You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Transformations to and from the Google's pseudo/web Mercator returns incorrect results.
The problem was introduced in #743 which fixed #22 but unfortunately also broke the web Mercator. This happens since the web Mercator handles ellipsoids incorrectly (calculating coordinates on the ellipsoid, but projecting them on a sphere).
So, on 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. We want the latter option to make sure that PROJ behaves similarly to previous version.
Until a new version is release the a work around for the problem is to remove +nadgrids=@null from the web Mercator proj-string. This might be necessary to do in the epsg file, where the system is known as:
Set flags for Intel compiler to prevent incorrect optimization of
arithmetic expressions OSGeo#826. Guard against nans in sincosdx OSGeo#834.
Issue OSGeo#831 is not addressed here (need more information...).
Reported by @jmckenna - se details below.
Transformations to and from the Google's pseudo/web Mercator returns incorrect results.
The problem was introduced in #743 which fixed #22 but unfortunately also broke the web Mercator. This happens since the web Mercator handles ellipsoids incorrectly (calculating coordinates on the ellipsoid, but projecting them on a sphere).
So, on 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. We want the latter option to make sure that PROJ behaves similarly to previous version.
Until a new version is release the a work around for the problem is to remove
+nadgrids=@null
from the web Mercator proj-string. This might be necessary to do in theepsg
file, where the system is known as:Change it to
and it should work again.
More here: http://lists.maptools.org/pipermail/proj/2018-February/008082.html
The text was updated successfully, but these errors were encountered: