fix(ci): skip parquet tests when pyarrow is absent and add checkout step

.github/workflows/cowork-auto-pr.yml

@@ -12,6 +12,14 @@ jobs:
   ensure-pr:
     runs-on: ubuntu-latest
     steps:
+      # gh pr create requires a local git checkout to diff head against base;
+      # without this step every run failed with "not a git repository" and no
+      # PR was ever opened (fleet-wide defect: 11/11 seeded copies lacked it).
+      - name: Check out the pushed branch
+        uses: actions/checkout@v4
+        with:
+          ref: ${{ github.ref_name }}
+          fetch-depth: 0
       - name: Open PR for this branch if none exists
         env:
           GH_TOKEN: ${{ github.token }}

tests/test_converters.py

@@ -18,6 +18,18 @@
     supported_formats,
 )
 
+try:
+    import pyarrow.parquet  # noqa: F401
+
+    _HAS_PYARROW = True
+except (ImportError, Exception):
+    _HAS_PYARROW = False
+
+requires_pyarrow = pytest.mark.skipif(
+    not _HAS_PYARROW,
+    reason="Parquet support requires pyarrow: pip install 'datamorph[parquet]'",
+)
+
 # ── Fixtures ──────────────────────────────────────────────────────────
 
 
@@ -235,6 +247,7 @@ def test_yaml_to_csv(self, sample_yaml, tmp_path):
 # ── Parquet ───────────────────────────────────────────────────────────
 
 
+@requires_pyarrow
 class TestParquetConversion:
     def test_csv_to_parquet(self, sample_csv, tmp_path):
         output = tmp_path / "output.parquet"
@@ -315,9 +328,7 @@ def test_avro_write_empty(self, tmp_path):
     def test_avro_nullable_first_row(self, tmp_path):
         """Avro should handle nullable fields even when the first row has nulls."""
         path = tmp_path / "data.csv"
-        path.write_text(
-            "name,age,email\nAlice,,alice@test.com\nBob,30,\nCharlie,25,charlie@test.com\n"
-        )
+        path.write_text("name,age,email\nAlice,,alice@test.com\nBob,30,\nCharlie,25,charlie@test.com\n")
         avro_file = tmp_path / "out.avro"
         result = convert(path, avro_file)
         assert not result.errors
@@ -464,15 +475,14 @@ def test_convert_with_format_override(self, runner, sample_csv, tmp_path):
         assert result.exit_code == 0
         assert "Converted" in result.output
 
+    @requires_pyarrow
     def test_convert_to_parquet(self, runner, sample_csv, tmp_path):
         output = tmp_path / "out.parquet"
         result = runner.invoke(cli, ["convert", str(sample_csv), str(output)])
         assert result.exit_code == 0
 
     def test_convert_nonexistent_input(self, runner, tmp_path):
-        result = runner.invoke(
-            cli, ["convert", "/nonexistent/file.csv", str(tmp_path / "out.json")]
-        )
+        result = runner.invoke(cli, ["convert", "/nonexistent/file.csv", str(tmp_path / "out.json")])
         assert result.exit_code != 0
 
     def test_formats_command(self, runner):
@@ -559,6 +569,7 @@ def test_batch_recursive(self, runner, tmp_path):
         # Verify nested directory structure is preserved
         assert any("nested" in str(f) for f in out_files)
 
+    @requires_pyarrow
     def test_batch_to_parquet(self, runner, sample_csv, tmp_path):
         """Batch convert CSV files to Parquet via CLI."""
         output_dir = tmp_path / "pq_out"
@@ -710,12 +721,7 @@ def test_large_json_array(self, tmp_path):
 class TestJsonlConversion:
     def test_jsonl_to_json(self, tmp_path):
         path = tmp_path / "data.jsonl"
-        path.write_text(
-            json.dumps({"name": "Alice", "age": 30})
-            + "\n"
-            + json.dumps({"name": "Bob", "age": 25})
-            + "\n"
-        )
+        path.write_text(json.dumps({"name": "Alice", "age": 30}) + "\n" + json.dumps({"name": "Bob", "age": 25}) + "\n")
         output = tmp_path / "out.json"
         result = convert(path, output)
         assert not result.errors
@@ -726,12 +732,7 @@ def test_jsonl_to_json(self, tmp_path):
 
     def test_jsonl_to_csv(self, tmp_path):
         path = tmp_path / "data.jsonl"
-        path.write_text(
-            json.dumps({"name": "Alice", "age": 30})
-            + "\n"
-            + json.dumps({"name": "Bob", "age": 25})
-            + "\n"
-        )
+        path.write_text(json.dumps({"name": "Alice", "age": 30}) + "\n" + json.dumps({"name": "Bob", "age": 25}) + "\n")
         output = tmp_path / "out.csv"
         result = convert(path, output)
         assert not result.errors
@@ -768,6 +769,7 @@ def test_jsonl_empty(self, tmp_path):
 class TestWriterKwargLeak:
     """Regression tests for delimiter kwarg leaking to non-CSV writers."""
 
+    @requires_pyarrow
     def test_csv_delimiter_to_parquet(self, tmp_path):
         """csv_delimiter should not crash when converting to Parquet."""
         csv_path = tmp_path / "data.csv"