Enhance text handling by adding emoji support and sanitization for curses rendering

Succumb to wcwidth

contact/ui/nav_utils.py

@@ -1,5 +1,6 @@
 import curses
 import re
+import unicodedata
 from typing import Any, Optional, List, Dict
 
 from wcwidth import wcwidth, wcswidth
@@ -9,6 +10,9 @@ from contact.utilities.i18n import t
 from contact.utilities.control_utils import transform_menu_path
 from contact.utilities.singleton import interface_state, ui_state
 
+ZWJ = "\u200d"
+KEYCAP = "\u20e3"
+
 
 def get_node_color(node_index: int, reverse: bool = False):
     node_num = ui_state.node_list[node_index]
@@ -328,7 +332,90 @@ def get_wrapped_help_text(
     return wrapped_help
 
 
+def _is_regional_indicator(char: str) -> bool:
+    codepoint = ord(char)
+    return 0x1F1E6 <= codepoint <= 0x1F1FF
+
+
+def _is_variation_selector(char: str) -> bool:
+    codepoint = ord(char)
+    return 0xFE00 <= codepoint <= 0xFE0F or 0xE0100 <= codepoint <= 0xE01EF
+
+
+def _is_emoji_modifier(char: str) -> bool:
+    codepoint = ord(char)
+    return 0x1F3FB <= codepoint <= 0x1F3FF
+
+
+def _is_display_modifier(char: str) -> bool:
+    return unicodedata.category(char) in {"Mn", "Mc", "Me"} or _is_emoji_modifier(char)
+
+
+def iter_display_units(text: str) -> List[str]:
+    """Split text into display units so emoji sequences stay intact."""
+    units: List[str] = []
+    index = 0
+
+    while index < len(text):
+        unit = text[index]
+        index += 1
+
+        if _is_regional_indicator(unit) and index < len(text) and _is_regional_indicator(text[index]):
+            unit += text[index]
+            index += 1
+
+        while index < len(text) and _is_display_modifier(text[index]):
+            unit += text[index]
+            index += 1
+
+        while index < len(text) and text[index] == ZWJ and index + 1 < len(text):
+            unit += text[index]
+            index += 1
+            unit += text[index]
+            index += 1
+
+            while index < len(text) and _is_display_modifier(text[index]):
+                unit += text[index]
+                index += 1
+
+        units.append(unit)
+
+    return units
+
+
+def sanitize_for_curses(text: str) -> str:
+    """Collapse complex emoji sequences to stable fallbacks before rendering."""
+    sanitized: List[str] = []
+
+    for unit in iter_display_units(text):
+        if ZWJ not in unit and KEYCAP not in unit and not any(
+            _is_variation_selector(char) or _is_emoji_modifier(char) for char in unit
+        ):
+            sanitized.append(unit)
+            continue
+
+        visible = [
+            char
+            for char in unit
+            if char != ZWJ and char != KEYCAP and not _is_variation_selector(char) and not _is_emoji_modifier(char)
+        ]
+
+        if KEYCAP in unit and visible:
+            sanitized.append(visible[0])
+        elif ZWJ in unit and visible:
+            sanitized.append(visible[0])
+        elif any(_is_emoji_modifier(char) for char in unit) and visible:
+            sanitized.append(visible[0])
+        elif visible:
+            sanitized.append("".join(visible))
+        else:
+            sanitized.append(unit)
+
+    return "".join(sanitized)
+
+
 def text_width(text: str) -> int:
+    text = sanitize_for_curses(text)
     width = wcswidth(text)
     if width >= 0:
         return width
@@ -337,12 +424,13 @@ def text_width(text: str) -> int:
 
 def slice_text_to_width(text: str, max_width: int) -> str:
     """Return the longest prefix that fits within max_width terminal cells."""
+    text = sanitize_for_curses(text)
     if max_width <= 0:
         return ""
 
     chunk = ""
-    for char in text:
-        candidate = chunk + char
+    for unit in iter_display_units(text):
+        candidate = chunk + unit
         if text_width(candidate) > max_width:
             break
         chunk = candidate
@@ -352,6 +440,8 @@ def slice_text_to_width(text: str, max_width: int) -> str:
 
 def fit_text(text: str, width: int, suffix: str = "") -> str:
     """Trim and pad text so its terminal display width fits exactly."""
+    text = sanitize_for_curses(text)
+    suffix = sanitize_for_curses(suffix)
     if width <= 0:
         return ""
 
@@ -366,17 +456,18 @@ def fit_text(text: str, width: int, suffix: str = "") -> str:
 
 def split_text_to_width(text: str, max_width: int) -> List[str]:
     """Split text into chunks that each fit within max_width terminal cells."""
+    text = sanitize_for_curses(text)
     if max_width <= 0:
         return [""]
 
     chunks: List[str] = []
     chunk = ""
 
-    for char in text:
-        candidate = chunk + char
+    for unit in iter_display_units(text):
+        candidate = chunk + unit
         if chunk and text_width(candidate) > max_width:
             chunks.append(chunk)
-            chunk = char
+            chunk = unit
         else:
             chunk = candidate
 
@@ -392,6 +483,7 @@ def split_text_to_width(text: str, max_width: int) -> List[str]:
 
 def wrap_text(text: str, wrap_width: int) -> List[str]:
     """Wraps text while preserving spaces and breaking long words."""
+    text = sanitize_for_curses(text)
 
     whitespace = "\t\n\x0b\x0c\r "
     whitespace_trans = dict.fromkeys(map(ord, whitespace), ord(" "))

pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "contact"
-version = "1.4.19"
+version = "1.4.20"
 description = "This Python curses client for Meshtastic is a terminal-based client designed to manage device settings, enable mesh chat communication, and handle configuration backups and restores."
 authors = [
     {name = "Ben Lipsey",email = "ben@pdxlocations.com"}