Trying to figure out how to convert these hex bytes into an ASCII string

Don M

I have 2 byte variables called Temp1 & Temp2

var

  byte Temp1[6]
  byte Temp2[14]

In Temp 1 I have these bytes- $00, $07, $80, $7F, $A8, $D3

I'm trying to figure out how to break apart the hex bytes to store the individual nibbles (?) as their ascii equivalent in Temp2

In Temp2 I want to have these bytes- $30, $30, $30, $37, $38, $30, $37, $46, $41, $38, $44, $33 so that I can work with it and display it as a string "0007807FA8D3".

I tried using this:

  Temp2[0] := ((byte[@Temp1][0] & $F) + $30)
  Temp2[1] := ((byte[@Temp1][0] >> 4) + $30)
  and so on...
  Temp2[12] := 0

But it won't work with the letters A - F.

Is there a way to do this with a lookdown or lookup?

Thanks.
Don

T Chap

One idea is to use CASE

PUB getAscii(val)
        Case Val
              $30 : 'display the ascii value 
              $31 : 'display the ascii value

etc

r.daneel

Since you want to use lookup or lookdown, how about:

repeat i from 0 to 5
  left := $30 + offset(byte[@temp1[i]] >> 4 & $f)
  right := $30 + offset(byte[@temp1[i]] & $f)

...
...


PRI offset(index)
  return lookupz(index: 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 17, 18, 19, 20, 21, 22)

You could add the $30 to the values in the lookup table if you wanted.

kwinn

For the ls nibble you and it with hexF and add $30
For the ms nibble shift right 4 and add $30

r.daneel

kwinn wrote: »

For the ls nibble you and it with hexF and add $30
For the ms nibble shift right 4 and add $30

That doesn't work for $a-$f. $30+$a = $3a which is the ascii ":" character, not "A".

Phil Pilgrim (PhiPi)

Don,

Do you ever refer to the Propeller Manual before you come to the forum for answers? If you had, you would be familiar with the lookdownz command. Look it up!

-Phil

kwinn

r.daneel wrote: »

That doesn't work for $a-$f. $30+$a = $3a which is the ascii ":" character, not "A".

You're right.
Should have been add $30
Compare to see if the result was greater than $39 and add $8 if so.

Don M

Thanks everyone. Got it sorted out.

  debug.str(string("String: ")) 
  repeat i from 0 to 5
    Temp2 [i*2] := lookupz((byte[@Temp1[i]] >> 4 & $f): $30, $31, $32, $33, $34, $35, $36, $37, $38, $39, $41, $42, $43, $44, $45, $46)
    Temp2 [(i*2)+1] := lookupz((byte[@Temp1[i]] & $f): $30, $31, $32, $33, $34, $35, $36, $37, $38, $39, $41, $42, $43, $44, $45, $46)
  Temp2[12] := 0
  debug.str(@Temp2)
  debug.tx(13)  

Don M

This works too...

  debug.str(string("String: ")) 
  repeat i from 0 to 5
    Temp2 [i*2] := lookupz((byte[@Temp1[i]] >> 4 & $f): "0123456789ABCDEF")
    Temp2 [(i*2)+1] := lookupz((byte[@Temp1[i]] & $f): "0123456789ABCDEF")
  Temp2[12] := 0
  debug.str(@Temp2)
  debug.tx(13)

Mike G

I must be missing something... Why create a secondary buffer to hold ASCII values when you can just convert the values on the fly? Less Smile to worry about.

CON
  _clkmode = xtal1 + pll16x     
  _xinfreq = 5_000_000

  CR            = 13

VAR

DAT
  workspace byte   $0[16]


OBJ
  pst             : "Parallax Serial Terminal"



PUB Main | wizver, p1

  pst.Start(115_200) 
  pause(500)

  InsertTestValues
  PrintHexValues


PUB PrintHexValues | i
  pst.str(string(CR, "Workspace = "))
  repeat i from 0 to 15
    pst.hex(workspace[i], 2)
  pst.char(CR)

PUB InsertTestValues | seed, i
  seed := cnt
  repeat i from 0 to 15
    workspace[i] := ?seed & $FF

PRI pause(Duration)  
  waitcnt(((clkfreq / 1_000 * Duration - 3932) #> 381) + cnt)
  return

Don M

Mike G wrote: »

I must be missing something... Why create a secondary buffer to hold ASCII values when you can just convert the values on the fly? Less Smile to worry about.

CON
  _clkmode = xtal1 + pll16x     
  _xinfreq = 5_000_000

  CR            = 13

VAR

DAT
  workspace byte   $0[16]


OBJ
  pst             : "Parallax Serial Terminal"



PUB Main | wizver, p1

  pst.Start(115_200) 
  pause(500)

  InsertTestValues
  PrintHexValues


PUB PrintHexValues | i
  pst.str(string(CR, "Workspace = "))
  repeat i from 0 to 15
    pst.hex(workspace[i], 2)
  pst.char(CR)

PUB InsertTestValues | seed, i
  seed := cnt
  repeat i from 0 to 15
    workspace[i] := ?seed & $FF

PRI pause(Duration)  
  waitcnt(((clkfreq / 1_000 * Duration - 3932) #> 381) + cnt)
  return

Mike,

Actually I do convert on the fly. I just use this type of example for forum posting and in some test programs on a Quickstart.

Mike G

Huh? So the original question is now to use the lookup and lookdown command...

r.daneel

Well, he did say:

Don M wrote: »

In Temp2 I want to have these bytes- $30, $30, $30, $37, $38, $30, $37, $46, $41, $38, $44, $33 so that I can work with it and display it as a string "0007807FA8D3".

So I assumed he wanted to work with the bytes in Temp2, not just display them.

info

Can you do
if nibble < $0A, add $30
if nibble > $09 add $37
?

info

BTW, you can also check how other programs do convertions. There are several programs on obex to convert number base, string, etc.

I used serial port driver from obex to display Dec, Hex and ASCII for i2c eeprom hex dump.

Mike G

So I assumed he wanted to work with the bytes in Temp2, not just display them.

Right, and my suggestion was intended to help Don create reusable and clean code without dependencies. Why work with Temp2 when you already have a the working values in Temp1?

Anyway, I guess I'm becoming a curmudgeon.

r.daneel

Mike G wrote: »

Right, and my suggestion was intended to help Don create reusable and clean code without dependencies. Why work with Temp2 when you already have a the working values in Temp1?

Anyway, I guess I'm becoming a curmudgeon.

Well, I can't say if you're becoming a curmudgeon, but what if Don wanted to use the converted bytes in some sort of loop millions of times, but retain the original values? Why wouldn't he convert once and put the result in a buffer like Temp2? Without knowing what he wants to do with the result I can't say if it would be better to convert on the fly or not...