Mail Archives: geda-user/2012/07/06/12:34:48
On Fri, Jul 06, 2012 at 11:06:36AM +0200, Gabriel Paubert wrote:
> >
> > Does tests/run_tests.sh still pass?
>
> It seems to, I have tested on two machines, which give the same
> results:the first 4 tests fail (the bom ones), but I don't see
> how they could even be remotely related to the patch.
>
No, I'm sure they're not. As I recall, there was some minor
reordering in the BOM exporter, hence the test breaking. But
I also recall the tests being fixed...perhaps I should take a
look at that.
> Note that with my current patch metric is on by default for Gerber
> generation, so this means that the files compare equal after
> going through gerbv as a png exporter and comparing through.
>
Cool!
> I realise that setting metric by default may break existing
> scripts, so maybe we should stay with imperial Gerber and
> drill files by default. Right now I have imperial in the
> golden subdirectory and metric in the outputs. They do not
> compare at all equal from diff -r, but they do after conversion
> to png. I believe that this is good enough for inclusion.
>
We should not change the defaults. From my experience in the CAD world,
scripts wind up hidden behind scripts, hidden behind cronjobs, hidden
on Mystery Servers that everyone uses but isn't really aware of it, and
if they were to suddenly break, it would cause much undue pain.
> [Making a few tests]
> Even if I stay with imperial, the differences between golden
> and outputs are large: aperture definitions are not the same, etc...
> If there is a bug, it is well hidden, except perhaps for octogonal
> apertures, which I don't use and don't seem to be covered by the
> test suite.
>
You mean to say: the text output is very different, but the actual
gerber content (as seen by gerbv and exported to PNG) is the same?
--
Andrew Poelstra
Email: asp11 at sfu.ca OR apoelstra at wpsoftware.net
Web: http://www.wpsoftware.net/andrew
"You shouldn't trust every quote you read on the Internet." -- Socrates
- Raw text -