Bugfix grapheme and East-Asian support by wcwidth>=0.5.1 integration

pyproject.toml

@@ -27,7 +27,7 @@ dynamic = ["version"]
 Homepage = "https://github.com/astanin/python-tabulate"
 
 [project.optional-dependencies]
-widechars = ["wcwidth"]
+widechars = ["wcwidth>=0.5.1"]
 
 [project.scripts]
 tabulate = "tabulate:_main"

tabulate/__init__.py

@@ -1116,6 +1116,9 @@ def _visible_width(s):
     # optional wide-character support
     if wcwidth is not None and WIDE_CHARS_MODE:
         len_fn = wcwidth.wcswidth
+        if hasattr(wcwidth, "width"):
+            # wcwidth >=0.3.0 handles ansi
+            return wcwidth.width(s)
     else:
         len_fn = len
     if isinstance(s, (str, bytes)):
@@ -1588,9 +1591,11 @@ def _normalize_tabular_data(tabular_data, headers, showindex="default"):
             if headers == "keys":
                 headers = field_names
             rows = [
-                [getattr(row, f) for f in field_names]
-                if not _is_separating_line(row)
-                else row
+                (
+                    [getattr(row, f) for f in field_names]
+                    if not _is_separating_line(row)
+                    else row
+                )
                 for row in rows
             ]
 
@@ -1638,7 +1643,13 @@ def _normalize_tabular_data(tabular_data, headers, showindex="default"):
     return rows, headers, headers_pad
 
 
-def _wrap_text_to_colwidths(list_of_lists, colwidths, numparses=True, break_long_words=_BREAK_LONG_WORDS, break_on_hyphens=_BREAK_ON_HYPHENS):
+def _wrap_text_to_colwidths(
+    list_of_lists,
+    colwidths,
+    numparses=True,
+    break_long_words=_BREAK_LONG_WORDS,
+    break_on_hyphens=_BREAK_ON_HYPHENS,
+):
     if len(list_of_lists):
         num_cols = len(list_of_lists[0])
     else:
@@ -1655,10 +1666,15 @@ def _wrap_text_to_colwidths(list_of_lists, colwidths, numparses=True, break_long
                 continue
 
             if width is not None:
-                wrapper = _CustomTextWrap(width=width, break_long_words=break_long_words, break_on_hyphens=break_on_hyphens)
+                wrapper_wrap = partial(
+                    _wrap_text,
+                    width=width,
+                    break_long_words=break_long_words,
+                    break_on_hyphens=break_on_hyphens,
+                )
                 casted_cell = str(cell)
                 wrapped = [
-                    "\n".join(wrapper.wrap(line))
+                    "\n".join(wrapper_wrap(line))
                     for line in casted_cell.splitlines()
                     if line.strip() != ""
                 ]
@@ -2258,7 +2274,11 @@ def tabulate(
 
         numparses = _expand_numparse(disable_numparse, num_cols)
         list_of_lists = _wrap_text_to_colwidths(
-            list_of_lists, maxcolwidths, numparses=numparses, break_long_words=break_long_words, break_on_hyphens=break_on_hyphens
+            list_of_lists,
+            maxcolwidths,
+            numparses=numparses,
+            break_long_words=break_long_words,
+            break_on_hyphens=break_on_hyphens,
         )
 
     if maxheadercolwidths is not None:
@@ -2272,7 +2292,11 @@ def tabulate(
 
         numparses = _expand_numparse(disable_numparse, num_cols)
         headers = _wrap_text_to_colwidths(
-            [headers], maxheadercolwidths, numparses=numparses, break_long_words=break_long_words, break_on_hyphens=break_on_hyphens
+            [headers],
+            maxheadercolwidths,
+            numparses=numparses,
+            break_long_words=break_long_words,
+            break_on_hyphens=break_on_hyphens,
         )[0]
 
     # empty values in the first column of RST tables should be escaped (issue #82)
@@ -2672,6 +2696,26 @@ def _format_table(
         return ""
 
 
+def _wrap_text(text, width, break_long_words=True, break_on_hyphens=True):
+    """Wrap text to width with wide character and ANSI code support."""
+    # wcwidth >= 0.5.0 has wrap() with proper grapheme cluster support and
+    # propagate_sgr=True by default, which handles ANSI code propagation natively.
+    if wcwidth is not None and hasattr(wcwidth, "wrap"):
+        return wcwidth.wrap(
+            text,
+            width,
+            break_long_words=break_long_words,
+            break_on_hyphens=break_on_hyphens,
+        )
+    else:
+        # Fallback for wcwidth < 0.3.0 or no wcwidth
+        return _CustomTextWrap(
+            width=width,
+            break_long_words=break_long_words,
+            break_on_hyphens=break_on_hyphens,
+        ).wrap(text)
+
+
 class _CustomTextWrap(textwrap.TextWrapper):
     """A custom implementation of CPython's textwrap.TextWrapper. This supports
     both wide characters (Korea, Japanese, Chinese)  - including mixed string.
@@ -2740,11 +2784,13 @@ def _handle_long_word(self, reversed_chunks, cur_line, cur_len, width):
         # of the next chunk onto the current line as will fit.
         if self.break_long_words:
             # Tabulate Custom: Build the string up piece-by-piece in order to
-            # take each charcter's width into account
+            # take each character's width into account
             chunk = reversed_chunks[-1]
             i = 1
             # Only count printable characters, so strip_ansi first, index later.
-            while len(_strip_ansi(chunk)[:i]) <= space_left:
+            # Use self._len() instead of len() to account for displayed width, eg.
+            # wide chars like CJK count as 2 when using wcwidth<0.3.0 without wrap()
+            while self._len(_strip_ansi(chunk)[:i]) <= space_left:
                 i = i + 1
             # Consider escape codes when breaking words up
             total_escape_len = 0

test/test_api.py

@@ -1,11 +1,8 @@
-"""API properties.
-
-"""
+"""API properties."""
 
 from tabulate import tabulate, tabulate_formats, simple_separated_format
 from common import skip
 
-
 try:
     from inspect import signature, _empty
 except ImportError:

test/test_cli.py

@@ -1,6 +1,4 @@
-"""Command-line interface.
-
-"""
+"""Command-line interface."""
 
 import os
 import sys
@@ -12,7 +10,6 @@
 
 from common import assert_equal
 
-
 SAMPLE_SIMPLE_FORMAT = "\n".join(
     [
         "-----  ------  -------------",