问题
I start with a list of failures that take place in locations
failureName, failureLocation
failure a, location 1
failure b, location 1
failure a, location 2
failure a, location 1
<etc>
I can transform that into this table by using summarize count() by location
failureName, failureLocation, count
failure a, location 1, 100
failure a, location 2, 50
failure b, location 1, 10
<etc>
I'd like to transform the counts into percent on a per. failure basis, so I need to add a sum per failure name. My goal is to end up with this table:
failureName, failureLocation, count, sumPerFailureName
failure a, location 1, 100, 150
failure a, location 2, 50, 150
failure b, location 1, 10, 10
<etc>
Suggestions?
回答1:
Try this, to take you from your 2nd table to the 3rd (and extend a calculated column of the percentage):
let T =
datatable(failureName:string, failureLocation:string, ['count']:long)
[
'failure a', 'location 1', 100,
'failure a', 'location 2', 50,
'failure b', 'location 1', 10,
]
;
T
| summarize sumPerFailureName = sum(['count']) by failureName
| join
(
T
) on failureName
| project failureName, failureLocation, ['count'], sumPerFailureName, percentage = round(100.0 * ['count'] / sumPerFailureName, 2)
来源：https://stackoverflow.com/questions/53163317/add-column-of-totals-pr-field-value