On Wed, Dec 30, 2015 at 12:03 AM, Vladimir Zhbanov (vzhbanov AT gmail DOT com) [via geda-user AT delorie DOT com] <geda-user AT delorie DOT com>= ; wrote:

On 12/29/15, Evan Foss (<= a href=3D"mailto:evanfoss AT gmail DOT com">evanfoss AT gmail DOT com) [via
geda-user AT delorie DOT com] <geda-user AT delorie DOT com> wrote:<= br> ...
>>> could be miss interpreting your plan here but= it sounds like you are
>>> going to replace more of the C with Scheme.
>>
>> Yeah, that's my plan :)
>
> Sorry but I oppose this plan and I think I can= gather a lot of support.

I was going to write an extended answer until I realized it would be overwhelmingly difficult for me due to I am not so fluent in
English as you are. Hence I'll try to manifest what I consider to
be done (and what I worked on so far). Please let me know which
things you consider to be harmful for gEDA. Now then, I list my
wishes here:

- each C-function in geda-gaf should be available for
=C2=A0 Scheme-scripting;
- all geda-gaf programs should be modularized and all their
=C2=A0 functions should be available for other programs
- all duplicated functionality should be unified and cleaned up.

Now, what I did so far (some of my stuff can be found on
htt= ps://github.com/vzh):
- rewrote pretty useless gsymcheck having lots of erroneous C
=C2=A0 functionality in Scheme, thus added
=C2=A0 - new checks;
=C2=A0 - representation of a symbol as a whole as a list having all
=C2=A0 =C2=A0 slotting and net info without conflicts in it;
- added a simple interface for using it in gschem where the user
=C2=A0 can:
=C2=A0 - check common symbol errors (in primitives or attributes);
=C2=A0 - select particular objects (or all them) and check what's wrong=
=C2=A0 =C2=A0 with them (if any)
- wrote a simple DSL to quickly parse our new ini-style config
=C2=A0 files in order to get all their structure in libgeda at once
=C2=A0 (did not finish yet); if I ever will work on it further, it
=C2=A0 could close one of the blueprints we have in the bug-tracker
=C2=A0 for geda
- played with gschem's undo to represent it as a plain Scheme list
=C2=A0 so user could just type something like "(get-undo)" in the= REPL
=C2=A0 and see all undo info (which is now buried in libgeda C structs)
=C2=A0 as a plain list (it is in another branch, didn't check it in
=C2=A0 because it has to be cleaned up, and my attempts to work on DSL
=C2=A0 are on top of it)
- added REPL in background terminal for gschem and found a way to
=C2=A0 add more REPLs which you would be able to use, say, in Emacs,
=C2=A0 using its shiny geiser environment
- did lots of other work (e.g. symbol generator which could work
=C2=A0 as a gschem function, that is could have a menu for calling it;
=C2=A0 whitespace separated netlist to schematic converter; and other
=C2=A0 things) which I'm not motivated to continue to work on until we<= br> =C2=A0 find the direction we'll go further.

I must emphasize one thing: all those Scheme functions/add-ons are
available on the C side as well.

Well it all sounds good and non-destructive given the details.

<= div>=C2=A0

Finally, I consider the position of "get rid of Scheme" to be a Bad Thing which slows down geda development.

Agreed.=C2=A0 It probably wouldn't deserve to be a pri= ority even if we didn't have some hard-core lisp types.=C2=A0 I would s= ay it would be best to avoid as much as possible adding additional dependen= ce on any script language, and if there's no choice it would be nice to= pick one of the most widely used ones.

=C2=A0

Britton