From 6190ae1071c8d07a9a2d10037801838718363e1f Mon Sep 17 00:00:00 2001
From: bptato <nincsnevem662@gmail.com>
Date: Sun, 5 Feb 2023 16:24:56 +0100
Subject: Update character width ranges

Use the Unicode 15 EastAsianWidth.txt file instead of an ancient wcwidth
implementation's ranges.
---
 src/data/charwidth.nim | 42 ++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 42 insertions(+)
 create mode 100644 src/data/charwidth.nim

(limited to 'src/data')

diff --git a/src/data/charwidth.nim b/src/data/charwidth.nim
new file mode 100644
index 00000000..daec31ee
--- /dev/null
+++ b/src/data/charwidth.nim
@@ -0,0 +1,42 @@
+import strutils
+
+const (
+  DoubleWidthRanges*,
+  DoubleWidthAmbiguousRanges*,
+  HalfWidthRanges*
+) = (func(): (
+    seq[(uint32, uint32)],
+    seq[(uint32, uint32)],
+    seq[(uint32, uint32)]
+  ) =
+  template add(firstcol: string, res: var seq[(uint32, uint32)]) =
+    if firstcol.contains(".."):
+      let fcs = firstcol.split("..")
+      let rstart = uint32(parseHexInt(fcs[0]))
+      let rend = uint32(parseHexInt(fcs[1]))
+      res.add((rstart, rend))
+    else:
+      let cp = uint32(parseHexInt(firstcol))
+      res.add((cp, cp))
+  const map = staticRead"res/map/EastAsianWidth.txt"
+  for line in map.split('\n'):
+    if line.len == 0 or line[0] == '#':
+      continue
+    var i = 0
+    var firstcol = ""
+    var status = ""
+    while i < line.len and line[i] notin {'#', ';'}:
+      if line[i] != ' ':
+        firstcol &= line[i]
+      inc i
+    if line[i] != '#':
+      inc i
+    while i < line.len and line[i] notin {'#', ';'}:
+      if line[i] != ' ':
+        status &= line[i]
+      inc i
+    case status
+    of "W": add(firstcol, result[0])
+    of "A": add(firstcol, result[1])
+    of "H": add(firstcol, result[2])
+)()
-- 
cgit 1.4.1-2-gfad0