TanWenyan/leaudit-platform-frontend
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Merge branch 'PingChuan' into shiy-login

Browse Source 
# Conflicts:
#	app/config/api-config.ts
This commit is contained in:
yorn
2025-12-03 12:10:21 +08:00
parent 6e2cd453f8 a13f3b3635
commit eca98fc540
33 changed files with 6104 additions and 1477 deletions
Download Patch File Download Diff File Expand all files Collapse all files
app/api/dify-dataset/api/documentApi.ts
+151 -20
View File
@@ -121,7 +121,7 @@ export async function toggleDocumentStatus(
}
/**
* 上传文件到知识库
* 上传文件到知识库（使用默认配置）
*
* @param datasetId - 知识库 ID
* @param file - 文件对象
@@ -132,19 +132,44 @@ export async function uploadDocument(
datasetId: string,
file: File,
onProgress?: (percent: number) => void
): Promise<any> {
): Promise<UploadDocumentResponse> {
return uploadDocumentWithConfig(
datasetId,
file,
{
indexing_technique: 'high_quality',
process_rule: { mode: 'automatic' },
},
onProgress
);
}
/**
* 上传文件到知识库（使用自定义配置）
*
* @param datasetId - 知识库 ID
* @param file - 文件对象
* @param config - 上传配置（索引模式、分段规则等）
* @param onProgress - 上传进度回调
* @returns 创建的文档信息，包含 batch 用于查询索引进度
*/
export async function uploadDocumentWithConfig(
datasetId: string,
file: File,
config: UploadDocumentConfig,
onProgress?: (percent: number) => void
): Promise<UploadDocumentResponse> {
const formData = new FormData();
formData.append('file', file);
formData.append('data', JSON.stringify({
indexing_technique: 'high_quality',
process_rule: {
mode: 'automatic',
},
}));
formData.append('data', JSON.stringify(config));
console.log('[Dataset Client] 上传文档:', { datasetId, fileName: file.name });
console.log('[Dataset Client] 上传文档:', {
datasetId,
fileName: file.name,
config,
});
const response = await axios.post(
const response = await axios.post<UploadDocumentResponse>(
`${API_URL}/datasets/${datasetId}/documents`,
formData,
{
@@ -200,21 +225,41 @@ export async function fetchUploadFileInfo(
return response.data;
}
/**
* 预处理规则 ID
*/
export type PreProcessingRuleId = 'remove_extra_spaces' | 'remove_urls_emails';
/**
* 预处理规则配置
*/
export interface PreProcessingRule {
id: PreProcessingRuleId;
enabled: boolean;
}
/**
* 分段配置
*/
export interface SegmentationConfig {
separator: string;
max_tokens: number;
}
/**
* 自定义处理规则
*/
export interface CustomRules {
pre_processing_rules?: PreProcessingRule[];
segmentation?: SegmentationConfig;
}
/**
* 文档处理规则配置
*/
export interface ProcessRule {
mode: 'automatic' | 'custom';
rules?: {
pre_processing_rules?: Array<{
id: 'remove_extra_spaces' | 'remove_urls_emails';
enabled: boolean;
}>;
segmentation?: {
separator: string;
max_tokens: number;
};
};
rules?: CustomRules;
}
/**
@@ -225,6 +270,45 @@ export interface UpdateDocumentSettings {
process_rule?: ProcessRule;
}
/**
* 上传文档配置参数
*/
export interface UploadDocumentConfig {
indexing_technique: 'high_quality' | 'economy';
process_rule: ProcessRule;
}
/**
* 上传文档响应
*/
export interface UploadDocumentResponse {
document: {
id: string;
position: number;
data_source_type: string;
data_source_info: {
upload_file_id: string;
};
dataset_process_rule_id: string;
name: string;
created_from: string;
created_by: string;
created_at: number;
tokens: number;
indexing_status: string;
error: string | null;
enabled: boolean;
disabled_at: number | null;
disabled_by: string | null;
archived: boolean;
display_status: string;
word_count: number;
hit_count: number;
doc_form: string;
};
batch: string;
}
/**
* 更新文档设置并重新处理
* 注意：Dify API 不直接支持修改已有文档的分段设置
@@ -252,3 +336,50 @@ export async function updateDocumentWithSettings(
);
return response.data;
}
/**
* 通过文件更新文档
* Dify API: POST /datasets/{dataset_id}/documents/{document_id}/update-by-file
*
* 用于在用户修改分段参数后，使用同一文件重新处理文档
*
* @param datasetId - 知识库 ID
* @param documentId - 文档 ID
* @param file - 文件对象（使用原上传的文件）
* @param config - 新的分段配置
* @param onProgress - 上传进度回调
* @returns 更新后的文档信息，包含新的 batch 用于查询索引进度
*/
export async function updateDocumentByFile(
datasetId: string,
documentId: string,
file: File,
config: UploadDocumentConfig,
onProgress?: (percent: number) => void
): Promise<UploadDocumentResponse> {
const formData = new FormData();
formData.append('file', file);
formData.append('data', JSON.stringify(config));
console.log('[Dataset Client] 通过文件更新文档:', {
datasetId,
documentId,
fileName: file.name,
config,
});
const response = await axios.post<UploadDocumentResponse>(
`${API_URL}/datasets/${datasetId}/documents/${documentId}/update-by-file`,
formData,
{
withCredentials: true,
onUploadProgress: (progressEvent) => {
if (progressEvent.total && onProgress) {
const percent = Math.round((progressEvent.loaded * 100) / progressEvent.total);
onProgress(percent);
}
},
}
);
return response.data;
}
app/api/dify-dataset/api/index.ts
+9
View File
@@ -18,10 +18,19 @@ export {
deleteDocument,
toggleDocumentStatus,
uploadDocument,
uploadDocumentWithConfig,
updateDocumentByFile,
fetchIndexingStatus,
fetchUploadFileInfo,
} from './documentApi';
// 文档 API 类型
export type {
ProcessRule,
UploadDocumentConfig,
UploadDocumentResponse,
} from './documentApi';
// 分段、子分段、检索 API
export {
fetchSegments,
app/api/dify-dataset/api/segmentApi.ts
+2 -2
View File
@@ -332,7 +332,7 @@ export async function deleteChildChunk(
*
* @param datasetId - 知识库 ID
* @param query - 检索关键词
* @param retrievalModel - 检索模型配置
* @param retrievalModel - 检索模型配置（完整的 Dify API 格式）
* @returns 检索结果
*/
export async function retrieveDataset(
@@ -340,7 +340,7 @@ export async function retrieveDataset(
query: string,
retrievalModel?: RetrieveRequest['retrieval_model']
): Promise<RetrieveResponse> {
console.log('[Dataset Client] 检索知识库:', { datasetId, query });
console.log('[Dataset Client] 检索知识库:', { datasetId, query, retrievalModel });
const requestBody: RetrieveRequest = {
query,
app/api/dify-dataset/type/index.ts
+1
View File
@@ -42,6 +42,7 @@ export type {
MetadataFilterCondition,
MetadataFilteringConditions,
RetrieveRequest,
RetrieveSegment,
RetrieveRecord,
RetrieveResponse,
} from './segmentTypes';
app/api/dify-dataset/type/segmentTypes.ts
+19 -12
View File
@@ -4,6 +4,11 @@
* @module api/dify-dataset/type/segmentTypes
*/
import type { RetrievalModel } from './datasetTypes';
// 重新导出以便其他模块使用
export type { RetrievalModel };
// ============================================================================
// 分段类型
// ============================================================================
@@ -154,25 +159,27 @@ export interface MetadataFilteringConditions {
*/
export interface RetrieveRequest {
query: string;
retrieval_model?: {
search_method: 'keyword_search' | 'semantic_search' | 'full_text_search' | 'hybrid_search';
reranking_enable?: boolean;
reranking_model?: {
reranking_provider_name: string;
reranking_model_name: string;
};
top_k?: number;
score_threshold_enabled?: boolean;
score_threshold?: number;
};
retrieval_model?: RetrievalModel;
metadata_filtering_conditions?: MetadataFilteringConditions;
}
/**
* 检索结果中的分段信息（包含关联文档）
*/
export interface RetrieveSegment extends Segment {
document?: {
id: string;
data_source_type: string;
name: string;
doc_type: string | null;
};
}
/**
* 检索结果记录
*/
export interface RetrieveRecord {
segment: Segment;
segment: RetrieveSegment;
score: number;
tsne_position?: {
x: number;