This project is read-only.

Representation of values


text, varchar(max), nvachar(max) of total length longer then about 2500 charaters to be done


text, varchar(max), nvachar(max) containing lines wider than about 4000 characters to be done


Batch / Transaction Controll

I need some way to split the data into chunks (of rows), to keep RAM requirements of input clients low. I.e. to make it practical usable.
It seems not resonable/possible to load complete big tables into tablevariables.

Support for SQL-Server 2000 or SQL-Server 2005

I just happy to forget about 2000 limitations (mostly missing analytical functions), but relative to this project 2000 or 2005 doesn't make much difference.
The main problem is that varchar(max) is missing and has to be replaced by either varchar(8000) or text.
For Tables with small columns similar results are possible. Main part is rewriting the printstringin_lines procedure to work with text, where it is more difficult to split it into lines.
In my working context, I just do not need it. I somebody else provides it, I would add it to the project.


One of the driving tasks for creating this project is my need to move tables with text ciolumns from MS-SQL to Oracle databases.

I tried DTS, SSIS, linked servers etc. All seem to share the same problem, being based on OLEDB they are not able to map CLOB to varchar(max).

I think I could add to SQLPSX a Powershell solution, but as the percentage of my coworker using PowerShell still is low, currently I enforce the pure T-SQL solution.

I didn't try any commercial tools. I guess there are some which attack these kinds of tasks.

Last edited Nov 6, 2010 at 12:57 PM by Bernd_K, version 3


No comments yet.