Austin/dify
1
0
Fork 0
mirror of https://github.com/langgenius/dify.git synced 2024-11-16 11:42:29 +08:00
Code Issues Actions 3 Packages Projects Releases Wiki Activity

feat(document_extractor): integrate unstructured API for PPTX extraction (#10180)

Browse Source
This commit is contained in:
-LAN- 2024-11-01 23:19:11 +08:00 committed by Joel
parent ba48754be6
commit 101d9798f0
1 changed files with 10 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

11
api/core/workflow/nodes/document_extractor/node.py
View File

@ -6,12 +6,14 @@ import docx
import pandas as pd
import pypdfium2
import yaml
from unstructured.partition.api import partition_via_api
from unstructured.partition.email import partition_email
from unstructured.partition.epub import partition_epub
from unstructured.partition.msg import partition_msg
from unstructured.partition.ppt import partition_ppt
from unstructured.partition.pptx import partition_pptx
from configs import dify_config
from core.file import File, FileTransferMethod, file_manager
from core.helper import ssrf_proxy
from core.variables import ArrayFileSegment
@ -263,7 +265,14 @@ def _extract_text_from_ppt(file_content: bytes) -> str:
def _extract_text_from_pptx(file_content: bytes) -> str:
try:
with io.BytesIO(file_content) as file:
elements = partition_pptx(file=file)
if dify_config.UNSTRUCTURED_API_URL and dify_config.UNSTRUCTURED_API_KEY:
elements = partition_via_api(
file=file,
api_url=dify_config.UNSTRUCTURED_API_URL,
api_key=dify_config.UNSTRUCTURED_API_KEY,
)
else:
elements = partition_pptx(file=file)
return "\n".join([getattr(element, "text", "") for element in elements])
except Exception as e:
raise TextExtractionError(f"Failed to extract text from PPTX: {str(e)}") from e