Enable DuckDB >= v0.8.0 (#114)

* enable duckdb v0.9.0 * update test with mixed order schema
cytomining · Oct 5, 2023 · a5b5895 · a5b5895
1 parent d64f954
commit a5b5895
Show file tree

Hide file tree

Showing 3 changed files with 48 additions and 59 deletions.
diff --git a/poetry.lock b/poetry.lock
diff --git a/pyproject.toml b/pyproject.toml
@@ -14,7 +14,7 @@ keywords = ["python", "cellprofiler","single-cell-analysis", "way-lab"]
 python = ">=3.8,<3.13"
 pyarrow = "^13.0.0"
 cloudpathlib = {extras = ["all"], version = "^0.15.0"}
-duckdb = "^0.8.0"
+duckdb = ">=0.8.0"
 parsl = ">=2023.9.18"
 
 [tool.poetry.dev-dependencies]

diff --git a/tests/test_convert.py b/tests/test_convert.py
@@ -567,9 +567,6 @@ def test_to_parquet(
 
     flattened_results = list(itertools.chain(*list(result.values())))
     for i, flattened_result in enumerate(flattened_results):
-        parquet_result = parquet.ParquetDataset(
-            path_or_paths=flattened_result["table"]
-        ).read()
         csv_source = (
             _duckdb_reader()
             .execute(
@@ -581,6 +578,11 @@ def test_to_parquet(
             )
             .arrow()
         )
+        parquet_result = parquet.ParquetDataset(
+            path_or_paths=flattened_result["table"],
+            # set the order of the columns uniformly for schema comparison
+            schema=csv_source.schema,
+        ).read()
         assert parquet_result.schema.equals(csv_source.schema)
         assert parquet_result.shape == csv_source.shape