wcwidth 0.9.0__tar.gz → 0.9.2__tar.gz

{wcwidth-0.9.0 → wcwidth-0.9.2}/.gitignore

@@ -14,7 +14,11 @@ wheelhouse
 *.so
 dist
 docs/_build
+# generated from the C headers by bin/update-docs.py, not committed
+docs/api_c.rst
 htmlcov
+htmlcov-c
+coverage-c.lcov
 .coveralls.yml
 data
 .DS_Store

{wcwidth-0.9.0 → wcwidth-0.9.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: wcwidth
-Version: 0.9.0
+Version: 0.9.2
 Summary: Measures the displayed width of unicode strings in a terminal
 Author-email: Jeff Quast <contact@jeffquast.com>
 Project-URL: Homepage, https://github.com/jquast/wcwidth
@@ -97,8 +97,9 @@ This library may be considered to presume the terminal is enabled for DEC Privat
 ("Grapheme Clustering") by default, which may require to be enabled by a TUI application but
 is often the default mode for those terminals that support it.
 
-This library does support any specific "legacy width" measurement by API, but it does provide
-Corrections_ for those terminals without grapheme support.
+This library does not support any specific "legacy width" measurement by API, but it does provide
+Corrections_ for terminals using "legacy width" measurements or are without basic emoji or grapheme
+support.
 
 See also:
 
@@ -154,7 +155,7 @@ anywhere in the string.
 wcstwidth()
 -----------
 
-Same behavior as `wcswidth()`_ with automatic terminal-specific Corrections_, reading
+`wcstwidth()`_ is a variant of `wcswidth()`_ that provides terminal-specific Corrections_, reading
 ``TERM_PROGRAM`` or ``TERM`` when ``True`` (default), or caller can provide terminal query
 XTVERSION_ or ENQ_ response:
 
@@ -550,7 +551,7 @@ https://wcwidth.readthedocs.io/en/latest/related.html
 .. _`str.rjust()`: https://docs.python.org/3/library/stdtypes.html#str.rjust
 .. _`str.center()`: https://docs.python.org/3/library/stdtypes.html#str.center
 .. _`str.expandtabs()`: https://docs.python.org/3/library/stdtypes.html#str.expandtabs
-.. _`General Tabulated Summary`: https://ucs-detect.readthedocs.io/results.html#tabulated-results
+.. _`General Tabulated Summary`: https://ucs-detect.readthedocs.io/results.html#general-tabulated-summary
 .. _`wcwidth()`: https://wcwidth.readthedocs.io/en/latest/api.html#wcwidth.wcwidth
 .. _`wcswidth()`: https://wcwidth.readthedocs.io/en/latest/api.html#wcwidth.wcswidth
 .. _`wcstwidth()`: https://wcwidth.readthedocs.io/en/latest/api.html#wcwidth.wcstwidth

{wcwidth-0.9.0 → wcwidth-0.9.2}/README.rst

@@ -66,8 +66,9 @@ This library may be considered to presume the terminal is enabled for DEC Privat
 ("Grapheme Clustering") by default, which may require to be enabled by a TUI application but
 is often the default mode for those terminals that support it.
 
-This library does support any specific "legacy width" measurement by API, but it does provide
-Corrections_ for those terminals without grapheme support.
+This library does not support any specific "legacy width" measurement by API, but it does provide
+Corrections_ for terminals using "legacy width" measurements or are without basic emoji or grapheme
+support.
 
 See also:
 
@@ -123,7 +124,7 @@ anywhere in the string.
 wcstwidth()
 -----------
 
-Same behavior as `wcswidth()`_ with automatic terminal-specific Corrections_, reading
+`wcstwidth()`_ is a variant of `wcswidth()`_ that provides terminal-specific Corrections_, reading
 ``TERM_PROGRAM`` or ``TERM`` when ``True`` (default), or caller can provide terminal query
 XTVERSION_ or ENQ_ response:
 
@@ -519,7 +520,7 @@ https://wcwidth.readthedocs.io/en/latest/related.html
 .. _`str.rjust()`: https://docs.python.org/3/library/stdtypes.html#str.rjust
 .. _`str.center()`: https://docs.python.org/3/library/stdtypes.html#str.center
 .. _`str.expandtabs()`: https://docs.python.org/3/library/stdtypes.html#str.expandtabs
-.. _`General Tabulated Summary`: https://ucs-detect.readthedocs.io/results.html#tabulated-results
+.. _`General Tabulated Summary`: https://ucs-detect.readthedocs.io/results.html#general-tabulated-summary
 .. _`wcwidth()`: https://wcwidth.readthedocs.io/en/latest/api.html#wcwidth.wcwidth
 .. _`wcswidth()`: https://wcwidth.readthedocs.io/en/latest/api.html#wcwidth.wcswidth
 .. _`wcstwidth()`: https://wcwidth.readthedocs.io/en/latest/api.html#wcwidth.wcstwidth

{wcwidth-0.9.0 → wcwidth-0.9.2}/bin/update-docs.py

@@ -6,7 +6,9 @@ from __future__ import annotations
 # std imports
 import os
 import re
+import sys
 import glob
+import argparse
 import textwrap
 import importlib.util
 
@@ -21,7 +23,7 @@ import jinja2
 # docs/unicode_version.rst, docs/libwcwidth.rst, and the list_term_programs()
 # example in README.rst.
 
-PATH_UP = os.path.relpath(os.path.join(os.path.dirname(__file__), os.path.pardir))
+PATH_UP = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
 JINJA_ENV = jinja2.Environment(
     loader=jinja2.FileSystemLoader(os.path.join(PATH_UP, 'code_templates')),
     keep_trailing_newline=True,
@@ -458,21 +460,29 @@ def report(label: str, changed: bool) -> None:
     print(f'{label}: {"updated" if changed else "up-to-date"}')
 
 
-def write_if_changed(path: str, content: str, label: str) -> None:
-    """Write *content* to *path* only if it differs, reporting the result."""
+def write_if_changed(path: str, content: str, label: str, check: bool = False) -> bool:
+    """
+    Write *content* to *path* only if it differs, reporting and returning whether it did.
+
+    With *check*, report the difference without writing, so a stale file fails the run.
+    """
     try:
         with open(path, encoding='utf-8') as fin:
             original = fin.read()
     except FileNotFoundError:
         original = ''
-    if content != original:
-        new_path = path + '.new'
-        with open(new_path, 'w', encoding='utf-8', newline='\n') as fout:
-            fout.write(content)
-        os.replace(new_path, path)
-        print(f'{label}: updated')
-    else:
+    if content == original:
         print(f'{label}: up-to-date')
+        return False
+    if check:
+        print(f'{label}: out of date, run bin/update-docs.py')
+        return True
+    new_path = path + '.new'
+    with open(new_path, 'w', encoding='utf-8', newline='\n') as fout:
+        fout.write(content)
+    os.replace(new_path, path)
+    print(f'{label}: updated')
+    return True
 
 
 def update_libwcwidth_term_programs() -> bool:
@@ -545,18 +555,31 @@ def unicode_version_page() -> str:
         source_headers=unicode_source_headers())
 
 
-def main() -> None:
-    """Regenerate all generated documentation files."""
-    write_if_changed(API_C_OUTPUT, render_api_doc(), 'docs/api_c.rst')
+def main(api_only: bool = False, check: bool = False) -> int:
+    """Regenerate all generated documentation files, or only docs/api_c.rst."""
+    stale = write_if_changed(API_C_OUTPUT, render_api_doc(), 'docs/api_c.rst', check)
+    if api_only:
+        # api_c.rst is parsed from the C headers, but everything below reads the
+        # Unicode sources under data/, which only `tox -e fetch` downloads.
+        return int(check and stale)
     write_if_changed(os.path.join(PATH_DOCS, 'unicode_version.rst'),
-                     unicode_version_page(), 'docs/unicode_version.rst')
+                     unicode_version_page(), 'docs/unicode_version.rst', check)
     report('docs/libwcwidth.rst: canonical terminal names',
            update_libwcwidth_term_programs())
     report('docs/libwcwidth.rst: Unicode version',
            update_libwcwidth_unicode_version())
     report('README.rst: list_term_programs() example',
            update_readme_term_programs())
+    return int(check and stale)
 
 
 if __name__ == '__main__':
-    main()
+    parser = argparse.ArgumentParser(description=__doc__)
+    parser.add_argument('--api-only', action='store_true',
+                        help='regenerate docs/api_c.rst only, which needs no Unicode data')
+    parser.add_argument('--check', action='store_true',
+                        help='report stale output instead of writing it, for CI')
+    args = parser.parse_args()
+    if args.check and not args.api_only:
+        parser.error('--check needs --api-only: the other outputs read the fetched Unicode data')
+    sys.exit(main(api_only=args.api_only, check=args.check))

{wcwidth-0.9.0 → wcwidth-0.9.2}/docs/developing.rst

@@ -18,6 +18,12 @@ To run tests with detailed coverage reporting showing missing lines::
 
    tox -epy314 -- --cov-report=term-missing
 
+The C11 library's coverage is measured separately, since coverage.py cannot see compiled
+code.  This merges two instrumented runs, the C11 test suite and the Python suite driving
+the C extension, and writes ``coverage-c.lcov`` and ``htmlcov-c/index.html``::
+
+   tox -e c_coverage
+
 Updating Unicode Version
 ------------------------
 

{wcwidth-0.9.0 → wcwidth-0.9.2}/docs/history.rst

@@ -2,7 +2,20 @@
 History
 =======
 
-0.9.0 *unreleased* ('jq/libwcwidth-shim-wheels' branch)
+0.9.2 *2026-10-04*
+  * **Bugfix** `wrap()`_ should omit an empty line as output, like `textwrap.wrap()`_ `PR #291`_.
+  * **Bugfix** `clip()`_ when using non-default argument, ``tabsize=0``, `PR #288`_.
+  * **Bugfix** regional-indicator pairs (flags) when corrected for terminals that show lone regional
+    indicators as narrow (xterm, wezterm, VTE, and others) in `wcstwidth()`_ and `width()`_, `PR
+    #284`.
+
+0.9.1 *2026-09-23*
+  * **Bugfix** `width()`_ in Python and libwcwidth mis-measured CSI sequences over 64 bytes, a 0.9.0
+    release regression! `PR #271`_.
+  * **Performance** improvement of ~15% for many Python API functions when using default arguments
+    like ``ambiguous_width`` and ``term_program``. `PR #273`_
+
+0.9.0 *2026-09-23*
   * **New** Optional CPython extension using libwcwidth_, improving performance 10-58x `PR #248`_.
   * **Drop Support** for Python 3.8, `PR #248`_.
 
@@ -233,7 +246,7 @@ https://www.cl.cam.ac.uk/~mgk25/ucs/wcwidth.c::
  * for any purpose and without fee is hereby granted. The author
  * disclaims all warranties with regard to this software.
 
-.. _`Spacing Combining Mark`: https://www.unicode.org/versions/latest/ch04.pdf#G134153
+.. _`Spacing Combining Mark`: https://www.unicode.org/versions/latest/core-spec/chapter-4/#G134153
 .. _`specification`: https://wcwidth.readthedocs.io/en/latest/specs.html
 .. _`prospector`: https://github.com/landscapeio/prospector
 .. _`combining`: https://en.wikipedia.org/wiki/Combining_character
@@ -299,6 +312,12 @@ https://www.cl.cam.ac.uk/~mgk25/ucs/wcwidth.c::
 .. _`PR #253`: https://github.com/jquast/wcwidth/pull/253
 .. _`PR #259`: https://github.com/jquast/wcwidth/pull/259
 .. _`PR #262`: https://github.com/jquast/wcwidth/pull/262
+.. _`PR #271`: https://github.com/jquast/wcwidth/pull/271
+.. _`PR #273`: https://github.com/jquast/wcwidth/pull/273
+.. _`PR #281`: https://github.com/jquast/wcwidth/pull/281
+.. _`PR #284`: https://github.com/jquast/wcwidth/pull/284
+.. _`PR #288`: https://github.com/jquast/wcwidth/pull/288
+.. _`PR #291`: https://github.com/jquast/wcwidth/pull/291
 .. _`Issue #101`: https://github.com/jquast/wcwidth/issues/101
 .. _`Issue #155`: https://github.com/jquast/wcwidth/issues/155
 .. _`Issue #211`: https://github.com/jquast/wcwidth/issues/211

{wcwidth-0.9.0 → wcwidth-0.9.2}/docs/intro.rst

@@ -66,8 +66,9 @@ This library may be considered to presume the terminal is enabled for DEC Privat
 ("Grapheme Clustering") by default, which may require to be enabled by a TUI application but
 is often the default mode for those terminals that support it.
 
-This library does support any specific "legacy width" measurement by API, but it does provide
-Corrections_ for those terminals without grapheme support.
+This library does not support any specific "legacy width" measurement by API, but it does provide
+Corrections_ for terminals using "legacy width" measurements or are without basic emoji or grapheme
+support.
 
 See also:
 
@@ -123,7 +124,7 @@ anywhere in the string.
 wcstwidth()
 -----------
 
-Same behavior as `wcswidth()`_ with automatic terminal-specific Corrections_, reading
+`wcstwidth()`_ is a variant of `wcswidth()`_ that provides terminal-specific Corrections_, reading
 ``TERM_PROGRAM`` or ``TERM`` when ``True`` (default), or caller can provide terminal query
 XTVERSION_ or ENQ_ response:
 
@@ -519,7 +520,7 @@ https://wcwidth.readthedocs.io/en/latest/related.html
 .. _`str.rjust()`: https://docs.python.org/3/library/stdtypes.html#str.rjust
 .. _`str.center()`: https://docs.python.org/3/library/stdtypes.html#str.center
 .. _`str.expandtabs()`: https://docs.python.org/3/library/stdtypes.html#str.expandtabs
-.. _`General Tabulated Summary`: https://ucs-detect.readthedocs.io/results.html#tabulated-results
+.. _`General Tabulated Summary`: https://ucs-detect.readthedocs.io/results.html#general-tabulated-summary
 .. _`wcwidth()`: https://wcwidth.readthedocs.io/en/latest/api.html#wcwidth.wcwidth
 .. _`wcswidth()`: https://wcwidth.readthedocs.io/en/latest/api.html#wcwidth.wcswidth
 .. _`wcstwidth()`: https://wcwidth.readthedocs.io/en/latest/api.html#wcwidth.wcstwidth

{wcwidth-0.9.0 → wcwidth-0.9.2}/docs/libwcwidth.rst

@@ -10,16 +10,20 @@ The Python documentation_ closely matches this C library.  The C API adds UTF-8
 array interfaces.
 
 The lowest-level functions are derived from POSIX.1-2001 and POSIX.1-2008 `wcwidth(3)`_ and
-`wcswidth(3)`_, which this library implements as `wcwidth_u32()`_ and `wcswidth_u32()`_.  These
-functions return -1 when C0 and C1 control codes other than NUL are present; NUL measures as
-zero-width.  They do not parse terminal escape sequences: any escape sequence contains control
-codes, so these functions return -1 for it.
+`wcswidth(3)`_, which this library implements as :ref:`wcwidth_u32() <libwcwidth-wcwidth>` and
+:ref:`wcswidth_u32() <libwcwidth-wcswidth>`.  These functions return -1 when C0 and C1 control
+codes other than NUL are present; NUL measures as zero-width.
+:ref:`wcstwidth_u8() <libwcwidth-wcstwidth>` applies corrections for a specific terminal program,
+as described in the Python Corrections_ documentation.
 
-`wcwidth_width_u8()`_ is a higher-level wrapper of `wcswidth_u8()`_ that also measures terminal
-control sequences, like colors, bold, tabstops, and horizontal cursor movement.
+:ref:`wcwidth_width_u8() <libwcwidth-width>` is a higher-level wrapper of
+:ref:`wcswidth_u8() <libwcwidth-wcswidth>` that also measures terminal control sequences, like
+colors, bold, tabstops, and horizontal cursor movement.
 
-`wcstwidth_u8()`_ applies corrections for a specific terminal program and version, as described
-in the Python Corrections_ documentation.
+The text transforms have their own sections:
+:ref:`wcwidth_ljust_u8() <libwcwidth-align>` with ``rjust`` and ``center``,
+:ref:`wcwidth_clip_u8() <libwcwidth-clip>`, :ref:`wcwidth_wrap_u8() <libwcwidth-wrap>`, and
+:ref:`wcwidth_escape_strip() <libwcwidth-escape-strip>`.
 
 Quick Start
 -----------
@@ -133,16 +137,15 @@ address:
 String length conventions
 ~~~~~~~~~~~~~~~~~~~~~~~~~
 
-Every string function takes an explicit length and reads exactly that many units:
+Every string function reads exactly the length you pass:
 
 ``_u32`` functions
     count codepoints in the array,
 ``_u8`` functions
     count bytes.
 
-There is no NUL-terminated sentinel form; pass ``strlen(text)`` when the text is a C string.  The
-length is authoritative: a NUL is an ordinary zero-width character that may appear anywhere, and
-it survives into transform output, whose ``*out_len`` is the true length.
+Pass ``strlen(text)`` when the text is a C string.  A NUL counts as an ordinary zero-width character
+in any position.  Transform functions copy it through and report the full length in ``*out_len``.
 
 Alternate encodings
 ~~~~~~~~~~~~~~~~~~~
@@ -187,6 +190,8 @@ throughout, or use the ``_u32`` forms and re-encode the result.  `wcwidth_encode
 Re-encoding to a legacy charset is the caller's iconv(3) or ICU (``ucnv_*``) call; a byte cast
 works when every codepoint fits the target, and iconv reports ``EILSEQ`` when one does not.
 
+.. _libwcwidth-wcwidth:
+
 wcwidth_u32()
 ~~~~~~~~~~~~~
 
@@ -206,6 +211,8 @@ U+2640 above is Ambiguous, so it answers to the second argument, while U+2630 is
 2 under either setting.  A single codepoint needs no ``_u8`` variant; use `wcswidth_u8()`_ to
 measure text.
 
+.. _libwcwidth-wcswidth:
+
 wcswidth_u32() and wcswidth_u8()
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
@@ -219,6 +226,8 @@ when any control code other than NUL is present:
     wcswidth_u32(family, 5, 1)   /* man ZWJ woman ZWJ girl */  2
     wcswidth_u8("café", 5, 1)                                  4
 
+.. _libwcwidth-wcstwidth:
+
 wcstwidth_u32() and wcstwidth_u8()
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
@@ -230,6 +239,8 @@ argument applies terminal-specific corrections:
     wcswidth_u8("☰", 3, 1)               /* U+2630, wide */  2
     wcstwidth_u8("☰", 3, 1, "vte")                            1
 
+.. _libwcwidth-width:
+
 wcwidth_width_u32() and wcwidth_width_u8()
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
@@ -261,6 +272,8 @@ codes:
         /* err is WCWIDTH_ERROR_VERTICAL_CTRL */
     }
 
+.. _libwcwidth-align:
+
 wcwidth_ljust_u8(), wcwidth_rjust_u8(), and wcwidth_center_u8()
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
@@ -288,6 +301,13 @@ returns a ``malloc``\ 'd NUL-terminated string the caller must ``free``:
     /* "*café*" */
     free(out);
 
+Tabs are measured as though text begins at column 0.  ``wcwidth_rjust_u8()`` and
+``wcwidth_center_u8()``  insert padding to the left of the ``text`` which move tabstops and return
+an incorrect result, always expand tabs or call `wcwidth_clip_u8()`_ to ``opts->dest_width`` first,
+which expands tabs before using these alignment functions.
+
+.. _libwcwidth-clip:
+
 wcwidth_clip_u8()
 ~~~~~~~~~~~~~~~~~
 
@@ -344,6 +364,8 @@ Movement is reported wherever it appears; the OSC sequences only where the clip
 `wcwidth_clip_u32()`_ is the codepoint-array form, returning a ``malloc``\ 'd array of ``*out_len``
 codepoints.
 
+.. _libwcwidth-wrap:
+
 wcwidth_wrap_u8() and wcwidth_wrap_u8_text()
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
@@ -383,6 +405,8 @@ unbreakable unit, so a line is never broken inside one, even at a space or hyphe
 text.  Python's `wrap()`_ behaves the same way.  `wcwidth_wrap_u32()`_ and `wcwidth_wrap_u32_text()`_ are the
 codepoint-array forms.
 
+.. _libwcwidth-escape-strip:
+
 wcwidth_escape_strip()
 ~~~~~~~~~~~~~~~~~~~~~~
 

{wcwidth-0.9.0 → wcwidth-0.9.2}/docs/specs.rst

@@ -4,10 +4,11 @@
 Specification
 =============
 
-This document defines how this Python wcwidth library measures the printable width of characters of
-a string. This is not meant to an official standard, but as a terse description of the lowest level
-API functions :func:`wcwidth.wcwidth` and  :func:`wcwidth.wcswidth` and its relation to higher level
-functions :func:`wcwidth.width` and :func:`wcwidth.iter_graphemes`.
+This document defines how the Python wcwidth and C11 libwcwidth_ libraries measure the printable
+width of characters of a string. This is not meant to be an official standard, but as a terse
+description of the lowest level API functions :func:`wcwidth.wcwidth` and :func:`wcwidth.wcswidth`
+and their relation to higher level functions :func:`wcwidth.width`, :func:`wcwidth.wcstwidth` and
+:func:`wcwidth.iter_graphemes`.
 
 Scope
 -----
@@ -19,10 +20,19 @@ its docstring and specifications of related control codes, `XTerm Control Sequen
 Text Sizing Protocol`_.
 
 Each string yielded by :func:`wcwidth.iter_graphemes` may be mapped to :func:`wcwidth.wcswidth` to
-accurately measure the width of a grapheme. Although :func:`wcwidth.iter_graphemes` matches
-behavior of Python 3.15 `uncodedata.iter_graphemes()`_ it differs in its return value,
-:func:`wcwidth.iter_graphemes` yields only strings, while :func:`unicodedata.iter_graphemes` yields
-``unicodedata.Segment`` class objects.
+accurately measure the width of a grapheme.  :func:`wcwidth.iter_graphemes` implements the `Unicode
+Standard Annex #29`_ rules of Unicode 18.0, the version of this library's tables, while
+`unicodedata.iter_graphemes()`_, new in Python 3.15, implements Unicode 17.0.  They also differ in
+return value: :func:`wcwidth.iter_graphemes` yields only strings, while
+:func:`unicodedata.iter_graphemes` yields ``unicodedata.Segment`` class objects.
+
+Tamil, Kannada and Sinhala text is measured differently than it is segmented, so
+:func:`wcwidth.clip` and :func:`wcwidth.wrap` do not measure these scripts correctly.  See `Virama
+Conjunct Formation`_.
+
+The C11 library libwcwidth_ implements this specification through parallel entry points:
+``wcswidth_u32()`` and ``wcswidth_u8()`` for :func:`wcwidth.wcswidth`, ``wcstwidth_u8()`` for
+:func:`wcwidth.wcstwidth`, and ``wcwidth_width_u8()`` for :func:`wcwidth.width`; see the `C11 API`_.
 
 Width of -1
 -----------
@@ -30,39 +40,38 @@ Width of -1
 The following have a column width of -1 for function :func:`wcwidth.wcwidth`
 
 - ``C0`` control characters (`U+0001`_ through `U+001F`_).
-- ``C1`` control characters and ``DEL`` (`U+007F`_ through `U+00A0`_).
+- ``C1`` control characters and ``DEL`` (`U+007F`_ through `U+009F`_).
 
 If any character in sequence contains ``C0`` or ``C1`` control characters, the final
-return value of :func:`wcwidth.wcswidth` is -1.
+return value of :func:`wcwidth.wcswidth` is -1, except a control character immediately following
+ZWJ (`U+200D`_), which `Width of 0`_ consumes.
 
 Width of 0
 ----------
 
 Any characters with the `Default_Ignorable_Code_Point`_ property in
-`DerivedCoreProperties.txt`_ files, 4,174 characters, excluding `U+00AD`_ SOFT HYPHEN
-(width 1) and `U+115F`_ HANGUL CHOSEONG FILLER (width 2).
+`DerivedCoreProperties.txt`_, excluding `U+00AD`_ SOFT HYPHEN (width 1) and
+`U+115F`_ HANGUL CHOSEONG FILLER (width 2).
 
-Any characters defined by `General Category`_ codes in `DerivedGeneralCategory.txt`_ files:
+Any characters defined by `General Category`_ codes in `DerivedGeneralCategory.txt`_:
 
-- 'Me': `Enclosing Mark`_, aprox. 13 characters.
-- 'Mn': `Nonspacing Mark`_, aprox. 1,839 characters.
+- 'Me': `Enclosing Mark`_.
+- 'Mn': `Nonspacing Mark`_.
 - 'Cf': `Format`_ control characters excluding `U+00AD`_ SOFT HYPHEN and
-  `Prepended_Concatenation_Mark`_ characters, aprox. 147 characters.
+  `Prepended_Concatenation_Mark`_ characters.
 - 'Zl': `U+2028`_ LINE SEPARATOR only
 - 'Zp': `U+2029`_ PARAGRAPH SEPARATOR only
-- 'Sk': `Modifier Symbol`_, aprox. 1 character with ``'FULLWIDTH'`` in comment
-  of `UnicodeData.txt`_ (see `Width of 2`_). `Emoji Modifier`_ Fitzpatrick
-  symbols (`U+1F3FB`_ through `U+1F3FF`_) are zero-width only when following
-  an emoji base character in sequence; see `Width of 2`_ for standalone.
+
+`Emoji Modifier`_ Fitzpatrick symbols (`U+1F3FB`_ through `U+1F3FF`_) are zero-width
+only when following an emoji base character in sequence; see `Width of 2`_ for standalone.
 
 The NULL character (`U+0000`_).
 
-Any character following ZWJ (`U+200D`_) when preceded by an emoji
-(`Extended_Pictographic`_ property) or `Regional Indicator`_ in sequence by
-function :func:`wcwidth.wcswidth`, following grapheme cluster boundary rules
-of `Unicode Standard Annex #29`_. When ZWJ follows a non-emoji character
-(including CJK), only the ZWJ itself is zero-width; the following character
-is measured normally.
+ZWJ (`U+200D`_) is zero-width, and in sequence by :func:`wcwidth.wcswidth` or
+:func:`wcwidth.width` the codepoint immediately following it is consumed as zero-width.
+For emoji (`Extended_Pictographic`_ property) and `Regional Indicator`_ sequences this follows the
+grapheme cluster boundary rules of `Unicode Standard Annex #29`_. After a virama, the following
+consonant continues the conjunct, see `Virama Conjunct Formation`_.
 
 The second `Regional Indicator`_ symbol (`U+1F1E6`_ through `U+1F1FF`_) in a
 consecutive pair, when measured in sequence by :func:`wcwidth.wcswidth` or
@@ -71,9 +80,9 @@ consecutive pair, when measured in sequence by :func:`wcwidth.wcswidth` or
 `Hangul Jamo`_ Jungseong and "Extended-B" code blocks, `U+1160`_ through
 `U+11FF`_ and `U+D7B0`_ through `U+D7FF`_.
 
-Any characters of category ``Mc`` (`Spacing Combining Mark`_), aprox. 443
-characters, for the single-character function :func:`wcwidth.wcwidth`.
-When measured in sequence by :func:`wcwidth.wcswidth`, see `Width of 2`_.
+Any characters of category ``Mc`` (`Spacing Combining Mark`_), for the
+single-character function :func:`wcwidth.wcwidth`.  When measured in sequence by
+:func:`wcwidth.wcswidth`, see `Width of 2`_.
 
 Width of 1
 ----------
@@ -93,8 +102,9 @@ Width of 2
 ----------
 
 Any character defined by `East Asian`_ (`Unicode Standard Annex #11`_) Fullwidth
-(``F``) or Wide (``W``) properties in `EastAsianWidth.txt`_ files, except those
-that are defined by the Category code of `Nonspacing Mark`_ (``Mn``).
+(``F``) or Wide (``W``) properties in `EastAsianWidth.txt`_, except those that
+are defined by the Category codes of `Nonspacing Mark`_ (``Mn``) or `Spacing
+Combining Mark`_ (``Mc``).
 
 `Regional Indicator`_ symbols (`U+1F1E6`_ through `U+1F1FF`_). Though
 classified as Neutral in `EastAsianWidth.txt`_, terminals universally render
@@ -106,7 +116,7 @@ measured standalone (not following an emoji base character). When following
 an emoji base, they combine with the base and add 0 to total width.
 
 Any characters of `Modifier Symbol`_ category, ``'Sk'`` where ``'FULLWIDTH'`` is
-present in comment of `UnicodeData.txt`_, aprox. 3 characters.
+present in comment of `UnicodeData.txt`_.
 
 Any character with `U+FE0F`_ (Variation Selector 16) defined as ``emoji style``
 in `emoji-variation-sequences.txt`_, per `UTS #51`_ and `Unicode Standard
@@ -116,13 +126,14 @@ making the pair width 2. Wide characters are unchanged.
 Any character of non-zero width followed by an ``Mc`` (`Spacing Combining Mark`_)
 character when measured in sequence by :func:`wcwidth.wcswidth` or
 :func:`wcwidth.width`. The ``Mc`` character caps the cluster width at 2,
-reflecting its *positive advance width* as defined in `General Category`_
-(Table 4-4). Zero-width combining marks (``Mn``) between the base character
-and the ``Mc`` do not break the association. For example, a consonant followed
+reflecting its *positive advance width* as defined in the `General Category Values`_
+table. Zero-width combining marks (``Mn``) between the base character and the ``Mc``
+do not break the association. For example, a consonant followed
 by a Nukta (``Mn``) and then a vowel sign (``Mc``) is measured as a cluster of
 width 2.
 
-Any grapheme cluster width is limited to 2 cells since 0.8.0, `PR #224`_.
+Any grapheme cluster width is limited to 2 cells since 0.8.0, `PR #224`_, which prevents the ``Mc``
+rule and the virama conjunct sum from exceeding 2 cells.
 
 Virama Conjunct Formation
 -------------------------
@@ -145,16 +156,19 @@ described in the Virama section header).
 - ZWJ (`U+200D`_) after a virama is consumed without breaking conjunct state,
   supporting explicit half-form requests (virama + ZWJ + consonant).
 
-See also: `L2/2023/23107`_ "Proper Complex Script Support in Text Terminals".
+See also: `L2/23-107`_ "Proper Complex Script Support in Text Terminals", the 2023 project
+proposal that records the missing terminal-width standard, and `L2/25-263`_ for the Text Terminal
+Working Group's latest report.
 
-.. _`Hyperlinks in Terminal Emulators`: https://gist.github.com/egmontkob/eb114294efbcd5adb1944c9f3cb5feda
 .. _`Kitty Text Sizing Protocol`: https://sw.kovidgoyal.net/kitty/text-sizing-protocol/
 .. _`XTerm Control Sequences`: https://invisible-island.net/xterm/ctlseqs/ctlseqs.html
+.. _`C11 API`: https://wcwidth.readthedocs.io/en/latest/api_c.html
+.. _libwcwidth: https://wcwidth.readthedocs.io/en/latest/libwcwidth.html
 .. _`U+0000`: https://codepoints.net/U+0000
 .. _`U+0001`: https://codepoints.net/U+0001
 .. _`U+001F`: https://codepoints.net/U+001F
 .. _`U+007F`: https://codepoints.net/U+007F
-.. _`U+00A0`: https://codepoints.net/U+00A0
+.. _`U+009F`: https://codepoints.net/U+009F
 .. _`U+00AD`: https://codepoints.net/U+00AD
 .. _`U+1160`: https://codepoints.net/U+1160
 .. _`U+11FF`: https://codepoints.net/U+11FF
@@ -172,12 +186,13 @@ See also: `L2/2023/23107`_ "Proper Complex Script Support in Text Terminals".
 .. _`Prepended_Concatenation_Mark`: https://www.unicode.org/reports/tr44/#Prepended_Concatenation_Mark
 .. _`Default_Ignorable_Code_Point`: https://www.unicode.org/reports/tr44/#Default_Ignorable_Code_Point
 .. _`General Category`: https://www.unicode.org/reports/tr44/#General_Category
+.. _`General Category Values`: https://www.unicode.org/reports/tr44/#GC_Values_Table
 .. _`Spacing Combining Mark`: https://www.unicode.org/versions/latest/core-spec/chapter-4/#G134153
 .. _`Enclosing Mark`: https://www.unicode.org/versions/latest/core-spec/chapter-4/#G134153
 .. _`Format`: https://www.unicode.org/versions/latest/core-spec/chapter-4/#G134153
 .. _`Modifier Symbol`: https://www.unicode.org/versions/latest/core-spec/chapter-4/#G134153
 .. _`Hangul Jamo`: https://www.unicode.org/charts/PDF/U1100.pdf
-.. _`U+D7FF`: https://codepoints.net/U+D7FF
+.. _`U+D7FF`: https://www.unicode.org/charts/PDF/UD7B0.pdf
 .. _`UnicodeData.txt`: https://www.unicode.org/Public/UCD/latest/ucd/UnicodeData.txt
 .. _`East Asian`: https://www.unicode.org/reports/tr11/
 .. _`Unicode Standard Annex #11`: https://www.unicode.org/reports/tr11/
@@ -187,18 +202,18 @@ See also: `L2/2023/23107`_ "Proper Complex Script Support in Text Terminals".
 .. _`U+1F3FF`: https://codepoints.net/U+1F3FF
 .. _`Regional Indicator`: https://www.unicode.org/charts/PDF/U1F100.pdf
 .. _`Emoji Modifier`: https://unicode.org/reports/tr51/#Emoji_Modifiers
-.. _`Extended_Pictographic`: https://www.unicode.org/reports/tr51/#def_extended_pictographic
+.. _`Extended_Pictographic`: https://www.unicode.org/Public/UCD/latest/ucd/emoji/emoji-data.txt
 .. _`UTS #51`: https://www.unicode.org/reports/tr51/
 .. _`Nonspacing Mark`: https://www.unicode.org/versions/latest/core-spec/chapter-4/#G134153
 .. _`IndicSyllabicCategory.txt`: https://www.unicode.org/Public/UCD/latest/ucd/IndicSyllabicCategory.txt
-.. _`Indic_Syllabic_Category`: https://www.unicode.org/reports/tr44/#Indic_Syllabic_Category
 .. _`Invisible_Stacker`: https://www.unicode.org/Public/UCD/latest/ucd/IndicSyllabicCategory.txt
 .. _`Brahmic scripts`: https://en.wikipedia.org/wiki/Brahmic_scripts
 .. _`Virama`: https://www.unicode.org/glossary/#virama
 .. _`conjunct`: https://www.unicode.org/glossary/#consonant_conjunct
 .. _`aksara`: https://www.unicode.org/glossary/#aksara
-.. _`L2/2023/23107`: https://www.unicode.org/L2/L2023/23107-terminal-suppt.pdf
+.. _`L2/23-107`: https://www.unicode.org/L2/L2023/23107-terminal-suppt.pdf
+.. _`L2/25-263`: https://www.unicode.org/L2/L2025/25263-ttwg-report-utc185.pdf
 .. _`Unicode Standard Annex #29`: https://www.unicode.org/reports/tr29/
-.. _`Unicode Standard Section 23.4`: https://www.unicode.org/versions/Unicode16.0.0/core-spec/chapter-23/#G77993
-.. _`uncodedata.iter_graphemes()`: https://docs.python.org/3.15/library/unicodedata.html#unicodedata.iter_graphemes
+.. _`Unicode Standard Section 23.4`: https://www.unicode.org/versions/latest/core-spec/chapter-23/#G19053
+.. _`unicodedata.iter_graphemes()`: https://docs.python.org/3.15/library/unicodedata.html#unicodedata.iter_graphemes
 .. _`PR #224`: https://github.com/jquast/wcwidth/pull/224

{wcwidth-0.9.0 → wcwidth-0.9.2}/libwcwidth/include/wcwidth/align.h

@@ -48,6 +48,11 @@ extern const wcwidth_align_opts_t WCWIDTH_ALIGN_OPTS_DEFAULT;
  *   opts:     alignment options, or NULL for defaults.
  *   out_len:  output: byte length of result (excluding NUL); may be NULL.
  *   error:    output: wcwidth_error_t, WCWIDTH_ERROR_NONE on success.
+ *
+ * Tabs are measured as though text begins at column 0.  'rjust' and 'center' operations insert
+ * padding to the left of the ``text`` which move tabstops and return an incorrect result, always
+ * expand tabs or clip to ``opts->dest_width``, which expands tabs before using those alignment
+ * functions.
  */
 char *wcwidth_ljust_u8(const char *text, size_t text_len, wcwidth_control_mode_t mode,
                        const wcwidth_align_opts_t *opts, size_t *out_len, int *error);