新增合同模板上传功能,支持选择PDF和Word格式文件,并实现上传逻辑及状态管理。

This commit is contained in:
2025-09-11 17:25:58 +08:00
parent c611d6621d
commit a5ca3a8261
3 changed files with 618 additions and 11 deletions
+79
View File
@@ -136,6 +136,85 @@ export async function uploadFileToBinary(file: File): Promise<ArrayBuffer> {
* @param jwtToken JWT token * @param jwtToken JWT token
* @returns 上传结果 * @returns 上传结果
*/ */
/**
* 上传合同模板(用于与合同文档结构对比)
* @param file 模板文件
* @param documentId 源合同文档ID
* @param comparisonId 已有对比记录ID(可选)
* @param jwtToken JWT token
* @returns 上传结果
*/
export async function uploadContractTemplate(
file: File,
documentId: number,
comparisonId?: number,
jwtToken?: string
): Promise<{data: FileUploadResponse; error?: never} | {data?: never; error: string; status?: number}> {
try {
console.log('【合同模板上传】开始上传模板:', { fileName: file.name, documentId, comparisonId });
// 创建FormData对象
const formData = new FormData();
// 添加文件
formData.append('file', file);
// 添加上传信息
const uploadInfo = {
document_id: documentId,
...(comparisonId && { comparison_id: comparisonId })
};
formData.append('upload_info', JSON.stringify(uploadInfo));
// 构建请求URL
const uploadUrl = `${UPLOAD_URL}/upload_contract_template`;
console.log('【合同模板上传】准备发送请求到服务器:', uploadUrl);
// 设置请求头
const headers: HeadersInit = {
'Accept': 'application/json'
};
if (jwtToken) {
headers['Authorization'] = `Bearer ${jwtToken}`;
}
// 发送请求
const response = await fetch(uploadUrl, {
method: 'POST',
headers,
body: formData
});
console.log('【合同模板上传】服务器响应状态:', response.status);
if (!response.ok) {
const errorText = await response.text();
console.error('【合同模板上传】服务器返回错误:', errorText);
return {
error: `服务器错误: ${response.status} ${response.statusText}`,
status: response.status
};
}
const result = await response.json();
console.log('【合同模板上传】服务器返回结果:', result);
if (result.success) {
return { data: result.result };
} else {
return { error: result.error || '合同模板上传失败' };
}
} catch (error) {
console.error('【合同模板上传】上传过程中发生错误:', error);
return {
error: error instanceof Error ? error.message : '合同模板上传过程中发生未知错误'
};
}
}
/** /**
* 合同文档追加附件并合并 * 合同文档追加附件并合并
* @param documentId 合同文档ID * @param documentId 合同文档ID
+182 -11
View File
@@ -17,6 +17,7 @@ import {
uploadFileToBinary, uploadFileToBinary,
uploadDocumentToServer, uploadDocumentToServer,
appendContractAttachments, appendContractAttachments,
uploadContractTemplate,
type Document, type Document,
type DocumentType, type DocumentType,
type FileUploadResponse, type FileUploadResponse,
@@ -313,6 +314,11 @@ export default function FilesUpload() {
const [attachmentMergeMode, setAttachmentMergeMode] = useState<'overwrite' | 'new'>('overwrite'); const [attachmentMergeMode, setAttachmentMergeMode] = useState<'overwrite' | 'new'>('overwrite');
const [attachmentRemark, setAttachmentRemark] = useState<string>(""); const [attachmentRemark, setAttachmentRemark] = useState<string>("");
const [attachmentUploading, setAttachmentUploading] = useState<boolean>(false); const [attachmentUploading, setAttachmentUploading] = useState<boolean>(false);
// 合同模板上传状态
const [showTemplateUpload, setShowTemplateUpload] = useState<boolean>(false);
const [templateFile, setTemplateFile] = useState<File | null>(null);
const [templateUploading, setTemplateUploading] = useState<boolean>(false);
const [uploadProgress, setUploadProgress] = useState(0); const [uploadProgress, setUploadProgress] = useState(0);
const [uploadSpeed, setUploadSpeed] = useState("0KB/s"); const [uploadSpeed, setUploadSpeed] = useState("0KB/s");
@@ -864,6 +870,75 @@ export default function FilesUpload() {
setAttachmentUploading(false); setAttachmentUploading(false);
} }
}; };
// 处理合同模板文件选择
const handleTemplateFileSelected = (files: FileList) => {
try {
console.log('【合同模板上传】开始处理模板文件选择, 文件数量:', files.length);
if (files.length > 0) {
const file = files[0];
// 验证文件类型,支持PDF和Word
const fileName = file.name.toLowerCase();
const isValidType =
file.type === 'application/pdf' || fileName.endsWith('.pdf') ||
file.type === 'application/msword' || fileName.endsWith('.doc') ||
file.type === 'application/vnd.openxmlformats-officedocument.wordprocessingml.document' || fileName.endsWith('.docx');
if (isValidType) {
setTemplateFile(file);
console.log('【合同模板上传】有效文件:', file.name);
} else {
messageService.error('只支持PDF、Word格式的文件', {
title: '文件类型错误',
confirmText: '确定',
cancelText: '',
});
}
}
} catch (error) {
console.error('【合同模板上传】处理文件选择时发生错误:', error);
}
};
// 处理合同模板上传
const handleTemplateUpload = async () => {
if (!selectedDocumentId || !templateFile) {
toastService.error('请选择文档和模板文件');
return;
}
try {
setTemplateUploading(true);
const result = await uploadContractTemplate(
templateFile,
selectedDocumentId,
undefined, // comparisonId
loaderData.userInfo?.token
);
if (result.error) {
throw new Error(result.error);
}
toastService.success('合同模板上传成功!');
// 重置状态
setTemplateFile(null);
setShowTemplateUpload(false);
setSelectedDocumentId(null);
// 刷新文档列表
await filterDocuments(reviewType);
} catch (error) {
console.error('【合同模板上传】上传失败:', error);
toastService.error(error instanceof Error ? error.message : '合同模板上传失败');
} finally {
setTemplateUploading(false);
}
};
// 检查并准备上传 // 检查并准备上传
const checkAndPrepareUpload = (mainFiles: File[], attachmentFiles: File[]) => { const checkAndPrepareUpload = (mainFiles: File[], attachmentFiles: File[]) => {
@@ -1785,17 +1860,30 @@ export default function FilesUpload() {
</Button> </Button>
{record.type_id === 1 && record.status === DocumentStatus.PROCESSED && ( {record.type_id === 1 && record.status === DocumentStatus.PROCESSED && (
<Button <>
type="primary" <Button
size="small" type="primary"
icon="ri-attachment-line" size="small"
onClick={() => { icon="ri-attachment-line"
setSelectedDocumentId(record.id); onClick={() => {
setShowAttachmentUpload(true); setSelectedDocumentId(record.id);
}} setShowAttachmentUpload(true);
> }}
>
</Button>
</Button>
<Button
type="default"
size="small"
icon="ri-file-copy-line"
onClick={() => {
setSelectedDocumentId(record.id);
setShowTemplateUpload(true);
}}
>
</Button>
</>
)} )}
</div> </div>
) )
@@ -2310,6 +2398,89 @@ export default function FilesUpload() {
</div> </div>
</div> </div>
)} )}
{/* 合同模板上传模态框 */}
{showTemplateUpload && (
<div className="fixed inset-0 bg-black bg-opacity-50 flex items-center justify-center z-50">
<div className="bg-white rounded-lg p-6 w-full max-w-lg mx-4">
<div className="flex justify-between items-center mb-4">
<h3 className="text-lg font-semibold"></h3>
<button
onClick={() => {
setShowTemplateUpload(false);
setSelectedDocumentId(null);
setTemplateFile(null);
}}
className="text-gray-400 hover:text-gray-600"
>
<i className="ri-close-line text-xl"></i>
</button>
</div>
<div className="space-y-4">
{/* 文档信息 */}
<div className="bg-gray-50 p-3 rounded">
<p className="text-sm text-gray-600">
ID: <span className="font-medium">{selectedDocumentId}</span>
</p>
<p className="text-xs text-gray-500 mt-1">
PDFWord格式
</p>
</div>
{/* 文件上传区域 */}
<div>
<label className="block text-sm font-medium text-gray-700 mb-2">
<span className="text-red-500">*</span>
</label>
<UploadArea
onFilesSelected={handleTemplateFileSelected}
multiple={false}
accept=".pdf,.doc,.docx"
tipText="支持PDF、Word格式"
mainText="选择模板文件"
buttonText="选择文件"
icon="ri-file-copy-line"
/>
{templateFile && (
<div className="mt-2">
<p className="text-sm text-green-600 mb-2">
<i className="ri-checkbox-circle-line"></i>
</p>
<div className="text-xs text-gray-600 bg-gray-50 p-2 rounded">
<i className="ri-file-line mr-1"></i>
{templateFile.name} ({formatFileSize(templateFile.size)})
</div>
</div>
)}
</div>
{/* 操作按钮 */}
<div className="flex justify-end gap-3 pt-4 border-t">
<Button
type="default"
onClick={() => {
setShowTemplateUpload(false);
setSelectedDocumentId(null);
setTemplateFile(null);
}}
disabled={templateUploading}
>
</Button>
<Button
type="primary"
onClick={handleTemplateUpload}
disabled={!templateFile || templateUploading}
icon={templateUploading ? "ri-loader-4-line" : "ri-upload-cloud-line"}
>
{templateUploading ? '上传中...' : '开始上传'}
</Button>
</div>
</div>
</div>
</div>
)}
</div> </div>
); );
} }
+357
View File
@@ -0,0 +1,357 @@
## 合同文档上传与接口说明(v2
本文档详细说明 DocAuditAI 中与“合同文档上传”相关的 API 接口、参数、请求示例、响应示例以及后端处理流程。路径均以后端实际路由挂载为准:所有业务接口统一前缀为 `/admin`,本文档涉及的文档管理接口统一前缀为 `/admin/documents`
---
### 鉴权与基础信息
- **基础前缀**: `/admin/documents`
- **认证方式**:
- 推荐:`Authorization: Bearer <JWT>`(后端中间件会将用户信息注入 `req.state.current_user`
- 参考文档:`docs/接口认证方式统一说明.md`
- **存储后端**: MinIO(参考 `core/storage/minio_client.py``core/config.py` 中的 `MINIO_CONFIG`
- **异步任务**: Celery,队列名称默认为 `f"{REDIS_KEY_PREFIX}_tasks"`
---
### 文档类型枚举(后端基准)
来自 `core/utils/enums.py`
- `1` 合同文档(代码: HT,完整中文: 合同文档)
- `2` 行政许可决定书(代码: XZXK,完整中文: 行政许可决定书)
- `3` 行政处罚决定书(代码: XZCF,完整中文: 行政处罚决定书)
- `4` 合同对比模板(代码: HTMB,完整中文: 合同对比模板)
- `99` 其他文档(代码: QT,完整中文: 其他文档)
> 说明:客户端在 `upload_info` 中通常只需要传 `type_id`,后端会基于枚举推导短代码与中文全称。
---
## 接口清单
#### 1) 上传通用文档(支持合同/许可/处罚/其他)
- 方法与路径: `POST /admin/documents/upload`
- 表单参数:
- `file`: 文件(支持 PDF、Word.doc/.docx 会自动转换为 PDF
- `upload_info`: JSON 字符串,示例:
```json
{
"type_id": 1,
"document_number": "HT_20250111_120000_001",
"is_test_document": true,
"remark": "示例备注",
"evaluation_level": "普通"
}
```
- cURL 示例:
```bash
curl -X POST "http://127.0.0.1:8008/admin/documents/upload" \
-H "Authorization: Bearer <YOUR_TOKEN>" \
-F "file=@./samples/contract.pdf" \
-F 'upload_info={"type_id":1,"remark":"示例备注","evaluation_level":"普通"}'
```
- JavaScript fetch 示例:
```javascript
const form = new FormData();
form.append('file', fileInput.files[0]);
form.append('upload_info', JSON.stringify({ type_id: 1, remark: '示例备注' }));
const res = await fetch('/admin/documents/upload', {
method: 'POST',
headers: { Authorization: 'Bearer ' + token },
body: form
});
const data = await res.json();
```
- 响应示例:
```json
{
"success": true,
"result": {
"id": 123,
"file_name": "contract.pdf",
"file_size": 1024000,
"file_url": "https://minio.example.com/bucket/documents/...",
"type_id": 1,
"type_description": "HT",
"type_full_description": "合同文档",
"document_number": "HT_20250111_120000_001",
"storage_type": "minio",
"background_processing": true,
"api_version": "v2",
"is_test_document": true,
"remark": "示例备注",
"evaluation_level": "普通"
}
}
```
> 说明:合同(type_id=1)、许可(2)、处罚(3)会自动投递 Celery 任务进行 OCR/抽取/评查。
---
#### 2) 上传合同模板(用于与合同文档结构对比)
- 方法与路径: `POST /admin/documents/upload_contract_template`
- 表单参数:
- `file`: 模板 PDF/WordWord将自动转PDF
- `upload_info`: JSON 字符串,字段:
- `document_id`(必填):源合同文档ID
- `comparison_id`(可选):已有对比记录ID,未提供将自动创建
- cURL 示例:
```bash
curl -X POST "http://127.0.0.1:8008/admin/documents/upload_contract_template" \
-H "Authorization: Bearer <YOUR_TOKEN>" \
-F "file=@./samples/contract_template.pdf" \
-F 'upload_info={"document_id":123}'
```
- 响应示例:
```json
{
"success": true,
"result": {
"id": 456, // comparison_id(新建或传入)
"document_id": 789, // 模板对应的新建文档ID
"file_name": "contract_template.pdf",
"file_size": 204800,
"file_url": "https://minio.example.com/...",
"template_path": "documents/.../contract_template.pdf",
"template_contract_name": "contract_template.pdf",
"status": "Waiting",
"api_version": "v2"
}
}
```
---
#### 3) 合同文档追加附件并合并
- 方法与路径: `POST /admin/documents/contracts/{document_id}/append_attachments`
- 路径参数:
- `document_id`: 合同文档IDtype_id 必须为 1
- 表单参数:
- `files`: 多个附件(支持 PDF、Word(doc/docx)、ZIP、RARZIP/RAR 内仅合并其中的 PDF
- `merge_mode`: `overwrite`(默认,覆盖原文档路径)或 `new`(新建一条文档记录)
- `is_reprocess`: 是否触发 OCR/抽取/评查(默认 true)
- `remark`: 备注
- cURL 示例(支持压缩包作为附件来源):
```bash
curl -X POST "http://127.0.0.1:8008/admin/documents/contracts/123/append_attachments" \
-H "Authorization: Bearer <YOUR_TOKEN>" \
-F "files=@./samples/appendix1.pdf" \
-F "files=@./samples/attachments.zip" \
-F "merge_mode=overwrite" -F "is_reprocess=true" -F "remark=附加材料"
```
- 响应示例(new 模式):
```json
{
"success": true,
"result": {
"id": 901, // 新文档ID
"file_name": "contract_merged_20250111123000.pdf",
"file_size": 512000,
"file_url": "https://minio.example.com/...",
"type_id": 1,
"document_number": "HT_20250111_...",
"storage_type": "minio",
"is_merged": true,
"attachments_count": 2,
"merge_mode": "new",
"background_processing": true,
"api_version": "v2"
}
}
```
---
#### 4) 合同模板记录追加附件并合并
- 方法与路径: `POST /admin/documents/contract_templates/{comparison_id}/append_attachments`
- 路径参数:
- `comparison_id`: 合同结构对比记录ID
- 表单参数:
- `files`: 多个 PDF 附件
- `is_reprocess`: 是否触发模板 OCR 与对比(默认 true)
- `remark`: 备注
- cURL 示例:
```bash
curl -X POST "http://127.0.0.1:8008/admin/documents/contract_templates/456/append_attachments" \
-H "Authorization: Bearer <YOUR_TOKEN>" \
-F "files=@./samples/template_appendix.pdf" \
-F "is_reprocess=true" -F "remark=补充分条款"
```
- 响应示例:
```json
{
"success": true,
"result": {
"id": 456,
"document_id": 790, // 新模板文档ID
"file_name": "template_merged_20250111123500.pdf",
"file_size": 256000,
"file_url": "https://minio.example.com/...",
"template_path": "documents/.../template_merged_20250111123500.pdf",
"status": "Waiting",
"api_version": "v2",
"is_merged": true,
"attachments_count": 1
}
}
```
---
#### 5) 一体化:上传并分配交叉评查任务(支持单 PDF 或压缩包)
- 方法与路径: `POST /admin/documents/cross_review/documents/upload_and_assign`
- 表单参数:
- `file`: 单个 PDF、Word 或压缩包(zip/rar/7z/tar
- `upload_info`: JSON 字符串(至少包含 `type_id`
- `assign_user_ids`: JSON 数组字符串(如 `[1,2,3]`
- cURL 示例(批量:zip 内多个 PDF 将逐个入库并触发处理):
```bash
curl -X POST "/admin/documents/cross_review/documents/upload_and_assign" \
-H "Authorization: Bearer <YOUR_TOKEN>" \
-F "file=@./samples/contracts.zip" \
-F 'upload_info={"type_id":1,"task_name":"自动分配任务","doc_type":"合同文档"}' \
-F 'assign_user_ids=[1,2,3]'
```
- 响应示例(压缩包):
```json
{
"success": true,
"message": "已处理3个PDF。",
"pdf_files": ["a.pdf","b.pdf","c.pdf"],
"results": [ {"success": true, "document_id": 101, ... }, ... ],
"assigned_users": { "result": { /* 任务分配详情 */ } }
}
```
---
#### 6) 获取文档列表(按用户权限过滤)
- 方法与路径: `GET /admin/documents/list`
- 查询参数:
- `page`(默认 1)、`page_size`(默认 20,最大 100
- `search`(可选,模糊匹配 name 或 document_number
- `type_id`(可选)、`status`(可选)
- 认证后端会基于 `req.state.current_user.user_id` 过滤用户可见文档
- cURL 示例:
```bash
curl -G "/admin/documents/list" \
-H "Authorization: Bearer <YOUR_TOKEN>" \
--data-urlencode "page=1" \
--data-urlencode "page_size=20" \
--data-urlencode "search=合同"
```
- 响应示例:
```json
{
"total": 42,
"items": [
{
"id": 123,
"name": "contract.pdf",
"document_number": "HT_20250111_120000_001",
"type_id": 1,
"status": "Processed",
"upload_time": "2025-01-11T12:01:02",
"created_at": "2025-01-11T12:00:00",
"updated_at": "2025-01-11T12:10:00"
}
],
"page": 1,
"page_size": 20
}
```
---
## 处理流程说明(后端链路)
> 关键实现文件:`app/routes/v2/documents/documents.py`
1. 校验与预处理
- 校验文件大小(`MAX_FILE_SIZE`)与类型
- Word 自动转换为 PDF`core/utils/office_convert.ensure_pdf_from_upload`
- 生成安全文件名
2. 生成存储路径并上传 MinIO
- 路径规则:`documents/{INSTANCE_NAME}/{中文类型}/{YYYY}/{MM月DD日}/{文件名_时分秒}/{文件名}`
- 上传后得到 `storage_path` 与 `file_url`
3. 写入数据库(PostgREST
- 组装文档数据(含 `user_id`、`file_size`、`remark` 等)
- `document_service.async_insert_document` 入库,返回文档ID
4. 追加合并详细逻辑(仅合同附件追加支持压缩包)
- 入口:`POST /admin/documents/contracts/{document_id}/append_attachments`
- 校验:`document_id` 必须为合同文档(type_id=1),否则返回错误
- 下载原 PDF:从 MinIO 将原文档 `path` 下载到本地临时文件
- 处理附件输入:
- 支持的外层类型:`.pdf`、`.doc`、`.docx`、`.zip`、`.rar`
- 当为 PDF:直接保存到临时目录,加入合并列表
- 当为 Word.doc/.docx):先转换为 PDF(同单文件上传逻辑),再加入合并列表
- 当为 ZIP/RAR:先落地为临时压缩包,再调用 `document_upload_service.extract_archive` 解压,仅收集其中的 PDF 路径加入合并列表
- 对于其他类型:返回错误(仅支持 PDF/ZIP/RAR/Word
- 若最终未收集到任何 PDF:返回错误
- 合并顺序:始终使用【原PDF】+【附件PDF列表】顺序,调用 `merge_pdfs` 生成合并结果临时文件
- 目标路径:沿用原 `path` 的父目录,文件名追加 `_merged_{时间戳}.pdf`
- 入库与更新:
- `merge_mode=overwrite`(默认):更新原文档 `path/name/file_size/status=Waiting/remark`
- `merge_mode=new`:调用文档写入流程新建记录,并带上 `remark`
- 触发处理:当 `is_reprocess=true` 时,读入合并后的字节内容,调用 `_submit_ocr_task_v2` 投递后续 OCR/抽取/评查
- 返回:新/原文档标识、URL、大小、是否合并、附件数量、`merge_mode` 等
5. 投递异步任务(Celery
- 合同(1):`process_contract_ocr`
- 行政许可(2)、行政处罚(3):`process_document_ocr_v2`
- 合同模板(4):`process_contract_template_ocr`
- 队列:`{REDIS_KEY_PREFIX}_tasks`,默认过期 24htime_limit 600s
6. 后处理与状态更新
- OCR完成后按需进入抽取与评查
- 文档状态:`Cutting`/`Extractioning`/`Evaluationing`/`Processed`/`Failed`
---
## 错误处理与返回
- 统一响应模型:
- 成功:`{"success": true, "result": { ... }}`
- 失败:`{"success": false, "error": "错误消息"}`
- 常见错误:
- `400``upload_info` 非法、文件为空或类型不支持
- `413`:文件过大(超过 `MAX_FILE_SIZE`
- `500`:存储/数据库/任务投递异常
---
## 使用建议与注意事项
- 合同类建议始终以 PDF 上传,可减少转换耗时与兼容性问题
- 压缩包仅支持解出 PDF 后逐一入库与处理(zip/rar/7z/tar
- 建议合理设置 `remark` 与 `evaluation_level` 便于后续筛选
- 若需追踪异步任务状态,请结合 Flower 与后端日志(`logs/`
---
## 变更记录
- 2025-09-11:首版编制,依据 `app/routes/v2/documents/documents.py` 与相关服务模块撰写