tSQLt and the INSERT EXEC Problem

  • Comments posted to this topic are about the item tSQLt and the INSERT EXEC Problem

  • Unfortunately you lost me here: "One of these was a procedure that created a prepared SQL query based on any combination of over 30 parameters. The procedure services at least 9 different applications and is executed more than 2 million times a day. "

    Having that many dependencies on a single stored procedure feels like bad practice to me.

  • "Since the SQLCLR works by mapping data from the result set to the target table by column name, the order of the columns doesn’t matter, nor does the number. If the target table has more or less columns than the result set, that doesn’t matter either."

    Isn't this the fatal flaw in the process from a QA perspective? If the addition of new columns to a table are not detected, it seems to me that you have a major weakness in the testing process.

  • Would a local loopback linked server be an easier solution for this?

  • +1

  • Thanks for the article.

Viewing 6 posts - 1 through 6 (of 6 total)

You must be logged in to reply to this topic. Login to reply