gEDA-dev: Broken slotting

[Gabriel Paubert]

On Tue, Sep 11, 2007 at 11:12:09AM +0100, Peter Clifton wrote:
> 
> 
> Thanks for the suggestions. Whilst multi-billion pin packages are a tad
> unlikely, it feels wrong to hard-code it in a way which could prevent
> them ;)

Indeed, the bug was to allocate gigbytes to store the pin number 
for such packages.

> 
> 
> I was thinking I'd replace it with:
> 
> ... = g_strdup_printf( "pinseq=%d", numpins*(slot-1)+pin_counter );

Looks fine, I did not know about that function, but why the "pinseq="
in the format? It was not in the original one.

> 
> The pain (assuming its the correct fix) to accept pinseq=4 as a hit for
> looking up "this slot's pin number 2" on a two pin component.
> 
> Code currently says "find me a pin on this component which matches
> pinseq=2". I need to turn that around... saying "iterate over all pins
> on this component, tell me what their pinseq is" - then test if the
> pinseq is a wraps around to what we want.
> 
> Not a big problem, I just wanted to confirm other developers thought
> this was the right fix (for now at least).

Probably, but you know the code much better than me.

	Regards,
	Gabriel