Re: gEDA-user: net labels question

[Dan McMahill <dan@xxxxxxxxxxxx>]

al davis wrote:
> On Sunday 06 April 2008 09:36, gene wrote:
>> Is it ok to name a net "3.3VD", for example, using the "."?  I
>> haven't tried it yet, so was just looking for a quick answer to save
>> some time.
> 
> It's better not to.  Even if gschem accepts it, sometimes the netlister 
> doesn't map things like that properly, and sometimes target tools don't 
> accept it.

I also try to use pretty simple names without any characters which might 
be considered as 'special' by any tools I may run into.  gnetlist does 
provide a facility for mapping gschem names into a possibly more 
restrictive name space needed by particular target netlist formats.  So 
in principle if the "." caused issues for some output netlist type, then 
the gnetlist backend could be fixed up to avoid this.  The switcap and 
pads backends use this capability and maybe some others do.  gnetlist 
does check for shorts caused by renames.  In otherwords if you have a 
backend that applies a toupper() (makes everything upper case) transform 
it will detect if you had MySig and MYSIG which now get shorted and it 
will complain.  So the biggest dangers are that

1)  you have more nets that have a different name in the output format 
than in the schematic which may be confusing.

2)  you increase the probability that you'll have to rename a bunch of 
nets later to avoid collisions

3)  the particular gnetlist backend in question doesn't properly 
implement the net aliasing stuff.  This is not hard to retrofit into an 
existing backend, but it may cause some delays.  I'm pretty certain that 
more backends are missing this than include it.

-Dan


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