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
The problem stems from an annotation about adding an out-of-panel position annotate. https://stackoverflow.com/questions/79222655/ggplot2-adds-a-second-image-title-on-the-right-side It has been suggested that ggplot 3.5.0 provides a more convenient location setup : ‘AsIs’ position aesthetics I() . However, when mapping using coords_sf, there are some problems: everything works fine without asserting default_crs in it. Once this parameter is supplied, the labeled position changes. It should go back to using the x,y axis coordinates as defined.
I think the current expectation is that AsIs works only with cartesian coordinates. This probably requires some new mechanism like the one proposed in #5609. But, probably this needs to be noted on some documentation at least.
I'm fine with adhering to npc coordinates when x and y are both 'AsIs' variables. I'm unsure what the 'principle of least surprise'-outcome would be when only one of x/y is an 'AsIs' variable, and the other is a regular coordinate.
The problem stems from an annotation about adding an out-of-panel position annotate. https://stackoverflow.com/questions/79222655/ggplot2-adds-a-second-image-title-on-the-right-side It has been suggested that ggplot 3.5.0 provides a more convenient location setup :
‘AsIs’ position aesthetics I()
. However, when mapping usingcoords_sf
, there are some problems: everything works fine without assertingdefault_crs
in it. Once this parameter is supplied, the labeled position changes. It should go back to using the x,y axis coordinates as defined.Expected results:

The position of the labeling has changed and seems to follow the definition of the axes:

The text was updated successfully, but these errors were encountered: