From 994a2196745637fd0805132711b1ed9354b75426 Mon Sep 17 00:00:00 2001
From: salmonik <salmonik@tuta.io>
Date: Fri, 31 Jul 2026 17:28:31 +0200
Subject: vt: decode UTF-8, render box-drawing/arrow glyphs procedurally
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The editor's gutter '│' (U+2502) appeared as '???' — write_bytes fed
each UTF-8 byte to the state machine, which mapped every byte >= 0x80
to '?', and the Latin-1 font has no box-drawing glyphs anyway.

- vt.write_bytes now decodes UTF-8 sequences; printable non-ASCII
  dispatches to the glyph path instead of the byte parser.
- gop: unicode_glyph_bits() procedurally rasterises box-drawing
  (U+2500-257F light/heavy/double), arrows (U+2190-2193) and the em
  dash (U+2014) — pure geometry, no font data needed; blit_glyph falls
  back to it for ch >= 256.

Verified via framebuffer dumps: editor gutter │ renders as a vertical
line, help screen ═ separators and ←→↑↓ all render as lines (no '?').
Shell banner em dash renders too.
---
 rsnix/src/vt.rs | 87 ++++++++++++++++++++++++++++++++++++++++++++++-----------
 1 file changed, 70 insertions(+), 17 deletions(-)

(limited to 'rsnix/src/vt.rs')

diff --git a/rsnix/src/vt.rs b/rsnix/src/vt.rs
index eb60e11..3b81d68 100644
--- a/rsnix/src/vt.rs
+++ b/rsnix/src/vt.rs
@@ -48,7 +48,11 @@ struct Cell {
     bg: u8,
 }
 
-const BLANK: Cell = Cell { ch: 32, fg: 7, bg: 0 };
+const BLANK: Cell = Cell {
+    ch: 32,
+    fg: 7,
+    bg: 0,
+};
 
 // ---- screen state ----------------------------------------------------------
 
@@ -115,7 +119,9 @@ pub fn size() -> (usize, usize) {
 }
 
 pub fn set_cursor_visible(v: bool) {
-    unsafe { CURSOR_VISIBLE = v; }
+    unsafe {
+        CURSOR_VISIBLE = v;
+    }
     flush();
 }
 
@@ -180,15 +186,7 @@ fn put_byte(b: u8) {
                     CURSOR_C = (CURSOR_C + n).min(COLS - 1);
                 }
                 0x00..=0x1f => {} // other control chars: ignored
-                _ => {
-                    let ch = if b < 0x80 { b as u16 } else { b'?' as u16 };
-                    put_cell(CURSOR_R, CURSOR_C, ch);
-                    CURSOR_C += 1;
-                    if CURSOR_C >= COLS {
-                        CURSOR_C = 0;
-                        newline();
-                    }
-                }
+                _ => put_printable(b as u16),
             },
             Parser::Escape => match b {
                 b'[' => {
@@ -234,6 +232,34 @@ fn param(i: usize, default: usize) -> usize {
     }
 }
 
+/// Place a printable glyph at the cursor and advance (wrapping/scrolling).
+/// `ch` is the character code: ASCII or a Unicode codepoint that the
+/// rasteriser can draw (box-drawing / arrows are rendered procedurally).
+fn put_printable(ch: u16) {
+    unsafe {
+        put_cell(CURSOR_R, CURSOR_C, ch);
+        CURSOR_C += 1;
+        if CURSOR_C >= COLS {
+            CURSOR_C = 0;
+            newline();
+        }
+    }
+}
+
+/// Dispatch one character to the terminal: ASCII goes through the byte
+/// state machine (control/escape handling), printable non-ASCII goes
+/// straight to the glyph path (replaced by '?' when not drawable).
+fn write_char(c: char) {
+    let u = c as u32;
+    if u < 0x80 {
+        put_byte(u as u8);
+    } else if (0x2500..=0x257f).contains(&u) || (0x2190..=0x2193).contains(&u) || u == 0x2014 {
+        put_printable(c as u16);
+    } else {
+        put_byte(b'?');
+    }
+}
+
 fn csi_final(b: u8) {
     match b {
         // Cursor position (1-based; 0 means 1)
@@ -392,19 +418,46 @@ fn scroll_up() {
 // public writing API
 // ---------------------------------------------------------------------------
 
-/// Write raw bytes (ANSI-capable).  Used by full-screen apps.
+/// Write raw bytes (ANSI-capable).  Used by full-screen apps.  UTF-8
+/// sequences are decoded so Unicode glyphs (box-drawing, arrows) render;
+/// undecodable bytes become '?'.
 pub fn write_bytes(b: &[u8]) {
-    for &x in b {
-        put_byte(x);
+    let mut i = 0;
+    while i < b.len() {
+        let byte = b[i];
+        if byte < 0x80 {
+            put_byte(byte);
+            i += 1;
+            continue;
+        }
+        // Decode the next UTF-8 character starting at i.
+        match core::str::from_utf8(&b[i..]) {
+            Ok(s) => {
+                let c = s.chars().next().unwrap_or('?');
+                write_char(c);
+                i += c.len_utf8();
+            }
+            Err(e) => {
+                let valid = e.valid_up_to();
+                if valid > 0 {
+                    let s = unsafe { core::str::from_utf8_unchecked(&b[i..i + valid]) };
+                    let c = s.chars().next().unwrap_or('?');
+                    write_char(c);
+                    i += c.len_utf8();
+                } else {
+                    put_byte(b'?');
+                    i += 1;
+                }
+            }
+        }
     }
     flush();
 }
 
-/// Write a UTF-8 string (plain text; non-ASCII becomes '?').
+/// Write a UTF-8 string (plain text; Unicode glyphs render, others '?').
 pub fn write_str(s: &str) {
     for c in s.chars() {
-        let b = if (c as u32) < 0x80 { c as u8 } else { b'?' };
-        put_byte(b);
+        write_char(c);
     }
     flush();
 }
-- 
cgit v1.2.3