fix: japanese characters not truncated correctly in tui terminal output

2 files changed, 25 insertions, 10 deletions

diff --git a/app/ui/tui.py b/app/ui/tui.py
index b22a186..60e2dda 100644
--- a/app/ui/tui.py
+++ b/app/ui/tui.py
@@ -37,6 +37,7 @@ import contextlib
 import os
 import re
 import textwrap
+import unicodedata
 from pathlib import Path
 from typing import Callable, List, Optional, Sequence, Tuple, Union
 
@@ -255,13 +256,31 @@ def _hline(scr, y: int, x: int, n: int, attr: int = 0) -> None:
         pass
 
 
+def _char_width(ch: str) -> int:
+    """Terminal cells CH occupies (East Asian Wide/Fullwidth count 2)."""
+    return 2 if unicodedata.east_asian_width(ch) in ("W", "F") else 1
+
+
+def _disp_width(text: str) -> int:
+    """Terminal cells TEXT occupies (East Asian Wide/Fullwidth count 2)."""
+    return sum(_char_width(ch) for ch in text)
+
+
 def _fit(text: str, width: int) -> str:
     """Truncate TEXT to WIDTH columns, appending '~' when cut."""
     if width < 1:
         return ""
-    if len(text) <= width:
+    if _disp_width(text) <= width:
         return text
-    return text[: max(0, width - 1)] + "~"
+    out: List[str] = []
+    used = 0
+    for ch in text:
+        w = _char_width(ch)
+        if used + w > width - 1:
+            break
+        out.append(ch)
+        used += w
+    return "".join(out) + "~"
 
 
 def _wrap_segments(segments: Sequence[Tuple[str, int]], width: int
@@ -283,13 +302,13 @@ def _wrap_segments(segments: Sequence[Tuple[str, int]], width: int
     current: List[Tuple[str, int]] = []
     used = 0
     for text, attr in tokens:
-        if current and used + len(text) > width:
+        if current and used + _disp_width(text) > width:
             lines.append(current)
             current, used = [], 0
             if text.isspace():
                 continue  # the break eats the whitespace it happens at
         current.append((text, attr))
-        used += len(text)
+        used += _disp_width(text)
     if current:
         lines.append(current)
     return lines or [[]]
diff --git a/app/ui/viewkit.py b/app/ui/viewkit.py
index 301a0af..e54aada 100644
--- a/app/ui/viewkit.py
+++ b/app/ui/viewkit.py
@@ -217,11 +217,7 @@ def _sep(scr, curses, theme, y, width) -> None:
 
 def _fit(text: str, width: int) -> str:
     """Truncate TEXT to WIDTH columns, appending '~' when cut."""
-    if width < 1:
-        return ""
-    if len(text) <= width:
-        return text
-    return text[: max(0, width - 1)] + "~"
+    return tui._fit(text, width)
 
 
 def _wrap(text: str, width: int) -> List[str]:
@@ -230,7 +226,7 @@ def _wrap(text: str, width: int) -> List[str]:
     current = ""
     for word in text.split():
         candidate = f"{current} {word}".strip()
-        if len(candidate) <= max(10, width):
+        if tui._disp_width(candidate) <= max(10, width):
             current = candidate
         else:
             if current: