0

My table is in SQL and contains only nvarchar columns. The columns collation is set to Latin1_General_CI_AS . I want to import from SSIS, using flat file connection, some data that contains german charachters like Ö, Ü, etc. In the flat file connection manager i have set the Code page:65001 (UTF-8) and the locale Germany, like in the attachement below and in the preview i have the data accordingly (Ex: Nürnberg). However, when I execute the task and check the data in SQL table , it appears Nürnberg.

Am I missing something in this process?

enter image description here

Pirvu Georgian
  • 657
  • 1
  • 12
  • 37
  • Is your file UTF-8? You might want to try Code page: 1252 (ANSI - Latin I)? – Jarle Bjørnbeth Dec 08 '14 at 08:54
  • What is the codepage in which the source data is encoded? I routinely import Japanese using SSIS. Locale is set to English(India) and codepage is set to 932 (ANSI/OEM - Japanese Shift-JIS). That is the codepage used to create the source file – Raj Dec 08 '14 at 08:55
  • @JarleBjørnbeth thanks for advice. I tried already with 1252 but the package will not be executed " cannot convert between unicode and non-unicode string data types." – Pirvu Georgian Dec 08 '14 at 09:02
  • Ok. Are your nvarchar columns required, or could you just as well use varchar columns in your table? – Jarle Bjørnbeth Dec 08 '14 at 09:15
  • @Raj Hi, I am using 65001 (UTF-8) as code page, with German (Germany) as locale. When I previe the datasource in SSIS, the description is correct with these settings. (Nürnberg not Nürnberg) – Pirvu Georgian Dec 08 '14 at 09:15
  • @JarleBjørnbeth there is no problem. I can also alter my columns to varchar – Pirvu Georgian Dec 08 '14 at 09:16

1 Answers1

1

I am assuming that your file is not Unicode. Since you get the "cannot convert between unicode and non-unicode string data types." error when using the 1252 ANSI code page, and you can change the datatype of your columns. I would just convert them to varchar. Since varchar is not Unicode the error should stop occurring.

The alternative would might be to do some thing like this: Import Package Error - Cannot Convert between Unicode and Non Unicode String Data Type, but I think it would be easier to use varchar columns.

Community
  • 1
  • 1
Jarle Bjørnbeth
  • 429
  • 3
  • 9