修BUG[0.2.8.16][ci]

elastic/es_connect.py

@@ -199,7 +199,8 @@ def get_registration_code(code: str):
 
 def list_registration_codes():
     try:
-        search = RegistrationCodeDocument.search()
+        # 增加 size=1000 以支持返回更多注册码
+        search = RegistrationCodeDocument.search()[:1000]
         body = {
             "sort": [{"created_at": {"order": "desc"}}],
             "query": {"exists": {"field": "code"}}
@@ -297,7 +298,8 @@ def search_data(query):
     """
     try:
         # 使用Django-elasticsearch-dsl进行搜索
-        search = AchievementDocument.search()
+        # 增加 size=10000 以支持返回更多结果（ES默认限制为10000，如需更多需分页）
+        search = AchievementDocument.search()[:10000]
         search = search.query("multi_match", query=query, fields=['*'])
         response = search.execute()
         
@@ -319,7 +321,8 @@ def search_data(query):
 def search_all():
     """获取所有文档"""
     try:
-        search = AchievementDocument.search()
+        # 增加 size=10000 以支持返回更多结果（ES默认限制为10000，如需更多需分页）
+        search = AchievementDocument.search()[:10000]
         search = search.query("match_all")
         response = search.execute()
         
@@ -421,7 +424,8 @@ def search_by_any_field(keyword):
         list: 包含搜索结果的列表
     """
     try:
-        search = AchievementDocument.search()
+        # 增加 size=10000 以支持返回更多结果（ES默认限制为10000，如需更多需分页）
+        search = AchievementDocument.search()[:10000]
         
         # 使用multi_match查询，在所有字段中搜索
         search = search.query("multi_match", 
@@ -988,7 +992,7 @@ def list_registration_code_manage_requests(status: str = None, limit: int = 200)
         if status:
             must.append({"term": {"status": str(status)}})
         body = {
-            "size": max(1, min(int(limit or 200), 500)),
+            "size": max(1, min(int(limit or 200), 2000)),
             "query": {"bool": {"must": must}},
             "sort": [{"created_at": {"order": "desc"}}],
         }

elastic/templates/elastic/manage.html

@@ -86,6 +86,7 @@
         <button class="btn btn-primary" onclick="performSearch('exact')">关键词搜索</button>
         <button class="btn btn-secondary" onclick="performSearch('fuzzy')">模糊搜索</button>
         <button class="btn" onclick="loadAllData()">显示全部</button>
+        <button class="btn btn-primary" onclick="exportAllData()">一键导出Excel</button>
         <button class="btn" onclick="clearSearch()">清空结果</button>
       </div>
 
@@ -507,6 +508,10 @@ function downloadReportCsv() {
   window.location.href = `/elastic/report/csv/?${params.toString()}`;
 }
 
+function exportAllData() {
+  window.location.href = "/elastic/export_achievements_csv/";
+}
+
 // 渲染表格
 function renderTable(data) {
   tableBody.innerHTML = '';

elastic/templates/elastic/upload.html

@@ -103,6 +103,7 @@
     <div class="upload-section" id="dropArea">
       <h3>上传文件</h3>
       <p>点击下方按钮选择图片或PDF文件，或拖拽文件到此区域</p>
+      <p style="margin: 8px 0 0; font-size: 13px; color: #64748b;">单次最多上传 {{ max_single_upload_count|default:"3" }} 个文件。</p>
       <form id="uploadForm" enctype="multipart/form-data">
         {% csrf_token %}
         <input type="file" id="fileInput" name="file" accept="image/*,.pdf" multiple />
@@ -155,6 +156,7 @@ const dropArea = document.getElementById('dropArea');
 const progressWrap = document.getElementById('progressWrap');
 const progressBar = document.getElementById('progressBar');
 const progressText = document.getElementById('progressText');
+const MAX_SINGLE_UPLOAD_COUNT = Number('{{ max_single_upload_count|default:"3" }}');
 
 let currentItems = []; // 存储当前待处理的所有文件结果
 let selectedFiles = [];
@@ -277,13 +279,21 @@ function updateFileHint() {
 function addFiles(files) {
   const incoming = Array.from(files || []).filter(f => f && (f.type.startsWith('image/') || f.name.toLowerCase().endsWith('.pdf')));
   const existingKeys = new Set(selectedFiles.map(f => `${f.name}|${f.size}|${f.lastModified}`));
+  const rejected = [];
   incoming.forEach(f => {
     const key = `${f.name}|${f.size}|${f.lastModified}`;
-    if (!existingKeys.has(key)) {
+    if (!existingKeys.has(key) && selectedFiles.length < MAX_SINGLE_UPLOAD_COUNT) {
       existingKeys.add(key);
       selectedFiles.push(f);
+    } else if (!existingKeys.has(key) && selectedFiles.length >= MAX_SINGLE_UPLOAD_COUNT) {
+      rejected.push(f.name);
     }
   });
+  if (rejected.length) {
+    uploadMsg.textContent = `单次最多上传 ${MAX_SINGLE_UPLOAD_COUNT} 个文件，以下文件未加入：${rejected.join('、')}`;
+    uploadMsg.className = 'status-message error';
+    uploadMsg.style.display = 'block';
+  }
   const urls = selectedFiles.map(f => {
     if (f.name.toLowerCase().endsWith('.pdf')) {
         return 'data:image/svg+xml;base64,PHN2ZyB4bWxucz0iaHR0cDovL3d3dy53My5vcmcvMjAwMC9zdmciIHdpZHRoPSI0OCIgaGVpZ2h0PSI0OCIgdmlld0JveD0iMCAwIDI0IDI0IiBmaWxsPSJub25lIiBzdHJva2U9IiNlZjQ0NDQiIHN0cm9rZS13aWR0aD0iMiIgc3Ryb2tlLWxpbmVjYXA9InJvdW5kIiBzdHJva2UtbGluZWpvaW49InJvdW5kIj48cGF0aCBkPSJNMTQgMmgyYTIgMiAwIDAgMSAyIDJ2MTZhMiAyIDAgMCAxLTIgMmgtMTJhMiAyIDAgMCAxLTItMlY0YTIgMiAwIDAgMSAyLTJoMiIvPjxwYXRoIGQ9Ik0xNCAydjRjMCAxLjEgLjkgMiAyIDJoNCIvPjxwYXRoIGQ9Ik03IDloNSIvPjxwYXRoIGQ9Ik03IDEzaDUiLz48cGF0aCBkPSJNNyAxN2g4Ii8+PC9zdmc+';
@@ -460,6 +470,12 @@ uploadForm.addEventListener('submit', async (e) => {
     uploadMsg.style.display = 'block';
     return;
   }
+  if (selectedFiles.length > MAX_SINGLE_UPLOAD_COUNT) {
+    uploadMsg.textContent = `单次最多上传 ${MAX_SINGLE_UPLOAD_COUNT} 个文件，请分批上传`;
+    uploadMsg.className = 'status-message error';
+    uploadMsg.style.display = 'block';
+    return;
+  }
 
   showProgress();
   setProgress(5, '预处理中');

elastic/urls.py

@@ -23,6 +23,7 @@ urlpatterns = [
     path('filter/', views.filter_view, name='filter'),
     path('report/', views.report_view, name='report'),
     path('report/csv/', views.report_csv_view, name='report_csv'),
+    path('export_achievements_csv/', views.export_achievements_csv, name='export_achievements_csv'),
     
     # 用户管理
     path('users/', views.get_users, name='get_users'),

elastic/views.py

@@ -8,6 +8,14 @@ import base64
 import json
 import csv
 import io
+import mimetypes
+try:
+    import openpyxl
+    from openpyxl.utils import get_column_letter
+    from openpyxl.drawing.image import Image as XLImage
+    HAS_OPENPYXL = True
+except ImportError:
+    HAS_OPENPYXL = False
 from datetime import datetime, timezone, timedelta
 import tempfile
 import concurrent.futures
@@ -40,6 +48,8 @@ except ImportError as e:
         HAS_PDF_SUPPORT = False
         PDF_ERROR = str(e)
 
+MAX_SINGLE_UPLOAD_COUNT = int(getattr(settings, "MAX_SINGLE_UPLOAD_COUNT", 3))
+
 
 def _filter_results_for_user(request, results):
     session_user_id = request.session.get("user_id")
@@ -614,6 +624,7 @@ def ocr_and_extract_info(image_path: str):
             return base64.b64encode(f.read()).decode("utf-8")
 
     base64_image = encode_image(image_path)
+    mime_type = mimetypes.guess_type(image_path)[0] or "image/jpeg"
 
     # api_key = getattr(settings, "AISTUDIO_API_KEY", "188f57db3766e02ed2c7e18373996d84f4112272")
     # base_url = getattr(settings, "OPENAI_BASE_URL", "https://aistudio.baidu.com/llm/lmapi/v3")
@@ -665,7 +676,7 @@ def ocr_and_extract_info(image_path: str):
                 "role": "user",
                 "content": [
                     {"type": "text", "text": f"请识别这张图片中的信息，将你认为重要的数据转换为不包含嵌套的json，不要显示其它信息以便于解析，直接输出json结果即可。使用“数据类型”字段表示这个东西的大致类型，除此之外你可以自行决定使用哪些json字段。“数据类型”的内容有严格规定，请查看{json.dumps(types, ensure_ascii=False)}中是否包含你所需要的类型，确定不包含后你才可以填入你觉得合适的大致分类。"},
-                    {"type": "image_url", "image_url": {"url": f"data:image/png;base64,{base64_image}"}},
+                    {"type": "image_url", "image_url": {"url": f"data:{mime_type};base64,{base64_image}"}},
                 ],
             },
         ],
@@ -715,6 +726,7 @@ def upload_page(request):
     context = {
         "user_id": user_id_qs or session_user_id,
         "username": me.get("username"),
+        "max_single_upload_count": MAX_SINGLE_UPLOAD_COUNT,
     }
     return render(request, "elastic/upload.html", context)
 
@@ -738,6 +750,14 @@ def upload(request):
                 files = [one]
         if not files:
             return JsonResponse({"status": "error", "message": "未选择文件"}, status=400)
+        if len(files) > MAX_SINGLE_UPLOAD_COUNT:
+            return JsonResponse(
+                {
+                    "status": "error",
+                    "message": f"单次最多上传 {MAX_SINGLE_UPLOAD_COUNT} 个文件，请分批上传",
+                },
+                status=400,
+            )
 
         images_dir = os.path.join(settings.MEDIA_ROOT, "images")
         os.makedirs(images_dir, exist_ok=True)
@@ -784,17 +804,20 @@ def upload(request):
                 abs_p, fname = img_info
                 try:
                     data = ocr_and_extract_info(abs_p)
-                    return data
-                except Exception:
-                    return None
+                    return data, None
+                except Exception as e:
+                    return None, f"{fname}: {str(e)}"
 
             group_data_list = []
+            group_errors = []
             with concurrent.futures.ThreadPoolExecutor(max_workers=min(len(group_images), 8)) as executor:
                 futures = [executor.submit(run_ocr, img_info) for img_info in group_images]
                 for future in concurrent.futures.as_completed(futures):
-                    res = future.result()
+                    res, err = future.result()
                     if res:
                         group_data_list.append(res)
+                    elif err:
+                        group_errors.append(err)
 
             merged_group_data = {}
             for item in group_data_list:
@@ -814,10 +837,34 @@ def upload(request):
                         merged_group_data[f"{base}_{idx}"] = v
 
             if not merged_group_data:
-                merged_group_data = {"文件名": f.name, "提示": "未识别到具体内容"}
+                merged_group_data = {
+                    "文件名": f.name,
+                    "提示": "未识别到具体内容" if not group_errors else "识别失败",
+                }
+                if group_errors:
+                    merged_group_data["错误信息"] = "；".join(group_errors[:3])
 
             rel_paths = [f"images/{img[1]}" for img in group_images]
-            image_urls = [request.build_absolute_uri(settings.MEDIA_URL + rp) for rp in rel_paths]
+            
+            # 改进：如果配置了 MinIO，则在上传阶段就同步到 MinIO，确保在线版本待处理列表能显示图片
+            image_urls = []
+            from minio_storage.minio_connect import is_minio_configured, upload_file, presigned_get_url
+            
+            minio_enabled = is_minio_configured()
+            for rp in rel_paths:
+                abs_p = os.path.join(settings.MEDIA_ROOT, rp)
+                if minio_enabled:
+                    try:
+                        # 上传到 MinIO
+                        upload_file(abs_p, rp)
+                        # 生成预签名 URL
+                        url = presigned_get_url(rp)
+                        image_urls.append(url)
+                    except Exception as e:
+                        print(f"上传临时图片到 MinIO 失败: {e}")
+                        image_urls.append(request.build_absolute_uri(settings.MEDIA_URL + rp))
+                else:
+                    image_urls.append(request.build_absolute_uri(settings.MEDIA_URL + rp))
 
             file_results.append({
                 "name": f.name,
@@ -1729,6 +1776,208 @@ def report_csv_view(request):
     out["Content-Disposition"] = 'attachment; filename="report.csv"'
     return out
 
+
+@require_http_methods(["GET"])
+def export_achievements_csv(request):
+    """一键导出所有可见成果为 Excel (如果支持) 或 CSV"""
+    try:
+        session_user_id = request.session.get("user_id")
+        if session_user_id is None:
+            return HttpResponse("Unauthorized", status=401)
+            
+        # 1. 获取所有数据
+        results = search_all()
+        # 2. 根据权限过滤
+        results = _filter_results_for_user(request, results)
+        # 3. 补充录入人姓名
+        results = _attach_writer_names(results)
+        
+        if not results:
+            return HttpResponse("No data to export", status=404)
+            
+        # 4. 解析数据并准备数据列表
+        parsed_data_list = []
+        all_data_keys = set()
+        
+        for item in results:
+            raw_data = item.get("data", "{}")
+            try:
+                if isinstance(raw_data, str):
+                    parsed_dict = json.loads(raw_data)
+                else:
+                    parsed_dict = raw_data
+            except Exception:
+                parsed_dict = {"原始数据": str(raw_data)}
+            
+            if not isinstance(parsed_dict, dict):
+                parsed_dict = {"数据内容": str(parsed_dict)}
+            
+            # 展平基础字段和动态数据字段
+            flat_item = {
+                "ID": item.get("_id", ""),
+                "录入人": item.get("writer_name") or item.get("writer_id", ""),
+                "时间": format_datetime_for_export(item.get("time")),
+            }
+            # 清理动态字段中的换行符
+            clean_parsed_dict = {}
+            for k, v in parsed_dict.items():
+                if isinstance(v, str):
+                    clean_parsed_dict[k] = v.replace('\r', '').replace('\n', ' ')
+                else:
+                    clean_parsed_dict[k] = v
+                    
+            flat_item.update(clean_parsed_dict)
+            # 保存原始图片引用以便导出 Excel 时使用
+            flat_item["_image_refs"] = _parse_image_refs(item.get("image", ""))
+            parsed_data_list.append(flat_item)
+            all_data_keys.update(clean_parsed_dict.keys())
+            
+        # 确定表头：基础字段 + 动态字段（按字母排序）
+        dynamic_headers = sorted(list(all_data_keys))
+        headers = ["ID", "录入人", "时间"] + dynamic_headers
+        
+        # 如果是 Excel 且支持图片，添加图片列
+        if HAS_OPENPYXL:
+            headers.append("成果图片")
+        
+        filename_base = f"achievements_export_{datetime.now().strftime('%Y%m%d_%H%M%S')}"
+
+        # 5. 如果安装了 openpyxl，生成 Excel
+        if HAS_OPENPYXL:
+            wb = openpyxl.Workbook()
+            ws = wb.active
+            ws.title = "成果数据"
+            
+            # 写入表头
+            for col_num, header in enumerate(headers, 1):
+                cell = ws.cell(row=1, column=col_num, value=header)
+                cell.font = openpyxl.styles.Font(bold=True)
+                cell.alignment = openpyxl.styles.Alignment(horizontal='center', vertical='center')
+            
+            # 写入数据
+            img_col_index = headers.index("成果图片") + 1 if "成果图片" in headers else None
+            
+            for row_num, row_data in enumerate(parsed_data_list, 2):
+                for col_num, header in enumerate(headers, 1):
+                    if header == "成果图片":
+                        continue # 图片单独处理
+                    ws.cell(row=row_num, column=col_num, value=row_data.get(header, ""))
+                
+                # 处理图片插入
+                if img_col_index and row_data.get("_image_refs"):
+                    first_ref = row_data["_image_refs"][0]
+                    img_bytes = _get_image_bytes(first_ref)
+                    if img_bytes:
+                        try:
+                            img = XLImage(img_bytes)
+                            # 调整图片大小以适应单元格 (假设高度 80 像素左右)
+                            aspect_ratio = img.width / img.height
+                            img.height = 80
+                            img.width = 80 * aspect_ratio
+                            
+                            # 计算插入位置
+                            cell_address = f"{get_column_letter(img_col_index)}{row_num}"
+                            ws.add_image(img, cell_address)
+                            # 设置行高以容纳图片 (80 像素约为 60 磅)
+                            ws.row_dimensions[row_num].height = 65
+                        except Exception as e:
+                            ws.cell(row=row_num, column=img_col_index, value=f"图片加载失败: {str(e)}")
+            
+            # 自动调整列宽
+            for i, column_cells in enumerate(ws.columns, 1):
+                header = headers[i-1]
+                if header == "成果图片":
+                    ws.column_dimensions[get_column_letter(i)].width = 20
+                    continue
+                length = max(len(str(cell.value or "")) for cell in column_cells)
+                ws.column_dimensions[get_column_letter(i)].width = min(length + 2, 50)
+
+            response = HttpResponse(
+                content_type='application/vnd.openxmlformats-officedocument.spreadsheetml.sheet',
+            )
+            response['Content-Disposition'] = f'attachment; filename="{filename_base}.xlsx"'
+            wb.save(response)
+            return response
+            
+        # 6. 否则回退到 CSV
+        output = io.StringIO()
+        output.write('\ufeff') # UTF-8 BOM
+        # 增加 extrasaction='ignore' 以忽略 _image_refs 等内部辅助字段
+        writer = csv.DictWriter(output, fieldnames=headers, extrasaction='ignore')
+        writer.writeheader()
+        for row in parsed_data_list:
+            writer.writerow(row)
+            
+        response = HttpResponse(output.getvalue(), content_type='text/csv; charset=utf-8')
+        response['Content-Disposition'] = f'attachment; filename="{filename_base}.csv"'
+        return response
+    except Exception as e:
+        import traceback
+        traceback.print_exc()
+        return HttpResponse(f"导出失败: {str(e)}", status=500)
+
+def format_datetime_for_export(t):
+    if not t: return ""
+    try:
+        if isinstance(t, datetime):
+            return t.strftime("%Y-%m-%d %H:%M:%S")
+        d = datetime.fromisoformat(str(t).replace('Z', '+00:00'))
+        return d.strftime("%Y-%m-%d %H:%M:%S")
+    except Exception:
+        return str(t)
+
+def _get_image_bytes(image_ref):
+    """根据 image_ref 获取图片字节流，并确保转换为 Excel 支持的格式 (如 JPEG/PNG)"""
+    s = str(image_ref or '').strip()
+    if not s:
+        return None
+
+    img_raw_bytes = None
+    if s.startswith('minio:'):
+        object_name = s[len('minio:'):].lstrip('/')
+        try:
+            from minio_storage.minio_connect import get_minio_client, get_bucket_name
+            client = get_minio_client()
+            bucket = get_bucket_name()
+            if client:
+                response = client.get_object(bucket, object_name)
+                img_raw_bytes = response.read()
+        except Exception:
+            pass
+
+    elif s.startswith('local:'):
+        rel_path = s[len('local:'):].lstrip('/')
+        abs_path = os.path.join(settings.MEDIA_ROOT, rel_path)
+        if os.path.isfile(abs_path):
+            try:
+                with open(abs_path, 'rb') as f:
+                    img_raw_bytes = f.read()
+            except Exception:
+                pass
+
+    if not img_raw_bytes:
+        return None
+
+    # 处理 WebP 或其他 openpyxl 可能不支持的格式
+    try:
+        from PIL import Image as PILImage
+        img_io = io.BytesIO(img_raw_bytes)
+        with PILImage.open(img_io) as pil_img:
+            # 如果是 WebP 或带有透明通道的图片，转换为 RGB 格式并存为 JPEG 或 PNG
+            # Excel 对 PNG 支持较好
+            output_io = io.BytesIO()
+            if pil_img.format == 'WEBP' or pil_img.mode in ('RGBA', 'LA', 'P'):
+                rgb_img = pil_img.convert('RGB')
+                rgb_img.save(output_io, format='JPEG', quality=85)
+            else:
+                pil_img.save(output_io, format=pil_img.format or 'JPEG')
+            output_io.seek(0)
+            return output_io
+    except Exception as e:
+        print(f"图片转换失败: {str(e)}")
+        return io.BytesIO(img_raw_bytes) # 尝试直接返回原始数据作为最后手段
+
+
 @require_http_methods(["POST"])
 @csrf_protect
 def revoke_registration_code_view(request):