Unihan Database from
	https://www.unicode.org/Public/UNIDATA/
Unihan_DictionaryLikeData.txt
	has all the four corners info

Logo genrrated here:
	https://www.zhuanshuti.cn/3

grep kFourCornerCode
delete comment line
:%s/\(.*\s.*\s\)\(.*\)\s\(.*\)/\1\2\r\1\3/
	removes duplicate four corners

clean it up so its
12345 U+212121
then turn to echo $'123445 \u12341'
using two seperate subsitutes for 4 and 5 char
change to 8 character length

then use the following to convert to actual unicode
:%s/^\(.*\)\t\(.*\)$/echo -e "\1\\t$(echo \2 |xxd -r -ps -u | iconv -f UTF-32BE -t UTF-8)"/

xxd -r -ps -u | iconv -f UTF-32BE -t UTF-8
	coverts fro U+code to normal, need to pad to 32bits

create tempalte according to usr/share/ibus-input/tables/template.txt


cat four | awk '{print $1}' | sort | uniq -c | sort -n
	counts 71 conflicts at most