From: anton.txt@g{oogle}mail.com   
      
   Hello, all.  The MSSQL documentation for `UPDATE FROM' is   
   not very clear on how the update source should be specified   
   and how its rows are matched against those of the table   
   being updated. I think the following methods are equivalent   
   and correct:   
      
      -- 1.   
      UPDATE my_alias   
      SET col = data_tab.col   
      FROM upd_tab my_alias   
      JOIN data_tab ON data_tab.code = my_alias.code   
      
      -- 2.   
      UPDATE upd_tab   
      SET col = data_tab.col   
      FROM upd_tab my_alias   
      JOIN data_tab ON data_tab.code = my_alias.code   
      
   But in some old code that seems to have been working for   
   about ten years I have found an `UPDATE' with the following   
   structure:   
      
      -- 3.   
      UPDATE upd_tab   
      SET col = data_tab.col   
      FROM data_tab   
      WHERE data_tab.code = my_alias.code   
      
   which, unlike the previous two commands, works non-   
   deterministically, although there are no more than one row   
   in data_tab from each row in upd_tab. Is it because the FROM   
   clause does not mention upd_tab, whereas it must?  If so,   
   how is the WHERE predicate above interpreted and how does it   
   affect the result?   
      
   --   
   ()  ascii ribbon campaign - against html e-mail   
   /\  http://preview.tinyurl.com/qcy6mjc [archived]   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)