Pro forma PR - do not merge

[busstoptaktik]

What happens to tests if we get rid of proj_def.dat? See #201

[kbevers]

You are not going to get away with removing WGS84 as the default ellipsoid :-)

[busstoptaktik]

@kbevers sang:

You are not going to get away with removing WGS84 as the default ellipsoid :-)

(see my recent comment over at #201)

[kbevers]

I meant for this test... Try changing the default to GRS80 and see how big the impact on the test data is.

[busstoptaktik]

I meant for this test

That will be my next commit. Until now, I just want to see how bad it gets (how many non-express assumptions are, at least, in the code - apparently calcofi depends hard on WGS84, which would make sense if talking about the datum, but not the ellipsoid)

[busstoptaktik]

Fails miserably, but only (?) due to the expected submillimetric differences in testvarious output file tv_out. Mostly a matter of creating a new tv_out.dist file.

[kbevers]

It doesn't look that bad actually. It is at most 5 or 6 test cases that are affected. Should be easy enough to check if they are something other deviations on the 8th decimal, which is most likely the case. An updated tv_out.dist is definitely the way to go then.

[busstoptaktik]

(Closing for now, having gained the insight, that solving the technical aspect of this is not a big problem. Deferring further work until this has been discussed on the mailing list - hopefully first after next release)