csvsmith 0.11.1__tar.gz → 0.12.0__tar.gz

{csvsmith-0.11.1/src/csvsmith.egg-info → csvsmith-0.12.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: csvsmith
-Version: 0.11.1
+Version: 0.12.0
 Summary: Small CSV utilities: row deduplication, classification, row filtering, and CLI helpers.
 Author-email: Eiichi YAMAMOTO <info@yeiichi.com>
 License: MIT License

{csvsmith-0.11.1 → csvsmith-0.12.0}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "csvsmith"
-version = "0.11.1"
+version = "0.12.0"
 description = "Small CSV utilities: row deduplication, classification, row filtering, and CLI helpers."
 readme = "README.rst"
 requires-python = ">=3.10"

{csvsmith-0.11.1 → csvsmith-0.12.0}/src/csvsmith/__init__.py

@@ -21,6 +21,7 @@ Public API:
 - concentrate_csv
 - rehydrate_csv
 - create_sample_csv
+- strip_utf8_bom
 - DataFrame
 - build_filter
 - infer_type
@@ -67,6 +68,7 @@ from .tools.row_dedup import (
     dedupe_with_report,
 )
 from .tools.sample_csv import SampleCSVResult, create_sample_csv
+from .tools.strip_bom import StripBomResult, strip_utf8_bom
 from .tools.strict_concat import save_csv, strict_concat_rows
 from .utils.clean_numeric import clean_numeric
 from .utils.distance import StringDistance, Relation, Result, analyze_pair
@@ -101,6 +103,8 @@ __all__ = [
     "rehydrate_csv",
     "SampleCSVResult",
     "create_sample_csv",
+    "StripBomResult",
+    "strip_utf8_bom",
     "DataFrame",
     "build_filter",
     "infer_type",

{csvsmith-0.11.1 → csvsmith-0.12.0}/src/csvsmith/cli.py

@@ -24,6 +24,7 @@ from .tools.sample_csv import (
     DEFAULT_ROW_COUNT,
     create_sample_csv,
 )
+from .tools.strip_bom import strip_utf8_bom
 from .tools.strict_concat import save_csv, strict_concat_rows
 from .tools.find_matches_in_csv import find_matches_in_csv
 from .tools.knapsack_csv import mark_knapsack_csv
@@ -204,6 +205,18 @@ def cmd_find_matches(args: argparse.Namespace) -> int:
     return 0
 
 
+def cmd_strip_bom(args: argparse.Namespace) -> int:
+    try:
+        result = strip_utf8_bom(args.input, args.output, in_place=args.in_place)
+    except (OSError, ValueError) as e:
+        print(f"Error: {e}", file=sys.stderr)
+        return 1
+
+    status = "removed BOM" if result.removed else "no BOM found"
+    print(f"Wrote CSV to: {result.output_path} ({status})")
+    return 0
+
+
 def cmd_strict_concat(args: argparse.Namespace) -> int:
     input_dir = Path(args.input_dir)
     if not input_dir.is_dir():
@@ -388,6 +401,25 @@ def _add_find_matches_parser(subparsers) -> None:
     parser.set_defaults(func=cmd_find_matches)
 
 
+def _add_strip_bom_parser(subparsers) -> None:
+    parser = subparsers.add_parser(
+        "strip-bom",
+        help="Remove a leading UTF-8 BOM from a CSV file.",
+    )
+    parser.add_argument("input", help="Input CSV file.")
+    parser.add_argument(
+        "-o",
+        "--output",
+        help="Output CSV file (default: <input-stem>.no-bom.csv).",
+    )
+    parser.add_argument(
+        "--in-place",
+        action="store_true",
+        help="Rewrite the input file instead of writing a separate output file.",
+    )
+    parser.set_defaults(func=cmd_strip_bom)
+
+
 def _add_strict_concat_parser(subparsers) -> None:
     parser = subparsers.add_parser(
         "strict-concat",
@@ -664,6 +696,7 @@ def build_parser() -> argparse.ArgumentParser:
 
     _add_row_duplicates_parser(subparsers)
     _add_find_matches_parser(subparsers)
+    _add_strip_bom_parser(subparsers)
     _add_strict_concat_parser(subparsers)
     _add_concentrate_parser(subparsers)
     _add_rehydrate_parser(subparsers)

csvsmith-0.12.0/src/csvsmith/tools/strip_bom.py

@@ -0,0 +1,52 @@
+from __future__ import annotations
+
+from dataclasses import dataclass
+from pathlib import Path
+
+
+UTF8_BOM = b"\xef\xbb\xbf"
+
+
+@dataclass(frozen=True)
+class StripBomResult:
+    input_path: Path
+    output_path: Path
+    removed: bool
+
+
+def default_output_path(csv_path: str | Path) -> Path:
+    """Return the default output path for a BOM-stripped CSV."""
+    path = Path(csv_path)
+    return path.with_name(f"{path.stem}.no-bom{path.suffix}")
+
+
+def strip_utf8_bom(
+    csv_path: str | Path,
+    output_path: str | Path | None = None,
+    *,
+    in_place: bool = False,
+) -> StripBomResult:
+    """Remove a leading UTF-8 BOM from a CSV file while preserving all other bytes."""
+    input_path = Path(csv_path)
+    if in_place and output_path is not None:
+        raise ValueError("--in-place cannot be used with --output")
+
+    if in_place:
+        resolved_output_path = input_path
+    elif output_path:
+        resolved_output_path = Path(output_path)
+    else:
+        resolved_output_path = default_output_path(input_path)
+
+    data = input_path.read_bytes()
+    removed = data.startswith(UTF8_BOM)
+    output_data = data[len(UTF8_BOM) :] if removed else data
+
+    resolved_output_path.parent.mkdir(parents=True, exist_ok=True)
+    resolved_output_path.write_bytes(output_data)
+
+    return StripBomResult(
+        input_path=input_path,
+        output_path=resolved_output_path,
+        removed=removed,
+    )

{csvsmith-0.11.1 → csvsmith-0.12.0/src/csvsmith.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: csvsmith
-Version: 0.11.1
+Version: 0.12.0
 Summary: Small CSV utilities: row deduplication, classification, row filtering, and CLI helpers.
 Author-email: Eiichi YAMAMOTO <info@yeiichi.com>
 License: MIT License

{csvsmith-0.11.1 → csvsmith-0.12.0}/src/csvsmith.egg-info/SOURCES.txt

@@ -21,6 +21,7 @@ src/csvsmith/tools/move_files.py
 src/csvsmith/tools/row_dedup.py
 src/csvsmith/tools/sample_csv.py
 src/csvsmith/tools/strict_concat.py
+src/csvsmith/tools/strip_bom.py
 src/csvsmith/utils/__init__.py
 src/csvsmith/utils/clean_numeric.py
 src/csvsmith/utils/distance.py
@@ -40,4 +41,5 @@ tests/test_normalize.py
 tests/test_row_dedup.py
 tests/test_sample_csv.py
 tests/test_strict_concat.py
-tests/test_string_distance.py
+tests/test_string_distance.py
+tests/test_strip_bom.py

{csvsmith-0.11.1 → csvsmith-0.12.0}/tests/test_cli.py

@@ -190,6 +190,26 @@ def test_cli_parses_find_matches_command():
     assert args.no_nfkc is True
 
 
+def test_cli_parses_strip_bom_command():
+    parser = build_parser()
+    args = parser.parse_args(["strip-bom", "input.csv", "-o", "output.csv"])
+
+    assert args.command == "strip-bom"
+    assert args.input == "input.csv"
+    assert args.output == "output.csv"
+    assert args.in_place is False
+
+
+def test_cli_parses_strip_bom_command_in_place():
+    parser = build_parser()
+    args = parser.parse_args(["strip-bom", "input.csv", "--in-place"])
+
+    assert args.command == "strip-bom"
+    assert args.input == "input.csv"
+    assert args.output is None
+    assert args.in_place is True
+
+
 def test_cli_parses_strict_concat_command():
     parser = build_parser()
     args = parser.parse_args(["strict-concat", "input_dir", "-o", "output.csv"])

csvsmith-0.12.0/tests/test_strip_bom.py

@@ -0,0 +1,63 @@
+import pytest
+
+from csvsmith.cli import main
+from csvsmith.tools.strip_bom import UTF8_BOM, default_output_path, strip_utf8_bom
+
+
+def test_strip_utf8_bom_writes_default_output_without_bom(tmp_path):
+    source = tmp_path / "input.csv"
+    source.write_bytes(UTF8_BOM + b"id,name\r\n1,Alice\r\n")
+
+    result = strip_utf8_bom(source)
+
+    assert result.output_path == tmp_path / "input.no-bom.csv"
+    assert result.removed is True
+    assert result.output_path.read_bytes() == b"id,name\r\n1,Alice\r\n"
+    assert source.read_bytes().startswith(UTF8_BOM)
+
+
+def test_strip_utf8_bom_preserves_file_without_bom(tmp_path):
+    source = tmp_path / "input.csv"
+    output = tmp_path / "output.csv"
+    source.write_bytes(b"id,name\n1,Alice\n")
+
+    result = strip_utf8_bom(source, output)
+
+    assert result.output_path == output
+    assert result.removed is False
+    assert output.read_bytes() == b"id,name\n1,Alice\n"
+
+
+def test_strip_utf8_bom_can_rewrite_in_place(tmp_path):
+    source = tmp_path / "input.csv"
+    source.write_bytes(UTF8_BOM + b"id,name\n1,Alice\n")
+
+    result = strip_utf8_bom(source, in_place=True)
+
+    assert result.output_path == source
+    assert result.removed is True
+    assert source.read_bytes() == b"id,name\n1,Alice\n"
+
+
+def test_strip_utf8_bom_rejects_in_place_with_output(tmp_path):
+    source = tmp_path / "input.csv"
+    source.write_bytes(UTF8_BOM + b"id,name\n")
+
+    with pytest.raises(ValueError, match="--in-place"):
+        strip_utf8_bom(source, tmp_path / "output.csv", in_place=True)
+
+
+def test_default_output_path_handles_multi_dot_names():
+    assert default_output_path("report.final.csv").name == "report.final.no-bom.csv"
+
+
+def test_strip_bom_cli_writes_requested_output(tmp_path, capsys):
+    source = tmp_path / "input.csv"
+    output = tmp_path / "clean.csv"
+    source.write_bytes(UTF8_BOM + b"id,name\n1,Alice\n")
+
+    exit_code = main(["strip-bom", str(source), "-o", str(output)])
+
+    assert exit_code == 0
+    assert output.read_bytes() == b"id,name\n1,Alice\n"
+    assert "removed BOM" in capsys.readouterr().out