Re: gEDA-user: symbol files

[Dave N6NZ <n6nz@xxxxxxxx>]

Peter TB Brett wrote:
> 
> Yep.  The reason this sort of feature isn't being developed (and hasn't 
> already *been* developed) is because it's neither clear how to get it right, 
> even if it's assumed that it's something that actually fits into the gEDA 
> paradigm!

Well, there is clearly a missing piece of data: mapping from logical 
function onto package pin number.  gschem symbols should not force 
direct physical pin numbers.  The symbols must allow for a pin number 
mapping depending on a package selection attribute, and the schematic 
should display the appropriately mapped pin number.  That part of the 
gschem symbol def is simply broken/weak. (And was addressed sometime in 
the last millennium by other cad packages... like 1980's time frame.)

The mapping of logical pin identifier onto physical pin identifier (not 
always a number either!) needs to be an extra set of mapping data.  For 
convenience, it could be optional so that a mapping of 1:1, 2:2... n:n 
doesn't have to be included in every symbol, it could be a default.

One of the things that gripes me about geda development is that we are 
still arguing about whether it is "doable" to do stuff I used in the 
1980's.  *sheesh*  There may not be any single "right" way to do it. 
But there are certainly ways.

And yes, it *is* easy to sit here whining and yet not submit any 
patches... thanks for asking :)

Reality check: Is this broken? Yup. Has this problem been solved before? 
Yup. Does it get in the way of my current work flow?  No, I simply make 
multiple symbols.  If the pins line up on the page, it is easy enough to 
delete one and drop in its cousin.  Of course, I'm doing small designs. 
    That work flow doesn't scale.

-dave


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