[Proj] invgeod results for small distances

[Hermann Peifer]

Hi All,

I am wondering if I am the only one who is somewhat puzzled with 
invgeod's results for small distances. If not: would it be worth fixing 
for the upcoming release?

Hermann


# This looks reasonable for a difference in the 6th decimal
[peifer at whitefish:~]> echo 0 0 0 .000001 | invgeod +ellps=WGS84
90d     -90d    0.134


# This should then be 0.013 for the 7th decimal, but it isn't :-(
[peifer at whitefish:~]> echo 0 0 0 .0000001 | invgeod +ellps=WGS84
-2147483648d-2147483648'-nan"   -2147483648d-2147483648'-nan"   -nan


# This isn't true in real life, but at least better than Not-A-Number
[peifer at whitefish:~]> echo 0 0 0 .0000001 | invgeod +ellps=sphere
90d     -90d    0.000