[fpc-devel]const sLineBreak

Darek Mazur darekm at emadar.com
Wed Sep 22 11:51:06 CEST 2004


Użytkownik Tomas Hajny napisał:
> From:           	Darek Mazur <darekm at emadar.com>
> To:             	fpc-devel at lists.freepascal.org
> Subject:        	Re: [fpc-devel]const sLineBreak
> Send reply to:  	fpc-devel at lists.freepascal.org
> 	<mailto:fpc-devel-request at lists.freepascal.org?subject=subscribe>
> 	<mailto:fpc-devel-request at lists.freepascal.org?subject=unsubscribe>
> Date sent:      	Thu, 16 Sep 2004 16:52:54 +0200
> 
> 
>>Użytkownik Tomas Hajny napisał:
>>
>>>On Wed, 15 Sep 2004 23:59:57 +0200, Darek Mazur wrote
>>>
>>>
>>>>Użytkownik Michael Van Canneyt napisał:
>>>>
> 
>  .
>  .
> 
>>>> And second I don't understand why You don't want change it. I 
>>>>don't see any true reason to left it unchanged. Everybody, who want
>>>>to control output must avoid WRITELN (I don't test READLN, but I use
>>>>it very rare).  I want do decide, how function work, not OS.
>>>>
>>>>Of course I can do it by hand, but I think that somebody help me (I
>>>>must change   more than 1000 places in my program)
>>>
>>>
>>>We will discuss this internally and let you know the result.
>>
>>Thx, that sounds good.
> 
> 
> A new procedure SetTextLineEnding (var T: text; S: string) was added 
> to System unit in CVS. This allows you to change line ending for 
> individual text files. After call to Assign the line ending for that 
> file is set to the default line ending on your platform, then you can 
> change the default value using the newly provided procedure - since 
> that time all WriteLn calls will use your new value. Changing the 
> default value system-wide would be dangerous because of output to 
> console, redirection, libraries possibly expecting particular 
> behaviour etc.

That is better solution, than I think .

But is function like
    GetTextLineEnding(vat T: text): string

I don check CVS yet

Darek



> 
> Tomas
> 
> 
> _______________________________________________
> fpc-devel maillist  -  fpc-devel at lists.freepascal.org
> http://lists.freepascal.org/mailman/listinfo/fpc-devel
> 
> 





More information about the fpc-devel mailing list