Non-ansi characters in object names result in false differences
jensgc
Posts: 10 Bronze 2
I am currently working with a database that was made outside of the IT department (made in MS Access, now being ported to SQL Server). The names of tables and views are made in "natural language" - and since the business users that made the database are Danish, the table and view names include lots of objects with the Danish letters æ, ø and å. This is not an issue in SQL Server - but it is giving me a headache in SQL Source Control.
The problem is that SQL Source Control reports a difference for each object having a non-standard letter in the name. I can try to solve the conflict - using either "keep mine" og "take theirs" - but it doesn't solve anything since the false difference will keep showing up.
I found a similar thread from 2009 (https://forum.red-gate.com/discussion/9754/use-of-a-symbol-results-in-false-differences) where the user was able to handle the false differences using a workaround (setting a different charset in RedGateDatabaseInfo.xml) - but this workaround doesn't help in my setup.
Could someone please let me know if this is a bug within Git or Sql Source Control - and if there is any way to get a working workaround (or a permanent fix).
I am well aware that use of non-standard characters in object naming is a bad idea - and part of my assignment is also to refactor the database in order to have a more solid naming structure. I was just hoping that the SQL Source Control would be able to handle the existing names so I can track whatever changes I make to the DB.
The problem is that SQL Source Control reports a difference for each object having a non-standard letter in the name. I can try to solve the conflict - using either "keep mine" og "take theirs" - but it doesn't solve anything since the false difference will keep showing up.
I found a similar thread from 2009 (https://forum.red-gate.com/discussion/9754/use-of-a-symbol-results-in-false-differences) where the user was able to handle the false differences using a workaround (setting a different charset in RedGateDatabaseInfo.xml) - but this workaround doesn't help in my setup.
Could someone please let me know if this is a bug within Git or Sql Source Control - and if there is any way to get a working workaround (or a permanent fix).
I am well aware that use of non-standard characters in object naming is a bad idea - and part of my assignment is also to refactor the database in order to have a more solid naming structure. I was just hoping that the SQL Source Control would be able to handle the existing names so I can track whatever changes I make to the DB.
