Bug report: issue with BNF to DM+D mappings or conversion

[StevenMaude]

Summary

It's possible that:

• importing the BNF to DM+D mappings is failing somewhere
• there's some issue with converting codelists.
• possibly more likely, the BNF data is outdated (see below)

What did you do and what did you expect to happen?

From Andrea on Slack to tech support:

Hello, I've created a series of codelists for BNF chapters (example for Chapter 1: BNF codelist) which I've also converted to dmd. However, when I look at the tree for the dmd codelists some items are excluded, which I'm pretty sure should not be. There is no way to manually add those codes to the dmd codelist, and I can't see how to "fix" the BNF codelist so they are included. Any help?

Impact of the bug

It's happening consistently, and preventing the DM+D codelist from matching the original BNF one.

[StevenMaude]

I'm not sure that it's that the mappings aren't imported.

First, there was an update posted to Slack on 2025-11-26 from this script on success. And that message should only fire if the import is successful.

It's possible there's some other condition where the import function fails, but still gives a 0 exit code, which might warrant investigation, but…

Second, locally running the import also worked, if I hacked the script considerably to:

• ensure the env variables were set and the virtualenv was active
• alter the local paths
• remove attempts to run dokku and print something instead
• remove calls to Sentry and Slack

So it's possible there's something else going on here than a simple failure of the existing machinery.

[StevenMaude]

Another thought: is this maybe related to #2435?

The BNF data is apparently currently from 2024:

Latest release: 86 (2024-09-02)

The latest BNF data is at release 88.

(I don't know enough about how all this works to know if that would cause the problem, but it's a thought. But Becky knows more about this and suggested it could be.)

[lucyb]

I agree, I think it's worth doing #2435 at this point. Thanks for looking into it.

[lucyb]

I'm going to take this off the board for now, given that we have #2435 in Next.