From 63d42405476741b35832ef34364de0d2685708ad Mon Sep 17 00:00:00 2001
From: Shubham Padkonde <shubhampadkonde12@gmail.com>
Date: Mon, 26 Oct 2026 16:06:09 +0530
Subject: [PATCH] fix: accept UTF-8 signatures in tabular input

---
 databusclient/filehandling/format.py |  4 ++--
 tests/test_tabular_bom.py            | 33 ++++++++++++++++++++++++++++
 2 files changed, 35 insertions(+), 2 deletions(-)
 create mode 100644 tests/test_tabular_bom.py

diff --git a/databusclient/filehandling/format.py b/databusclient/filehandling/format.py
index 262f61e..c1cd4ec 100644
--- a/databusclient/filehandling/format.py
+++ b/databusclient/filehandling/format.py
@@ -411,7 +411,7 @@ def read(self, source: str, input_format: str) -> list:
                 f"Supported: {list(TABULAR_FORMATS)}"
             )
         delimiter = TABULAR_FORMATS[input_format]
-        with open(source, "r", newline="", encoding="utf-8") as f:
+        with open(source, "r", newline="", encoding="utf-8-sig") as f:
             reader = csv.reader(f, delimiter=delimiter)
             return list(reader)
 
@@ -592,4 +592,4 @@ def convert_file(
     raise ValueError(
         f"Conversion from '{input_format}' ({input_class}) to "
         f"'{convert_format}' ({output_class}) is not supported."
-    )
\ No newline at end of file
+    )
diff --git a/tests/test_tabular_bom.py b/tests/test_tabular_bom.py
new file mode 100644
index 0000000..6a6ab21
--- /dev/null
+++ b/tests/test_tabular_bom.py
@@ -0,0 +1,33 @@
+"""UTF-8 signatures must not become part of tabular column names."""
+
+import csv
+
+import pytest
+from rdflib import Graph, Literal, URIRef
+
+from databusclient.filehandling.format import TSDHandler
+from databusclient.filehandling.mapping import convert_csv_to_rdf
+
+
+@pytest.mark.parametrize("input_format,delimiter", [("csv", ","), ("tsv", "\t")])
+@pytest.mark.parametrize("encoding", ["utf-8", "utf-8-sig"])
+def test_tabular_conversion_accepts_utf8_bom(tmp_path, input_format, delimiter, encoding):
+    source = tmp_path / f"input.{input_format}"
+    rows = [["resource", "https://example.org/name"], ["alice", "Zoë\ufeff Smith"]]
+    with source.open("w", encoding=encoding, newline="") as stream:
+        csv.writer(stream, delimiter=delimiter).writerows(rows)
+
+    assert TSDHandler().read(str(source), input_format) == rows
+
+    output = tmp_path / "output.ttl"
+    convert_csv_to_rdf(
+        str(source), str(output), input_format, "turtle", "https://example.org"
+    )
+    graph = Graph().parse(output, format="turtle")
+    assert set(graph) == {
+        (
+            URIRef("https://example.org/alice"),
+            URIRef("https://example.org/name"),
+            Literal("Zoë\ufeff Smith"),
+        )
+    }