In article <4070963a@news.povray.org>, dne### [at] sanrrcom says...
> Patrick Elliott wrote:
> > You know.. I never quite understood why such and option even existed.
> 
> Because FTP (and hence many other internet standards) use CR followed by 
> LF to indicate the end of a line, while UNIX went askew and decided to 
> drop the CR off the ends of lines to make it easier to program.
> 
> Hence, if you copy a Unix-based file with only LFs to a Windows-based OS 
> (or, for that matter, any of the many other OSes popular around the time 
> such things were being created), it needs to stuff a CR before each LF 
> in order to make it a valid text file.  If it's something like an 
> executable or a .gif or something, doing that will corrupt the file, 
> since those files aren't supposed to have CRs before every LF.
> 
> Had UNIX stuck with CRLF meaning end of line, or had internet standards 
> mandated that CR, LF, or CRLF would all be valid end-of-line markers in 
> a byte stream, we wouldn't need that.
> 
Yeah. I get that, but like anything else, it is a matter of designing the 
programs on the machine to handle the problem, not assuming that the guy 
downloading it will be on whatever system the machine storing it uses. I 
mean, lets say I upload a Windows file to a unix server, then later 
download it again to a Windows machine. Only in the last step I forget to 
do it in text mode. Oops! It creates problems, not solving it.

-- 
void main () {

    call functional_code()
  else
    call crash_windows();
}

Post a reply to this message