[Oberon] all in one git tree

[Jörg Straube]

> 
> On 2020-12-31 11:17, D EMARD wrote:
>> ... if once it wants to exchange data over network, conversion of text to
>> plain ascii seems like a good idea.
> 
> Right, it's important to distinguish between interchange formats and internal ones, especially since Project Oberon deliberately and provocatively challenges the status quo (even now!), providing corresponding, but not necessarily compatible, formats, protocols and algorithms compared with prevailing systems. I often notice that Project Oberon is criticised as if it should somehow be one of the latter type of systems... :)
> 

> On the other hand, for convenience, sometimes external formats are supported even if they don't quite fit with the philosophy. When publishing the source code on the web, we use .Mod.txt because this is Oberon text converted to ASCII text for the purposes of interoperability. But in fact, if you use PCLink to transfer such an ASCII file into the Oberon system verbatim, it will indeed compile, if it has DOS (CRLF) line endings, because opening this kind of text in the Oberon system is treated as a special case (and always has been).
> 
> However, it may be that when downloading the ASCII file onto a Unix/Linux system it gets LF line endings, so then this compatibility inside Oberon doesn't work. The world is complicated - Oberon is not. ;-)

I transfer file to and from Oberon with tftp (RFC1350, tftp client on OberonOS, tftp server on other side).
tftp offers two transfer modes „netascii" and „octet“. In netascii mode, every OS has to translate the newline code to its internal convention.

br
Jörg