key="tts_file")
+    if uploaded:
+        text = uploaded.read().decode('utf-8')
+        st.text_area("文件内容预览", text[:500] + "..." if len(text) > 500 else text, height=100, key="tts_preview")
+
+# 设置选项
+st.subheader("脚本设置")
+col1, col2 = st.columns(2)
+
+with col1:
+    speed = st.selectbox("默认语速", ["慢速", "正常", "快速"], index=1, key="speed")
+    add_pauses = st.checkbox("自动添加停顿标记", value=True, key="pauses")
+
+with col2:
+    add_emphasis = st.checkbox("自动标记数字和强调", value=True, key="emphasis")
+    output_format = st.selectbox("输出格式", ["纯文本", "带标记文本", "SSML"], key="format")
+
+# 生成
+if st.button("生成语音脚本", key="gen_tts"):
+    if not text.strip():
+        st.warning("请输入或上传文本内容")
+    else:
+        speed_map = {"慢速": "slow", "正常": "normal", "快速": "fast"}
+
+        result = generate_tts_script(
+            text,
+            speed=speed_map[speed],
+            add_pauses=add_pauses,
+            add_emphasis=add_emphasis
+        )
+
+        st.subheader("生成结果")
+        st.code(result, language="text")
+
+        # 统计信息
+        line_count = len(result.strip().split("\n"))
+        char_count = len(result)
+        st.caption(f"共 {line_count} 行，{char_count} 字符")
+
+        col1, col2 = st.columns(2)
+        with col1:
+            st.download_button(
+                "下载脚本", result, "tts_script.txt", "text/plain"
+            )
+        with col2:
+            # 估算朗读时间（中文约每分钟200字）
+            est_minutes = char_count / 200
+            st.info(f"预计朗读时长：约 {est_minutes:.1f} 分钟")

pages/8_tts_audio.py

@@ -0,0 +1,101 @@
+import streamlit as st
+import sys
+sys.path.append('..')
+
+st.set_page_config(page_title="语音合成", page_icon="🔊", layout="wide")
+
+try:
+    from utils.tts_mimo import text_to_speech, PRESET_VOICES
+except Exception as e:
+    st.error(f"模块加载失败: {e}")
+    st.stop()
+
+st.title("🔊 文本转语音")
+st.markdown("---")
+
+st.markdown("输入文本，使用 mimo-v2.5-tts 模型直接生成可播放、可下载的语音文件")
+
+# 输入
+tab1, tab2 = st.tabs(["手动输入", "上传文件"])
+
+text = ""
+with tab1:
+    text = st.text_area(
+        "输入要转换的文本",
+        height=200,
+        placeholder="在这里输入或粘贴文本...",
+        key="tts_input"
+    )
+
+with tab2:
+    uploaded = st.file_uploader("上传文本文件", type=['txt', 'md'], key="tts_file_upload")
+    if uploaded:
+        text = uploaded.read().decode('utf-8')
+        st.text_area("文件预览", text[:300] + "..." if len(text) > 300 else text, height=80, key="tts_file_preview")
+
+# 参数设置
+st.subheader("语音设置")
+col1, col2 = st.columns(2)
+
+with col1:
+    voice_options = list(PRESET_VOICES.keys())
+    voice_labels = [f"{name} ({info['lang']}·{info['gender']})" for name, info in PRESET_VOICES.items()]
+    voice_idx = st.selectbox(
+        "音色",
+        range(len(voice_options)),
+        format_func=lambda i: voice_labels[i],
+        index=0,
+        help="选择预置音色",
+        key="voice"
+    )
+    voice = voice_options[voice_idx]
+
+with col2:
+    style = st.text_input(
+        "风格指令（可选）",
+        placeholder="例如：温柔、活泼、东北话、悲伤、磁性...",
+        help="自然语言描述发音风格，留空使用默认风格",
+        key="style"
+    )
+
+# 高级选项
+with st.expander("高级选项"):
+    st.markdown("""
+    **风格控制提示：**
+    - 基础情绪：开心 / 悲伤 / 愤怒 / 惊讶 / 兴奋 / 平静
+    - 复合情绪：怅然 / 欣慰 / 无奈 / 愧疚 / 释然
+    - 整体语调：温柔 / 高冷 / 活泼 / 严肃 / 慵懒 / 俏皮
+    - 音色定位：磁性 / 醇厚 / 清亮 / 空灵 / 甜美 / 沙哑
+    - 方言：东北话 / 四川话 / 河南话 / 粤语
+    - 也可以在文本中插入标签，如 `(温柔)` `(东北话)` `(唱歌)`
+    """)
+
+# 字符统计
+if text:
+    st.caption(f"文本长度：{len(text)} 字符")
+
+# 生成
+if st.button("生成语音", key="gen_audio"):
+    if not text.strip():
+        st.warning("请输入文本内容")
+    elif len(text) > 4096:
+        st.warning(f"文本过长（{len(text)}字），TTS 单次最多支持 4096 字符，请分段生成")
+    else:
+        with st.spinner("正在生成语音，请稍候..."):
+            try:
+                audio_bytes = text_to_speech(text, voice=voice, style=style)
+
+                st.success("生成完成！")
+
+                # 播放
+                st.audio(audio_bytes, format="audio/wav")
+
+                # 下载
+                st.download_button(
+                    "下载音频",
+                    audio_bytes,
+                    "speech.wav",
+                    "audio/wav"
+                )
+            except Exception as e:
+                st.error(f"生成失败: {str(e)}")

requirements.txt

@@ -0,0 +1,12 @@
+streamlit>=1.28.0
+pandas>=2.0.0
+openpyxl>=3.1.0
+xlrd>=2.0.0
+PyPDF2>=3.0.0
+python-docx>=1.0.0
+reportlab>=4.0.0
+qrcode[pil]>=7.4.0
+Pillow>=10.0.0
+xlsxwriter>=3.1.0
+python-pptx>=0.6.21
+openai>=1.0.0

utils/__init__.py

@@ -0,0 +1 @@
+# Utils package

utils/ai_utils.py

@@ -0,0 +1,145 @@
+import streamlit as st
+import os
+from openai import OpenAI
+
+def get_mimo_client():
+    """获取mimo API客户端"""
+    # 优先从环境变量读取
+    api_key = os.getenv("MIMO_API_KEY")
+    base_url = os.getenv("MIMO_BASE_URL")
+
+    # 如果环境变量没有，尝试从st.secrets读取
+    if not api_key:
+        try:
+            api_key = st.secrets["MIMO_API_KEY"]
+        except:
+            pass
+
+    if not base_url:
+        try:
+            base_url = st.secrets["MIMO_BASE_URL"]
+        except:
+            base_url = "https://api.xiaomimimo.com/v1"
+
+    if not api_key:
+        return None
+
+    return OpenAI(
+        api_key=api_key,
+        base_url=base_url
+    )
+
+def summarize_document(text: str, max_length: int = 500) -> str:
+    """文档摘要"""
+    client = get_mimo_client()
+    if not client:
+        return "错误：未配置 MIMO API Key，请在 Settings → Secrets 中配置"
+
+    try:
+        response = client.chat.completions.create(
+            model="mimo-v2.5-pro",
+            messages=[
+                {"role": "system", "content": "你是一个专业的文档摘要助手，请用简洁的语言总结文档核心内容。"},
+                {"role": "user", "content": f"请为以下文档生成摘要，控制在{max_length}字以内：\n\n{text}"}
+            ],
+            max_tokens=1000
+        )
+        return response.choices[0].message.content
+    except Exception as e:
+        return f"API调用失败: {str(e)}"
+
+def translate_document(text: str, target_lang: str = "英文") -> str:
+    """文档翻译"""
+    client = get_mimo_client()
+    if not client:
+        return "错误：未配置 MIMO API Key，请在 Settings → Secrets 中配置"
+
+    try:
+        response = client.chat.completions.create(
+            model="mimo-v2.5-pro",
+            messages=[
+                {"role": "system", "content": f"你是一个专业的翻译助手，请将文档翻译成{target_lang}，保持原文格式和专业术语。"},
+                {"role": "user", "content": f"请翻译以下文档：\n\n{text}"}
+            ],
+            max_tokens=2000
+        )
+        return response.choices[0].message.content
+    except Exception as e:
+        return f"API调用失败: {str(e)}"
+
+def organize_meeting_notes(notes: str) -> str:
+    """会议纪要整理"""
+    client = get_mimo_client()
+    if not client:
+        return "错误：未配置 MIMO API Key，请在 Settings → Secrets 中配置"
+
+    try:
+        response = client.chat.completions.create(
+            model="mimo-v2.5-pro",
+            messages=[
+                {"role": "system", "content": "你是一个专业的会议纪要整理助手，请将杂乱的会议笔记整理成结构化的会议纪要格式，包含：会议主题、参会人员、主要议题、讨论要点、决议事项、待办事项。"},
+                {"role": "user", "content": f"请整理以下会议笔记：\n\n{notes}"}
+            ],
+            max_tokens=1500
+        )
+        return response.choices[0].message.content
+    except Exception as e:
+        return f"API调用失败: {str(e)}"
+
+def explain_contract_terms(text: str) -> str:
+    """合同条款解读"""
+    client = get_mimo_client()
+    if not client:
+        return "错误：未配置 MIMO API Key，请在 Settings → Secrets 中配置"
+
+    try:
+        response = client.chat.completions.create(
+            model="mimo-v2.5-pro",
+            messages=[
+                {"role": "system", "content": "你是一个法律顾问助手，请用通俗易懂的语言解释合同条款，指出可能的风险点和注意事项。"},
+                {"role": "user", "content": f"请解读以下合同条款：\n\n{text}"}
+            ],
+            max_tokens=1500
+        )
+        return response.choices[0].message.content
+    except Exception as e:
+        return f"API调用失败: {str(e)}"
+
+def generate_ppt_content(topic: str, slides_count: int = 10, style: str = "商务") -> dict:
+    """生成PPT内容"""
+    client = get_mimo_client()
+    if not client:
+        return None
+
+    prompt = f"""请为以下主题生成PPT内容，要求：
+- 主题：{topic}
+- 页数：{slides_count}页
+- 风格：{style}
+
+请以JSON格式返回，结构如下：
+{{
+    "title": "PPT标题",
+    "slides": [
+        {{
+            "title": "页面标题",
+            "content": ["要点1", "要点2", "要点3"],
+            "notes": "演讲者备注"
+        }}
+    ]
+}}"""
+
+    try:
+        response = client.chat.completions.create(
+            model="mimo-v2.5-pro",
+            messages=[
+                {"role": "system", "content": "你是一个专业的PPT内容创作助手，请生成结构清晰、内容专业的PPT大纲。"},
+                {"role": "user", "content": prompt}
+            ],
+            max_tokens=3000
+        )
+
+        import json
+        return json.loads(response.choices[0].message.content)
+    except Exception as e:
+        st.error(f"API调用失败: {str(e)}")
+        return None

utils/doc_utils.py

@@ -0,0 +1,143 @@
+from docx import Document
+from docx.shared import Inches
+import io
+import qrcode
+from typing import Dict
+
+def generate_contract(template_type: str, data: Dict) -> bytes:
+    """生成合同文档"""
+    doc = Document()
+
+    if template_type == "采购合同":
+        doc.add_heading('采购合同', 0)
+        doc.add_paragraph(f'合同编号：{data.get("contract_no", "")}')
+        doc.add_paragraph(f'甲方（采购方）：{data.get("party_a", "")}')
+        doc.add_paragraph(f'乙方（供应方）：{data.get("party_b", "")}')
+        doc.add_paragraph(f'签订日期：{data.get("date", "")}')
+        doc.add_paragraph('')
+        doc.add_paragraph(f'一、采购物品：{data.get("item", "")}')
+        doc.add_paragraph(f'二、数量：{data.get("quantity", "")}')
+        doc.add_paragraph(f'三、单价：{data.get("unit_price", "")}')
+        doc.add_paragraph(f'四、总金额：{data.get("total", "")}')
+        doc.add_paragraph(f'五、交货日期：{data.get("delivery_date", "")}')
+        doc.add_paragraph('')
+        doc.add_paragraph('甲方签字：________________     乙方签字：________________')
+
+    elif template_type == "服务合同":
+        doc.add_heading('服务合同', 0)
+        doc.add_paragraph(f'合同编号：{data.get("contract_no", "")}')
+        doc.add_paragraph(f'甲方（委托方）：{data.get("party_a", "")}')
+        doc.add_paragraph(f'乙方（服务方）：{data.get("party_b", "")}')
+        doc.add_paragraph(f'签订日期：{data.get("date", "")}')
+        doc.add_paragraph('')
+        doc.add_paragraph(f'一、服务内容：{data.get("service_content", "")}')
+        doc.add_paragraph(f'二、服务期限：{data.get("service_period", "")}')
+        doc.add_paragraph(f'三、服务费用：{data.get("fee", "")}')
+        doc.add_paragraph(f'四、付款方式：{data.get("payment_method", "")}')
+        doc.add_paragraph('')
+        doc.add_paragraph('甲方签字：________________     乙方签字：________________')
+
+    elif template_type == "劳动合同":
+        doc.add_heading('劳动合同', 0)
+        doc.add_paragraph(f'甲方（用人单位）：{data.get("company", "")}')
+        doc.add_paragraph(f'乙方（劳动者）：{data.get("employee", "")}')
+        doc.add_paragraph(f'身份证号：{data.get("id_card", "")}')
+        doc.add_paragraph(f'签订日期：{data.get("date", "")}')
+        doc.add_paragraph('')
+        doc.add_paragraph(f'一、工作岗位：{data.get("position", "")}')
+        doc.add_paragraph(f'二、工作地点：{data.get("location", "")}')
+        doc.add_paragraph(f'三、合同期限：{data.get("contract_period", "")}')
+        doc.add_paragraph(f'四、薪资待遇：{data.get("salary", "")}')
+        doc.add_paragraph('')
+        doc.add_paragraph('甲方盖章：________________     乙方签字：________________')
+
+    buffer = io.BytesIO()
+    doc.save(buffer)
+    return buffer.getvalue()
+
+def mail_merge(template_text: str, data_list: list) -> list:
+    """批量邮件合并"""
+    results = []
+    for data in data_list:
+        result = template_text
+        for key, value in data.items():
+            result = result.replace(f'{{{{{key}}}}}', str(value))
+        results.append(result)
+    return results
+
+def amount_to_chinese(amount: float) -> str:
+    """金额转换为中文大写"""
+    digits = ['零', '壹', '贰', '叁', '肆', '伍', '陆', '柒', '捌', '玖']
+    units = ['', '拾', '佰', '仟']
+    big_units = ['', '万', '亿']
+
+    if amount == 0:
+        return '零元整'
+
+    # 分离整数和小数部分
+    amount = round(amount, 2)
+    integer_part = int(amount)
+    decimal_part = round((amount - integer_part) * 100)
+
+    result = ''
+
+    # 处理整数部分
+    if integer_part > 0:
+        str_int = str(integer_part)
+        length = len(str_int)
+        for i, char in enumerate(str_int):
+            digit = int(char)
+            pos = length - i - 1
+            big_pos = pos // 4
+            unit_pos = pos % 4
+
+            if digit != 0:
+                result += digits[digit] + units[unit_pos]
+            else:
+                if unit_pos == 0 and big_pos > 0:
+                    result += big_units[big_pos]
+                elif result and not result.endswith('零'):
+                    result += '零'
+
+            if unit_pos == 0 and big_pos > 0:
+                result += big_units[big_pos]
+
+        result = result.rstrip('零') + '元'
+
+    # 处理小数部分
+    if decimal_part > 0:
+        jiao = decimal_part // 10
+        fen = decimal_part % 10
+        if jiao > 0:
+            result += digits[jiao] + '角'
+        if fen > 0:
+            result += digits[fen] + '分'
+    else:
+        result += '整'
+
+    return result
+
+def generate_qrcode(data: str, size: int = 10) -> bytes:
+    """生成二维码"""
+    qr = qrcode.QRCode(version=1, box_size=size, border=2)
+    qr.add_data(data)
+    qr.make(fit=True)
+    img = qr.make_image(fill_color="black", back_color="white")
+    buffer = io.BytesIO()
+    img.save(buffer, format='PNG')
+    return buffer.getvalue()
+
+def generate_barcode(data: str, barcode_type: str = "code128") -> bytes:
+    """生成条形码"""
+    try:
+        import barcode
+        from barcode.writer import ImageWriter
+
+        code_class = barcode.get_barcode_class(barcode_type)
+        code = code_class(data, writer=ImageWriter())
+        buffer = io.BytesIO()
+        code.write(buffer)
+        return buffer.getvalue()
+    except ImportError:
+        # 如果没有python-barcode库，返回None
+        return None

utils/excel_utils.py

@@ -0,0 +1,50 @@
+import pandas as pd
+import io
+from typing import List, Optional
+
+def merge_excel_files(files: List, merge_type: str = "rows") -> pd.DataFrame:
+    """合并多个Excel文件"""
+    all_dfs = []
+    for file in files:
+        if file.name.endswith('.csv'):
+            df = pd.read_csv(file)
+        else:
+            df = pd.read_excel(file)
+        all_dfs.append(df)
+
+    if merge_type == "rows":
+        return pd.concat(all_dfs, ignore_index=True)
+    else:
+        return pd.concat(all_dfs, axis=1)
+
+def remove_duplicates(df: pd.DataFrame, columns: Optional[List] = None) -> pd.DataFrame:
+    """去除重复行"""
+    return df.drop_duplicates(subset=columns, keep='first')
+
+def clean_data(df: pd.DataFrame, remove_empty_rows: bool = True, remove_empty_cols: bool = True) -> pd.DataFrame:
+    """清洗数据"""
+    if remove_empty_rows:
+        df = df.dropna(how='all')
+    if remove_empty_cols:
+        df = df.dropna(axis=1, how='all')
+    return df.reset_index(drop=True)
+
+def convert_format(df: pd.DataFrame, target_format: str) -> bytes:
+    """转换数据格式"""
+    buffer = io.BytesIO()
+    if target_format == 'csv':
+        df.to_csv(buffer, index=False, encoding='utf-8-sig')
+    elif target_format == 'excel':
+        df.to_excel(buffer, index=False, engine='openpyxl')
+    elif target_format == 'json':
+        df.to_json(buffer, orient='records', force_ascii=False, indent=2)
+    return buffer.getvalue()
+
+def create_pivot_table(df: pd.DataFrame, index_col: str, columns_col: str, values_col: str, aggfunc: str = 'sum') -> pd.DataFrame:
+    """创建透视表"""
+    return pd.pivot_table(df, index=index_col, columns=columns_col, values=values_col, aggfunc=aggfunc)
+
+def vlookup_merge(left_df: pd.DataFrame, right_df: pd.DataFrame, key_col: str, value_col: str) -> pd.DataFrame:
+    """类似VLOOKUP的合并"""
+    right_subset = right_df[[key_col, value_col]]
+    return left_df.merge(right_subset, on=key_col, how='left')

utils/image_utils.py

@@ -0,0 +1,63 @@
+import io
+from PIL import Image
+
+
+def merge_images(images, direction="horizontal", gap=0, bg_color=(255, 255, 255)):
+    """拼接多张图片"""
+    if not images:
+        raise ValueError("没有图片")
+
+    imgs = [Image.open(f) for f in images]
+
+    if direction == "horizontal":
+        total_w = sum(im.width for im in imgs) + gap * (len(imgs) - 1)
+        max_h = max(im.height for im in imgs)
+        result = Image.new("RGB", (total_w, max_h), bg_color)
+        x = 0
+        for im in imgs:
+            y = (max_h - im.height) // 2
+            result.paste(im, (x, y))
+            x += im.width + gap
+
+    elif direction == "vertical":
+        max_w = max(im.width for im in imgs)
+        total_h = sum(im.height for im in imgs) + gap * (len(imgs) - 1)
+        result = Image.new("RGB", (max_w, total_h), bg_color)
+        y = 0
+        for im in imgs:
+            x = (max_w - im.width) // 2
+            result.paste(im, (x, y))
+            y += im.height + gap
+
+    elif direction == "grid":
+        cols = _calc_cols(len(imgs))
+        rows = (len(imgs) + cols - 1) // cols
+
+        max_w = max(im.width for im in imgs)
+        max_h = max(im.height for im in imgs)
+        total_w = cols * max_w + gap * (cols - 1)
+        total_h = rows * max_h + gap * (rows - 1)
+        result = Image.new("RGB", (total_w, total_h), bg_color)
+
+        for idx, im in enumerate(imgs):
+            r, c = divmod(idx, cols)
+            x = c * (max_w + gap) + (max_w - im.width) // 2
+            y = r * (max_h + gap) + (max_h - im.height) // 2
+            result.paste(im, (x, y))
+    else:
+        raise ValueError(f"不支持的方向: {direction}")
+
+    buf = io.BytesIO()
+    result.save(buf, format="PNG")
+    return buf.getvalue()
+
+
+def _calc_cols(n):
+    """计算宫格列数"""
+    if n <= 1:
+        return 1
+    if n <= 4:
+        return 2
+    if n <= 9:
+        return 3
+    return 4

utils/pdf_utils.py

@@ -0,0 +1,96 @@
+import PyPDF2
+import io
+from typing import List
+
+def merge_pdfs(files: List) -> bytes:
+    """合并多个PDF文件"""
+    merger = PyPDF2.PdfMerger()
+    for file in files:
+        merger.append(file)
+    buffer = io.BytesIO()
+    merger.write(buffer)
+    merger.close()
+    return buffer.getvalue()
+
+def split_pdf(file, pages: str = None) -> List[bytes]:
+    """拆分PDF文件"""
+    reader = PyPDF2.PdfReader(file)
+    total_pages = len(reader.pages)
+
+    if pages:
+        page_list = parse_page_range(pages, total_pages)
+    else:
+        page_list = list(range(total_pages))
+
+    results = []
+    for page_num in page_list:
+        writer = PyPDF2.PdfWriter()
+        writer.add_page(reader.pages[page_num])
+        buffer = io.BytesIO()
+        writer.write(buffer)
+        results.append(buffer.getvalue())
+
+    return results
+
+def parse_page_range(pages_str: str, total_pages: int) -> List[int]:
+    """解析页码范围字符串"""
+    result = []
+    parts = pages_str.split(',')
+    for part in parts:
+        part = part.strip()
+        if '-' in part:
+            start, end = part.split('-')
+            start = int(start) - 1
+            end = int(end) if end else total_pages
+            result.extend(range(start, end))
+        else:
+            result.append(int(part) - 1)
+    return result
+
+def add_watermark(pdf_file, watermark_text: str) -> bytes:
+    """添加文字水印"""
+    from reportlab.lib.pagesizes import letter
+    from reportlab.pdfgen import canvas
+
+    reader = PyPDF2.PdfReader(pdf_file)
+    writer = PyPDF2.PdfWriter()
+
+    for page in reader.pages:
+        # 创建水印
+        watermark_buffer = io.BytesIO()
+        c = canvas.Canvas(watermark_buffer, pagesize=letter)
+        c.setFont("Helvetica", 50)
+        c.setFillAlpha(0.3)
+        c.saveState()
+        c.translate(300, 400)
+        c.rotate(45)
+        c.drawCentredString(0, 0, watermark_text)
+        c.restoreState()
+        c.save()
+
+        # 添加水印到页面
+        watermark_buffer.seek(0)
+        watermark_reader = PyPDF2.PdfReader(watermark_buffer)
+        watermark_page = watermark_reader.pages[0]
+        page.merge_page(watermark_page)
+        writer.add_page(page)
+
+    buffer = io.BytesIO()
+    writer.write(buffer)
+    return buffer.getvalue()
+
+def add_password(pdf_file, user_password: str, owner_password: str = None) -> bytes:
+    """添加密码保护"""
+    reader = PyPDF2.PdfReader(pdf_file)
+    writer = PyPDF2.PdfWriter()
+
+    for page in reader.pages:
+        writer.add_page(page)
+
+    if not owner_password:
+        owner_password = user_password
+
+    writer.encrypt(user_password, owner_password)
+    buffer = io.BytesIO()
+    writer.write(buffer)
+    return buffer.getvalue()

utils/ppt_utils.py

@@ -0,0 +1,138 @@
+from pptx import Presentation
+from pptx.util import Inches, Pt
+from pptx.enum.text import PP_ALIGN
+from pptx.dml.color import RGBColor
+import io
+
+def create_ppt(title: str, slides_data: list, template_style: str = "商务") -> bytes:
+    """创建PPT"""
+    prs = Presentation()
+
+    # 设置模板颜色
+    colors = get_template_colors(template_style)
+
+    # 添加标题页
+    add_title_slide(prs, title, colors)
+
+    # 添加内容页
+    for slide_data in slides_data:
+        add_content_slide(prs, slide_data, colors)
+
+    # 保存到内存
+    buffer = io.BytesIO()
+    prs.save(buffer)
+    return buffer.getvalue()
+
+def get_template_colors(style: str) -> dict:
+    """获取模板颜色"""
+    templates = {
+        "商务": {
+            "primary": RGBColor(0, 82, 136),      # 深蓝
+            "secondary": RGBColor(0, 120, 215),    # 蓝色
+            "accent": RGBColor(255, 185, 0),       # 金色
+            "text": RGBColor(51, 51, 51),          # 深灰
+            "background": RGBColor(255, 255, 255)  # 白色
+        },
+        "科技": {
+            "primary": RGBColor(88, 86, 214),      # 紫色
+            "secondary": RGBColor(0, 200, 150),    # 青色
+            "accent": RGBColor(255, 100, 100),     # 红色
+            "text": RGBColor(30, 30, 30),          # 近黑
+            "background": RGBColor(245, 245, 250)  # 浅灰蓝
+        },
+        "教育": {
+            "primary": RGBColor(46, 139, 87),      # 绿色
+            "secondary": RGBColor(70, 130, 180),   # 钢蓝
+            "accent": RGBColor(255, 165, 0),       # 橙色
+            "text": RGBColor(51, 51, 51),          # 深灰
+            "background": RGBColor(255, 255, 255)  # 白色
+        },
+        "简约": {
+            "primary": RGBColor(60, 60, 60),       # 深灰
+            "secondary": RGBColor(120, 120, 120),  # 中灰
+            "accent": RGBColor(200, 50, 50),       # 红色
+            "text": RGBColor(51, 51, 51),          # 深灰
+            "background": RGBColor(255, 255, 255)  # 白色
+        }
+    }
+    return templates.get(style, templates["商务"])
+
+def add_title_slide(prs, title: str, colors: dict):
+    """添加标题页"""
+    slide_layout = prs.slide_layouts[0]  # 标题布局
+    slide = prs.slides.add_slide(slide_layout)
+
+    # 设置标题
+    title_shape = slide.shapes.title
+    title_shape.text = title
+    title_shape.text_frame.paragraphs[0].alignment = PP_ALIGN.CENTER
+    title_shape.text_frame.paragraphs[0].font.size = Pt(44)
+    title_shape.text_frame.paragraphs[0].font.color.rgb = colors["primary"]
+
+    # 设置副标题
+    if len(slide.placeholders) > 1:
+        subtitle = slide.placeholders[1]
+        subtitle.text = "自动生成的演示文稿"
+        subtitle.text_frame.paragraphs[0].alignment = PP_ALIGN.CENTER
+        subtitle.text_frame.paragraphs[0].font.size = Pt(20)
+        subtitle.text_frame.paragraphs[0].font.color.rgb = colors["secondary"]
+
+def add_content_slide(prs, slide_data: dict, colors: dict):
+    """添加内容页"""
+    slide_layout = prs.slide_layouts[1]  # 标题和内容布局
+    slide = prs.slides.add_slide(slide_layout)
+
+    # 设置标题
+    title_shape = slide.shapes.title
+    title_shape.text = slide_data.get("title", "")
+    title_shape.text_frame.paragraphs[0].font.size = Pt(32)
+    title_shape.text_frame.paragraphs[0].font.color.rgb = colors["primary"]
+
+    # 设置内容
+    if len(slide.placeholders) > 1:
+        content = slide.placeholders[1]
+        content.text = ""
+
+        for point in slide_data.get("content", []):
+            p = content.text_frame.add_paragraph()
+            p.text = point
+            p.font.size = Pt(18)
+            p.font.color.rgb = colors["text"]
+            p.space_after = Pt(12)
+            p.level = 0
+
+    # 添加备注
+    if slide_data.get("notes"):
+        notes_slide = slide.notes_slide
+        notes_slide.notes_text_frame.text = slide_data["notes"]
+
+def create_ppt_from_outline(outline: str, template_style: str = "商务") -> bytes:
+    """从大纲文本创建PPT"""
+    lines = outline.strip().split('\n')
+
+    title = lines[0] if lines else "演示文稿"
+    slides_data = []
+
+    current_slide = None
+    for line in lines[1:]:
+        line = line.strip()
+        if not line:
+            continue
+
+        # 判断是否是标题行（以数字开头或以#开头）
+        if line[0].isdigit() or line.startswith('#') or line.startswith('**'):
+            if current_slide:
+                slides_data.append(current_slide)
+            # 提取标题文本
+            slide_title = line.lstrip('0123456789.#* ').strip()
+            current_slide = {"title": slide_title, "content": []}
+        elif current_slide:
+            # 作为要点添加
+            point = line.lstrip('- •* ').strip()
+            if point:
+                current_slide["content"].append(point)
+
+    if current_slide:
+        slides_data.append(current_slide)
+
+    return create_ppt(title, slides_data, template_style)

utils/text_utils.py

@@ -0,0 +1,54 @@
+import difflib
+import json
+import re
+from typing import List
+
+def compare_texts(text1: str, text2: str) -> str:
+    """对比两段文本，返回差异"""
+    lines1 = text1.splitlines(keepends=True)
+    lines2 = text2.splitlines(keepends=True)
+
+    diff = difflib.unified_diff(lines1, lines2, lineterm='')
+    return ''.join(diff)
+
+def batch_replace(text: str, replacements: dict) -> str:
+    """批量文本替换"""
+    result = text
+    for old, new in replacements.items():
+        result = result.replace(old, new)
+    return result
+
+def format_json(json_str: str, indent: int = 2) -> str:
+    """格式化JSON"""
+    try:
+        data = json.loads(json_str)
+        return json.dumps(data, indent=indent, ensure_ascii=False)
+    except json.JSONDecodeError as e:
+        return f"JSON解析错误: {str(e)}"
+
+def minify_json(json_str: str) -> str:
+    """压缩JSON"""
+    try:
+        data = json.loads(json_str)
+        return json.dumps(data, separators=(',', ':'), ensure_ascii=False)
+    except json.JSONDecodeError as e:
+        return f"JSON解析错误: {str(e)}"
+
+def count_chars(text: str) -> dict:
+    """统计字数"""
+    total_chars = len(text)
+    chinese_chars = len(re.findall(r'[一-鿿]', text))
+    english_chars = len(re.findall(r'[a-zA-Z]', text))
+    digits = len(re.findall(r'\d', text))
+    spaces = len(re.findall(r'\s', text))
+    punctuation = total_chars - chinese_chars - english_chars - digits - spaces
+
+    return {
+        '总字符数': total_chars,
+        '中文字数': chinese_chars,
+        '英文字母': english_chars,
+        '数字': digits,
+        '空格': spaces,
+        '标点符号': punctuation,
+        '总行数': len(text.splitlines())
+    }

utils/tts_mimo.py

@@ -0,0 +1,81 @@
+import base64
+import os
+import streamlit as st
+from openai import OpenAI
+
+
+def get_mimo_tts_client():
+    """获取 mimo TTS 客户端（复用同一套密钥）"""
+    api_key = os.getenv("MIMO_API_KEY")
+    base_url = os.getenv("MIMO_BASE_URL")
+
+    if not api_key:
+        try:
+            api_key = st.secrets["MIMO_API_KEY"]
+        except Exception:
+            pass
+
+    if not base_url:
+        try:
+            base_url = st.secrets["MIMO_BASE_URL"]
+        except Exception:
+            base_url = "https://api.xiaomimimo.com/v1"
+
+    if not api_key:
+        return None
+
+    return OpenAI(api_key=api_key, base_url=base_url)
+
+
+# mimo-v2.5-tts 预置音色列表
+PRESET_VOICES = {
+    "冰糖": {"id": "冰糖", "lang": "中文", "gender": "女性"},
+    "茉莉": {"id": "茉莉", "lang": "中文", "gender": "女性"},
+    "苏打": {"id": "苏打", "lang": "中文", "gender": "男性"},
+    "白桦": {"id": "白桦", "lang": "中文", "gender": "男性"},
+    "Mia": {"id": "Mia", "lang": "英文", "gender": "女性"},
+    "Chloe": {"id": "Chloe", "lang": "英文", "gender": "女性"},
+    "Milo": {"id": "Milo", "lang": "英文", "gender": "男性"},
+    "Dean": {"id": "Dean", "lang": "英文", "gender": "男性"},
+}
+
+
+def text_to_speech(
+    text: str,
+    voice: str = "冰糖",
+    style: str = "",
+) -> bytes:
+    """调用 mimo-v2.5-tts 模型生成语音
+
+    Args:
+        text: 要合成的文本（放在 assistant 消息中）
+        voice: 预置音色名称
+        style: 风格指令（放在 user 消息中，可选）
+
+    Returns:
+        WAV 格式的音频字节数据
+    """
+    client = get_mimo_tts_client()
+    if not client:
+        raise RuntimeError("未配置 MIMO API Key")
+
+    messages = []
+
+    # user 消息：风格指令（可选）
+    if style:
+        messages.append({"role": "user", "content": style})
+    else:
+        messages.append({"role": "user", "content": ""})
+
+    # assistant 消息：要合成的目标文本
+    messages.append({"role": "assistant", "content": text})
+
+    completion = client.chat.completions.create(
+        model="mimo-v2.5-tts",
+        messages=messages,
+        audio={"format": "wav", "voice": voice},
+    )
+
+    message = completion.choices[0].message
+    audio_bytes = base64.b64decode(message.audio.data)
+    return audio_bytes

utils/tts_utils.py

@@ -0,0 +1,52 @@
+import re
+
+
+def generate_tts_script(text, speed="normal", add_pauses=True, add_emphasis=True):
+    """生成带停顿和标记的朗读脚本"""
+    lines = text.strip().split("\n")
+    result_lines = []
+
+    for line in lines:
+        line = line.strip()
+        if not line:
+            if add_pauses:
+                result_lines.append("[停顿 0.5秒]")
+            continue
+
+        # 检测标题行（数字开头或全大写或短行）
+        is_title = (
+            re.match(r'^[\d一二三四五六七八九十]+[、.．)\s]', line)
+            or (len(line) <= 15 and not line.endswith(("。", "，", "；", "！", "？")))
+        )
+
+        if is_title and add_emphasis:
+            result_lines.append(f"[强调] {line} [/强调]")
+            if add_pauses:
+                result_lines.append("[停顿 1秒]")
+            continue
+
+        # 处理标点停顿
+        processed = line
+        if add_pauses:
+            processed = processed.replace("。", "。[停顿 0.3秒]")
+            processed = processed.replace("！", "！[停顿 0.3秒]")
+            processed = processed.replace("？", "？[停顿 0.3秒]")
+            processed = processed.replace("；", "；[停顿 0.2秒]")
+
+        # 检测数字和英文，标记需放慢
+        if add_emphasis:
+            processed = re.sub(
+                r'(\d+(?:\.\d+)?(?:%|万|亿|元|个|条|项|次|天|年|月|日)?)',
+                r'[\1]',
+                processed
+            )
+
+        result_lines.append(processed)
+
+    script = "\n".join(result_lines)
+
+    # 添加头部说明
+    speed_map = {"slow": "慢速", "normal": "正常", "fast": "快速"}
+    header = f"[语速: {speed_map.get(speed, speed)}]\n\n"
+
+    return header + script