ASCII Chars

[SamMan]

This is a *very* simple question, I know, but how would you either display,
or check for a value by using an ASCII character?

Thanks,
Sam

 

[Marco Schmidt]

SamMan:

This is a *very* simple question, I know, but how would you either display,
or check for a value by using an ASCII character?

Just compare if the character falls into the right interval. Values 0
to 127 are ASCII, but the first 32 are non-printing
<http://www.asciitable.com/>. So this may be what you want:

public static boolean isAscii(char c) {
return c >= 32 && c < 128;
}

Maybe

return (c & 224) == 96;

works as well... Can't think, must eat first.

Regards,
Marco

 

[Jon A. Cruz]

This is a *very* simple question, I know, but how would you either display,
or check for a value by using an ASCII character?

What exactly do you mean by "ASCII character?"

I ask because it's often a misused label, and with Java it does indeed
make a difference.

 

[John C. Bollinger]

SamMan:

This is a *very* simple question, I know, but how would you either display,
or check for a value by using an ASCII character?

[...]

Maybe

return (c & 224) == 96;

works as well... Can't think, must eat first.

Nope. (How about c == 65 ?). I don't see how to do it with a single
bitwise operation, but I think this would work:

return (((c & 160) == 32) || ((c & 192) == 64));

You could also try

return ((c & 224) ^ 96) != 96;

if that appeals to you.


John Bollinger
(e-mail address removed)

 

[SamMan]

What exactly do you mean by "ASCII character?"

I ask because it's often a misused label, and with Java it does indeed
make a difference.

Sorry... I should have said ASCII value. As in the Dec, Hx or Oct value of,
say #.

Sam

 

[Jon A. Cruz]

SamMan:

This is a *very* simple question, I know, but how would you either
display,
or check for a value by using an ASCII character?

[...]

Maybe
return (c & 224) == 96;

works as well... Can't think, must eat first.

Nope. (How about c == 65 ?). I don't see how to do it with a single
bitwise operation, but I think this would work:

[SNIP]

Uh oh.


You just mixed the concept of encoded character with language
conversions. Might not always work properly.

 

[Jon A. Cruz]

Sorry... I should have said ASCII value. As in the Dec, Hx or Oct value of,
say #.

Again, that sounds like you are not using the term correctly.


Let's try two different expansions:


A)
how would you either display, or check for a value by using a 7-bit
value ranging from 0 through 127 that is only mainly good for US
English, Hawaiian, Swahili and Latin value?


B)
how would you either display, or check for a value by using the numeric
value of some character?



I think the latter is what you mean.


Remember, MS Windows does not do "ASCII". It does "ANSI". Java does not
use ASCII, it uses Unicode, which has 16-bit unsigned values ranging
from 0 through 65536.

There is no ASCII value for 'ñ', nor for 'µ', '±', 'ß', '£', '¥', '¢'...

However, there are Latin-1 values, Unicode values, Cp1252 values and
MacRoman values for all of those.




Now, if you mean "numerical value" and not "ASCII", then it's simple.

Java characters are 16-bit unsigned integers.

int i = 'A';
char c = 65;
char c2 = 'A'

if ( 'A' == i )
....

 

[SamMan]

Again, that sounds like you are not using the term correctly.

snip...

Thanks for your help. I just realized (I think) that I would be trying to
compare a substring to a char, in this case the numerical value of one. I
would have to cast the substring to a char before making the comparison,
right? If this is the case, it would be simpler, less confusing and less
code to just compare Strings.

Sam

 

[Jon A. Cruz]

Thanks for your help. I just realized (I think) that I would be trying to
compare a substring to a char, in this case the numerical value of one. I
would have to cast the substring to a char before making the comparison,
right? If this is the case, it would be simpler, less confusing and less
code to just compare Strings.

Depends on what the *meaning* of what you are trying to do is.

First of all, a numerical value of 1 is very suspicious. Sounds like
you've brought over the bad C/C++ habit of mixing byte data in strings.

Is the information you have truely a sequence of characters? Or is it a
sequence of byte values? If the former, use a String; if it is the
latter, use a byte[].


You can't cast a substring to a char, however you can call several handy
methods on the string, including indexOf().

Check the API docs:

http://java.sun.com/j2se/1.4.2/docs/api/java/lang/String.html


Remember, if you convert to or from bytes, BEWARE!!!! That's veeeeery
easy to have corrupted data in doing so.

 

[Tukla Ratte]

SamMan:

This is a *very* simple question, I know, but how would you either display,
or check for a value by using an ASCII character?

[...]

Maybe

return (c & 224) == 96;

works as well... Can't think, must eat first.

Nope. (How about c == 65 ?). I don't see how to do it with a single
bitwise operation, but I think this would work:

return (((c & 160) == 32) || ((c & 192) == 64));

You could also try

return ((c & 224) ^ 96) != 96;

if that appeals to you.

Oy vey.

If I ran across any code like that without some sort comment
explaining what it's for, someone would have to die. >

 

[Joona I Palaste]

Again, that sounds like you are not using the term correctly.

Let's try two different expansions:

A)
how would you either display, or check for a value by using a 7-bit
value ranging from 0 through 127 that is only mainly good for US
English, Hawaiian, Swahili and Latin value?

B)
how would you either display, or check for a value by using the numeric
value of some character?

I think the latter is what you mean.

Remember, MS Windows does not do "ASCII". It does "ANSI". Java does not
use ASCII, it uses Unicode, which has 16-bit unsigned values ranging
from 0 through 65536.

There is no ASCII value for 'ñ', nor for 'µ', '±', 'ß', '£', '¥', '¢'...

However, there are Latin-1 values, Unicode values, Cp1252 values and
MacRoman values for all of those.

I know at least one person who advocates the use of "ASCII" to mean the
very concept of organising character glyphs in a deterministic countable
order.

--
/-- Joona Palaste ([email protected]) ---------------------------\
| Kingpriest of "The Flying Lemon Tree" G++ FR FW+ M- #108 D+ ADA N+++|
| http://www.helsinki.fi/~palaste W++ B OP+ |
\----------------------------------------- Finland rules! ------------/
"That's no raisin - it's an ALIEN!"
- Tourist in MTV's Oddities

 

[Jon Skeet]

I know at least one person who advocates the use of "ASCII" to mean the
very concept of organising character glyphs in a deterministic countable
order.

On what grounds?

 

[Joona I Palaste]

On what grounds?

He didn't say. But he flamed me for correcting his post where he
referred to UTF-8 as ASCII.

--
/-- Joona Palaste ([email protected]) ---------------------------\
| Kingpriest of "The Flying Lemon Tree" G++ FR FW+ M- #108 D+ ADA N+++|
| http://www.helsinki.fi/~palaste W++ B OP+ |
\----------------------------------------- Finland rules! ------------/
"B-but Angus! You're a dragon!"
- Mickey Mouse

 

[Jon A. Cruz]

He didn't say. But he flamed me for correcting his post where he
referred to UTF-8 as ASCII.

Point things out next time you have problems.

Among other things, "ASCII" is pretty well defined by a few standards.
And it even seems to be used correctly by the RFC's.

 

[Roedy Green]

If the legal values are unaccented, you can use a rough bounds check
followed by a lookup in a BitSet or boolean[theChar-lowerlegalBound]
to determine if the character is legal. If there are a handful of
legal chars, you can use a CASE, and let the compiler figure out how
to optimise it.