Description
It would be great to have a visualization mode when the first column is considered as key and all other columns are considered as values.
Example. User has a table with Top-5 CPU consumers:
+---------+----+----+-----+----+-------+------+ |timestamp|java|rngd|rsync|Xorg|Adapter|Others| +---------+----+----+-----+----+-------+------+ | 1|10.0|20.3| 0.0| 5.0| 0.0| 5.0| | 2|31.0|40.0| 0.0| 0.0| 0.0| 28.0| | 3|25.0|40.0| 6.0| 0.0| 0.0| 28.0| | 4|25.0|40.0| 6.0| 0.0| 0.0| 28.0| | 5|25.0|40.0| 6.0| 0.0| 0.0| 19.0| | 6|25.0|40.0| 6.0| 0.0| 0.0| 24.0| | 7|25.0|40.0| 6.0| 0.0| 0.0| 28.0| | 8|25.0|40.0| 6.0| 0.0| 0.0| 26.0| | 9|25.0|40.0| 6.0| 0.0| 0.0| 28.0| | 10|15.0| 0.0| 0.0|10.0| 20.2| 35.0| +---------+----+----+-----+----+-------+------+
User plots this table as a stacked bar-chart (see attached image).
User has to manually assign all columns with process names as values. The problem is that on next execution user has a different set in his Top-5. So user has to repeat this exercise with mouse dragging-dropping every time he wants to visualize new data.
The idea of the improvement is to have a mode first column is always key, others are always values. With this mode on, Zeppelin will update automatically the set of values to dispay each time it plots the data. In general, this mode will be helpful when column names or even their number is not known upfront.