Unicode input kitten: Add symbols from NERD font

These are mostly Private Use symbols not in any standard,
however they are common enough to be useful.

Fixes #2972

docs/changelog.rst

@@ -28,6 +28,9 @@ To update |kitty|, :doc:`follow the instructions <binary>`.
 - Implement special rendering for various characters from the set of "Symbols
   for Legacy Computing" from the Unicode 13 standard
 
+- Unicode input kitten: Allow choosing symbols from the NERD font as well.
+  These are mostly Private Use symbols not in any standard, however are common. (:iss:`2972`)
+
 - Allow specifying border sizes in either pts or pixels. Change the default to
   0.5pt borders as this works best with the new minimal border style
 

gen-wcwidth.py

@@ -98,6 +98,18 @@ def parse_ucd() -> None:
             elif category.startswith('S'):
                 all_symbols.add(codepoint)
 
+    with open('nerd-fonts-glyphs.txt') as f:
+        for line in f:
+            line = line.strip()
+            if not line or line.startswith('#'):
+                continue
+            code, category, name = line.split(' ', 2)
+            codepoint = int(code, 16)
+            if name and codepoint not in name_map:
+                name_map[codepoint] = name.upper()
+                for word in name.lower().split():
+                    add_word(word, codepoint)
+
     # Some common synonyms
     word_search_map['bee'] |= word_search_map['honeybee']
     word_search_map['lambda'] |= word_search_map['lamda']

kitty/emoji.h

@@ -1,4 +1,4 @@
-// unicode data, built from the unicode standard on: 2020-08-06
+// unicode data, built from the unicode standard on: 2020-09-22
 // see gen-wcwidth.py
 #pragma once
 #include "data-types.h"

kitty/unicode-data.c

@@ -1,4 +1,4 @@
-// unicode data, built from the unicode standard on: 2020-08-06
+// unicode data, built from the unicode standard on: 2020-09-22
 // see gen-wcwidth.py
 #include "data-types.h"
 

kitty/wcwidth-std.h

@@ -1,4 +1,4 @@
-// unicode data, built from the unicode standard on: 2020-08-06
+// unicode data, built from the unicode standard on: 2020-09-22
 // see gen-wcwidth.py
 #pragma once
 #include "data-types.h"