Re: gEDA-user: Thoughts using gschem

[John Griessen <john@xxxxxxxxxxxxxx>]

Sean D'Epagnier wrote:

> Ok, so whenever you place a symbol that isn't in the project's library,
> it should copy it into the project's library and switch to the file
> in the library.  Then  you can modify that element directly by loading it
> as a symbol.
> 
> If you select a symbol not from the project's directories should it create
> a second copy of that symbol and place it in the project's directory?

That is a good simplifying assumptioin to code up first.  Many people would
operate that way, which is, "keep all symbols local and unmodified by libraries
unless I say so by replacing the symbol."

That brings up a possible conflict.
When replacing a symbol where there are multiple placements of it,
how to replace all with the same name, or how to identify that one or several but not all
come from a different library?  This gets back to the complexity that chip
design methods handle....the defacto way of doing that is to define libraries in a hierarchy of directories
and define which library search path to use when placing new symbols.  The usual chip design method has
all representations in the custom library -- schematics, symbols, simulation models... and
with a custom symbol, you get a custom schematic, which could have the same module name, but be on a different
path (e.g. /home/john/geda/circuits/proj-x/custom-lib/lvds-buffer/schem and 
/home/john/geda/circuits/proj-x/custom-lib/lvds-buffer/symbol and
/home/john/geda/circuits/proj-x/custom-lib/lvds-buffer/gnucap)

John Griessen

> Or should it only make a second copy if the first one was edited?
> 
> Would the locations for the project's directories be specified in gafrc?
Yes, that's the right purpose for gafrc.

> Should the symbols be places in "component-library" or should there be a new
> tag for them?
Places.


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