remove deprecated write_df()

was slated for removal 2 years ago in v0.2.8 on 2022-06-29
ruff unignore FBT001 FBT002 DTZ005 PLR ARG001 and fix existing errors

.pre-commit-config.yaml

@@ -7,19 +7,19 @@ default_install_hook_types: [pre-commit, commit-msg]
 
 repos:
   - repo: https://github.com/astral-sh/ruff-pre-commit
-    rev: v0.3.4
+    rev: v0.4.8
     hooks:
       - id: ruff
         args: [--fix]
       - id: ruff-format
 
   - repo: https://github.com/pre-commit/mirrors-mypy
-    rev: v1.9.0
+    rev: v1.10.0
     hooks:
       - id: mypy
 
   - repo: https://github.com/pre-commit/pre-commit-hooks
-    rev: v4.5.0
+    rev: v4.6.0
     hooks:
       - id: check-case-conflict
       - id: check-symlinks
@@ -31,7 +31,7 @@ repos:
       - id: trailing-whitespace
 
   - repo: https://github.com/codespell-project/codespell
-    rev: v2.2.6
+    rev: v2.3.0
     hooks:
       - id: codespell
         stages: [commit, commit-msg]

pyproject.toml

@@ -69,27 +69,25 @@ include = ["**/pyproject.toml", "*.ipynb", "*.py", "*.pyi"]
 [tool.ruff.lint]
 select = ["ALL"]
 ignore = [
-  "ANN101", # Missing type annotation for self in method
+  "ANN101",  # Missing type annotation for self in method
   "ANN401",
-  "ARG001",
   "C901",
   "COM812",
-  "D100",   # Missing docstring in public module
-  "D205",   # 1 blank line required between summary line and description
-  "DTZ005",
+  "D100",    # Missing docstring in public module
+  "D205",    # 1 blank line required between summary line and description
   "EM101",
   "EM102",
-  "FBT001",
-  "FBT002",
-  "PLR",    # pylint refactor
-  "PT006",  # pytest-parametrize-names-wrong-type
+  "PLR0912",
+  "PLR0913",
+  "PLR0915",
+  "PT006",   # pytest-parametrize-names-wrong-type
   "PTH",
   "T201",
   "TRY003",
 ]
 pydocstyle.convention = "google"
 
 [tool.ruff.lint.per-file-ignores]
-"tests/*" = ["D103", "D104", "INP001", "S101"]
+"tests/*" = ["D103", "D104", "FBT001", "INP001", "S101"]
 "__init__.py" = ["F401"]
 "examples/*" = ["D102", "D103", "D107", "E402", "FA102"]

tensorboard_reducer/__init__.py

@@ -8,7 +8,7 @@
 from tensorboard_reducer.load import load_tb_events
 from tensorboard_reducer.main import main
 from tensorboard_reducer.reduce import reduce_events
-from tensorboard_reducer.write import write_data_file, write_df, write_tb_events
+from tensorboard_reducer.write import write_data_file, write_tb_events
 
 try:  # noqa: SIM105
     __version__ = version("tensorboard-reducer")

tensorboard_reducer/load.py

@@ -13,6 +13,7 @@
 
 def load_tb_events(
     input_dirs: list[str],
+    *,
     strict_tags: bool = True,
     strict_steps: bool = True,
     handle_dup_steps: HandleDupSteps | None = None,
@@ -192,7 +193,9 @@ def load_tb_events(
             for tag in list(out_dict)[:50]:
                 df_scalar = out_dict[tag]
                 print(f"- '{tag}': {df_scalar.shape}")
-            if len(out_dict) > 50:
+
+            max_tags_to_print = 50
+            if len(out_dict) > max_tags_to_print:
                 print("...")
 
     return out_dict

tensorboard_reducer/main.py

@@ -117,10 +117,11 @@ def main(argv: list[str] | None = None) -> int:
 
     reduced_events = reduce_events(events_dict, reduce_ops, verbose=args.verbose)
 
+    common_kwds = {"overwrite": overwrite, "verbose": args.verbose}
     if out_path.endswith(".csv"):
-        write_data_file(reduced_events, out_path, overwrite, verbose=args.verbose)
+        write_data_file(reduced_events, out_path, **common_kwds)
     else:
-        write_tb_events(reduced_events, out_path, overwrite, verbose=args.verbose)
+        write_tb_events(reduced_events, out_path, **common_kwds)
     return 0
 
 

tensorboard_reducer/reduce.py

@@ -9,6 +9,7 @@
 def reduce_events(
     events_dict: dict[str, pd.DataFrame],
     reduce_ops: Sequence[str],
+    *,
     verbose: bool = False,
 ) -> dict[str, dict[str, pd.DataFrame]]:
     """Perform numpy reduce operations along the last dimension of each array in a

tensorboard_reducer/write.py

@@ -2,15 +2,14 @@
 
 import os
 import sys
-from typing import Any
 
 import pandas as pd
 from tqdm import tqdm
 
 _known_extensions = (".csv", ".json", ".xlsx")
 
 
-def _rm_rf_or_raise(path: str, overwrite: bool) -> None:
+def _rm_rf_or_raise(path: str, *, overwrite: bool) -> None:
     """Remove the directory tree below dir if overwrite is True.
 
     Args:
@@ -48,6 +47,7 @@ def _rm_rf_or_raise(path: str, overwrite: bool) -> None:
 def write_tb_events(
     data_to_write: dict[str, dict[str, pd.DataFrame]],
     out_dir: str,
+    *,
     overwrite: bool = False,
     verbose: bool = False,
 ) -> list[str]:
@@ -97,7 +97,7 @@ def write_tb_events(
             if verbose:
                 print(f"Writing mean{symbol}std reduction to disk...", file=sys.stderr)
 
-            _rm_rf_or_raise(std_out_dir, overwrite)
+            _rm_rf_or_raise(std_out_dir, overwrite=overwrite)
             out_dirs.append(std_out_dir)
 
             writer = SummaryWriter(std_out_dir)
@@ -116,7 +116,7 @@ def write_tb_events(
         op_out_dir = f"{out_dir}{out_dir_op_connector}{op}"
         out_dirs.append(op_out_dir)
 
-        _rm_rf_or_raise(op_out_dir, overwrite)
+        _rm_rf_or_raise(op_out_dir, overwrite=overwrite)
 
         writer = SummaryWriter(op_out_dir)
 
@@ -136,16 +136,10 @@ def write_tb_events(
     return out_dirs
 
 
-def write_df(*args: Any) -> None:
-    """Inform users of breaking change if they try to use the old API."""
-    raise NotImplementedError(
-        "write_df() was renamed to write_data_file() in tensorboard-reducer v0.2.8"
-    )
-
-
 def write_data_file(
     data_to_write: dict[str, dict[str, pd.DataFrame]],
     out_path: str,
+    *,
     overwrite: bool = False,
     verbose: bool = False,
 ) -> str:
@@ -168,7 +162,7 @@ def write_data_file(
     Returns:
         str: Path to the new data file.
     """
-    _rm_rf_or_raise(out_path, overwrite)
+    _rm_rf_or_raise(out_path, overwrite=overwrite)
 
     # create multi-index dataframe from event data with reduce op names as 1st-level col
     # names and tag names as 2nd level

tests/test_load.py

@@ -13,7 +13,7 @@
 
 @pytest.mark.parametrize("verbose", [True, False])
 def test_load_tb_events_strict(
-    verbose: bool, capsys: pytest.CaptureFixture[str]
+    *, verbose: bool, capsys: pytest.CaptureFixture[str]
 ) -> None:
     """Test load_tb_events for strict input data, i.e. without any of the special cases
     below.
@@ -32,13 +32,14 @@ def test_load_tb_events_strict(
     assert actual_keys == ["strict/foo"], assert_keys
 
     n_steps, n_runs = events_dict["strict/foo"].shape
+    n_expected = 100
     assert_len = (
-        f"load_tb_events() returned TB event with {n_steps} steps, expected 100"
+        f"load_tb_events() returned TB event with {n_steps} steps, {n_expected=}"
     )
-    assert n_steps == 100, assert_len
+    assert n_steps == n_expected, assert_len
 
     assert_len = f"load_tb_events() returned {n_runs} TB runs, expected 3"
-    assert n_steps == 100, assert_len
+    assert n_steps == n_expected, assert_len
 
     # columns correspond to different runs for the same tag, the mean across a run is
     # meaningless and only used for asserting value constancy

tests/test_write.py

@@ -103,8 +103,3 @@ def test_write_data_file_with_bad_ext(
 ) -> None:
     with pytest.raises(ValueError, match="has unknown extension, should be one of"):
         tbr.write_data_file(reduced_events, "foo.bad_ext")
-
-
-def test_write_df() -> None:
-    with pytest.raises(NotImplementedError, match=r"write_df\(\) was renamed"):
-        tbr.write_df(None, "foo.csv")