From a0278729b693a1e563ed317e732e67926410835f Mon Sep 17 00:00:00 2001
From: DTB <trinity@trinity.moe>
Date: Sat, 2 Sep 2023 13:00:50 -0400
Subject: [PATCH] does not work, stuck on problem

---
 toki/toki_sitelen | 88 +++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 88 insertions(+)
 create mode 100755 toki/toki_sitelen

diff --git a/toki/toki_sitelen b/toki/toki_sitelen
new file mode 100755
index 0000000..b9f8fd7
--- /dev/null
+++ b/toki/toki_sitelen
@@ -0,0 +1,88 @@
+#!/bin/sh
+
+# codepoint -> utf8
+while read -r codepoint; do
+	# normalize to U+000000
+	codepoint="$(printf '%s\n' "$codepoint" | sed 's/^U+//')"
+	codepoint="U+$( \
+		dd if=/dev/zero bs=1 count=$( \
+			printf '%s\n' "$codepoint" \
+				| wc -c \
+				| xargs printf '7 - %s\n' \
+				| bc \
+			) 2>/dev/null | tr '\0' 0)$codepoint"
+
+	codepoint_bin="$(printf '%s\n' "$codepoint" \
+		| sed \
+			-e 's/^U+//' -e 's/0/0000/g' -e 's/1/0001/g' \
+			-e 's/2/0010/g' -e 's/3/0011/g' -e 's/4/0100/g' \
+			-e 's/5/0101/g' -e 's/6/0110/g' -e 's/7/0111/g' \
+			-e 's/8/1000/g' -e 's/9/1001/g' -e 's/A/1010/g' \
+			-e 's/B/1011/g' -e 's/C/1100/g' -e 's/D/1101/g' \
+			-e 's/E/1110/g' -e 's/F/1111/g')"
+
+	printf '%s\n' "$codepoint_bin" \
+		| dd bs=17 count=1 2>/dev/null \
+		| grep 1 2>/dev/null 1>&2 \
+			|| bytes=1
+	printf '%s\n' "$codepoint_bin" \
+		| dd bs=13 count=1 2>/dev/null \
+		| grep 1 2>/dev/null 1>&2\
+			|| bytes=2
+	printf '%s\n' "$codepoint_bin" \
+		| dd bs=8 count=1 2>/dev/null \
+		| grep 1 2>/dev/null 1>&2 \
+			&& bytes=4 \
+			|| bytes=3
+
+	# TODO: How to bring bin,oct, or hex to actual binary in POSIX?
+
+	utf8_bin="$(case $bytes in \
+	1) printf '0%s\n' "$(printf '%s\n' "$codepoint_bin" \
+			| tail -c 7)" ;; \
+	2) printf '110%s10%s\n' "$(printf '%s\n' "$codepoint_bin" \
+			| tail -c 11 | head -c 5)" \
+		"$(printf '%s\n' "$codepoint_bin" \
+			| tail -c 6)" ;; \
+	3) printf '1110%s10%s10%s\n' "$(printf '%s\n' "$codepoint_bin" \
+			| tail -c 16 | head -c 4)" \
+		"$(printf '%s\n' "$codepoint_bin" \
+			| tail -c 12 | head -c 6)" \
+		"$(printf '%s\n' "$codepoint_bin" \
+			| tail -c 6)" ;; \
+	4) printf '11110%s10%s10%s10%s\n' "$(printf '%s\n' "$codepoint_bin" \
+			| tail -c 21 | head -c 3)" \
+		"$(printf '%s\n' "$codepoint_bin" \
+			| tail -c 18 | head -c 6)" \
+		"$(printf '%s\n' "$codepoint_bin" \
+			| tail -c 12 | head -c 6)" \
+		"$(printf '%s\n' "$codepoint_bin" \
+			| tail -c 6)" ;; \
+	esac)"
+
+	utf8_oct="$(dd if=/dev/zero bs=1 count=$( \
+			printf '%s\n' "$utf8_bin" \
+				| wc -c \
+				| xargs printf '34 - %s\n' \
+				| bc \
+			) 2>/dev/null \
+		| tr '\0' 0 \
+		| xargs printf "%s$utf8_bin\n" \
+		| sed 's/.../& /g' \
+		| sed \
+			-e 's/000/0/g' -e 's/001/1/g' -e 's/010/2/g' \
+			-e 's/011/3/g' -e 's/100/4/g' -e 's/101/5/g' \
+			-e 's/110/6/g' -e 's/111/7/g' \
+		| tr -d ' ')"
+
+	utf8_hex="$(printf '%s\n' "$utf8_bin" \
+		| sed \
+			-e 's/0000/0/g' -e 's/0001/1/g' -e 's/0010/2/g' \
+			-e 's/0011/3/g' -e 's/0100/4/g' -e 's/0101/5/g' \
+			-e 's/0110/6/g' -e 's/0111/7/g' -e 's/1000/8/g' \
+			-e 's/1001/9/g' -e 's/1010/A/g' -e 's/1011/B/g' \
+			-e 's/1100/C/g' -e 's/1101/D/g' -e 's/1110/E/g' \
+			-e 's/1111/F/g')"
+
+	shift
+done