Initial implementation of generic Excel-to-DB import tool

Supports .xls and .xlsx, Oracle and PostgreSQL via SQLAlchemy.
Includes CLI (run/inspect/generate-config), YAML config, auto schema
detection, and append/replace/upsert modes.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

tests/test_config.py

@@ -0,0 +1,39 @@
+from pathlib import Path
+import pytest
+import yaml
+
+from excel_import.config import ImportConfig
+
+
+@pytest.fixture
+def config_file(tmp_path: Path) -> Path:
+    cfg = {
+        "dsn": "postgresql+psycopg2://u:p@localhost/db",
+        "sheets": [
+            {
+                "sheet": "Artikel",
+                "header_row": 0,
+                "target_table": "artikel",
+                "mode": "replace",
+                "columns": [
+                    {"source": "Artikelnummer", "target": "art_nr", "dtype": "VARCHAR(50)"},
+                    {"source": "Preis", "target": "preis"},
+                ],
+            }
+        ],
+    }
+    path = tmp_path / "config.yaml"
+    path.write_text(yaml.dump(cfg))
+    return path
+
+
+def test_load_from_yaml(config_file: Path):
+    cfg = ImportConfig.from_yaml(config_file)
+    assert cfg.dsn == "postgresql+psycopg2://u:p@localhost/db"
+    assert len(cfg.sheets) == 1
+    sheet = cfg.sheets[0]
+    assert sheet.sheet == "Artikel"
+    assert sheet.target_table == "artikel"
+    assert sheet.mode == "replace"
+    assert len(sheet.columns) == 2
+    assert sheet.columns[0].dtype == "VARCHAR(50)"