-
-
Notifications
You must be signed in to change notification settings - Fork 100
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Workaround for transformation from epsg:3857 with wrong results #1562
Conversation
What problem is being solved exactly by adding an additional test dependency (com.sun.xml.ws:jaxws-rt)? |
@copierrj The error is a This is unrelated to the CRS transformation issue, I just needed it to run the tests. As mentioned in the related commit message tests would fail for me otherwise:
|
Also observed this with Java 11 (OpenJDK) and Maven 3.9:
So no idea how the tests run on other systems, but for me a large amount of tests fail without the additional dependencies. |
Any feedback on the actual topic of the PR? |
Java 17 is not supported yet by deegree, but any Java 11 JDK should be sufficient to build deegree (and run the tests) without any additional dependencies needed. If you remove the commit introducing the new dependency and Jenkins subsequently is able to successfully build and test it we're ready to merge your PR. |
To work around problems when transforming from World Mercator to other reference system use WGS 84 as intermediate CRS to transform to.
Thanks! I updated the PR. |
FYI: we're currently experiencing problems on our build server. We'll get back to this PR when these problems are resolved. |
We were experimenting a bit with transformations from EPSG:3857 that are producing wrong results, specifically to EPSG:31468 / EPSG:4314 where data would be moved by ~20 km.
Example:
![grafik-20230317-111301](https://private-user-images.githubusercontent.com/375475/261556703-dc56514b-ff96-4ee2-8b1f-5eb956c1015a.png?jwt=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJpc3MiOiJnaXRodWIuY29tIiwiYXVkIjoicmF3LmdpdGh1YnVzZXJjb250ZW50LmNvbSIsImtleSI6ImtleTUiLCJleHAiOjE3Mzg4OTA2NjgsIm5iZiI6MTczODg5MDM2OCwicGF0aCI6Ii8zNzU0NzUvMjYxNTU2NzAzLWRjNTY1MTRiLWZmOTYtNGVlMi04YjFmLTVlYjk1NmMxMDE1YS5wbmc_WC1BbXotQWxnb3JpdGhtPUFXUzQtSE1BQy1TSEEyNTYmWC1BbXotQ3JlZGVudGlhbD1BS0lBVkNPRFlMU0E1M1BRSzRaQSUyRjIwMjUwMjA3JTJGdXMtZWFzdC0xJTJGczMlMkZhd3M0X3JlcXVlc3QmWC1BbXotRGF0ZT0yMDI1MDIwN1QwMTA2MDhaJlgtQW16LUV4cGlyZXM9MzAwJlgtQW16LVNpZ25hdHVyZT1jNmJmM2I2Y2UyNzdmODY0YWI5NzA3NDNjMWEwMTI1OTY0NTU0MWI3NGU5OTY2NDc0MWUxNThjZDgxZTdmNjU2JlgtQW16LVNpZ25lZEhlYWRlcnM9aG9zdCJ9.bs41HXmZQ0q8W12UyXEBPO3HCqmEcFpqsPgo5Pf0KIc)
Debugging the transformation we could not really pinpoint where the things go wrong.
So for now we here implemented a workaround to get correct results, by having WGS 84 as an intermediate step in the transformation.
This is definitely not pretty and an actual fix would be preferable, but I was wondering if something like this workaround would in principle be something that you would consider to be added to the codebase. Thanks!