Feat/dataset notion import (#392)

Co-authored-by: StyleZhang <jasonapring2015@outlook.com>
Co-authored-by: JzoNg <jzongcode@gmail.com>

api/.env.example

@@ -22,6 +22,7 @@ CELERY_BROKER_URL=redis://:difyai123456@localhost:6379/1
 # redis configuration
 REDIS_HOST=localhost
 REDIS_PORT=6379
+REDIS_USERNAME: ''
 REDIS_PASSWORD=difyai123456
 REDIS_DB=0
 

api/app.py

@@ -20,7 +20,7 @@ from extensions.ext_database import db
 from extensions.ext_login import login_manager
 
 # DO NOT REMOVE BELOW
-from models import model, account, dataset, web, task
+from models import model, account, dataset, web, task, source
 from events import event_handlers
 # DO NOT REMOVE ABOVE
 

api/config.py

@@ -187,6 +187,9 @@ class Config:
         # For temp use only
         # set default LLM provider, default is 'openai', support `azure_openai`
         self.DEFAULT_LLM_PROVIDER = get_env('DEFAULT_LLM_PROVIDER')
+        # notion import setting
+        self.NOTION_CLIENT_ID = get_env('NOTION_CLIENT_ID')
+        self.NOTION_CLIENT_SECRET = get_env('NOTION_CLIENT_SECRET')
 
 class CloudEditionConfig(Config):
 

api/controllers/console/__init__.py

@@ -12,10 +12,10 @@ from . import setup, version, apikey, admin
 from .app import app, site, completion, model_config, statistic, conversation, message, generator
 
 # Import auth controllers
-from .auth import login, oauth
+from .auth import login, oauth, data_source_oauth
 
 # Import datasets controllers
-from .datasets import datasets, datasets_document, datasets_segments, file, hit_testing
+from .datasets import datasets, datasets_document, datasets_segments, file, hit_testing, data_source
 
 # Import workspace controllers
 from .workspace import workspace, members, providers, account

api/controllers/console/auth/data_source_oauth.py

+import logging
+from datetime import datetime
+from typing import Optional
+
+import flask_login
+import requests
+from flask import request, redirect, current_app, session
+from flask_login import current_user, login_required
+from flask_restful import Resource
+from werkzeug.exceptions import Forbidden
+from libs.oauth_data_source import NotionOAuth
+from controllers.console import api
+from ..setup import setup_required
+from ..wraps import account_initialization_required
+
+
+def get_oauth_providers():
+    with current_app.app_context():
+        notion_oauth = NotionOAuth(client_id=current_app.config.get('NOTION_CLIENT_ID'),
+                                   client_secret=current_app.config.get(
+                                       'NOTION_CLIENT_SECRET'),
+                                   redirect_uri=current_app.config.get(
+                                       'CONSOLE_URL') + '/console/api/oauth/data-source/callback/notion')
+
+        OAUTH_PROVIDERS = {
+            'notion': notion_oauth
+        }
+        return OAUTH_PROVIDERS
+
+
+class OAuthDataSource(Resource):
+    def get(self, provider: str):
+        # The role of the current user in the table must be admin or owner
+        if current_user.current_tenant.current_role not in ['admin', 'owner']:
+            raise Forbidden()
+        OAUTH_DATASOURCE_PROVIDERS = get_oauth_providers()
+        with current_app.app_context():
+            oauth_provider = OAUTH_DATASOURCE_PROVIDERS.get(provider)
+            print(vars(oauth_provider))
+        if not oauth_provider:
+            return {'error': 'Invalid provider'}, 400
+
+        auth_url = oauth_provider.get_authorization_url()
+        return redirect(auth_url)
+
+
+class OAuthDataSourceCallback(Resource):
+    def get(self, provider: str):
+        OAUTH_DATASOURCE_PROVIDERS = get_oauth_providers()
+        with current_app.app_context():
+            oauth_provider = OAUTH_DATASOURCE_PROVIDERS.get(provider)
+        if not oauth_provider:
+            return {'error': 'Invalid provider'}, 400
+        if 'code' in request.args:
+            code = request.args.get('code')
+            try:
+                oauth_provider.get_access_token(code)
+            except requests.exceptions.HTTPError as e:
+                logging.exception(
+                    f"An error occurred during the OAuthCallback process with {provider}: {e.response.text}")
+                return {'error': 'OAuth data source process failed'}, 400
+
+            return redirect(f'{current_app.config.get("CONSOLE_URL")}?oauth_data_source=success')
+        elif 'error' in request.args:
+            error = request.args.get('error')
+            return redirect(f'{current_app.config.get("CONSOLE_URL")}?oauth_data_source={error}')
+        else:
+            return redirect(f'{current_app.config.get("CONSOLE_URL")}?oauth_data_source=access_denied')
+
+
+class OAuthDataSourceSync(Resource):
+    @setup_required
+    @login_required
+    @account_initialization_required
+    def get(self, provider, binding_id):
+        provider = str(provider)
+        binding_id = str(binding_id)
+        OAUTH_DATASOURCE_PROVIDERS = get_oauth_providers()
+        with current_app.app_context():
+            oauth_provider = OAUTH_DATASOURCE_PROVIDERS.get(provider)
+        if not oauth_provider:
+            return {'error': 'Invalid provider'}, 400
+        try:
+            oauth_provider.sync_data_source(binding_id)
+        except requests.exceptions.HTTPError as e:
+            logging.exception(
+                f"An error occurred during the OAuthCallback process with {provider}: {e.response.text}")
+            return {'error': 'OAuth data source process failed'}, 400
+
+        return {'result': 'success'}, 200
+
+
+api.add_resource(OAuthDataSource, '/oauth/data-source/<string:provider>')
+api.add_resource(OAuthDataSourceCallback, '/oauth/data-source/callback/<string:provider>')
+api.add_resource(OAuthDataSourceSync, '/oauth/data-source/<string:provider>/<uuid:binding_id>/sync')

api/controllers/console/datasets/datasets.py

@@ -12,8 +12,9 @@ from controllers.console.wraps import account_initialization_required
 from core.indexing_runner import IndexingRunner
 from libs.helper import TimestampField
 from extensions.ext_database import db
+from models.dataset import DocumentSegment, Document
 from models.model import UploadFile
-from services.dataset_service import DatasetService
+from services.dataset_service import DatasetService, DocumentService
 
 dataset_detail_fields = {
     'id': fields.String,
@@ -217,17 +218,31 @@ class DatasetIndexingEstimateApi(Resource):
     @login_required
     @account_initialization_required
     def post(self):
-        segment_rule = request.get_json()
-        file_detail = db.session.query(UploadFile).filter(
-            UploadFile.tenant_id == current_user.current_tenant_id,
-            UploadFile.id == segment_rule["file_id"]
-        ).first()
-
-        if file_detail is None:
-            raise NotFound("File not found.")
-
-        indexing_runner = IndexingRunner()
-        response = indexing_runner.indexing_estimate(file_detail, segment_rule['process_rule'])
+        parser = reqparse.RequestParser()
+        parser.add_argument('info_list', type=dict, required=True, nullable=True, location='json')
+        parser.add_argument('process_rule', type=dict, required=True, nullable=True, location='json')
+        args = parser.parse_args()
+        # validate args
+        DocumentService.estimate_args_validate(args)
+        if args['info_list']['data_source_type'] == 'upload_file':
+            file_ids = args['info_list']['file_info_list']['file_ids']
+            file_details = db.session.query(UploadFile).filter(
+                UploadFile.tenant_id == current_user.current_tenant_id,
+                UploadFile.id.in_(file_ids)
+            ).all()
+
+            if file_details is None:
+                raise NotFound("File not found.")
+
+            indexing_runner = IndexingRunner()
+            response = indexing_runner.file_indexing_estimate(file_details, args['process_rule'])
+        elif args['info_list']['data_source_type'] == 'notion_import':
+
+            indexing_runner = IndexingRunner()
+            response = indexing_runner.notion_indexing_estimate(args['info_list']['notion_info_list'],
+                                                                args['process_rule'])
+        else:
+            raise ValueError('Data source type not support')
         return response, 200
 
 
@@ -274,8 +289,54 @@ class DatasetRelatedAppListApi(Resource):
         }, 200
 
 
+class DatasetIndexingStatusApi(Resource):
+    document_status_fields = {
+        'id': fields.String,
+        'indexing_status': fields.String,
+        'processing_started_at': TimestampField,
+        'parsing_completed_at': TimestampField,
+        'cleaning_completed_at': TimestampField,
+        'splitting_completed_at': TimestampField,
+        'completed_at': TimestampField,
+        'paused_at': TimestampField,
+        'error': fields.String,
+        'stopped_at': TimestampField,
+        'completed_segments': fields.Integer,
+        'total_segments': fields.Integer,
+    }
+
+    document_status_fields_list = {
+        'data': fields.List(fields.Nested(document_status_fields))
+    }
+
+    @setup_required
+    @login_required
+    @account_initialization_required
+    def get(self, dataset_id):
+        dataset_id = str(dataset_id)
+        documents = db.session.query(Document).filter(
+            Document.dataset_id == dataset_id,
+            Document.tenant_id == current_user.current_tenant_id
+        ).all()
+        documents_status = []
+        for document in documents:
+            completed_segments = DocumentSegment.query.filter(DocumentSegment.completed_at.isnot(None),
+                                                              DocumentSegment.document_id == str(document.id),
+                                                              DocumentSegment.status != 're_segment').count()
+            total_segments = DocumentSegment.query.filter(DocumentSegment.document_id == str(document.id),
+                                                          DocumentSegment.status != 're_segment').count()
+            document.completed_segments = completed_segments
+            document.total_segments = total_segments
+            documents_status.append(marshal(document, self.document_status_fields))
+        data = {
+            'data': documents_status
+        }
+        return data
+
+
 api.add_resource(DatasetListApi, '/datasets')
 api.add_resource(DatasetApi, '/datasets/<uuid:dataset_id>')
 api.add_resource(DatasetQueryApi, '/datasets/<uuid:dataset_id>/queries')
-api.add_resource(DatasetIndexingEstimateApi, '/datasets/file-indexing-estimate')
+api.add_resource(DatasetIndexingEstimateApi, '/datasets/indexing-estimate')
 api.add_resource(DatasetRelatedAppListApi, '/datasets/<uuid:dataset_id>/related-apps')
+api.add_resource(DatasetIndexingStatusApi, '/datasets/<uuid:dataset_id>/indexing-status')

api/controllers/service_api/dataset/document.py

@@ -69,12 +69,16 @@ class DocumentListApi(DatasetApiResource):
         document_data = {
             'data_source': {
                 'type': 'upload_file',
-                'info': upload_file.id
+                'info': [
+                    {
+                        'upload_file_id': upload_file.id
+                    }
+                ]
             }
         }
 
         try:
-            document = DocumentService.save_document_with_dataset_id(
+            documents, batch = DocumentService.save_document_with_dataset_id(
                 dataset=dataset,
                 document_data=document_data,
                 account=dataset.created_by_account,
@@ -83,7 +87,7 @@ class DocumentListApi(DatasetApiResource):
             )
         except ProviderTokenNotInitError:
             raise ProviderNotInitializeError()
-
+        document = documents[0]
         if doc_type and doc_metadata:
             metadata_schema = DocumentService.DOCUMENT_METADATA_SCHEMA[doc_type]
 

api/libs/oauth.py

+import json
 import urllib.parse
 from dataclasses import dataclass
 
 import requests
+from flask_login import current_user
+
+from extensions.ext_database import db
+from models.source import DataSourceBinding
 
 
 @dataclass
@@ -134,3 +139,5 @@ class GoogleOAuth(OAuth):
             name=None,
             email=raw_info['email']
         )
+
+

api/migrations/versions/e32f6ccb87c6_e08af0a69ccefbb59fa80c778efee300bb780980.py

+"""e08af0a69ccefbb59fa80c778efee300bb780980
+
+Revision ID: e32f6ccb87c6
+Revises: a45f4dfde53b
+Create Date: 2023-06-06 19:58:33.103819
+
+"""
+from alembic import op
+import sqlalchemy as sa
+from sqlalchemy.dialects import postgresql
+
+# revision identifiers, used by Alembic.
+revision = 'e32f6ccb87c6'
+down_revision = '614f77cecc48'
+branch_labels = None
+depends_on = None
+
+
+def upgrade():
+    # ### commands auto generated by Alembic - please adjust! ###
+    op.create_table('data_source_bindings',
+    sa.Column('id', postgresql.UUID(), server_default=sa.text('uuid_generate_v4()'), nullable=False),
+    sa.Column('tenant_id', postgresql.UUID(), nullable=False),
+    sa.Column('access_token', sa.String(length=255), nullable=False),
+    sa.Column('provider', sa.String(length=255), nullable=False),
+    sa.Column('source_info', postgresql.JSONB(astext_type=sa.Text()), nullable=False),
+    sa.Column('created_at', sa.DateTime(), server_default=sa.text('CURRENT_TIMESTAMP(0)'), nullable=False),
+    sa.Column('updated_at', sa.DateTime(), server_default=sa.text('CURRENT_TIMESTAMP(0)'), nullable=False),
+    sa.Column('disabled', sa.Boolean(), server_default=sa.text('false'), nullable=True),
+    sa.PrimaryKeyConstraint('id', name='source_binding_pkey')
+    )
+    with op.batch_alter_table('data_source_bindings', schema=None) as batch_op:
+        batch_op.create_index('source_binding_tenant_id_idx', ['tenant_id'], unique=False)
+        batch_op.create_index('source_info_idx', ['source_info'], unique=False, postgresql_using='gin')
+
+    # ### end Alembic commands ###
+
+
+def downgrade():
+    # ### commands auto generated by Alembic - please adjust! ###
+    with op.batch_alter_table('data_source_bindings', schema=None) as batch_op:
+        batch_op.drop_index('source_info_idx', postgresql_using='gin')
+        batch_op.drop_index('source_binding_tenant_id_idx')
+
+    op.drop_table('data_source_bindings')
+    # ### end Alembic commands ###

api/models/dataset.py

@@ -190,7 +190,7 @@ class Document(db.Model):
     doc_type = db.Column(db.String(40), nullable=True)
     doc_metadata = db.Column(db.JSON, nullable=True)
 
-    DATA_SOURCES = ['upload_file']
+    DATA_SOURCES = ['upload_file', 'notion_import']
 
     @property
     def display_status(self):
@@ -242,6 +242,8 @@ class Document(db.Model):
                             'created_at': file_detail.created_at.timestamp()
                         }
                     }
+            elif self.data_source_type == 'notion_import':
+                return json.loads(self.data_source_info)
         return {}
 
     @property

api/models/source.py

+from sqlalchemy.dialects.postgresql import UUID
+
+from extensions.ext_database import db
+from sqlalchemy.dialects.postgresql import JSONB
+
+class DataSourceBinding(db.Model):
+    __tablename__ = 'data_source_bindings'
+    __table_args__ = (
+        db.PrimaryKeyConstraint('id', name='source_binding_pkey'),
+        db.Index('source_binding_tenant_id_idx', 'tenant_id'),
+        db.Index('source_info_idx', "source_info", postgresql_using='gin')
+    )
+
+    id = db.Column(UUID, server_default=db.text('uuid_generate_v4()'))
+    tenant_id = db.Column(UUID, nullable=False)
+    access_token = db.Column(db.String(255), nullable=False)
+    provider = db.Column(db.String(255), nullable=False)
+    source_info = db.Column(JSONB, nullable=False)
+    created_at = db.Column(db.DateTime, nullable=False, server_default=db.text('CURRENT_TIMESTAMP(0)'))
+    updated_at = db.Column(db.DateTime, nullable=False, server_default=db.text('CURRENT_TIMESTAMP(0)'))
+    disabled = db.Column(db.Boolean, nullable=True, server_default=db.text('false'))

api/tasks/clean_notion_document_task.py

+import logging
+import time
+from typing import List
+
+import click
+from celery import shared_task
+
+from core.index.keyword_table_index import KeywordTableIndex
+from core.index.vector_index import VectorIndex
+from extensions.ext_database import db
+from models.dataset import DocumentSegment, Dataset, Document
+
+
+@shared_task
+def clean_notion_document_task(document_ids: List[str], dataset_id: str):
+    """
+    Clean document when document deleted.
+    :param document_ids: document ids
+    :param dataset_id: dataset id
+
+    Usage: clean_notion_document_task.delay(document_ids, dataset_id)
+    """
+    logging.info(click.style('Start clean document when import form notion document deleted: {}'.format(dataset_id), fg='green'))
+    start_at = time.perf_counter()
+
+    try:
+        dataset = db.session.query(Dataset).filter(Dataset.id == dataset_id).first()
+
+        if not dataset:
+            raise Exception('Document has no dataset')
+
+        vector_index = VectorIndex(dataset=dataset)
+        keyword_table_index = KeywordTableIndex(dataset=dataset)
+        for document_id in document_ids:
+            document = db.session.query(Document).filter(
+                Document.id == document_id
+            ).first()
+            db.session.delete(document)
+            segments = db.session.query(DocumentSegment).filter(DocumentSegment.document_id == document_id).all()
+            index_node_ids = [segment.index_node_id for segment in segments]
+
+            # delete from vector index
+            vector_index.del_nodes(index_node_ids)
+
+            # delete from keyword index
+            if index_node_ids:
+                keyword_table_index.del_nodes(index_node_ids)
+
+            for segment in segments:
+                db.session.delete(segment)
+        db.session.commit()
+        end_at = time.perf_counter()
+        logging.info(
+            click.style('Clean document when import form notion document deleted end :: {} latency: {}'.format(
+                dataset_id, end_at - start_at),
+                        fg='green'))
+    except Exception:
+        logging.exception("Cleaned document when import form notion document deleted  failed")

api/tasks/document_indexing_sync_task.py

+import datetime
+import logging
+import time
+
+import click
+from celery import shared_task
+from werkzeug.exceptions import NotFound
+
+from core.data_source.notion import NotionPageReader
+from core.index.keyword_table_index import KeywordTableIndex
+from core.index.vector_index import VectorIndex
+from core.indexing_runner import IndexingRunner, DocumentIsPausedException
+from core.llm.error import ProviderTokenNotInitError
+from extensions.ext_database import db
+from models.dataset import Document, Dataset, DocumentSegment
+from models.source import DataSourceBinding
+
+
+@shared_task
+def document_indexing_sync_task(dataset_id: str, document_id: str):
+    """
+    Async update document
+    :param dataset_id:
+    :param document_id:
+
+    Usage: document_indexing_sync_task.delay(dataset_id, document_id)
+    """
+    logging.info(click.style('Start sync document: {}'.format(document_id), fg='green'))
+    start_at = time.perf_counter()
+
+    document = db.session.query(Document).filter(
+        Document.id == document_id,
+        Document.dataset_id == dataset_id
+    ).first()
+
+    if not document:
+        raise NotFound('Document not found')
+
+    data_source_info = document.data_source_info_dict
+    if document.data_source_type == 'notion_import':
+        if not data_source_info or 'notion_page_id' not in data_source_info \
+                or 'notion_workspace_id' not in data_source_info:
+            raise ValueError("no notion page found")
+        workspace_id = data_source_info['notion_workspace_id']
+        page_id = data_source_info['notion_page_id']
+        page_edited_time = data_source_info['last_edited_time']
+        data_source_binding = DataSourceBinding.query.filter(
+            db.and_(
+                DataSourceBinding.tenant_id == document.tenant_id,
+                DataSourceBinding.provider == 'notion',
+                DataSourceBinding.disabled == False,
+                DataSourceBinding.source_info['workspace_id'] == f'"{workspace_id}"'
+            )
+        ).first()
+        if not data_source_binding:
+            raise ValueError('Data source binding not found.')
+        reader = NotionPageReader(integration_token=data_source_binding.access_token)
+        last_edited_time = reader.get_page_last_edited_time(page_id)
+        # check the page is updated
+        if last_edited_time != page_edited_time:
+            document.indexing_status = 'parsing'
+            document.processing_started_at = datetime.datetime.utcnow()
+            db.session.commit()
+
+            # delete all document segment and index
+            try:
+                dataset = db.session.query(Dataset).filter(Dataset.id == dataset_id).first()
+                if not dataset:
+                    raise Exception('Dataset not found')
+
+                vector_index = VectorIndex(dataset=dataset)
+                keyword_table_index = KeywordTableIndex(dataset=dataset)
+
+                segments = db.session.query(DocumentSegment).filter(DocumentSegment.document_id == document_id).all()
+                index_node_ids = [segment.index_node_id for segment in segments]
+
+                # delete from vector index
+                vector_index.del_nodes(index_node_ids)
+
+                # delete from keyword index
+                if index_node_ids:
+                    keyword_table_index.del_nodes(index_node_ids)
+
+                for segment in segments:
+                    db.session.delete(segment)
+
+                end_at = time.perf_counter()
+                logging.info(
+                    click.style('Cleaned document when document update data source or process rule: {} latency: {}'.format(document_id, end_at - start_at), fg='green'))
+            except Exception:
+                logging.exception("Cleaned document when document update data source or process rule failed")
+            try:
+                indexing_runner = IndexingRunner()
+                indexing_runner.run([document])
+                end_at = time.perf_counter()
+                logging.info(click.style('update document: {} latency: {}'.format(document.id, end_at - start_at), fg='green'))
+            except DocumentIsPausedException:
+                logging.info(click.style('Document update paused, document id: {}'.format(document.id), fg='yellow'))
+            except ProviderTokenNotInitError as e:
+                document.indexing_status = 'error'
+                document.error = str(e.description)
+                document.stopped_at = datetime.datetime.utcnow()
+                db.session.commit()
+            except Exception as e:
+                logging.exception("consume update document failed")
+                document.indexing_status = 'error'
+                document.error = str(e)
+                document.stopped_at = datetime.datetime.utcnow()
+                db.session.commit()

api/tasks/document_indexing_task.py

@@ -13,32 +13,36 @@ from models.dataset import Document
 
 
 @shared_task
-def document_indexing_task(dataset_id: str, document_id: str):
+def document_indexing_task(dataset_id: str, document_ids: list):
     """
     Async process document
     :param dataset_id:
-    :param document_id:
+    :param document_ids:
 
     Usage: document_indexing_task.delay(dataset_id, document_id)
     """
-    logging.info(click.style('Start process document: {}'.format(document_id), fg='green'))
-    start_at = time.perf_counter()
-
-    document = db.session.query(Document).filter(
-        Document.id == document_id,
-        Document.dataset_id == dataset_id
-    ).first()
-
-    if not document:
-        raise NotFound('Document not found')
-
-    document.indexing_status = 'parsing'
-    document.processing_started_at = datetime.datetime.utcnow()
+    documents = []
+    for document_id in document_ids:
+        logging.info(click.style('Start process document: {}'.format(document_id), fg='green'))
+        start_at = time.perf_counter()
+
+        document = db.session.query(Document).filter(
+            Document.id == document_id,
+            Document.dataset_id == dataset_id
+        ).first()
+
+        if not document:
+            raise NotFound('Document not found')
+
+        document.indexing_status = 'parsing'
+        document.processing_started_at = datetime.datetime.utcnow()
+        documents.append(document)
+        db.session.add(document)
     db.session.commit()
 
     try:
         indexing_runner = IndexingRunner()
-        indexing_runner.run(document)
+        indexing_runner.run(documents)
         end_at = time.perf_counter()
         logging.info(click.style('Processed document: {} latency: {}'.format(document.id, end_at - start_at), fg='green'))
     except DocumentIsPausedException:

api/tasks/document_indexing_update_task.py

@@ -67,7 +67,7 @@ def document_indexing_update_task(dataset_id: str, document_id: str):
         logging.exception("Cleaned document when document update data source or process rule failed")
     try:
         indexing_runner = IndexingRunner()
-        indexing_runner.run(document)
+        indexing_runner.run([document])
         end_at = time.perf_counter()
         logging.info(click.style('update document: {} latency: {}'.format(document.id, end_at - start_at), fg='green'))
     except DocumentIsPausedException:

api/tasks/recover_document_indexing_task.py

@@ -34,7 +34,7 @@ def recover_document_indexing_task(dataset_id: str, document_id: str):
     try:
         indexing_runner = IndexingRunner()
         if document.indexing_status in ["waiting", "parsing", "cleaning"]:
-            indexing_runner.run(document)
+            indexing_runner.run([document])
         elif document.indexing_status == "splitting":
             indexing_runner.run_in_splitting_status(document)
         elif document.indexing_status == "indexing":

web/app/(commonLayout)/datasets/(datasetDetailLayout)/[datasetId]/layout.tsx

@@ -28,6 +28,7 @@ import Indicator from '@/app/components/header/indicator'
 import AppIcon from '@/app/components/base/app-icon'
 import Loading from '@/app/components/base/loading'
 import DatasetDetailContext from '@/context/dataset-detail'
+import { DataSourceType } from '@/models/datasets'
 
 // import { fetchDatasetDetail } from '@/service/datasets'
 
@@ -162,7 +163,7 @@ const DatasetDetailLayout: FC<IAppDetailLayoutProps> = (props) => {
         desc={datasetRes?.description || '--'}
         navigation={navigation}
         extraInfo={<ExtraInfo />}
-        iconType='dataset'
+        iconType={datasetRes?.data_source_type === DataSourceType.NOTION ? 'notion' : 'dataset'}
       />}
       <DatasetDetailContext.Provider value={{
         indexingTechnique: datasetRes?.indexing_technique,

web/app/components/app-sidebar/basic.tsx

@@ -14,7 +14,7 @@ export function randomString(length: number) {
 }
 
 export type IAppBasicProps = {
-  iconType?: 'app' | 'api' | 'dataset' | 'webapp'
+  iconType?: 'app' | 'api' | 'dataset' | 'webapp' | 'notion'
   icon?: string
   icon_background?: string
   name: string
@@ -40,11 +40,25 @@ const WebappSvg = <svg width="16" height="18" viewBox="0 0 16 18" fill="none" xm
   <path d="M14.375 5.45825L7.99998 8.99992M7.99998 8.99992L1.62498 5.45825M7.99998 8.99992L8 16.1249M14.75 12.0439V5.95603C14.75 5.69904 14.75 5.57055 14.7121 5.45595C14.6786 5.35457 14.6239 5.26151 14.5515 5.18299C14.4697 5.09424 14.3574 5.03184 14.1328 4.90704L8.58277 1.8237C8.37007 1.70553 8.26372 1.64645 8.15109 1.62329C8.05141 1.60278 7.9486 1.60278 7.84891 1.62329C7.73628 1.64645 7.62993 1.70553 7.41723 1.8237L1.86723 4.90704C1.64259 5.03184 1.53026 5.09424 1.44847 5.18299C1.37612 5.26151 1.32136 5.35457 1.28786 5.45595C1.25 5.57055 1.25 5.69904 1.25 5.95603V12.0439C1.25 12.3008 1.25 12.4293 1.28786 12.5439C1.32136 12.6453 1.37612 12.7384 1.44847 12.8169C1.53026 12.9056 1.64259 12.968 1.86723 13.0928L7.41723 16.1762C7.62993 16.2943 7.73628 16.3534 7.84891 16.3766C7.9486 16.3971 8.05141 16.3971 8.15109 16.3766C8.26372 16.3534 8.37007 16.2943 8.58277 16.1762L14.1328 13.0928C14.3574 12.968 14.4697 12.9056 14.5515 12.8169C14.6239 12.7384 14.6786 12.6453 14.7121 12.5439C14.75 12.4293 14.75 12.3008 14.75 12.0439Z" stroke="#155EEF" strokeWidth="1.5" strokeLinecap="round" strokeLinejoin="round"/>
 </svg>
 
+const NotionSvg = <svg width="24" height="24" viewBox="0 0 24 24" fill="none" xmlns="http://www.w3.org/2000/svg">
+  <g clip-path="url(#clip0_6294_13848)">
+    <path fill-rule="evenodd" clip-rule="evenodd" d="M4.287 21.9133L1.70748 18.6999C1.08685 17.9267 0.75 16.976 0.75 15.9974V4.36124C0.75 2.89548 1.92269 1.67923 3.43553 1.57594L15.3991 0.759137C16.2682 0.699797 17.1321 0.930818 17.8461 1.41353L22.0494 4.25543C22.8018 4.76414 23.25 5.59574 23.25 6.48319V19.7124C23.25 21.1468 22.0969 22.3345 20.6157 22.4256L7.3375 23.243C6.1555 23.3158 5.01299 22.8178 4.287 21.9133Z" fill="white"/>
+    <path d="M8.43607 10.1842V10.0318C8.43607 9.64564 8.74535 9.32537 9.14397 9.29876L12.0475 9.10491L16.0628 15.0178V9.82823L15.0293 9.69046V9.6181C15.0293 9.22739 15.3456 8.90501 15.7493 8.88433L18.3912 8.74899V9.12918C18.3912 9.30765 18.2585 9.46031 18.0766 9.49108L17.4408 9.59861V18.0029L16.6429 18.2773C15.9764 18.5065 15.2343 18.2611 14.8527 17.6853L10.9545 11.803V17.4173L12.1544 17.647L12.1377 17.7583C12.0853 18.1069 11.7843 18.3705 11.4202 18.3867L8.43607 18.5195C8.39662 18.1447 8.67758 17.8093 9.06518 17.7686L9.45771 17.7273V10.2416L8.43607 10.1842Z" fill="black"/>
+    <path fill-rule="evenodd" clip-rule="evenodd" d="M15.5062 2.22521L3.5426 3.04201C2.82599 3.09094 2.27051 3.66706 2.27051 4.36136V15.9975C2.27051 16.6499 2.49507 17.2837 2.90883 17.7992L5.48835 21.0126C5.90541 21.5322 6.56174 21.8183 7.24076 21.7765L20.519 20.9591C21.1995 20.9172 21.7293 20.3716 21.7293 19.7125V6.48332C21.7293 6.07557 21.5234 5.69348 21.1777 5.45975L16.9743 2.61784C16.546 2.32822 16.0277 2.1896 15.5062 2.22521ZM4.13585 4.54287C3.96946 4.41968 4.04865 4.16303 4.25768 4.14804L15.5866 3.33545C15.9476 3.30956 16.3063 3.40896 16.5982 3.61578L18.8713 5.22622C18.9576 5.28736 18.9171 5.41935 18.8102 5.42516L6.8129 6.07764C6.44983 6.09739 6.09144 5.99073 5.80276 5.77699L4.13585 4.54287ZM6.25018 8.12315C6.25018 7.7334 6.56506 7.41145 6.9677 7.38952L19.6523 6.69871C20.0447 6.67734 20.375 6.97912 20.375 7.35898V18.8141C20.375 19.2031 20.0613 19.5247 19.6594 19.5476L7.05516 20.2648C6.61845 20.2896 6.25018 19.954 6.25018 19.5312V8.12315Z" fill="black"/>
+  </g>
+  <defs>
+    <clipPath id="clip0_6294_13848">
+      <rect width="24" height="24" fill="white"/>
+    </clipPath>
+  </defs>
+</svg>
+
 const ICON_MAP = {
   app: <AppIcon className='border !border-[rgba(0,0,0,0.05)]' />,
   api: <AppIcon innerIcon={ApiSvg} className='border !bg-purple-50 !border-purple-200' />,
   dataset: <AppIcon innerIcon={DatasetSvg} className='!border-[0.5px] !border-indigo-100 !bg-indigo-25' />,
   webapp: <AppIcon innerIcon={WebappSvg} className='border !bg-primary-100 !border-primary-200' />,
+  notion: <AppIcon innerIcon={NotionSvg} className='!border-[0.5px] !border-indigo-100 !bg-white' />,
 }
 
 export default function AppBasic({ icon, icon_background, name, type, hoverTip, textStyle, iconType = 'app' }: IAppBasicProps) {

web/app/components/app-sidebar/index.tsx

@@ -4,7 +4,7 @@ import NavLink from './navLink'
 import AppBasic from './basic'
 
 export type IAppDetailNavProps = {
-  iconType?: 'app' | 'dataset'
+  iconType?: 'app' | 'dataset' | 'notion'
   title: string
   desc: string
   icon: string
@@ -18,7 +18,6 @@ export type IAppDetailNavProps = {
   extraInfo?: React.ReactNode
 }
 
-
 const AppDetailNav: FC<IAppDetailNavProps> = ({ title, desc, icon, icon_background, navigation, extraInfo, iconType = 'app' }) => {
   return (
     <div className="flex flex-col w-56 overflow-y-auto bg-white border-r border-gray-200 shrink-0">

web/app/components/base/checkbox/assets/check.svg

+<svg width="12" height="12" viewBox="0 0 12 12" fill="none" xmlns="http://www.w3.org/2000/svg">
+<path d="M10 3L4.5 8.5L2 6" stroke="white" stroke-width="2" stroke-linecap="round" stroke-linejoin="round"/>
+</svg>

web/app/components/base/checkbox/index.module.css

+.wrapper {
+  border-color: #d0d5dd;
+}
+
+.checked {
+  background: #155eef url(./assets/check.svg) center center no-repeat;
+  background-size: 12px 12px;
+  border-color: #155eef;
+}
\ No newline at end of file

web/app/components/base/checkbox/index.tsx

+import cn from 'classnames'
+import s from './index.module.css'
+
+type CheckboxProps = {
+  checked?: boolean
+  onCheck?: () => void
+  className?: string
+}
+
+const Checkbox = ({ checked, onCheck, className }: CheckboxProps) => {
+  return (
+    <div
+      className={cn(s.wrapper, checked && s.checked, 'w-4 h-4 border rounded border-gray-300', className)}
+      onClick={onCheck}
+    />
+  )
+}
+
+export default Checkbox

web/app/components/base/notion-icon/index.module.css

+.default-page-icon {
+  width: 20px;
+  height: 20px;
+  background: url(../notion-page-selector/assets/notion-page.svg) center center no-repeat;
+  background-size: cover;
+}
\ No newline at end of file

web/app/components/base/notion-icon/index.tsx

+import cn from 'classnames'
+import s from './index.module.css'
+import type { DataSourceNotionPage } from '@/models/common'
+
+type IconTypes = 'workspace' | 'page'
+type NotionIconProps = {
+  type?: IconTypes
+  name?: string | null
+  className?: string
+  src?: string | null | Pick<DataSourceNotionPage, 'page_icon'>['page_icon']
+}
+const NotionIcon = ({
+  type = 'workspace',
+  src,
+  name,
+  className,
+}: NotionIconProps) => {
+  if (type === 'workspace') {
+    if (typeof src === 'string') {
+      if (src.startsWith('https://') || src.startsWith('http://')) {
+        return (
+          <img
+            alt='workspace icon'
+            src={src}
+            className={cn('block object-cover w-5 h-5', className)}
+          />
+        )
+      }
+      return (
+        <div className={cn('flex items-center justify-center w-5 h-5', className)}>{src}</div>
+      )
+    }
+    return (
+      <div className={cn('flex items-center justify-center w-5 h-5 bg-gray-200 text-xs font-medium text-gray-500 rounded', className)}>{name?.[0].toLocaleUpperCase()}</div>
+    )
+  }
+
+  if (typeof src === 'object' && src !== null) {
+    if (src?.type === 'url') {
+      return (
+        <img
+          alt='page icon'
+          src={src.url || ''}
+          className={cn('block object-cover w-5 h-5', className)}
+        />
+      )
+    }
+    return (
+      <div className={cn('flex items-center justify-center w-5 h-5', className)}>{src?.emoji}</div>
+    )
+  }
+
+  return (
+    <div className={cn(s['default-page-icon'], className)} />
+  )
+}
+
+export default NotionIcon

web/app/components/base/notion-page-selector/assets/clear.svg

+<svg width="16" height="16" viewBox="0 0 16 16" fill="none" xmlns="http://www.w3.org/2000/svg">
+<path fill-rule="evenodd" clip-rule="evenodd" d="M8 2.5C4.96243 2.5 2.5 4.96243 2.5 8C2.5 11.0376 4.96243 13.5 8 13.5C11.0376 13.5 13.5 11.0376 13.5 8C13.5 4.96243 11.0376 2.5 8 2.5ZM9.85355 6.14645C10.0488 6.34171 10.0488 6.65829 9.85355 6.85355L8.70711 8L9.85355 9.14645C10.0488 9.34171 10.0488 9.65829 9.85355 9.85355C9.65829 10.0488 9.34171 10.0488 9.14645 9.85355L8 8.70711L6.85355 9.85355C6.65829 10.0488 6.34171 10.0488 6.14645 9.85355C5.95118 9.65829 5.95118 9.34171 6.14645 9.14645L7.29289 8L6.14645 6.85355C5.95118 6.65829 5.95118 6.34171 6.14645 6.14645C6.34171 5.95118 6.65829 5.95118 6.85355 6.14645L8 7.29289L9.14645 6.14645C9.34171 5.95118 9.65829 5.95118 9.85355 6.14645Z" fill="#98A2B3"/>
+</svg>

web/app/components/base/notion-page-selector/assets/down-arrow.svg

+<svg width="12" height="12" viewBox="0 0 12 12" fill="none" xmlns="http://www.w3.org/2000/svg">
+<path d="M3 4.5L6 7.5L9 4.5" stroke="#344054" stroke-width="1.5" stroke-linecap="round" stroke-linejoin="round"/>
+</svg>

web/app/components/base/notion-page-selector/assets/notion-empty-page.svg

+<svg width="20" height="21" viewBox="0 0 20 21" fill="none" xmlns="http://www.w3.org/2000/svg">
+<path d="M5.49939 19.1498H13.6897C15.3354 19.1498 16.1891 18.2807 16.1891 16.6273V9.6521C16.1891 8.58313 16.0507 8.09095 15.3816 7.41418L11.3441 3.30749C10.6981 2.65381 10.1675 2.5 9.20618 2.5H5.49939C3.85363 2.5 3 3.36902 3 5.02246V16.6273C3 18.2884 3.85363 19.1498 5.49939 19.1498ZM5.62243 17.6424C4.87646 17.6424 4.50732 17.2502 4.50732 16.5351V5.11475C4.50732 4.40722 4.87646 4.00732 5.62243 4.00732H8.89856V8.22168C8.89856 9.32142 9.44457 9.85205 10.5366 9.85205H14.6818V16.5351C14.6818 17.2502 14.3049 17.6424 13.5589 17.6424H5.62243ZM10.675 8.52929C10.3597 8.52929 10.229 8.39087 10.229 8.07556V4.21496L14.4741 8.52929H10.675Z" fill="#37352F" fill-opacity="0.45"/>
+</svg>

web/app/components/base/notion-page-selector/assets/notion-page.svg

+<svg width="20" height="21" viewBox="0 0 20 21" fill="none" xmlns="http://www.w3.org/2000/svg">
+<path d="M5.49939 19.1498H13.6897C15.3354 19.1498 16.1891 18.2807 16.1891 16.6273V9.6521C16.1891 8.58313 16.0507 8.09095 15.3816 7.41418L11.3441 3.30749C10.6981 2.65381 10.1675 2.5 9.20618 2.5H5.49939C3.85363 2.5 3 3.36902 3 5.02246V16.6273C3 18.2884 3.85363 19.1498 5.49939 19.1498ZM5.62243 17.6424C4.87645 17.6424 4.50732 17.2502 4.50732 16.5351V5.11475C4.50732 4.40722 4.87645 4.00732 5.62243 4.00732H8.89856V8.22168C8.89856 9.32142 9.44457 9.85205 10.5366 9.85205H14.6818V16.5351C14.6818 17.2502 14.3049 17.6424 13.5589 17.6424H5.62243ZM10.675 8.52929C10.3597 8.52929 10.229 8.39087 10.229 8.07556V4.21496L14.4741 8.52929H10.675ZM12.3362 11.8746H6.70678C6.41454 11.8746 6.2069 12.09 6.2069 12.3591C6.2069 12.636 6.41454 12.8513 6.70678 12.8513H12.3362C12.613 12.8513 12.8207 12.636 12.8207 12.3591C12.8207 12.09 12.613 11.8746 12.3362 11.8746ZM12.3362 14.4587H6.70678C6.41454 14.4587 6.2069 14.674 6.2069 14.9509C6.2069 15.22 6.41454 15.4276 6.70678 15.4276H12.3362C12.613 15.4276 12.8207 15.22 12.8207 14.9509C12.8207 14.674 12.613 14.4587 12.3362 14.4587Z" fill="#37352F" fill-opacity="0.45"/>
+</svg>

web/app/components/base/notion-page-selector/assets/search.svg

+<svg width="14" height="14" viewBox="0 0 14 14" fill="none" xmlns="http://www.w3.org/2000/svg">
+<g id="Icon">
+<path id="Icon_2" d="M12.25 12.25L10.2084 10.2083M11.6667 6.70833C11.6667 9.44675 9.44675 11.6667 6.70833 11.6667C3.96992 11.6667 1.75 9.44675 1.75 6.70833C1.75 3.96992 3.96992 1.75 6.70833 1.75C9.44675 1.75 11.6667 3.96992 11.6667 6.70833Z" stroke="#344054" stroke-width="1.25" stroke-linecap="round" stroke-linejoin="round"/>
+</g>
+</svg>

web/app/components/base/notion-page-selector/assets/setting.svg

+<svg width="14" height="14" viewBox="0 0 14 14" fill="none" xmlns="http://www.w3.org/2000/svg">
+<g clip-path="url(#clip0_5943_4745)">
+<path d="M6.99984 8.74984C7.96634 8.74984 8.74984 7.96634 8.74984 6.99984C8.74984 6.03334 7.96634 5.24984 6.99984 5.24984C6.03334 5.24984 5.24984 6.03334 5.24984 6.99984C5.24984 7.96634 6.03334 8.74984 6.99984 8.74984Z" stroke="#667085" stroke-width="1.25" stroke-linecap="round" stroke-linejoin="round"/>
+<path d="M10.9241 8.59075C10.8535 8.75069 10.8324 8.92812 10.8636 9.10015C10.8948 9.27218 10.9768 9.43092 11.0991 9.5559L11.1309 9.58772C11.2295 9.68622 11.3077 9.80319 11.3611 9.93195C11.4145 10.0607 11.442 10.1987 11.442 10.3381C11.442 10.4775 11.4145 10.6155 11.3611 10.7442C11.3077 10.873 11.2295 10.99 11.1309 11.0885C11.0324 11.1871 10.9154 11.2653 10.7867 11.3187C10.6579 11.3721 10.5199 11.3995 10.3805 11.3995C10.2411 11.3995 10.1031 11.3721 9.97437 11.3187C9.84561 11.2653 9.72864 11.1871 9.63014 11.0885L9.59832 11.0567C9.47334 10.9344 9.3146 10.8524 9.14257 10.8212C8.97055 10.79 8.79312 10.8111 8.63317 10.8817C8.47632 10.9489 8.34256 11.0605 8.24833 11.2028C8.15411 11.345 8.10355 11.5118 8.10287 11.6824V11.7726C8.10287 12.0539 7.99112 12.3236 7.79222 12.5225C7.59332 12.7214 7.32355 12.8332 7.04226 12.8332C6.76097 12.8332 6.4912 12.7214 6.2923 12.5225C6.0934 12.3236 5.98166 12.0539 5.98166 11.7726V11.7248C5.97755 11.5493 5.92073 11.3791 5.81859 11.2363C5.71645 11.0935 5.57371 10.9847 5.40893 10.9241C5.24898 10.8535 5.07155 10.8324 4.89953 10.8636C4.7275 10.8948 4.56876 10.9768 4.44378 11.0991L4.41196 11.1309C4.31346 11.2295 4.19648 11.3077 4.06773 11.3611C3.93897 11.4145 3.80096 11.442 3.66158 11.442C3.5222 11.442 3.38419 11.4145 3.25543 11.3611C3.12668 11.3077 3.0097 11.2295 2.9112 11.1309C2.81259 11.0324 2.73436 10.9154 2.68099 10.7867C2.62761 10.6579 2.60014 10.5199 2.60014 10.3805C2.60014 10.2411 2.62761 10.1031 2.68099 9.97437C2.73436 9.84561 2.81259 9.72864 2.9112 9.63014L2.94302 9.59832C3.06527 9.47334 3.14728 9.3146 3.17848 9.14257C3.20967 8.97055 3.18861 8.79312 3.11802 8.63317C3.0508 8.47632 2.93918 8.34256 2.7969 8.24833C2.65463 8.15411 2.48791 8.10355 2.31726 8.10287H2.22711C1.94582 8.10287 1.67605 7.99112 1.47715 7.79222C1.27825 7.59332 1.1665 7.32355 1.1665 7.04226C1.1665 6.76097 1.27825 6.4912 1.47715 6.2923C1.67605 6.0934 1.94582 5.98166 2.22711 5.98166H2.27484C2.45036 5.97755 2.6206 5.92073 2.7634 5.81859C2.90621 5.71645 3.01499 5.57371 3.07559 5.40893C3.14619 5.24898 3.16724 5.07155 3.13605 4.89953C3.10486 4.7275 3.02285 4.56876 2.90059 4.44378L2.86878 4.41196C2.77017 4.31346 2.69194 4.19648 2.63856 4.06773C2.58519 3.93897 2.55772 3.80096 2.55772 3.66158C2.55772 3.5222 2.58519 3.38419 2.63856 3.25543C2.69194 3.12668 2.77017 3.0097 2.86878 2.9112C2.96728 2.81259 3.08425 2.73436 3.21301 2.68099C3.34176 2.62761 3.47978 2.60014 3.61916 2.60014C3.75854 2.60014 3.89655 2.62761 4.0253 2.68099C4.15406 2.73436 4.27103 2.81259 4.36953 2.9112L4.40135 2.94302C4.52633 3.06527 4.68507 3.14728 4.8571 3.17848C5.02913 3.20967 5.20656 3.18861 5.3665 3.11802H5.40893C5.56578 3.0508 5.69954 2.93918 5.79377 2.7969C5.88799 2.65463 5.93855 2.48791 5.93923 2.31726V2.22711C5.93923 1.94582 6.05097 1.67605 6.24988 1.47715C6.44878 1.27825 6.71855 1.1665 6.99984 1.1665C7.28113 1.1665 7.5509 1.27825 7.7498 1.47715C7.9487 1.67605 8.06044 1.94582 8.06044 2.22711V2.27484C8.06112 2.44548 8.11169 2.6122 8.20591 2.75448C8.30013 2.89675 8.4339 3.00837 8.59075 3.07559C8.75069 3.14619 8.92812 3.16724 9.10015 3.13605C9.27218 3.10486 9.43092 3.02285 9.5559 2.90059L9.58772 2.86878C9.68622 2.77017 9.80319 2.69194 9.93195 2.63856C10.0607 2.58519 10.1987 2.55772 10.3381 2.55772C10.4775 2.55772 10.6155 2.58519 10.7442 2.63856C10.873 2.69194 10.99 2.77017 11.0885 2.86878C11.1871 2.96728 11.2653 3.08425 11.3187 3.21301C11.3721 3.34176 11.3995 3.47978 11.3995 3.61916C11.3995 3.75854 11.3721 3.89655 11.3187 4.0253C11.2653 4.15406 11.1871 4.27103 11.0885 4.36953L11.0567 4.40135C10.9344 4.52633 10.8524 4.68507 10.8212 4.8571C10.79 5.02913 10.8111 5.20656 10.8817 5.3665V5.40893C10.9489 5.56578 11.0605 5.69954 11.2028 5.79377C11.345 5.88799 11.5118 5.93855 11.6824 5.93923H11.7726C12.0539 5.93923 12.3236 6.05097 12.5225 6.24988C12.7214 6.44878 12.8332 6.71855 12.8332 6.99984C12.8332 7.28113 12.7214 7.5509 12.5225 7.7498C12.3236 7.9487 12.0539 8.06044 11.7726 8.06044H11.7248C11.5542 8.06112 11.3875 8.11169 11.2452 8.20591C11.1029 8.30013 10.9913 8.4339 10.9241 8.59075Z" stroke="#667085" stroke-width="1.25" stroke-linecap="round" stroke-linejoin="round"/>
+</g>
+<defs>
+<clipPath id="clip0_5943_4745">
+<rect width="14" height="14" fill="white"/>
+</clipPath>
+</defs>
+</svg>

web/app/components/base/notion-page-selector/base.module.css

+.setting-icon {
+  background: url(./assets/setting.svg) center center no-repeat;
+  background-size: 14px 14px;
+}
\ No newline at end of file

web/app/components/base/notion-page-selector/base.tsx

+import { useCallback, useEffect, useMemo, useState } from 'react'
+import useSWR from 'swr'
+import cn from 'classnames'
+import s from './base.module.css'
+import WorkspaceSelector from './workspace-selector'
+import SearchInput from './search-input'
+import PageSelector from './page-selector'
+import { preImportNotionPages } from '@/service/datasets'
+import AccountSetting from '@/app/components/header/account-setting'
+import { NotionConnector } from '@/app/components/datasets/create/step-one'
+import type { DataSourceNotionPage, DataSourceNotionPageMap, DataSourceNotionWorkspace } from '@/models/common'
+
+export type NotionPageSelectorValue = DataSourceNotionPage & { workspace_id: string }
+
+type NotionPageSelectorProps = {
+  value?: string[]
+  onSelect: (selectedPages: NotionPageSelectorValue[]) => void
+  canPreview?: boolean
+  previewPageId?: string
+  onPreview?: (selectedPage: NotionPageSelectorValue) => void
+  datasetId?: string
+}
+
+const NotionPageSelector = ({
+  value,
+  onSelect,
+  canPreview,
+  previewPageId,
+  onPreview,
+  datasetId = '',
+}: NotionPageSelectorProps) => {
+  const { data, mutate } = useSWR({ url: '/notion/pre-import/pages', datasetId }, preImportNotionPages)
+  const [prevData, setPrevData] = useState(data)
+  const [searchValue, setSearchValue] = useState('')
+  const [showDataSourceSetting, setShowDataSourceSetting] = useState(false)
+  const [currentWorkspaceId, setCurrentWorkspaceId] = useState('')
+
+  const notionWorkspaces = useMemo(() => {
+    return data?.notion_info || []
+  }, [data?.notion_info])
+  const firstWorkspaceId = notionWorkspaces[0]?.workspace_id
+  const currentWorkspace = notionWorkspaces.find(workspace => workspace.workspace_id === currentWorkspaceId)
+
+  const getPagesMapAndSelectedPagesId: [DataSourceNotionPageMap, Set<string>] = useMemo(() => {
+    const selectedPagesId = new Set<string>()
+    const pagesMap = notionWorkspaces.reduce((prev: DataSourceNotionPageMap, next: DataSourceNotionWorkspace) => {
+      next.pages.forEach((page) => {
+        if (page.is_bound)
+          selectedPagesId.add(page.page_id)
+        prev[page.page_id] = {
+          ...page,
+          workspace_id: next.workspace_id,
+        }
+      })
+
+      return prev
+    }, {})
+    return [pagesMap, selectedPagesId]
+  }, [notionWorkspaces])
+  const defaultSelectedPagesId = [...Array.from(getPagesMapAndSelectedPagesId[1]), ...(value || [])]
+  const [selectedPagesId, setSelectedPagesId] = useState<Set<string>>(new Set(defaultSelectedPagesId))
+
+  if (prevData !== data) {
+    setPrevData(data)
+    setSelectedPagesId(new Set(defaultSelectedPagesId))
+  }
+
+  const handleSearchValueChange = useCallback((value: string) => {
+    setSearchValue(value)
+  }, [])
+  const handleSelectWorkspace = useCallback((workspaceId: string) => {
+    setCurrentWorkspaceId(workspaceId)
+  }, [])
+  const handleSelecPages = (selectedPagesId: Set<string>) => {
+    setSelectedPagesId(new Set(Array.from(selectedPagesId)))
+    const selectedPages = Array.from(selectedPagesId).map(pageId => getPagesMapAndSelectedPagesId[0][pageId])
+    onSelect(selectedPages)
+  }
+  const handlePreviewPage = (previewPageId: string) => {
+    if (onPreview)
+      onPreview(getPagesMapAndSelectedPagesId[0][previewPageId])
+  }
+
+  useEffect(() => {
+    setCurrentWorkspaceId(firstWorkspaceId)
+  }, [firstWorkspaceId])
+
+  return (
+    <div className='bg-gray-25 border border-gray-200 rounded-xl'>
+      {
+        data?.notion_info?.length
+          ? (
+            <>
+              <div className='flex items-center pl-[10px] pr-2 h-11 bg-white border-b border-b-gray-200 rounded-t-xl'>
+                <WorkspaceSelector
+                  value={currentWorkspaceId || firstWorkspaceId}
+                  items={notionWorkspaces}
+                  onSelect={handleSelectWorkspace}
+                />
+                <div className='mx-1 w-[1px] h-3 bg-gray-200' />
+                <div
+                  className={cn(s['setting-icon'], 'w-6 h-6 cursor-pointer')}
+                  onClick={() => setShowDataSourceSetting(true)}
+                />
+                <div className='grow' />
+                <SearchInput
+                  value={searchValue}
+                  onChange={handleSearchValueChange}
+                />
+              </div>
+              <div className='rounded-b-xl overflow-hidden'>
+                <PageSelector
+                  value={selectedPagesId}
+                  searchValue={searchValue}
+                  list={currentWorkspace?.pages || []}
+                  pagesMap={getPagesMapAndSelectedPagesId[0]}
+                  onSelect={handleSelecPages}
+                  canPreview={canPreview}
+                  previewPageId={previewPageId}
+                  onPreview={handlePreviewPage}
+                />
+              </div>
+            </>
+          )
+          : (
+            <NotionConnector onSetting={() => setShowDataSourceSetting(true)} />
+          )
+      }
+      {
+        showDataSourceSetting && (
+          <AccountSetting activeTab='data-source' onCancel={() => {
+            setShowDataSourceSetting(false)
+            mutate()
+          }} />
+        )
+      }
+    </div>
+  )
+}
+
+export default NotionPageSelector

web/app/components/base/notion-page-selector/index.tsx

+export { default as NotionPageSelectorModal } from './notion-page-selector-modal'
+export { default as NotionPageSelector } from './base'

web/app/components/base/notion-page-selector/notion-page-selector-modal/index.module.css

+.modal {
+  width: 600px !important;
+  max-width: 600px !important;
+  padding: 24px 32px !important;
+}
+
+.operate {
+  padding: 0 8px;
+  min-width: 96px;
+  height: 36px;
+  line-height: 36px;
+  text-align: center;
+  background-color: #ffffff;
+  box-shadow: 0px 1px 2px rgba(16, 24, 40, 0.05);
+  border-radius: 8px;
+  border: 0.5px solid #eaecf0;
+  font-size: 14px;
+  font-weight: 500;
+  color: #667085;
+  cursor: pointer;
+}
+
+.operate-save {
+  margin-left: 8px;
+  border-color: #155eef;
+  background-color: #155eef;
+  color: #ffffff;
+}
\ No newline at end of file

web/app/components/base/notion-page-selector/notion-page-selector-modal/index.tsx

+import { useState } from 'react'
+import { useTranslation } from 'react-i18next'
+import cn from 'classnames'
+import { XMarkIcon } from '@heroicons/react/24/outline'
+import NotionPageSelector from '../base'
+import type { NotionPageSelectorValue } from '../base'
+import s from './index.module.css'
+import Modal from '@/app/components/base/modal'
+
+type NotionPageSelectorModalProps = {
+  isShow: boolean
+  onClose: () => void
+  onSave: (selectedPages: NotionPageSelectorValue[]) => void
+  datasetId: string
+}
+const NotionPageSelectorModal = ({
+  isShow,
+  onClose,
+  onSave,
+  datasetId,
+}: NotionPageSelectorModalProps) => {
+  const { t } = useTranslation()
+  const [selectedPages, setSelectedPages] = useState<NotionPageSelectorValue[]>([])
+
+  const handleClose = () => {
+    onClose()
+  }
+  const handleSelectPage = (newSelectedPages: NotionPageSelectorValue[]) => {
+    setSelectedPages(newSelectedPages)
+  }
+  const handleSave = () => {
+    onSave(selectedPages)
+  }
+
+  return (
+    <Modal
+      className={s.modal}
+      isShow={isShow}
+      onClose={() => {}}
+    >
+      <div className='flex items-center justify-between mb-6 h-8'>
+        <div className='text-xl font-semibold text-gray-900'>{t('common.dataSource.notion.selector.addPages')}</div>
+        <div
+          className='flex items-center justify-center -mr-2 w-8 h-8 cursor-pointer'
+          onClick={handleClose}>
+          <XMarkIcon className='w-4 h-4' />
+        </div>
+      </div>
+      <NotionPageSelector
+        onSelect={handleSelectPage}
+        canPreview={false}
+        datasetId={datasetId}
+      />
+      <div className='mt-8 flex justify-end'>
+        <div className={s.operate} onClick={handleClose}>{t('common.operation.cancel')}</div>
+        <div className={cn(s.operate, s['operate-save'])} onClick={handleSave}>{t('common.operation.save')}</div>
+      </div>
+    </Modal>
+  )
+}
+
+export default NotionPageSelectorModal

web/app/components/base/notion-page-selector/page-selector/index.module.css

+.arrow {
+  width: 20px;
+  height: 20px;
+  background: url(../assets/down-arrow.svg) center center no-repeat;
+  background-size: 16px 16px;
+  transform: rotate(-90deg);
+}
+
+.arrow-expand {
+  transform: rotate(0);
+}
+
+.preview-item {
+  background-color: #eff4ff;
+  border: 1px solid #D1E0FF;
+  box-shadow: 0px 1px 2px rgba(16, 24, 40, 0.05);
+}
\ No newline at end of file

web/app/components/base/notion-page-selector/page-selector/index.tsx

+import { memo, useMemo, useState } from 'react'
+import { useTranslation } from 'react-i18next'
+import { FixedSizeList as List, areEqual } from 'react-window'
+import type { ListChildComponentProps } from 'react-window'
+import cn from 'classnames'
+import Checkbox from '../../checkbox'
+import NotionIcon from '../../notion-icon'
+import s from './index.module.css'
+import type { DataSourceNotionPage, DataSourceNotionPageMap } from '@/models/common'
+
+type PageSelectorProps = {
+  value: Set<string>
+  searchValue: string
+  pagesMap: DataSourceNotionPageMap
+  list: DataSourceNotionPage[]
+  onSelect: (selectedPagesId: Set<string>) => void
+  canPreview?: boolean
+  previewPageId?: string
+  onPreview?: (selectedPageId: string) => void
+}
+type NotionPageTreeItem = {
+  children: Set<string>
+  descendants: Set<string>
+  deepth: number
+  ancestors: string[]
+} & DataSourceNotionPage
+type NotionPageTreeMap = Record<string, NotionPageTreeItem>
+type NotionPageItem = {
+  expand: boolean
+  deepth: number
+} & DataSourceNotionPage
+
+const recursivePushInParentDescendants = (
+  pagesMap: DataSourceNotionPageMap,
+  listTreeMap: NotionPageTreeMap,
+  current: NotionPageTreeItem,
+  leafItem: NotionPageTreeItem,
+) => {
+  const parentId = current.parent_id
+  const pageId = current.page_id
+
+  if (!parentId || !pageId)
+    return
+
+  if (parentId !== 'root' && pagesMap[parentId]) {
+    if (!listTreeMap[parentId]) {
+      const children = new Set([pageId])
+      const descendants = new Set([pageId, leafItem.page_id])
+      listTreeMap[parentId] = {
+        ...pagesMap[parentId],
+        children,
+        descendants,
+        deepth: 0,
+        ancestors: [],
+      }
+    }
+    else {
+      listTreeMap[parentId].children.add(pageId)
+      listTreeMap[parentId].descendants.add(pageId)
+      listTreeMap[parentId].descendants.add(leafItem.page_id)
+    }
+    leafItem.deepth++
+    leafItem.ancestors.unshift(listTreeMap[parentId].page_name)
+
+    if (listTreeMap[parentId].parent_id !== 'root')
+      recursivePushInParentDescendants(pagesMap, listTreeMap, listTreeMap[parentId], leafItem)
+  }
+}
+
+const Item = memo(({ index, style, data }: ListChildComponentProps<{
+  dataList: NotionPageItem[]
+  handleToggle: (index: number) => void
+  checkedIds: Set<string>
+  handleCheck: (index: number) => void
+  canPreview?: boolean
+  handlePreview: (index: number) => void
+  listMapWithChildrenAndDescendants: NotionPageTreeMap
+  searchValue: string
+  previewPageId: string
+  pagesMap: DataSourceNotionPageMap
+}>) => {
+  const { t } = useTranslation()
+  const { dataList, handleToggle, checkedIds, handleCheck, canPreview, handlePreview, listMapWithChildrenAndDescendants, searchValue, previewPageId, pagesMap } = data
+  const current = dataList[index]
+  const currentWithChildrenAndDescendants = listMapWithChildrenAndDescendants[current.page_id]
+  const hasChild = currentWithChildrenAndDescendants.descendants.size > 0
+  const ancestors = currentWithChildrenAndDescendants.ancestors
+  const breadCrumbs = ancestors.length ? [...ancestors, current.page_name] : [current.page_name]
+
+  const renderArrow = () => {
+    if (hasChild) {
+      return (
+        <div
+          className={cn(s.arrow, current.expand && s['arrow-expand'], 'shrink-0 mr-1 w-5 h-5 hover:bg-gray-200 rounded-md')}
+          style={{ marginLeft: current.deepth * 8 }}
+          onClick={() => handleToggle(index)}
+        />
+      )
+    }
+    if (current.parent_id === 'root' || !pagesMap[current.parent_id]) {
+      return (
+        <div></div>
+      )
+    }
+    return (
+      <div className='shrink-0 mr-1 w-5 h-5' style={{ marginLeft: current.deepth * 8 }} />
+    )
+  }
+
+  return (
+    <div
+      className={cn('group flex items-center pl-2 pr-[2px] rounded-md border border-transparent hover:bg-gray-100 cursor-pointer', previewPageId === current.page_id && s['preview-item'])}
+      style={{ ...style, top: style.top as number + 8, left: 8, right: 8, width: 'calc(100% - 16px)' }}
+    >
+      <Checkbox
+        className='shrink-0 mr-2 group-hover:border-primary-600 group-hover:border-[2px]'
+        checked={checkedIds.has(current.page_id)}
+        onCheck={() => handleCheck(index)}
+      />
+      {!searchValue && renderArrow()}
+      <NotionIcon
+        className='shrink-0 mr-1'
+        type='page'
+        src={current.page_icon}
+      />
+      <div
+        className='grow text-sm font-medium text-gray-700 truncate'
+        title={current.page_name}
+      >
+        {current.page_name}
+      </div>
+      {
+        canPreview && (
+          <div
+            className='shrink-0 hidden group-hover:flex items-center ml-1 px-2 h-6 rounded-md text-xs font-medium text-gray-500 cursor-pointer hover:bg-gray-50 hover:text-gray-700'
+            onClick={() => handlePreview(index)}>
+            {t('common.dataSource.notion.selector.preview')}
+          </div>
+        )
+      }
+      {
+        searchValue && (
+          <div
+            className='shrink-0 ml-1 max-w-[120px] text-xs text-gray-400 truncate'
+            title={breadCrumbs.join(' / ')}
+          >
+            {breadCrumbs.join(' / ')}
+          </div>
+        )
+      }
+    </div>
+  )
+}, areEqual)
+
+const PageSelector = ({
+  value,
+  searchValue,
+  pagesMap,
+  list,
+  onSelect,
+  canPreview = true,
+  previewPageId,
+  onPreview,
+}: PageSelectorProps) => {
+  const { t } = useTranslation()
+  const [prevDataList, setPrevDataList] = useState(list)
+  const [dataList, setDataList] = useState<NotionPageItem[]>([])
+  const [localPreviewPageId, setLocalPreviewPageId] = useState('')
+  if (prevDataList !== list) {
+    setPrevDataList(list)
+    setDataList(list.filter(item => item.parent_id === 'root' || !pagesMap[item.parent_id]).map((item) => {
+      return {
+        ...item,
+        expand: false,
+        deepth: 0,
+      }
+    }))
+  }
+  const searchDataList = list.filter((item) => {
+    return item.page_name.includes(searchValue)
+  }).map((item) => {
+    return {
+      ...item,
+      expand: false,
+      deepth: 0,
+    }
+  })
+  const currentDataList = searchValue ? searchDataList : dataList
+  const currentPreviewPageId = previewPageId === undefined ? localPreviewPageId : previewPageId
+
+  const listMapWithChildrenAndDescendants = useMemo(() => {
+    return list.reduce((prev: NotionPageTreeMap, next: DataSourceNotionPage) => {
+      const pageId = next.page_id
+      if (!prev[pageId])
+        prev[pageId] = { ...next, children: new Set(), descendants: new Set(), deepth: 0, ancestors: [] }
+
+      recursivePushInParentDescendants(pagesMap, prev, prev[pageId], prev[pageId])
+      return prev
+    }, {})
+  }, [list, pagesMap])
+
+  const handleToggle = (index: number) => {
+    const current = dataList[index]
+    const pageId = current.page_id
+    const currentWithChildrenAndDescendants = listMapWithChildrenAndDescendants[pageId]
+    const descendantsIds = Array.from(currentWithChildrenAndDescendants.descendants)
+    const childrenIds = Array.from(currentWithChildrenAndDescendants.children)
+    let newDataList = []
+
+    if (current.expand) {
+      current.expand = false
+
+      newDataList = [...dataList.filter(item => !descendantsIds.includes(item.page_id))]
+    }
+    else {
+      current.expand = true
+
+      newDataList = [
+        ...dataList.slice(0, index + 1),
+        ...childrenIds.map(item => ({
+          ...pagesMap[item],
+          expand: false,
+          deepth: listMapWithChildrenAndDescendants[item].deepth,
+        })),
+        ...dataList.slice(index + 1)]
+    }
+    setDataList(newDataList)
+  }
+
+  const handleCheck = (index: number) => {
+    const current = currentDataList[index]
+    const pageId = current.page_id
+    const currentWithChildrenAndDescendants = listMapWithChildrenAndDescendants[pageId]
+
+    if (value.has(pageId)) {
+      if (!searchValue) {
+        for (const item of currentWithChildrenAndDescendants.descendants)
+          value.delete(item)
+      }
+
+      value.delete(pageId)
+    }
+    else {
+      if (!searchValue) {
+        for (const item of currentWithChildrenAndDescendants.descendants)
+          value.add(item)
+      }
+
+      value.add(pageId)
+    }
+
+    onSelect(new Set([...value]))
+  }
+
+  const handlePreview = (index: number) => {
+    const current = currentDataList[index]
+    const pageId = current.page_id
+
+    setLocalPreviewPageId(pageId)
+
+    if (onPreview)
+      onPreview(pageId)
+  }
+
+  if (!currentDataList.length) {
+    return (
+      <div className='flex items-center justify-center h-[296px] text-[13px] text-gray-500'>
+        {t('common.dataSource.notion.selector.noSearchResult')}
+      </div>
+    )
+  }
+
+  return (
+    <List
+      className='py-2'
+      height={296}
+      itemCount={currentDataList.length}
+      itemSize={28}
+      width='100%'
+      itemKey={(index, data) => data.dataList[index].page_id}
+      itemData={{
+        dataList: currentDataList,
+        handleToggle,
+        checkedIds: value,
+        handleCheck,
+        canPreview,
+        handlePreview,
+        listMapWithChildrenAndDescendants,
+        searchValue,
+        previewPageId: currentPreviewPageId,
+        pagesMap,
+      }}
+    >
+      {Item}
+    </List>
+  )
+}
+
+export default PageSelector

web/app/components/base/notion-page-selector/search-input/index.module.css

+.search-icon {
+  background: url(../assets/search.svg) center center;
+  background-size: 14px 14px;
+}
+
+.clear-icon {
+  background: url(../assets/clear.svg) center center;
+  background-size: contain;
+}
+
+.input-wrapper {
+  flex-basis: 200px;
+  width: 0;
+  box-shadow: 0px 1px 2px rgba(16, 24, 40, 0.05);
+}
\ No newline at end of file

web/app/components/base/notion-page-selector/search-input/index.tsx

+import { useCallback } from 'react'
+import type { ChangeEvent } from 'react'
+import { useTranslation } from 'react-i18next'
+import cn from 'classnames'
+import s from './index.module.css'
+
+type SearchInputProps = {
+  value: string
+  onChange: (v: string) => void
+}
+const SearchInput = ({
+  value,
+  onChange,
+}: SearchInputProps) => {
+  const { t } = useTranslation()
+
+  const handleClear = useCallback(() => {
+    onChange('')
+  }, [onChange])
+
+  return (
+    <div className={cn(s['input-wrapper'], 'flex items-center px-2 h-7 rounded-md', `${value ? 'bg-white' : 'bg-gray-100'}`)}>
+      <div className={cn(s['search-icon'], 'mr-[6px] w-4 h-4')} />
+      <input
+        className='grow text-[13px] bg-inherit border-0 outline-0 appearance-none'
+        value={value}
+        onChange={(e: ChangeEvent<HTMLInputElement>) => onChange(e.target.value)}
+        placeholder={t('common.dataSource.notion.selector.searchPages') || ''}
+      />
+      {
+        value && (
+          <div
+            className={cn(s['clear-icon'], 'ml-1 w-4 h-4 cursor-pointer')}
+            onClick={handleClear}
+          />
+        )
+      }
+    </div>
+  )
+}
+
+export default SearchInput

web/app/components/base/notion-page-selector/workspace-selector/index.module.css

+.down-arrow {
+  background: url(../assets/down-arrow.svg) center center no-repeat;
+  background-size: cover;
+}
+
+.popup {
+  box-shadow: 0px 12px 16px -4px rgba(16, 24, 40, 0.08), 0px 4px 6px -2px rgba(16, 24, 40, 0.03);
+  z-index: 10;
+}
\ No newline at end of file

web/app/components/base/notion-page-selector/workspace-selector/index.tsx

+'use client'
+import { useTranslation } from 'react-i18next'
+import { Fragment } from 'react'
+import { Menu, Transition } from '@headlessui/react'
+import cn from 'classnames'
+import NotionIcon from '../../notion-icon'
+import s from './index.module.css'
+import type { DataSourceNotionWorkspace } from '@/models/common'
+
+type WorkspaceSelectorProps = {
+  value: string
+  items: Omit<DataSourceNotionWorkspace, 'total'>[]
+  onSelect: (v: string) => void
+}
+export default function WorkspaceSelector({
+  value,
+  items,
+  onSelect,
+}: WorkspaceSelectorProps) {
+  const { t } = useTranslation()
+  const currentWorkspace = items.find(item => item.workspace_id === value)
+
+  return (
+    <Menu as="div" className="relative inline-block text-left">
+      {
+        ({ open }) => (
+          <>
+            <Menu.Button className={`flex items-center justify-center h-7 rounded-md hover:bg-gray-50 ${open && 'bg-gray-50'} cursor-pointer`}>
+              <NotionIcon
+                className='ml-1 mr-2'
+                src={currentWorkspace?.workspace_icon}
+                name={currentWorkspace?.workspace_name}
+              />
+              <div className='mr-1 w-[90px] text-left text-sm font-medium text-gray-700 truncate' title={currentWorkspace?.workspace_name}>{currentWorkspace?.workspace_name}</div>
+              <div className='mr-1 px-1 h-[18px] bg-primary-50 rounded-lg text-xs font-medium text-primary-600'>{currentWorkspace?.pages.length}</div>
+              <div className={cn(s['down-arrow'], 'mr-2 w-3 h-3')} />
+            </Menu.Button>
+            <Transition
+              as={Fragment}
+              enter="transition ease-out duration-100"
+              enterFrom="transform opacity-0 scale-95"
+              enterTo="transform opacity-100 scale-100"
+              leave="transition ease-in duration-75"
+              leaveFrom="transform opacity-100 scale-100"
+              leaveTo="transform opacity-0 scale-95"
+            >
+              <Menu.Items
+                className={cn(
+                  s.popup,
+                  `absolute left-0 top-8 w-80
+                  origin-top-right rounded-lg bg-white
+                  border-[0.5px] border-gray-200`,
+                )}
+              >
+                <div className="p-1 max-h-50 overflow-auto">
+                  {
+                    items.map(item => (
+                      <Menu.Item key={item.workspace_id}>
+                        <div
+                          className='flex items-center px-3 h-9 hover:bg-gray-50 cursor-pointer'
+                          onClick={() => onSelect(item.workspace_id)}
+                        >
+                          <NotionIcon
+                            className='shrink-0 mr-2'
+                            src={item.workspace_icon}
+                            name={item.workspace_name}
+                          />
+                          <div className='grow mr-2 text-sm text-gray-700 truncate' title={item.workspace_name}>{item.workspace_name}</div>
+                          <div className='shrink-0 text-xs font-medium text-primary-600'>
+                            {item.pages.length} {t('common.dataSource.notion.selector.pageSelected')}
+                          </div>
+                        </div>
+                      </Menu.Item>
+                    ))
+                  }
+                </div>
+              </Menu.Items>
+            </Transition>
+          </>
+        )
+      }
+    </Menu>
+  )
+}

web/app/components/base/progress-bar/index.tsx

+type ProgressBarProps = {
+  percent: number
+}
+const ProgressBar = ({
+  percent = 0,
+}: ProgressBarProps) => {
+  return (
+    <div className='flex items-center'>
+      <div className='mr-2 w-[100px] bg-gray-100 rounded-lg'>
+        <div
+          className='h-1 bg-[#2970FF] rounded-lg'
+          style={{ width: `${percent}%` }}
+        />
+      </div>
+      <div className='text-xs font-medium text-gray-500'>{percent}%</div>
+    </div>
+  )
+}
+
+export default ProgressBar

web/app/components/datasets/create/assets/Icon-3-dots.svg

+<svg width="16" height="16" viewBox="0 0 16 16" fill="none" xmlns="http://www.w3.org/2000/svg">
+<path d="M5 6.5V5M8.93934 7.56066L10 6.5M10.0103 11.5H11.5103" stroke="#374151" stroke-width="2" stroke-linecap="round" stroke-linejoin="round"/>
+</svg>

web/app/components/datasets/create/assets/normal.svg

+<svg width="12" height="12" viewBox="0 0 12 12" fill="none" xmlns="http://www.w3.org/2000/svg">
+<path d="M5.75 4.5C6.7165 4.5 7.5 3.7165 7.5 2.75C7.5 1.7835 6.7165 1 5.75 1C4.7835 1 4 1.7835 4 2.75C4 3.7165 4.7835 4.5 5.75 4.5Z" fill="#444CE7"/>
+<path d="M3.48775 4.314C3.36842 4.14172 3.30875 4.05558 3.24448 4.02712C3.18679 4.00157 3.12605 3.99844 3.06603 4.01794C2.99918 4.03965 2.94661 4.10099 2.84146 4.22367C2.41951 4.71598 2.13172 5.32705 2.03543 6.00009H2C1.72386 6.00009 1.5 5.77623 1.5 5.50009C1.5 5.31565 1.59961 5.15388 1.75036 5.06668C1.98939 4.9284 2.07107 4.62254 1.9328 4.38351C1.79453 4.14448 1.48867 4.0628 1.24964 4.20107C0.802591 4.45967 0.5 4.94425 0.5 5.50009C0.5 6.32852 1.17157 7.00009 2 7.00009H2.03545C2.14342 7.75422 2.49192 8.43113 2.99997 8.94961L2.99997 10.1117C2.99994 10.1712 2.99992 10.2424 3.00504 10.305C3.01097 10.3776 3.02619 10.4816 3.08171 10.5906C3.15362 10.7317 3.26835 10.8465 3.40948 10.9184C3.51845 10.9739 3.62245 10.9891 3.69505 10.9951C3.7577 11.0002 3.82881 11.0001 3.88836 11.0001H4.86154C4.92109 11.0001 4.99224 11.0002 5.05488 10.9951C5.12749 10.9891 5.23149 10.9739 5.34046 10.9184C5.48158 10.8465 5.59632 10.7317 5.66822 10.5906C5.72375 10.4816 5.73897 10.3776 5.7449 10.305C5.75002 10.2424 5.75 10.1712 5.74997 10.1117L5.74997 10.0001H6.24998L6.24997 10.1115C6.24995 10.1711 6.24992 10.2422 6.25504 10.3048C6.26097 10.3775 6.2762 10.4815 6.33172 10.5904C6.40363 10.7315 6.51836 10.8463 6.65948 10.9182C6.76846 10.9737 6.87245 10.9889 6.94506 10.9949C7.0077 11 7.0788 11 7.13835 10.9999H8.11159C8.17113 11 8.24229 11 8.30493 10.9949C8.37753 10.9889 8.48153 10.9737 8.5905 10.9182C8.73162 10.8463 8.84636 10.7315 8.91827 10.5904C8.97379 10.4815 8.98901 10.3775 8.99494 10.3048C9.00006 10.2422 9.00004 10.1711 9.00001 10.1115L9.00001 9.66299C9.55312 9.40029 10.0258 8.99721 10.3726 8.49993L10.6116 8.49994C10.6711 8.49996 10.7423 8.49999 10.8049 8.49487C10.8775 8.48893 10.9815 8.47371 11.0905 8.41819C11.2316 8.34628 11.3464 8.23155 11.4183 8.09043C11.4738 7.98145 11.489 7.87746 11.4949 7.80485C11.5001 7.74221 11.5 7.67109 11.5 7.61154V5.88181C11.5 5.82509 11.5001 5.75735 11.4954 5.69761C11.49 5.62851 11.4763 5.5294 11.4257 5.42448C11.352 5.27143 11.2285 5.14794 11.0755 5.07422C10.9705 5.02369 10.8714 5.00992 10.8023 5.00454C10.7577 5.00106 10.7087 5.0002 10.6631 4.99999C10.4953 4.64662 10.2702 4.32616 10 4.05044L10 3.51615C10 3.43874 10.0001 3.35111 9.99335 3.27574C9.98593 3.19252 9.96656 3.06385 9.88754 2.93633C9.78902 2.77733 9.63465 2.66089 9.4547 2.60984C9.31038 2.56889 9.18134 2.58561 9.09929 2.60134C9.02497 2.61559 8.94073 2.63969 8.8663 2.66098L8.78839 2.68324C8.6859 2.71252 8.63465 2.72716 8.59861 2.75356C8.5638 2.77904 8.54252 2.80415 8.52309 2.84266C8.50297 2.88255 8.49603 2.94339 8.48215 3.06506C8.32585 4.43546 7.16224 5.5 5.75 5.5C4.81225 5.5 3.98413 5.03063 3.48775 4.314Z" fill="#444CE7"/>
+</svg>

web/app/components/datasets/create/assets/star.svg

+<svg width="12" height="12" viewBox="0 0 12 12" fill="none" xmlns="http://www.w3.org/2000/svg">
+<path d="M6 0.5C6.27614 0.5 6.5 0.723858 6.5 1V2C6.5 2.27614 6.27614 2.5 6 2.5C5.72386 2.5 5.5 2.27614 5.5 2V1C5.5 0.723858 5.72386 0.5 6 0.5Z" fill="#FB6514"/>
+<path d="M2.81791 2.11092C2.62265 1.91566 2.30606 1.91566 2.1108 2.11092C1.91554 2.30619 1.91554 2.62277 2.1108 2.81803L2.81791 3.52514C3.01317 3.7204 3.32975 3.7204 3.52502 3.52514C3.72028 3.32988 3.72028 3.01329 3.52502 2.81803L2.81791 2.11092Z" fill="#FB6514"/>
+<path d="M0.5 6C0.5 5.72386 0.723858 5.5 1 5.5H2C2.27614 5.5 2.5 5.72386 2.5 6C2.5 6.27614 2.27614 6.5 2 6.5H1C0.723858 6.5 0.5 6.27614 0.5 6Z" fill="#FB6514"/>
+<path d="M10 5.5C9.72386 5.5 9.5 5.72386 9.5 6C9.5 6.27614 9.72386 6.5 10 6.5H11C11.2761 6.5 11.5 6.27614 11.5 6C11.5 5.72386 11.2761 5.5 11 5.5H10Z" fill="#FB6514"/>
+<path d="M9.18192 8.47482C8.98666 8.27955 8.67008 8.27955 8.47482 8.47482C8.27955 8.67008 8.27955 8.98666 8.47482 9.18192L9.18192 9.88903C9.37718 10.0843 9.69377 10.0843 9.88903 9.88903C10.0843 9.69377 10.0843 9.37718 9.88903 9.18192L9.18192 8.47482Z" fill="#FB6514"/>
+<path d="M9.88903 2.81803C10.0843 2.62277 10.0843 2.30619 9.88903 2.11092C9.69377 1.91566 9.37718 1.91566 9.18192 2.11092L8.47482 2.81803C8.27955 3.01329 8.27955 3.32988 8.47482 3.52514C8.67008 3.7204 8.98666 3.7204 9.18192 3.52514L9.88903 2.81803Z" fill="#FB6514"/>
+<path d="M6 9.5C6.27614 9.5 6.5 9.72386 6.5 10V11C6.5 11.2761 6.27614 11.5 6 11.5C5.72386 11.5 5.5 11.2761 5.5 11V10C5.5 9.72386 5.72386 9.5 6 9.5Z" fill="#FB6514"/>
+<path d="M3.52502 9.18192C3.72028 8.98666 3.72028 8.67008 3.52502 8.47482C3.32975 8.27955 3.01317 8.27955 2.81791 8.47482L2.1108 9.18192C1.91554 9.37718 1.91554 9.69377 2.1108 9.88903C2.30606 10.0843 2.62265 10.0843 2.81791 9.88903L3.52502 9.18192Z" fill="#FB6514"/>
+<path d="M6.44837 3.27869C6.36413 3.10804 6.19032 3 6.00001 3C5.8097 3 5.6359 3.10804 5.55166 3.27869L4.89538 4.60823L3.4277 4.82276C3.23942 4.85028 3.08308 4.98228 3.02439 5.16328C2.9657 5.34429 3.01484 5.54291 3.15115 5.67568L4.21275 6.70968L3.96221 8.17048C3.93004 8.35807 4.00716 8.54766 4.16115 8.65953C4.31514 8.77139 4.51928 8.78613 4.68774 8.69754L6.00001 8.00742L7.31229 8.69754C7.48075 8.78613 7.68489 8.77139 7.83888 8.65953C7.99287 8.54766 8.06999 8.35807 8.03782 8.17048L7.78728 6.70968L8.84888 5.67568C8.98519 5.54291 9.03433 5.34429 8.97564 5.16328C8.91695 4.98228 8.76061 4.85028 8.57233 4.82276L7.10465 4.60823L6.44837 3.27869Z" fill="#FB6514"/>
+</svg>

web/app/components/datasets/create/embedding-process/index.module.css

+.progressContainer {
+  @apply relative pb-4 w-full;
+  border-bottom: 0.5px solid #EAECF0;
+}
+.sourceItem {
+  position: relative;
+  display: flex;
+  justify-content: space-between;
+  align-items: center;
+  margin-bottom: 4px;
+  padding: 0 4px;
+  height: 24px;
+  background: #EFF4FF;
+  border-radius: 6px;
+  overflow: hidden;
+}
+.sourceItem.error {
+  background: #FEE4E2;
+}
+.sourceItem.success {
+  background: #D1FADF;
+}
+.progressbar {
+  position: absolute;
+  top: 0;
+  left: 0;
+  height: 100%;
+  background-color: #B2CCFF;
+}
+.sourceItem .info {
+  display: flex;
+  align-items: center;
+}
+.sourceItem .info .name {
+  font-weight: 500;
+  font-size: 12px;
+  line-height: 18px;
+  color: #101828;
+}
+.sourceItem.success .info .name {
+  color: #05603A;
+}
+.sourceItem .percent {
+  font-weight: 500;
+  font-size: 12px;
+  line-height: 18px;
+  color: #344054;
+}
+.sourceItem .error {
+  color: #D92D20;
+}
+.sourceItem .success {
+  color: #05603A;
+}
+
+
+.cost {
+  @apply flex justify-between items-center text-xs text-gray-700;
+}
+.embeddingStatus {
+  @apply flex items-center justify-between text-gray-900 font-medium text-sm mr-2;
+}
+.commonIcon {
+  @apply w-3 h-3 mr-1 inline-block align-middle;
+}
+.highIcon {
+  mask-image: url(../assets/star.svg);
+  @apply bg-orange-500;
+}
+.economyIcon {
+  background-color: #444ce7;
+  mask-image: url(../assets/normal.svg);
+}
+.tokens {
+  @apply text-xs font-medium px-1;
+}
+.price {
+  color: #f79009;
+  @apply text-xs font-medium;
+}
+
+.fileIcon {
+  @apply w-4 h-4 mr-1 bg-center bg-no-repeat;
+  background-image: url(../assets/unknow.svg);
+  background-size: 16px;
+}
+.fileIcon.csv {
+  background-image: url(../assets/csv.svg);
+}
+
+.fileIcon.xlsx,
+.fileIcon.xls {
+  background-image: url(../assets/xlsx.svg);
+}
+.fileIcon.pdf {
+  background-image: url(../assets/pdf.svg);
+}
+.fileIcon.html,
+.fileIcon.htm {
+  background-image: url(../assets/html.svg);
+}
+.fileIcon.md,
+.fileIcon.markdown {
+  background-image: url(../assets/md.svg);
+}
+.fileIcon.txt {
+  background-image: url(../assets/txt.svg);
+}
+.fileIcon.json {
+  background-image: url(../assets/json.svg);
+}

web/app/components/datasets/create/embedding-process/index.tsx

+import type { FC } from 'react'
+import React, { useCallback, useEffect, useMemo } from 'react'
+import useSWR from 'swr'
+import { useRouter } from 'next/navigation'
+import { useTranslation } from 'react-i18next'
+import { omit } from 'lodash-es'
+import { ArrowRightIcon } from '@heroicons/react/24/solid'
+import { useGetState } from 'ahooks'
+import cn from 'classnames'
+import s from './index.module.css'
+import { FieldInfo } from '@/app/components/datasets/documents/detail/metadata'
+import Button from '@/app/components/base/button'
+import type { FullDocumentDetail, IndexingStatusResponse, ProcessRuleResponse } from '@/models/datasets'
+import { formatNumber } from '@/utils/format'
+import { fetchIndexingStatusBatch as doFetchIndexingStatus, fetchIndexingEstimateBatch, fetchProcessRule } from '@/service/datasets'
+import { DataSourceType } from '@/models/datasets'
+import NotionIcon from '@/app/components/base/notion-icon'
+
+type Props = {
+  datasetId: string
+  batchId: string
+  documents?: FullDocumentDetail[]
+  indexingType?: string
+}
+
+const RuleDetail: FC<{ sourceData?: ProcessRuleResponse }> = ({ sourceData }) => {
+  const { t } = useTranslation()
+
+  const segmentationRuleMap = {
+    mode: t('datasetDocuments.embedding.mode'),
+    segmentLength: t('datasetDocuments.embedding.segmentLength'),
+    textCleaning: t('datasetDocuments.embedding.textCleaning'),
+  }
+
+  const getRuleName = (key: string) => {
+    if (key === 'remove_extra_spaces')
+      return t('datasetCreation.stepTwo.removeExtraSpaces')
+
+    if (key === 'remove_urls_emails')
+      return t('datasetCreation.stepTwo.removeUrlEmails')
+
+    if (key === 'remove_stopwords')
+      return t('datasetCreation.stepTwo.removeStopwords')
+  }
+
+  const getValue = useCallback((field: string) => {
+    let value: string | number | undefined = '-'
+    switch (field) {
+      case 'mode':
+        value = sourceData?.mode === 'automatic' ? (t('datasetDocuments.embedding.automatic') as string) : (t('datasetDocuments.embedding.custom') as string)
+        break
+      case 'segmentLength':
+        value = sourceData?.rules?.segmentation?.max_tokens
+        break
+      default:
+        value = sourceData?.mode === 'automatic'
+          ? (t('datasetDocuments.embedding.automatic') as string)
+          // eslint-disable-next-line array-callback-return
+          : sourceData?.rules?.pre_processing_rules?.map((rule) => {
+            if (rule.enabled)
+              return getRuleName(rule.id)
+          }).filter(Boolean).join(';')
+        break
+    }
+    return value
+  }, [sourceData])
+
+  return <div className='flex flex-col pt-8 pb-10 first:mt-0'>
+    {Object.keys(segmentationRuleMap).map((field) => {
+      return <FieldInfo
+        key={field}
+        label={segmentationRuleMap[field as keyof typeof segmentationRuleMap]}
+        displayedValue={String(getValue(field))}
+      />
+    })}
+  </div>
+}
+
+const EmbeddingProcess: FC<Props> = ({ datasetId, batchId, documents = [], indexingType }) => {
+  const { t } = useTranslation()
+
+  const getFirstDocument = documents[0]
+
+  const [indexingStatusBatchDetail, setIndexingStatusDetail, getIndexingStatusDetail] = useGetState<IndexingStatusResponse[]>([])
+  const fetchIndexingStatus = async () => {
+    const status = await doFetchIndexingStatus({ datasetId, batchId })
+    setIndexingStatusDetail(status.data)
+  }
+
+  const [runId, setRunId, getRunId] = useGetState<any>(null)
+
+  const stopQueryStatus = () => {
+    clearInterval(getRunId())
+  }
+
+  const startQueryStatus = () => {
+    const runId = setInterval(() => {
+      const indexingStatusBatchDetail = getIndexingStatusDetail()
+      const isCompleted = indexingStatusBatchDetail.every(indexingStatusDetail => ['completed', 'error'].includes(indexingStatusDetail.indexing_status))
+      if (isCompleted) {
+        stopQueryStatus()
+        return
+      }
+      fetchIndexingStatus()
+    }, 2500)
+    setRunId(runId)
+  }
+
+  useEffect(() => {
+    fetchIndexingStatus()
+    startQueryStatus()
+    return () => {
+      stopQueryStatus()
+    }
+  }, [])
+
+  // get rule
+  const { data: ruleDetail, error: ruleError } = useSWR({
+    action: 'fetchProcessRule',
+    params: { documentId: getFirstDocument.id },
+  }, apiParams => fetchProcessRule(omit(apiParams, 'action')), {
+    revalidateOnFocus: false,
+  })
+  // get cost
+  const { data: indexingEstimateDetail, error: indexingEstimateErr } = useSWR({
+    action: 'fetchIndexingEstimateBatch',
+    datasetId,
+    batchId,
+  }, apiParams => fetchIndexingEstimateBatch(omit(apiParams, 'action')), {
+    revalidateOnFocus: false,
+  })
+
+  const router = useRouter()
+  const navToDocumentList = () => {
+    router.push(`/datasets/${datasetId}/documents`)
+  }
+
+  const isEmbedding = useMemo(() => {
+    return indexingStatusBatchDetail.some((indexingStatusDetail: { indexing_status: any }) => ['indexing', 'splitting', 'parsing', 'cleaning'].includes(indexingStatusDetail?.indexing_status || ''))
+  }, [indexingStatusBatchDetail])
+  const isEmbeddingCompleted = useMemo(() => {
+    return indexingStatusBatchDetail.every((indexingStatusDetail: { indexing_status: any }) => ['completed', 'error'].includes(indexingStatusDetail?.indexing_status || ''))
+  }, [indexingStatusBatchDetail])
+
+  const getSourceName = (id: string) => {
+    const doc = documents.find(document => document.id === id)
+    return doc?.name
+  }
+  const getFileType = (name?: string) => name?.split('.').pop() || 'txt'
+  const getSourcePercent = (detail: IndexingStatusResponse) => {
+    const completedCount = detail.completed_segments || 0
+    const totalCount = detail.total_segments || 0
+    if (totalCount === 0)
+      return 0
+    const percent = Math.round(completedCount * 100 / totalCount)
+    return percent > 100 ? 100 : percent
+  }
+  const getSourceType = (id: string) => {
+    const doc = documents.find(document => document.id === id)
+    return doc?.data_source_type as DataSourceType
+  }
+  const getIcon = (id: string) => {
+    const doc = documents.find(document => document.id === id) as any // TODO type fix
+
+    return doc.data_source_info.notion_page_icon
+  }
+  const isSourceEmbedding = (detail: IndexingStatusResponse) => ['indexing', 'splitting', 'parsing', 'cleaning', 'waiting'].includes(detail.indexing_status || '')
+
+  return (
+    <>
+      <div className='h-5 flex justify-between items-center mb-5'>
+        <div className={s.embeddingStatus}>
+          {isEmbedding && t('datasetDocuments.embedding.processing')}
+          {isEmbeddingCompleted && t('datasetDocuments.embedding.completed')}
+        </div>
+        <div className={s.cost}>
+          {indexingType === 'high_quaility' && (
+            <div className='flex items-center'>
+              <div className={cn(s.commonIcon, s.highIcon)} />
+              {t('datasetDocuments.embedding.highQuality')} · {t('datasetDocuments.embedding.estimate')}
+              <span className={s.tokens}>{formatNumber(indexingEstimateDetail?.tokens || 0)}</span>tokens
+              (<span className={s.price}>${formatNumber(indexingEstimateDetail?.total_price || 0)}</span>)
+            </div>
+          )}
+          {indexingType === 'economy' && (
+            <div className='flex items-center'>
+              <div className={cn(s.commonIcon, s.economyIcon)} />
+              {t('datasetDocuments.embedding.economy')} · {t('datasetDocuments.embedding.estimate')}
+              <span className={s.tokens}>0</span>tokens
+            </div>
+          )}
+        </div>
+      </div>
+      <div className={s.progressContainer}>
+        {indexingStatusBatchDetail.map(indexingStatusDetail => (
+          <div className={cn(
+            s.sourceItem,
+            indexingStatusDetail.indexing_status === 'error' && s.error,
+            indexingStatusDetail.indexing_status === 'completed' && s.success,
+          )}>
+            {isSourceEmbedding(indexingStatusDetail) && (
+              <div className={s.progressbar} style={{ width: `${getSourcePercent(indexingStatusDetail)}%` }}/>
+            )}
+            <div className={s.info}>
+              {getSourceType(indexingStatusDetail.id) === DataSourceType.FILE && (
+                <div className={cn(s.fileIcon, s[getFileType(getSourceName(indexingStatusDetail.id))])}/>
+              )}
+              {getSourceType(indexingStatusDetail.id) === DataSourceType.NOTION && (
+                <NotionIcon
+                  className='shrink-0 mr-1'
+                  type='page'
+                  src={getIcon(indexingStatusDetail.id)}
+                />
+              )}
+              <div className={s.name}>{getSourceName(indexingStatusDetail.id)}</div>
+            </div>
+            <div className='shrink-0'>
+              {isSourceEmbedding(indexingStatusDetail) && (
+                <div className={s.percent}>{`${getSourcePercent(indexingStatusDetail)}%`}</div>
+              )}
+              {indexingStatusDetail.indexing_status === 'error' && (
+                <div className={cn(s.percent, s.error)}>Error</div>
+              )}
+              {indexingStatusDetail.indexing_status === 'completed' && (
+                <div className={cn(s.percent, s.success)}>100%</div>
+              )}
+            </div>
+          </div>
+        ))}
+      </div>
+      <RuleDetail sourceData={ruleDetail} />
+      <div className='flex items-center gap-2 mt-10'>
+        <Button className='w-fit' type='primary' onClick={navToDocumentList}>
+          <span>{t('datasetCreation.stepThree.navTo')}</span>
+          <ArrowRightIcon className='h-4 w-4 ml-2 stroke-current stroke-1' />
+        </Button>
+      </div>
+    </>
+  )
+}
+
+export default EmbeddingProcess

web/app/components/datasets/create/file-preview/index.module.css

@@ -11,6 +11,9 @@
   }
   
   .previewHeader .title {
+    display: flex;
+    justify-content: space-between;
+    align-items: center;
     color: #101828;
     font-weight: 600;
     font-size: 18px;

web/app/components/datasets/create/file-preview/index.tsx

 'use client'
-import React, { useState, useEffect } from 'react'
+import React, { useEffect, useState } from 'react'
 import { useTranslation } from 'react-i18next'
-import type { File } from '@/models/datasets'
-import { fetchFilePreview } from '@/service/common'
-
 import cn from 'classnames'
+import { XMarkIcon } from '@heroicons/react/20/solid'
 import s from './index.module.css'
+import type { File } from '@/models/datasets'
+import { fetchFilePreview } from '@/service/common'
 
 type IProps = {
-  file?: File,
+  file?: File
+  notionPage?: any
+  hidePreview: () => void
 }
 
 const FilePreview = ({
   file,
+  hidePreview,
 }: IProps) => {
   const { t } = useTranslation()
   const [previewContent, setPreviewContent] = useState('')
@@ -28,23 +31,27 @@ const FilePreview = ({
   }
 
   const getFileName = (currentFile?: File) => {
-    if (!currentFile) {
+    if (!currentFile)
       return ''
-    }
+
     const arr = currentFile.name.split('.')
     return arr.slice(0, -1).join()
   }
 
   useEffect(() => {
-    if (file) {
+    if (file)
       getPreviewContent(file.id)
-    }
   }, [file])
 
   return (
     <div className={cn(s.filePreview)}>
       <div className={cn(s.previewHeader)}>
-        <div className={cn(s.title)}>{t('datasetCreation.stepOne.filePreview')}</div>
+        <div className={cn(s.title)}>
+          <span>{t('datasetCreation.stepOne.filePreview')}</span>
+          <div className='flex items-center justify-center w-6 h-6 cursor-pointer' onClick={hidePreview}>
+            <XMarkIcon className='h-4 w-4'></XMarkIcon>
+          </div>
+        </div>
         <div className={cn(s.fileName)}>
           <span>{getFileName(file)}</span><span className={cn(s.filetype)}>.{file?.extension}</span>
         </div>

web/app/components/datasets/create/index.tsx

 'use client'
-import React, { useState, useCallback, useEffect } from 'react'
+import React, { useCallback, useEffect, useState } from 'react'
 import { useTranslation } from 'react-i18next'
 import { useBoolean } from 'ahooks'
-import type { DataSet, File, createDocumentResponse } from '@/models/datasets'
-import { fetchTenantInfo } from '@/service/common'
-import { fetchDataDetail } from '@/service/datasets'
-
+import AppUnavailable from '../../base/app-unavailable'
 import StepsNavBar from './steps-nav-bar'
 import StepOne from './step-one'
 import StepTwo from './step-two'
 import StepThree from './step-three'
+import { DataSourceType } from '@/models/datasets'
+import type { DataSet, File, createDocumentResponse } from '@/models/datasets'
+import { fetchDataSource, fetchTenantInfo } from '@/service/common'
+import { fetchDataDetail } from '@/service/datasets'
+import type { DataSourceNotionPage } from '@/models/common'
+
 import AccountSetting from '@/app/components/header/account-setting'
-import AppUnavailable from '../../base/app-unavailable'
+
+type Page = DataSourceNotionPage & { workspace_id: string }
 
 type DatasetUpdateFormProps = {
-  datasetId?: string;
+  datasetId?: string
 }
 
 const DatasetUpdateForm = ({ datasetId }: DatasetUpdateFormProps) => {
   const { t } = useTranslation()
   const [hasSetAPIKEY, setHasSetAPIKEY] = useState(true)
   const [isShowSetAPIKey, { setTrue: showSetAPIKey, setFalse: hideSetAPIkey }] = useBoolean()
+  const [hasConnection, setHasConnection] = useState(true)
+  const [isShowDataSourceSetting, { setTrue: showDataSourceSetting, setFalse: hideDataSourceSetting }] = useBoolean()
+  const [dataSourceType, setDataSourceType] = useState<DataSourceType>(DataSourceType.FILE)
   const [step, setStep] = useState(1)
   const [indexingTypeCache, setIndexTypeCache] = useState('')
   const [file, setFile] = useState<File | undefined>()
   const [result, setResult] = useState<createDocumentResponse | undefined>()
   const [hasError, setHasError] = useState(false)
 
+  const [notionPages, setNotionPages] = useState<Page[]>([])
+  const updateNotionPages = (value: Page[]) => {
+    setNotionPages(value)
+  }
+
   const updateFile = (file?: File) => {
     setFile(file)
   }
@@ -50,9 +62,15 @@ const DatasetUpdateForm = ({ datasetId }: DatasetUpdateFormProps) => {
     const hasSetKey = data.providers.some(({ is_valid }) => is_valid)
     setHasSetAPIKEY(hasSetKey)
   }
+  const checkNotionConnection = async () => {
+    const { data } = await fetchDataSource({ url: '/data-source/integrates' })
+    const hasConnection = data.filter(item => item.provider === 'notion') || []
+    setHasConnection(hasConnection.length > 0)
+  }
 
   useEffect(() => {
     checkAPIKey()
+    checkNotionConnection()
   }, [])
 
   const [detail, setDetail] = useState<DataSet | null>(null)
@@ -62,16 +80,16 @@ const DatasetUpdateForm = ({ datasetId }: DatasetUpdateFormProps) => {
         try {
           const detail = await fetchDataDetail(datasetId)
           setDetail(detail)
-        } catch (e) {
+        }
+        catch (e) {
           setHasError(true)
         }
       }
     })()
   }, [datasetId])
 
-  if (hasError) {
+  if (hasError)
     return <AppUnavailable code={500} unknownReason={t('datasetCreation.error.unavailable') as string} />
-  }
 
   return (
     <div className='flex' style={{ height: 'calc(100vh - 56px)' }}>
@@ -80,9 +98,16 @@ const DatasetUpdateForm = ({ datasetId }: DatasetUpdateFormProps) => {
       </div>
       <div className="grow bg-white">
         {step === 1 && <StepOne
+          hasConnection={hasConnection}
+          onSetting={showDataSourceSetting}
           datasetId={datasetId}
+          dataSourceType={dataSourceType}
+          dataSourceTypeDisable={!!detail?.data_source_type}
+          changeType={setDataSourceType}
           file={file}
           updateFile={updateFile}
+          notionPages={notionPages}
+          updateNotionPages={updateNotionPages}
           onStepChange={nextStep}
         />}
         {(step === 2 && (!datasetId || (datasetId && !!detail))) && <StepTwo
@@ -90,7 +115,9 @@ const DatasetUpdateForm = ({ datasetId }: DatasetUpdateFormProps) => {
           onSetting={showSetAPIKey}
           indexingType={detail?.indexing_technique || ''}
           datasetId={datasetId}
+          dataSourceType={dataSourceType}
           file={file}
+          notionPages={notionPages}
           onStepChange={changeStep}
           updateIndexingTypeCache={updateIndexingTypeCache}
           updateResultCache={updateResultCache}
@@ -106,6 +133,7 @@ const DatasetUpdateForm = ({ datasetId }: DatasetUpdateFormProps) => {
         await checkAPIKey()
         hideSetAPIkey()
       }} />}
+      {isShowDataSourceSetting && <AccountSetting activeTab="data-source" onCancel={hideDataSourceSetting}/>}
     </div>
   )
 }

web/app/components/datasets/create/notion-page-preview/index.module.css

+.filePreview {
+    @apply flex flex-col border-l border-gray-200 shrink-0;
+    width: 528px;
+    background-color: #fcfcfd;
+  }
+  
+  .previewHeader {
+    @apply border-b border-gray-200 shrink-0;
+    margin: 42px 32px 0;
+    padding-bottom: 16px;
+  }
+  
+  .previewHeader .title {
+    display: flex;
+    justify-content: space-between;
+    align-items: center;
+    color: #101828;
+    font-weight: 600;
+    font-size: 18px;
+    line-height: 28px;
+  }
+  
+  .previewHeader .fileName {
+    display: flex;
+    align-items: center;
+    font-weight: 400;
+    font-size: 12px;
+    line-height: 18px;
+    color: #1D2939;
+  }
+  
+  .previewHeader .filetype {
+    color: #667085;
+  }
+  
+  .previewContent {
+    @apply overflow-y-auto grow;
+    padding: 20px 32px;
+    font-weight: 400;
+    font-size: 16px;
+    line-height: 24px;
+    color: #344054;
+  }
+  
+  .previewContent .loading {
+    width: 100%;
+    height: 180px;
+    background: #f9fafb center no-repeat url(../assets/Loading.svg);
+    background-size: contain;
+  }
+  .fileContent {
+    white-space: pre-line;
+  }
+  
\ No newline at end of file

web/app/components/datasets/create/notion-page-preview/index.tsx

+'use client'
+import React, { useEffect, useState } from 'react'
+import { useTranslation } from 'react-i18next'
+import cn from 'classnames'
+import { XMarkIcon } from '@heroicons/react/20/solid'
+import s from './index.module.css'
+import type { DataSourceNotionPage } from '@/models/common'
+import NotionIcon from '@/app/components/base/notion-icon'
+import { fetchNotionPagePreview } from '@/service/datasets'
+
+type Page = DataSourceNotionPage & { workspace_id: string }
+type IProps = {
+  currentPage?: Page
+  hidePreview: () => void
+}
+
+const NotionPagePreview = ({
+  currentPage,
+  hidePreview,
+}: IProps) => {
+  const { t } = useTranslation()
+  const [previewContent, setPreviewContent] = useState('')
+  const [loading, setLoading] = useState(true)
+
+  const getPreviewContent = async () => {
+    if (!currentPage)
+      return
+    try {
+      const res = await fetchNotionPagePreview({
+        workspaceID: currentPage.workspace_id,
+        pageID: currentPage.page_id,
+        pageType: currentPage.type,
+      })
+      setPreviewContent(res.content)
+      setLoading(false)
+    }
+    catch {}
+  }
+
+  useEffect(() => {
+    if (currentPage) {
+      setLoading(true)
+      getPreviewContent()
+    }
+  }, [currentPage])
+
+  return (
+    <div className={cn(s.filePreview)}>
+      <div className={cn(s.previewHeader)}>
+        <div className={cn(s.title)}>
+          <span>{t('datasetCreation.stepOne.pagePreview')}</span>
+          <div className='flex items-center justify-center w-6 h-6 cursor-pointer' onClick={hidePreview}>
+            <XMarkIcon className='h-4 w-4'></XMarkIcon>
+          </div>
+        </div>
+        <div className={cn(s.fileName)}>
+          <NotionIcon
+            className='shrink-0 mr-1'
+            type='page'
+            src={currentPage?.page_icon}
+          />
+          {currentPage?.page_name}
+        </div>
+      </div>
+      <div className={cn(s.previewContent)}>
+        {loading && <div className={cn(s.loading)}/>}
+        {!loading && (
+          <div className={cn(s.fileContent)}>{previewContent}</div>
+        )}
+      </div>
+    </div>
+  )
+}
+
+export default NotionPagePreview

web/app/components/datasets/create/step-one/index.module.css

@@ -107,3 +107,53 @@
   background: center no-repeat url(../assets/folder-plus.svg);
   background-size: contain;
 }
+
+.notionConnectionTip {
+  display: flex;
+  flex-direction: column;
+  align-items: flex-start;
+  padding: 24px;
+  max-width: 640px;
+  background: #F9FAFB;
+  border-radius: 16px;
+}
+
+.notionIcon {
+  display: flex;
+  padding: 12px;
+  width: 48px;
+  height: 48px;
+  background: #fff center no-repeat url(../assets/notion.svg);
+  background-size: 24px;
+  border: 0.5px solid #EAECF5;
+  box-shadow: 0px 12px 16px -4px rgba(16, 24, 40, 0.08), 0px 4px 6px -2px rgba(16, 24, 40, 0.03);
+  border-radius: 12px;
+}
+
+.notionConnectionTip .title {
+  position: relative;
+  margin: 24px 0 4px;
+  font-style: normal;
+  font-weight: 600;
+  font-size: 16px;
+  line-height: 24px;
+  color: #374151;
+}
+.notionConnectionTip .title::after {
+  content: '';
+  position: absolute;
+  top: -6px;
+  right: -12px;
+  width: 16px;
+  height: 16px;
+  background: center no-repeat url(../assets/Icon-3-dots.svg);
+  background-size: contain;
+}
+.notionConnectionTip .tip {
+  margin-bottom: 20px;
+  font-style: normal;
+  font-weight: 400;
+  font-size: 13px;
+  line-height: 18px;
+  color: #6B7280;
+}

web/app/components/datasets/create/step-one/index.tsx

 'use client'
 import React, { useState } from 'react'
 import { useTranslation } from 'react-i18next'
-import type { File } from '@/models/datasets'
+import cn from 'classnames'
 import FilePreview from '../file-preview'
 import FileUploader from '../file-uploader'
+import NotionPagePreview from '../notion-page-preview'
 import EmptyDatasetCreationModal from '../empty-dataset-creation-modal'
-import Button from '@/app/components/base/button'
-
-import cn from 'classnames'
 import s from './index.module.css'
+import type { File } from '@/models/datasets'
+import type { DataSourceNotionPage } from '@/models/common'
+import { DataSourceType } from '@/models/datasets'
+import Button from '@/app/components/base/button'
+import { NotionPageSelector } from '@/app/components/base/notion-page-selector'
 
 type IStepOneProps = {
-  datasetId?: string,
-  file?: File,
-  updateFile: (file?: File) => void,
-  onStepChange: () => void,
+  datasetId?: string
+  dataSourceType: DataSourceType
+  dataSourceTypeDisable: Boolean
+  hasConnection: boolean
+  onSetting: () => void
+  file?: File
+  updateFile: (file?: File) => void
+  notionPages?: any[]
+  updateNotionPages: (value: any[]) => void
+  onStepChange: () => void
+  changeType: (type: DataSourceType) => void
+}
+
+type Page = DataSourceNotionPage & { workspace_id: string }
+
+type NotionConnectorProps = {
+  onSetting: () => void
+}
+export const NotionConnector = ({ onSetting }: NotionConnectorProps) => {
+  const { t } = useTranslation()
+
+  return (
+    <div className={s.notionConnectionTip}>
+      <span className={s.notionIcon}/>
+      <div className={s.title}>{t('datasetCreation.stepOne.notionSyncTitle')}</div>
+      <div className={s.tip}>{t('datasetCreation.stepOne.notionSyncTip')}</div>
+      <Button className='h-8' type='primary' onClick={onSetting}>{t('datasetCreation.stepOne.connect')}</Button>
+    </div>
+  )
 }
 
 const StepOne = ({
   datasetId,
+  dataSourceType,
+  dataSourceTypeDisable,
+  changeType,
+  hasConnection,
+  onSetting,
   onStepChange,
   file,
   updateFile,
+  notionPages = [],
+  updateNotionPages,
 }: IStepOneProps) => {
-  const [dataSourceType, setDataSourceType] = useState('FILE')
   const [showModal, setShowModal] = useState(false)
+  const [showFilePreview, setShowFilePreview] = useState(true)
+  const [currentNotionPage, setCurrentNotionPage] = useState<Page | undefined>()
   const { t } = useTranslation()
 
+  const hidePreview = () => setShowFilePreview(false)
+
   const modalShowHandle = () => setShowModal(true)
 
   const modalCloseHandle = () => setShowModal(false)
 
+  const updateCurrentPage = (page: Page) => {
+    setCurrentNotionPage(page)
+  }
+
+  const hideNotionPagePreview = () => {
+    setCurrentNotionPage(undefined)
+  }
+
   return (
     <div className='flex w-full h-full'>
       <div className='grow overflow-y-auto relative'>
@@ -38,41 +84,76 @@ const StepOne = ({
         <div className={s.form}>
           <div className={s.dataSourceTypeList}>
             <div
-              className={cn(s.dataSourceItem, dataSourceType === 'FILE' && s.active)}
-              onClick={() => setDataSourceType('FILE')}
+              className={cn(
+                s.dataSourceItem,
+                dataSourceType === DataSourceType.FILE && s.active,
+                dataSourceTypeDisable && dataSourceType !== DataSourceType.FILE && s.disabled,
+              )}
+              onClick={() => {
+                if (dataSourceTypeDisable)
+                  return
+                changeType(DataSourceType.FILE)
+                hidePreview()
+              }}
             >
-              <span className={cn(s.datasetIcon)}/>
+              <span className={cn(s.datasetIcon)} />
               {t('datasetCreation.stepOne.dataSourceType.file')}
             </div>
             <div
-              className={cn(s.dataSourceItem, s.disabled, dataSourceType === 'notion' && s.active)}
-              // onClick={() => setDataSourceType('notion')}
+              className={cn(
+                s.dataSourceItem,
+                dataSourceType === DataSourceType.NOTION && s.active,
+                dataSourceTypeDisable && dataSourceType !== DataSourceType.NOTION && s.disabled,
+              )}
+              onClick={() => {
+                if (dataSourceTypeDisable)
+                  return
+                changeType(DataSourceType.NOTION)
+                hidePreview()
+              }}
             >
-              <span className={s.comingTag}>Coming soon</span>
-              <span className={cn(s.datasetIcon, s.notion)}/>
+              <span className={cn(s.datasetIcon, s.notion)} />
               {t('datasetCreation.stepOne.dataSourceType.notion')}
             </div>
             <div
-              className={cn(s.dataSourceItem, s.disabled, dataSourceType === 'web' && s.active)}
-              // onClick={() => setDataSourceType('web')}
+              className={cn(s.dataSourceItem, s.disabled, dataSourceType === DataSourceType.WEB && s.active)}
+            // onClick={() => changeType(DataSourceType.WEB)}
             >
               <span className={s.comingTag}>Coming soon</span>
-              <span className={cn(s.datasetIcon, s.web)}/>  
+              <span className={cn(s.datasetIcon, s.web)} />
               {t('datasetCreation.stepOne.dataSourceType.web')}
             </div>
           </div>
-          <FileUploader onFileUpdate={updateFile} file={file} />
-          <Button disabled={!file} className={s.submitButton} type='primary' onClick={onStepChange}>{t('datasetCreation.stepOne.button')}</Button>
+          {dataSourceType === DataSourceType.FILE && (
+            <>
+              <FileUploader onFileUpdate={updateFile} file={file} />
+              <Button disabled={!file} className={s.submitButton} type='primary' onClick={onStepChange}>{t('datasetCreation.stepOne.button')}</Button>
+            </>
+          )}
+          {dataSourceType === DataSourceType.NOTION && (
+            <>
+              {!hasConnection && <NotionConnector onSetting={onSetting} />}
+              {hasConnection && (
+                <>
+                  <div className='mb-8 w-[640px]'>
+                    <NotionPageSelector value={notionPages.map(page => page.page_id)} onSelect={updateNotionPages} onPreview={updateCurrentPage} />
+                  </div>
+                  <Button disabled={!notionPages.length} className={s.submitButton} type='primary' onClick={onStepChange}>{t('datasetCreation.stepOne.button')}</Button>
+                </>
+              )}
+            </>
+          )}
           {!datasetId && (
             <>
-              <div className={s.dividerLine}/>
+              <div className={s.dividerLine} />
               <div onClick={modalShowHandle} className={s.OtherCreationOption}>{t('datasetCreation.stepOne.emptyDatasetCreation')}</div>
             </>
           )}
         </div>
-        <EmptyDatasetCreationModal show={showModal} onHide={modalCloseHandle}/>
+        <EmptyDatasetCreationModal show={showModal} onHide={modalCloseHandle} />
       </div>
-      {file && <FilePreview file={file} />}
+      {file && showFilePreview && <FilePreview file={file} hidePreview={hidePreview} />}
+      {currentNotionPage && <NotionPagePreview currentPage={currentNotionPage} hidePreview={hideNotionPagePreview} />}
     </div>
   )
 }

web/app/components/datasets/create/step-three/index.tsx

 'use client'
 import React from 'react'
 import { useTranslation } from 'react-i18next'
-import type { createDocumentResponse } from '@/models/datasets'
-import EmbeddingDetail from '../../documents/detail/embedding'
-
 import cn from 'classnames'
+import EmbeddingProcess from '../embedding-process'
+
 import s from './index.module.css'
+import type { FullDocumentDetail, createDocumentResponse } from '@/models/datasets'
 
 type StepThreeProps = {
-  datasetId?: string,
-  datasetName?: string,
-  indexingType?: string,
+  datasetId?: string
+  datasetName?: string
+  indexingType?: string
   creationCache?: createDocumentResponse
 }
 
@@ -38,12 +38,11 @@ const StepThree = ({ datasetId, datasetName, indexingType, creationCache }: Step
               <div className={s.content}>{`${t('datasetCreation.stepThree.additionP1')} ${datasetName || creationCache?.dataset?.name} ${t('datasetCreation.stepThree.additionP2')}`}</div>
             </div>
           )}
-          <EmbeddingDetail
-            datasetId={datasetId || creationCache?.dataset?.id}
-            documentId={creationCache?.document.id}
+          <EmbeddingProcess
+            datasetId={datasetId || creationCache?.dataset?.id || ''}
+            batchId={creationCache?.batch || ''}
+            documents={creationCache?.documents as FullDocumentDetail[]}
             indexingType={indexingType || creationCache?.dataset?.indexing_technique}
-            stopPosition='bottom'
-            detail={creationCache?.document}
           />
         </div>
       </div>
@@ -58,4 +57,4 @@ const StepThree = ({ datasetId, datasetName, indexingType, creationCache }: Step
   )
 }
 
-export default StepThree;
+export default StepThree

web/app/components/datasets/create/step-two/index.module.css

@@ -14,9 +14,26 @@
 }
 
 .fixed {
+  padding-top: 12px;
+  font-size: 12px;
+  line-height: 18px;
   background: rgba(255, 255, 255, 0.9);
   border-bottom: 0.5px solid #EAECF0;
   backdrop-filter: blur(4px);
+  animation: fix 0.5s;
+}
+
+@keyframes fix {
+  from {
+    padding-top: 42px;
+    font-size: 18px;
+    line-height: 28px;
+  }
+  to {
+    padding-top: 12px;
+    font-size: 12px;
+    line-height: 18px;
+  }
 }
 
 .form {
@@ -273,11 +290,11 @@
   @apply bg-gray-100 caret-primary-600 hover:bg-gray-100 focus:ring-1 focus:ring-inset focus:ring-gray-200 focus-visible:outline-none focus:bg-white placeholder:text-gray-400;
 }
 
-.file {
+.source {
   @apply flex justify-between items-center mt-8 px-6 py-4 rounded-xl bg-gray-50;
 }
 
-.file .divider {
+.source .divider {
   @apply shrink-0 mx-4 w-px bg-gray-200;
   height: 42px;
 }
@@ -318,9 +335,19 @@
 .fileIcon.json {
   background-image: url(../assets/json.svg);
 }
-
-.fileContent {
-  flex: 1 1 50%;
+.sourceContent {
+  flex: 1 1 auto;
+}
+.sourceCount {
+  @apply shrink-0 ml-1;
+  font-weight: 500;
+  font-size: 13px;
+  line-height: 18px;
+  color: #667085;
+}
+.segmentCount {
+  flex: 1 1 30%;
+  max-width: 120px;
 }
 
 .divider {

web/app/components/datasets/create/step-two/index.tsx

@@ -6,9 +6,10 @@ import { useBoolean } from 'ahooks'
 import { XMarkIcon } from '@heroicons/react/20/solid'
 import cn from 'classnames'
 import Link from 'next/link'
+import { groupBy } from 'lodash-es'
 import PreviewItem from './preview-item'
 import s from './index.module.css'
-import type { CreateDocumentReq, File, FullDocumentDetail, FileIndexingEstimateResponse as IndexingEstimateResponse, PreProcessingRule, Rules, createDocumentResponse } from '@/models/datasets'
+import type { CreateDocumentReq, File, FullDocumentDetail, FileIndexingEstimateResponse as IndexingEstimateResponse, NotionInfo, PreProcessingRule, Rules, createDocumentResponse } from '@/models/datasets'
 import {
   createDocument,
   createFirstDocument,
@@ -20,6 +21,11 @@ import Loading from '@/app/components/base/loading'
 
 import Toast from '@/app/components/base/toast'
 import { formatNumber } from '@/utils/format'
+import type { DataSourceNotionPage } from '@/models/common'
+import { DataSourceType } from '@/models/datasets'
+import NotionIcon from '@/app/components/base/notion-icon'
+
+type Page = DataSourceNotionPage & { workspace_id: string }
 
 type StepTwoProps = {
   isSetting?: boolean
@@ -28,7 +34,9 @@ type StepTwoProps = {
   onSetting: () => void
   datasetId?: string
   indexingType?: string
+  dataSourceType: DataSourceType
   file?: File
+  notionPages?: Page[]
   onStepChange?: (delta: number) => void
   updateIndexingTypeCache?: (type: string) => void
   updateResultCache?: (res: createDocumentResponse) => void
@@ -52,7 +60,9 @@ const StepTwo = ({
   onSetting,
   datasetId,
   indexingType,
+  dataSourceType,
   file,
+  notionPages = [],
   onStepChange,
   updateIndexingTypeCache,
   updateResultCache,
@@ -169,12 +179,54 @@ const StepTwo = ({
     return processRule
   }
 
+  const getNotionInfo = () => {
+    const workspacesMap = groupBy(notionPages, 'workspace_id')
+    const workspaces = Object.keys(workspacesMap).map((workspaceId) => {
+      return {
+        workspaceId,
+        pages: workspacesMap[workspaceId],
+      }
+    })
+    return workspaces.map((workspace) => {
+      return {
+        workspace_id: workspace.workspaceId,
+        pages: workspace.pages.map((page) => {
+          const { page_id, page_name, page_icon, type } = page
+          return {
+            page_id,
+            page_name,
+            page_icon,
+            type,
+          }
+        }),
+      }
+    }) as NotionInfo[]
+  }
+
   const getFileIndexingEstimateParams = () => {
-    const params = {
-      file_id: file?.id,
-      dataset_id: datasetId,
-      indexing_technique: getIndexing_technique(),
-      process_rule: getProcessRule(),
+    let params
+    if (dataSourceType === DataSourceType.FILE) {
+      params = {
+        info_list: {
+          data_source_type: dataSourceType,
+          file_info_list: {
+            // TODO multi files
+            file_ids: [file?.id || ''],
+          },
+        },
+        indexing_technique: getIndexing_technique(),
+        process_rule: getProcessRule(),
+      }
+    }
+    if (dataSourceType === DataSourceType.NOTION) {
+      params = {
+        info_list: {
+          data_source_type: dataSourceType,
+          notion_info_list: getNotionInfo(),
+        },
+        indexing_technique: getIndexing_technique(),
+        process_rule: getProcessRule(),
+      }
     }
     return params
   }
@@ -190,13 +242,22 @@ const StepTwo = ({
     else {
       params = {
         data_source: {
-          type: 'upload_file',
-          info: file?.id,
-          name: file?.name,
+          type: dataSourceType,
+          info_list: {
+            data_source_type: dataSourceType,
+          },
         },
         indexing_technique: getIndexing_technique(),
         process_rule: getProcessRule(),
       } as CreateDocumentReq
+      if (dataSourceType === DataSourceType.FILE) {
+        params.data_source.info_list.file_info_list = {
+          // TODO multi files
+          file_ids: [file?.id || ''],
+        }
+      }
+      if (dataSourceType === DataSourceType.NOTION)
+        params.data_source.info_list.notion_info_list = getNotionInfo()
     }
     return params
   }
@@ -249,9 +310,7 @@ const StepTwo = ({
           body: params,
         })
         updateIndexingTypeCache && updateIndexingTypeCache(indexType)
-        updateResultCache && updateResultCache({
-          document: res,
-        })
+        updateResultCache && updateResultCache(res)
       }
       onStepChange && onStepChange(+1)
       isSetting && onSave && onSave()
@@ -319,7 +378,6 @@ const StepTwo = ({
         <div className={cn(s.form)}>
           <div className={s.label}>{t('datasetCreation.stepTwo.segmentation')}</div>
           <div className='max-w-[640px]'>
-
             <div
               className={cn(
                 s.radioItem,
@@ -467,16 +525,41 @@ const StepTwo = ({
                 <Link className='text-[#155EEF]' href={`/datasets/${datasetId}/settings`}>{t('datasetCreation.stepTwo.datasetSettingLink')}</Link>
               </div>
             )}
-            <div className={s.file}>
-              <div className={s.fileContent}>
-                <div className='mb-2 text-xs font-medium text-gray-500'>{t('datasetCreation.stepTwo.fileName')}</div>
-                <div className='flex items-center text-sm leading-6 font-medium text-gray-800'>
-                  <span className={cn(s.fileIcon, file && s[file.extension])} />
-                  {getFileName(file?.name || '')}
-                </div>
+            {/* TODO multi files */}
+            <div className={s.source}>
+              <div className={s.sourceContent}>
+                {dataSourceType === DataSourceType.FILE && (
+                  <>
+                    <div className='mb-2 text-xs font-medium text-gray-500'>{t('datasetCreation.stepTwo.fileSource')}</div>
+                    <div className='flex items-center text-sm leading-6 font-medium text-gray-800'>
+                      <span className={cn(s.fileIcon, file && s[file.extension])} />
+                      {getFileName(file?.name || '')}
+                    </div>
+                  </>
+                )}
+                {dataSourceType === DataSourceType.NOTION && (
+                  <>
+                    <div className='mb-2 text-xs font-medium text-gray-500'>{t('datasetCreation.stepTwo.notionSource')}</div>
+                    <div className='flex items-center text-sm leading-6 font-medium text-gray-800'>
+                      <NotionIcon
+                        className='shrink-0 mr-1'
+                        type='page'
+                        src={notionPages[0]?.page_icon}
+                      />
+                      {notionPages[0]?.page_name}
+                      {notionPages.length > 1 && (
+                        <span className={s.sourceCount}>
+                          <span>{t('datasetCreation.stepTwo.other')}</span>
+                          <span>{notionPages.length - 1}</span>
+                          <span>{t('datasetCreation.stepTwo.notionUnit')}</span>
+                        </span>
+                      )}
+                    </div>
+                  </>
+                )}
               </div>
               <div className={s.divider} />
-              <div className={s.fileContent}>
+              <div className={s.segmentCount}>
                 <div className='mb-2 text-xs font-medium text-gray-500'>{t('datasetCreation.stepTwo.emstimateSegment')}</div>
                 <div className='flex items-center text-sm leading-6 font-medium text-gray-800'>
                   {

web/app/components/datasets/documents/detail/index.tsx

@@ -8,15 +8,16 @@ import { useTranslation } from 'react-i18next'
 import { useRouter } from 'next/navigation'
 import { omit } from 'lodash-es'
 import cn from 'classnames'
-import Divider from '@/app/components/base/divider'
-import Loading from '@/app/components/base/loading'
-import { fetchDocumentDetail, MetadataType } from '@/service/datasets'
 import { OperationAction, StatusItem } from '../list'
+import s from '../style.module.css'
 import Completed from './completed'
 import Embedding from './embedding'
 import Metadata from './metadata'
-import s from '../style.module.css'
 import style from './style.module.css'
+import Divider from '@/app/components/base/divider'
+import Loading from '@/app/components/base/loading'
+import type { MetadataType } from '@/service/datasets'
+import { fetchDocumentDetail } from '@/service/datasets'
 
 export const BackCircleBtn: FC<{ onClick: () => void }> = ({ onClick }) => {
   return (
@@ -29,11 +30,11 @@ export const BackCircleBtn: FC<{ onClick: () => void }> = ({ onClick }) => {
 export const DocumentContext = createContext<{ datasetId?: string; documentId?: string }>({})
 
 type DocumentTitleProps = {
-  extension?: string;
-  name?: string;
-  iconCls?: string;
-  textCls?: string;
-  wrapperCls?: string;
+  extension?: string
+  name?: string
+  iconCls?: string
+  textCls?: string
+  wrapperCls?: string
 }
 
 export const DocumentTitle: FC<DocumentTitleProps> = ({ extension, name, iconCls, textCls, wrapperCls }) => {
@@ -58,15 +59,16 @@ const DocumentDetail: FC<Props> = ({ datasetId, documentId }) => {
     action: 'fetchDocumentDetail',
     datasetId,
     documentId,
-    params: { metadata: 'without' as MetadataType }
+    params: { metadata: 'without' as MetadataType },
   }, apiParams => fetchDocumentDetail(omit(apiParams, 'action')))
 
   const { data: documentMetadata, error: metadataErr, mutate: metadataMutate } = useSWR({
     action: 'fetchDocumentDetail',
     datasetId,
     documentId,
-    params: { metadata: 'only' as MetadataType }
-  }, apiParams => fetchDocumentDetail(omit(apiParams, 'action')))
+    params: { metadata: 'only' as MetadataType },
+  }, apiParams => fetchDocumentDetail(omit(apiParams, 'action')),
+  )
 
   const backToPrev = () => {
     router.push(`/datasets/${datasetId}/documents`)
@@ -77,6 +79,13 @@ const DocumentDetail: FC<Props> = ({ datasetId, documentId }) => {
 
   const embedding = ['queuing', 'indexing', 'paused'].includes((documentDetail?.display_status || '').toLowerCase())
 
+  const handleOperate = (operateName?: string) => {
+    if (operateName === 'delete')
+      backToPrev()
+    else
+      detailMutate()
+  }
+
   return (
     <DocumentContext.Provider value={{ datasetId, documentId }}>
       <div className='flex flex-col h-full'>
@@ -90,10 +99,10 @@ const DocumentDetail: FC<Props> = ({ datasetId, documentId }) => {
             detail={{
               enabled: documentDetail?.enabled || false,
               archived: documentDetail?.archived || false,
-              id: documentId
+              id: documentId,
             }}
             datasetId={datasetId}
-            onUpdate={detailMutate}
+            onUpdate={handleOperate}
             className='!w-[216px]'
           />
           <button
@@ -102,8 +111,9 @@ const DocumentDetail: FC<Props> = ({ datasetId, documentId }) => {
           />
         </div>
         <div className='flex flex-row flex-1' style={{ height: 'calc(100% - 4rem)' }}>
-          {isDetailLoading ? <Loading type='app' /> :
-            <div className={`box-border h-full w-full overflow-y-scroll ${embedding ? 'py-12 px-16' : 'pb-[30px] pt-3 px-6'}`}>
+          {isDetailLoading
+            ? <Loading type='app' />
+            : <div className={`box-border h-full w-full overflow-y-scroll ${embedding ? 'py-12 px-16' : 'pb-[30px] pt-3 px-6'}`}>
               {embedding ? <Embedding detail={documentDetail} /> : <Completed />}
             </div>
           }

web/app/components/datasets/documents/detail/settings/index.tsx

 'use client'
-import React, { useEffect, useState } from 'react'
+import React, { useEffect, useMemo, useState } from 'react'
 import { useTranslation } from 'react-i18next'
 import { useBoolean } from 'ahooks'
 import { useContext } from 'use-context-selector'
@@ -43,6 +43,15 @@ const DocumentSettings = ({ datasetId, documentId }: DocumentSettingsProps) => {
   }, [])
 
   const [documentDetail, setDocumentDetail] = useState<FullDocumentDetail | null>(null)
+  const currentPage = useMemo(() => {
+    return {
+      workspace_id: documentDetail?.data_source_info.notion_workspace_id,
+      page_id: documentDetail?.data_source_info.notion_page_id,
+      page_name: documentDetail?.name,
+      page_icon: documentDetail?.data_source_info.notion_page_icon,
+      type: documentDetail?.data_source_info.type,
+    }
+  }, [documentDetail])
   useEffect(() => {
     (async () => {
       try {
@@ -71,6 +80,8 @@ const DocumentSettings = ({ datasetId, documentId }: DocumentSettingsProps) => {
             hasSetAPIKEY={hasSetAPIKEY}
             onSetting={showSetAPIKey}
             datasetId={datasetId}
+            dataSourceType={documentDetail.data_source_type}
+            notionPages={[currentPage]}
             indexingType={indexingTechnique || ''}
             isSetting
             documentDetail={documentDetail}

web/app/components/datasets/documents/index.tsx

@@ -4,7 +4,7 @@ import React, { useMemo, useState } from 'react'
 import useSWR from 'swr'
 import { useTranslation } from 'react-i18next'
 import { useRouter } from 'next/navigation'
-import { debounce, omit } from 'lodash-es'
+import { debounce, groupBy, omit } from 'lodash-es'
 // import Link from 'next/link'
 import { PlusIcon } from '@heroicons/react/24/solid'
 import List from './list'
@@ -14,7 +14,12 @@ import Button from '@/app/components/base/button'
 import Input from '@/app/components/base/input'
 import Pagination from '@/app/components/base/pagination'
 import { get } from '@/service/base'
-import { fetchDocuments } from '@/service/datasets'
+import { createDocument, fetchDocuments } from '@/service/datasets'
+import { useDatasetDetailContext } from '@/context/dataset-detail'
+import { NotionPageSelectorModal } from '@/app/components/base/notion-page-selector'
+import type { DataSourceNotionPage } from '@/models/common'
+import type { CreateDocumentReq } from '@/models/datasets'
+import { DataSourceType } from '@/models/datasets'
 
 // Custom page count is not currently supported.
 const limit = 15
@@ -75,20 +80,63 @@ const Documents: FC<IDocumentsProps> = ({ datasetId }) => {
   const [searchValue, setSearchValue] = useState<string>('')
   const [currPage, setCurrPage] = React.useState<number>(0)
   const router = useRouter()
+  const { dataset } = useDatasetDetailContext()
+  const [notionPageSelectorModalVisible, setNotionPageSelectorModalVisible] = useState(false)
+  const [timerCanRun, setTimerCanRun] = useState(true)
+  const isDataSourceNotion = dataset?.data_source_type === DataSourceType.NOTION
 
   const query = useMemo(() => {
-    return { page: currPage + 1, limit, keyword: searchValue }
-  }, [searchValue, currPage])
-
-  const { data: documentsRes, error, mutate } = useSWR({
-    action: 'fetchDocuments',
-    datasetId,
-    params: query,
-  }, apiParams => fetchDocuments(omit(apiParams, 'action')))
+    return { page: currPage + 1, limit, keyword: searchValue, fetch: isDataSourceNotion ? true : '' }
+  }, [searchValue, currPage, isDataSourceNotion])
+
+  const { data: documentsRes, error, mutate } = useSWR(
+    {
+      action: 'fetchDocuments',
+      datasetId,
+      params: query,
+    },
+    apiParams => fetchDocuments(omit(apiParams, 'action')),
+    { refreshInterval: (isDataSourceNotion && timerCanRun) ? 2500 : 0 },
+  )
 
+  const documentsWithProgress = useMemo(() => {
+    let completedNum = 0
+    let percent = 0
+    const documentsData = documentsRes?.data?.map((documentItem) => {
+      const { indexing_status, completed_segments, total_segments } = documentItem
+      const isEmbeddinged = indexing_status === 'completed' || indexing_status === 'paused' || indexing_status === 'error'
+
+      if (isEmbeddinged)
+        completedNum++
+
+      const completedCount = completed_segments || 0
+      const totalCount = total_segments || 0
+      if (totalCount === 0 && completedCount === 0) {
+        percent = isEmbeddinged ? 100 : 0
+      }
+      else {
+        const per = Math.round(completedCount * 100 / totalCount)
+        percent = per > 100 ? 100 : per
+      }
+      return {
+        ...documentItem,
+        percent,
+      }
+    })
+    if (completedNum === documentsRes?.data?.length)
+      setTimerCanRun(false)
+    return {
+      ...documentsRes,
+      data: documentsData,
+    }
+  }, [documentsRes])
   const total = documentsRes?.total || 0
 
   const routeToDocCreate = () => {
+    if (isDataSourceNotion) {
+      setNotionPageSelectorModalVisible(true)
+      return
+    }
     router.push(`/datasets/${datasetId}/documents/create`)
   }
 
@@ -96,6 +144,54 @@ const Documents: FC<IDocumentsProps> = ({ datasetId }) => {
 
   const isLoading = !documentsRes && !error
 
+  const handleSaveNotionPageSelected = async (selectedPages: (DataSourceNotionPage & { workspace_id: string })[]) => {
+    const workspacesMap = groupBy(selectedPages, 'workspace_id')
+    const workspaces = Object.keys(workspacesMap).map((workspaceId) => {
+      return {
+        workspaceId,
+        pages: workspacesMap[workspaceId],
+      }
+    })
+    const params = {
+      data_source: {
+        type: dataset?.data_source_type,
+        info_list: {
+          data_source_type: dataset?.data_source_type,
+          notion_info_list: workspaces.map((workspace) => {
+            return {
+              workspace_id: workspace.workspaceId,
+              pages: workspace.pages.map((page) => {
+                const { page_id, page_name, page_icon, type } = page
+                return {
+                  page_id,
+                  page_name,
+                  page_icon,
+                  type,
+                }
+              }),
+            }
+          }),
+        },
+      },
+      indexing_technique: dataset?.indexing_technique,
+      process_rule: {
+        rules: {},
+        mode: 'automatic',
+      },
+    } as CreateDocumentReq
+
+    await createDocument({
+      datasetId,
+      body: params,
+    })
+    mutate()
+    setTimerCanRun(true)
+    // mutateDatasetIndexingStatus(undefined, { revalidate: true })
+    setNotionPageSelectorModalVisible(false)
+  }
+
+  const documentsList = isDataSourceNotion ? documentsWithProgress?.data : documentsRes?.data
+
   return (
     <div className='flex flex-col h-full overflow-y-auto'>
       <div className='flex flex-col justify-center gap-1 px-6 pt-4'>
@@ -113,19 +209,29 @@ const Documents: FC<IDocumentsProps> = ({ datasetId }) => {
           />
           <Button type='primary' onClick={routeToDocCreate} className='!h-8 !text-[13px]'>
             <PlusIcon className='h-4 w-4 mr-2 stroke-current' />
-            {t('datasetDocuments.list.addFile')}
+            {
+              isDataSourceNotion
+                ? t('datasetDocuments.list.addPages')
+                : t('datasetDocuments.list.addFile')
+            }
           </Button>
         </div>
         {isLoading
           ? <Loading type='app' />
           : total > 0
-            ? <List documents={documentsRes?.data || []} datasetId={datasetId} onUpdate={mutate} />
+            ? <List documents={documentsList || []} datasetId={datasetId} onUpdate={mutate} />
             : <EmptyElement onClick={routeToDocCreate} />
         }
         {/* Show Pagination only if the total is more than the limit */}
         {(total && total > limit)
           ? <Pagination current={currPage} onChange={setCurrPage} total={total} limit={limit} />
           : null}
+        <NotionPageSelectorModal
+          isShow={notionPageSelectorModalVisible}
+          onClose={() => setNotionPageSelectorModalVisible(false)}
+          onSave={handleSaveNotionPageSelected}
+          datasetId={dataset?.id || ''}
+        />
       </div>
     </div>
   )

web/app/components/datasets/documents/style.module.css

@@ -75,6 +75,9 @@
 .markdownIcon {
   background-image: url(./assets/md.svg);
 }
+.mdIcon {
+  background-image: url(./assets/md.svg);
+}
 .xlsIcon {
   background-image: url(./assets/xlsx.svg);
 }

web/app/components/header/account-dropdown/index.tsx

@@ -9,9 +9,9 @@ import { Menu, Transition } from '@headlessui/react'
 import Indicator from '../indicator'
 import AccountSetting from '../account-setting'
 import AccountAbout from '../account-about'
+import WorkplaceSelector from './workplace-selector'
 import type { LangGeniusVersionResponse, UserProfileResponse } from '@/models/common'
 import I18n from '@/context/i18n'
-import WorkplaceSelector from './workplace-selector'
 import Avatar from '@/app/components/base/avatar'
 
 type IAppSelectorProps = {