Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > comp.lang.clipper.visual-objects > #5283 > unrolled thread

Shortint calculations

Started bykvstegemann@web.de
First post2013-05-27 08:53 -0700
Last post2013-05-31 03:43 -0700
Articles 20 on this page of 24 — 8 participants

Back to article view | Back to comp.lang.clipper.visual-objects


Contents

  Shortint calculations kvstegemann@web.de - 2013-05-27 08:53 -0700
    Re: Shortint calculations Karl Faller <mail_withoutthat_@_drei-f_andthat_.de> - 2013-05-27 18:09 +0200
    Re: Shortint calculations dlzc <dlzc1@cox.net> - 2013-05-27 10:21 -0700
      Re: Shortint calculations Karl Faller <mail_withoutthat_@_drei-f_andthat_.de> - 2013-05-27 20:32 +0200
      Re: Shortint calculations kvstegemann@web.de - 2013-05-27 11:46 -0700
        Re: Shortint calculations Karl Faller <mail_withoutthat_@_drei-f_andthat_.de> - 2013-05-27 23:40 +0200
        Re: Shortint calculations "Wolfgang Riedmann" <wriedmann@gmail.com> - 2013-05-28 08:02 +0200
    Re: Shortint calculations Mathias <mathias.hakansson@consultec.se> - 2013-05-27 22:39 -0700
      Re: Shortint calculations Karl Faller <mail_withoutthat_@_drei-f_andthat_.de> - 2013-05-28 09:01 +0200
        Re: Shortint calculations kvstegemann@web.de - 2013-05-28 01:02 -0700
          Re: Shortint calculations Karl Faller <mail_withoutthat_@_drei-f_andthat_.de> - 2013-05-28 10:32 +0200
            Re: Shortint calculations kvstegemann@web.de - 2013-05-28 01:46 -0700
      Re: Shortint calculations Meinhard Schnoor-Matriciani <msm@appfact.de> - 2013-05-28 11:45 +0200
        Re: Shortint calculations Mathias <mathias.hakansson@consultec.se> - 2013-05-28 23:45 -0700
          Re: Shortint calculations "Er!k \\/!$$Er" <nospam@nospam.url> - 2013-05-29 16:57 +0200
            Re: Shortint calculations dlzc <dlzc1@cox.net> - 2013-05-29 09:33 -0700
    Re: Shortint calculations dlzc <dlzc1@cox.net> - 2013-05-28 07:15 -0700
      Re: Shortint calculations Mathias <mathias.hakansson@consultec.se> - 2013-05-29 04:44 -0700
        Re: Shortint calculations dlzc <dlzc1@cox.net> - 2013-05-29 07:25 -0700
    Re: Shortint calculations "Jamal" <jamal@examples.net> - 2013-05-29 17:18 -0400
      Re: Shortint calculations kvstegemann@web.de - 2013-05-30 01:47 -0700
        Re: Shortint calculations "Jamal" <jamal@examples.net> - 2013-05-30 13:19 -0400
    Re: Shortint calculations Meinhard Schnoor-Matriciani <msm@appfact.de> - 2013-05-30 22:05 +0200
      Re: Shortint calculations kvstegemann@web.de - 2013-05-31 03:43 -0700

Page 1 of 2  [1] 2  Next page →


#5283 — Shortint calculations

Fromkvstegemann@web.de
Date2013-05-27 08:53 -0700
SubjectShortint calculations
Message-ID<aa22a440-8e92-444c-9538-3488ccce999d@googlegroups.com>
Hi out there,

I'm getting strange results with simple calculations with shortints. Consider this mini-test-app:

FUNCTION Start()

	LOCAL nA, nB, nC AS SHORTINT
	
	nA := 748
	nB := 22
	nC := 35
	
	IF nA > ( nB - nC )
		? "748 is greater than (22 - 35)"
	ELSE
		? "748 is not greater than (22 - 35). Oops!"
	ENDIF
	
	_Wait("")
	

Try it and tell me what happens here. When the three variables are declared as longint, everything seems ok. But when they are shortints, the result is obviously wrong. It seems to me that the results here are well within the scope of a shortint, so what am I missing? (Using VO 2.8 Sp2)

Thanks for input
Kay-Viktor

[toc] | [next] | [standalone]


#5284

FromKarl Faller <mail_withoutthat_@_drei-f_andthat_.de>
Date2013-05-27 18:09 +0200
Message-ID<h717q8hp17ish4pm5nus3rmrojj97fhd4h@4ax.com>
In reply to#5283
Kay,
strange indeed - i see it with Sp 4 but not with VN2...

Karl

>Hi out there,
>
>I'm getting strange results with simple calculations with shortints. Consider this mini-test-app:
>
>FUNCTION Start()
>
>	LOCAL nA, nB, nC AS SHORTINT
>	
>	nA := 748
>	nB := 22
>	nC := 35
>	
>	IF nA > ( nB - nC )
>		? "748 is greater than (22 - 35)"
>	ELSE
>		? "748 is not greater than (22 - 35). Oops!"
>	ENDIF
>	
>	_Wait("")
>	
>
>Try it and tell me what happens here. When the three variables are declared as longint, everything seems ok. But when they are shortints, the result is obviously wrong. It seems to me that the results here are well within the scope of a shortint, so what am I missing? (Using VO 2.8 Sp2)
>
>Thanks for input
>Kay-Viktor

[toc] | [prev] | [next] | [standalone]


#5285

Fromdlzc <dlzc1@cox.net>
Date2013-05-27 10:21 -0700
Message-ID<ca879809-f20c-4d98-a225-44d4e73ae01e@googlegroups.com>
In reply to#5283
Dear kvste...:

On Monday, May 27, 2013 8:53:43 AM UTC-7, kvste...@web.de wrote:
...
> 	LOCAL nA, nB, nC AS SHORTINT
> 
> 	nA := 748
> 	nB := 22
> 	nC := 35
> 
> 	IF nA > ( nB - nC )
> 		? "748 is greater than (22 - 35)"
> 	ELSE
> 		? "748 is not greater than (22 - 35). Oops!"
> 	ENDIF
...
> 
> When the three variables are declared as longint,
> everything seems ok. But when they are shortints,
> the result is obviously wrong.

Are shortints signed?  Can you print (nB - nC)?

> It seems to me that the results here are well
> within the scope of a shortint, so what am I missing?
> (Using VO 2.8 Sp2)

Shortint might be unsigned, so the amount that nC is greater than nB, is deducted from 65536.

David A. Smith

[toc] | [prev] | [next] | [standalone]


#5286

FromKarl Faller <mail_withoutthat_@_drei-f_andthat_.de>
Date2013-05-27 20:32 +0200
Message-ID<lj97q8poflhnip8r9iefigh2oce3bhni35@4ax.com>
In reply to#5285
>Are shortints signed?  Can you print (nB - nC)?
gives -13, as it should...

Karl
>> It seems to me that the results here are well
>> within the scope of a shortint, so what am I missing?
>> (Using VO 2.8 Sp2)
>
>Shortint might be unsigned, so the amount that nC is greater than nB, is deducted from 65536.
>
>David A. Smith

[toc] | [prev] | [next] | [standalone]


#5287

Fromkvstegemann@web.de
Date2013-05-27 11:46 -0700
Message-ID<fc9ab794-eb2e-462d-a633-8df7bfa03f26@googlegroups.com>
In reply to#5285
> Are shortints signed?  Can you print (nB - nC)?

David,

shortints are signed and should have no problems with negative values in this range. I've had enough headaches with word/dword values within calculations therefore I preferred shortints, but now this ...

Kay-Viktor

[toc] | [prev] | [next] | [standalone]


#5288

FromKarl Faller <mail_withoutthat_@_drei-f_andthat_.de>
Date2013-05-27 23:40 +0200
Message-ID<kik7q8le7ja7i438je3bpuspmfn0jl7fra@4ax.com>
In reply to#5287
Kay-Viktor,
>values within calculations therefore I preferred shortints, but now this ...
any reason to use shorts? Do you such number crunching that the size
really matters vs "regular" ints?
Not saying, that your results are pointing on some bug, which should
be corrected, but in this case i think the workaround relatively easy
;-)

Karl

[toc] | [prev] | [next] | [standalone]


#5290

From"Wolfgang Riedmann" <wriedmann@gmail.com>
Date2013-05-28 08:02 +0200
Message-ID<b0ivk9Fn4c6U1@mid.individual.net>
In reply to#5287
Hi Kay-Viktor,

> shortints are signed and should have no problems with negative values
> in this range. I've had enough headaches with word/dword values
> within calculations therefore I preferred shortints, but now this ...

on a 32 bit system it is better to user int and dword because they are
32 bit values. Shortint and word are 16 bit values and have a
performance penalty because they need to be aligned to 32 bit.

Wolfgang

P.S. I think that VO behaves erratically in your sample

-- 

[toc] | [prev] | [next] | [standalone]


#5289

FromMathias <mathias.hakansson@consultec.se>
Date2013-05-27 22:39 -0700
Message-ID<456106ce-3950-46bd-9800-de8543f5df72@googlegroups.com>
In reply to#5283
This sample may be en explanation, but I still Think it's very confusing. It seems like if all the numbers are longints. But maybe only nb - nC is a longint and something goes wrong when comparing a shortint and a longint in nA > ( nB - nC )?

/Mathias

function Start()
   local nA, nB, nC as shortint 
         
   nA := 748 
   nB := 22 
   nC := 35 
        


	? "Shortint?"
	? UsualType(nB - nC) == shortint
	? "Longint?"
	? UsualType(nB - nC) == longint
	
	? "Shortint?"
	? UsualType(nA) == shortint
	? "Longint?"
	? UsualType(nA) == longint
	
   ? nB - nC
   ? nA
	? nA > ( nB - nC )           
	? longint(nA) > ( nB - nC )
         
        if longint(nA) > nB - nC 
                ? "748 is greater than (22 - 35)" 
        else 
                ? "748 is not greater than (22 - 35). Oops!" 
        endif 
         
        _Wait("") 

	
return nil	


Den måndagen den 27:e maj 2013 kl. 17:53:43 UTC+2 skrev kvste...@web.de:
> Hi out there,
> 
> 
> 
> I'm getting strange results with simple calculations with shortints. Consider this mini-test-app:
> 
> 
> 
> FUNCTION Start()
> 
> 
> 
> 	LOCAL nA, nB, nC AS SHORTINT
> 
> 	
> 
> 	nA := 748
> 
> 	nB := 22
> 
> 	nC := 35
> 
> 	
> 
> 	IF nA > ( nB - nC )
> 
> 		? "748 is greater than (22 - 35)"
> 
> 	ELSE
> 
> 		? "748 is not greater than (22 - 35). Oops!"
> 
> 	ENDIF
> 
> 	
> 
> 	_Wait("")
> 
> 	
> 
> 
> 
> Try it and tell me what happens here. When the three variables are declared as longint, everything seems ok. But when they are shortints, the result is obviously wrong. It seems to me that the results here are well within the scope of a shortint, so what am I missing? (Using VO 2.8 Sp2)
> 
> 
> 
> Thanks for input
> 
> Kay-Viktor

[toc] | [prev] | [next] | [standalone]


#5291

FromKarl Faller <mail_withoutthat_@_drei-f_andthat_.de>
Date2013-05-28 09:01 +0200
Message-ID<lsk8q8tlk92eoqvu0pfkrqb44fc0spqkgp@4ax.com>
In reply to#5289
Mathias & all,
just had a look, my compiler options has set (as per default)
"Overflow checking", so i'd expect a compiletime error at nA := 748,
not a "silent" cast.  Stranger still, why, if it does the casts, the
compare goes wrong. Strangest, that the same code in VN2 gives the
correct compare, but also no warning at the assignment of nA...

Karl

>This sample may be en explanation, but I still Think it's very confusing. It seems like if all the numbers are longints. But maybe only nb - nC is a longint and something goes wrong when comparing a shortint and a longint in nA > ( nB - nC )?
>
>/Mathias
>
>function Start()
>   local nA, nB, nC as shortint 
>         
>   nA := 748 
>   nB := 22 
>   nC := 35 
>        
>
>
>	? "Shortint?"
>	? UsualType(nB - nC) == shortint
>	? "Longint?"
>	? UsualType(nB - nC) == longint
>	
>	? "Shortint?"
>	? UsualType(nA) == shortint
>	? "Longint?"
>	? UsualType(nA) == longint
>	
>   ? nB - nC
>   ? nA
>	? nA > ( nB - nC )           
>	? longint(nA) > ( nB - nC )
>         
>        if longint(nA) > nB - nC 
>                ? "748 is greater than (22 - 35)" 
>        else 
>                ? "748 is not greater than (22 - 35). Oops!" 
>        endif 
>         
>        _Wait("") 
>
>	
>return nil	
>
>
>Den måndagen den 27:e maj 2013 kl. 17:53:43 UTC+2 skrev kvste...@web.de:
>> Hi out there,
>> 
>> 
>> 
>> I'm getting strange results with simple calculations with shortints. Consider this mini-test-app:
>> 
>> 
>> 
>> FUNCTION Start()
>> 
>> 
>> 
>> 	LOCAL nA, nB, nC AS SHORTINT
>> 
>> 	
>> 
>> 	nA := 748
>> 
>> 	nB := 22
>> 
>> 	nC := 35
>> 
>> 	
>> 
>> 	IF nA > ( nB - nC )
>> 
>> 		? "748 is greater than (22 - 35)"
>> 
>> 	ELSE
>> 
>> 		? "748 is not greater than (22 - 35). Oops!"
>> 
>> 	ENDIF
>> 
>> 	
>> 
>> 	_Wait("")
>> 
>> 	
>> 
>> 
>> 
>> Try it and tell me what happens here. When the three variables are declared as longint, everything seems ok. But when they are shortints, the result is obviously wrong. It seems to me that the results here are well within the scope of a shortint, so what am I missing? (Using VO 2.8 Sp2)
>> 
>> 
>> 
>> Thanks for input
>> 
>> Kay-Viktor

[toc] | [prev] | [next] | [standalone]


#5292

Fromkvstegemann@web.de
Date2013-05-28 01:02 -0700
Message-ID<28c0ea56-652e-4746-b71c-c33c4a45199d@googlegroups.com>
In reply to#5291
Karl,

why should there be any sort of error at nA := 748 ? A shortint should be perfectly capable of holding a value like 748. The range of shortints is -32767 to +32768, correct me if I'm wrong. Therefore all the mentioned operations should work perfectly.

To all: The original reason why I used shortints sometimes (in this case the whole thing is about window size calculations) was because some Windows API functions require shortint parameters. But I would never have suspected that shortint computations are broken in VO (after all those years ...). I will change them to longints and hope the headache goes away.

Have a nice day
Kay-Viktor

[toc] | [prev] | [next] | [standalone]


#5293

FromKarl Faller <mail_withoutthat_@_drei-f_andthat_.de>
Date2013-05-28 10:32 +0200
Message-ID<kfq8q8tqv2ggo5gqsurp2qu3ncso0gqba4@4ax.com>
In reply to#5292
Kay,
>why should there be any sort of error at nA := 748 ? A shortint should be perfectly capable of holding a value like 748. The range of shortints is -32767 to +32768, correct me if I'm wrong. 
that was my thinking, too, but then i googled and found:
 * delphibasics.com says: "ShortInt type is the smallest form of
signed integer, occupying 8 bits (1 byte) of storage. It supports only
the integers from -128 to 127. "

* wikipedia says: Short/short, Integer 	signed 	-32.768 	32.767 	

* VO pre release manuals <g>, says 16bit int, as in Ansi C short

Good, that i never used them...

Karl

[toc] | [prev] | [next] | [standalone]


#5294

Fromkvstegemann@web.de
Date2013-05-28 01:46 -0700
Message-ID<d239ac8c-895b-491d-ae7c-07ac9b4aebd1@googlegroups.com>
In reply to#5293
Am Dienstag, 28. Mai 2013 10:32:43 UTC+2 schrieb Karl Faller:
> that was my thinking, too, but then i googled and found:
> 
>  * delphibasics.com says: "ShortInt type is the smallest form of
> 
> signed integer, occupying 8 bits (1 byte) of storage. It supports only
> 
> the integers from -128 to 127. "
> 
> 
> 
> * wikipedia says: Short/short, Integer 	signed 	-32.768 	32.767 	
> 
> 
> 
> * VO pre release manuals , says 16bit int, as in Ansi C short

Hi Karl,

VO shortints are 16 bit as stated in the manual, therefore there should be no problem storing the values. The bug must be in the operations somewhere. This is definitely a VO bug.

Kay-Viktor

[toc] | [prev] | [next] | [standalone]


#5295

FromMeinhard Schnoor-Matriciani <msm@appfact.de>
Date2013-05-28 11:45 +0200
Message-ID<51a47d2d$0$6643$3ca0ca56@newsspool2.vodafone-ip.de>
In reply to#5289
Mathias,

this approach must lead to confusions, because UsualType would never 
return something like Shortint ( see documentation ).

I am going to have a look at this.

Regards
Meinhard

Am 28.05.2013 07:39, schrieb Mathias:
> This sample may be en explanation, but I still Think it's very confusing. It seems like if all the numbers are longints. But maybe only nb - nC is a longint and something goes wrong when comparing a shortint and a longint in nA > ( nB - nC )?
>
> /Mathias
>
> function Start()
>     local nA, nB, nC as shortint
>
>     nA := 748
>     nB := 22
>     nC := 35
>
>
>
> 	? "Shortint?"
> 	? UsualType(nB - nC) == shortint
> 	? "Longint?"
> 	? UsualType(nB - nC) == longint
> 	
> 	? "Shortint?"
> 	? UsualType(nA) == shortint
> 	? "Longint?"
> 	? UsualType(nA) == longint
> 	
>     ? nB - nC
>     ? nA
> 	? nA > ( nB - nC )
> 	? longint(nA) > ( nB - nC )
>
>          if longint(nA) > nB - nC
>                  ? "748 is greater than (22 - 35)"
>          else
>                  ? "748 is not greater than (22 - 35). Oops!"
>          endif
>
>          _Wait("")
>
> 	
> return nil	
>
>
> Den måndagen den 27:e maj 2013 kl. 17:53:43 UTC+2 skrev kvste...@web.de:
>> Hi out there,
>>
>>
>>
>> I'm getting strange results with simple calculations with shortints. Consider this mini-test-app:
>>
>>
>>
>> FUNCTION Start()
>>
>>
>>
>> 	LOCAL nA, nB, nC AS SHORTINT
>>
>> 	
>>
>> 	nA := 748
>>
>> 	nB := 22
>>
>> 	nC := 35
>>
>> 	
>>
>> 	IF nA > ( nB - nC )
>>
>> 		? "748 is greater than (22 - 35)"
>>
>> 	ELSE
>>
>> 		? "748 is not greater than (22 - 35). Oops!"
>>
>> 	ENDIF
>>
>> 	
>>
>> 	_Wait("")
>>
>> 	
>>
>>
>>
>> Try it and tell me what happens here. When the three variables are declared as longint, everything seems ok. But when they are shortints, the result is obviously wrong. It seems to me that the results here are well within the scope of a shortint, so what am I missing? (Using VO 2.8 Sp2)
>>
>>
>>
>> Thanks for input
>>
>> Kay-Viktor
>


-- 
Meinhard Schnoor-Matriciani
Grafx Development Team

[toc] | [prev] | [next] | [standalone]


#5298

FromMathias <mathias.hakansson@consultec.se>
Date2013-05-28 23:45 -0700
Message-ID<018d375d-3c3f-4ba3-ab4f-a5075a8f4072@googlegroups.com>
In reply to#5295
Hi Meinard,

I suspected that usuals don't carry shortints. However the whole thing is very confusing and it gets even more confusing if I modify the testprogram like below. It seems like it works if I calculate the subtraction before comparing.

function Start()
	local nA, nB, nC, nD as shortint 
         
	nA := 748 
	nB := 22 
	nC := 35 
	                    
	nD := nB - nC
	                    
	? nB - nC	// -13
	? nD		// -13
	? nA		// 748
	? nA > nB - nC	// .F.
	? nA > nD	// .T.
	? longint(nA) > ( nB - nC )	// .T.
         
        if longint(nA) > nB - nC 
                ? "748 is greater than (22 - 35)" 
        else 
                ? "748 is not greater than (22 - 35). Oops!" 
        endif 
         
        _Wait("") 

	
return nil	



Den tisdagen den 28:e maj 2013 kl. 11:45:31 UTC+2 skrev Meinhard Schnoor-Matriciani:
> Mathias,
> 
> 
> 
> this approach must lead to confusions, because UsualType would never 
> 
> return something like Shortint ( see documentation ).
> 
> 
> 
> I am going to have a look at this.
> 
> 
> 
> Regards
> 
> Meinhard
> 
> 
> 
> Am 28.05.2013 07:39, schrieb Mathias:
> 
> > This sample may be en explanation, but I still Think it's very confusing. It seems like if all the numbers are longints. But maybe only nb - nC is a longint and something goes wrong when comparing a shortint and a longint in nA > ( nB - nC )?
> 
> >
> 
> > /Mathias
> 
> >
> 
> > function Start()
> 
> >     local nA, nB, nC as shortint
> 
> >
> 
> >     nA := 748
> 
> >     nB := 22
> 
> >     nC := 35
> 
> >
> 
> >
> 
> >
> 
> > 	? "Shortint?"
> 
> > 	? UsualType(nB - nC) == shortint
> 
> > 	? "Longint?"
> 
> > 	? UsualType(nB - nC) == longint
> 
> > 	
> 
> > 	? "Shortint?"
> 
> > 	? UsualType(nA) == shortint
> 
> > 	? "Longint?"
> 
> > 	? UsualType(nA) == longint
> 
> > 	
> 
> >     ? nB - nC
> 
> >     ? nA
> 
> > 	? nA > ( nB - nC )
> 
> > 	? longint(nA) > ( nB - nC )
> 
> >
> 
> >          if longint(nA) > nB - nC
> 
> >                  ? "748 is greater than (22 - 35)"
> 
> >          else
> 
> >                  ? "748 is not greater than (22 - 35). Oops!"
> 
> >          endif
> 
> >
> 
> >          _Wait("")
> 
> >
> 
> > 	
> 
> > return nil	
> 
> >
> 
> >
> 
> > Den m�ndagen den 27:e maj 2013 kl. 17:53:43 UTC+2 skrev kvste...@web.de:
> 
> >> Hi out there,
> 
> >>
> 
> >>
> 
> >>
> 
> >> I'm getting strange results with simple calculations with shortints. Consider this mini-test-app:
> 
> >>
> 
> >>
> 
> >>
> 
> >> FUNCTION Start()
> 
> >>
> 
> >>
> 
> >>
> 
> >> 	LOCAL nA, nB, nC AS SHORTINT
> 
> >>
> 
> >> 	
> 
> >>
> 
> >> 	nA := 748
> 
> >>
> 
> >> 	nB := 22
> 
> >>
> 
> >> 	nC := 35
> 
> >>
> 
> >> 	
> 
> >>
> 
> >> 	IF nA > ( nB - nC )
> 
> >>
> 
> >> 		? "748 is greater than (22 - 35)"
> 
> >>
> 
> >> 	ELSE
> 
> >>
> 
> >> 		? "748 is not greater than (22 - 35). Oops!"
> 
> >>
> 
> >> 	ENDIF
> 
> >>
> 
> >> 	
> 
> >>
> 
> >> 	_Wait("")
> 
> >>
> 
> >> 	
> 
> >>
> 
> >>
> 
> >>
> 
> >> Try it and tell me what happens here. When the three variables are declared as longint, everything seems ok. But when they are shortints, the result is obviously wrong. It seems to me that the results here are well within the scope of a shortint, so what am I missing? (Using VO 2.8 Sp2)
> 
> >>
> 
> >>
> 
> >>
> 
> >> Thanks for input
> 
> >>
> 
> >> Kay-Viktor
> 
> >
> 
> 
> 
> 
> 
> -- 
> 
> Meinhard Schnoor-Matriciani
> 
> Grafx Development Team

[toc] | [prev] | [next] | [standalone]


#5303

From"Er!k \\/!$$Er" <nospam@nospam.url>
Date2013-05-29 16:57 +0200
Message-ID<ko5503$2lr5$1@adenine.netfront.net>
In reply to#5298
Mathias,

I have tried this code on VO 2.6, VO 2.7 build 2773 and VO 2.8 build 2837 
and it works fine here.

Erik

-13
-13
748
.F.
.T.
.T.
748 is greater than (22 - 35)




"Mathias"  schreef in bericht 
news:018d375d-3c3f-4ba3-ab4f-a5075a8f4072@googlegroups.com...

Hi Meinard,

I suspected that usuals don't carry shortints. However the whole thing is 
very confusing and it gets even more confusing if I modify the testprogram 
like below. It seems like it works if I calculate the subtraction before 
comparing.

function Start()
local nA, nB, nC, nD as shortint

nA := 748
nB := 22
nC := 35

nD := nB - nC

? nB - nC // -13
? nD // -13
? nA // 748
? nA > nB - nC // .F.
? nA > nD // .T.
? longint(nA) > ( nB - nC ) // .T.

        if longint(nA) > nB - nC
                ? "748 is greater than (22 - 35)"
        else
                ? "748 is not greater than (22 - 35). Oops!"
        endif

        _Wait("")


return nil



Den tisdagen den 28:e maj 2013 kl. 11:45:31 UTC+2 skrev Meinhard 
Schnoor-Matriciani:
> Mathias,
>
>
>
> this approach must lead to confusions, because UsualType would never
>
> return something like Shortint ( see documentation ).
>
>
>
> I am going to have a look at this.
>
>
>
> Regards
>
> Meinhard
>
>
>
> Am 28.05.2013 07:39, schrieb Mathias:
>
> > This sample may be en explanation, but I still Think it's very 
> > confusing. It seems like if all the numbers are longints. But maybe only 
> > nb - nC is a longint and something goes wrong when comparing a shortint 
> > and a longint in nA > ( nB - nC )?
>
> >
>
> > /Mathias
>
> >
>
> > function Start()
>
> >     local nA, nB, nC as shortint
>
> >
>
> >     nA := 748
>
> >     nB := 22
>
> >     nC := 35
>
> >
>
> >
>
> >
>
> > ? "Shortint?"
>
> > ? UsualType(nB - nC) == shortint
>
> > ? "Longint?"
>
> > ? UsualType(nB - nC) == longint
>
> >
>
> > ? "Shortint?"
>
> > ? UsualType(nA) == shortint
>
> > ? "Longint?"
>
> > ? UsualType(nA) == longint
>
> >
>
> >     ? nB - nC
>
> >     ? nA
>
> > ? nA > ( nB - nC )
>
> > ? longint(nA) > ( nB - nC )
>
> >
>
> >          if longint(nA) > nB - nC
>
> >                  ? "748 is greater than (22 - 35)"
>
> >          else
>
> >                  ? "748 is not greater than (22 - 35). Oops!"
>
> >          endif
>
> >
>
> >          _Wait("")
>
> >
>
> >
>
> > return nil
>
> >
>
> >
>
> > Den m�ndagen den 27:e maj 2013 kl. 17:53:43 UTC+2 skrev 
> > kvste...@web.de:
>
> >> Hi out there,
>
> >>
>
> >>
>
> >>
>
> >> I'm getting strange results with simple calculations with shortints. 
> >> Consider this mini-test-app:
>
> >>
>
> >>
>
> >>
>
> >> FUNCTION Start()
>
> >>
>
> >>
>
> >>
>
> >> LOCAL nA, nB, nC AS SHORTINT
>
> >>
>
> >>
>
> >>
>
> >> nA := 748
>
> >>
>
> >> nB := 22
>
> >>
>
> >> nC := 35
>
> >>
>
> >>
>
> >>
>
> >> IF nA > ( nB - nC )
>
> >>
>
> >> ? "748 is greater than (22 - 35)"
>
> >>
>
> >> ELSE
>
> >>
>
> >> ? "748 is not greater than (22 - 35). Oops!"
>
> >>
>
> >> ENDIF
>
> >>
>
> >>
>
> >>
>
> >> _Wait("")
>
> >>
>
> >>
>
> >>
>
> >>
>
> >>
>
> >> Try it and tell me what happens here. When the three variables are 
> >> declared as longint, everything seems ok. But when they are shortints, 
> >> the result is obviously wrong. It seems to me that the results here are 
> >> well within the scope of a shortint, so what am I missing? (Using VO 
> >> 2.8 Sp2)
>
> >>
>
> >>
>
> >>
>
> >> Thanks for input
>
> >>
>
> >> Kay-Viktor
>
> >
>
>
>
>
>
> -- 
>
> Meinhard Schnoor-Matriciani
>
> Grafx Development Team 


--- news://freenews.netfront.net/ - complaints: news@netfront.net ---

[toc] | [prev] | [next] | [standalone]


#5304

Fromdlzc <dlzc1@cox.net>
Date2013-05-29 09:33 -0700
Message-ID<18f12c97-3ae8-438a-b7cd-9483a6867afa@googlegroups.com>
In reply to#5303
Dear Er!k /!$$Er:

On Wednesday, May 29, 2013 7:57:06 AM UTC-7, Er!k /!$$Er wrote:
> Mathias,
> 
> I have tried this code on VO 2.6, VO 2.7 build 2773
> and VO 2.8 build 2837 and it works fine here.
> 
> 
> -13
> -13
> 748
> .F.

This above says 748 is not greater than -13.  Do you want to revise your "fine"?

> .T.
> .T.
> 748 is greater than (22 - 35)
...
> function Start()
> 
> local nA, nB, nC, nD as shortint
> 
> nA := 748
> nB := 22
> nC := 35
> 
> nD := nB - nC
> 
> ? nB - nC // -13
> ? nD // -13
> ? nA // 748
> ? nA > nB - nC // .F.
> ? nA > nD // .T.
> ? longint(nA) > ( nB - nC ) // .T.
> 
>         if longint(nA) > nB - nC
>                 ? "748 is greater than (22 - 35)"
>         else
>                 ? "748 is not greater than (22 - 35). Oops!"
>         endif
> 
>         _Wait("")
> 
> return nil

Erik's results say that this has been broken for a long time.  And it looks like a negative longint result is downcoverted incorrectly to a shortint.

I wonder what:
? shortint( longint( -13 )) // looks like?


David A. Smith

[toc] | [prev] | [next] | [standalone]


#5297

Fromdlzc <dlzc1@cox.net>
Date2013-05-28 07:15 -0700
Message-ID<0d515ec5-a6d5-4e1f-9ea8-13c5128f0798@googlegroups.com>
In reply to#5283
Dear kvste...:

On Monday, May 27, 2013 8:53:43 AM UTC-7, kvste...@web.de wrote:
...
> 	LOCAL nA, nB, nC AS SHORTINT
> 
> 	nA := 748
> 	nB := 22
> 	nC := 35
> 
> 	IF nA > ( nB - nC )

Here is the problem, the result of subtraction is NOT a shortint.  It is (probably) a long integer.  When compared to the shortint nA, the result of subtraction may be being truncated, rather than the result being demoted to a shortint, or nA being promoted to a long int.

> 		? "748 is greater than (22 - 35)"
> 	ELSE
> 		? "748 is not greater than (22 - 35). Oops!"
> 	ENDIF

It sure looks like a bug to me.

David A. Smith

[toc] | [prev] | [next] | [standalone]


#5301

FromMathias <mathias.hakansson@consultec.se>
Date2013-05-29 04:44 -0700
Message-ID<418e57c0-80db-420d-a98d-c3da4c19c2d9@googlegroups.com>
In reply to#5297
Hi David,

no it's more complicated than this...

	? shortint(nA) > nD				// .T.
No surprise, as nA > nD also was true. nA is allready a shortint so this is expected.

	? shortint(nA) > nB - nC		// .F.
Expected...

	? nA > shortint(nD)				// .F.
What? nD is a shortint. Converting to a shortint should have no effect as nD allready is a shortint. 

	? shortint(nA) > shortint(nD)	// .F.

Strange ideed...

/MathiasH

Den tisdagen den 28:e maj 2013 kl. 16:15:04 UTC+2 skrev dlzc:
> Dear kvste...:
> 
> 
> 
> On Monday, May 27, 2013 8:53:43 AM UTC-7, kvste...@web.de wrote:
> 
> ...
> 
> > 	LOCAL nA, nB, nC AS SHORTINT
> 
> > 
> 
> > 	nA := 748
> 
> > 	nB := 22
> 
> > 	nC := 35
> 
> > 
> 
> > 	IF nA > ( nB - nC )
> 
> 
> 
> Here is the problem, the result of subtraction is NOT a shortint.  It is (probably) a long integer.  When compared to the shortint nA, the result of subtraction may be being truncated, rather than the result being demoted to a shortint, or nA being promoted to a long int.
> 
> 
> 
> > 		? "748 is greater than (22 - 35)"
> 
> > 	ELSE
> 
> > 		? "748 is not greater than (22 - 35). Oops!"
> 
> > 	ENDIF
> 
> 
> 
> It sure looks like a bug to me.
> 
> 
> 
> David A. Smith

[toc] | [prev] | [next] | [standalone]


#5302

Fromdlzc <dlzc1@cox.net>
Date2013-05-29 07:25 -0700
Message-ID<6e14b938-f3b6-46ac-b90e-5e051460a891@googlegroups.com>
In reply to#5301
Dear Mathias:

On Wednesday, May 29, 2013 4:44:22 AM UTC-7, Mathias wrote:
...
> no it's more complicated than this...
...

Really not, but you did find the exact cause, I think.

> 	? shortint(nA) > nB - nC		// .F.
> 
> Expected...

... since nA was already a shortint, and (nB - nC) is a long int (or worse)

> 	? nA > shortint(nD)				// .F.
> 
> What? nD is a shortint. Converting to a shortint
> should have no effect as nD allready is a shortint. 

The problem is sign extending in shortint conversion.  Apparently conversion occurs even if the data type is already shortint.

> 	? shortint(nA) > shortint(nD)	// .F.
> 
> Strange ideed...

The conversion fails miserably on negatives.

David A. Smith

[toc] | [prev] | [next] | [standalone]


#5305

From"Jamal" <jamal@examples.net>
Date2013-05-29 17:18 -0400
Message-ID<ko5raf$hba$1@speranza.aioe.org>
In reply to#5283
Hi Kay-Victor,

It seems to be a bug the > operator internal handling of Short INT.

If you reverse the test to use the <  (less than operator):

IF ( nB - nC ) < nA
    ? "(22 - 35)  is less than 748"
ENDIF

It works as expected. So try putting the complex expression part in left 
hand of the comparison.

Jamal

wrote in message 
news:aa22a440-8e92-444c-9538-3488ccce999d@googlegroups.com...

Hi out there,

I'm getting strange results with simple calculations with shortints. 
Consider this mini-test-app:

FUNCTION Start()

LOCAL nA, nB, nC AS SHORTINT

nA := 748
nB := 22
nC := 35

IF nA > ( nB - nC )
? "748 is greater than (22 - 35)"
ELSE
? "748 is not greater than (22 - 35). Oops!"
ENDIF

_Wait("")


Try it and tell me what happens here. When the three variables are declared 
as longint, everything seems ok. But when they are shortints, the result is 
obviously wrong. It seems to me that the results here are well within the 
scope of a shortint, so what am I missing? (Using VO 2.8 Sp2)

Thanks for input
Kay-Viktor 

[toc] | [prev] | [next] | [standalone]


Page 1 of 2  [1] 2  Next page →

Back to top | Article view | comp.lang.clipper.visual-objects


csiph-web