Re: gEDA-user: Matching footprints with symbols

[Armin Faltl <armin.faltl@xxxxxx>]

Hello Vladimir,

the point in not using a text file format for this but a relational database
with SQL is not the data storage but the capabilities of the database 
server.
It allows modeling of relations and more important, relational queries.
This can look something like:

SELECT elem.footprint FROM elem, part
WHERE part.name LIKE "TLC555%" AND elem.package = part.package AND 
elem.process = "reflow";

listing all the known footprints of all known packages for parts 
starting with "TLC555",
that are good for reflow soldering.
The real tables will be somewhat more complicated - it's just to give an 
idea.
The DB avoids all the hassle with file parsing, inventing file formats 
and in-memory datastructures
because all this is done by the server. All one needs is a simple list 
to receive the answer.

The raw table data can be inserted and retrieved using text files, 
that's formats are well
documented and extremely well tested. For what they do DB-servers are 
even very fast.

Armin


Vladimir Zhbanov wrote:
> DJ Delorie wrote:
>   
> > > between footprints and its instances on a board and am able to think
> > > of things like SQL-databases providing a clear, yet flexible mapping
> > > between
> > >       
> > Perhaps this idea of mine is relevent?
> >
> > http://www.delorie.com/pcb/component-dbs.html
> >
> >     
>
>
> Hello.
> I am newbie here, too.
> My suggestion is to make database in text format for gschem, pcb and
> probably others. Idea is that central object is real component from
> programs point of view. There should be a way for users to choose
> components for their symbols (for example for pin mapping) and/or
> symbols/footprints for their components. I intentionally use only
> general elements and imply initial schematic capture so DB should be
> rather light than heavy.
> DB format could be simple and it is stuff to discuss. Simple example:
>
> (component_name (pcb_name1 pcb_name2 ...) (gschem_name1
> gschem_name1 ...) ...)
>
> or maybe so:
>
> component=component_name {
> 	company=company_name
> 	...
> 	footprint=footprint_name1
> 	footprint=footprint_name2
> 	...
> 	symbol=sym_name1
> 	symbol=sym_name2
> 	...
> 	whatelse=whatelse1
> 	...
> }
>
> and so on.
>
> The database could be distributed and all gEDA programs could use it.
> For example I'd like to have common gEDA database and my own local
> database which contents line (it is just silly example):
> ...
> (7400 (7400-1.sym 7400-2.sym) (DIP14 DIP14N))
> ...
>
> That way programs could know what symbols user may use for component.
> Then instead of symbol selecting in gschem user could select real
> component's name ('component=' attribute?) and preferred symbol and
> footprint for it. And in pcb he/she could select required component and
> then footprints for it only instead of searching footprint firstly in
> datasheet and then in pcb library.
>
> All above could be compromiss for disctinct programs of gEDA project
> until new better format will be accepted.
>
> And that way distinct databases for different users, sizes, standarts,
> locales and so on could be created.
>
> VZh
>
>
>
>
> _______________________________________________
> geda-user mailing list
> geda-user@xxxxxxxxxxxxxx
> http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
>
>   


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