Heapers - what is the MS SQL Server safe comparison function for to_number(blah) - if blah is text, it just produces NULL - if you try that in PostgreSQL, your SQL will fail - looking for "safe" conversion function...
Riddle me this: is there a way for a set-based solution (without recursion) to update a row and then refer to the updated version in the next row? Sort of like running total
This is the challenge https://adventofcode.com/2022/day/5
Well, a dynamic sequence of UPDATE statements, or nested SELECTs performing one operation per statement or level doesn't seem very set based. But yes, ok, I understand.
Not as far as I know, but I've only been using KQL for a few weeks now. But I would be surprised if it was there, given the lack of other control commands
For others: Cluster - There is a free tier provided by azure, and there are a few sample ones. Then a client, either through azure data explorer or download a Kusto explorer client locally.
Well, this was a fun trip down to Oracle nonsense. SELECT * FROM <table> WHERE Create_Date = '2022-10-08' yields NULL, even though 1. Create_Date is DATE format 2. There is a row with '2022-10-08'. HOWEVER, SELECT * FROM <table> WHERE TRUNC(Create_Date) = '2022-10-08' does return rows.
@ypercubeᵀᴹ I blame someone not knowing what they were doing when they inserted the data.
Ok, it's that way for most of the data... wtf
Ok, I see now - all the rows for the same date get the same timepart and 90% of the rows were created on one day.
Well, considering the time zone of whatever that is and what data it's being joined to I don't think anything of value is lost if I truncate it on my end.
@ypercubeᵀᴹ Thanks for enlightening me to the pitfalls of Oracle's "date" type
