Hi.

Although I was trying implements of Japanese Lexer to 6.8 beta, I have noticed that the result of a ft:mark() function shifts. 

55 line of https://github.com/BaseXdb/basex/blob/master/src/main/java/org/basex/query/util/DataFTBuilder.java  

final FTLexer lex = new FTLexer(ftopt).sc().init(nd.atom()); 

FTLexer will generate the tokenizer of the default language instead of an original language, if it does not perform in this way. 

How is it? 

In code which I am correcting, ft:mark() operated normally by carrying out like this. 

thanks.

--

Toshio HIRAI