Have EvTable width correctly handle asian alphabets

CHANGELOG.md

@@ -60,6 +60,7 @@ without arguments starts a full interactive Python console.
   to `spawn` command to extract the raw prototype dict for manual editing.
 - `list_to_string` is now `iter_to_string` (but old name still works as legacy alias). It will 
   now accept any input, including generators and single values. 
+- EvTable should now correctly handle columns with wider asian-characters in them.
    
 
 

evennia/utils/evtable.py

@@ -114,7 +114,7 @@ table string.
 from django.conf import settings
 from textwrap import TextWrapper
 from copy import deepcopy, copy
-from evennia.utils.utils import m_len, is_iter
+from evennia.utils.utils import is_iter, display_len as d_len
 from evennia.utils.ansi import ANSIString
 
 _DEFAULT_WIDTH = settings.CLIENT_DEFAULT_WIDTH
@@ -228,7 +228,7 @@ class ANSITextWrapper(TextWrapper):
                 indent = self.initial_indent
 
             # Maximum width for this line.
-            width = self.width - m_len(indent)
+            width = self.width - d_len(indent)
 
             # First chunk on line is whitespace -- drop it, unless this
             # is the very beginning of the text (ie. no lines started yet).
@@ -236,7 +236,7 @@ class ANSITextWrapper(TextWrapper):
                 del chunks[-1]
 
             while chunks:
-                l = m_len(chunks[-1])
+                l = d_len(chunks[-1])
 
                 # Can at least squeeze this chunk onto the current line.
                 if cur_len + l <= width:
@@ -249,7 +249,7 @@ class ANSITextWrapper(TextWrapper):
 
             # The current line is full, and the next chunk is too big to
             # fit on *any* line (not just this one).
-            if chunks and m_len(chunks[-1]) > width:
+            if chunks and d_len(chunks[-1]) > width:
                 self._handle_long_word(chunks, cur_line, cur_len, width)
 
             # If the last chunk on this line is all whitespace, drop it.
@@ -439,7 +439,7 @@ class EvCell(object):
         self.valign = kwargs.get("valign", "c")
 
         self.data = self._split_lines(_to_ansi(data))
-        self.raw_width = max(m_len(line) for line in self.data)
+        self.raw_width = max(d_len(line) for line in self.data)
         self.raw_height = len(self.data)
 
         # this is extra trimming required for cels in the middle of a table only
@@ -478,9 +478,9 @@ class EvCell(object):
             width (int): The width to crop `text` to.
 
         """
-        if m_len(text) > width:
+        if d_len(text) > width:
             crop_string = self.crop_string
-            return text[: width - m_len(crop_string)] + crop_string
+            return text[: width - d_len(crop_string)] + crop_string
         return text
 
     def _reformat(self):
@@ -521,7 +521,7 @@ class EvCell(object):
         width = self.width
         adjusted_data = []
         for line in data:
-            if 0 < width < m_len(line):
+            if 0 < width < d_len(line):
                 # replace_whitespace=False, expand_tabs=False is a
                 # fix for ANSIString not supporting expand_tabs/translate
                 adjusted_data.extend(
@@ -564,7 +564,7 @@ class EvCell(object):
             text (str): Centered text.
 
         """
-        excess = width - m_len(text)
+        excess = width - d_len(text)
         if excess <= 0:
             return text
         if excess % 2:
@@ -603,13 +603,13 @@ class EvCell(object):
                     if line.startswith(" ") and not line.startswith("  ")
                     else line
                 )
-                + hfill_char * (width - m_len(line))
+                + hfill_char * (width - d_len(line))
                 for line in data
             ]
             return lines
         elif align == "r":
             return [
-                hfill_char * (width - m_len(line))
+                hfill_char * (width - d_len(line))
                 + (
                     " " + line.rstrip(" ")
                     if line.endswith(" ") and not line.endswith("  ")
@@ -750,7 +750,7 @@ class EvCell(object):
             natural_width (int): Width of cell.
 
         """
-        return m_len(self.formatted[0])  # if self.formatted else 0
+        return d_len(self.formatted[0])  # if self.formatted else 0
 
     def replace_data(self, data, **kwargs):
         """
@@ -765,7 +765,7 @@ class EvCell(object):
 
         """
         self.data = self._split_lines(_to_ansi(data))
-        self.raw_width = max(m_len(line) for line in self.data)
+        self.raw_width = max(d_len(line) for line in self.data)
         self.raw_height = len(self.data)
         self.reformat(**kwargs)
 

evennia/utils/tests/test_utils.py

@@ -98,6 +98,34 @@ class TestMLen(TestCase):
         self.assertEqual(utils.m_len({"hello": True, "Goodbye": False}), 2)
 
 
+class TestDisplayLen(TestCase):
+    """
+    Verifies that display_len behaves like m_len in all situations except those
+    where asian characters are involved.
+    """
+
+    def test_non_mxp_string(self):
+        self.assertEqual(utils.display_len("Test_string"), 11)
+
+    def test_mxp_string(self):
+        self.assertEqual(utils.display_len("|lclook|ltat|le"), 2)
+
+    def test_mxp_ansi_string(self):
+        self.assertEqual(utils.display_len(ANSIString("|lcl|gook|ltat|le|n")), 2)
+
+    def test_non_mxp_ansi_string(self):
+        self.assertEqual(utils.display_len(ANSIString("|gHello|n")), 5)
+
+    def test_list(self):
+        self.assertEqual(utils.display_len([None, None]), 2)
+
+    def test_dict(self):
+        self.assertEqual(utils.display_len({"hello": True, "Goodbye": False}), 2)
+
+    def test_east_asian(self):
+        self.assertEqual(utils.display_len("서서서"), 6)
+
+
 class TestANSIString(TestCase):
     """
     Verifies that ANSIString's string-API works as intended.

evennia/utils/utils.py

@@ -9,7 +9,7 @@ be of use when designing your own game.
 import os
 import gc
 import sys
-import copy 
+import copy
 import types
 import math
 import re
@@ -20,6 +20,7 @@ import traceback
 import importlib
 import importlib.util
 import importlib.machinery
+from unicodedata import east_asian_width
 from twisted.internet.task import deferLater
 from twisted.internet.defer import returnValue  # noqa - used as import target
 from os.path import join as osjoin
@@ -1819,7 +1820,7 @@ def m_len(target):
     back to normal len for other objects.
 
     Args:
-        target (string): A string with potential MXP components
+        target (str): A string with potential MXP components
             to search.
 
     Returns:
@@ -1834,6 +1835,35 @@ def m_len(target):
     return len(target)
 
 
+def display_len(target):
+    """
+    Calculate the 'visible width' of text. This is not necessarily the same as the
+    number of characters in the case of certain asian characters. This will also
+    strip MXP patterns.
+
+    Args:
+        target (string): A string with potential MXP components
+            to search.
+
+    Return:
+        int: The visible width of the target.
+
+    """
+    # Would create circular import if in module root.
+    from evennia.utils.ansi import ANSI_PARSER
+
+    if inherits_from(target, str):
+        # str or ANSIString
+        target = ANSI_PARSER.strip_mxp(target)
+        target = ANSI_PARSER.parse_ansi(target, strip_ansi=True)
+        extra_wide = ("F", "W")
+        return sum(2 if east_asian_width(char) in extra_wide else 1 for char in target)
+    else:
+        return len(target)
+
+
+
+
 # -------------------------------------------------------------------
 # Search handler function
 # -------------------------------------------------------------------