2

app.py

@@ -1,223 +1,341 @@
 """
-图表问答数据集审核系统 - Gradio 5.x 稳定版
+图表问答数据集审核系统 - Gradio 5.x 应用
 """
 import gradio as gr
-from data_manager import data_manager # 假设你已经实例化了 data_manager
+from data_manager import DataManager, data_manager
+from typing import Dict, List, Optional, Tuple, Any
 import json
+import os
 import base64
-from typing import Dict, List, Optional
+import tempfile
 
-# ============== 全局状态管理 ==============
+# ============== 全局状态 ==============
 
 class AppState:
     def __init__(self):
-        self.current_source = ""
-        self.current_chart_type = ""
-        self.current_chart_id = ""
-        self.current_model = ""
-        self.all_paths = []
-        self.current_index = -1
+        self.current_source: str = ""
+        self.current_chart_type: str = ""
+        self.current_chart_id: str = ""
+        self.current_model: str = ""
+        self.all_paths: List[Dict] = []
+        self.current_index: int = -1
         self.refresh_paths()
     
     def refresh_paths(self):
         self.all_paths = data_manager.get_all_chart_paths()
     
-    def set_position(self, source, chart_type, chart_id, model):
+    def get_current_path(self) -> Optional[Dict]:
+        if 0 <= self.current_index < len(self.all_paths):
+            return self.all_paths[self.current_index]
+        return None
+    
+    def set_position(self, source: str, chart_type: str, chart_id: str, model: str):
         self.current_source = source
         self.current_chart_type = chart_type
         self.current_chart_id = chart_id
         self.current_model = model
         for i, path in enumerate(self.all_paths):
-            if (path['source'] == source and path['chart_type'] == chart_type and 
-                path['chart_id'] == chart_id and path['model'] == model):
+            if (path['source'] == source and 
+                path['chart_type'] == chart_type and 
+                path['chart_id'] == chart_id and 
+                path['model'] == model):
                 self.current_index = i
                 break
     
-    def navigate(self, direction):
+    def navigate(self, direction: int) -> bool:
         new_index = self.current_index + direction
         if 0 <= new_index < len(self.all_paths):
             self.current_index = new_index
             path = self.all_paths[new_index]
-            return path['source'], path['chart_type'], path['chart_id'], path['model']
-        return None
+            self.current_source = path['source']
+            self.current_chart_type = path['chart_type']
+            self.current_chart_id = path['chart_id']
+            self.current_model = path['model']
+            return True
+        return False
 
 state = AppState()
 
-# ============== 辅助逻辑 ==============
+# 当前 QA 数据缓存（使用简单全局变量，避免 gr.State 的问题）
+_current_qa_list: List = []
+_current_reviews: Dict = {}
 
-def create_embedded_html(html_content: str) -> str:
-    if not html_content:
-        return '<div style="height:600px;display:flex;align-items:center;justify-content:center;background:#eee;">请选择图表进行加载</div>'
-    # 使用 Base64 嵌入以避免跨域和字符编码问题
-    html_base64 = base64.b64encode(html_content.encode('utf-8')).decode('utf-8')
-    return f'<iframe src="data:text/html;base64,{html_base64}" style="width:100%;height:700px;border:none;background:white;"></iframe>'
+# ============== UI 更新函数 ==============
 
-# ============== UI 事件处理 ==============
+def init_dataset():
+    structure = data_manager.get_dataset_structure()
+    sources = list(structure.get('sources', {}).keys())
+    return gr.Dropdown(choices=sources, value=sources[0] if sources else None)
 
-def on_source_change(source):
+def update_chart_type_dropdown(source: str):
+    state.current_source = source
     structure = data_manager.get_dataset_structure()
-    types = list(structure.get('sources', {}).get(source, {}).get('chart_types', {}).keys())
-    return gr.update(choices=types, value=types[0] if types else None)
+    chart_types = list(structure.get('sources', {}).get(source, {}).get('chart_types', {}).keys())
+    return gr.Dropdown(choices=chart_types, value=chart_types[0] if chart_types else None)
 
-def on_type_change(source, chart_type):
+def update_chart_dropdown(source: str, chart_type: str):
+    state.current_source = source
+    state.current_chart_type = chart_type
     charts = data_manager.get_chart_list(source, chart_type)
     structure = data_manager.get_dataset_structure()
-    models = structure.get('sources', {}).get(source, {}).get('chart_types', {}).get(chart_type, {}).get('models', [])
+    ct_data = structure.get('sources', {}).get(source, {}).get('chart_types', {}).get(chart_type, {})
+    models = ct_data.get('models', [])
     return (
-        gr.update(choices=charts, value=charts[0] if charts else None),
-        gr.update(choices=models, value=models[0] if models else None)
+        gr.Dropdown(choices=charts, value=charts[0] if charts else None),
+        gr.Dropdown(choices=models, value=models[0] if models else None)
     )
 
-def load_data(source, chart_type, chart_id, model):
+def create_embedded_html(html_content: str, chart_id: str = "") -> str:
+    if not html_content:
+        return f'''<div style="display:flex;flex-direction:column;min-height:750px;align-items:center;justify-content:center;background:#fafafa;border:2px dashed #ddd;border-radius:12px;">
+            <div style="font-size:48px;margin-bottom:16px;">📭</div>
+            <div style="font-size:18px;color:#666;">暂无图表内容</div>
+            <div style="font-size:14px;color:#999;margin-top:8px;">ID: {chart_id}</div>
+        </div>'''
+    html_base64 = base64.b64encode(html_content.encode('utf-8')).decode('utf-8')
+    return f'<iframe src="data:text/html;base64,{html_base64}" style="width:100%;height:750px;border:1px solid #e0e0e0;border-radius:8px;background:#fff;" sandbox="allow-scripts allow-same-origin"></iframe>'
+
+def load_chart_data(source: str, chart_type: str, chart_id: str, model: str):
+    global _current_qa_list, _current_reviews
+    
     if not all([source, chart_type, chart_id, model]):
-        return ["", "", "[]", "等待选择...", "0/0", "{}", gr.update(choices=[]), "未选择"]
+        _current_qa_list = []
+        _current_reviews = {}
+        return (
+            create_embedded_html(""),
+            "### 请选择图表",
+            "等待加载...",
+            "请选择图表",
+            gr.Radio(choices=[], value=None),
+            ""
+        )
     
     state.set_position(source, chart_type, chart_id, model)
     chart_data = data_manager.get_chart_data(source, chart_type, chart_id)
+    html_content = chart_data.get('html_content', '')
+    label_info = chart_data.get('label_info', {})
+    
+    embedded_html = create_embedded_html(html_content, chart_id)
+    
+    # 格式化标签
+    if label_info:
+        label_lines = ["### 📝 图表信息", ""]
+        for k, v in label_info.items():
+            label_lines.append(f"**{k}**: {v}")
+        label_text = "\n".join(label_lines)
+    else:
+        label_text = "### ⚠️ 暂无标签信息"
+    
+    # 获取 QA 数据
     qa_list = data_manager.get_qa_list(source, chart_type, model, chart_id)
-    reviews = {r['qa_id']: r for r in data_manager.get_reviews_by_chart(chart_id, model)}
+    _current_qa_list = [{"id": qa.id, "question": qa.question, "answer": qa.answer} for qa in qa_list]
+    
+    # 获取审核记录
+    reviews = data_manager.get_reviews_by_chart(chart_id, model)
+    _current_reviews = {r['qa_id']: r for r in reviews}
+    
+    # 统计
     stats = data_manager.get_review_stats()
+    status_msg = f"已审核: {stats['total']} | ✅正确: {stats['correct']} | ❌错误: {stats['incorrect']} | ✏️需修改: {stats['needs_modification']}"
+    progress_msg = f"进度: {state.current_index + 1} / {len(state.all_paths)}"
     
-    # 构造返回值
-    html_frame = create_embedded_html(chart_data.get('html_content', ''))
-    label_md = "\n".join([f"- **{k}**: {v}" for k, v in chart_data.get('label_info', {}).items()])
-    qa_json = json.dumps([{"id": q.id, "question": q.question, "answer": q.answer} for q in qa_list])
-    stats_msg = f"✅{stats['correct']} | ❌{stats['incorrect']} | 总{stats['total']}"
-    progress_msg = f"{state.current_index + 1} / {len(state.all_paths)}"
-    qa_choices = [f"Q{i+1}: {q.question[:30]}..." for i, q in enumerate(qa_list)]
+    qa_choices = [f"Q{i+1}: {qa.question[:40]}..." for i, qa in enumerate(qa_list)]
     
-    return [
-        html_frame, label_md, qa_json, stats_msg, progress_msg, 
-        json.dumps(reviews), gr.update(choices=qa_choices, value=qa_choices[0] if qa_choices else None),
-        f"{source}/{chart_type}/{chart_id}"
-    ]
+    debug_info = f"📁 {source}/{chart_type}/{chart_id} | HTML: {len(html_content)} 字符"
+    
+    return (
+        embedded_html,
+        label_text,
+        status_msg,
+        progress_msg,
+        gr.Radio(choices=qa_choices, value=qa_choices[0] if qa_choices else None),
+        debug_info
+    )
 
-# ============== 界面构建 ==============
+def on_qa_select(qa_index_str: str):
+    """选择 QA 后更新审核面板"""
+    global _current_qa_list, _current_reviews
+    
+    if not qa_index_str or not _current_qa_list:
+        return ("", "", "", "correct", "", "", "", "")
+    
+    try:
+        idx = int(qa_index_str.split(":")[0].replace("Q", "")) - 1
+        qa = _current_qa_list[idx]
+        review = _current_reviews.get(qa['id'], {})
+        
+        return (
+            qa['id'],
+            qa['question'],
+            qa['answer'],
+            review.get('status', 'correct'),
+            review.get('issue_type', ''),
+            review.get('modified_question', ''),
+            review.get('modified_answer', ''),
+            review.get('comment', '')
+        )
+    except Exception as e:
+        print(f"Error: {e}")
+        return ("", "", "", "correct", "", "", "", "")
+
+def navigate_prev():
+    if state.navigate(-1):
+        path = state.get_current_path()
+        if path:
+            return (
+                gr.Dropdown(value=path['source']),
+                gr.Dropdown(value=path['chart_type']),
+                gr.Dropdown(value=path['chart_id']),
+                gr.Dropdown(value=path['model'])
+            )
+    return (gr.Dropdown(), gr.Dropdown(), gr.Dropdown(), gr.Dropdown())
+
+def navigate_next():
+    if state.navigate(1):
+        path = state.get_current_path()
+        if path:
+            return (
+                gr.Dropdown(value=path['source']),
+                gr.Dropdown(value=path['chart_type']),
+                gr.Dropdown(value=path['chart_id']),
+                gr.Dropdown(value=path['model'])
+            )
+    return (gr.Dropdown(), gr.Dropdown(), gr.Dropdown(), gr.Dropdown())
+
+def save_review_handler(qa_id, chart_id, source, chart_type, model, orig_q, orig_a, status, mod_q, mod_a, issue_type, comment, reviewer):
+    if not qa_id:
+        return "❌ 请先选择问答对"
+    
+    data_manager.save_review({
+        "qa_id": qa_id,
+        "chart_id": chart_id,
+        "source": source,
+        "chart_type": chart_type,
+        "model": model,
+        "original_question": orig_q,
+        "original_answer": orig_a,
+        "status": status,
+        "modified_question": mod_q,
+        "modified_answer": mod_a,
+        "issue_type": issue_type,
+        "comment": comment,
+        "reviewer": reviewer
+    })
+    
+    # 更新缓存
+    global _current_reviews
+    _current_reviews[qa_id] = {
+        "qa_id": qa_id, "status": status, "issue_type": issue_type,
+        "modified_question": mod_q, "modified_answer": mod_a, "comment": comment
+    }
+    
+    stats = data_manager.get_review_stats()
+    return f"✅ 已保存! 总计: {stats['total']} | ✅正确: {stats['correct']} | ❌错误: {stats['incorrect']}"
+
+def export_reviews_handler():
+    reviews = data_manager.get_all_reviews()
+    if not reviews:
+        return None
+    temp_path = os.path.join(tempfile.gettempdir(), "reviews_export.json")
+    with open(temp_path, 'w', encoding='utf-8') as f:
+        json.dump(reviews, f, ensure_ascii=False, indent=2)
+    return temp_path
+
+# ============== 创建 Gradio 界面 ==============
 
 def create_ui():
-    with gr.Blocks(title="图表 QA 审核系统", theme=gr.themes.Default()) as app:
-        # 隐藏状态存储
-        qa_store = gr.State("[]")
-        review_store = gr.State("{}")
+    custom_css = """
+    .chart-container { min-height: 780px; }
+    .control-panel { background: #f8f9fa; padding: 12px; border-radius: 8px; margin-bottom: 8px; }
+    """
+    
+    with gr.Blocks(title="图表问答数据集审核系统", theme=gr.themes.Soft(), css=custom_css) as app:
+        gr.Markdown("# 📊 图表问答数据集审核系统\n\n审核图表问答对，使用 ← → 按钮切换图表")
         
-        gr.Markdown("# 📊 图表问答数据集审核")
+        # 顶部状态
+        with gr.Row():
+            status_text = gr.Textbox(label="统计", interactive=False, scale=2)
+            progress_text = gr.Textbox(label="进度", interactive=False, scale=1)
         
+        # 导航栏
         with gr.Row():
-            with gr.Column(scale=3):
-                with gr.Row():
-                    source_drop = gr.Dropdown(label="数据源", choices=[], allow_custom_value=True)
-                    type_drop = gr.Dropdown(label="类型", choices=[], allow_custom_value=True)
-                    id_drop = gr.Dropdown(label="图表 ID", choices=[], allow_custom_value=True)
-                    model_drop = gr.Dropdown(label="模型", choices=[], allow_custom_value=True)
-                
-                html_view = gr.HTML()
-                path_display = gr.Text(label="当前路径", interactive=False)
-                
-            with gr.Column(scale=1):
-                stats_view = gr.Textbox(label="审核统计", interactive=False)
-                progress_view = gr.Textbox(label="总进度", interactive=False)
+            source_drop = gr.Dropdown(label="数据来源", choices=[], interactive=True, scale=1)
+            type_drop = gr.Dropdown(label="图表类型", choices=[], interactive=True, scale=1)
+            id_drop = gr.Dropdown(label="图表ID", choices=[], interactive=True, scale=1)
+            model_drop = gr.Dropdown(label="模型", choices=[], interactive=True, scale=1)
+            reviewer = gr.Textbox(label="审核人", value="admin", interactive=True, scale=1)
+        
+        with gr.Row():
+            prev_btn = gr.Button("⬅️ 上一个", size="sm")
+            next_btn = gr.Button("➡️ 下一个", size="sm")
+            export_btn = gr.Button("📦 导出记录", variant="secondary", size="sm")
+            download_file = gr.File(label="下载", visible=False)
+        
+        # 主内容
+        with gr.Row():
+            # 左侧：图表
+            with gr.Column(scale=6):
+                html_display = gr.HTML(elem_classes=["chart-container"])
+                debug_info = gr.Textbox(label="调试", interactive=False, show_label=False)
+            
+            # 右侧：审核
+            with gr.Column(scale=4):
+                with gr.Accordion("📝 图表元数据", open=False):
+                    label_display = gr.Markdown()
                 
-                with gr.Accordion("图表元数据", open=False):
-                    meta_view = gr.Markdown()
+                gr.Markdown("### ❓ 问答审核")
+                qa_selector = gr.Radio(label="选择问答对", choices=[], interactive=True)
                 
-                gr.Markdown("### 审核区")
-                qa_select = gr.Radio(label="选择题目", choices=[])
-                curr_qa_id = gr.Text(visible=False)
-                q_text = gr.Textbox(label="问题", interactive=False, lines=2)
-                a_text = gr.Textbox(label="答案", interactive=False)
+                current_qa_id = gr.Textbox(visible=False, value="")
+                orig_q = gr.Textbox(label="原始问题", interactive=False, lines=2)
+                orig_a = gr.Textbox(label="原始答案", interactive=False)
                 
+                gr.Markdown("---")
                 status_radio = gr.Radio(
-                    label="判定", 
-                    choices=[("正确", "correct"), ("错误", "incorrect"), ("改后通过", "modified")],
-                    value="correct"
+                    label="判定结果",
+                    choices=[("✅ 正确", "correct"), ("❌ 错误", "incorrect"), ("✏️ 需修改", "needs_modification")],
+                    value="correct",
+                    interactive=True
+                )
+                issue_type = gr.Dropdown(
+                    label="错误类型",
+                    choices=["答案错误", "问题模糊", "图文不符", "其他"],
+                    interactive=True
                 )
-                issue_type = gr.Dropdown(label="错误类型", choices=["无", "答案错误", "问题模糊", "图表读取失败"])
-                mod_q = gr.Textbox(label="修改问题(选填)")
-                mod_a = gr.Textbox(label="修改答案(选填)")
-                comment = gr.Textbox(label="备注")
+                mod_q = gr.Textbox(label="修改后问题", lines=2, interactive=True)
+                mod_a = gr.Textbox(label="修改后答案", interactive=True)
+                comment = gr.Textbox(label="备注", lines=2, interactive=True)
                 
                 save_btn = gr.Button("💾 保存审核", variant="primary")
-                with gr.Row():
-                    prev_btn = gr.Button("⬅️ 上一个")
-                    next_btn = gr.Button("➡️ 下一个")
-                
-                export_btn = gr.Button("📦 导出 JSON")
-                download_file = gr.File(label="下载地址", visible=False)
-
-        # --- 交互绑定 ---
+                save_msg = gr.Textbox(label="", visible=False)
         
-        # 初始化
-        def init_ui():
-            structure = data_manager.get_dataset_structure()
-            sources = list(structure.get('sources', {}).keys())
-            return gr.update(choices=sources, value=sources[0] if sources else None)
+        # ===== 事件绑定 =====
+        app.load(fn=init_dataset, outputs=[source_drop])
         
-        app.load(init_ui, outputs=[source_drop])
+        source_drop.change(fn=update_chart_type_dropdown, inputs=[source_drop], outputs=[type_drop])
+        type_drop.change(fn=update_chart_dropdown, inputs=[source_drop, type_drop], outputs=[id_drop, model_drop])
         
-        # 联动逻辑
-        source_drop.change(on_source_change, inputs=[source_drop], outputs=[type_drop])
-        type_drop.change(on_type_change, inputs=[source_drop, type_drop], outputs=[id_drop, model_drop])
+        load_outputs = [html_display, label_display, status_text, progress_text, qa_selector, debug_info]
+        id_drop.change(fn=load_chart_data, inputs=[source_drop, type_drop, id_drop, model_drop], outputs=load_outputs)
+        model_drop.change(fn=load_chart_data, inputs=[source_drop, type_drop, id_drop, model_drop], outputs=load_outputs)
+        
+        qa_outputs = [current_qa_id, orig_q, orig_a, status_radio, issue_type, mod_q, mod_a, comment]
+        qa_selector.change(fn=on_qa_select, inputs=[qa_selector], outputs=qa_outputs)
+        
+        nav_outputs = [source_drop, type_drop, id_drop, model_drop]
+        prev_btn.click(fn=navigate_prev, outputs=nav_outputs)
+        next_btn.click(fn=navigate_next, outputs=nav_outputs)
         
-        # 加载数据
-        load_inputs = [source_drop, type_drop, id_drop, model_drop]
-        load_outputs = [html_view, meta_view, qa_store, stats_view, progress_view, review_store, qa_select, path_display]
-        id_drop.change(load_data, inputs=load_inputs, outputs=load_outputs)
-        model_drop.change(load_data, inputs=load_inputs, outputs=load_outputs)
-
-        # 题目切换
-        def switch_qa(sel, qa_json, rev_json):
-            if not sel or qa_json == "[]": return [""] * 8
-            qas = json.loads(qa_json)
-            revs = json.loads(rev_json)
-            idx = int(sel.split(":")[0][1:]) - 1
-            curr = qas[idx]
-            r = revs.get(curr['id'], {})
-            return [
-                curr['id'], curr['question'], curr['answer'],
-                r.get('status', 'correct'), r.get('issue_type', '无'),
-                r.get('modified_question', ''), r.get('modified_answer', ''),
-                r.get('comment', '')
-            ]
-
-        qa_select.change(
-            switch_qa, 
-            inputs=[qa_select, qa_store, review_store], 
-            outputs=[curr_qa_id, q_text, a_text, status_radio, issue_type, mod_q, mod_a, comment]
-        )
-
-        # 导航
-        def nav(direction):
-            res = state.navigate(direction)
-            if res:
-                return [gr.update(value=v) for v in res]
-            return [gr.update() for _ in range(4)]
-
-        prev_btn.click(lambda: nav(-1), outputs=load_inputs)
-        next_btn.click(lambda: nav(1), outputs=load_inputs)
-
-        # 保存
-        def save_handler(q_id, c_id, src, tp, mdl, o_q, o_a, stat, m_q, m_a, it, cmt):
-            data_manager.save_review({
-                "qa_id": q_id, "chart_id": c_id, "source": src, "chart_type": tp, "model": mdl,
-                "original_question": o_q, "original_answer": o_a, "status": stat,
-                "modified_question": m_q, "modified_answer": m_a, "issue_type": it, "comment": cmt
-            })
-            return "已保存"
-
         save_btn.click(
-            save_handler,
-            inputs=[curr_qa_id, id_drop, source_drop, type_drop, model_drop, q_text, a_text, status_radio, mod_q, mod_a, issue_type, comment],
-            outputs=[gr.Textbox(visible=False)] # 仅触发
-        )
+            fn=save_review_handler,
+            inputs=[current_qa_id, id_drop, source_drop, type_drop, model_drop, orig_q, orig_a, status_radio, mod_q, mod_a, issue_type, comment, reviewer],
+            outputs=[save_msg]
+        ).then(fn=lambda: gr.Textbox(visible=True), outputs=[save_msg])
         
-        export_btn.click(
-            fn=lambda: gr.update(value=data_manager.export_reviews(), visible=True),
-            outputs=[download_file]
-        )
-
-        return app # 必须要 return app 
+        export_btn.click(fn=export_reviews_handler, outputs=[download_file]).then(fn=lambda: gr.File(visible=True), outputs=[download_file])
+    
+    return app
 
 if __name__ == "__main__":
-    demo = create_ui()
-    # show_api=False 是解决那个 TypeError 的关键
-    demo.launch(server_name="0.0.0.0", server_port=7860, show_api=False)
+    app = create_ui()
+    app.launch(server_name="0.0.0.0", server_port=7860, show_api=False)