What is the most accurate encoding detector? [closed]
I’ve checked juniversalchardet and ICU4J on some CSV files, and the results are inconsistent: juniversalchardet had better results: UTF-8: Both detected. Windows-1255: juniversalchardet detected when it had enough hebrew letters, ICU4J still thought it was ISO-8859-1. With even more hebrew letters, ICU4J detected it as ISO-8859-8 which is the other hebrew encoding(and so the text … Read more