Date help

[marc_brown98]

I was wondering if anybody has seen this situation. I am using the following: DATE.TAG(@DATE) to insert into 2 different columns, one has the correct date, for instance '2004-05-04 00:00:00.000' and the next column ends up with '4513-02-25 00:00:00.000' the target db is sql server 2000, both fields are defined exactly the same in the db and defined in xfm the same. Any help or work arounds would be greatly appreciated.

Thanks,
Marc

[ds_developer]

I would use a stage variable for your date and then write it to both the columns. It seems that would keep them the same.

John

[marc_brown98]

thanks. I will take a look at doing that.

[chulett]

When you do, consider putting the derivation (DATE.TAG(@DATE)) in the 'Initial Value' area when you define the variable. Then simply use it in your Transformation derivations. This way you incur the overhead of the system call to get the current date once and then simply leverage it after that. In your original layout, you are calling it two times for every record through the job. And it probably isn't changing.

[marc_brown98]

Craig,
Yup, I wasn't even thinking about that, your right. mucho thanks.

[ray.wurlod]

DATE.TAG has been around for a long time. It simply performs Oconv(%Arg1%, "D-YMD[4,2,2]"), so should not have any side effects. So this is very odd behaviour indeed.
Is it reproducible? If so take a look at the generated Transformer stage code (in the RT_BPnnn directory, where nnn is your job number) to see whether you can spot anything in the job logic.

[marc_brown98]

Ray,
I got it straightened out at least it is not doing it, I think I was parsing the file wrong, etc...Thanks for the help.