Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 4 additions & 0 deletions .semversioner/next-release/patch-20260225001919068435.json
Original file line number Diff line number Diff line change
@@ -0,0 +1,4 @@
{
"type": "patch",
"description": "fix csv reader"
}
3 changes: 2 additions & 1 deletion packages/graphrag-input/graphrag_input/csv.py
Original file line number Diff line number Diff line change
Expand Up @@ -4,6 +4,7 @@
"""A module containing 'CSVFileReader' model."""

import csv
import io
import logging
import sys

Expand Down Expand Up @@ -39,6 +40,6 @@ async def read_file(self, path: str) -> list[TextDocument]:
"""
file = await self._storage.get(path, encoding=self._encoding)

reader = csv.DictReader(file.splitlines())
reader = csv.DictReader(io.StringIO(file))
rows = list(reader)
return await self.process_data_columns(rows, path)
23 changes: 23 additions & 0 deletions tests/unit/indexing/input/test_csv_loader.py
Original file line number Diff line number Diff line change
Expand Up @@ -54,3 +54,26 @@ async def test_csv_loader_multiple_files():
reader = create_input_reader(config, storage)
documents = await reader.read_files()
assert len(documents) == 4


async def test_csv_loader_preserves_multiline_fields(tmp_path):
"""Multiline quoted CSV fields must retain their internal newlines."""
csv_content = (
"title,text\r\n"
'"Post 1","Line one.\nLine two.\nLine three."\r\n'
'"Post 2","Single line."\r\n'
)
(tmp_path / "input.csv").write_text(csv_content, encoding="utf-8")
config = InputConfig(
type=InputType.Csv,
text_column="text",
title_column="title",
)
storage = create_storage(StorageConfig(base_dir=str(tmp_path)))
reader = create_input_reader(config, storage)
documents = await reader.read_files()
assert len(documents) == 2
assert documents[0].title == "Post 1"
assert documents[0].text == "Line one.\nLine two.\nLine three."
assert documents[1].title == "Post 2"
assert documents[1].text == "Single line."
Loading