Hannes Mehnert

Hey, I was wondering - some months later - what the status of this is?

related https://discuss.ocaml.org/t/reproducible-builds-across-different-architectures/14799 //cc @dra27

Further discussion at https://discuss.ocaml.org/t/dune-lockfiles-are-not-cross-platform-how-should-we-approach-checking-them-into-version-control/14992/19 What introduced this is the PR https://github.com/ocaml/opam-repository/pull/25861 > The key rule is that `arch-` and `system-` should never be used in opam files because there are...

As far as I can read, initially Tags were a set of strings, since #2 there's been the GADT. From my current perspective, a tag should be a string as...

Another piece of motivation is that e.g. influxdb uses strings there as well https://docs.influxdata.com/influxdb/v2/reference/syntax/line-protocol/#tag-set

My point of view is: - we have announced this change here - we also have announced it at https://discuss.ocaml.org/t/ann-opam-repository-policy-change-checksums-no-md5-and-no-extra-files/14720/1 - macOS homebrew carries opam 2.1.6 which has the gpatch...

And finally, I don't see a way to have a failsafe and smooth transition. Of course we could wait (forever?) until everyone updated to 2.1.6. We could also push for...

I was wondering whether such a patch would be sensible (and would work): ```diff diff --git a/repo b/repo index ff07c1813e..8e7eb9533c 100644 --- a/repo +++ b/repo @@ -8,4 +8,7 @@ announce:...

> > * we also have announced it at https://discuss.ocaml.org/t/ann-opam-repository-policy-change-checksums-no-md5-and-no-extra-files/14720/1 > > The discuss post announces this as a policy change, which i think is fair to understand this way....

> yes i think it would be nice, however currently users of the default opam-repository (via opam.ocaml.org) use a different `repo` file generated here: https://github.com/ocaml-opam/opam2web/blob/d3d36b81174cbbb9df9dfe525cd2cfe77fd86206/bin/opam-web.sh#L26 so this one would probably...