2

app.py

@@ -1,341 +1,208 @@
-"""
-图表问答数据集审核系统 - Gradio 5.x 应用
-"""
 import gradio as gr
-from data_manager import DataManager, data_manager
-from typing import Dict, List, Optional, Tuple, Any
 import json
-import os
 import base64
-import tempfile
+from data_manager import data_manager
 
-# ============== 全局状态 ==============
-
-class AppState:
+# ============== 状态管理 (封装逻辑) ==============
+class ReviewState:
     def __init__(self):
-        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()
+        self.all_paths = []
+        self.current_idx = -1
     
-    def refresh_paths(self):
+    def sync_paths(self):
         self.all_paths = data_manager.get_all_chart_paths()
     
-    def get_current_path(self) -> Optional[Dict]:
-        if 0 <= self.current_index < len(self.all_paths):
-            return self.all_paths[self.current_index]
+    def get_nav_target(self, direction):
+        new_idx = self.current_idx + direction
+        if 0 <= new_idx < len(self.all_paths):
+            self.current_idx = new_idx
+            return self.all_paths[new_idx]
         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):
-                self.current_index = i
-                break
-    
-    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]
-            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()
 
-# 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 = {}
+nav_state = ReviewState()
 
-# ============== UI 更新函数 ==============
-
-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 update_chart_type_dropdown(source: str):
-    state.current_source = source
-    structure = data_manager.get_dataset_structure()
-    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 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()
-    ct_data = structure.get('sources', {}).get(source, {}).get('chart_types', {}).get(chart_type, {})
-    models = ct_data.get('models', [])
-    return (
-        gr.Dropdown(choices=charts, value=charts[0] if charts else None),
-        gr.Dropdown(choices=models, value=models[0] if models else None)
-    )
-
-def create_embedded_html(html_content: str, chart_id: str = "") -> str:
+# ============== 工具函数 ==============
+def to_html_frame(html_content):
     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]):
-        _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)
-    _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)}"
-    
-    qa_choices = [f"Q{i+1}: {qa.question[:40]}..." for i, qa in enumerate(qa_list)]
-    
-    debug_info = f"📁 {source}/{chart_type}/{chart_id} | HTML: {len(html_content)} 字符"
-    
+        return '<div style="padding:20px;text-align:center;">请选择数据进行加载</div>'
+    b64_content = base64.b64encode(html_content.encode('utf-8')).decode('utf-8')
+    return f'<iframe src="data:text/html;base64,{b64_content}" style="width:100%;height:600px;border:none;"></iframe>'
+
+# ============== 交互逻辑 ==============
+def handle_source_change(source):
+    struct = data_manager.get_dataset_structure()
+    types = list(struct.get('sources', {}).get(source, {}).get('chart_types', {}).keys())
+    # 默认选中第一个
+    return gr.update(choices=types, value=types[0] if types else None)
+
+def handle_type_change(source, c_type):
+    charts = data_manager.get_chart_list(source, c_type)
+    struct = data_manager.get_dataset_structure()
+    models = struct.get('sources', {}).get(source, {}).get('chart_types', {}).get(c_type, {}).get('models', [])
     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
+        gr.update(choices=charts, value=charts[0] if charts else None),
+        gr.update(choices=models, value=models[0] if models else None)
     )
 
-def on_qa_select(qa_index_str: str):
-    """选择 QA 后更新审核面板"""
-    global _current_qa_list, _current_reviews
+def handle_load(source, c_type, c_id, model):
+    if not all([source, c_type, c_id, model]):
+        # 必须返回 8 个 update 对象以匹配 outputs
+        return [gr.update()] * 8
     
-    if not qa_index_str or not _current_qa_list:
-        return ("", "", "", "correct", "", "", "", "")
+    # 获取数据
+    chart_data = data_manager.get_chart_data(source, c_type, c_id)
+    qa_list = data_manager.get_qa_list(source, c_type, model, c_id)
+    stats = data_manager.get_review_stats()
     
+    # 同步索引位置
+    nav_state.sync_paths()
+    for i, p in enumerate(nav_state.all_paths):
+        if p['chart_id'] == c_id and p['model'] == model:
+            nav_state.current_idx = i
+            break
+
+    # 准备 UI 数据
+    html_code = to_html_frame(chart_data.get('html_content', ''))
+    meta_md = "\n".join([f"- **{k}**: {v}" for k, v in chart_data.get('label_info', {}).items()])
+    qa_json = json.dumps([{"id": q.id, "q": q.question, "a": q.answer} for q in qa_list])
+    stats_str = f"✅{stats['correct']} | ❌{stats['incorrect']} | 总{stats['total']}"
+    prog_str = f"{nav_state.current_idx + 1} / {len(nav_state.all_paths)}"
+    radio_choices = [f"Q{i+1}: {q.question[:20]}..." for i, q in enumerate(qa_list)]
+    
+    return [
+        html_code, 
+        meta_md, 
+        qa_json, 
+        stats_str, 
+        prog_str, 
+        f"{source}/{c_type}/{c_id}",
+        gr.update(choices=radio_choices, value=radio_choices[0] if radio_choices else None),
+        json.dumps({}) # 占位 review_store
+    ]
+
+def handle_qa_select(selection, qa_json):
+    if not selection or not qa_json:
+        return [""] * 8
     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 ==============
-
+        qas = json.loads(qa_json)
+        idx = int(selection.split(":")[0][1:]) - 1
+        curr = qas[idx]
+        return [curr['id'], curr['q'], curr['a'], "correct", "无", "", "", ""]
+    except:
+        return [""] * 8
+
+# ============== UI 布局 ==============
 def create_ui():
-    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")
+    with gr.Blocks(title="审核系统 V2", theme=gr.themes.Soft()) as demo:
+        # 显式初始化 State，避免布尔值陷阱
+        qa_store = gr.State(value="[]")
+        review_store = gr.State(value="{}")
         
-        # 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():
-            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)
+        gr.Markdown("## 📑 图表问答数据集审核终端")
         
         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.Row():
+                    src_dd = gr.Dropdown(label="数据源", choices=["None"])
+                    typ_dd = gr.Dropdown(label="图表类型")
+                    id_dd = gr.Dropdown(label="图表 ID")
+                    mdl_dd = gr.Dropdown(label="评估模型")
                 
-                gr.Markdown("### ❓ 问答审核")
-                qa_selector = gr.Radio(label="选择问答对", choices=[], interactive=True)
+                chart_view = gr.HTML(value='<div style="height:500px; background:#f0f0f0;"></div>')
+                path_info = gr.Text(label="当前路径", 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)
+            with gr.Column(scale=2):
+                with gr.Group():
+                    stats_txt = gr.Text(label="统计信息", interactive=False)
+                    prog_txt = gr.Text(label="审核进度", interactive=False)
                 
+                with gr.Accordion("元数据解析", open=False):
+                    meta_md = gr.Markdown()
+
                 gr.Markdown("---")
-                status_radio = gr.Radio(
-                    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
-                )
-                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)
+                qa_radio = gr.Radio(label="题目列表", choices=[])
                 
-                save_btn = gr.Button("💾 保存审核", variant="primary")
-                save_msg = gr.Textbox(label="", visible=False)
-        
-        # ===== 事件绑定 =====
-        app.load(fn=init_dataset, 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])
-        
-        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)
+                with gr.Group():
+                    curr_qid = gr.Text(visible=False)
+                    q_disp = gr.Text(label="问题内容", lines=2)
+                    a_disp = gr.Text(label="标准答案")
+                    
+                    status_opt = gr.Radio(
+                        label="审核结论", 
+                        choices=[("正确", "correct"), ("错误", "incorrect"), ("优化", "modified")],
+                        value="correct"
+                    )
+                    err_type = gr.Dropdown(label="错误分类", choices=["无", "事实错误", "逻辑错误", "图表无法读取"])
+                    
+                comment = gr.Text(label="审核备注")
+                save_btn = gr.Button("💾 提交单条审核", variant="primary")
+                
+                with gr.Row():
+                    prev_btn = gr.Button("⬅️ 上一个图表")
+                    next_btn = gr.Button("➡️ 下一个图表")
+
+        # --- 事件绑定 ---
         
-        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)
+        # 初始化第一级
+        demo.load(
+            fn=lambda: gr.update(choices=list(data_manager.get_dataset_structure().get('sources', {}).keys())),
+            outputs=[src_dd]
+        )
+
+        # 联动逻辑
+        src_dd.change(handle_source_change, inputs=[src_dd], outputs=[typ_dd])
+        typ_dd.change(handle_type_change, inputs=[src_dd, typ_dd], outputs=[id_dd, mdl_dd])
         
+        # 加载数据 (严格匹配 8 个 output)
+        load_event_outputs = [chart_view, meta_md, qa_store, stats_txt, prog_txt, path_info, qa_radio, review_store]
+        id_dd.change(handle_load, inputs=[src_dd, typ_dd, id_dd, mdl_dd], outputs=load_event_outputs)
+        mdl_dd.change(handle_load, inputs=[src_dd, typ_dd, id_dd, mdl_dd], outputs=load_event_outputs)
+
+        # 题目切换
+        qa_radio.change(
+            handle_qa_select, 
+            inputs=[qa_radio, qa_store], 
+            outputs=[curr_qid, q_disp, a_disp, status_opt, err_type, gr.State(), gr.State(), comment] 
+        )
+
+        # 导航逻辑
+        def navigate(direction):
+            target = nav_state.get_nav_target(direction)
+            if target:
+                return [
+                    gr.update(value=target['source']),
+                    gr.update(value=target['chart_type']),
+                    gr.update(value=target['chart_id']),
+                    gr.update(value=target['model'])
+                ]
+            return [gr.update()] * 4
+
+        prev_btn.click(lambda: navigate(-1), outputs=[src_dd, typ_dd, id_dd, mdl_dd])
+        next_btn.click(lambda: navigate(1), outputs=[src_dd, typ_dd, id_dd, mdl_dd])
+
+        # 保存逻辑
+        def quick_save(qid, cid, src, status, cmt):
+            if not qid: return "无效操作"
+            data_manager.save_review({
+                "qa_id": qid, "chart_id": cid, "source": src,
+                "status": status, "comment": cmt
+            })
+            return "已保存"
+
         save_btn.click(
-            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=export_reviews_handler, outputs=[download_file]).then(fn=lambda: gr.File(visible=True), outputs=[download_file])
-    
-    return app
+            quick_save, 
+            inputs=[curr_qid, id_dd, src_dd, status_opt, comment],
+            outputs=[gr.Text(visible=False)]
+        )
+
+    return demo
 
 if __name__ == "__main__":
     app = create_ui()
-    app.launch(server_name="0.0.0.0", server_port=7860, show_api=False)
+    # 强制禁用所有 API 预览功能
+    app.launch(
+        server_name="0.0.0.0", 
+        server_port=7860, 
+        show_api=False,
+        max_threads=10
+    )