handle creator parsing with comma

src/croissant_baker/__main__.py

@@ -1,5 +1,6 @@
 """Command-line interface for Croissant Baker."""
 
+import csv
 import typer
 from pathlib import Path
 import importlib.metadata
@@ -163,21 +164,27 @@ def main(
             parsed_creators = []
             if creator:
                 for creator_info in creator:
-                    # Parse format: "Name[,Email[,URL]]"
-                    parts = [part.strip() for part in creator_info.split(",")]
+                    creator_info = creator_info.strip()
 
-                    if not parts[0]:  # Empty name
+                    # Preferred: semicolon
+                    if ";" in creator_info:
+                        creator_parts = [p.strip() for p in creator_info.split(";")]
+
+                    else:
+                        # Use CSV parsing for comma cases (handles quotes properly)
+                        creator_parts = next(csv.reader([creator_info]))
+                        creator_parts = [p.strip() for p in creator_parts]
+
+                    if not creator_parts or not creator_parts[0]:
                         continue
 
-                    creator_obj = {"name": parts[0]}  # Name is required
+                    creator_obj = {"name": creator_parts[0]}
 
-                    # Add optional email if provided and not empty
-                    if len(parts) > 1 and parts[1]:
-                        creator_obj["email"] = parts[1]
+                    if len(creator_parts) > 1 and creator_parts[1]:
+                        creator_obj["email"] = creator_parts[1]
 
-                    # Add optional URL if provided and not empty
-                    if len(parts) > 2 and parts[2]:
-                        creator_obj["url"] = parts[2]
+                    if len(creator_parts) > 2 and creator_parts[2]:
+                        creator_obj["url"] = creator_parts[2]
 
                     parsed_creators.append(creator_obj)
 

tests/test_cli.py

@@ -135,6 +135,60 @@ def test_missing_creator_required(csv_dataset: Path, tmp_path: Path) -> None:
     assert "Example:" in result.stderr
 
 
+def test_creator_parsing_variants(csv_dataset: Path, tmp_path: Path) -> None:
+    """Test creator parsing for comma, quoted, and semicolon formats."""
+
+    test_cases = [
+        # (input, expected_strings)
+        ('"Google, LLC"', ["Google, LLC"]),
+        ('"Google, LLC",info@google.com', ["Google, LLC", "info@google.com"]),
+        (
+            '"Google, LLC",info@google.com,https://google.com',
+            ["Google, LLC", "info@google.com", "https://google.com"],
+        ),
+        (
+            '"Doe, Jr., John",john@example.com',
+            ["Doe, Jr., John", "john@example.com"],
+        ),
+        # Backward compatibility
+        ("Alice Smith", ["Alice Smith"]),
+        ("Alice Smith,alice@example.com", ["Alice Smith", "alice@example.com"]),
+        (
+            "Alice Smith,alice@example.com,https://example.com",
+            ["Alice Smith", "alice@example.com", "https://example.com"],
+        ),
+        # Semicolon format
+        (
+            "Google, LLC;info@google.com;https://google.com",
+            ["Google, LLC", "info@google.com", "https://google.com"],
+        ),
+    ]
+
+    for creator_input, expected_values in test_cases:
+        output = tmp_path / f"output_{hash(creator_input)}.jsonld"
+
+        result = runner.invoke(
+            app,
+            [
+                "--input",
+                str(csv_dataset),
+                "--output",
+                str(output),
+                "--creator",
+                creator_input,
+            ],
+        )
+
+        assert result.exit_code == 0, f"Failed for input: {creator_input}"
+
+        content = output.read_text()
+
+        for expected in expected_values:
+            assert expected in content, (
+                f"Missing '{expected}' for input: {creator_input}"
+            )
+
+
 def test_invalid_date_format(csv_dataset: Path, tmp_path: Path) -> None:
     """Test that invalid date format gives clear error message."""
     output = tmp_path / "output.jsonld"