perf: drop un-flatten-tree dep & optimize Tree Data utils by ~3x perf

[ghiscoding]

vibe coded with copilot using Claude Sonnet 4.6

Summary

Removes the un-flatten-tree third-party dependency from @slickgrid-universal/common and rewrites or optimizes all tree data utility functions for significantly better performance, particularly at scale (50k–250k nodes). This makes the Tree Data to parse about ~3x faster when converting from flat to tree and vice versa (flatten/unflatten) which is how Tree Data was implemented in SlickGrid (when using Tree Data in SlickGrid we always keep a flat data copy and its equivalent tree data).

---

Changes

package.json

• Removed "un-flatten-tree" from dependencies

utilities.ts

flattenToParentChildArray (previously used un-flatten-tree)

• Replaced the library call with a custom O(n) iterative DFS using two parallel stacks (nodeStack + parentStack) — avoids tuple allocation per node
• Uses a shallow copy per node ({ ...node } then delete flatNode[childrenPropName]) to preserve the original hierarchical dataset intact (shared object references between flat and hierarchical arrays require this)
• ~3x faster than the previous implementation at 250k nodes (~35ms vs ~100ms)

findItemInTreeStructure (previously recursive)

• Replaced with an O(n) iterative DFS using a stack — eliminates call-stack depth risk on deep trees and is faster than the two-pass recursive approach

unflattenParentChildArrayToTree

• Replaced the plain object lookup with a Map — avoids prototype-chain overhead and Object.keys() allocation
• Replaced forEach with indexed for loops

addTreeLevelByMutation

• Replaced getTreeDataOptionPropName() calls with direct options.childrenPropName access
• Replaced forEach with indexed for loop; cached children to a local variable (single Array.isArray check)

addTreeLevelAndAggregatorsByMutation

• Same optimizations as above; also caches isParent to avoid double Array.isArray check

treeData.service.ts

• Added 5 cached instance fields (_childrenPropName, _collapsedPropName, _hasChildrenPropName, _lazyLoadingPropName, _levelPropName) initialized once in init() — eliminates ~20 repeated getTreeDataOptionPropName() call sites across the service, each of which previously chained two getter calls plus a switch statement
• Removed all per-call getTreeDataOptionPropName() lookups from: handleOnCellClick, handleOnKeyDown, toggleTreeDataCollapse, updateToggledItem, applyToggledItemStateChanges, initHierarchicalTree, recalculateTreeTotals, getItemCount
• toggleAllFlatDataItems: inlined the updateToggledItem(item, collapsing, false) body directly into the loop — eliminates one function call + two getTreeDataOptionPropName lookups per item during full collapse/expand of large datasets

utilities.spec.ts

• Updated 4 unflattenParentChildArrayToTree test expectations to match Map insertion order (previously relied on plain-object auto-sorting of integer keys)
• Updated 3 flattenToParentChildArray tests to reflect that children property is no longer present on flat items (was previously leaking through)
• Added 7 new unit tests: 3 for addTreeLevelByMutation, 2 for addTreeLevelAndAggregatorsByMutation, 2 for findItemInTreeStructure

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 100.0%. Comparing base (9dcc1ff) to head (570bb6b).
⚠️ Report is 1 commits behind head on master.

Additional details and impacted files

@@           Coverage Diff           @@
##           master    #2601   +/-   ##
=======================================
  Coverage   100.0%   100.0%           
=======================================
  Files         199      199           
  Lines       25168    25182   +14     
  Branches     8908     8909    +1     
=======================================
+ Hits        25168    25182   +14     

Flag	Coverage Δ
angular	100.0% <ø> (ø)
universal	100.0% <100.0%> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ 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.

[pkg-pr-new]

angular-slickgrid

npm i https://pkg.pr.new/angular-slickgrid@2601

aurelia-slickgrid

npm i https://pkg.pr.new/aurelia-slickgrid@2601

slickgrid-react

npm i https://pkg.pr.new/slickgrid-react@2601

slickgrid-vue

npm i https://pkg.pr.new/slickgrid-vue@2601

@slickgrid-universal/angular-row-detail-plugin

npm i https://pkg.pr.new/@slickgrid-universal/angular-row-detail-plugin@2601

@slickgrid-universal/aurelia-row-detail-plugin

npm i https://pkg.pr.new/@slickgrid-universal/aurelia-row-detail-plugin@2601

@slickgrid-universal/react-row-detail-plugin

npm i https://pkg.pr.new/@slickgrid-universal/react-row-detail-plugin@2601

@slickgrid-universal/vue-row-detail-plugin

npm i https://pkg.pr.new/@slickgrid-universal/vue-row-detail-plugin@2601

@slickgrid-universal/binding

npm i https://pkg.pr.new/@slickgrid-universal/binding@2601

@slickgrid-universal/common

npm i https://pkg.pr.new/@slickgrid-universal/common@2601

@slickgrid-universal/composite-editor-component

npm i https://pkg.pr.new/@slickgrid-universal/composite-editor-component@2601

@slickgrid-universal/custom-footer-component

npm i https://pkg.pr.new/@slickgrid-universal/custom-footer-component@2601

@slickgrid-universal/custom-tooltip-plugin

npm i https://pkg.pr.new/@slickgrid-universal/custom-tooltip-plugin@2601

@slickgrid-universal/empty-warning-component

npm i https://pkg.pr.new/@slickgrid-universal/empty-warning-component@2601

@slickgrid-universal/event-pub-sub

npm i https://pkg.pr.new/@slickgrid-universal/event-pub-sub@2601

@slickgrid-universal/excel-export

npm i https://pkg.pr.new/@slickgrid-universal/excel-export@2601

@slickgrid-universal/graphql

npm i https://pkg.pr.new/@slickgrid-universal/graphql@2601

@slickgrid-universal/odata

npm i https://pkg.pr.new/@slickgrid-universal/odata@2601

@slickgrid-universal/pagination-component

npm i https://pkg.pr.new/@slickgrid-universal/pagination-component@2601

@slickgrid-universal/pdf-export

npm i https://pkg.pr.new/@slickgrid-universal/pdf-export@2601

@slickgrid-universal/row-detail-view-plugin

npm i https://pkg.pr.new/@slickgrid-universal/row-detail-view-plugin@2601

@slickgrid-universal/rxjs-observable

npm i https://pkg.pr.new/@slickgrid-universal/rxjs-observable@2601

@slickgrid-universal/sql

npm i https://pkg.pr.new/@slickgrid-universal/sql@2601

@slickgrid-universal/text-export

npm i https://pkg.pr.new/@slickgrid-universal/text-export@2601

@slickgrid-universal/utils

npm i https://pkg.pr.new/@slickgrid-universal/utils@2601

@slickgrid-universal/vanilla-bundle

npm i https://pkg.pr.new/@slickgrid-universal/vanilla-bundle@2601

@slickgrid-universal/vanilla-force-bundle

npm i https://pkg.pr.new/@slickgrid-universal/vanilla-force-bundle@2601

commit: 570bb6b

[ghiscoding]

@zewa666 another win for the AI usage... by looking at the dependency "un-flatten-tree", I found this benchmark created by a user a few years ago: https://github.com/philipstanislaus/array-to-tree-benchmarks

then after asking the AI if I could switch to the more performant lib, Claude AI actually identified that we could use native code (aka drop dependency) and improve the perf by about ~3x (or 300%) with O(n) instead of O(n2)... wow this is just awesome... go ahead Claude 🚀 That's also great because that was 1 of the last package that wasn't fully ESM, which is why it was on my list of TODOs eventually.

my company increased our plan to Premium for a month because of all the token outcry, many employees got 75% token usage on day 1, so after a few days they bumped our account to Premium but it's just for a month, so better take advantage of it while it last 😆 .... EDIT: ah wait, they changed their mind and reverted to regular access and now I'm out of credit, it was fun while it lasted lol

---

also just let me know when you're done reviewing the other 2 PRs