Linestring clipping bug

[artemp]

No description provided.

[springmeyer]

@artemp do you know why the valid=false was there originally? Did we think it was needed for something?

[artemp]

@artemp do you know why the valid=false was there originally? Did we think it was needed for something?

No entirely sure ^. This looks like copy&paste mistake after refactoring. The intention always was to collect all valid geometries in clipped set and skip invalid ones. With that extra valid=false inside the loop the logic became : if last geometry is invalid all set is declared as invalid. Which doesn't make sense really. Let me know if this explains clearly enough. /cc @springmeyer

[springmeyer]

This looks like copy&paste mistake after refactoring.

Okay. Looks like it landed in e589565

Let me know if this explains clearly enough

Yes, thanks for the more detailed explanation. Sounds good 👍

[springmeyer]

codecov/project — 99.37% (-0.01%) compared to baf3b42

Note: there is a bug in codecov that we've reported to them (and they've confirmed) that makes PRs like this fail that remove code, when actually the coverage did not change. So, we can ignore this.

[springmeyer]

👍

[mapsam]

Thanks @artemp!