encoding issues

rbennet8

I have two screen shots that I can send if you will explain how to attach.

The first shows the text as it appears in a stored procedure script created in SSMS. The second shows the comparison of two versions of a migration script. As you can see the single quote has changed to ? in the migration script. Because we stored translation text for many languages in the database, we always use encode the scripts with UNICODE - codepage 1200. I did not see n option to set this for migrations

Brian Donahue

SQL Source Control should be saving migration scripts as UTF-8, so anything you use to view them (SSMS, notepad) should display international characters correctly. As far as I am aware, you cannot save codepage information in a Unicode document? Maybe I am wrong.

The best explanation I have is that either your source control system is modifying the UTF-8 header or whatever you use to view the text is not interpreting UTF-8 properly.

rbennet8

The issues seems to occur only when I create a deploy script from any 2 versions of the source control projects. We use TFS as our source control server.
This is a snippet of code that is incorrect
--09/23/08 rsb WI 24915 FR2.5 - Add Auto Classify using �x� number of recordings rather than %

It should be
--09/23/08 rsb WI 24915 FR2.5 - Add Auto Classify using ‘x’ number of recordings rather than %
This is the code from both versions of source control, but the deploy script has � instead of '.

Brian Donahue

I double-checked and actually, the .sql files created by the migrations feature do not have a UTF-8 preamble, so that would explain why they do not view correctly. I've loged a bug about this (SOC-3901).