Skip to content

Conversation

@greglucas
Copy link
Contributor

When image tile requests crossing a dateline happen, there is a multipolygon extent that gets passed to the find images function. We don't necessarily want just the tiles that are within those geometries, but also the ones that could be between them.

closes #2579

When image tile requests crossing a dateline happen, there is
a multipolygon that gets passed to the find images function. We
don't necessarily want just the tiles that are within those
geometries, but also the ones that could be between them.
@greglucas greglucas force-pushed the fix-add-image-multipolygon branch from cc2c7dd to 12bf5b4 Compare November 15, 2025 20:11
@greglucas greglucas marked this pull request as draft December 1, 2025 15:24
@greglucas
Copy link
Contributor Author

Thinking about this a bit more, this might expand the domain a little bit too much depending on the bounds of the polygons and turn two halves of an oval shape into a rectangle unnecessarily. I'll try and look into whether buffering the multipolygon is more appropriate here because I am now wondering if this has to do with the inexact boundary of Orthographic (or any curved boundaries).

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Coastlines shifted when using add_image with Orthographic projection centered on dateline

1 participant