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
We have a client app that's setting a selected record based on a "last viewed" preference / route and ensuring that gets set on the relevant Grid / GridModel via an autorun call to gridModel.selModel.select().
The app code has long had a delay: 1 opt on the auto-run "to account for grid rendering." Updated to latest hoist-react 11.0.0-SNAPSHOT and now that code is throwing within Grid.selectionReaction() as the node is not found. As a result, the row is not visible selected within the grid, and an error ends up on the console.
Trial and error showed that upping the delay to e.g. 300 works OK, but 100 or 200 does not - i.e. this is actually a time-based thing and not simply a matter of pushing to the end of the event loop.
Ideally we would not need any delays at all in client app code, and definitely want to avoid any fuzzy delays based on variable rendering performance. Curious though as to what changed - and then happy to help brainstorm ideas to make more reliable all around at the Hoist level.
The text was updated successfully, but these errors were encountered:
We have a client app that's setting a selected record based on a "last viewed" preference / route and ensuring that gets set on the relevant Grid / GridModel via an autorun call to
gridModel.selModel.select()
.The app code has long had a
delay: 1
opt on the auto-run "to account for grid rendering." Updated to latest hoist-react 11.0.0-SNAPSHOT and now that code is throwing withinGrid.selectionReaction()
as the node is not found. As a result, the row is not visible selected within the grid, and an error ends up on the console.Trial and error showed that upping the delay to e.g. 300 works OK, but 100 or 200 does not - i.e. this is actually a time-based thing and not simply a matter of pushing to the end of the event loop.
Ideally we would not need any delays at all in client app code, and definitely want to avoid any fuzzy delays based on variable rendering performance. Curious though as to what changed - and then happy to help brainstorm ideas to make more reliable all around at the Hoist level.
The text was updated successfully, but these errors were encountered: