Re: Re: source-to-source translators

On Thu, 2005-06-16 at 22:54 -0400, Stefan Seefeld wrote:

> > In a sense, it does source (C++) to source (C++ again) translation of
> > templates to working C++ code.
> 
> Sure, but what relevance does that have for source-to-source translation
> in the context of synopsis ?


For example, you could create a generic template-like language for
expressing source files in one language in terms of abstract semantic
constructs in Synopsis' parse tree.  Ugly example for something
attempting to translate to Python:


sub translate_class(c) ^indent_to_current(
[class ^translaterate_identifier(c.name) ( ^commalist( transliterate( base.name for base in c.bases ) ) ):
^add_indent(emit_memfunction(func) for func in c.memberfunc)


etc...

That looks kinda ugly, but I think you see what I mean.  This would
allow you to create custom ways of implementing particular language
features in one language that don't have a good mapping into another
language.  For example, what do you do with non-virtual C++ functions
when you translate them to Python?

I guess that relates to some of the XSLT ideas being thrown around
before.

Have fun (if at all possible),
-- 
The best we can hope for concerning the people at large is that they
be properly armed.  -- Alexander Hamilton
-- Eric Hopper ([email protected]  http://www.omnifarious.org/~hopper) --

_______________________________________________
Synopsis-devel mailing list
[email protected]
http://lists.fresco.org/cgi-bin/listinfo/synopsis-devel

-----BEGIN PGP MESSAGE-----
Version: GnuPG v1.2.6 (GNU/Linux)

iD8DBQBCtYdBjtsvlOwvazYRAtzQAJ47teZF+Czy3tnpxJ0VBpwpJmz3GQCfWG28
I4GxqsvfZ5zsJyQNyGVqBiY=
=9aUC
-----END PGP MESSAGE-----