An Example of Test-Driven Development, Part 2

• icocks

  Mr or Mrs. 500

  Points: 557

  More actions

  June 17, 2009 at 2:12 am

  #1010688

  Thanks Andy for an interesting article, like some of the others I'm not sure I can see myself going down the test-driven route in the same way, but it's good to see how it's done.

  Interesting discussion about Alter vs Drop/Create. Here we Alter on the dev db, but deploy with drop/create (with an appropriate permissions script at the end). Only in extremis do we resort to backup/restore or shipping mdfs; apart from being poor practice in that it can mask broken code, there's generally too much hassle with realigning the permissions & getting round collation issues.

  I know this is taking the discussion off-topic but I think there could be some mileage in a SQL Server Mythbusting article to debunk those things everybody 'knows' about SQL Server. Obviously the Exists thing is a prime candidate. Other ones that come to mind are 'Table Variables are always held in memory/Temp tables are always written to disk', 'Subqueries perform worse than joins', 'Cross Joins are always bad' and 'Cursors are always bad' (note the 'always' - I'm not disputing that they're abused an awful lot of the time )

  Cheers,

  Ian

• Andy Leonard

  SSChampion

  Points: 10151

  More actions

  June 17, 2009 at 6:24 am

  #1010834

  Hi Ian,

  That's a good candidate for a Steve Jones Friday poll: What SQL Server myths did you once believe?

  :{> Andy

  Andy Leonard, Chief Data Engineer, Enterprise Data & Analytics

• Adam Gojdas

  SSCommitted

  Points: 1766

  More actions

  July 13, 2009 at 9:43 am

  #1023988

  I know this is taking the discussion off-topic but I think there could be some mileage in a SQL Server Mythbusting article to debunk those things everybody 'knows' about SQL Server. Obviously the Exists thing is a prime candidate. Other ones that come to mind are 'Table Variables are always held in memory/Temp tables are always written to disk', 'Subqueries perform worse than joins', 'Cross Joins are always bad' and 'Cursors are always bad' (note the 'always' - I'm not disputing that they're abused an awful lot of the time )

  I would really like to see such a mythbusting article along with examples that disprove them. I know I have been guilty of believing some of them at one time or another before actually investigating them for myself.

• asimsubedi

  SSC Rookie

  Points: 40

  More actions

  October 18, 2010 at 2:02 am

  #1238254

  nice

• Robert Davis

  One Orange Chip

  Points: 28027

  More actions

  August 16, 2012 at 10:12 am

  #1526705

  Adam Gojdas (7/13/2009)

  ---

  I would really like to see such a mythbusting article along with examples that disprove them. I know I have been guilty of believing some of them at one time or another before actually investigating them for myself.

  If you're not already reading Paul Randal's blog, this is a good reason to start. Paul Randal is the best myth buster out there, in my opinion, and has done a lot of posts on it. In fact, he did a full month of daily posts busting myths last year.

  His blog: http://www.sqlskills.com/blogs/paul/

  Myth busting category: http://www.sqlskills.com/BLOGS/PAUL/category/Misconceptions.aspx

  ---

  My blog: SQL Soldier[/url]
  SQL Server Best Practices: SQL Server Best Practices
  Twitter: @SQLSoldier
  My book: Pro SQL Server 2008 Mirroring[/url]
  Microsoft Certified Master: SQL Server, Data Platform MVP
  Database Engineer at BlueMountain Capital Management[/url]