From df83a9a07c17dc16ccca12292503284d6b4d2814 Mon Sep 17 00:00:00 2001 From: FFXN <31929997+FFXN@users.noreply.github.com> Date: Tue, 3 Feb 2026 22:02:28 +0800 Subject: [PATCH] feat: summary index for knowledge. (#670) * feat: summary index for knowledge. * feat: summary index for knowledge. * align terminology * add zh and ja translation --------- Co-authored-by: Riskey --- en/api-reference/openapi_knowledge.json | 56 +++++- ja/api-reference/openapi_knowledge.json | 234 +++++++++++++++++++----- zh/api-reference/openapi_knowledge.json | 234 +++++++++++++++++++----- 3 files changed, 441 insertions(+), 83 deletions(-) diff --git a/en/api-reference/openapi_knowledge.json b/en/api-reference/openapi_knowledge.json index e99d8e17..f1737651 100644 --- a/en/api-reference/openapi_knowledge.json +++ b/en/api-reference/openapi_knowledge.json @@ -330,7 +330,7 @@ "data": { "type": "string", "description": "A JSON string containing document metadata and processing rules. See `CreateDocumentByFileRequestData` schema for details.", - "example": "{\"indexing_technique\":\"high_quality\",\"process_rule\":{\"mode\":\"custom\", \"rules\": { \"segmentation\": {\"separator\":\"###\", \"max_tokens\":500}}}}" + "example": "{\"indexing_technique\":\"high_quality\",\"process_rule\":{\"mode\":\"custom\", \"rules\": { \"segmentation\": {\"separator\":\"###\", \"max_tokens\":500}}},\"summary_index_setting\":{\"enable\":true,\"model_name\":\"qwen-vl-plus\",\"model_provider_name\":\"langgenius/tongyi/tongyi\",\"summary_prompt\":\"summarize the text\"}}" }, "file": { "type": "string", @@ -1815,6 +1815,31 @@ } } }, + "SummaryIndexSetting": { + "type": "object", + "description": "Configuration for Summary Auto-Gen.", + "properties": { + "enable": { + "type": "boolean", + "description": "Whether to enable Summary Auto-Gen." + }, + "model_name": { + "type": "string", + "description": "Name of the model to use for generating summaries." + }, + "model_provider_name": { + "type": "string", + "description": "Provider of the summary model." + }, + "summary_prompt": { + "type": "string", + "description": "Prompt template for generating summaries." + } + }, + "required": [ + "enable" + ] + }, "PreprocessingRule": { "type": "object", "description": "A rule for preprocessing document content.", @@ -1957,6 +1982,10 @@ "embedding_model_provider": { "type": "string", "description": "Provider of the embedding model." + }, + "summary_index_setting": { + "$ref": "#/components/schemas/SummaryIndexSetting", + "description": "Configuration for Summary Auto-Gen." } } }, @@ -1999,6 +2028,10 @@ }, "embedding_model_provider": { "type": "string" + }, + "summary_index_setting": { + "$ref": "#/components/schemas/SummaryIndexSetting", + "description": "Configuration for Summary Auto-Gen." } } }, @@ -2175,6 +2208,10 @@ }, "retrieval_model": { "$ref": "#/components/schemas/RetrievalModel" + }, + "summary_index_setting": { + "$ref": "#/components/schemas/SummaryIndexSetting", + "description": "Configuration for Summary Auto-Gen." } } }, @@ -2242,6 +2279,11 @@ "embedding_available": { "type": "boolean", "nullable": true + }, + "summary_index_setting": { + "$ref": "#/components/schemas/SummaryIndexSetting", + "description": "Summary Auto-Gen configurations for this dataset.", + "nullable": true } } }, @@ -2328,6 +2370,10 @@ "retrieval_model": { "$ref": "#/components/schemas/RetrievalModel" }, + "summary_index_setting": { + "$ref": "#/components/schemas/SummaryIndexSetting", + "description": "Configuration for Summary Auto-Gen." + }, "partial_member_list": { "type": "array", "description": "List of members for 'partial_members' permission.", @@ -2552,6 +2598,10 @@ "type": "integer", "format": "int64", "nullable": true + }, + "summary": { + "type": "string", + "nullable": true } } }, @@ -2656,6 +2706,10 @@ "type": "string" } }, + "summary": { + "type": "string", + "nullable": true + }, "enabled": { "type": "boolean" }, diff --git a/ja/api-reference/openapi_knowledge.json b/ja/api-reference/openapi_knowledge.json index 121861c3..d55eecd8 100644 --- a/ja/api-reference/openapi_knowledge.json +++ b/ja/api-reference/openapi_knowledge.json @@ -330,7 +330,7 @@ "data": { "type": "string", "description": "ドキュメントメタデータと処理ルールを含むJSON文字列。詳細は`CreateDocumentByFileRequestData`スキーマを参照してください。", - "example": "{\"indexing_technique\":\"high_quality\",\"process_rule\":{\"mode\":\"custom\", \"rules\": { \"segmentation\":{\"separator\":\"###\", \"max_tokens\":500}}}}" + "example": "{\"indexing_technique\":\"high_quality\",\"process_rule\":{\"mode\":\"custom\", \"rules\": { \"segmentation\":{\"separator\":\"###\", \"max_tokens\":500}}},\"summary_index_setting\":{\"enable\":true,\"model_name\":\"qwen-vl-plus\",\"model_provider_name\":\"langgenius/tongyi/tongyi\",\"summary_prompt\":\"summarize the text\"}}" }, "file": { "type": "string", @@ -1410,7 +1410,9 @@ "application/json": { "schema": { "type": "object", - "required": ["name"], + "required": [ + "name" + ], "properties": { "name": { "type": "string", @@ -1471,7 +1473,10 @@ "application/json": { "schema": { "type": "object", - "required": ["tag_id", "name"], + "required": [ + "tag_id", + "name" + ], "properties": { "tag_id": { "type": "string", @@ -1514,7 +1519,9 @@ "application/json": { "schema": { "type": "object", - "required": ["tag_id"], + "required": [ + "tag_id" + ], "properties": { "tag_id": { "type": "string", @@ -1547,7 +1554,10 @@ "application/json": { "schema": { "type": "object", - "required": ["target_id", "tag_ids"], + "required": [ + "target_id", + "tag_ids" + ], "properties": { "target_id": { "type": "string", @@ -1588,7 +1598,10 @@ "application/json": { "schema": { "type": "object", - "required": ["target_id", "tag_id"], + "required": [ + "target_id", + "tag_id" + ], "properties": { "target_id": { "type": "string", @@ -1645,8 +1658,13 @@ "items": { "type": "object", "properties": { - "id": { "type": "string", "format": "uuid" }, - "name": { "type": "string" } + "id": { + "type": "string", + "format": "uuid" + }, + "name": { + "type": "string" + } } } }, @@ -1815,6 +1833,31 @@ } } }, + "SummaryIndexSetting": { + "type": "object", + "description": "要約自動生成の設定。", + "properties": { + "enable": { + "type": "boolean", + "description": "要約自動生成を有効にするかどうか。" + }, + "model_name": { + "type": "string", + "description": "要約生成に使用するモデル名。" + }, + "model_provider_name": { + "type": "string", + "description": "要約モデルのプロバイダー。" + }, + "summary_prompt": { + "type": "string", + "description": "要約生成に使用するプロンプトテンプレート。" + } + }, + "required": [ + "enable" + ] + }, "PreprocessingRule": { "type": "object", "description": "ドキュメントコンテンツの前処理ルール。", @@ -1957,6 +2000,10 @@ "embedding_model_provider": { "type": "string", "description": "埋め込みモデルのプロバイダー。" + }, + "summary_index_setting": { + "$ref": "#/components/schemas/SummaryIndexSetting", + "description": "要約自動生成の設定。" } } }, @@ -1999,6 +2046,10 @@ }, "embedding_model_provider": { "type": "string" + }, + "summary_index_setting": { + "$ref": "#/components/schemas/SummaryIndexSetting", + "description": "要約自動生成の設定。" } } }, @@ -2175,6 +2226,10 @@ }, "retrieval_model": { "$ref": "#/components/schemas/RetrievalModel" + }, + "summary_index_setting": { + "$ref": "#/components/schemas/SummaryIndexSetting", + "description": "要約自動生成の設定。" } } }, @@ -2242,6 +2297,11 @@ "embedding_available": { "type": "boolean", "nullable": true + }, + "summary_index_setting": { + "$ref": "#/components/schemas/SummaryIndexSetting", + "description": "このデータセットの要約自動生成設定。", + "nullable": true } } }, @@ -2343,6 +2403,10 @@ } } } + }, + "summary_index_setting": { + "$ref": "#/components/schemas/SummaryIndexSetting", + "description": "要約自動生成の設定。" } } }, @@ -2552,6 +2616,10 @@ "type": "integer", "format": "int64", "nullable": true + }, + "summary": { + "type": "string", + "nullable": true } } }, @@ -2870,25 +2938,64 @@ "type": "object", "description": "階層セグメント化における子チャンクを表します。", "properties": { - "id": { "type": "string", "format": "uuid" }, - "segment_id": { "type": "string", "format": "uuid" }, - "content": { "type": "string" }, - "word_count": { "type": "integer" }, - "tokens": { "type": "integer" }, - "index_node_id": { "type": "string" }, - "index_node_hash": { "type": "string" }, - "status": { "type": "string" }, - "created_by": { "type": "string", "format": "uuid" }, - "created_at": { "type": "integer", "format": "int64" }, - "indexing_at": { "type": "integer", "format": "int64" }, - "completed_at": { "type": "integer", "format": "int64" }, - "error": { "type": "string", "nullable": true }, - "stopped_at": { "type": "integer", "format": "int64", "nullable": true } + "id": { + "type": "string", + "format": "uuid" + }, + "segment_id": { + "type": "string", + "format": "uuid" + }, + "content": { + "type": "string" + }, + "word_count": { + "type": "integer" + }, + "tokens": { + "type": "integer" + }, + "index_node_id": { + "type": "string" + }, + "index_node_hash": { + "type": "string" + }, + "status": { + "type": "string" + }, + "created_by": { + "type": "string", + "format": "uuid" + }, + "created_at": { + "type": "integer", + "format": "int64" + }, + "indexing_at": { + "type": "integer", + "format": "int64" + }, + "completed_at": { + "type": "integer", + "format": "int64" + }, + "error": { + "type": "string", + "nullable": true + }, + "stopped_at": { + "type": "integer", + "format": "int64", + "nullable": true + } } }, "CreateChildChunkRequest": { "type": "object", - "required": ["content"], + "required": [ + "content" + ], "properties": { "content": { "type": "string", @@ -2898,7 +3005,9 @@ }, "UpdateChildChunkRequest": { "type": "object", - "required": ["content"], + "required": [ + "content" + ], "properties": { "content": { "type": "string", @@ -2923,33 +3032,76 @@ "$ref": "#/components/schemas/ChildChunk" } }, - "total": { "type": "integer" }, - "total_pages": { "type": "integer" }, - "page": { "type": "integer" }, - "limit": { "type": "integer" } + "total": { + "type": "integer" + }, + "total_pages": { + "type": "integer" + }, + "page": { + "type": "integer" + }, + "limit": { + "type": "integer" + } } }, "UploadFileResponse": { "type": "object", "properties": { - "id": { "type": "string", "format": "uuid" }, - "name": { "type": "string" }, - "size": { "type": "integer" }, - "extension": { "type": "string" }, - "url": { "type": "string", "format": "uri", "description": "ファイルのプレビューURL。" }, - "download_url": { "type": "string", "format": "uri", "description": "ファイルのダウンロードURL。" }, - "mime_type": { "type": "string" }, - "created_by": { "type": "string", "format": "uuid" }, - "created_at": { "type": "integer", "format": "int64" } + "id": { + "type": "string", + "format": "uuid" + }, + "name": { + "type": "string" + }, + "size": { + "type": "integer" + }, + "extension": { + "type": "string" + }, + "url": { + "type": "string", + "format": "uri", + "description": "ファイルのプレビューURL。" + }, + "download_url": { + "type": "string", + "format": "uri", + "description": "ファイルのダウンロードURL。" + }, + "mime_type": { + "type": "string" + }, + "created_by": { + "type": "string", + "format": "uuid" + }, + "created_at": { + "type": "integer", + "format": "int64" + } } }, "Tag": { "type": "object", "properties": { - "id": { "type": "string", "format": "uuid" }, - "name": { "type": "string" }, - "type": { "type": "string", "example": "knowledge" }, - "binding_count": { "type": "integer" } + "id": { + "type": "string", + "format": "uuid" + }, + "name": { + "type": "string" + }, + "type": { + "type": "string", + "example": "knowledge" + }, + "binding_count": { + "type": "integer" + } } } } diff --git a/zh/api-reference/openapi_knowledge.json b/zh/api-reference/openapi_knowledge.json index 0efeb4cf..0c1e92e6 100644 --- a/zh/api-reference/openapi_knowledge.json +++ b/zh/api-reference/openapi_knowledge.json @@ -330,7 +330,7 @@ "data": { "type": "string", "description": "包含文档元数据和处理规则的 JSON 字符串。详情请参见 `CreateDocumentByFileRequestData` 模式。", - "example": "{\"indexing_technique\":\"high_quality\",\"process_rule\":{\"mode\":\"custom\", \"rules\": { \"segmentation\":{\"separator\":\"###\", \"max_tokens\":500}}}}" + "example": "{\"indexing_technique\":\"high_quality\",\"process_rule\":{\"mode\":\"custom\", \"rules\": { \"segmentation\":{\"separator\":\"###\", \"max_tokens\":500}}},\"summary_index_setting\":{\"enable\":true,\"model_name\":\"qwen-vl-plus\",\"model_provider_name\":\"langgenius/tongyi/tongyi\",\"summary_prompt\":\"summarize the text\"}}" }, "file": { "type": "string", @@ -1410,7 +1410,9 @@ "application/json": { "schema": { "type": "object", - "required": ["name"], + "required": [ + "name" + ], "properties": { "name": { "type": "string", @@ -1471,7 +1473,10 @@ "application/json": { "schema": { "type": "object", - "required": ["tag_id", "name"], + "required": [ + "tag_id", + "name" + ], "properties": { "tag_id": { "type": "string", @@ -1514,7 +1519,9 @@ "application/json": { "schema": { "type": "object", - "required": ["tag_id"], + "required": [ + "tag_id" + ], "properties": { "tag_id": { "type": "string", @@ -1547,7 +1554,10 @@ "application/json": { "schema": { "type": "object", - "required": ["target_id", "tag_ids"], + "required": [ + "target_id", + "tag_ids" + ], "properties": { "target_id": { "type": "string", @@ -1588,7 +1598,10 @@ "application/json": { "schema": { "type": "object", - "required": ["target_id", "tag_id"], + "required": [ + "target_id", + "tag_id" + ], "properties": { "target_id": { "type": "string", @@ -1645,8 +1658,13 @@ "items": { "type": "object", "properties": { - "id": { "type": "string", "format": "uuid" }, - "name": { "type": "string" } + "id": { + "type": "string", + "format": "uuid" + }, + "name": { + "type": "string" + } } } }, @@ -1815,6 +1833,31 @@ } } }, + "SummaryIndexSetting": { + "type": "object", + "description": "摘要自动生成配置。", + "properties": { + "enable": { + "type": "boolean", + "description": "是否启用摘要自动生成。" + }, + "model_name": { + "type": "string", + "description": "用于生成摘要的模型名称。" + }, + "model_provider_name": { + "type": "string", + "description": "摘要模型的提供商。" + }, + "summary_prompt": { + "type": "string", + "description": "用于生成摘要的提示词模板。" + } + }, + "required": [ + "enable" + ] + }, "PreprocessingRule": { "type": "object", "description": "文档内容预处理规则。", @@ -1957,6 +2000,10 @@ "embedding_model_provider": { "type": "string", "description": "嵌入模型的提供商。" + }, + "summary_index_setting": { + "$ref": "#/components/schemas/SummaryIndexSetting", + "description": "摘要自动生成配置。" } } }, @@ -1999,6 +2046,10 @@ }, "embedding_model_provider": { "type": "string" + }, + "summary_index_setting": { + "$ref": "#/components/schemas/SummaryIndexSetting", + "description": "摘要自动生成配置。" } } }, @@ -2175,6 +2226,10 @@ }, "retrieval_model": { "$ref": "#/components/schemas/RetrievalModel" + }, + "summary_index_setting": { + "$ref": "#/components/schemas/SummaryIndexSetting", + "description": "摘要自动生成配置。" } } }, @@ -2242,6 +2297,11 @@ "embedding_available": { "type": "boolean", "nullable": true + }, + "summary_index_setting": { + "$ref": "#/components/schemas/SummaryIndexSetting", + "description": "该数据集的摘要自动生成配置。", + "nullable": true } } }, @@ -2343,6 +2403,10 @@ } } } + }, + "summary_index_setting": { + "$ref": "#/components/schemas/SummaryIndexSetting", + "description": "摘要自动生成配置。" } } }, @@ -2552,6 +2616,10 @@ "type": "integer", "format": "int64", "nullable": true + }, + "summary": { + "type": "string", + "nullable": true } } }, @@ -2870,25 +2938,64 @@ "type": "object", "description": "表示分层分割中的子块。", "properties": { - "id": { "type": "string", "format": "uuid" }, - "segment_id": { "type": "string", "format": "uuid" }, - "content": { "type": "string" }, - "word_count": { "type": "integer" }, - "tokens": { "type": "integer" }, - "index_node_id": { "type": "string" }, - "index_node_hash": { "type": "string" }, - "status": { "type": "string" }, - "created_by": { "type": "string", "format": "uuid" }, - "created_at": { "type": "integer", "format": "int64" }, - "indexing_at": { "type": "integer", "format": "int64" }, - "completed_at": { "type": "integer", "format": "int64" }, - "error": { "type": "string", "nullable": true }, - "stopped_at": { "type": "integer", "format": "int64", "nullable": true } + "id": { + "type": "string", + "format": "uuid" + }, + "segment_id": { + "type": "string", + "format": "uuid" + }, + "content": { + "type": "string" + }, + "word_count": { + "type": "integer" + }, + "tokens": { + "type": "integer" + }, + "index_node_id": { + "type": "string" + }, + "index_node_hash": { + "type": "string" + }, + "status": { + "type": "string" + }, + "created_by": { + "type": "string", + "format": "uuid" + }, + "created_at": { + "type": "integer", + "format": "int64" + }, + "indexing_at": { + "type": "integer", + "format": "int64" + }, + "completed_at": { + "type": "integer", + "format": "int64" + }, + "error": { + "type": "string", + "nullable": true + }, + "stopped_at": { + "type": "integer", + "format": "int64", + "nullable": true + } } }, "CreateChildChunkRequest": { "type": "object", - "required": ["content"], + "required": [ + "content" + ], "properties": { "content": { "type": "string", @@ -2898,7 +3005,9 @@ }, "UpdateChildChunkRequest": { "type": "object", - "required": ["content"], + "required": [ + "content" + ], "properties": { "content": { "type": "string", @@ -2923,33 +3032,76 @@ "$ref": "#/components/schemas/ChildChunk" } }, - "total": { "type": "integer" }, - "total_pages": { "type": "integer" }, - "page": { "type": "integer" }, - "limit": { "type": "integer" } + "total": { + "type": "integer" + }, + "total_pages": { + "type": "integer" + }, + "page": { + "type": "integer" + }, + "limit": { + "type": "integer" + } } }, "UploadFileResponse": { "type": "object", "properties": { - "id": { "type": "string", "format": "uuid" }, - "name": { "type": "string" }, - "size": { "type": "integer" }, - "extension": { "type": "string" }, - "url": { "type": "string", "format": "uri", "description": "文件的预览 URL。" }, - "download_url": { "type": "string", "format": "uri", "description": "文件的下载 URL。" }, - "mime_type": { "type": "string" }, - "created_by": { "type": "string", "format": "uuid" }, - "created_at": { "type": "integer", "format": "int64" } + "id": { + "type": "string", + "format": "uuid" + }, + "name": { + "type": "string" + }, + "size": { + "type": "integer" + }, + "extension": { + "type": "string" + }, + "url": { + "type": "string", + "format": "uri", + "description": "文件的预览 URL。" + }, + "download_url": { + "type": "string", + "format": "uri", + "description": "文件的下载 URL。" + }, + "mime_type": { + "type": "string" + }, + "created_by": { + "type": "string", + "format": "uuid" + }, + "created_at": { + "type": "integer", + "format": "int64" + } } }, "Tag": { "type": "object", "properties": { - "id": { "type": "string", "format": "uuid" }, - "name": { "type": "string" }, - "type": { "type": "string", "example": "knowledge" }, - "binding_count": { "type": "integer" } + "id": { + "type": "string", + "format": "uuid" + }, + "name": { + "type": "string" + }, + "type": { + "type": "string", + "example": "knowledge" + }, + "binding_count": { + "type": "integer" + } } } }