CRS bug fix with temporary subset_geom clipping object, original get_target_crs function #125
Conversation
There was a problem hiding this comment.
The change here overlaps with the #126 PR to which I committed a further edit.
There was a problem hiding this comment.
I remember talking about this change, but I don't remember enough to feel confident this is the entire solution we wanted. Plus, we talked more recently about target CRS versus user-provided CRS, so does that affect this change here?
There was a problem hiding this comment.
The actual CRS bug was fixed previously in this commit: 3dc613b. The CRS related-change here improves upon that fix by creating a temporary geometry instead of modifying the request object's geometry to prevent downstream issues. Switching to using separate target_crs and user_geom_crs parameters will be a rather time consuming overhaul of the code, so I propose we fix this small issue now, and fix the larger issue of separating target_crs into two different parameters once we have the capacity to take that on.
No description provided.