Report superceded/replaced drafts in search

ietf-tools / datatracker

The day-to-day front-end to the IETF database for people who work on IETF standards.

https://datatracker.ietf.org

BSD 3-Clause "New" or "Revised" License

624 stars 391 forks source link

Report superceded/replaced drafts in search #6276

Open ekr opened 1 year ago

ekr commented 1 year ago

Description

By default the datatracker doesn't show expired drafts. However, when a draft has been superceded by some other draft which is still active or an RFC I think we should show it by default.

Right now the behavior is inconsistent:

When I search for "rescorla" in datatracker, I get two RFCs: 5430 and 4347, both of which were individual submissions, even though the drafts themselves are expired.
By contrast, draft-rescorla-tls-dtls13/, which then became draft-ietf-tls-dtls13 which then became RFC 9147 is not shown.

I would propose that if a draft or its descendent is still active or an RFC, then we should show it, perhaps by redirecting you to that descendent. This helps answer the question that people want to know, which is "what happened with draft-xyz"

Code of Conduct

[X] I agree to follow the IETF's Code of Conduct

rjsparks commented 1 year ago

That https://datatracker.ietf.org/doc/draft-rescorla-tls-dtls13/ doesn't show RFC9147 in the header graphic is unexpected, and whatever is at the root of that may also be affecting the search results.

Given that https://datatracker.ietf.org/doc/search?rfcs=on&activedrafts=on&by=author&author=rescorla returns such a wildly different result, this is likely not an issue with just -tls-dtls13.

Once that's worked out, then at:

perhaps by redirecting you to that descendant

We would be showing the descendants as the search results already (as we did for RFC5430 above. Are you proposing something different?

ekr commented 1 year ago

Yes, it's not not just a problem with dtls13. That was merely an example. For instance,

https://datatracker.ietf.org/doc/draft-thomson-ohai-ohttp/ behaves the same way (search for "thomson"). Note that in that case the superseding draft is live and there is no RFC.

I think what 5430 does is fine for things that became an RFC but note that you also need to handle the case of drafts that are now WG drafts, etc.

larseggert commented 1 year ago

The endpoint that creates doc.json (which the js that makes the timeline graph works on) fails to pick up the RFC as a related document.