Daniel Huppmann

Thanks for raising this issue, see a few comments below. Let's please try to keep issues and discussions narrow and start new issues where possible. **Cross-reference to legacy variables/regions or...

Please rebase this branch to `main` or start a new branch and carry over the relevant changes. Also keep in mind that #188 added **Emissions|{Level-3 Species}|AFOLU|Land|Fires** and subcategories, so these...

> For the second point, I'd very much integrate it into `dsd.validate()`. For the variables we already test for units and so we've already broken the pattern of only outputting...

Quick recap of yesterday's discussion - would be great to have this PR merged and create a new release by end of this week... 1. Calling `definition.validate(df)` should raise an...

Thanks @dc-almeida, but I'm confused why this would raise an error. Shouldn't we expect that an empty dataframe is ok?

> The error raising is not added in the PR, it is the default behaviour of the region processor whenever processing results in an empty dataframe. I can change it...

The comment by @phackstock above https://github.com/IAMconsortium/nomenclature/pull/495#pullrequestreview-2850809242 seems to have taken this PR in a wrong direction. It probably wasn't helpful that we have a test dataframe with only one variable....

Thanks @dc-almeida, that's a useful illustration of a corner case and explanation for the confusion. My suggestion would be the following, also going in the direction of @phackstock's intuition: 1....

Following our discussion, my understanding of the preferred solution that only requires updating some GCAM mapping files. 1. For **native-regions**, distinguish between rename and keep-as-is items - handle them separately...

Thanks @dc-almeida for following this rabbit hole all the way to the end! UPDATE: removed earlier answer I guess your option 3 (excluding region_B) is the right course of action....