hz/ebiz-dify-ai
1
0
Fork 0
mirror of http://112.124.100.131/huang.ze/ebiz-dify-ai.git synced 2025-12-09 10:56:52 +08:00
Code Issues Packages Projects Releases Wiki Activity

refactor(rag): switch to dify_config. (#6410)

Browse Source 
Co-authored-by: -LAN- <laipz8200@outlook.com>
This commit is contained in:
Poorandy
2024-07-18 18:40:36 +08:00
committed by GitHub
parent 27c8deb4ec
commit c8f5dfcf17
18 changed files with 121 additions and 131 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
api/core/rag/extractor/extract_processor.py
View File

@@ -5,8 +5,8 @@ from typing import Union
from urllib.parse import unquote
import requests
from flask import current_app
from configs import dify_config
from core.rag.extractor.csv_extractor import CSVExtractor
from core.rag.extractor.entity.datasource_type import DatasourceType
from core.rag.extractor.entity.extract_setting import ExtractSetting
@@ -94,9 +94,9 @@ class ExtractProcessor:
storage.download(upload_file.key, file_path)
input_file = Path(file_path)
file_extension = input_file.suffix.lower()
etl_type = current_app.config['ETL_TYPE']
unstructured_api_url = current_app.config['UNSTRUCTURED_API_URL']
unstructured_api_key = current_app.config['UNSTRUCTURED_API_KEY']
etl_type = dify_config.ETL_TYPE
unstructured_api_url = dify_config.UNSTRUCTURED_API_URL
unstructured_api_key = dify_config.UNSTRUCTURED_API_KEY
if etl_type == 'Unstructured':
if file_extension == '.xlsx' or file_extension == '.xls':
extractor = ExcelExtractor(file_path)

4
api/core/rag/extractor/notion_extractor.py
View File

@@ -3,8 +3,8 @@ import logging
from typing import Any, Optional
import requests
from flask import current_app
from configs import dify_config
from core.rag.extractor.extractor_base import BaseExtractor
from core.rag.models.document import Document
from extensions.ext_database import db
@@ -49,7 +49,7 @@ class NotionExtractor(BaseExtractor):
self._notion_access_token = self._get_access_token(tenant_id,
self._notion_workspace_id)
if not self._notion_access_token:
integration_token = current_app.config.get('NOTION_INTEGRATION_TOKEN')
integration_token = dify_config.NOTION_INTEGRATION_TOKEN
if integration_token is None:
raise ValueError(
"Must specify `integration_token` or set environment "

7
api/core/rag/extractor/word_extractor.py
View File

@@ -8,8 +8,8 @@ from urllib.parse import urlparse
import requests
from docx import Document as DocxDocument
from flask import current_app
from configs import dify_config
from core.rag.extractor.extractor_base import BaseExtractor
from core.rag.models.document import Document
from extensions.ext_database import db
@@ -96,10 +96,9 @@ class WordExtractor(BaseExtractor):
storage.save(file_key, rel.target_part.blob)
# save file to db
config = current_app.config
upload_file = UploadFile(
tenant_id=self.tenant_id,
storage_type=config['STORAGE_TYPE'],
storage_type=dify_config.STORAGE_TYPE,
key=file_key,
name=file_key,
size=0,
@@ -114,7 +113,7 @@ class WordExtractor(BaseExtractor):
db.session.add(upload_file)
db.session.commit()
image_map[rel.target_part] = f"![image]({current_app.config.get('CONSOLE_API_URL')}/files/{upload_file.id}/image-preview)"
image_map[rel.target_part] = f"![image]({dify_config.CONSOLE_API_URL}/files/{upload_file.id}/image-preview)"
return image_map