feat(nanoviews): add as_ to hand a row through a transform

[dangreen]

A row that needs a transform cannot be an expression. From this repo's own weather example:

for_(props.$suggestions, trackBy('label'))(
  ($city, index) => {
    const city = record($city)

    return li({ role: 'presentation' })(
      // …
    )
  }
)

Four lines of scaffolding — the braces, the const, the blank line, the return — around a body that is one expression. as_ gives the expression back:

for_(props.$suggestions, trackBy('label'))(
  as_(record, (city, index) => li({ role: 'presentation' })(
    // …
  ))
)

The whole thing

/* @__NO_SIDE_EFFECTS__ */
export function as_<Item, Index, Value>(
  as: (item: Item) => Value,
  each_: (value: Value, index: Index) => Child
) {
  return (item: Item, index: Index) => each_(as(item), index)
}

The transform is any function of the row, not record specifically. The row and index types are deliberately unconstrained, so one signature serves all three shapes for_ hands out — the writable row, the readable one, and the static one.

The thing that decided this was worth having is that inference survives. record is itself generic, so the row type has to flow through two generic calls in a row before it reaches the callback. It does, with no type arguments written anywhere:

for_($cities, trackById)(
  as_(record, (city, index) => li()(city.$name, ':', index))
)

city.$name is WritableSignal<string> there — checked by giving it to a number and reading the error, not by looking at it.

Tests

Two, both rendering for real: a row goes through record, the index arrives second, and a change to the array repaints; and — the one that matters — a row transformed by as_ is still the row, so normalising $player.$name inside the body writes back into the items array.

Size

+17 B gzip on all publics, and nothing on the average-usage bundle: as_ is a separate export, so a consumer who does not import it does not pay for it. One brotli pin moves up a step; the rest stay.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 85.33%. Comparing base (215f157) to head (8f98203).

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #207      +/-   ##
==========================================
+ Coverage   85.29%   85.33%   +0.03%     
==========================================
  Files         139      139              
  Lines        3142     3143       +1     
  Branches      591      591              
==========================================
+ Hits         2680     2682       +2     
  Misses        332      332              
+ Partials      130      129       -1     

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.