Further explanation of second part of the ananke query

[LynxMaxim]

Hello,

I'm not very certain about what the purpose is of the second part of the .sql file (where you indicate "Improvement of <vocabulary_id>") and I would love to understand it better! :)

To reiterate (and as far as I understand) the following happens:

1. The first part of the SQL file generates a temp table based on the matching of UMLS atom codes and ATHENA concept codes. Here a filter is set on: language to be English and on vocabularies (to ensure matching).
2. In the second part you look for concept_ids which do not have a match in the generated table from set 1. These are then the ones which did not have an English atom, or do I see this wrong? --> But then, why not remove the language filter in the first part?

I would be very grateful if you could explain a bit how the second part works.
Thanks a lot!