Re: gEDA-dev: Parts DB API: the story so far

[Levente <leventelist-at-NOSPAMgmail_dot_com>]

DJ Delorie <dj@delorie.com> wrote:
> 
>> http://archives.seul.org/geda/user/Dec-2007/msg00497.html
> 
> Thanks.
> 
>> Or just search your inbox for that message ID.
> 
> My "inbox" goes back nine years, about 100k messages per year.  That's
> nearly a million messages.  "Just search" is not trivial.
> 
> Hmmm.... I'm going to have to write a search engine for that some day.
> 
> 
> Comments on your email:
> 
> We can't convince you that embedded stops people from doing things
> because it doesn't.  However, some operations are easier when they're
> not embedded, some are easier when they are.  I don't think we should
> make any changes that *require* embedded or non-embedded.
> 
> Perhaps the "unique name" for symbols could be, say, an url?  We can
> make up our own protocols (http, ftp... psql, m4, etc) to point at the
> backends we offer.  Protocol chooses a plugin, host choses a source
> within that plugin, path chooses a symbol.
> 
> For searching I've suggested something like what xfontsel does.
> Similar technique, different details.
> 
> I think the workflow needs to be:
> 
> * user picks generic symbol for initial schematics
> 
> * user picks details that work for that generic symbol.
> 
> Thus, choosing resistor-1.sym leads you to the specific values,
> vendors, etc.  that can be used with that symbol.

I think choosing a vendor is out of the scope of an electronic design. I think
part must mean an electronic part. After the design phase, there must be a
post processor where you can decide from what vendor you order stuff. However,
the concept written here is good, I believe.

> My view of the database matches generic symbol names with specific
> manufacturer parts.  However, I did consider that multiple generic
> symbols could map to one "generic symbol name".  I think that's what
> your "compatible" is.  Thus, many .../resistor-*.sym could map to a
> "resistor" symbol, then "resistor" could map to all the manufacturer
> parts.
> 
> "Old fashioned" flat file symbols can still support this if the
> compatibility database is separate from the symbol database.
> 
> 
> _______________________________________________
> geda-dev mailing list
> geda-dev@moria.seul.org
> http://www.seul.org/cgi-bin/mailman/listinfo/geda-dev
> 

-- 
Levente
http://web.interware.hu/lekovacs



_______________________________________________
geda-dev mailing list
geda-dev@moria.seul.org
http://www.seul.org/cgi-bin/mailman/listinfo/geda-dev