Reduce re-derivation of body rows and columns

[bryanmylee]

The original strategy had plugins expose a row transformation function in a Readable, to which useTable then subscribed to transform originalRows into the final set of rows to display.

const rows = derived([originalRows, ...transforms], ...);

However, this meant that any updates to plugin state would cause the entire derivation to recompute from originalRows to rows.

We can improve plugin performance – especially if plugin state being updated is at the end of the transformations – by creating a chain of derived stores.

The goal is to only recompute rows from the point of a plugin that updated and onwards.

originalRows
  .map(sort)      <- do not update this
  .map(filter)    <- plugin state changes
  .map(paginate)  <- only update this

To do so, each plugin receives a Readable<BodyRow[]> store which it uses to derive a transformed Readable<BodyRow[]> store. Plugins can then be chained together by passing the result of one plugin to the next. By creating this chain, the only dependencies of a plugin's derived rows are its own plugin state and the previous plugin's derived rows, minimizing the re-computation required.

const useSortBy = () => {
  ...
  return {
    deriveRows: (rows: Readable<BodyRow[]>) => derived([rows, state], ...),
  }
}

const useTable = () => {
  ...
  let rows = originalRows;
  deriveRowsFns.forEach(fn => {
    rows = fn(rows);
  });
  ...
}

[coveralls]

Pull Request Test Coverage Report for Build 2315042871

Warning: This coverage report may be inaccurate.

This pull request's base commit is no longer the HEAD commit of its target branch. This means it includes changes from outside the original pull request, including, potentially, unrelated coverage changes.

• For more information on this, see Tracking coverage changes with pull request builds.
• To avoid this issue with future PRs, see these Recommended CI Configurations.
• For a quick fix, rebase this PR at GitHub. Your next report should be accurate.

Details

• 0 of 0 changed or added relevant lines in 0 files are covered.
• No unchanged relevant lines lost coverage.
• Overall coverage remained the same at 81.529%

---

Totals
Change from base Build 2311762508:	0.0%
Covered Lines:	217
Relevant Lines:	260

---

💛 - Coveralls