how do i convert a VAR string to an Array? NOW a MATH problem

grasshopper · 2008-06-18 18:56

How can i convert a string that is a long into a array. I wish to convert a number 2263 into 22.63? so I was thinking that i convert 2263 into an array then manuplate the array so that i have 22.63.

Post Edited (grasshopper) : 6/18/2008 10:13:19 PM GMT

Coley · 2008-06-18 20:01

Hi,

Have a look at the format object in the object exchange, I think you'll find some routines in there that will do what you want.

obex.parallax.com/objects/230/

Regards,

Coley

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
PropGFX Forums - The home of the Hybrid Development System and PropGFX Lite

grasshopper · 2008-06-18 22:10

Well thanks for the link however i found a more suiting object, the FloatString. I remain boggled still in the fact that i can use this code and it works fine

pub Calc_Slope | B, M, subtracted, slope, invert

  M :=  135
  B :=  35693
  subtracted := f.FSub(lm135, B)                        'Y - B
   
  slope := f.FDiv(subtracted,M)                         'Y - B / M

  lcd.pos(1,1)
  lcd.writestr(@screen1)
  
  lcd.writestr(fs.Floattostring(slope))
  lcd.pos(1,15)
  lcd.writestr(@screen0)

But when I replace M with the value of


  M :=  135.25

My math is all incorrect.

Please help me i have tryed several varations of the


 slope := f.FDiv(subtracted,M)                         'Y - B / M

i.e.


 slope := f.foat(f.FDiv(subtracted,M))                         'Y - B / M

Mike Green · 2008-06-18 22:13

Your problem is that you're trying to do floating point arithmetic on 32-bit integers. Try assigning 35693.0 to B as well as 135.0 to M.

grasshopper · 2008-06-18 22:21

By the way i am using the Float32 object.

Mike my life saver:

I did as you asked but no luck. I got a negative [noparse][[/noparse] -264. ] when the number should be around 23.62.

Any thing else? Please help

grasshopper · 2008-06-18 22:30

** SOLVED thanks Mike you always trigger something that allows me to solve the problem


  M :=  135.5
  B :=  35693
  
  subtracted := F.FSub(lm135,B)                        'Y - B
  
  slope := F.FDiv(f.ffloat(subtracted),M)                        'Y - B / M
  
  lcd.pos(1,1)
  lcd.writestr(@screen1)
  
  lcd.writestr(fs.Floattostring(slope))
  lcd.pos(1,15)
  lcd.writestr(@screen0)

Mike Green · 2008-06-18 22:35

The problem is not solved. You still have things wrong. Even though the result may look correct, it's wrong. You need:

M := 135.5
B := 35693.0
subtracted := F.FSub(lm135,B) ' assuming lm135 is floating point
slope := F.FDiv(subtracted,M)

grasshopper · 2008-06-19 01:41

Mike:

I think it works fine the lm135 is not floating point, so i assume that i can just subtract the two. Please let me know your concerns. And thanks a million for the help.

Mike Green · 2008-06-19 01:47

You can't use F.FSub to subtract non-floating point values. It may work sometimes, but it's a fluke of the way that floating point values are represented and manipulated. Use the integer operators like "+", "-", "*", "/", then use F.FFloat to convert the result to floating point.

grasshopper · 2008-06-19 02:10

Mike:

It works as you requested and again i thank you for the wisdom that you reflect upon me. For the record, perhaps others will benefit the final code is below.


pub Calc_Slope | B, M, subtracted, slope


  M :=  135.51
  B :=  35693
  
  subtracted := lm135 - B                                                 'Y - B
  
  slope := F.FDiv(f.ffloat(subtracted),M)                                       'Y - B / M
  
  lcd.pos(1,1)
  lcd.writestr(@screen1)
  
  lcd.writestr(fs.Floattostring(slope))
  lcd.pos(1,15)
  lcd.writestr(@screen0)

** as a side note the 2x20 display i am using does not like the F.S.Floattostring(slope) statement. It displays 5-6 characters after the number is shown. The characters are not any thing i have ever seen but i fix this by clearing the letters with more code after the F.F.Floattostring(slope) statement **

Mike Green · 2008-06-19 02:16

Did you make sure to change the "FloatMath" in "FloatString" to match what you're using for the other floating point?

You didn't show how you positioned the output string on the LCD display. The "writestr" just puts out what's in the string and anything left on the display from before is left unchanged. You at least need to erase the rest of the current line. When I've done similar things, I save the pointer in a local variable and do STRSIZE on it to get the number of characters. I then output enough spaces to clear the rest of the current display line.

Post Edited (Mike Green) : 6/19/2008 2:21:56 AM GMT

grasshopper · 2008-06-19 02:20

Yes i replaced the F with a G and still weird characters. Good catch though.

Mike Green · 2008-06-19 02:26

As is usually true, you would have to post your whole program to get specific advice on this. This sort of problem is usually due to something not shown in the little fragments you've posted.

grasshopper · 2008-06-19 02:33

Ok here it is

_clkmode = xtal1 + pll16x
    _xinfreq = 6_000_000 
  
CON     


  AD_CS         = 9             'AD Chip Select
  AD_Din        = 10            'AD Data IN
  AD_Dout       = 11            'AD Data out
  AD_Dclk       = 8             'AD clock
  
  Chan0         = %10010111     'board temp
  Chan1         = %11010111     'lm135                            
  Chan2         = %10100111     'l                             
  Chan3         = %11100111     'e
  
  eos           = $ff
  cr            = $0D
                 
  
var

  long  counter
  long  Lm135
  long  setpoint
  long  slope  
  long  temp_data 

  
OBJ

  BASIC         : "BS2_Functions"
  lcd           : "LcdDriver"
  G             : "Floatmath"
  fs            : "floatString"
    

pub start | X

  init
 
  repeat
     
      repeat x from 0 to 25
      
          AtoD
          Calc_Slope
          


      Display_it
      
                       
Pub Init
  
 ' G.start                                               'Start math obj
  lcd.init                                              'Init display
  lcd.cls                                               'Clear display

  
Pub AtoD | Temp, X                        ' Read A to D

    temp := 0 
    lm135 := 0 
   repeat X from 0 to 24
      outa[noparse][[/noparse]AD_CS]~                                                                'Chip Select active low "ON"   
      Basic.SHIFTOUT(AD_Din,AD_DClk,Chan1,BASIC#MSBFIRST,8)                       'Chan 1 is Lm134                        
      Temp := BASIC.SHIFTIN(AD_Dout,AD_DCLK,BASIC#MSBPOST,16)                     'Shift in 16 bit
      outa[noparse][[/noparse]AD_CS]~~                                                               'Turn it OFF" 
      lm135 := Temp + lm135

  lm135 := lm135 / 25
    
Pub Calc_Slope | B, M, subtracted


  M :=  135.51
  B :=  35693
  
  subtracted := lm135 - B                                                 'Y - B
  
  slope := G.FDiv(G.ffloat(subtracted),M)                                       'Y - B / M

    
Pub Display_it

  lcd.cls
  lcd.home
  lcd.pos(1,1)
  lcd.writestr(@screen1)
  
  lcd.writestr(fs.Floattostring(slope))
  lcd.pos(1,15)
  lcd.writestr(@screen3)

                 
dat

  screen1      byte "Holder = ",eos
  screen2      byte "Target = ",eos
  screen3      byte " ",$DF,"C",eos

Mike Green · 2008-06-19 02:39

The problem you're running into is that strings are terminated with a zero byte and you're using $ff as "eos". Change the definition of "eos" to zero and things should work better.

grasshopper · 2008-06-19 03:38

hate to be a pain but that made it real bad. I got some real cool space age flashing characters flopping around real fast.


con 

eos   = $00

Then i fixed it in the LCD object and now works great. Thanks again Mike. I owe you a few...

Post Edited (grasshopper) : 6/19/2008 3:44:26 AM GMT

grasshopper · 2008-06-28 20:17

Mike I have a little problem with the change you mentioned 00 for eos vs. $FF for eos. The display reads a correct value but it looks horrible for example

lcd reads 20.3 instead of 20.30 i want it to look uniform throughout the numbers

Any ideas ?

Mike Green · 2008-06-28 21:30

Use FloatToFormat instead of FloatToString. Read the comments in FloatString for details.

grasshopper · 2008-06-29 02:08

Thanks Mr.Green. Apparently I had an outdated object.

Again Thanks

how do i convert a VAR string to an Array? NOW a MATH problem

Comments