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
[$250] Expense - Attendee row appears after merging from Reports when disabled in Rules #83165
Select the expense and second option in all section and continue
Note in confirmation page, "attendee" row is not shown
Go to reports - expense
Long press & select same expense (step 5) and merge selecting second option in all section and continue
Note in confirmation page, "attendee" row is shown
Go to inbox
Open workspace chat and open same expense used in step 5
Tap more - merge selecting expense & second option in all section and continue
Note in confirmation page, "attendee" row is shown
Expected Result:
When Attendees section is disabled in Rules:
Attendee row should NOT be displayed across all merge entry points [Expense screen
/ Reports list (long press) ].
Behavior must be consistent regardless of navigation path.
Actual Result:
Initial merge from workspace, attendee row is not shown correctly.
When merge from report, attendee row appears incorrectly.
After that, workspace merge also shows attendee row incorrectly.
👋 Friendly reminder that deploy blockers are time-sensitive ⏱ issues! Check out the open `StagingDeployCash` deploy checklist to see the list of PRs included in this release, then work quickly to do one of the following:
Identify the pull request that introduced this issue and revert it.
Classification: Frontend bug Causing PR: #71244 - "[Merge] Merging tax rates" by @dominictb (High confidence) Related Issues: None
Recommendation: REVERT
The PR introduced a regression in the merge transaction flow where the policy is incorrectly determined when merging from certain entry points (reports list, search), causing the attendee row to appear even when attendee tracking is disabled in the workspace rules.
Labels: Keeping DeployBlockerCash (Frontend bug - App PR needs to be reverted)
📋 Detailed Analysis
Evidence
PR #71244 added support for merging tax rates in expense merge transactions. As part of this change, it modified how policies are passed to the MoneyRequestView component during the merge confirmation step, changing from a generic policy to targetTransactionPolicy which is determined by the useMergeTransactions hook.
The issue is reproducible only on staging (where PR #71244 is deployed) and not on production, and the PR is directly mentioned in the issue report by the automated melvin-bot comment.
Verification
The bug manifests in the merge transaction confirmation page behavior:
Confirmation: PR [Merge] Merging tax rates #71244 modified this file to add targetTransactionPolicy and sourceTransactionPolicy logic
Root Cause
In src/hooks/useMergeTransactions.ts, lines 78-80, the code overwrites the policy from Onyx when in a search context:
// If we're on search, search snapshot policies are more up to datetargetTransactionPolicy=currentSearchResults?.data[`${ONYXKEYS.COLLECTION.POLICY}${targetTransactionPolicyID}`];sourceTransactionPolicy=currentSearchResults?.data[`${ONYXKEYS.COLLECTION.POLICY}${targetTransactionPolicyID}`];
This differs from how reports are handled (lines 76-77), which use the nullish coalescing operator ?? to fall back to Onyx data:
When merging from the reports list (which uses search context), if the policy is not present in currentSearchResults.data, the policy becomes undefined.
The shouldShowAttendees function in src/libs/TransactionUtils/index.ts (line 581) returns true by default when the policy is undefined:
returnpolicy?.isAttendeeTrackingEnabled??true;
This causes the attendee row to incorrectly appear even when attendee tracking is disabled in the workspace rules.
Bug flow:
Merge from expense screen (not search context) → policy loaded from Onyx → correct behavior
Merge from reports list (search context) → policy overwritten with undefined → attendee row appears incorrectly
Subsequent merges from any entry point continue showing attendee row due to persisted state
I submitted a proposal for this issue after the Help Wanted label was added. My proposal was reviewed, and both @hoangzinh and @mananjadhav confirmed that the identified root cause and proposed solution were correct.
I understand from the contribution guidelines that “If you propose solutions to jobs before the Help Wanted label is applied, you do so at your own risk.” In this case, however, my proposal was made after the label had already been added.
For that reason, it was surprising to see the issue later assigned to someone else. Since I invested time investigating and preparing the solution under the assumption that the issue was open for contribution, this situation doesn’t seem entirely fair.
Could you please clarify the reasoning behind the reassignment and how I should proceed? I would appreciate your guidance.
Activity
melvin-bot commentedon Feb 22, 2026
@youssef-lr recently merged #71244, but they are OOO right now, so I'll auto-assign someone else.
melvin-bot commentedon Feb 22, 2026
Triggered auto assignment to @greg-schroeder (
Bug), see https://stackoverflow.com/c/expensify/questions/14418 for more details. Please add this bug to a GH project, as outlined in the SO.melvin-bot commentedon Feb 22, 2026
Triggered auto assignment to @MonilBhavsar (
DeployBlockerCash), see https://stackoverflowteams.com/c/expensify/questions/9980/ for more details.melvin-bot commentedon Feb 22, 2026
💬 A slack conversation has been started in #expensify-open-source
github-actions commentedon Feb 22, 2026
👋 Friendly reminder that deploy blockers are time-sensitive ⏱ issues! Check out the open `StagingDeployCash` deploy checklist to see the list of PRs included in this release, then work quickly to do one of the following:
OSBotify commentedon Feb 22, 2026
🔍 Investigation Summary
Classification: Frontend bug
Causing PR: #71244 - "[Merge] Merging tax rates" by @dominictb (High confidence)
Related Issues: None
Recommendation: REVERT
The PR introduced a regression in the merge transaction flow where the policy is incorrectly determined when merging from certain entry points (reports list, search), causing the attendee row to appear even when attendee tracking is disabled in the workspace rules.
Labels: Keeping
DeployBlockerCash(Frontend bug - App PR needs to be reverted)📋 Detailed Analysis
Evidence
PR #71244 added support for merging tax rates in expense merge transactions. As part of this change, it modified how policies are passed to the
MoneyRequestViewcomponent during the merge confirmation step, changing from a genericpolicytotargetTransactionPolicywhich is determined by theuseMergeTransactionshook.The issue is reproducible only on staging (where PR #71244 is deployed) and not on production, and the PR is directly mentioned in the issue report by the automated melvin-bot comment.
Verification
The bug manifests in the merge transaction confirmation page behavior:
src/hooks/useMergeTransactions.ts(lines 78-80)targetTransactionPolicyandsourceTransactionPolicylogicRoot Cause
In
src/hooks/useMergeTransactions.ts, lines 78-80, the code overwrites the policy from Onyx when in a search context:This differs from how reports are handled (lines 76-77), which use the nullish coalescing operator
??to fall back to Onyx data:When merging from the reports list (which uses search context), if the policy is not present in
currentSearchResults.data, the policy becomesundefined.The
shouldShowAttendeesfunction insrc/libs/TransactionUtils/index.ts(line 581) returnstrueby default when the policy isundefined:This causes the attendee row to incorrectly appear even when attendee tracking is disabled in the workspace rules.
Bug flow:
undefined→ attendee row appears incorrectly31 remaining items
fix: attendee row appears when merging in Reports even when disabled E…
marufsharifi commentedon Mar 1, 2026
@youssef-lr
I submitted a proposal for this issue after the Help Wanted label was added. My proposal was reviewed, and both @hoangzinh and @mananjadhav confirmed that the identified root cause and proposed solution were correct.
I understand from the contribution guidelines that “If you propose solutions to jobs before the Help Wanted label is applied, you do so at your own risk.” In this case, however, my proposal was made after the label had already been added.
For that reason, it was surprising to see the issue later assigned to someone else. Since I invested time investigating and preparing the solution under the assumption that the issue was open for contribution, this situation doesn’t seem entirely fair.
Could you please clarify the reasoning behind the reassignment and how I should proceed? I would appreciate your guidance.
Thank you for your time and consideration.
cc @greg-schroeder
hoangzinh commentedon Mar 2, 2026
Hi @marufsharifi. I appreciate your effort on this issue. Let me clarify this part:
That’s why I suggested assigning this issue to me and @dominictb.
melvin-bot commentedon Mar 13, 2026
If you are the assigned CME please investigate whether the linked PR caused a regression and leave a comment with the results.
If a regression has occurred and you are the assigned CM follow the instructions here.
If this regression could have been avoided please consider also proposing a recommendation to the PR checklist so that we can avoid it in the future.
melvin-bot commentedon Mar 13, 2026
If you are the assigned CME please investigate whether the linked PR caused a regression and leave a comment with the results.
If a regression has occurred and you are the assigned CM follow the instructions here.
If this regression could have been avoided please consider also proposing a recommendation to the PR checklist so that we can avoid it in the future.
melvin-bot commentedon Mar 13, 2026
If you are the assigned CME please investigate whether the linked PR caused a regression and leave a comment with the results.
If a regression has occurred and you are the assigned CM follow the instructions here.
If this regression could have been avoided please consider also proposing a recommendation to the PR checklist so that we can avoid it in the future.
melvin-bot commentedon Mar 13, 2026
If you are the assigned CME please investigate whether the linked PR caused a regression and leave a comment with the results.
If a regression has occurred and you are the assigned CM follow the instructions here.
If this regression could have been avoided please consider also proposing a recommendation to the PR checklist so that we can avoid it in the future.
mananjadhav commentedon Mar 15, 2026
@youssef-lr Can you unassign me from this issue?
dominictb commentedon Mar 24, 2026
PR reach prod