[Oberon] PET ? IDE & syntax coloUring ?

easlab at absamail.co.za easlab at absamail.co.za
Mon Nov 13 17:12:06 MET 2006


soren.renner top-posted:
> Thomas Frey is working on TFPET, a refactoring PET, but it doesn't
> currently include the features you mention. A beautifier module used
> to exist for NO which automatically colored and formatted code.
> 
Is PET the editor, which was in v2.3.6 ?

goog: ETH+oberon+"coloring+syntax" doesn't find it.
   and it doesn't know colouring.
Please provide a pointer.

FindFile.Panel  is an example of an utility which was in V2.3.2,
but dropped from later versions.  I've recently found it very useful.
If it's got some problem, that should be made known, for possible
later fixing, rather than discarding the utility.

> 
> > I wonder if there are plans to provide PET with such luxuries as
> >         - code completion
> >         - inline help (through F1 function key for instance)

! Wow: has Aos/BB abandoned the NO mouse-input ?

> >         - suntax coloring
> >         - pretty / auto formatting
> 
Syntax coloring should be data-driven.
Ie. like linux > vim, where any language can be handled, by having
its own data-file.

And this moves towards a syntax editor utility, which I've previously
raised.  Again, the wiki-like incremental approach is applicable:
rather than hoping for a sparkling-complete software; a usable
tool can circulate, which wiki-like can be incrementally refined.

I find syntax coloring and bracket/quotes-matching-coloring
very useful on linux[-oberon].

And this is not far removed from a 'translator' between
languages, which need not be deterministically perfect;
but could be Bablefish-like heuristic.

You can't avoid having to work with multiple languages these
days. So, you should be able to write eg. a device driver by:
* get the linux C/C++ source,
* examine it with the 'multiple language translator tool',
* evolve the oberon code.

The multi language translator tool can also remove the pain
of N-O 'asm-level' syntax eg. <funny N-O syntax using sets>
==> 'PortN AND mask -> PortM'

All this is only possible with a more open, collaborative 
approach.

== Chris Glur.




More information about the Oberon mailing list