feat: allow to select a subset of columns

Cargo.toml

@@ -11,7 +11,8 @@ crate-type = ["cdylib"]
 [dependencies]
 calamine = { version = "0.24.0", features = ["dates"] }
 chrono = { version = "0.4.34", default-features = false }
-pyo3 = { version = "0.20.3", features = ["extension-module", "abi3-py38"] }
+# NOTE: "extension-module" is actually required, see comments on features below
+pyo3 = { version = "0.20.3", features = ["abi3-py38"] }
 
 [dependencies.arrow]
 version = "50.0.0"
@@ -20,4 +21,14 @@ default-features = false
 features = ["pyarrow"]
 
 [dev-dependencies]
+pretty_assertions = "1.4.0"
 rstest = { version = "0.18.2", default-features = false }
+
+# NOTE: This is a hack to bypass pyo3 limitations when testing:
+# https://pyo3.rs/v0.20.3/faq.html#i-cant-run-cargo-test-or-i-cant-build-in-a-cargo-workspace-im-having-linker-issues-like-symbol-not-found-or-undefined-reference-to-_pyexc_systemerror
+[features]
+extension-module = ["pyo3/extension-module"]
+default = ["extension-module"]
+# feature for tests only. This makes Python::with_gil auto-initialize Python
+# interpreters, which allows us ot instantiate Python objects in tests
+tests = ["pyo3/auto-initialize"]

Makefile

@@ -9,21 +9,30 @@ pytest	= pytest -v
 ## Rust
 clippy		= cargo clippy
 fmt		= cargo fmt
-cargo-test	= cargo test
+cargo-test	= cargo test --no-default-features --features tests
 ## Docs
 pdoc	= pdoc -o docs python/fastexcel
 
-lint:
+lint-python:
 	$(ruff)
 	$(format)  --check --diff
+
+lint-rust:
 	$(mypy)
 	$(clippy)
-format:
+
+lint: lint-rust lint-python
+
+format-python:
 	$(ruff) --fix
 	$(format)
+
+format-rust:
 	$(fmt)
 	$(clippy) --fix --lib -p fastexcel --allow-dirty --allow-staged
 
+format: format-rust format-python
+
 install-test-requirements:
 	pip install -U -r test-requirements.txt -r build-requirements.txt
 
@@ -39,10 +48,14 @@ dev-install:
 prod-install:
 	./prod_install.sh
 
-test:
+test-rust:
 	$(cargo-test)
+
+test-python:
 	$(pytest)
 
+test: test-rust test-python
+
 doc:
 	$(pdoc)
 

python/fastexcel/__init__.py

@@ -17,6 +17,7 @@
     CalamineCellError,
     CalamineError,
     CannotRetrieveCellDataError,
+    ColumnNotFoundError,
     FastExcelError,
     InvalidParametersError,
     SheetNotFoundError,
@@ -101,6 +102,7 @@ def load_sheet_by_name(
         skip_rows: int = 0,
         n_rows: int | None = None,
         schema_sample_rows: int | None = 1_000,
+        use_columns: list[str] | list[int] | None = None,
     ) -> ExcelSheet:
         """Loads a sheet by name.
 
@@ -126,6 +128,7 @@ def load_sheet_by_name(
                 skip_rows=skip_rows,
                 n_rows=n_rows,
                 schema_sample_rows=schema_sample_rows,
+                use_columns=use_columns,
             )
         )
 
@@ -138,6 +141,7 @@ def load_sheet_by_idx(
         skip_rows: int = 0,
         n_rows: int | None = None,
         schema_sample_rows: int | None = 1_000,
+        use_columns: list[str] | list[int] | None = None,
     ) -> ExcelSheet:
         """Loads a sheet by index.
 
@@ -165,6 +169,7 @@ def load_sheet_by_idx(
                 skip_rows=skip_rows,
                 n_rows=n_rows,
                 schema_sample_rows=schema_sample_rows,
+                use_columns=use_columns,
             )
         )
 
@@ -177,6 +182,7 @@ def load_sheet(
         skip_rows: int = 0,
         n_rows: int | None = None,
         schema_sample_rows: int | None = 1_000,
+        use_columns: list[str] | list[int] | None = None,
     ) -> ExcelSheet:
         """Loads a sheet by name if a string is passed or by index if an integer is passed.
 
@@ -190,6 +196,7 @@ def load_sheet(
                 skip_rows=skip_rows,
                 n_rows=n_rows,
                 schema_sample_rows=schema_sample_rows,
+                use_columns=use_columns,
             )
             if isinstance(idx_or_name, int)
             else self.load_sheet_by_name(
@@ -199,6 +206,7 @@ def load_sheet(
                 skip_rows=skip_rows,
                 n_rows=n_rows,
                 schema_sample_rows=schema_sample_rows,
+                use_columns=use_columns,
             )
         )
 
@@ -224,6 +232,7 @@ def read_excel(path: Path | str) -> ExcelReader:
     "CalamineCellError",
     "CalamineError",
     "SheetNotFoundError",
+    "ColumnNotFoundError",
     "ArrowError",
     "InvalidParametersError",
     "UnsupportedColumnTypeCombinationError",

python/fastexcel/_fastexcel.pyi

@@ -33,6 +33,7 @@ class _ExcelReader:
         skip_rows: int = 0,
         n_rows: int | None = None,
         schema_sample_rows: int | None = 1_000,
+        use_columns: list[str] | list[int] | None = None,
     ) -> _ExcelSheet: ...
     def load_sheet_by_idx(
         self,
@@ -43,16 +44,7 @@ class _ExcelReader:
         skip_rows: int = 0,
         n_rows: int | None = None,
         schema_sample_rows: int | None = 1_000,
-    ) -> _ExcelSheet: ...
-    def load_sheet(
-        self,
-        idx_or_name: int | str,
-        *,
-        header_row: int | None = 0,
-        column_names: list[str] | None = None,
-        skip_rows: int = 0,
-        n_rows: int | None = None,
-        schema_sample_rows: int | None = 1_000,
+        use_columns: list[str] | list[int] | None = None,
     ) -> _ExcelSheet: ...
     @property
     def sheet_names(self) -> list[str]: ...
@@ -69,5 +61,6 @@ class CannotRetrieveCellDataError(FastExcelError): ...
 class CalamineCellError(FastExcelError): ...
 class CalamineError(FastExcelError): ...
 class SheetNotFoundError(FastExcelError): ...
+class ColumnNotFoundError(FastExcelError): ...
 class ArrowError(FastExcelError): ...
 class InvalidParametersError(FastExcelError): ...