gardel/lightrag
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Merge pull request #659 from danielaskdd/cvs_robustness

Browse Source 
Enhance robustness of CVS processing ,Fix potential CSV parsing issues
This commit is contained in:
zrguo
2025-01-27 23:12:43 +08:00
committed by GitHub
parent 3b3e0719a7 c8d384f15f
commit ba40a8de8b
2 changed files with 5 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
lightrag/api/lightrag_server.py
View File

@@ -918,7 +918,7 @@ def create_app(args):
case ".pptx": case ".pptx":
if not pm.is_installed("pptx"): if not pm.is_installed("pptx"):
pm.install("pptx") pm.install("pptx")
from pptx import Presentation from pptx import Presentation # type: ignore
# PowerPoint handling # PowerPoint handling
prs = Presentation(file_path) prs = Presentation(file_path)
@@ -1216,7 +1216,7 @@ def create_app(args):
case ".pptx": case ".pptx":
if not pm.is_installed("pptx"): if not pm.is_installed("pptx"):
pm.install("pptx") pm.install("pptx")
from pptx import Presentation from pptx import Presentation # type: ignore
from io import BytesIO from io import BytesIO
# Read PPTX from memory # Read PPTX from memory
@@ -1320,7 +1320,7 @@ def create_app(args):
case ".pptx": case ".pptx":
if not pm.is_installed("pptx"): if not pm.is_installed("pptx"):
pm.install("pptx") pm.install("pptx")
from pptx import Presentation from pptx import Presentation # type: ignore
from io import BytesIO from io import BytesIO
pptx_content = await file.read() pptx_content = await file.read()

4
lightrag/utils.py
View File

@@ -237,13 +237,13 @@ def truncate_list_by_token_size(list_data: list, key: callable, max_token_size:
def list_of_list_to_csv(data: List[List[str]]) -> str: def list_of_list_to_csv(data: List[List[str]]) -> str:
output = io.StringIO() output = io.StringIO()
writer = csv.writer(output) writer = csv.writer(output, quoting=csv.QUOTE_ALL)
writer.writerows(data) writer.writerows(data)
return output.getvalue() return output.getvalue()
def csv_string_to_list(csv_string: str) -> List[List[str]]: def csv_string_to_list(csv_string: str) -> List[List[str]]:
output = io.StringIO(csv_string) output = io.StringIO(csv_string.replace("\x00", ""))
reader = csv.reader(output) reader = csv.reader(output)
return [row for row in reader] return [row for row in reader]