📌 Summary

Fixed a race condition in HdsAdvancedTable where re-rendering the table via updating the @model argumentcaused the internal state of selectable rows to be incorrectly wiped out.

🛠️ Detailed description

The Problem

When the table@model updates, it inserts the new DOM elements before destroying the old ones. This creates a momentary overlap where this._selectableRows contains two entries for the same row (the stale instance and the fresh instance) sharing the same selectionKey.

Previously, the willDestroyRowCheckbox method used .filter() to remove the row being destroyed. Because the old and new rows shared the same selectionKey during this overlap window, the filter logic aggressively removed both entries. This left the internal tracking array empty.

The Solution

Replaced the .filter() logic with a more targeted .findIndex() and .splice() approach.

Find: We locate the first index matching the selectionKey (which corresponds to the old row).

Splice: We remove exactly one item at that index.

The Results

This change ensures that during the lifecycle overlap, we only remove the stale DOM reference while keeping the newly inserted row intact. The cleanup now correctly leaves one valid entry remaining, preserving the integrity of the selection state.

🔗 External links

Jira ticket: HDS-5636

👀 Component checklist

• Percy was checked for any visual regression
• A changelog entry was added via Changesets if needed (see templates here)

💬 Please consider using conventional comments when reviewing this PR.