🦋 Changeset detected

Latest commit: 147c0e1

The changes in this PR will be included in the next version bump.

Not sure what this means? Click here to learn what changesets are.

Click here if you're a maintainer who wants to add another changeset to this PR

this is neat! can we rely on browser and development conditions in exports maps to always be correctly resolved.
As long as it happens in the vite build step i think that is the case, but what about edge runtimes, do they always get output of vite or another bundler? or would there be a point where we have to rely on them to do the right thing with these conditions?

using !browser for ssr a good step for compatibility 👍 otherwise node vs deno vs x would be playing catchup.

what about edge runtimes, do they always get output of vite or another bundler?

Yes. If they weren't then the code currently in main wouldn't work since those environments wouldn't know how to handle import.meta.env either

We still have __SVELTEKIT_DEV__ - do we need to keep that because it won't treeshake correctly or is the DEV equivalent from esm-env true at the wrong times?

Thanks! I wasn't aware of __SVELTEKIT_DEV__. I've replaced it as well. However, doing so causes two tests to fail that I believe is due to a bug that Rich discovered in Vite 3 (vitejs/vite#9203). I fixed it for Vite 4, so after we upgrade, I'll update this branch against master and I imagine those tests should start passing. I'll mark this as a draft until then