Gediminas Rimša

👍 I'll update the PR sometime over the next few days. @codecholeric In the meantime - have you seen this https://github.com/TNG/ArchUnit/issues/919#issuecomment-1189998749? Maybe a larger-scale change would be desired.

Hey, thanks a lot for moving on this, and sorry for dropping the ball :) I will check this today.

I noticed this when trying to implement a "a package should not depend on its subpackages" rule (reasoning: https://javadevguy.wordpress.com/2017/12/18/happy-packaging/). What I was trying to do was something like this: ```...

I ended up implementing the rule I wanted in a slightly different way (shared here: https://github.com/TNG/ArchUnit/discussions/921), which did require the methods of `JavaPackage` discussed here, so my itch can be...

Opened a PR with the proposed fix. Though it seems the root problem here might be that the concept of `JavaPackage` is a little muddled - in some cases it...

> [Jackson Coreutils](https://github.com/fge/jackson-coreutils) might be helpful here as it provides numeric equivalence. Note: `jackson-coreutils` moved under `java-json-tools` group ID: https://github.com/java-json-tools/jackson-coreutils

In case anyone needs a workaround, below you can find a code snippet that rotates the exterior ring by half the ring size and simplifies again (note: interior rings are...

Ran into a similar issue of union of polygons with very high precision coordinates producing an empty result. Works fine when serialized to WKT and loaded into test builder. Did...

First of all, thank you for all the comments - they are immensely useful. > Yes, the buffer boundary artifacts are probably caused by a robustness bug in the buffer...

(Somewhat related to the original issue) @dr-jts I wanted to share one more test case related to negative buffer operation and high precision coordinates. Input polygon looks like this: ...