import streamlit as st from datetime import datetime import json import os import re from collections import Counter # 1. 页面配置 st.set_page_config(page_title="学习笔记整理器", page_icon="📚", layout="wide") st.title("📚 学习笔记整理器") # 文本提炼函数 def extract_key_content(content, max_sentences=3, max_keywords=5): """提炼笔记的关键内容""" if not content or len(content.strip()) < 10: return {"summary": "内容太短,无法提炼", "keywords": []} # 分句 sentences = re.split(r'[。!?\n]', content) sentences = [s.strip() for s in sentences if s.strip()] # 提取关键句子(基于句子长度和关键词密度) sentence_scores = [] for sentence in sentences: score = len(sentence) # 检查是否包含常见关键词 keywords = ['重要', '关键', '核心', '主要', '注意', '必须', '应该', '需要', '总结', '结论'] for kw in keywords: if kw in sentence: score += 20 sentence_scores.append((sentence, score)) # 按分数排序,取前几句 sentence_scores.sort(key=lambda x: x[1], reverse=True) top_sentences = [s[0] for s in sentence_scores[:max_sentences]] # 提取关键词 words = re.findall(r'[\u4e00-\u9fa5]{2,}', content) word_freq = Counter(words) # 过滤常见词 stop_words = ['的', '是', '在', '有', '和', '了', '不', '也', '都', '这', '那', '就', '会', '要', '可', '以', '我', '你', '他', '她', '它', '我们', '你们', '他们'] keywords = [w for w, c in word_freq.most_common(max_keywords * 2) if w not in stop_words][:max_keywords] return { "summary": "。".join(top_sentences) + "。" if top_sentences else "无法提取摘要", "keywords": keywords } # 2. 数据持久化函数 def save_notes(): """保存笔记到 JSON 文件""" data = { "notes": st.session_state.notes, "categories": st.session_state.categories } with open('notes_data.json', 'w', encoding='utf-8') as f: json.dump(data, f, ensure_ascii=False, indent=2) def load_notes(): """从 JSON 文件加载笔记""" if os.path.exists('notes_data.json'): with open('notes_data.json', 'r', encoding='utf-8') as f: data = json.load(f) st.session_state.notes = data.get("notes", []) st.session_state.categories = data.get("categories", ["编程", "数学", "英语", "其他"]) # 3. 初始化数据 if "notes" not in st.session_state: st.session_state.notes = [] load_notes() if "categories" not in st.session_state: st.session_state.categories = ["编程", "数学", "英语", "其他"] load_notes() # 初始化编辑状态 if "editing_note" not in st.session_state: st.session_state.editing_note = None # 4. 左侧侧边栏 - 添加笔记和管理类型 with st.sidebar: st.header("📝 添加笔记") # 笔记标题 title = st.text_input("标题", placeholder="输入笔记标题") # 分类选择(支持自定义) category = st.selectbox( "选择分类", st.session_state.categories, help="选择笔记的分类" ) # 自定义分类输入 new_category = st.text_input( "或创建新分类", placeholder="输入新分类名称,按回车添加", key="new_category_input" ) # 添加新分类按钮 if st.button("➕ 添加新分类"): if new_category and new_category not in st.session_state.categories: st.session_state.categories.append(new_category) save_notes() st.success(f"已添加新分类:{new_category}") st.rerun() elif new_category in st.session_state.categories: st.warning("该分类已存在") # 笔记内容 content = st.text_area( "笔记内容", placeholder="输入笔记内容...", height=200 ) # 标签输入 tags_input = st.text_input( "标签", placeholder="输入标签,用逗号分隔 (例如: 重要, 复习, 考试)", help="为笔记添加多个标签,便于分类和搜索" ) tags = [tag.strip() for tag in tags_input.split(",") if tag.strip()] # 实时预览提炼结果 if content and len(content.strip()) > 10: with st.expander("🔍 预览提炼结果", expanded=False): extracted = extract_key_content(content) st.markdown("**摘要预览:**") st.info(extracted['summary']) st.markdown("**关键词预览:**") if extracted['keywords']: keywords_html = " ".join([f'{kw}' for kw in extracted['keywords']]) st.markdown(keywords_html, unsafe_allow_html=True) # 添加按钮 if st.button("添加笔记", type="primary"): if title and content: note = { "id": len(st.session_state.notes) + 1, "title": title, "category": category, "content": content, "tags": tags, "created_at": datetime.now().strftime("%Y-%m-%d %H:%M:%S"), "updated_at": datetime.now().strftime("%Y-%m-%d %H:%M:%S") } st.session_state.notes.append(note) save_notes() st.success("笔记添加成功!") else: st.warning("请填写标题和内容") st.divider() # 显示所有分类 st.subheader("📂 所有分类") for cat in st.session_state.categories: st.markdown(f"- {cat}") # 5. 主界面 - 笔记展示和搜索 st.header("📖 我的笔记") # 搜索和排序区域 col1, col2, col3, col4, col5 = st.columns([2, 1, 1, 1, 1]) with col1: # 搜索关键词 search_query = st.text_input( "🔍 搜索关键词", placeholder="输入关键词...", help="支持模糊搜索" ) with col2: # 搜索范围选择 search_scope = st.selectbox( "搜索范围", ["全部", "仅标题", "仅内容", "仅类型", "仅标签"], help="选择搜索的范围" ) with col3: # 分类筛选 filter_category = st.selectbox( "筛选分类", ["全部"] + st.session_state.categories, help="按分类筛选笔记" ) with col4: # 标签筛选 all_tags = list(set(tag for note in st.session_state.notes for tag in note.get("tags", []))) filter_tag = st.selectbox( "筛选标签", ["全部"] + sorted(all_tags), help="按标签筛选笔记" ) with col5: # 排序方式 sort_by = st.selectbox( "排序方式", ["最新", "最旧", "标题A-Z", "标题Z-A", "类型A-Z", "类型Z-A"], help="选择笔记的排序方式" ) # 批量提炼功能 st.divider() col_batch1, col_batch2 = st.columns([3, 1]) with col_batch1: st.markdown("### ✨ 批量提炼") with col_batch2: if st.button("🔄 提炼所有笔记", key="batch_extract"): for note in st.session_state.notes: if len(note['content'].strip()) > 10: extracted = extract_key_content(note['content']) st.session_state[f'extracted_{note["id"]}'] = extracted st.success(f"已提炼 {len(st.session_state.notes)} 条笔记的关键内容!") st.rerun() # 6. 显示笔记 if st.session_state.notes: # 筛选和搜索逻辑 filtered_notes = [] for note in st.session_state.notes: # 分类筛选 category_match = filter_category == "全部" or note["category"] == filter_category # 标签筛选 tag_match = filter_tag == "全部" or filter_tag in note.get("tags", []) # 关键词搜索(根据选择范围) search_match = True if search_query: query = search_query.lower() if search_scope == "全部": search_match = ( query in note["title"].lower() or query in note["content"].lower() or query in note["category"].lower() or any(query in tag.lower() for tag in note.get("tags", [])) ) elif search_scope == "仅标题": search_match = query in note["title"].lower() elif search_scope == "仅内容": search_match = query in note["content"].lower() elif search_scope == "仅类型": search_match = query in note["category"].lower() elif search_scope == "仅标签": search_match = any(query in tag.lower() for tag in note.get("tags", [])) if category_match and tag_match and search_match: filtered_notes.append(note) # 排序逻辑 if sort_by == "最新": filtered_notes.sort(key=lambda x: x["created_at"], reverse=True) elif sort_by == "最旧": filtered_notes.sort(key=lambda x: x["created_at"]) elif sort_by == "标题A-Z": filtered_notes.sort(key=lambda x: x["title"].lower()) elif sort_by == "标题Z-A": filtered_notes.sort(key=lambda x: x["title"].lower(), reverse=True) elif sort_by == "类型A-Z": filtered_notes.sort(key=lambda x: x["category"].lower()) elif sort_by == "类型Z-A": filtered_notes.sort(key=lambda x: x["category"].lower(), reverse=True) # 显示筛选结果 if filtered_notes: st.info(f"找到 {len(filtered_notes)} 条笔记") # 使用两列布局展示笔记 cols = st.columns(2) for i, note in enumerate(filtered_notes): with cols[i % 2]: # 检查是否正在编辑此笔记 if st.session_state.editing_note == note['id']: with st.expander(f"✏️ 编辑: {note['title']}", expanded=True): # 编辑表单 edit_title = st.text_input("标题", value=note['title'], key=f"edit_title_{note['id']}") edit_category = st.selectbox( "分类", st.session_state.categories, index=st.session_state.categories.index(note['category']) if note['category'] in st.session_state.categories else 0, key=f"edit_category_{note['id']}" ) edit_content = st.text_area("内容", value=note['content'], height=200, key=f"edit_content_{note['id']}") # 编辑标签 current_tags = ", ".join(note.get("tags", [])) edit_tags = st.text_input( "标签", value=current_tags, key=f"edit_tags_{note['id']}", help="输入标签,用逗号分隔" ) edit_tags_list = [tag.strip() for tag in edit_tags.split(",") if tag.strip()] # 保存和取消按钮 col_save, col_cancel = st.columns(2) with col_save: if st.button("💾 保存修改", key=f"save_{note['id']}", type="primary"): note['title'] = edit_title note['category'] = edit_category note['content'] = edit_content note['tags'] = edit_tags_list note['updated_at'] = datetime.now().strftime("%Y-%m-%d %H:%M:%S") st.session_state.editing_note = None save_notes() st.success("笔记已更新!") st.rerun() with col_cancel: if st.button("❌ 取消", key=f"cancel_{note['id']}"): st.session_state.editing_note = None st.rerun() else: # 正常显示笔记 with st.expander(f"📌 {note['title']}", expanded=False): # 显示笔记元信息 st.markdown(f"**分类:** {note['category']}") # 显示标签 if note.get("tags"): tags_html = " ".join([f'{tag}' for tag in note["tags"]]) st.markdown(f"**标签:** {tags_html}", unsafe_allow_html=True) st.markdown(f"**创建时间:** {note['created_at']}") if 'updated_at' in note: st.markdown(f"**更新时间:** {note['updated_at']}") st.markdown("---") # 提炼内容按钮 if st.button("✨ 提炼内容", key=f"extract_{note['id']}"): extracted = extract_key_content(note['content']) st.session_state[f'extracted_{note["id"]}'] = extracted # 显示提炼结果 if f'extracted_{note["id"]}' in st.session_state: extracted = st.session_state[f'extracted_{note["id"]}'] with st.expander("📝 提炼结果", expanded=True): st.markdown("**摘要:**") st.info(extracted['summary']) st.markdown("**关键词:**") if extracted['keywords']: keywords_html = " ".join([f'{kw}' for kw in extracted['keywords']]) st.markdown(keywords_html, unsafe_allow_html=True) else: st.warning("未找到关键词") # 显示笔记内容 st.markdown(note['content']) # 操作按钮 col_edit, col_delete = st.columns(2) with col_edit: if st.button("✏️ 编辑", key=f"edit_{note['id']}"): st.session_state.editing_note = note['id'] st.rerun() with col_delete: if st.button("🗑️ 删除", key=f"delete_{note['id']}"): st.session_state.notes.remove(note) st.session_state.editing_note = None save_notes() st.rerun() else: st.warning("没有找到匹配的笔记") else: st.info("还没有笔记,请在左侧添加您的第一条笔记!") # 7. 统计信息 st.divider() st.subheader("📊 统计信息") # 笔记和分类统计 col_stats1, col_stats2 = st.columns(2) with col_stats1: st.metric("总笔记数", len(st.session_state.notes)) with col_stats2: all_tags = list(set(tag for note in st.session_state.notes for tag in note.get("tags", []))) st.metric("总标签数", len(all_tags)) # 分类统计 st.markdown("### 分类统计") cat_stat_cols = st.columns(min(5, len(st.session_state.categories))) for i, cat in enumerate(st.session_state.categories): if i < len(cat_stat_cols): with cat_stat_cols[i]: count = len([n for n in st.session_state.notes if n["category"] == cat]) st.metric(cat, count) # 标签统计 if all_tags: st.markdown("### 标签统计") tag_stat_cols = st.columns(min(5, len(all_tags))) for i, tag in enumerate(sorted(all_tags)): if i < len(tag_stat_cols): with tag_stat_cols[i]: count = len([n for n in st.session_state.notes if tag in n.get("tags", [])]) st.metric(tag, count) # 8. 数据管理 st.divider() st.subheader("💾 数据管理") # 导入JSON文件 st.markdown("### 📥 导入笔记") tab_json, tab_txt = st.tabs(["JSON文件", "文本文件"]) with tab_json: uploaded_file = st.file_uploader("选择JSON文件", type=['json'], help="选择之前导出的JSON文件", key="json_uploader") if uploaded_file is not None: try: # 读取并解析JSON文件 import_data = json.load(uploaded_file) # 显示导入预览 st.info(f"文件包含 {len(import_data.get('notes', []))} 条笔记") # 显示笔记列表预览 with st.expander("📋 查看文件内容", expanded=False): for i, note in enumerate(import_data.get('notes', [])[:5]): st.markdown(f"**{i+1}. {note.get('title', '无标题')}**") st.markdown(f" 分类: {note.get('category', '未知')}") st.markdown(f" 时间: {note.get('created_at', '未知')}") st.markdown("---") if len(import_data.get('notes', [])) > 5: st.info(f"还有 {len(import_data.get('notes', [])) - 5} 条笔记...") # 导入选项 col_merge, col_replace = st.columns(2) with col_merge: if st.button("🔄 合并导入", type="primary", key="merge_json"): # 合并导入:保留现有笔记,添加新笔记 new_notes = import_data.get('notes', []) for note in new_notes: note['id'] = len(st.session_state.notes) + 1 st.session_state.notes.append(note) # 合并分类 new_categories = import_data.get('categories', []) for cat in new_categories: if cat not in st.session_state.categories: st.session_state.categories.append(cat) save_notes() st.success(f"成功导入 {len(new_notes)} 条笔记!") st.rerun() with col_replace: if st.button("⚠️ 替换导入", key="replace_json"): # 替换导入:清空现有笔记,导入新笔记 st.session_state.notes = import_data.get('notes', []) st.session_state.categories = import_data.get('categories', ["编程", "数学", "英语", "其他"]) st.session_state.editing_note = None save_notes() st.success(f"成功替换为 {len(st.session_state.notes)} 条笔记!") st.rerun() except Exception as e: st.error(f"文件解析失败: {str(e)}") with tab_txt: # 导入文本文件 txt_file = st.file_uploader("选择文本文件", type=['txt'], help="选择要导入的文本文件", key="txt_uploader") if txt_file is not None: try: # 读取文本文件内容 content = txt_file.read().decode('utf-8') # 显示文件内容预览 st.info(f"文件大小: {len(content)} 字符") with st.expander("📋 查看文件内容", expanded=False): st.text(content[:1000]) if len(content) > 1000: st.info(f"还有 {len(content) - 1000} 个字符...") # 设置导入参数 col_title, col_cat = st.columns(2) with col_title: import_title = st.text_input("笔记标题", value=txt_file.name.replace('.txt', ''), help="为导入的笔记设置标题") with col_cat: import_category = st.selectbox("笔记分类", st.session_state.categories) # 导入按钮 if st.button("📥 导入文本文件", type="primary"): note = { "id": len(st.session_state.notes) + 1, "title": import_title, "category": import_category, "content": content, "created_at": datetime.now().strftime("%Y-%m-%d %H:%M:%S"), "updated_at": datetime.now().strftime("%Y-%m-%d %H:%M:%S") } st.session_state.notes.append(note) save_notes() st.success(f"成功导入笔记:{import_title}") st.rerun() except Exception as e: st.error(f"文件读取失败: {str(e)}") st.divider() # 导出功能 st.markdown("### 📤 导出笔记") export_tab_json, export_tab_txt = st.tabs(["JSON格式", "文本格式"]) with export_tab_json: if st.button("📤 导出为JSON", key="export_json"): st.download_button( label="下载 JSON 文件", data=json.dumps({"notes": st.session_state.notes, "categories": st.session_state.categories}, ensure_ascii=False, indent=2), file_name=f"notes_backup_{datetime.now().strftime('%Y%m%d_%H%M%S')}.json", mime="application/json" ) with export_tab_txt: # 导出选项 export_option = st.radio("导出选项", ["导出所有笔记", "导出单个笔记"]) if export_option == "导出所有笔记": if st.button("📤 导出所有笔记为文本", type="primary", key="export_all_txt"): # 合并所有笔记为一个文本文件 all_content = "=" * 50 + "\n" all_content += f"学习笔记备份 - {datetime.now().strftime('%Y-%m-%d %H:%M:%S')}\n" all_content += "=" * 50 + "\n\n" for note in st.session_state.notes: all_content += "-" * 40 + "\n" all_content += f"标题: {note['title']}\n" all_content += f"分类: {note['category']}\n" all_content += f"创建时间: {note['created_at']}\n" all_content += "-" * 40 + "\n\n" all_content += note['content'] + "\n\n" st.download_button( label="下载文本文件", data=all_content, file_name=f"notes_all_{datetime.now().strftime('%Y%m%d_%H%M%S')}.txt", mime="text/plain" ) else: # 导出单个笔记 if st.session_state.notes: note_titles = [f"{note['id']}. {note['title']}" for note in st.session_state.notes] selected_note = st.selectbox("选择要导出的笔记", note_titles) if st.button("📤 导出选中笔记", type="primary", key="export_single_txt"): # 找到选中的笔记 note_id = int(selected_note.split('.')[0]) note = next((n for n in st.session_state.notes if n['id'] == note_id), None) if note: # 格式化笔记内容 note_content = "=" * 50 + "\n" note_content += f"标题: {note['title']}\n" note_content += f"分类: {note['category']}\n" note_content += f"创建时间: {note['created_at']}\n" if 'updated_at' in note: note_content += f"更新时间: {note['updated_at']}\n" note_content += "=" * 50 + "\n\n" note_content += note['content'] st.download_button( label="下载文本文件", data=note_content, file_name=f"note_{note['id']}_{note['title']}.txt", mime="text/plain" ) else: st.info("没有笔记可导出") st.divider() # 清空功能 col_clear = st.columns(1)[0] with col_clear: if st.button("🗑️ 清空所有笔记"): if st.session_state.notes: st.warning("⚠️ 此操作不可恢复!") if st.button("确认清空"): st.session_state.notes = [] st.session_state.editing_note = None save_notes() st.success("所有笔记已清空!") st.rerun() else: st.info("没有笔记可清空")