add STRING_AGG transform to reporter
Bug #2020914 reported by
Llewellyn Marshall
This bug affects 2 people
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
Evergreen |
Confirmed
|
Wishlist
|
Unassigned |
Bug Description
Evergreen 3.12
I've created a small patch that adds String Aggregate (STRING_AGG) as a transform to the reporter module. This allows users to aggregate all of the rows from a query within a comma separated string. Data is automatically cast as text so that any datatype except Boolean can be used with this transform. Duplicate and blank values are skipped.
Changed in evergreen: | |
status: | New → Confirmed |
importance: | Undecided → Wishlist |
tags: | added: reports |
Changed in evergreen: | |
milestone: | none → 3.12-beta |
Changed in evergreen: | |
assignee: | nobody → Terran McCanna (tmccanna) |
To post a comment you must log in.
Llewellyn, I like this a lot.
I have some feedback that may make future-us happy. What do you think of adding two versions of this aggregate:
* "Comma-separated list" -- what you have now, but renamed and relabeled
* "String aggregate with delimiter" -- a version of this that specifies one parameter to collect, which would be the delimiter to use instead of a comma.
I would be in favor of a "include empty strings" version for each that skips the NULLIF() part, but not so strongly in favor that I'd stop inclusion without that.
Thanks!