jandres/dify
1
0
Fork 0
mirror of https://github.com/langgenius/dify.git synced 2026-01-06 06:26:00 +00:00
Code Issues Actions 8 Packages Projects Releases Wiki Activity

feat: knowledge pipeline (#25360)

Browse Source 
Signed-off-by: -LAN- <laipz8200@outlook.com>
Co-authored-by: twwu <twwu@dify.ai>
Co-authored-by: crazywoola <100913391+crazywoola@users.noreply.github.com>
Co-authored-by: jyong <718720800@qq.com>
Co-authored-by: Wu Tianwei <30284043+WTW0313@users.noreply.github.com>
Co-authored-by: QuantumGhost <obelisk.reg+git@gmail.com>
Co-authored-by: lyzno1 <yuanyouhuilyz@gmail.com>
Co-authored-by: quicksand <quicksandzn@gmail.com>
Co-authored-by: Jyong <76649700+JohnJyong@users.noreply.github.com>
Co-authored-by: lyzno1 <92089059+lyzno1@users.noreply.github.com>
Co-authored-by: zxhlyh <jasonapring2015@outlook.com>
Co-authored-by: Yongtao Huang <yongtaoh2022@gmail.com>
Co-authored-by: autofix-ci[bot] <114827586+autofix-ci[bot]@users.noreply.github.com>
Co-authored-by: Joel <iamjoel007@gmail.com>
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
Co-authored-by: nite-knite <nkCoding@gmail.com>
Co-authored-by: Hanqing Zhao <sherry9277@gmail.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
Co-authored-by: Harry <xh001x@hotmail.com>
This commit is contained in:
-LAN-
2025-09-18 12:49:10 +08:00
committed by GitHub
parent 7dadb33003
commit 85cda47c70
1772 changed files with 102407 additions and 31710 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
api/controllers/service_api/dataset/dataset.py
View File

@@ -13,13 +13,13 @@ from controllers.service_api.wraps import (
validate_dataset_token,
)
from core.model_runtime.entities.model_entities import ModelType
from core.plugin.entities.plugin import ModelProviderID
from core.provider_manager import ProviderManager
from fields.dataset_fields import dataset_detail_fields
from fields.tag_fields import build_dataset_tag_fields
from libs.login import current_user
from models.account import Account
from models.dataset import Dataset, DatasetPermissionEnum
from models.provider_ids import ModelProviderID
from services.dataset_service import DatasetPermissionService, DatasetService, DocumentService
from services.entities.knowledge_entities.knowledge_entities import RetrievalModel
from services.tag_service import TagService

35
api/controllers/service_api/dataset/document.py
View File

@@ -124,7 +124,12 @@ class DocumentAddByTextApi(DatasetApiResource):
args.get("retrieval_model").get("reranking_model").get("reranking_model_name"),
)
upload_file = FileService.upload_text(text=str(text), text_name=str(name))
if not current_user:
raise ValueError("current_user is required")
upload_file = FileService(db.engine).upload_text(
text=str(text), text_name=str(name), user_id=current_user.id, tenant_id=tenant_id
)
data_source = {
"type": "upload_file",
"info_list": {"data_source_type": "upload_file", "file_info_list": {"file_ids": [upload_file.id]}},
@@ -134,6 +139,9 @@ class DocumentAddByTextApi(DatasetApiResource):
# validate args
DocumentService.document_create_args_validate(knowledge_config)
if not current_user:
raise ValueError("current_user is required")
try:
documents, batch = DocumentService.save_document_with_dataset_id(
dataset=dataset,
@@ -199,7 +207,11 @@ class DocumentUpdateByTextApi(DatasetApiResource):
name = args.get("name")
if text is None or name is None:
raise ValueError("Both text and name must be strings.")
upload_file = FileService.upload_text(text=str(text), text_name=str(name))
if not current_user:
raise ValueError("current_user is required")
upload_file = FileService(db.engine).upload_text(
text=str(text), text_name=str(name), user_id=current_user.id, tenant_id=tenant_id
)
data_source = {
"type": "upload_file",
"info_list": {"data_source_type": "upload_file", "file_info_list": {"file_ids": [upload_file.id]}},
@@ -301,8 +313,9 @@ class DocumentAddByFileApi(DatasetApiResource):
if not isinstance(current_user, EndUser):
raise ValueError("Invalid user account")
upload_file = FileService.upload_file(
if not current_user:
raise ValueError("current_user is required")
upload_file = FileService(db.engine).upload_file(
filename=file.filename,
content=file.read(),
mimetype=file.mimetype,
@@ -390,10 +403,14 @@ class DocumentUpdateByFileApi(DatasetApiResource):
if not file.filename:
raise FilenameNotExistsError
if not current_user:
raise ValueError("current_user is required")
if not isinstance(current_user, EndUser):
raise ValueError("Invalid user account")
try:
if not isinstance(current_user, EndUser):
raise ValueError("Invalid user account")
upload_file = FileService.upload_file(
upload_file = FileService(db.engine).upload_file(
filename=file.filename,
content=file.read(),
mimetype=file.mimetype,
@@ -571,7 +588,7 @@ class DocumentApi(DatasetApiResource):
response = {"id": document.id, "doc_type": document.doc_type, "doc_metadata": document.doc_metadata_details}
elif metadata == "without":
dataset_process_rules = DatasetService.get_process_rules(dataset_id)
document_process_rules = document.dataset_process_rule.to_dict()
document_process_rules = document.dataset_process_rule.to_dict() if document.dataset_process_rule else {}
data_source_info = document.data_source_detail_dict
response = {
"id": document.id,
@@ -604,7 +621,7 @@ class DocumentApi(DatasetApiResource):
}
else:
dataset_process_rules = DatasetService.get_process_rules(dataset_id)
document_process_rules = document.dataset_process_rule.to_dict()
document_process_rules = document.dataset_process_rule.to_dict() if document.dataset_process_rule else {}
data_source_info = document.data_source_detail_dict
response = {
"id": document.id,

6
api/controllers/service_api/dataset/error.py
View File

@@ -47,3 +47,9 @@ class DatasetInUseError(BaseHTTPException):
error_code = "dataset_in_use"
description = "The dataset is being used by some apps. Please remove the dataset from the apps before deleting it."
code = 409
class PipelineRunError(BaseHTTPException):
error_code = "pipeline_run_error"
description = "An error occurred while running the pipeline."
code = 500

4
api/controllers/service_api/dataset/metadata.py
View File

@@ -133,7 +133,7 @@ class DatasetMetadataServiceApi(DatasetApiResource):
return 204
@service_api_ns.route("/datasets/metadata/built-in")
@service_api_ns.route("/datasets/<uuid:dataset_id>/metadata/built-in")
class DatasetMetadataBuiltInFieldServiceApi(DatasetApiResource):
@service_api_ns.doc("get_built_in_fields")
@service_api_ns.doc(description="Get all built-in metadata fields")
@@ -143,7 +143,7 @@ class DatasetMetadataBuiltInFieldServiceApi(DatasetApiResource):
401: "Unauthorized - invalid API token",
}
)
def get(self, tenant_id):
def get(self, tenant_id, dataset_id):
"""Get all built-in metadata fields."""
built_in_fields = MetadataService.get_built_in_fields()
return {"fields": built_in_fields}, 200

0
api/controllers/service_api/dataset/rag_pipeline/__init__.py Normal file
View File

242
api/controllers/service_api/dataset/rag_pipeline/rag_pipeline_workflow.py Normal file
View File

@@ -0,0 +1,242 @@
import string
import uuid
from collections.abc import Generator
from typing import Any
from flask import request
from flask_restx import reqparse
from flask_restx.reqparse import ParseResult, RequestParser
from werkzeug.exceptions import Forbidden
import services
from controllers.common.errors import FilenameNotExistsError, NoFileUploadedError, TooManyFilesError
from controllers.service_api import service_api_ns
from controllers.service_api.dataset.error import PipelineRunError
from controllers.service_api.wraps import DatasetApiResource
from core.app.apps.pipeline.pipeline_generator import PipelineGenerator
from core.app.entities.app_invoke_entities import InvokeFrom
from libs import helper
from libs.login import current_user
from models.account import Account
from models.dataset import Pipeline
from models.engine import db
from services.errors.file import FileTooLargeError, UnsupportedFileTypeError
from services.file_service import FileService
from services.rag_pipeline.entity.pipeline_service_api_entities import DatasourceNodeRunApiEntity
from services.rag_pipeline.pipeline_generate_service import PipelineGenerateService
from services.rag_pipeline.rag_pipeline import RagPipelineService
@service_api_ns.route(f"/datasets/{uuid:dataset_id}/pipeline/datasource-plugins")
class DatasourcePluginsApi(DatasetApiResource):
"""Resource for datasource plugins."""
@service_api_ns.doc(shortcut="list_rag_pipeline_datasource_plugins")
@service_api_ns.doc(description="List all datasource plugins for a rag pipeline")
@service_api_ns.doc(
path={
"dataset_id": "Dataset ID",
}
)
@service_api_ns.doc(
params={
"is_published": "Whether to get published or draft datasource plugins "
"(true for published, false for draft, default: true)"
}
)
@service_api_ns.doc(
responses={
200: "Datasource plugins retrieved successfully",
401: "Unauthorized - invalid API token",
}
)
def get(self, tenant_id: str, dataset_id: str):
"""Resource for getting datasource plugins."""
# Get query parameter to determine published or draft
is_published: bool = request.args.get("is_published", default=True, type=bool)
rag_pipeline_service: RagPipelineService = RagPipelineService()
datasource_plugins: list[dict[Any, Any]] = rag_pipeline_service.get_datasource_plugins(
tenant_id=tenant_id, dataset_id=dataset_id, is_published=is_published
)
return datasource_plugins, 200
@service_api_ns.route(f"/datasets/{uuid:dataset_id}/pipeline/datasource/nodes/{string:node_id}/run")
class DatasourceNodeRunApi(DatasetApiResource):
"""Resource for datasource node run."""
@service_api_ns.doc(shortcut="pipeline_datasource_node_run")
@service_api_ns.doc(description="Run a datasource node for a rag pipeline")
@service_api_ns.doc(
path={
"dataset_id": "Dataset ID",
}
)
@service_api_ns.doc(
body={
"inputs": "User input variables",
"datasource_type": "Datasource type, e.g. online_document",
"credential_id": "Credential ID",
"is_published": "Whether to get published or draft datasource plugins "
"(true for published, false for draft, default: true)",
}
)
@service_api_ns.doc(
responses={
200: "Datasource node run successfully",
401: "Unauthorized - invalid API token",
}
)
def post(self, tenant_id: str, dataset_id: str, node_id: str):
"""Resource for getting datasource plugins."""
# Get query parameter to determine published or draft
parser: RequestParser = reqparse.RequestParser()
parser.add_argument("inputs", type=dict, required=True, nullable=False, location="json")
parser.add_argument("datasource_type", type=str, required=True, location="json")
parser.add_argument("credential_id", type=str, required=False, location="json")
parser.add_argument("is_published", type=bool, required=True, location="json")
args: ParseResult = parser.parse_args()
datasource_node_run_api_entity: DatasourceNodeRunApiEntity = DatasourceNodeRunApiEntity(**args)
assert isinstance(current_user, Account)
rag_pipeline_service: RagPipelineService = RagPipelineService()
pipeline: Pipeline = rag_pipeline_service.get_pipeline(tenant_id=tenant_id, dataset_id=dataset_id)
return helper.compact_generate_response(
PipelineGenerator.convert_to_event_stream(
rag_pipeline_service.run_datasource_workflow_node(
pipeline=pipeline,
node_id=node_id,
user_inputs=datasource_node_run_api_entity.inputs,
account=current_user,
datasource_type=datasource_node_run_api_entity.datasource_type,
is_published=datasource_node_run_api_entity.is_published,
credential_id=datasource_node_run_api_entity.credential_id,
)
)
)
@service_api_ns.route(f"/datasets/{uuid:dataset_id}/pipeline/run")
class PipelineRunApi(DatasetApiResource):
"""Resource for datasource node run."""
@service_api_ns.doc(shortcut="pipeline_datasource_node_run")
@service_api_ns.doc(description="Run a datasource node for a rag pipeline")
@service_api_ns.doc(
path={
"dataset_id": "Dataset ID",
}
)
@service_api_ns.doc(
body={
"inputs": "User input variables",
"datasource_type": "Datasource type, e.g. online_document",
"datasource_info_list": "Datasource info list",
"start_node_id": "Start node ID",
"is_published": "Whether to get published or draft datasource plugins "
"(true for published, false for draft, default: true)",
"streaming": "Whether to stream the response(streaming or blocking), default: streaming",
}
)
@service_api_ns.doc(
responses={
200: "Pipeline run successfully",
401: "Unauthorized - invalid API token",
}
)
def post(self, tenant_id: str, dataset_id: str):
"""Resource for running a rag pipeline."""
parser: RequestParser = reqparse.RequestParser()
parser.add_argument("inputs", type=dict, required=True, nullable=False, location="json")
parser.add_argument("datasource_type", type=str, required=True, location="json")
parser.add_argument("datasource_info_list", type=list, required=True, location="json")
parser.add_argument("start_node_id", type=str, required=True, location="json")
parser.add_argument("is_published", type=bool, required=True, default=True, location="json")
parser.add_argument(
"response_mode",
type=str,
required=True,
choices=["streaming", "blocking"],
default="blocking",
location="json",
)
args: ParseResult = parser.parse_args()
if not isinstance(current_user, Account):
raise Forbidden()
rag_pipeline_service: RagPipelineService = RagPipelineService()
pipeline: Pipeline = rag_pipeline_service.get_pipeline(tenant_id=tenant_id, dataset_id=dataset_id)
try:
response: dict[Any, Any] | Generator[str, Any, None] = PipelineGenerateService.generate(
pipeline=pipeline,
user=current_user,
args=args,
invoke_from=InvokeFrom.PUBLISHED if args.get("is_published") else InvokeFrom.DEBUGGER,
streaming=args.get("response_mode") == "streaming",
)
return helper.compact_generate_response(response)
except Exception as ex:
raise PipelineRunError(description=str(ex))
@service_api_ns.route("/datasets/pipeline/file-upload")
class KnowledgebasePipelineFileUploadApi(DatasetApiResource):
"""Resource for uploading a file to a knowledgebase pipeline."""
@service_api_ns.doc(shortcut="knowledgebase_pipeline_file_upload")
@service_api_ns.doc(description="Upload a file to a knowledgebase pipeline")
@service_api_ns.doc(
responses={
201: "File uploaded successfully",
400: "Bad request - no file or invalid file",
401: "Unauthorized - invalid API token",
413: "File too large",
415: "Unsupported file type",
}
)
def post(self, tenant_id: str):
"""Upload a file for use in conversations.
Accepts a single file upload via multipart/form-data.
"""
# check file
if "file" not in request.files:
raise NoFileUploadedError()
if len(request.files) > 1:
raise TooManyFilesError()
file = request.files["file"]
if not file.mimetype:
raise UnsupportedFileTypeError()
if not file.filename:
raise FilenameNotExistsError
if not current_user:
raise ValueError("Invalid user account")
try:
upload_file = FileService(db.engine).upload_file(
filename=file.filename,
content=file.read(),
mimetype=file.mimetype,
user=current_user,
)
except services.errors.file.FileTooLargeError as file_too_large_error:
raise FileTooLargeError(file_too_large_error.description)
except services.errors.file.UnsupportedFileTypeError:
raise UnsupportedFileTypeError()
return {
"id": upload_file.id,
"name": upload_file.name,
"size": upload_file.size,
"extension": upload_file.extension,
"mime_type": upload_file.mime_type,
"created_by": upload_file.created_by,
"created_at": upload_file.created_at,
}, 201