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
This feels related to the discussion we were having in #1649 on channel values supplied as arrays of values, and transforms breaking that association. I’m guessing that the isTextual heuristic is wrong here (related #1646): the heuristic is being applied to the mark’s raw data, but that data is subsequently transformed (by the group transform), and so maybe the heuristic shouldn’t apply?
The text was updated successfully, but these errors were encountered:
I think the fix is pretty simple—we just need to opt-out of the additional defaults when constructing the derived tip mark. We never want to compute “extra” channels when deriving a tip.
Ref. https://observablehq.com/d/42b72061ec8200e4
Screen.Recording.2023-05-30.at.11.59.56.AM.mov
It appears to have something to do with the default identity title.
If you change the definition to avoid this behavior, the tip shows the correct value:
This feels related to the discussion we were having in #1649 on channel values supplied as arrays of values, and transforms breaking that association. I’m guessing that the isTextual heuristic is wrong here (related #1646): the heuristic is being applied to the mark’s raw data, but that data is subsequently transformed (by the group transform), and so maybe the heuristic shouldn’t apply?
The text was updated successfully, but these errors were encountered: