You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
[Due for payment 2026-06-25] Search selection: move write-side off onto provider/controller #93677
The Search screen owned the entire selection write-path inline inside a ~800-line component (toggle row/group/all, reconcile stale keys, FlashList extraData, mobile selection mode). Every checkbox press re-rendered the whole screen subtree (SearchPage → Search → list), causing 8–15 row re-renders per toggle.
Solution
Moved selection write-side out of Search into single-responsibility units on dedicated contexts: pure selectionBuilders.ts, state-only SearchSelectionProvider with applySelection, new SearchWriteActionsProvider for toggle/reconcile effects, and self-subscribing SearchSelectionFooter. Rows and list pull actions from context; selectedTransactions removed from FlashList extraData. Checkbox presses now re-render only affected row(s) and the footer.
changed the title [-]Search selection: move write-side off onto provider/controller[/-][+][Due for payment 2026-06-25] Search selection: move write-side off onto provider/controller[/+]on Jun 18, 2026
The solution for this issue has been 🚀 deployed to production 🚀 in version 9.4.14-1 and is now subject to a 7-day regression period 📆. Here is the list of pull requests that resolve this issue:
[Contributor] The offending PR and associated issue have been commented on, pointing out the bug it caused and why, so the author and reviewers can learn from the mistake.
It's a refactoring issue
[Contributor] If the regression was CRITICAL (e.g. interrupts a core flow) A discussion in #expensify-open-source has been started about whether any other steps should be taken (e.g. updating the PR review checklist) in order to catch this type of bug sooner.
Link to discussion: NA
[Contributor] If it was decided to create a regression test for the bug, please propose the regression test steps using the template below to ensure the same bug will not reach production again.
Regression Test Proposal
Test:
Open Search (Reports / Expenses).
Tap a single expense row's checkbox — verify it selects, the bulk-action footer appears with the correct count/total, and only that row + footer update (no full-list flicker).
Tap a report/group checkbox — verify all of its transactions select together.
Tap "Select all" — verify every selectable row selects and the footer reflects the matching-items total; untoggle one row and verify "select all" turns off.
With a selection active, trigger a data change (e.g. let an Onyx update add/remove a row) — verify the selection reconciles and stale keys are dropped, with no console "Maximum update depth" errors.
On narrow/mobile layout: long-press a row to enter selection mode, select/deselect down to zero — verify selection mode turns off automatically.
Verify bulk actions (hold/unhold, approve, pay, change report, etc.) still operate on the selected set.
Activity
[-]Search selection: move write-side off onto provider/controller[/-][+][Due for payment 2026-06-25] Search selection: move write-side off onto provider/controller[/+]melvin-bot commentedon Jun 18, 2026
Reviewinglabel has been removed, please complete the "BugZero Checklist".melvin-bot commentedon Jun 18, 2026
The solution for this issue has been 🚀 deployed to production 🚀 in version 9.4.14-1 and is now subject to a 7-day regression period 📆. Here is the list of pull requests that resolve this issue:
If no regressions arise, payment will be issued on 2026-06-25. 🎊
For reference, here are some details about the assignees on this issue:
ZhenjaHorbach commentedon Jun 23, 2026
BugZero Checklist:
It's a refactoring issue
[Contributor] If the regression was CRITICAL (e.g. interrupts a core flow) A discussion in #expensify-open-source has been started about whether any other steps should be taken (e.g. updating the PR review checklist) in order to catch this type of bug sooner.
Link to discussion: NA
[Contributor] If it was decided to create a regression test for the bug, please propose the regression test steps using the template below to ensure the same bug will not reach production again.
Regression Test Proposal
Test:
Do we agree 👍 or 👎
Zapier Logs
Run ID: 00040eee-248b-a88e-3ffe-3740e4524844melvin-bot commentedon Jun 23, 2026
Triggered auto assignment to @mallenexpensify (
Awaiting Payment)melvin-bot commentedon Jun 24, 2026
Payment Summary
Upwork Job
BugZero Checklist (@mallenexpensify)
mallenexpensify commentedon Jun 27, 2026
Payment Summary
Contributor+: @ZhenjaHorbach due $250 via NewDot
^ Test case created. Thx
garrettmknight commentedon Jun 29, 2026
$250 approved for @ZhenjaHorbach