I have a table in an SQL-Server database, containing a column, called XmlMsg, being of the datatype Xml:
"Columns" definition:
I would like to filter on that column, which first I did using "simple" text filtering:
WHERE XmlMsg LIKE '%status%'
This resulted in following error message:
Argument ...
"Avoid starting your XPaths with "//")" that is interesting. I have not seen that at all. Using the parent axis ".." is really bad but not that. Do you have any sample query that shows it is bad? — Mikael Eriksson36 mins ago
This would be a simple solution. Though OP, keep in mind you said "The big point is that there are over 250,000 entries in that table and I don't want to overload the database by fetching such large query results." and doing an explicit cast to string like this will end up needing to fetch every row first before filtering, in the execution plan. So it is not any more performant than just selecting all rows, for the most part. — J.D.4 mins ago
There's no doubt an active grift out there than relies on telling people result order depends on the clustering key. Buy our merch to support this important cause
I wonder if ol' 300-baud quirky update man ever sold t-shirts
At least I found a use for BIT_COUNT: set-uniqueness constraint dbfiddle.uk/Mr8LPZF6
If only STRING_AGG worked in indexed views. There is no logical reason why it couldn't, it just needs a specialized storage implementation to store the PK with each value, then DML can modify it by taking out any particular value.
@ErikDarling Would that be a bad thing? Seems that the timezones line up well for American users to be up late and its comfortable my afternoon/evening, rather than flags or vandalism or spam posts being left up till the morning.
okay - that's one of the quickest ways to remind people that "the other side of the fence is not always greener" when they wax on about how NZ is awesome
On the other hand, switzerland is $18 USD/gallon as the most expensive fuel in the world.