format-docstring 0.2.6__tar.gz → 0.2.7__tar.gz

{format_docstring-0.2.6 → format_docstring-0.2.7}/CHANGELOG.md

@@ -6,6 +6,13 @@ The format is based on
 [Keep a Changelog](https://keepachangelog.com/en/1.1.0/), and this project
 adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## [0.2.7] - 2025-12-04
+
+- Fixed
+  - Bug where non-ASCII characters would mess up auto formatting
+- Full diff
+  - https://github.com/jsh9/format-docstring/compare/0.2.6...0.2.7
+
 ## [0.2.6] - 2025-11-21
 
 - Fixed

{format_docstring-0.2.6 → format_docstring-0.2.7}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: format-docstring
-Version: 0.2.6
+Version: 0.2.7
 Summary: A Python formatter to wrap/adjust docstring lines
 Author-email: jsh9 <25124332+jsh9@users.noreply.github.com>
 Maintainer-email: jsh9 <25124332+jsh9@users.noreply.github.com>

{format_docstring-0.2.6 → format_docstring-0.2.7}/format_docstring/docstring_rewriter.py

@@ -427,8 +427,23 @@ def build_replacement_docstring(
     if not hasattr(val, 'lineno') or not hasattr(val, 'end_lineno'):
         return None
 
-    start: int = calc_abs_pos(line_starts, val.lineno, val.col_offset)
-    end: int = calc_abs_pos(line_starts, val.end_lineno, val.end_col_offset)  # type: ignore[arg-type]
+    # ``end_lineno``/``end_col_offset`` are optional on older AST nodes or
+    # when running under tooling that strips positional info, so bail out if
+    # they are missing to avoid slicing with ``None`` later.
+    end_lineno: int | None = getattr(val, 'end_lineno', None)
+    end_col_offset: int | None = getattr(val, 'end_col_offset', None)
+    if end_lineno is None or end_col_offset is None:
+        return None
+
+    start: int = calc_abs_pos(
+        source_code, line_starts, val.lineno, val.col_offset
+    )
+    end: int = calc_abs_pos(
+        source_code,
+        line_starts,
+        end_lineno,
+        end_col_offset,
+    )
     original_literal = source_code[start:end]
 
     if _has_inline_no_format_comment(source_code, end):
@@ -524,12 +539,19 @@ def find_docstring(node: ModuleClassOrFunc) -> ast.Expr | None:
     return None
 
 
-def calc_abs_pos(line_starts: list[int], lineno: int, col: int) -> int:
+def calc_abs_pos(
+        source_code: str, line_starts: list[int], lineno: int, col: int
+) -> int:
     """
     Convert a (lineno, col) pair to an absolute index.
 
     Parameters
     ----------
+    source_code : str
+        Full source text for computing character offsets. AST column offsets
+        are byte-based, so we need the actual text to translate them back to
+        character indices when multi-byte Unicode code points (e.g., 😄, é, 文)
+        are present.
     line_starts : list[int]
         Precomputed start offsets for each line, from :func:`_line_starts`.
     lineno : int
@@ -542,7 +564,29 @@ def calc_abs_pos(line_starts: list[int], lineno: int, col: int) -> int:
     int
         The absolute character index into the source string.
     """
-    return line_starts[lineno - 1] + col
+    line_idx = lineno - 1
+    line_start = line_starts[line_idx]
+    next_line_start = (
+        line_starts[line_idx + 1]
+        if line_idx + 1 < len(line_starts)
+        else len(source_code)
+    )
+    line_segment = source_code[line_start:next_line_start]
+
+    # Column offsets from the AST are measured in bytes, so convert them back
+    # to character offsets when slicing the original ``str`` source. Iterate
+    # through the current line until reaching the requested byte position.
+    byte_count = 0
+    char_offset = 0
+    for char in line_segment:
+        if byte_count >= col:
+            break
+
+        byte_count += len(char.encode('utf-8'))
+        char_offset += 1
+
+    # Clamp to the line length in case the reported byte offset overshoots.
+    return line_start + min(char_offset, len(line_segment))
 
 
 def rebuild_literal(original_literal: str, content: str) -> str | None:

{format_docstring-0.2.6 → format_docstring-0.2.7}/format_docstring.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: format-docstring
-Version: 0.2.6
+Version: 0.2.7
 Summary: A Python formatter to wrap/adjust docstring lines
 Author-email: jsh9 <25124332+jsh9@users.noreply.github.com>
 Maintainer-email: jsh9 <25124332+jsh9@users.noreply.github.com>

{format_docstring-0.2.6 → format_docstring-0.2.7}/format_docstring.egg-info/SOURCES.txt

@@ -58,6 +58,7 @@ tests/test_data/end_to_end/numpy/mismatched_underlines_two_dashes.txt
 tests/test_data/end_to_end/numpy/module_level_docstring.txt
 tests/test_data/end_to_end/numpy/new_lines_before_and_after.txt
 tests/test_data/end_to_end/numpy/no_format_docstring_comment.txt
+tests/test_data/end_to_end/numpy/non_ascii_docstrings.txt
 tests/test_data/end_to_end/numpy/param_signature_without_type.txt
 tests/test_data/end_to_end/numpy/parameters_returns_raises_wrapping.txt
 tests/test_data/end_to_end/numpy/rST_cross_reference.txt
@@ -98,6 +99,7 @@ tests/test_data/line_wrap/numpy/mismatched_underlines.txt
 tests/test_data/line_wrap/numpy/mismatched_underlines_one_dash.txt
 tests/test_data/line_wrap/numpy/mismatched_underlines_two_dashes.txt
 tests/test_data/line_wrap/numpy/module_level_docstring.txt
+tests/test_data/line_wrap/numpy/non_ascii_docstrings.txt
 tests/test_data/line_wrap/numpy/param_signature_without_type.txt
 tests/test_data/line_wrap/numpy/parameters_returns_raises_wrapping.txt
 tests/test_data/line_wrap/numpy/returns_signature_and_description.txt

{format_docstring-0.2.6 → format_docstring-0.2.7}/pyproject.toml

@@ -4,7 +4,7 @@ requires = ["setuptools-scm[toml]>=6.2", "setuptools>=45"]
 
 [project]
 name = "format-docstring"
-version = "0.2.6"
+version = "0.2.7"
 dependencies = [
   "click>=8.0",
   "jupyter-notebook-parser>=0.1.4",

format_docstring-0.2.7/tests/test_data/end_to_end/numpy/non_ascii_docstrings.txt

@@ -0,0 +1,89 @@
+LINE_LENGTH: 79
+
+**********
+
+# Non-ASCII characters should not interfere with line length calculation.
+
+def unicode_docstring_pythagoras() -> None:
+    """a² + b² = c²"""
+    return None
+
+def unicode_docstring_quadratic() -> None:
+    """Δ = √(b² - 4ac)"""
+    return None
+
+def unicode_docstring_emoji() -> None:
+    """emoji 😄🚀 with sparkle ✨"""
+    return None
+
+def unicode_docstring_accents() -> None:
+    """accents naïve façade jalapeño año"""
+    return None
+
+def unicode_docstring_greek() -> None:
+    """Greek letters αβγδεζηθ"""
+    return None
+
+def unicode_docstring_cyrillic() -> None:
+    """Cyrillic Привет мир"""
+    return None
+
+def unicode_docstring_cjk() -> None:
+    """CJK 面積は平方メートルです"""
+    return None
+
+def unicode_docstring_pi_tau() -> None:
+    """Math π·r² + τ is about 6.283"""
+    return None
+
+def unicode_docstring_currency() -> None:
+    """Currency mix € £ ¥ ₽"""
+    return None
+
+def unicode_docstring_hebrew() -> None:
+    """Hebrew שלום עליכם"""
+    return None
+
+**********
+
+# Non-ASCII characters should not interfere with line length calculation.
+
+def unicode_docstring_pythagoras() -> None:
+    """a² + b² = c²"""
+    return None
+
+def unicode_docstring_quadratic() -> None:
+    """Δ = √(b² - 4ac)"""
+    return None
+
+def unicode_docstring_emoji() -> None:
+    """emoji 😄🚀 with sparkle ✨"""
+    return None
+
+def unicode_docstring_accents() -> None:
+    """accents naïve façade jalapeño año"""
+    return None
+
+def unicode_docstring_greek() -> None:
+    """Greek letters αβγδεζηθ"""
+    return None
+
+def unicode_docstring_cyrillic() -> None:
+    """Cyrillic Привет мир"""
+    return None
+
+def unicode_docstring_cjk() -> None:
+    """CJK 面積は平方メートルです"""
+    return None
+
+def unicode_docstring_pi_tau() -> None:
+    """Math π·r² + τ is about 6.283"""
+    return None
+
+def unicode_docstring_currency() -> None:
+    """Currency mix € £ ¥ ₽"""
+    return None
+
+def unicode_docstring_hebrew() -> None:
+    """Hebrew שלום עליכם"""
+    return None

format_docstring-0.2.7/tests/test_data/line_wrap/numpy/non_ascii_docstrings.txt

@@ -0,0 +1,89 @@
+LINE_LENGTH: 79
+
+**********
+
+Non-ASCII characters should not interfere with line length calculation.
+
+Function f_square
+-----------------
+a² + b² = c²
+
+Function f_delta
+----------------
+Δ = √(b² - 4ac)
+
+Function f_emoji
+----------------
+emoji 😄🚀 with sparkle ✨
+
+Function f_accents
+------------------
+accents naïve façade jalapeño año
+
+Function f_greek
+----------------
+Greek letters αβγδεζηθ
+
+Function f_cyrillic
+-------------------
+Cyrillic Привет мир
+
+Function f_cjk
+---------------
+CJK 面積は平方メートルです
+
+Function f_pi_tau
+-----------------
+Math π·r² + τ is about 6.283
+
+Function f_currency
+-------------------
+Currency mix € £ ¥ ₽
+
+Function f_hebrew
+-----------------
+Hebrew שלום עליכם
+
+**********
+
+Non-ASCII characters should not interfere with line length calculation.
+
+Function f_square
+-----------------
+a² + b² = c²
+
+Function f_delta
+----------------
+Δ = √(b² - 4ac)
+
+Function f_emoji
+----------------
+emoji 😄🚀 with sparkle ✨
+
+Function f_accents
+------------------
+accents naïve façade jalapeño año
+
+Function f_greek
+----------------
+Greek letters αβγδεζηθ
+
+Function f_cyrillic
+-------------------
+Cyrillic Привет мир
+
+Function f_cjk
+---------------
+CJK 面積は平方メートルです
+
+Function f_pi_tau
+-----------------
+Math π·r² + τ is about 6.283
+
+Function f_currency
+-------------------
+Currency mix € £ ¥ ₽
+
+Function f_hebrew
+-----------------
+Hebrew שלום עליכם

{format_docstring-0.2.6 → format_docstring-0.2.7}/tests/test_docstring_rewriter.py

@@ -28,12 +28,20 @@ def test_calc_line_starts(src: str, expected: list[int]) -> None:
         ('a\n\nxyz\n', 2, 0, 2),
         ('a\n\nxyz\n', 3, 2, 5),
         ('one\ntwo\nthree', 2, 1, 5),
+        (
+            'def f():\n    """π"""\n    pass\n',
+            2,
+            len('    """π"""'.encode()),
+            len('def f():\n') + len('    """π"""'),
+        ),
     ],
 )
 def test_calc_abs_pos(src: str, lineno: int, col: int, expected: int) -> None:
     """Convert (lineno, col) to absolute indices using starts mapping."""
     starts = docstring_rewriter.calc_line_starts(src)
-    assert docstring_rewriter.calc_abs_pos(starts, lineno, col) == expected
+    assert (
+        docstring_rewriter.calc_abs_pos(src, starts, lineno, col) == expected
+    )
 
 
 @pytest.mark.parametrize(