Your columns should be as of type nvarchar and not varchar. An nvarchar column can store any Unicode data. A varchar column is restricted to an 8-bit codepage.
Here is an sniplet from our PWORKSPACEOBJECT Table:
Good afternoon, yes, I agree with you that you need to use NVARCHAR2 Datatype or VARCHAR2 with Character semantics.
In the Oracle documentation this is described:
But Teamcenter creates a database with UTF-8 encoding using VARCHAR2 with Byte semantics.
I applied to GTAC and they have already confirmed this behavior, but it seems to me that this is a big limitation for using multiple languages in Teamcenter
I will continue to work with GTAC on this issue.
Michael_Ruhnke Your screenshot is not visible
Michael_Ruhnke, thank you. As far as I understand, Teamcenter itself created a database with this type of field for characters? And you can enter 128 characters in the name in any language? (for example in Japanese.)
I just did a quick test how many Chinese or Japanese characters I can enter in a type nvarchar(128) field of a UTF-8 MSSQL Database:
Simplified Chinese characters = 42
Japanese characters = 42
But we are faced with the fact that Teamcenter creates a database for UTF-8 incorrectly, it creates it also if it was not a Unicode encoding, VARCHAR2 (128 BYTE) instead of VARCHAR2 (128 CHAR), which leads to the fact that languages, which use more than one byte will be limited in length, for example, if you use Chinese, then instead of 128 characters you can enter three times less. This is a big problem for using multiple languages in Teamcenter, and it's very strange why this is not implemented well in Teamcenter, and nothing is written about it in the documentation. I think we need to create an IR on the GTAC.
Did you already open an IR at GTAC?
Yes, I already created IR on the GTAC, I wrote about it here
I created my table in Oracle and three fields with different types of characters:
1) VARCHAR2 (128 BYTE)
2) VARCHAR2 (128 CHAR)
3) NVARCHAR2 (128)
I checked the input from SQLDeveloper, not from Teamcenter and tried to enter 128 characters in Japanese.
The first field does not do this because each character in Japanese takes 3 bytes (only 42 characters can be entered). The other two allow you to enter 128 characters in Japanese. But Teamcenter creates fields with symbols using the first option (BYTE).
Michael_Ruhnke, is your Teamcenter corporate server installed on a Windows host? I read the documentation for installing Teamcenter 11.4 Windows Server Installation and there is a Warning in the UTF-8 installation section:
Perhaps someone else installed Teamcenter in UTF-8? Can you confirm this? Our customization for the server (DLLs) will not work in Linux and we would not like to recompile them for Linux
Have you found anything about this warning?! I'm also confused about this section in the installation doc.
I have already installed UTF-8 tc on windows host and did some tests but I'm worried about future ...
I really appreciate your help.
Good afternoon @JFK1963, no, I did not find the answer to this question, we were not satisfied with the restrictions on the length of characters and we refused to install the UTF-8 encoding for Teamcenter. If we decided to use UTF-8 encoding, I would most likely clarify this issue in GTAC.
I think you need to create a IR in GTAC if you are going to use UTF-8 and Teamcenter 11.x
Interestingly, I do not find this warning in the Teamcenter 12 documentation: