-
Reproduce the issue:
at this point the graph will only show 1 of the lines, and going into edit wont let you choose multi lines again either ( seems to be in an unrecoverable state at this point ) sample graph: |
Beta Was this translation helpful? Give feedback.
Replies: 3 comments 1 reply
-
Use There is a bug here in import pandas
x = pandas.DataFrame( { "a" : list(range(10)), "b" : list(x*x for x in range(10))})
perspective.PerspectiveWidget( x, plugin='d3_y_line', columns=["a", "b", "index"]) |
Beta Was this translation helpful? Give feedback.
-
I see, ctrl+shift does seem to let me add them back in, but its odd to see it drop them all by switching to datagrid and then back ( also, I may be missing it, but why the need for ctrl+?) |
Beta Was this translation helpful? Give feedback.
-
I suspect this would be more obvious on a non-trivial example with more than 3 numeric columns, e.g. Superstore example has 21 columns, only 5 of which are numeric and only 2 of those share a logic axis. Perspective will not change arguments supplied explicitly as
|
Beta Was this translation helpful? Give feedback.
Use
shift+click
on the inactive columns to additional columns when in chart mode, or drag/drop them from bottom to top (this is just the reverse of the behavior ofclick
andshift+click
in Datagrid mode).There is a bug here in
PerspectiveWidget
though - settingplugin
to"d3_y_line"
without also explicitly settingcolumns
should not default to all columns (as it does currently), but rather should display the behavior shown in your screenshot, e.g. choose a single numeric column if one is not already in the 1st position of the column selector. To get the chart you want, the (intended) API would be: