feat: implemented custom errors

.github/workflows/CI.yml

@@ -110,3 +110,31 @@ jobs:
         command: build
         args: "-o dist --interpreter python${{ matrix.python-version }}"
         target: ${{ steps.target.outputs.target }}
+
+  check-docs:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+      - name: Set up Python
+        uses: actions/setup-python@v5
+        with:
+          python-version: "3.11"
+      - name: Set up rust toolchain
+        uses: actions-rs/toolchain@v1
+        with:
+          profile: minimal
+          toolchain: stable
+          override: true
+      - run: |
+          git config user.name github-actions
+          git config user.email [email protected]
+
+          # venv required by maturin
+          python3 -m venv .venv
+          source .venv/bin/activate
+
+          make install-test-requirements
+          make install-doc-requirements
+          # Required for pdoc to be able to import the sources
+          make dev-install
+          make doc

Cargo.toml

@@ -9,10 +9,9 @@ name = "fastexcel"
 crate-type = ["cdylib"]
 
 [dependencies]
-anyhow = "1.0.80"
 calamine = { version = "0.24.0", features = ["dates"] }
 chrono = { version = "0.4.34", default-features = false }
-pyo3 = { version = "0.20.3", features = ["extension-module", "anyhow", "abi3-py38"] }
+pyo3 = { version = "0.20.3", features = ["extension-module", "abi3-py38"] }
 
 [dependencies.arrow]
 version = "50.0.0"

Makefile

@@ -22,6 +22,7 @@ format:
 	$(ruff) --fix
 	$(format)
 	$(fmt)
+	$(clippy) --fix --lib -p fastexcel --allow-dirty --allow-staged
 
 install-test-requirements:
 	pip install -U -r test-requirements.txt -r build-requirements.txt

python/fastexcel/__init__.py

@@ -12,7 +12,19 @@
 
 import pyarrow as pa
 
-from ._fastexcel import __version__, _ExcelReader, _ExcelSheet
+from ._fastexcel import (
+    ArrowError,
+    CalamineCellError,
+    CalamineError,
+    CannotRetrieveCellDataError,
+    FastExcelError,
+    InvalidParametersError,
+    SheetNotFoundError,
+    UnsupportedColumnTypeCombinationError,
+    __version__,
+    _ExcelReader,
+    _ExcelSheet,
+)
 from ._fastexcel import read_excel as _read_excel
 
 
@@ -202,4 +214,17 @@ def read_excel(path: Path | str) -> ExcelReader:
     return ExcelReader(_read_excel(expanduser(path)))
 
 
-__all__ = ("ExcelReader", "ExcelSheet", "read_excel", "__version__")
+__all__ = (
+    "__version__",
+    "read_excel",
+    "ExcelReader",
+    "ExcelSheet",
+    "FastExcelError",
+    "CannotRetrieveCellDataError",
+    "CalamineCellError",
+    "CalamineError",
+    "SheetNotFoundError",
+    "ArrowError",
+    "InvalidParametersError",
+    "UnsupportedColumnTypeCombinationError",
+)

python/fastexcel/_fastexcel.pyi

@@ -61,3 +61,13 @@ def read_excel(path: str) -> _ExcelReader:
     """Reads an excel file and returns an ExcelReader"""
 
 __version__: str
+
+# Exceptions
+class FastExcelError(Exception): ...
+class UnsupportedColumnTypeCombinationError(FastExcelError): ...
+class CannotRetrieveCellDataError(FastExcelError): ...
+class CalamineCellError(FastExcelError): ...
+class CalamineError(FastExcelError): ...
+class SheetNotFoundError(FastExcelError): ...
+class ArrowError(FastExcelError): ...
+class InvalidParametersError(FastExcelError): ...

python/tests/test_errors.py

@@ -0,0 +1,60 @@
+from __future__ import annotations
+
+import fastexcel
+import pytest
+from utils import path_for_fixture
+
+
+def test_does_not_exist() -> None:
+    expected_message = """calamine error: Cannot detect file format
+Context:
+    0: Could not open workbook at path_does_not_exist.nope
+    1: could not load excel file at path_does_not_exist.nope"""
+
+    with pytest.raises(fastexcel.CalamineError, match=expected_message) as exc_info:
+        fastexcel.read_excel("path_does_not_exist.nope")
+
+    assert exc_info.value.__doc__ == "Generic calamine error"
+
+    # Should also work with the base error type
+    with pytest.raises(fastexcel.FastExcelError, match=expected_message):
+        fastexcel.read_excel("path_does_not_exist.nope")
+
+
+def test_sheet_not_found_error() -> None:
+    excel_reader = fastexcel.read_excel(path_for_fixture("fixture-single-sheet.xlsx"))
+    expected_message = """sheet at index 42 not found
+Context:
+    0: Sheet index 42 is out of range. File has 1 sheets"""
+
+    with pytest.raises(fastexcel.SheetNotFoundError, match=expected_message) as exc_info:
+        excel_reader.load_sheet(42)
+
+    assert exc_info.value.__doc__ == "Sheet was not found"
+
+    # Should also work with the base error type
+    with pytest.raises(fastexcel.FastExcelError, match=expected_message):
+        excel_reader.load_sheet(42)
+
+
+@pytest.mark.parametrize(
+    "exc_class, expected_docstring",
+    [
+        (fastexcel.FastExcelError, "The base class for all fastexcel errors"),
+        (
+            fastexcel.UnsupportedColumnTypeCombinationError,
+            "Column contains an unsupported type combination",
+        ),
+        (fastexcel.CannotRetrieveCellDataError, "Data for a given cell cannot be retrieved"),
+        (
+            fastexcel.CalamineCellError,
+            "calamine returned an error regarding the content of the cell",
+        ),
+        (fastexcel.CalamineError, "Generic calamine error"),
+        (fastexcel.SheetNotFoundError, "Sheet was not found"),
+        (fastexcel.ArrowError, "Generic arrow error"),
+        (fastexcel.InvalidParametersError, "Provided parameters are invalid"),
+    ],
+)
+def test_docstrings(exc_class: type[Exception], expected_docstring: str) -> None:
+    assert exc_class.__doc__ == expected_docstring

python/tests/test_fastexcel.py

@@ -387,7 +387,9 @@ def test_sheet_with_pagination_out_of_bound():
     excel_reader = fastexcel.read_excel(path_for_fixture("fixture-single-sheet-with-types.xlsx"))
     assert excel_reader.sheet_names == ["Sheet1"]
 
-    with pytest.raises(RuntimeError, match="To many rows skipped. Max height is 4"):
+    with pytest.raises(
+        fastexcel.InvalidParametersError, match="Too many rows skipped. Max height is 4"
+    ):
         excel_reader.load_sheet(
             0,
             skip_rows=1000000,