At 07:41 PM 2/6/00 +0100, Stefan Matthias Aust wrote:
>I'd suggest to not depend on any encoding.  It might even change within the
>file.  All you can depend on is, that CR will prepend LF.  And don't depend
>the line end encoding on the platform!  I want to read unix files on
>Windows, too.
>
>For one of my parsers I use the following methods
[excellent exaple snipped]

I read the files in a manner similar to your examples.  The difficulty
surfaces in writing files that will be used by other tools such as RCS, text
editors, etc.

I wanted to write the files using the appropriate platform line termination
convention unless specifically set to use a different one.  This is how I've
seen it done on other dialects, seems reasonable to me, and would ease using
the file with other tools.

Thanks,

--
Richard A. Harmon          "The only good zombie is a dead zombie"
harmonra at webname.com           E. G. McCarthy
Spencer, Iowa