Refactor index_data.painless script and params

[ellisandrews-toast]

Summary

This is a preparatory refactoring of the index_data Painless update script to make it ready for nested sourced_from support. No behavior change.

• Restructure the flat script into helper functions
• Rename params.data → params.topLevelFields to clarify intent (since putAll only applies top-level fields)

Script structure after refactoring

setup(...)                // Initializes bookkeeping structures (__sources, __versions, __counts)
validateSource(...)       // Checks for relationship mutation and version staleness (throws if stale)
applyTopLevelFields(...)  // Does putAll for top-level fields and merges __counts
recordSource(...)         // Sets id, records version, maintains __sources sorted list

Why rename params.data → params.topLevelFields?

When nested sourced_from support is added, the script will also receive params.nestedFields (data that gets stored in __nested_sourced_data and applied to nested elements). Renaming data to topLevelFields now makes the distinction clear and avoids a confusing rename later when both params coexist.

[myronmarston]

Looking good! Left some minor suggestions.

[myronmarston]

On the surface, it looks like there may be a risk of a null pointer exception here--previously we did map?.get(params.sourceId) which would evaluate to null if map was null; nwo we do map.get(sourceId) which will throw an NPE if map is null.

But looking at the setup function, it looks like we initialize source.__versions[relationship] if it's not already set, so I'm thinking it's impossible for the map to be null here.

Can you confirm that analysis matches your own?

[myronmarston]

Setting source.id doesn't feel like it should be part of recordSource. Can we move this into applyTopLevelFields? After all, source.id = id is setting the top-level id field.

[myronmarston]

Worth renaming data = to top_level_fields =?

[myronmarston]

Suggested change

	without_id_or_data = params.except("id", "topLevelFields")
	without_id_or_top_level_fields = params.except("id", "topLevelFields")

(And you'll need to update the reference to it below).

[myronmarston]

Worth renaming data = to top_level_fields = ?