From 4a128b8969fe4d720f50c1fdb68f0265af8a7117 Mon Sep 17 00:00:00 2001
From: Thijs Schreijer <thijs@thijsschreijer.nl>
Date: Thu, 6 Feb 2025 16:52:08 +0100
Subject: feat(terminal): also accept codepoint integers for width check

Lua utf8 functions return codepoints, hence it makes sense to accept
those, instead of having to convert to utf8 string and back again.
---
 CHANGELOG.md          |  2 ++
 spec/04-term_spec.lua | 25 +++++++++++++++++++++++++
 src/term.c            | 47 ++++++++++++++++++++++++++++++++---------------
 3 files changed, 59 insertions(+), 15 deletions(-)

diff --git a/CHANGELOG.md b/CHANGELOG.md
index 7ae7189..25114c3 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -30,6 +30,8 @@ The scope of what is covered by the version number excludes:
 
 ### unreleased
 
+- Feat: when detecting character display width, also accept unicode codepoints (integers),
+  since the Lua utf8 library returns codepoints, not strings
 - Fix: NetBSD fix compilation, undeclared directives
 
 ### version 0.4.5, released 18-Dec-2024
diff --git a/spec/04-term_spec.lua b/spec/04-term_spec.lua
index 813947a..907f903 100644
--- a/spec/04-term_spec.lua
+++ b/spec/04-term_spec.lua
@@ -511,11 +511,18 @@ describe("Terminal:", function()
 
   describe("utf8cwidth()", function()
 
+    -- utf-8 strings
     local ch1 = string.char(226, 130, 172)       -- "€"   single
     local ch2 = string.char(240, 159, 154, 128)  -- "🚀"  double
     local ch3 = string.char(228, 189, 160)       -- "你"  double
     local ch4 = string.char(229, 165, 189)       -- "好"  double
 
+    -- unicode codepoints
+    local cp1 = 8364    -- "€"   single
+    local cp2 = 128640  -- "🚀"  double
+    local cp3 = 20320   -- "你"  double
+    local cp4 = 22909   -- "好"  double
+
     it("handles zero width characters", function()
       assert.same({0}, {system.utf8cwidth("")}) -- empty string returns 0-size
       assert.same({nil, 'Character width determination failed'}, {system.utf8cwidth("\a")})  -- bell character
@@ -539,6 +546,24 @@ describe("Terminal:", function()
       assert.same({2}, {system.utf8cwidth(ch2 .. ch3 .. ch4)})
     end)
 
+    it("handles integer codepoints", function()
+      assert.same({1}, {system.utf8cwidth(cp1)})
+      assert.same({2}, {system.utf8cwidth(cp2)})
+      assert.same({2}, {system.utf8cwidth(cp3)})
+      assert.same({2}, {system.utf8cwidth(cp4)})
+    end)
+
+    it("returns an error on bad argument", function()
+      assert.has.error(function()
+        system.utf8cwidth(true)
+      end, "bad argument #1 to 'utf8cwidth' (Expected UTF-8-string or codepoint-integer as first argument)")
+    end)
+
+    it("returns an error on bad unicode values", function()
+      assert.same({nil, "Invalid Unicode codepoint"}, {system.utf8cwidth(-10)})
+      assert.same({nil, "Invalid Unicode codepoint"}, {system.utf8cwidth(999999999999)})
+    end)
+
   end)
 
 
diff --git a/src/term.c b/src/term.c
index 8c2b87a..2375080 100644
--- a/src/term.c
+++ b/src/term.c
@@ -953,30 +953,47 @@ int utf8_to_wchar(const char *utf8, size_t len, mk_wchar_t *codepoint) {
 /***
 Get the width of a utf8 character for terminal display.
 @function utf8cwidth
-@tparam string utf8_char the utf8 character to check, only the width of the first character will be returned
+@tparam string|int utf8_char the utf8 character, or unicode codepoint, to check, only the width of the first character will be returned
 @treturn[1] int the display width in columns of the first character in the string (0 for an empty string)
 @treturn[2] nil
 @treturn[2] string error message
 */
 int lst_utf8cwidth(lua_State *L) {
-    const char *utf8_char;
-    size_t utf8_len;
-    utf8_char = luaL_checklstring(L, 1, &utf8_len);
     int width = 0;
-
     mk_wchar_t wc;
 
-    if (utf8_len == 0) {
-        lua_pushinteger(L, 0);
-        return 1;
-    }
+    if (lua_type(L, 1) == LUA_TSTRING) {
+        // Handle UTF8 as string input
+        const char *utf8_char;
+        size_t utf8_len;
+        utf8_char = luaL_checklstring(L, 1, &utf8_len);
 
-    // Convert the UTF-8 string to a wide character
-    int bytes_processed = utf8_to_wchar(utf8_char, utf8_len, &wc);
-    if (bytes_processed == -1) {
-        lua_pushnil(L);
-        lua_pushstring(L, "Invalid UTF-8 character");
-        return 2;
+        if (utf8_len == 0) {
+            lua_pushinteger(L, 0);
+            return 1;
+        }
+
+        // Convert the UTF-8 string to a wide character
+        int bytes_processed = utf8_to_wchar(utf8_char, utf8_len, &wc);
+        if (bytes_processed == -1) {
+            lua_pushnil(L);
+            lua_pushstring(L, "Invalid UTF-8 character");
+            return 2;
+        }
+
+    } else if (lua_type(L, 1) == LUA_TNUMBER) {
+        // Handle codepoint input
+        int codepoint = luaL_checkinteger(L, 1);
+
+        if (codepoint < 0 || codepoint > 0x10FFFF) {
+            lua_pushnil(L);
+            lua_pushstring(L, "Invalid Unicode codepoint");
+            return 2;
+        }
+        wc = (mk_wchar_t)codepoint;
+
+    } else {
+        return luaL_argerror(L, 1, "Expected UTF-8-string or codepoint-integer as first argument");
     }
 
     // Get the width of the wide character
-- 
cgit v1.2.3-55-g6feb