1. bytes_to_int
what is the highest floating point the Eu, bytes_to_int function can return?
2. Re: bytes_to_int
your average Joe wrote:
>
> what is the highest floating point the Eu, bytes_to_int function can return?
The bytes_to_int function returns a 32 bit unsigned integer, thus it's largest
value will be #FFFFFFFF or 4294967295.
Larry Miller
3. Re: bytes_to_int
that is reakky strange, because that is the number my bytes to int function
resets at
i cant go larger than 42.94967295
note the decimal place, thats were it cuts and goes back to zero + the remainder
4. Re: bytes_to_int
- Posted by Greg Haberek <ghaberek at gmail.com>
Feb 08, 2006
-
Last edited Feb 09, 2006
> i cant go larger than 42.94967295
You can't use bytes_to_int() for floating-point numbers, silly!
They're not integers! You need to use float32_to_atom() or
float64_to_atom() to convert bytes into floating-point numbers.
~Greg
5. bytes_to_int
Hi,
I seem to be having a problem reading in from an array of stored values and
using the bytes_to_int function.
A massive chunk of the data i read in is stored in 2 byte sequences so what
i do is use the algorithm
object void
for i = 1 to length(array) do
if length(array[i] = 2) then
void = append(array[i],0)
void = append(array[i],0)
end if
end for
* this sets up the array so that i can read a 4 byte sequence into the
bytes_to_int
function.
* will padding the 2 byte sequence with two zeros so that i can pass it to the
bytes_to_int
function alter the real value returned ?
* it seems to work fine for a majority of the data i pass, but it seems the
value is distorted
once it gets to a certain magnitude.
* is there another way of padding a 2byte sequence so that it can be read
unaltered by bytes_to_int?
thanx,
Joe.
6. Re: bytes_to_int
basically, how can i convert a sequence like this
sequence = {199,235}
into a 2 byte signed integer
7. Re: bytes_to_int
your average Joe wrote:
>
> basically, how can i convert a sequence like this
>
> sequence = {199,235}
>
> into a 2 byte signed integer
Doesn't
integer int
sequence = {199,235}
sequence = {199,235,0,0}
int = bytes_to_int(sequence)
bytes_to_int(sequence)
work?
I get 60359.
8. Re: bytes_to_int
Evan Marshall wrote:
>
> your average Joe wrote:
> >
> > basically, how can i convert a sequence like this
> >
> > sequence = {199,235}
> >
> > into a 2 byte signed integer
>
>
> Doesn't
>
> }}}
<eucode>
> integer int
> sequence = {199,235}
> sequence = {199,235,0,0}
> int = bytes_to_int(sequence)
> bytes_to_int(sequence)
> </eucode>
{{{
>
> work?
>
> I get 60359.
or rather
sequence seq
integer int
seq = {199,235}
seq = {199,235,0,0}
int = bytes_to_int(seq)
>
9. Re: bytes_to_int
yes i have been using that method, and as far as i can tell, there seems to be
some sort of clipping going on. by clipping i mean once i reach a certain
magnitude it is not converting properly, if that makes sense. i can see it
visually because i am mapping the returned values into a heightfield in opengl
and they are all clipped at the exact same height so the tops of the hills at
a certain elevation appear in the right spot x,z wise but appear below the other
terrain points that arent clipped. The only thing i can trace it back to is
this method of using {0,0} padding to convert them, note that the tops of the
hills
elevation wise map propely they just all have the same negative clipped
magnitude
10. Re: bytes_to_int
your average Joe wrote:
>
> yes i have been using that method, and as far as i can tell, there seems to
> be
> some sort of clipping going on. by clipping i mean once i reach a certain
> magnitude it is not converting properly, if that makes sense. i can see it
> visually because i am mapping the returned values into a heightfield in opengl
> and they are all clipped at the exact same height so the tops of the hills at
> a certain elevation appear in the right spot x,z wise but appear below the
> other
> terrain points that arent clipped. The only thing i can trace it back to is
> this method of using {0,0} padding to convert them, note that the tops of the
> hills
> elevation wise map propely they just all have the same negative clipped
> magnitude
Why don't you put a temporary check in the code where you are appending
the 2 zeros to print an error if any byte in your 4 byte sequence
exceeds the 256 size limit ?
Bernie
My files in archive:
motor.eu w32engin.ew mixedlib.e eu_engin.e win32eru.exw
Can be downloaded here:
http://www.rapideuphoria.com/cgi-bin/asearch.exu?dos=on&win=on&lnx=on&gen=on&keywords=bernie+ryan
11. Re: bytes_to_int
your average Joe wrote:
>
> Hi,
>
> I seem to be having a problem reading in from an array of stored values and
>
> using the bytes_to_int function.
>
> A massive chunk of the data i read in is stored in 2 byte sequences so what
> i do is use the algorithm
>
> object void
> for i = 1 to length(array) do
>
> if length(array[i] = 2) then
>
> void = append(array[i],0)
> void = append(array[i],0)
>
> end if
>
> end for
>
> * this sets up the array so that i can read a 4 byte sequence into the
> bytes_to_int
> function.
>
> * will padding the 2 byte sequence with two zeros so that i can pass it to the
> bytes_to_int
> function alter the real value returned ?
>
> * it seems to work fine for a majority of the data i pass, but it seems the
> value is distorted
> once it gets to a certain magnitude.
>
> * is there another way of padding a 2byte sequence so that it can be read
> unaltered
> by bytes_to_int?
>
> thanx,
> Joe.
Hi there Joe,
Here's something that works...
--start of file--------------------------
include misc.e
include machine.e
sequence array,intbytes
array={{1,2},{3,4},{5,6}}
intbytes=repeat(0,length(array))
for i = 1 to length(array) do
if length(array[i])=2 then
--array[i]=reverse(array[i]) --(optional)
array[i]=array[i]&{0,0}
intbytes[i]=bytes_to_int(array[i])
end if
end for
?array
?intbytes
sleep(100)
--end of file----------------------------
Of course if you dont need to store the 4 byte values of array[i]
then you can use a dummy variable to hold the data and then
just convert it and store the resulting integer in 'intbytes'.
Just one last note...
Depending on the order of your two-byte data you may need to use
'reverse' to change the order before appending zeros (shown above
as 'optional'.
Take care,
Al
And, good luck with your Euphoria programming!
My bumper sticker: "I brake for LED's"
12. Re: bytes_to_int
Evan Marshall wrote:
>
> Evan Marshall wrote:
> >
> > your average Joe wrote:
> > >
> > > basically, how can i convert a sequence like this
> > >
> > > sequence = {199,235}
> > >
> > > into a 2 byte signed integer
> >
> >
> > Doesn't
> >
> > }}}
<eucode>
> > integer int
> > sequence = {199,235}
> > sequence = {199,235,0,0}
> > int = bytes_to_int(sequence)
> > bytes_to_int(sequence)
> > </eucode>
{{{
> >
> > work?
> >
> > I get 60359.
>
>
> or rather
> }}}
<eucode>
> sequence seq
> integer int
>
> seq = {199,235}
> seq = {199,235,0,0}
> int = bytes_to_int(seq)
> > </eucode>
{{{
Actually, I think he wants to create a signed 16-bit integer. I'm not testing
this, but here is how I would do it:
-- tested
-- takes a two-element sequence in low-byte/hi-byte order
-- and returns a signed 16-bit integer
function bytes_to_short(sequence bytes)
integer short
if length(bytes) < 2 then
if length(bytes) < 1 then short = 0 -- empty sequence
else short = bytes[1]
end if
return short
end if
short = bytes[1] + 256 * bytes[2]
if short > #7FFF then -- change sign
short = and_bits(not_bits(short) + 1, #FFFF)
short = -short
end if
return short
end function -- bytes to short
? bytes_to_short({127}) -- 127
? bytes_to_short({255, 127}) -- 32767
? bytes_to_short({0, 128}) -- (-1)
? bytes_to_short({255, 255}) -- (-32768)
More error checking could be added, such as making sure each element of bytes is
not greater than 255 (or using and_bits(bytes, #FF)).
You could also change this to take longer sequences and break them into two, or
(easily enough) take an atom and convert it to a 16-bit signed integer, etc.
If you need to reverse the byte-order, then just switch this statement:
short = bytes[1] * 256 + bytes[2]
HTH
--
"Any programming problem can be solved by adding a level of indirection."
--anonymous
"Any performance problem can be solved by removing a level of indirection."
--M. Haertel
j.
13. Re: bytes_to_int
Using this code:
atom fn
fn = open("Results.txt","w")
for x = 0 to 255 do
for y = 0 to 255 do
printf(fn,"%d\n",bytes_to_int({y,x,0,0}))
end for
end for
gives a file listing all integers from 0 to 65535
14. Re: bytes_to_int
your average Joe wrote:
>
> basically, how can i convert a sequence like this
>
> sequence = {199,235}
>
> into a 2 byte signed integer
With the code I posted previously, I got -5177. Is that what you are looking
for?
--
"Any programming problem can be solved by adding a level of indirection."
--anonymous
"Any performance problem can be solved by removing a level of indirection."
--M. Haertel
j.
15. Re: bytes_to_int
Aha, thx for all the replies, lots of ideas and food for thought there, exactly
what i needed :)
I just noticed exactly where i am having the problem by slowing my
converter and printing each one in all its forms so i can see what's happening.
It is apparent to me now that the sequence {6,0} is suppose to return a value
higher than 65535, that's exactly where i am getting the clipping problem on all
elevations once
that exceed that number which to my undestanding is the max i can output using
the 2byte padding technique.
Ill try the methods listed and let you know if i can get it working
Thanks
Joe
16. Re: bytes_to_int
your average Joe wrote:
>
> Aha, thx for all the replies, lots of ideas and food for thought there,
> exactly
> what i needed :)
>
> I just noticed exactly where i am having the problem by slowing my
> converter and printing each one in all its forms so i can see what's
> happening.
>
> It is apparent to me now that the sequence {6,0} is suppose to return a value
> higher than 65535, that's exactly where i am getting the clipping problem on
> all elevations once
> that exceed that number which to my undestanding is the max i can output using
> the 2byte padding technique.
>
> Ill try the methods listed and let you know if i can get it working
>
> Thanks
> Joe
Glad to help. What exact range and byte-order are you looking for? In an earlier
post, you said you needed to convert them to 16-bit signed integers. What is the
actual min and max output that you need? Do the 16-bit numbers need to be scaled
to 32-bits?
--
"Any programming problem can be solved by adding a level of indirection."
--anonymous
"Any performance problem can be solved by removing a level of indirection."
--M. Haertel
j.
17. Re: bytes_to_int
i ran the data where i was getting the clip problem through your algorithm and
it appears to work very nicely Jay, stupid terragen information for developers
webpage claims that they are 2 byte signed integers lol
cheers,
Joe
18. Re: bytes_to_int
Problem solved, i just ran it through my new modified pre converter and into
opengl
for rendering and was pleasantly surprised that is maps PERFECTLY now :). That
was my problem that it needed to be a signed value. Oh the relief, ive been
beating my head on the wall for 3 days on that one hehe.
Thanx again Jay
