Robert Hunt (432ae66d) at 18 Mar 16:38

Robert Hunt (4f86a3f6) at 18 Mar 16:37

... and 1 more commit

What does this MR do and why?

Improve parser by allowing any bare word ( and underscore ) to be recognised as valid token:

1. Moves validation to the right layer — the analyzer knows field semantics, the parser shouldn't need to
2. Eliminates manual token maintenance — no more editing the allowlist for new sources/values
3. Enables EnumLike everywhere — sources can use EnumLike instead of StringEnumLike for enum fields. Previously words with special chars e.g. waiting_for_something would need to be a string

#112

How to set up and validate locally

NA

What does this MR do and why?

Improve parser by allowing any bare word ( and underscore ) to be recognised as valid token:

1. Moves validation to the right layer — the analyzer knows field semantics, the parser shouldn't need to
2. Eliminates manual token maintenance — no more editing the allowlist for new sources/values
3. Enables EnumLike everywhere — sources can use EnumLike instead of StringEnumLike for enum fields. Previously words with special chars e.g. waiting_for_something would need to be a string

#112

How to set up and validate locally

NA

question: @drosse the MR mentions:

Enables EnumLike everywhere — sources can use EnumLike instead of StringEnumLike for enum fields. Previously words with special chars e.g. waiting_for_something would need to be a string

But we don't action on it in this MR. I just wanted to confirm if this was missing, or it's more an FYI?

Robert Hunt (b9c10f68) at 18 Mar 15:55

... and 5 more commits

Robert Hunt (175c6535) at 18 Mar 15:44

What does this MR do and why?

• Adds Duo Code Suggestions aggregation engine to the GraphQL API.
• Renames the AiUsage GraphQL type to Analytics for a more generic naming. Updates all references across group/project types, resolvers, spec files, documentation, and the GraphQL introspection result. No functional changes — purely a rename/refactor

References

Screenshots or screen recordings

How to set up and validate locally

1. Setup ClickHouse in GDK
2. Run db seeds with FILTER=ai_usage_stats bundle exec rake db:seed_fu
3. Execute a query similar to the one shown in the screenshot above.
4. Validate output.

MR acceptance checklist

Evaluate this MR against the MR acceptance checklist. It helps you analyze changes to reduce risks in quality, performance, reliability, security, and maintainability.

@pshutsin I pulled this branch down to my GDK, generated a GLQL gem via the GLQL branch I'm currently working on, and verified that GLQL generates the right queries for this endpoint, and that the response is expected 🎉.

For transparency, I used test_queries.rb to run some basic tests. You can pretty much ignore this though, as it's specific to the GQL use-case 🙂.

Comparing this against CodeSuggestion: end-to-end GLQL aggregation int... (gitlab-org#21212), I can confirm that the endpoint is matching the expected I/O 👍

question (non-blocking): @pshutsin the examples given are all group-level only. I just want to confirm that we expect duoCodeSuggestions to return results for project-level queries too?

Robert Hunt (57251876) at 18 Mar 14:50

Robert Hunt (3379b2ca) at 18 Mar 14:35

@pshutsin specific to code suggestions, we're hoping to have this type released in %18.11 as soon as we've validated end-to-end. The existing dashboards won't be using the new endpoint yet though, so we need to maintain the existing endpoints.

Robert Hunt (aa9ee4b0) at 18 Mar 11:56

Robert Hunt (3be9d601) at 18 Mar 11:25

Robert Hunt (71605ec3) at 18 Mar 11:19

Robert Hunt (2da95ad5) at 18 Mar 09:20

Robert Hunt (2a0922f7) at 18 Mar 09:18

... and 22 more commits

This is fine

Robert Hunt (46c1904c) at 17 Mar 23:28

... and 3 more commits