Anton Strogonoff

@ronaldtse to clarify, if we really wanted to switch to non-numbered directories _without_ altering the contents provided from IETF side by rsync, we can describe this alias mapping in bibxml-data-archive...

We will not be renaming any `bibxml-data-archive` paths, by design they must *exactly* match xml2rfc paths for fallback purposes. This issue will be closed when there is a JSON file...

@ronaldtse Fallback purposes. Here is an example: https://github.com/ietf-ribose/bibxml-service/issues/170. When Relaton data was silently changed, `rfcsubseries` xml2rfc paths stopped resolving to Relaton data; but fallback to xml2rfc data still worked (which...

@ronaldtse The mapping tool (https://dev.bibxml.org/management/xml2rfc-compat/bibxml2/) was intended as the primary way for BibXML service operators to create and review automated and manual mappings.

@ronaldtse It’s trying to do that now. The goal is to let a human gradually inspect all available xml2rfc paths (to list which we are using `bibxml-data-archive` repository contents) throughout...

> I want to run a process that tells me if any mapping is invalid I can see how this could be addressed by adding a filter to mapping management...

Here’s some info on implementation, intended for service operators and maintainers: [xml2rfc-style path resolution](https://dev.bibxml.org/static/docs/topics/xml2rfc-compat.html#xml2rfc-style-path-resolution-algorithm)

Yes, we discussed using an OIDC library. The Django OIDC integration library maintained by Mozilla was judged to cost excess overhead for this project, but I’ll revisit that. Either way,...

Since you mentioned Datatracker login may be unnecessary, I’ll see if a settings flag can be implemented to control the behavior.

Automatic token refresh appears to work now (tested in staging), but I’ll leave this issue open until Datatracker auth is configurable too…