Reorg project (#251)

CMakeLists.txt

@@ -56,4 +56,4 @@ else()
   message("Could not find clang-tidy installation - checks disabled")
 endif()
 
-add_subdirectory(pantab/src)
+add_subdirectory(src/pantab)

CONTRIBUTING.md

@@ -57,9 +57,22 @@ For an editable install of pantab you can simply run `pip install -ve .` from th
 Tests are required for new changes and no code will be accepted without them. You should first set up your test in the appropriate module in the `pantab/tests` directory. You can then run the test suite with
 
 ```sh
-pytest pantab
+pytest tests
 ```
 
+For more advanced use cases where you may want to debug compiled extensions, you _may_ need to build the extension in the source tree and invoke pytest from the src folder. As an example:
+
+```sh
+cmake -S . -B . -DCMAKE_EXPORT_COMPILE_COMMANDS=ON -DCMAKE_BUILD_TYPE=Debug
+cmake --build .
+cd src
+python -m pytest ../tests
+```
+
+Will work as well. You may want to run ``git clean -xfd`` when done with the debugging to clean up the source tree.
+
+pantab would love contributions to make this work *out of source* if it is possible!
+
 ### Style guidelines for code changes
 
 Note that `pantab` uses `black`, `flake8` and `isort` to manage code style. Simply run pre-commit. If pre-commit modifies files, simply add them and run pre-commit again. Note, if you've already run `pre-commit install` it will automatically run before every commit regardless.

pyproject.toml

@@ -48,8 +48,9 @@ dev = ["pytest"]
     filename = "NEWS.rst"
 
 [tool.pytest.ini_options]
-testpaths = [
-    "pantab/tests",
+testpaths = ["tests"]
+addopts = [
+    "--import-mode=importlib",
 ]
 
 [tool.mypy]
@@ -67,7 +68,7 @@ known_first_party = "pantab"
 build = "cp39-*64 cp310-*64 cp311-*64 cp312-*64"
 skip = "*musllinux*"
 
-test-command = "pytest --import-mode=importlib {project}/pantab/tests"
+test-command = "pytest {project}/tests"
 test-requires = ["pytest", "pandas>=2.0.0", "polars", "numpy"]
 
 [tool.ruff]

pantab/src/CMakeLists.txt → src/pantab/CMakeLists.txt

@@ -1,16 +1,16 @@
-nanobind_add_module(pantab NOMINSIZE pantab.cpp numpy_datetime.cpp)
-target_include_directories(pantab PUBLIC ${Python_NumPy_INCLUDE_DIRS})
-target_link_libraries(pantab
+nanobind_add_module(libpantab NOMINSIZE libpantab.cpp numpy_datetime.cpp)
+target_include_directories(libpantab PUBLIC ${Python_NumPy_INCLUDE_DIRS})
+target_link_libraries(libpantab
   PRIVATE Tableau::tableauhyperapi-cxx
   PRIVATE nanoarrow
 )
-set_target_properties(pantab PROPERTIES CXX_CLANG_TIDY "${CLANG_TIDY_COMMAND}")
+set_target_properties(libpantab PROPERTIES CXX_CLANG_TIDY "${CLANG_TIDY_COMMAND}")
 set_target_properties(nanoarrow
                       PROPERTIES POSITION_INDEPENDENT_CODE
                       ON)
 
-install(TARGETS pantab
-  LIBRARY DESTINATION ${SKBUILD_PROJECT_NAME}/src)
+install(TARGETS libpantab
+  LIBRARY DESTINATION ${SKBUILD_PROJECT_NAME}/)
 
 if(WIN32)
   set(HYPERAPI_LIB_NAME "tableauhyperapi.lib")
@@ -26,6 +26,6 @@ endif()
 # Auditwheel doesn't know how to handle the cmake dependencies
 # so we manually install here and exclude from auditwheel
 install(FILES ${tableauhyperapi-cxx_SOURCE_DIR}/lib/${HYPERAPI_LIB_NAME}
-   DESTINATION ${SKBUILD_PROJECT_NAME}/src)
+   DESTINATION ${SKBUILD_PROJECT_NAME}/)
 install(DIRECTORY "${tableauhyperapi-cxx_SOURCE_DIR}/${HYPERAPI_BIN_LOC}/"
-  DESTINATION ${SKBUILD_PROJECT_NAME}/src/hyper)
+  DESTINATION ${SKBUILD_PROJECT_NAME}/hyper)

pantab/__init__.py → src/pantab/__init__.py

@@ -1,9 +1,8 @@
 __version__ = "4.0.0rc"
 
 
-from ._reader import frame_from_hyper, frame_from_hyper_query, frames_from_hyper
-from ._tester import test
-from ._writer import frame_to_hyper, frames_to_hyper
+from pantab._reader import frame_from_hyper, frame_from_hyper_query, frames_from_hyper
+from pantab._writer import frame_to_hyper, frames_to_hyper
 
 __all__ = [
     "__version__",
@@ -12,5 +11,4 @@
     "frames_from_hyper",
     "frame_to_hyper",
     "frames_to_hyper",
-    "test",
 ]

pantab/_reader.py → src/pantab/_reader.py

@@ -7,7 +7,7 @@
 import tableauhyperapi as tab_api
 
 import pantab._types as pantab_types
-import pantab.src.pantab as libpantab  # type: ignore
+import pantab.libpantab as libpantab  # type: ignore
 
 
 def frame_from_hyper_query(

pantab/_writer.py → src/pantab/_writer.py

@@ -7,7 +7,7 @@
 import tableauhyperapi as tab_api
 
 import pantab._types as pantab_types
-import pantab.src.pantab as libpantab  # type: ignore
+import pantab.libpantab as libpantab  # type: ignore
 
 
 def _validate_table_mode(table_mode: Literal["a", "w"]) -> None:

pantab/src/pantab.cpp → src/pantab/libpantab.cpp

@@ -1086,7 +1086,7 @@ auto read_from_hyper_query(const std::string &path, const std::string &query)
   return result;
 }
 
-NB_MODULE(pantab, m) { // NOLINT
+NB_MODULE(libpantab, m) { // NOLINT
   m.def("write_to_hyper", &write_to_hyper, nb::arg("dict_of_capsules"),
         nb::arg("path"), nb::arg("table_mode"), nb::arg("json_columns"),
         nb::arg("geo_columns"))

pantab/tests/test_reader.py → tests/test_reader.py

@@ -2,16 +2,16 @@
 import pandas.testing as tm
 import tableauhyperapi as tab_api
 
-import pantab as pt
+import pantab
 
 
 def test_read_doesnt_modify_existing_file(frame, tmp_hyper):
-    pt.frame_to_hyper(frame, tmp_hyper, table="test")
+    pantab.frame_to_hyper(frame, tmp_hyper, table="test")
     last_modified = tmp_hyper.stat().st_mtime
 
     # Try out our read methods
-    pt.frame_from_hyper(tmp_hyper, table="test")
-    pt.frames_from_hyper(tmp_hyper)
+    pantab.frame_from_hyper(tmp_hyper, table="test")
+    pantab.frames_from_hyper(tmp_hyper)
 
     # Neither should not update file stats
     assert last_modified == tmp_hyper.stat().st_mtime
@@ -46,16 +46,16 @@ def test_reads_nullable_columns(tmp_hyper, compat):
                 inserter.add_rows([[1], [2]])
                 inserter.execute()
 
-    result = pt.frame_from_hyper(tmp_hyper, table=table_name)
+    result = pantab.frame_from_hyper(tmp_hyper, table=table_name)
     expected = pd.DataFrame([[1], [2]], dtype="int32[pyarrow]", columns=[column_name])
     compat.assert_frame_equal(result, expected)
 
 
 def test_read_query(frame, tmp_hyper):
-    pt.frame_to_hyper(frame, tmp_hyper, table="test")
+    pantab.frame_to_hyper(frame, tmp_hyper, table="test")
 
     query = "SELECT int16 AS i, '_' || int32 AS _i2 FROM test"
-    result = pt.frame_from_hyper_query(tmp_hyper, query)
+    result = pantab.frame_from_hyper_query(tmp_hyper, query)
 
     expected = pd.DataFrame([[1, "_2"], [6, "_7"], [0, "_0"]], columns=["i", "_i2"])
     expected = expected.astype({"i": "int16[pyarrow]", "_i2": "large_string[pyarrow]"})
@@ -95,5 +95,5 @@ def test_read_varchar(tmp_hyper):
         [["foo"], ["bar"]], columns=[column_name], dtype="large_string[pyarrow]"
     )
 
-    result = pt.frame_from_hyper(tmp_hyper, table=table_name)
+    result = pantab.frame_from_hyper(tmp_hyper, table=table_name)
     tm.assert_frame_equal(result, expected)

pantab/tests/test_roundtrip.py → tests/test_roundtrip.py

@@ -1,7 +1,7 @@
 import pyarrow as pa
 from tableauhyperapi import TableName
 
-import pantab as pt
+import pantab
 
 
 def test_basic(frame, roundtripped, tmp_hyper, table_name, table_mode, compat):
@@ -11,15 +11,15 @@ def test_basic(frame, roundtripped, tmp_hyper, table_name, table_mode, compat):
         expected = compat.drop_columns(expected, ["interval"])
 
     # Write twice; depending on mode this should either overwrite or duplicate entries
-    pt.frame_to_hyper(
+    pantab.frame_to_hyper(
         frame,
         tmp_hyper,
         table=table_name,
         table_mode=table_mode,
         json_columns={"json"},
         geo_columns={"geography"},
     )
-    pt.frame_to_hyper(
+    pantab.frame_to_hyper(
         frame,
         tmp_hyper,
         table=table_name,
@@ -28,7 +28,9 @@ def test_basic(frame, roundtripped, tmp_hyper, table_name, table_mode, compat):
         geo_columns={"geography"},
     )
 
-    result = pt.frame_from_hyper(tmp_hyper, table=table_name, return_type=return_type)
+    result = pantab.frame_from_hyper(
+        tmp_hyper, table=table_name, return_type=return_type
+    )
 
     if table_mode == "a":
         expected = compat.concat_frames(expected, expected)
@@ -45,22 +47,22 @@ def test_multiple_tables(
         expected = compat.drop_columns(expected, ["interval"])
 
     # Write twice; depending on mode this should either overwrite or duplicate entries
-    pt.frames_to_hyper(
+    pantab.frames_to_hyper(
         {table_name: frame, "table2": frame},
         tmp_hyper,
         table_mode=table_mode,
         json_columns={"json"},
         geo_columns={"geography"},
     )
-    pt.frames_to_hyper(
+    pantab.frames_to_hyper(
         {table_name: frame, "table2": frame},
         tmp_hyper,
         table_mode=table_mode,
         json_columns={"json"},
         geo_columns={"geography"},
     )
 
-    result = pt.frames_from_hyper(tmp_hyper, return_type=return_type)
+    result = pantab.frames_from_hyper(tmp_hyper, return_type=return_type)
 
     if table_mode == "a":
         expected = compat.concat_frames(expected, expected)
@@ -85,15 +87,15 @@ def test_empty_roundtrip(
     # object case is by definition vague, so lets punt that for now
     frame = compat.drop_columns(frame, ["object"])
     empty = compat.empty_like(frame)
-    pt.frame_to_hyper(
+    pantab.frame_to_hyper(
         empty,
         tmp_hyper,
         table=table_name,
         table_mode=table_mode,
         json_columns={"json"},
         geo_columns={"geography"},
     )
-    pt.frame_to_hyper(
+    pantab.frame_to_hyper(
         empty,
         tmp_hyper,
         table=table_name,
@@ -102,7 +104,9 @@ def test_empty_roundtrip(
         geo_columns={"geography"},
     )
 
-    result = pt.frame_from_hyper(tmp_hyper, table=table_name, return_type=return_type)
+    result = pantab.frame_from_hyper(
+        tmp_hyper, table=table_name, return_type=return_type
+    )
 
     expected = compat.drop_columns(expected, ["object"])
     expected = compat.empty_like(expected)