Java \u-escape sequence Conversion

To our Java Gurus - how can I convert ASCII String that contains strictly \u-escape sequence into *true* UTF-8 (multi) bytes?

I've tried SUN's sample from here. The sample works as it is, my issue however is that in there Java Compiler apparently
takes care of \u-escape Literals. If I supply true ASCII source from File or DB run-time (like for example "\u00ea\u00f1\u00fc") the
resulting Byte sequence is identical to the Source (ASCII to ASCII, no surprise here)

Hints, Tips, RTFMs warmly welcomed

Find more posts tagged with

Comments

Migrateduser

Java compiler does not run the byte code, java virtual machine runs the byte code. So not exactly sure what is the issue in your original post. You can use a binary stream to get the raw data from a file.

ISCBorisB

Java compiler does not run the byte code, java virtual machine runs the byte code.

Wow, who would have thought... Thanks for revelation.

So not exactly sure what is the issue in your original post

If you do not understand the question, why do you feel obliged to chime in? Do you even know what u-escape sequence is?

nipper

Wow, who would have thought... Thanks for revelation.If you do not understand the question, why do you feel obliged to chime in? Do you even know what u-escape sequence is?

Ahhh yes, someone else from the Smitty school for making friends and influencing people. Smiley Very Happy

ISCBorisB

Ahhh yes, someone else from the Smitty school for making friends and influencing people.

I've graduated from Smitty school with honors! Admit it Andy... It's hard to resist. You ask a very specific questions
and somebody chimes in with Java trivia, how to read File and all that nice and useless Jazz

nipper

I've graduated from Smitty school with honors! Admit it Andy... It's hard to resist. You ask a very specific questions
and somebody chimes in with Java trivia, how to read File and all that nice and useless Jazz

Well, we will have to wait to see if Professor Smitty agrees with the "honors" title.

Migrateduser

If I supply true ASCII source from File or DB run-time (like for example "\u00ea\u00f1\u00fc") the
resulting Byte sequence is identical to the Source (ASCII to ASCII, no surprise here)

Could you try a Reader/FileInputStreamReader for UTF-8 for 'files' example;

http://snippets.dzone.com/posts/show/371

For the DB assuming you are using good old result sets, if the corresponding column is mapped to Java 'String' you shouldn't have a problem, unless I'm missing something?

ISCBorisB

The issue has been resolved.

I ended up doing it by brute force: split up ASCII strings by \U, translate chunks to Unicode scalar
values (code points) and finally converting to UTF-8 bytes. To my surprise Sun does not provide
Class(es) even for the last step, sigh...