Let's pin 2.0.0; it has been in use for a year now, right?

Let's pin 2.0.0; it has been in use for a year now, right?

But it does not work for me with 2.0.0.
1.3.0 is the maximum that works.

Take a looks at 1.3.0 (works fine) and 2.0.0 (Loading... infinitely)
2-cases.zip

Let's pin 2.0.0; it has been in use for a year now, right?

No, it hasn't been: in 2.0, voyager.min.js was removed, so https://cdn.jsdelivr.net/npm/graphql-voyager/dist/voyager.min.js (without version) still points to the 1.3 version.

However, the corresponding CSS file is still present in 2.0, so there was CSS from 2.0 and JS from 1.3 for a year.
And in 2.1, it seems, the CSS finally became incompatible with the JS v1.3.

Switching to the new voyager.standalone.js from 2.0 (supersedes voyager.min.js) doesn't help because of some configuration changes in 2.0.

Ok sounds good. Let's add the sha hash to the tag to ensure it does not change in the future, and I'll merge it.

Ok sounds good. Let's add the sha hash to the tag to ensure it does not change in the future, and I'll merge it.

Sorry, I am not sure what this is. Is it something I should do?

https://developer.mozilla.org/en-US/docs/Web/Security/Subresource_Integrity

Probably some website like this can calculate it.

https://www.srihash.org/

I assume the hosting code needs to be updated for the 2.0 version to work?
Like this: https://github.com/graphql-kit/graphql-voyager/blob/c644fb529f8aa1f987c505aef4527207a0c752fd/example/cdn/index.html#L31-L51

@alexander-jesner-AP Yes, I think so and hope this will be eventually the case, the proper fix. That's why I mentioned the same example in #1152

@Shane32 I have engaged for the quick fix of the regression, to make things working again. I now see it's getting more involved and it's not really my area of expertise. At the moment I am not able to implement what you are requesting. Thank you for your help.

I'll issue a release with it pinned to 1.3.0

If someone wants to post a PR with it properly upgraded to 2.x, we can merge that in subsequently.