Mail Archives: geda-user/2012/12/08/22:17:00
On Sun, 9 Dec 2012, Peter Clifton wrote:
> On Sat, 2012-12-08 at 11:26 +0100, Levente Kovacs wrote:
>> Please note that when I start to draw a line from a copper object, highlighting
>> works as before. This might help find the problem.
>
> That behaviour was deliberately left, as it is required by
> auto-enforce-drc. (You will also find the "Find" button in the GTK HID
> netlist window still has the old behaviour too I think).
>
>
> I have worked up a test-case for the original issue, and having looked
> at it closely, one key issue we should fix is that bad rat-lines are
> being added which compound the shorted mess.
>
>
That problem has hit me multiple times: a mistake in a big board shorting
gnd and vcc, two random orange and a lot of legal-looking green and rats.
I think just looking at the current state it is impossible to decide which
part of the connection is wrong (which object to highlight in orange).
I mean sometimes I start moving a two terminal component along a trace:
when I remove the coponent from it's original place I connect the trace
over the gap and I intend to break it somewhere else to insert the
component again. During this process I have a short which could be
resolved multiple ways. The final resolution does not depend on objective
data of the current layout (shortest paths, closest pads, etc), but on my
plans about moving the component.
However, if we rephase the question from "what causes the short" to "what
user modification introduced the short", it could highlight something that
may be useful in more often. Proper implementation would be complicated
(need to save history of nets), but I see a reasonable cheat here: use the
undo buffer. This assumes there was no short at a point after loading the
board and the user is interested in which object he touched that first
caused the short.
Regards,
Tibor
- Raw text -