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
Release v2.20.0 includes PR #2600 which redraws every table column slot on any changes to the table source data. This introduces significant performance impact on large tables and makes it difficult for users to interact with inputs within tables since the redraw will lose cursor position. This also makes it impossible to tab between columns when data changes.
I intend to submit a PR which reverts the changes made in #2600.
Additional context
I demonstrate in #2004 (comment) that the issue reported which prompted PR #2006 was a misunderstanding and was not related to table behavior at all. Therefore, that PR is unnecessary and does not directly address the issue, yet it introduces unintended side effects.
The text was updated successfully, but these errors were encountered:
Environment
Version
v2.20.0 and later
Reproduction
I could not get Stackblitz to work when changing Nuxt UI versions due to internal 503 errors. Please forgive me for using codesandbox instead.
Nuxt UI v2.19.1 - Working as expected. You can edit and tab between inputs.
https://codesandbox.io/p/devbox/nuxt-v2-19-1-lz5zvk
Nuxt UI v2.20.0 - Cursor position lost when typing. Second column (email) uses
.lazy
which allows typing but loses focus when tabbing after edit.https://codesandbox.io/p/devbox/nuxt-v2-20-0-x9xylj
Description
Release v2.20.0 includes PR #2600 which redraws every table column slot on any changes to the table source data. This introduces significant performance impact on large tables and makes it difficult for users to interact with inputs within tables since the redraw will lose cursor position. This also makes it impossible to tab between columns when data changes.
I intend to submit a PR which reverts the changes made in #2600.
Additional context
I demonstrate in #2004 (comment) that the issue reported which prompted PR #2006 was a misunderstanding and was not related to table behavior at all. Therefore, that PR is unnecessary and does not directly address the issue, yet it introduces unintended side effects.
The text was updated successfully, but these errors were encountered: