csvsmith 0.4.0__tar.gz → 0.6.0__tar.gz

{csvsmith-0.4.0/src/csvsmith.egg-info → csvsmith-0.6.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: csvsmith
-Version: 0.4.0
+Version: 0.6.0
 Summary: Small CSV utilities: row deduplication, classification, row filtering, and CLI helpers.
 Author-email: Eiichi YAMAMOTO <info@yeiichi.com>
 License: MIT License
@@ -50,7 +50,7 @@ csvsmith
    :target: https://pypi.org/project/csvsmith/
 
 .. image:: https://img.shields.io/pypi/l/csvsmith.svg
-   :target: https://pypi.org/project/ccsvsmith/
+   :target: https://pypi.org/project/csvsmith/
 
 Introduction
 ------------
@@ -77,6 +77,7 @@ Features
 - rollback support via manifest
 - row filtering by substring
 - Excel worksheet to CSV conversion
+- numeric string cleaning and normalization
 - file moving by suffix
 - string distance and similarity analysis
 - a single command-line entrypoint with subcommands
@@ -153,7 +154,26 @@ Analyze string distance
    print(result.jaro_winkler_score)
    print(result.similarity_percentage)
 
-CLI Usage
+Clean numeric values
+~~~~~~~~~~~~~~~~~~~~
+
+.. code-block:: python
+
+   from csvsmith import clean_numeric
+
+   # Handles separators and parentheses for negative numbers
+   val1 = clean_numeric("1,234.56")
+   val2 = clean_numeric("(100.00)")
+   val3 = clean_numeric("1 234,56", sep=" ", decimal=",")
+
+   # Use relaxed mode to return original value on failure instead of raising ValueError
+   val4 = clean_numeric("not a number", relaxed=True)
+
+   print(val1)  # 1234.56
+   print(val2)  # -100.0
+   print(val3)  # 1234.56
+   print(val4)  # "not a number"
+
 Drop rows in a CSV by column name
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
@@ -203,7 +223,7 @@ Move files by suffix
    print(f"Moved {moved_count} files.")
 
 CSV File Classification (Python)
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
 .. code-block:: python
 
@@ -315,6 +335,25 @@ Move files by suffix
 This moves files whose suffix matches one of the given values. The suffixes can
 be written with or without a leading dot, and matching is case-insensitive.
 
+Clean numeric values
+~~~~~~~~~~~~~~~~~~~~
+
+.. code-block:: bash
+
+   csvsmith clean-numeric "1,234.56"
+
+Using custom separators:
+
+.. code-block:: bash
+
+   csvsmith clean-numeric "1 234,56" --sep " " --decimal ","
+
+Relaxed mode (returns original string on failure):
+
+.. code-block:: bash
+
+   csvsmith clean-numeric "not a number" --relaxed
+
 Drop CSV rows
 ~~~~~~~~~~~~~
 
@@ -379,7 +418,7 @@ Example
 
 Suppose you have a CSV like this:
 
-.. code-block:: csv
+.. code-block:: text
 
    id,name,notes
    1,Alice,ok
@@ -394,7 +433,7 @@ Running:
 
 produces a filtered file containing:
 
-.. code-block:: csv
+.. code-block:: text
 
    id,name,notes
    1,Alice,ok

{csvsmith-0.4.0 → csvsmith-0.6.0}/README.rst

@@ -8,7 +8,7 @@ csvsmith
    :target: https://pypi.org/project/csvsmith/
 
 .. image:: https://img.shields.io/pypi/l/csvsmith.svg
-   :target: https://pypi.org/project/ccsvsmith/
+   :target: https://pypi.org/project/csvsmith/
 
 Introduction
 ------------
@@ -35,6 +35,7 @@ Features
 - rollback support via manifest
 - row filtering by substring
 - Excel worksheet to CSV conversion
+- numeric string cleaning and normalization
 - file moving by suffix
 - string distance and similarity analysis
 - a single command-line entrypoint with subcommands
@@ -111,7 +112,26 @@ Analyze string distance
    print(result.jaro_winkler_score)
    print(result.similarity_percentage)
 
-CLI Usage
+Clean numeric values
+~~~~~~~~~~~~~~~~~~~~
+
+.. code-block:: python
+
+   from csvsmith import clean_numeric
+
+   # Handles separators and parentheses for negative numbers
+   val1 = clean_numeric("1,234.56")
+   val2 = clean_numeric("(100.00)")
+   val3 = clean_numeric("1 234,56", sep=" ", decimal=",")
+
+   # Use relaxed mode to return original value on failure instead of raising ValueError
+   val4 = clean_numeric("not a number", relaxed=True)
+
+   print(val1)  # 1234.56
+   print(val2)  # -100.0
+   print(val3)  # 1234.56
+   print(val4)  # "not a number"
+
 Drop rows in a CSV by column name
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
@@ -161,7 +181,7 @@ Move files by suffix
    print(f"Moved {moved_count} files.")
 
 CSV File Classification (Python)
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
 .. code-block:: python
 
@@ -273,6 +293,25 @@ Move files by suffix
 This moves files whose suffix matches one of the given values. The suffixes can
 be written with or without a leading dot, and matching is case-insensitive.
 
+Clean numeric values
+~~~~~~~~~~~~~~~~~~~~
+
+.. code-block:: bash
+
+   csvsmith clean-numeric "1,234.56"
+
+Using custom separators:
+
+.. code-block:: bash
+
+   csvsmith clean-numeric "1 234,56" --sep " " --decimal ","
+
+Relaxed mode (returns original string on failure):
+
+.. code-block:: bash
+
+   csvsmith clean-numeric "not a number" --relaxed
+
 Drop CSV rows
 ~~~~~~~~~~~~~
 
@@ -337,7 +376,7 @@ Example
 
 Suppose you have a CSV like this:
 
-.. code-block:: csv
+.. code-block:: text
 
    id,name,notes
    1,Alice,ok
@@ -352,7 +391,7 @@ Running:
 
 produces a filtered file containing:
 
-.. code-block:: csv
+.. code-block:: text
 
    id,name,notes
    1,Alice,ok

{csvsmith-0.4.0 → csvsmith-0.6.0}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "csvsmith"
-version = "0.4.0"
+version = "0.6.0"
 description = "Small CSV utilities: row deduplication, classification, row filtering, and CLI helpers."
 readme = "README.rst"
 requires-python = ">=3.10"

{csvsmith-0.4.0 → csvsmith-0.6.0}/src/csvsmith/__init__.py

@@ -21,6 +21,7 @@ Compatibility aliases:
 - CSVCleaner
 
 Submodules:
+- csvsmith.clean_numeric
 - csvsmith.string_distance
 - csvsmith.row_dedup
 - csvsmith.classify
@@ -30,8 +31,9 @@ Submodules:
 - csvsmith.cli (CLI entrypoint)
 """
 
-__version__ = "0.4.0"
+__version__ = "0.6.0"
 
+from .clean_numeric import clean_numeric
 from .row_dedup import (
     count_duplicates_sorted,
     add_row_digest,
@@ -61,4 +63,5 @@ __all__ = [
     "Relation",
     "Result",
     "analyze_pair",
+    "clean_numeric",
 ]

csvsmith-0.6.0/src/csvsmith/clean_numeric.py

@@ -0,0 +1,124 @@
+import re
+from typing import Any
+
+NON_BREAKING_SPACE = "\xa0"
+SEPARATOR_PATTERN = re.compile(r"[ _\xa0]")
+NUMBER_PATTERN = re.compile(r"^-?(?:\d+|\d*\.\d+)$")
+
+
+def _normalize_numeric_text(value: Any, *, sep: str, decimal: str) -> str:
+    """
+    Normalize a numeric text string for consistent formatting.
+
+    Converts a given value to a string representation and ensures normalization of numeric formatting,
+    such as removing group separators, converting localized decimal separators, and handling negative
+    values enclosed in parentheses.
+
+    :param value: The value to be normalized, which may be of any type.
+    :type value: Any
+    :param sep: The character used as a group separator in the input value, which will be removed
+        during normalization.
+    :type sep: str
+    :param decimal: The character used as the decimal separator in the input value, which will
+        be replaced with a standard period ('.') during normalization.
+    :type decimal: str
+    :return: A normalized numeric string with consistent formatting.
+    :rtype: str
+    """
+    numeric_text = str(value).strip()
+
+    if numeric_text.startswith("(") and numeric_text.endswith(")"):
+        numeric_text = f"-{numeric_text[1:-1]}"
+
+    if sep:
+        numeric_text = numeric_text.replace(sep, "")
+
+    if decimal != ".":
+        numeric_text = numeric_text.replace(decimal, ".")
+
+    return numeric_text
+
+
+def _has_valid_grouping(numeric_text: str, *, decimal: str) -> bool:
+    """
+    Checks whether a numeric text string has valid grouping based on a specified decimal character.
+
+    This function validates the structure of the given numeric text to determine if it adheres to allowed
+    grouping conventions. It ensures the string does not contain invalid or misplaced group separators,
+    decimal points, or spacing characters.
+
+    :param numeric_text: A string representing the numeric text to be validated.
+    :type numeric_text: str
+    :param decimal: A string representing the character used as the decimal point.
+    :type decimal: str
+    :return: True if the numeric text satisfies the grouping rules; otherwise, False.
+    :rtype: bool
+    """
+    if not numeric_text:
+        return False
+
+    unsigned_text = numeric_text[1:] if numeric_text.startswith("-") else numeric_text
+
+    if unsigned_text.count(decimal) > 1:
+        return False
+
+    integer_text, _, fraction_text = unsigned_text.partition(decimal)
+
+    if not integer_text and not fraction_text:
+        return False
+
+    for part in (integer_text, fraction_text):
+        if not part:
+            continue
+        if part.startswith("_") or part.endswith("_"):
+            return False
+        if part.startswith(" ") or part.endswith(" "):
+            return False
+        if part.startswith(NON_BREAKING_SPACE) or part.endswith(NON_BREAKING_SPACE):
+            return False
+        if "__" in part or "  " in part or NON_BREAKING_SPACE * 2 in part:
+            return False
+
+    stripped_text = SEPARATOR_PATTERN.sub("", numeric_text)
+    return bool(NUMBER_PATTERN.fullmatch(stripped_text))
+
+
+def clean_numeric(
+    value: Any, *, sep: str = ",", decimal: str = ".", relaxed: bool = False
+) -> float | Any:
+    """
+    Cleans and converts a given input to a float by normalizing its numeric representation.
+    Handles separators and decimal points based on the provided arguments. If the input
+    value is invalid or cannot be converted, a ValueError is raised unless relaxed mode
+    is enabled.
+
+    :param value: The input value to be cleaned and converted.
+    :type value: Any
+    :param sep: The character used as a thousands separator in the input value. Default is ",".
+    :type sep: str
+    :param decimal: The character used as a decimal point in the input value. Default is ".".
+    :type decimal: str
+    :param relaxed: If True, return the original input when it is not numeric.
+    :type relaxed: bool
+    :return: The cleaned and converted numeric value as a float, or the original value in relaxed mode.
+    :rtype: float | Any
+    :raises ValueError: If the input value cannot be converted to a valid number and relaxed is False.
+    """
+    if value is None:
+        return 0.0
+
+    normalized_number_text = _normalize_numeric_text(value, sep=sep, decimal=decimal)
+
+    if not _has_valid_grouping(normalized_number_text, decimal=decimal):
+        if relaxed:
+            return value
+        raise ValueError(f"Could not convert {value!r} to a valid number.")
+
+    numeric_text = SEPARATOR_PATTERN.sub("", normalized_number_text)
+
+    try:
+        return float(numeric_text)
+    except ValueError as exc:
+        if relaxed:
+            return value
+        raise ValueError(f"Could not convert {value!r} to a valid number.") from exc

{csvsmith-0.4.0 → csvsmith-0.6.0}/src/csvsmith/cli.py

@@ -9,6 +9,7 @@ from typing import Optional, Sequence
 from . import __version__
 from .classify import CSVClassifier
 from .excel2csv import excel_to_csv
+from .clean_numeric import clean_numeric
 from .filter_rows import DropRowsBySubstring
 from .move_files import move_by_suffix
 from .row_dedup import (
@@ -148,6 +149,21 @@ def cmd_string_distance(args: argparse.Namespace) -> int:
     return 0
 
 
+def cmd_clean_numeric(args: argparse.Namespace) -> int:
+    try:
+        cleaned = clean_numeric(
+            args.value,
+            sep=args.sep,
+            decimal=args.decimal,
+            relaxed=args.relaxed,
+        )
+        print(cleaned)
+    except ValueError as e:
+        print(f"Error: {e}", file=sys.stderr)
+        return 1
+    return 0
+
+
 def _add_row_duplicates_parser(subparsers) -> None:
     parser = subparsers.add_parser("row-duplicates", help="Find duplicate rows in a CSV.")
     parser.add_argument("input", help="Input CSV file.")
@@ -230,6 +246,21 @@ def _add_string_distance_parser(subparsers) -> None:
     parser.set_defaults(func=cmd_string_distance)
 
 
+def _add_clean_numeric_parser(subparsers) -> None:
+    parser = subparsers.add_parser(
+        "clean-numeric", help="Clean and convert a numeric string to float."
+    )
+    parser.add_argument("value", help="Numeric value to clean.")
+    parser.add_argument("--sep", default=",", help="Group separator (default: ,).")
+    parser.add_argument("--decimal", default=".", help="Decimal separator (default: .).")
+    parser.add_argument(
+        "--relaxed",
+        action="store_true",
+        help="Return the original input when it is not numeric.",
+    )
+    parser.set_defaults(func=cmd_clean_numeric)
+
+
 def build_parser() -> argparse.ArgumentParser:
     parser = argparse.ArgumentParser(
         prog="csvsmith",
@@ -249,6 +280,7 @@ def build_parser() -> argparse.ArgumentParser:
     _add_excel_to_csv_parser(subparsers)
     _add_drop_rows_parser(subparsers)
     _add_string_distance_parser(subparsers)
+    _add_clean_numeric_parser(subparsers)
 
     return parser
 

{csvsmith-0.4.0 → csvsmith-0.6.0}/src/csvsmith/string_distance.py

@@ -75,7 +75,13 @@ class StringDistance:
 
     @staticmethod
     def strip_all(s: str) -> str:
-        return "".join(ch for ch in s if not ch.isspace())
+        """
+        Remove all whitespace characters from a string.
+
+        Uses split/join logic so any whitespace character acts as a separator,
+        including spaces, tabs, and newlines.
+        """
+        return "".join(s.split())
 
     @staticmethod
     def equals_ignore_case(a: str, b: str) -> bool:

{csvsmith-0.4.0 → csvsmith-0.6.0/src/csvsmith.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: csvsmith
-Version: 0.4.0
+Version: 0.6.0
 Summary: Small CSV utilities: row deduplication, classification, row filtering, and CLI helpers.
 Author-email: Eiichi YAMAMOTO <info@yeiichi.com>
 License: MIT License
@@ -50,7 +50,7 @@ csvsmith
    :target: https://pypi.org/project/csvsmith/
 
 .. image:: https://img.shields.io/pypi/l/csvsmith.svg
-   :target: https://pypi.org/project/ccsvsmith/
+   :target: https://pypi.org/project/csvsmith/
 
 Introduction
 ------------
@@ -77,6 +77,7 @@ Features
 - rollback support via manifest
 - row filtering by substring
 - Excel worksheet to CSV conversion
+- numeric string cleaning and normalization
 - file moving by suffix
 - string distance and similarity analysis
 - a single command-line entrypoint with subcommands
@@ -153,7 +154,26 @@ Analyze string distance
    print(result.jaro_winkler_score)
    print(result.similarity_percentage)
 
-CLI Usage
+Clean numeric values
+~~~~~~~~~~~~~~~~~~~~
+
+.. code-block:: python
+
+   from csvsmith import clean_numeric
+
+   # Handles separators and parentheses for negative numbers
+   val1 = clean_numeric("1,234.56")
+   val2 = clean_numeric("(100.00)")
+   val3 = clean_numeric("1 234,56", sep=" ", decimal=",")
+
+   # Use relaxed mode to return original value on failure instead of raising ValueError
+   val4 = clean_numeric("not a number", relaxed=True)
+
+   print(val1)  # 1234.56
+   print(val2)  # -100.0
+   print(val3)  # 1234.56
+   print(val4)  # "not a number"
+
 Drop rows in a CSV by column name
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
@@ -203,7 +223,7 @@ Move files by suffix
    print(f"Moved {moved_count} files.")
 
 CSV File Classification (Python)
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
 .. code-block:: python
 
@@ -315,6 +335,25 @@ Move files by suffix
 This moves files whose suffix matches one of the given values. The suffixes can
 be written with or without a leading dot, and matching is case-insensitive.
 
+Clean numeric values
+~~~~~~~~~~~~~~~~~~~~
+
+.. code-block:: bash
+
+   csvsmith clean-numeric "1,234.56"
+
+Using custom separators:
+
+.. code-block:: bash
+
+   csvsmith clean-numeric "1 234,56" --sep " " --decimal ","
+
+Relaxed mode (returns original string on failure):
+
+.. code-block:: bash
+
+   csvsmith clean-numeric "not a number" --relaxed
+
 Drop CSV rows
 ~~~~~~~~~~~~~
 
@@ -379,7 +418,7 @@ Example
 
 Suppose you have a CSV like this:
 
-.. code-block:: csv
+.. code-block:: text
 
    id,name,notes
    1,Alice,ok
@@ -394,7 +433,7 @@ Running:
 
 produces a filtered file containing:
 
-.. code-block:: csv
+.. code-block:: text
 
    id,name,notes
    1,Alice,ok

{csvsmith-0.4.0 → csvsmith-0.6.0}/src/csvsmith.egg-info/SOURCES.txt

@@ -3,6 +3,7 @@ README.rst
 pyproject.toml
 src/csvsmith/__init__.py
 src/csvsmith/classify.py
+src/csvsmith/clean_numeric.py
 src/csvsmith/cli.py
 src/csvsmith/excel2csv.py
 src/csvsmith/filter_rows.py
@@ -16,6 +17,7 @@ src/csvsmith.egg-info/entry_points.txt
 src/csvsmith.egg-info/requires.txt
 src/csvsmith.egg-info/top_level.txt
 tests/test_classify.py
+tests/test_clean_numeric.py
 tests/test_cli.py
 tests/test_excel2csv.py
 tests/test_filter_rows.py

csvsmith-0.6.0/tests/test_clean_numeric.py

@@ -0,0 +1,36 @@
+import pytest
+from src.csvsmith.clean_numeric import clean_numeric
+
+
+def test_clean_numeric_with_valid_integer_string() -> None:
+    assert clean_numeric("1,234") == 1234.0
+
+
+def test_clean_numeric_with_valid_float_string() -> None:
+    assert clean_numeric("1,234.56") == 1234.56
+
+
+def test_clean_numeric_with_custom_separator() -> None:
+    assert clean_numeric("1.234,56", sep=".", decimal=",") == 1234.56
+
+
+def test_clean_numeric_with_negative_value() -> None:
+    assert clean_numeric("-1,234.56") == -1234.56
+
+
+def test_clean_numeric_with_parenthesis_for_negative() -> None:
+    assert clean_numeric("(1,234.56)") == -1234.56
+
+
+def test_clean_numeric_with_invalid_value_raises_valueerror() -> None:
+    with pytest.raises(ValueError, match="Could not convert 'abc'.*"):
+        clean_numeric("abc")
+
+
+def test_clean_numeric_with_multiple_decimal_points_raises_valueerror() -> None:
+    with pytest.raises(ValueError, match="Could not convert '1,234.5.6'.*"):
+        clean_numeric("1,234.5.6")
+
+
+def test_clean_numeric_with_none_value() -> None:
+    assert clean_numeric(None) == 0.0

{csvsmith-0.4.0 → csvsmith-0.6.0}/tests/test_cli.py

@@ -126,3 +126,34 @@ def test_cli_parses_move_files_command_with_suffixes_without_dots():
     assert args.source == "src_dir"
     assert args.dest == "dst_dir"
     assert args.suffixes == "csv,pdf"
+
+
+def test_cli_parses_clean_numeric_command():
+    parser = build_parser()
+    args = parser.parse_args(["clean-numeric", "1,234.56", "--sep", ",", "--decimal", "."])
+
+    assert args.command == "clean-numeric"
+    assert args.value == "1,234.56"
+    assert args.sep == ","
+    assert args.decimal == "."
+
+
+def test_cli_parses_clean_numeric_command_with_relaxed_mode():
+    parser = build_parser()
+    args = parser.parse_args(
+        ["clean-numeric", "not-a-number", "--relaxed"]
+    )
+
+    assert args.command == "clean-numeric"
+    assert args.value == "not-a-number"
+    assert args.relaxed is True
+
+
+def test_cli_parses_string_distance_command():
+    parser = build_parser()
+    args = parser.parse_args(["string-distance", "a", "b", "--ignore-case"])
+
+    assert args.command == "string-distance"
+    assert args.string_a == "a"
+    assert args.string_b == "b"
+    assert args.ignore_case is True