Hello:
good afternoon.
When compiling a prg using Context and using PSPad editors (unfortunately VXbase++ does not let us 
compile a single prg) it has an error in the line marked in the attached file that Xpp does not 
recognise. The command line and Pbuild recognize. Do somebody know why?

Beling
TIA

On Fri, 18 Dec 2009 16:16:06 -0200, Carlos Antonio Beling wrote:

> Hello:
> good afternoon.
> When compiling a prg using Context and using PSPad editors (unfortunately VXbase++ does not let us 
> compile a single prg) it has an error in the line marked in the attached file that Xpp does not 
> recognise. The command line and Pbuild recognize. Do somebody know why?
> 
> Beling
> TIA

Should the line in the if statement be:

nMarginBottom := cabCRT_MARGIN_BOTTOM    

not >=

Mark Sergent
InnSource Solutions Inc.
mark@innsourcesolutions.com

Hello Mark:
good afternoon.
Many thanks.

You are right. The problem is that Xpp did not reconized it as an error.

Beling
Happy New Year

Mark Sergent escreveu:
> On Fri, 18 Dec 2009 16:16:06 -0200, Carlos Antonio Beling wrote:
> 
>> Hello:
>> good afternoon.
>> When compiling a prg using Context and using PSPad editors (unfortunately VXbase++ does not let us 
>> compile a single prg) it has an error in the line marked in the attached file that Xpp does not 
>> recognise. The command line and Pbuild recognize. Do somebody know why?
>>
>> Beling
>> TIA
> 
> Should the line in the if statement be:
> 
> nMarginBottom := cabCRT_MARGIN_BOTTOM    
> 
> not >=
> 
> 
>

Carlos Antonio Beling wrote:

> You are right. The problem is that Xpp did not reconized it as an error.

This is caused by the /s commandline switch that just does a syntax check
without creating any object file.

No sure if this is the intended behaviour.

Thomas

Hello Thomas:
good morning.
It was intended. I think XPP should detected the statment as an error and did not it.
I am attaching an example of error that I sent to Alaska. Please compile the program B.prg and you 
will see errors that might be shown and was not.

Beling
Cheers

Thomas Braun escreveu:
> Carlos Antonio Beling wrote:
> 
>> You are right. The problem is that Xpp did not reconized it as an error.
> 
> This is caused by the /s commandline switch that just does a syntax check
> without creating any object file.
> 
> No sure if this is the intended behaviour.
> 
> Thomas

Carlos Antonio Beling wrote:

> I am attaching an example of error that I sent to Alaska. Please compile the program B.prg and you 
> will see errors that might be shown and was not.

I know, I tried it with your sample code myself.

With "intended" I meant that I can't tell if Alaska designed the /s command
line switch the way it is now on purpose or if this is just an unwanted
feature/side effect (also known as "bug" 

regards
Thomas

Hello,

If you look for compiler switches in the help file, you find:

/s   Only test syntax
  If the /s switch is used, the compiler merely performs a syntax check of 
the source code file and does not create an OBJ file.

Regards,
Jack Duijf

The line is not syntacticly wrong. It is logically wrong.

It is just a comparison of values that returns a logical. Since no
assignment is made the return value is just popped off the stack. It is
logically equivalent to a NOP.

Jim




On 01/07/2010 01:24 PM, Jack Duijf wrote:
> Hello,
> 
> If you look for compiler switches in the help file, you find:
> 
> /s   Only test syntax
>  If the /s switch is used, the compiler merely performs a syntax check
> of the source code file and does not create an OBJ file.
> 
> Regards,
> Jack Duijf

Hello Jack and Jim:
good afternoon.

Sorry if I disagree with you: the coded line is "flying": it does not have neither a sign operator 
nor is between parentheses nor is a command.  This way I think that the compiler should, at least, 
warn that something may be wrong.  

Beling
Cheers


James Loughner escreveu:
> The line is not syntacticly wrong. It is logically wrong.
> 
> It is just a comparison of values that returns a logical. Since no
> assignment is made the return value is just popped off the stack. It is
> logically equivalent to a NOP.
> 
> Jim
> 
> 
> 
> 
> On 01/07/2010 01:24 PM, Jack Duijf wrote:
>> Hello,
>>
>> If you look for compiler switches in the help file, you find:
>>
>> /s   Only test syntax
>>  If the /s switch is used, the compiler merely performs a syntax check
>> of the source code file and does not create an OBJ file.
>>
>> Regards,
>> Jack Duijf
>

No it is not

Someproced() is fine
SomeFunction() is fine

A+B is fine

They all just evaluate the values and do not assign them to anything

Perfectly legal even if it is not what YOU want it to do.

Jim



On 01/08/2010 01:59 PM, Carlos Antonio Beling wrote:
> Hello Jack and Jim:
> good afternoon.
> 
> Sorry if I disagree with you: the coded line is "flying": it does not
> have neither a sign operator nor is between parentheses nor is a
> command.  This way I think that the compiler should, at least, warn that
> something may be wrong.  
> 
> Beling
> Cheers
> 
> 
> James Loughner escreveu:
>> The line is not syntacticly wrong. It is logically wrong.
>>
>> It is just a comparison of values that returns a logical. Since no
>> assignment is made the return value is just popped off the stack. It is
>> logically equivalent to a NOP.
>>
>> Jim
>>
>>
>>
>>
>> On 01/07/2010 01:24 PM, Jack Duijf wrote:
>>> Hello,
>>>
>>> If you look for compiler switches in the help file, you find:
>>>
>>> /s   Only test syntax
>>>  If the /s switch is used, the compiler merely performs a syntax check
>>> of the source code file and does not create an OBJ file.
>>>
>>> Regards,
>>> Jack Duijf
>>

Hello Thomas:
good afternoon.

I think preprocessor has another one feature (???):
please compile the attached file with the option /p for known it.

Beling
Regards


Thomas Braun escreveu:
> Carlos Antonio Beling wrote:
> 
>> I am attaching an example of error that I sent to Alaska. Please compile the program B.prg and you 
>> will see errors that might be shown and was not.
> 
> I know, I tried it with your sample code myself.
> 
> With "intended" I meant that I can't tell if Alaska designed the /s command
> line switch the way it is now on purpose or if this is just an unwanted
> feature/side effect (also known as "bug" 
> 
> regards
> Thomas

Should not have a semicolon in the #define.

Except for the preprocessor commands them selves no syntax checks are
done. Again it is up to the programmer to do this.

Jim



On 01/08/2010 01:50 PM, Carlos Antonio Beling wrote:
> Hello Thomas:
> good afternoon.
> 
> I think preprocessor has another one feature (???):
> please compile the attached file with the option /p for known it.
> 
> Beling
> Regards
> 
> 
> Thomas Braun escreveu:
>> Carlos Antonio Beling wrote:
>>
>>> I am attaching an example of error that I sent to Alaska. Please
>>> compile the program B.prg and you will see errors that might be shown
>>> and was not.
>>
>> I know, I tried it with your sample code myself.
>>
>> With "intended" I meant that I can't tell if Alaska designed the /s
>> command
>> line switch the way it is now on purpose or if this is just an unwanted
>> feature/side effect (also known as "bug" 
>>
>> regards
>> Thomas