On Fri, Nov 25, 2016, at 11:04 AM, Gene Heskett wrote:
> On Friday 25 November 2016 10:30:35 John Kasunich wrote:
> 

> > Part 2 - many if not most pin names start with the instance name.
> > For example, component "or2_0" has pins named "or2_0.in0",
> > "or2_0.in1", and "or2_0.out".  Part 1 above displays the instance
> > name at the top of the component.  So it is redundant to display
> > it on every pin.  For any pin, drop all leading characters that match
> > the component instance name before displaying the pin name.
> > That means instead of "or2_0.in0", the pin name will be "in0".
> > And "near_speed.out" becomes "out".  These much shorter pin
> > names will allow either a larger font or smaller boxes.


> > Again comparing to an electrical schematic - if you want to refer
> > to a pin in the overall context, you might say "U1 pin 7".  But on
> > the schematic itself, the pin is just labeled "7", because it is
> > clearly part of the U1 symbol.
> 
> This could be confused by the pin numerology of our modules not having a 
> consistent base1/base0 start. And I sure don't have a clue how that 
> could be fixed without a total compatibility breakage between versions.  
> So we're stuck with it till the rapture I guess.
> 

Something got lost in translation I think.

I am in no way proposing that HAL start using integers to identify
pins.  So it doesn't matter if some modules use "in0, in1, in2",
some use "in1, in2, in3", and other use "foo, bar, blat".

In my first paragraph I simply propose that _inside each symbol_ 
there is no need to display the instance name prefix portion of 
the pin name, since it is the same for every pin.

In the second paragraph I talked about pin numbers as integers,
but that entire paragraph is not part of the proposal, it is simply
pointing out that existing schematic tools do the same thing as
my proposal - they drop the instance name (reference designator)
when it isn't needed for clarity.

-- 
  John Kasunich
  jmkasun...@fastmail.fm

------------------------------------------------------------------------------
_______________________________________________
Emc-developers mailing list
Emc-developers@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-developers

Reply via email to