Skip to content

armgit / dify

Public Imported from GitHub
0

Build Agentic workflows, RAG pipelines, with rich AI model and tool support on one collaborative workspace. Deploy on cloud, VPC, or self-hosted, so teams move from prototype to production without rebuilding the stack.

fix(rag): render integer CSV column labels in CSVExtractor (#43505)

Sourav Rajviec809d436ad6
diff --git a/api/core/rag/extractor/csv_extractor.py b/api/core/rag/extractor/csv_extractor.py
index b836e789c2d..ed8b26d7939 100644
--- a/api/core/rag/extractor/csv_extractor.py
+++ b/api/core/rag/extractor/csv_extractor.py
@@ -69,7 +69,7 @@ class CSVExtractor(BaseExtractor):
# create document objects
for i, row in df.iterrows():
- content = ";".join(f"{col.strip()}: {str(row[col]).strip()}" for col in df.columns)
+ content = ";".join(f"{str(col).strip()}: {str(row[col]).strip()}" for col in df.columns)
source = row[self.source_column] if self.source_column else ""
metadata = {"source": source, "row": i}
doc = Document(page_content=content, metadata=metadata)
diff --git a/api/tests/unit_tests/core/rag/extractor/test_csv_extractor.py b/api/tests/unit_tests/core/rag/extractor/test_csv_extractor.py
index 023c28d5d14..941b6971136 100644
--- a/api/tests/unit_tests/core/rag/extractor/test_csv_extractor.py
+++ b/api/tests/unit_tests/core/rag/extractor/test_csv_extractor.py
@@ -38,6 +38,25 @@ class TestCSVExtractor:
assert docs[0].page_content == f"value: {value};body: reference"
assert docs[0].metadata["source"] == value
+ def test_extract_header_none_uses_integer_column_labels(self, tmp_path: Path) -> None:
+ file_path = tmp_path / "headerless.csv"
+ file_path.write_text("00123,NA\n00456,NULL\n", encoding="utf-8")
+
+ docs = CSVExtractor(str(file_path), encoding="utf-8", csv_args={"header": None}).extract()
+
+ assert len(docs) == 2
+ assert docs[0].page_content == "0: 00123;1: NA"
+ assert docs[1].page_content == "0: 00456;1: NULL"
+
+ def test_extract_explicit_integer_names_in_csv_args(self, tmp_path: Path) -> None:
+ file_path = tmp_path / "named.csv"
+ file_path.write_text("a,b\n", encoding="utf-8")
+
+ docs = CSVExtractor(str(file_path), encoding="utf-8", csv_args={"header": None, "names": [10, 20]}).extract()
+
+ assert len(docs) == 1
+ assert docs[0].page_content == "10: a;20: b"
+
def test_extract_honors_explicit_csv_args(self, tmp_path: Path) -> None:
file_path = tmp_path / "data.csv"
file_path.write_text("value;body\n00123;NA\n", encoding="utf-8")