[Proj] PROJ 5.0.0RC5

[Kristian Evers]

This is a difficult nut to crack. On the hand one cs2cs/pj_transform can behave geodetically correct
OR on the other hand it can behave in a slightly wrong manner but return the expected
coordinates when the web Mercator is used.

In https://github.com/OSGeo/proj.4/pull/743 the first situation was instated without realizing the
full impact. Historically cs2cs has done the latter and should continue to do so in order to not screw
up more or less all web mapping applications on the internet. I have reverted the fix in PR #743 here:

https://github.com/OSGeo/proj.4/pull/830

Granted that all tests pass the fix will be included in the final release of 5.0.0.

Thanks for digging in to this Jeff!

/Kristian


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/proj/attachments/20180301/9051d0db/attachment.html>