UTF vs Unicode (JQ+)

[Mariusz Szurnacki]

Question ID :988397479953
Which of the following encoding schemes is used by the jvm internally for storing identifiers etc.?
- Unicode
- UTF8
- ASCII
- 8859_1
- It depends on the platform.
I answered Unicode, but it's the wrong answer: according to JQ+ it should be UTF8. But I think I was right, cos
inside JVM, text is represented in 16 bit Unicode and for I/O, UTF is used.
Could you put me right?
Have a nice day,
Mariusz

[Fei Ng]

For efficient reason they use UTF8 since unicode is not
particularly space efficient. But VM does translate them externally back from UTF8 to Unicode efficiently.
Correct me if i am wrong.

[Marcus Green]

Characters stored as Unicode always occupy 2 bytes. UTF8 is a way of storing both Unicode and ASCII text. If the text is within the ASCII range it will occupy 1 byte, if it is larger than the 1 byte range of ASCII it will use the Unicode encoding scheme and occupy more than 1 byte.
As much of the worlds text is stored within the range of ASCII UTF8 offers considerable space saving whilst allowing the huge character representation of the Unicode encoding scheme.
Marcus

------------------

http://www.jchq.net Mock Exams, FAQ,
Tutorial, Links, Book reviews
Java 2 Exam Prep, 2nd Edition by Bill Brogden and Marcus Green
=================================================
JCHQ, Almost as good as JavaRanch
=================================================

[Mariusz Szurnacki]

Hi again!!!
Thanks for your answers. I know what is Unicode and UTF, but I'm still not sure about the right answer to my question: "Which of the following encoding schemes (Unicode or UTF) is used by the jvm internally for storing identifiers etc.?".
According to RHE:
"Java uses two kinds of text representation:
- Unicode for internal representation of characters and strings
- UTF for input and output.
(...)
The outside-the-computer format for Unicode is known as UTF.".
So I think we all are sure that Java�s char data type uses Unicode encoding (and in this way String class too), and UTF is used for I/O.
But which encoding is used by the jvm internally for storing identifiers?
Have a nice day,
Mariusz

[Paul Anilprem]

Please read section "4.4.7 The CONSTANT_Utf8_info Structure
" of JVM spec. ( http://java.sun.com/docs/books/vmspec/2nd-edition/html/ClassFile.doc.html#7963 )
-Paul.
------------------
SCJP2, SCWCD Resources, Free Question A Day, Mock Exam Results and More!
www.jdiscuss.com
Get Certified, Guaranteed!
JQPlus - For SCJP2
JWebPlus - For SCWCD
JDevPlus - For SCJD

[Paul Anilprem]

The method names, field names etc. are all represented using this CONSTANT_Utf8_info .
-Paul.

[Marcus Green]

I have read that explanation in RHE several times and I have concluded it doesn't tell me much at all. I did lot of research on the web to find supporting information without any luck at all. I have a copy of the excellent Rusty Harold I/O book and that doesn't throw much light on the topic.
I have not heard of this topic coming up in the exam, even though the objectives imply that it might.
Mr Earnest?
Marcus
------------------

<A HREF="http://www.jchq.net</A>" TARGET=_blank>http://www.jchq.net[/URL] Mock Exams, FAQ,
Tutorial, Links, Book reviews
Java 2 Exam Prep, 2nd Edition by Bill Brogden and Marcus Green
=================================================
JCHQ, Almost as good as JavaRanch
=================================================
[This message has been edited by Marcus Green (edited November 06, 2001).]

[Mariusz Szurnacki]

Thanks Paul!!!

[Jose Botella]

But UTF8 is not used by the JVM but a modified version of it, so I don't think the answer UTF8 is right either.
I guess this question has not an exact aswer, and besides that maybe it is not likely to appear in the exam:

I think that a SCJP-to-be should know that the source of a java program can utilize Unicode for String and character literals, identifiers and comments. But I don't think that she/he should know the exact format by which descriptors, special strings and the content of Strings is stored whithin the JVM.

[Paul Anilprem]

Well yes, the spec. does say that this is a little different than "standard" UTF8 but all over the place they still call it UTF8.