DSXchange

You sure you want to do that. Why exactly do you want to do that?
A compiled job will only return a NOTRUNNING(99) status.

What do you mean by "compiled from command line". You can only reset a job via a shell using dsjob command.

ds_developer wrote:View Data does not use user defined environment variables well. You have to type in the values when it prompts, you can't just use $PROJ_DEF.

Thats why I asked for DataStage version. They fixed that in 7.5 and the OP is on 7.5.1, so that should not be an issue.

Did you try giving method a shot. Its easy and painless with no sorting involved. It will retain the first record of a key in the order it came in.

Thats odd. What version of DataStage are you using? Sounds like a bug.

Why isnt reverse sorting possible? Any who,
Another approach is , that you can write to a hashed file and reference the same hashed file. If the key is found dont write it again. Hence your constraint will be reflink.NOTFOUND.
Make sure your hashed file is locked for updates.

You can write a short before job subroutine that checks what instance it is and pause accordingly. SLEEP command can be used.

First of all, its bad ettiquetes to use all caps as its a sign of shouting. Second of all, you should have mentioned this before that its a fixed width file. A fixed width file will be loaded with spaces to honor the length of the field. But that is if the type is set as 'char'. Keep the spaces as i...

vi the file, press the $ key by holding down the Esc key, does the currsor end up on the last character or after the last character? You might be seeing spaces in the view data but thats just the display definition. There might not be a space.

What is your source datatype? Specify both source and target as varchar (not char). Also try with just TRIM(FIELD)

Trim by setting the type as varchar.

you need to look in the .bad file for more log messages.

This is the phenomenon that Ken likes to call 'back pressure'. Your target stage can only handle the numbers you see.

I think this is not a problem. As noted, a lot of factors are involved. If you dont like this, extract your data onto a file and bulk load the data. That otta make you happy.

Its a transform functioin. Its syntax is exactly what I gave you.

where in.Col is your incoming column.
You can also use OCONV with MCN.