Open jthomasmock opened 4 months ago
Thanks @smach for the suggestion!
Especially for numeric-esque columns we have aspirations to build a fast "between" brushed filter in the summary panel, similar to RStudio's filter tool but largely focused in the Summary Panel to allow quick vertical scrolling across many columns.
I think we can definitely refine the interactions for filtering to make it faster and easier to quickly add filters. Having better keyboard-driven interactions (so you can drive populating the filter modal quickly via the keyboard without clicking) would be another direction to invest some owrk
Given the feedback on UI title:
Would it be possible to implement a sort of variable selection in the data explorer? Let me explain better: When viewing tables with many variables, it would be very useful to have the ability to click on the variables that you want to show/hide in the menu with the list where the percentage of missing values is also indicated. This, in addition to simpler filters to apply, would reduce the need to create an intermediate dataset just for the visualization of variables that are not close to each other
Would it be possible to implement a sort of variable selection in the data explorer? Let me explain better: When viewing tables with many variables, it would be very useful to have the ability to click on the variables that you want to show/hide in the menu with the list where the percentage of missing values is also indicated.
We can discuss in a spin-off issue, but we have indeed been planning to add column filters (search-and-select) as well as column pinning (so you can select a particular column to pin to the left of the data explore). See https://github.com/posit-dev/positron/issues/2372 for a top-level discussion of some thoughts that we've had about this. See #2580 for an issue about column pinning.
Very similar feedback in https://github.com/posit-dev/positron/discussions/4334.
One more thought before the private beta winds down: Opinion only, and others may differ, but I prefer the RStudio approach to filtering data in the viewer pane. I realize the Positron approach is more like what Excel uses and will be familiar to everyone. But if you are filtering multiple columns, it is so much easier to click once and get multiple boxes to fill in than to have to click twice for each column. Plus in RStudio it's easier to see and modify the filters.
Originally posted by @smach in https://github.com/posit-dev/positron-beta/discussions/234