I'm sorry this took so long to get together, but I just posted a PR against
the IFNDR branch with the various instances of UB that Timur identified as
not yet being in the annex.

https://github.com/cplusplus/draft/pull/7864

I did not do these as separate PRs because there's some touch-ups spread
across the existing entries as well as a lot of stuff here. If there are
any bits that require extensive additional discussion i'll put those in
separate PRs.

One thing I did note is that we have some cases where we have UB in
multiple paragraphs that is covered by a single entry in the annex. I've
added \ubdef entries there, because I think having stuff in the main body
where we don't clearly identify that the UB is in the appendix will just
make it harder to identify if we have been thorough in indexing
everything. This leaves us with some dangling references, and we should
resolve (editorially) how exactly we want to capture this state.

Any specific issues with the contents of the PR should be handled in github.

On Sat, Apr 12, 2025 at 3:14 PM Jens Maurer via Core <core_at_[hidden]>
wrote:

>
> As discussed in Hagenberg, there is interest in having a UB/IFNDR Annex
> as a first step towards issuing recommendations on how to mitigate
> those.
>
> We take Shafik's P3075 and his work on a LaTeX patch as a basis for that.
> His work (plus fixes) has been migrated to the main github repository for
> the
> Working Draft, represented in this pull request:
>
> https://github.com/cplusplus/draft/pull/7826
>
> A PDF rendering of the current state is available here:
>
> https://github.com/cplusplus/draft/actions/runs/14405692379/artifacts/2928265136
> but that link will become stale as updates are merged.
> For those who can't access github, I've uploaded a PDF snapshot here:
> https://wiki.edg.com/pub/Wg21telecons2025/SG12/std-ub-ifndr.pdf
>
> An initial collection of editing instructions is here:
> https://github.com/cplusplus/draft/wiki/UB-and-IFNDR-Annex
>
> A sample pull request against the "working branch" is here:
> https://github.com/cplusplus/draft/pull/7828
>
> Going forward, all discussion should happen on the "sg12" reflector
> (as agreed by Herb and JF).
>
> Thanks,
> Jens
> _______________________________________________
> Core mailing list
> Core_at_[hidden]
> Subscription: https://lists.isocpp.org/mailman/listinfo.cgi/core
> Link to this post: http://lists.isocpp.org/core/2025/04/17845.php
>