Mail Archives: geda-user/2013/04/18/14:52:56
On Thu, Apr 18, 2013 at 11:09:56AM -0700, Andrew Poelstra wrote:
> On Thu, Apr 18, 2013 at 07:57:59PM +0200, Gabriel Paubert wrote:
> >
> > To be precise the original redundant check was:
> >
> > while (printf_spec[i] == '%' || isdigit(printf_spec[i]) ||
> > printf_spec[i] == '-' || printf_spec[i] == '+' ||
> > printf_spec[i] == '#' || printf_spec[i] == '0')
> >
> > but "is_digit(x)" is a superset of "x == '0'", so I eliminated the latter.
> >
>
> *facepalm* yes, of course.
>
> >
> > > Maybe somebody with more experience with gcode can check over
> > > the gcode changes. But they seem reasonable to me.
> >
> > Which gcode? I have not touched the gcode exporter AFAIK.
> >
>
> Oops! I mean't gerber.
>
> > >
> > > When I get a chance to check on that check, and assuming no NAK's
> > > come in, I'll push these.
> >
> > Thanks for taking care of it.
> >
>
> I'd like somebody more awake than I am to check on it first. ;)
Don't worry. Tomorrow I'll send a new set taking into account the couple
of remarks I've got.
This said, make check does not find any error. I have also checked
a single (simple 2 layer) board, the photoplotter files are identical as expected
(except for the date in the header). The drill files were obviously different,
since the last digit is no more zero. Under gerbv with xor drawing it was
esaily visible.
This said, there are visible, but minor, differences under gerbv for at least
text output. I have the same result on 2 machines, so I suspect that the golden
may have to be updated.
Gabriel
- Raw text -